MeghaMukimMarkRobertsEditorsMeghaMukimMarkRobertsEditorsMakingCitiesGreen,Resilient,andInclusiveinaChangingClimateTHRIVINGTHRIVINGMakingCitiesGreen,Resilient,andInclusiveinaChangingClimateTHRIVINGMeghaMukimMarkRobertsEditors©2023InternationalBankforReconstructionandDevelopment/TheWorldBank1818HStreetNW,Washington,DC20433Telephone:202-473-1000;Internet:www.worldbank.orgSomerightsreserved123426252423ThisworkisaproductofthestaffofTheWorldBankwithexternalcontributions.Thefindings,interpretations,andconclusionsexpressedinthisworkdonotnecessarilyreflecttheviewsofTheWorldBank,itsBoardofExecutiveDirectors,orthegovernmentstheyrepresent.TheWorldBankdoesnotguaranteetheaccuracy,completeness,orcurrencyofthedataincludedinthisworkanddoesnotassumeresponsibilityforanyerrors,omissions,ordiscrepanciesintheinforma-tion,orliabilitywithrespecttotheuseoforfailuretousetheinformation,methods,processes,orconclusionssetforth.Theboundaries,colors,denominations,andotherinformationshownonanymapinthisworkdonotimplyanyjudgmentonthepartofTheWorldBankconcerningthelegalstatusofanyterritoryortheendorsementoracceptanceofsuchboundaries.NothinghereinshallconstituteorbeconstruedorconsideredtobealimitationuponorwaiveroftheprivilegesandimmunitiesofTheWorldBank,allofwhicharespecificallyreserved.RightsandPermissionsThisworkisavailableundertheCreativeCommonsAttribution3.0IGOlicense(CCBY3.0IGO)http://creativecommons.org/licenses/by/3.0/igo.UndertheCreativeCommonsAttributionlicense,youarefreetocopy,distribute,transmit,andadaptthiswork,includingforcommercialpurposes,underthefollowingconditions:Attribution—Pleasecitetheworkasfollows:Mukim,Megha,andMarkRoberts,editors.2023.Thriving:MakingCitiesGreen,Resilient,andInclusiveinaChangingClimate.Washington,DC:WorldBank.doi:10.1596/978-1-4648-1935-3.License:CreativeCommonsAttributionCCBY3.0IGOTranslations—Ifyoucreateatranslationofthiswork,pleaseaddthefollowingdisclaimeralongwiththeattribution:ThistranslationwasnotcreatedbyTheWorldBankandshouldnotbeconsideredanofficialWorldBanktranslation.TheWorldBankshallnotbeliableforanycontentorerrorinthistranslation.Adaptations—Ifyoucreateanadaptationofthiswork,pleaseaddthefollowingdisclaimeralongwiththeattribution:ThisisanadaptationofanoriginalworkbyTheWorldBank.ViewsandopinionsexpressedintheadaptationarethesoleresponsibilityoftheauthororauthorsoftheadaptationandarenotendorsedbyTheWorldBank.Third-partycontent—TheWorldBankdoesnotnecessarilyowneachcomponentofthecontentcontainedwithinthework.TheWorldBankthereforedoesnotwarrantthattheuseofanythird-party-ownedindividualcomponentorpartcontainedintheworkwillnotinfringeontherightsofthosethirdparties.Theriskofclaimsresultingfromsuchinfringementrestssolelywithyou.Ifyouwishtore-useacomponentofthework,itisyourresponsibilitytodeterminewhetherpermissionisneededforthatre-useandtoobtainpermissionfromthecopyrightowner.Examplesofcomponentscaninclude,butarenotlimitedto,tables,figures,orimages.AllqueriesonrightsandlicensesshouldbeaddressedtoWorldBankPublications,TheWorldBankGroup,1818HStreetNW,Washington,DC20433,USA;e-mail:pubrights@worldbank.org.ISBN(paper):978-1-4648-1935-3ISBN(electronic):978-1-4648-1936-0DOI:10.1596/978-1-4648-1935-3Coverdesign:Voilà:LibraryofCongressControlNumber:2022919547vCONTENTSForeword..............................................................................................................................................................xiiiAcknowledgments.................................................................................................................................................xvMainMessages...................................................................................................................................................xviiAbbreviations......................................................................................................................................................xxiOverview......................................................................................................................1Introduction......................................................................................................................................................1Tailoredpolicyoptionsbytypeofcityandinstrument..............................................................41Notes.................................................................................................................................................................45References.......................................................................................................................................................48Introduction..............................................................................................................53Overview.........................................................................................................................................................53Aframeworkforassessingthegreenness,resilience,andinclusivenessofacity’sdevelopment...................................................................................................................................56Notes.................................................................................................................................................................58References.......................................................................................................................................................58PART1:WHEREAREWENOW?Chapter1:TheStylizedRelationships..................................................................63Introduction...................................................................................................................................................64Achangingclimate......................................................................................................................................66Howgreenarecitiestoday?....................................................................................................................74Howresilientarecitiestoday?...............................................................................................................89Howinclusivearecitiestoday?..............................................................................................................96Summaryandconclusions.....................................................................................................................110Annex1A:Spatialdistributionsofextremeweatheranomalies.............................................111Notes...............................................................................................................................................................115References.....................................................................................................................................................117Chapter2:AGlobalTypologyofCities................................................................125Introduction.................................................................................................................................................125Howcurrentchallengesvaryacrossdifferenttypesofcities..................................................134Howclimatechange–relatedhazardsvaryacrosscitiesglobally.........................................143Summaryandconclusions.....................................................................................................................150Annex2A:Methodologyfordefiningaglobaltypologyofcities............................................152Notes...............................................................................................................................................................156References.....................................................................................................................................................157viTHRIVINGSpotlight:MultidimensionalExclusionandExposuretoAirPollutioninPeruvianCities..................................................................................161Introduction.................................................................................................................................................161Methodology...............................................................................................................................................161Measurement...............................................................................................................................................162Profilesofmultidimensionalexclusion............................................................................................162Experiencewithmultidimensionalexclusionandexposuretooutdoorairpollution......163Notes..............................................................................................................................................................164References.....................................................................................................................................................165PART2:WHATDOWEKNOW?Chapter3:TheImpactsofClimateandEnvironmentalChangeonCities.....169Introduction.................................................................................................................................................169Howdoesclimatechangeaffectcitiesdirectly?...........................................................................170Howdoesclimatechangeaffectcitiesindirectly?......................................................................180Howdoesclimatechangeaffectdifferentpopulationsincities?..........................................188Summaryandconclusions....................................................................................................................207Notes..............................................................................................................................................................208References....................................................................................................................................................210Chapter4:TheImpactofCitiesonClimateandtheEnvironment................227Introduction................................................................................................................................................228Pancakesorpyramids—Howdocitiesaroundtheworldevolveinform?........................229Theimpactsofverticaldevelopmentonlandconsumption,theenvironment,andproductivity.......................................................................................................................................233Whatroledoes(horizontal)transportationplayinshapingcitiesandtheirenvironmentalimpacts?........................................................................................................................239Howwillurbanexpansionaffectagriculturallandandproduction?..................................246Howwillurbangrowthaffectcompetitionforwaterbetweencitiesandagriculturallands?....................................................................................................................................253Summaryandconclusions....................................................................................................................258Notes..............................................................................................................................................................260References....................................................................................................................................................262PART3:HOWDOWEGETITDONE?Chapter5:PoliciesforPromotingGreen,Resilient,andInclusiveUrbanDevelopment..............................................................................271Introduction................................................................................................................................................272Whatarethechoices?.............................................................................................................................272viiContentsWhomakesthechoices?.......................................................................................................................293Howtomakechoices?............................................................................................................................304Summaryandconclusions.....................................................................................................................313Notes..............................................................................................................................................................314References.....................................................................................................................................................316BoxesO.1Definingaglobaltypologyofcities..............................................................................................3O.2Howthisreportdefinesinclusiveness.........................................................................................17O.3ThefiveI’s:Information,incentives,insurance,integration,andinvestments.......31I.1Defininggreenness,resilience,andinclusiveness................................................................53I.2ThefiveI’s:Information,incentives,insurance,integration,andinvestments.......551.1Comparingappleswithapples:Howthisreportdefinescities.......................................651.2Whatisaweatheranomaly,andhowdoesthischapterdefineit?...............................671.3Whendoesaweteventbecomeaflood?.................................................................................721.4Hotinthecity:Thecausesandimpactsoftheurbanheatislandeffect.....................741.5Greenhousegasaccountingandthefourscopesofemissions.......................................771.6Shouldacity’scarbondioxideemissionsbeadjustedforinternationalandinternaltrade?............................................................................................................................781.7HowdidLondongetridof(mostof)itssmog?.....................................................................871.8Beijing’safforestationmiracle......................................................................................................901.9Comparinglevelsofsocialinclusioninurbanversusruralareas...................................981.10Citiestendtohaveslightlylowerlevelsofsocialcohesionthanruralareas..........1001.11Unhealthydietsandpoorlygovernedfoodsystemsforurbanconsumers.............1082.1Definingandapplyingtheglobaltypologyofcities..........................................................1273.1Howdoclimateshocksandstressorsdiffer?.......................................................................1713.2Measurementissues.......................................................................................................................1763.3Citybudgetsbearthebruntofclimateevents....................................................................1783.4Doesclimatechangeleadtoviolenceandmigration?....................................................1863.5Howwatersuppliesfromproximatenaturalsourcesarestrugglingtokeepupwiththeneedsofnearbycitiesandtowns..........................................................1873.6IntensifyingurbanheatinJohannesburgwilldisproportionatelyaffectthepoor..................................................................................................................................1893.7Measuringenvironmentalriskincities..................................................................................1963.8Theeconomicandsocietalconsequencesofurbanheatarepervasive.....................1974.1Abriefhistoryoftheelevatoror“verticalrailway”...........................................................2304.2Whatdohigh-risebuildingsmeanforurbaninclusion?................................................2354.3High-occupancyvehiclepolicyinJakarta............................................................................2434.4Willelectriccarssavetheday?..................................................................................................2434.5FiveSharedSocioeconomicPathways...................................................................................2474.6Pandemicrisksattheinterfaceofhumans,animals,andtheenvironment...........2504.7Sinkingcities:Groundwaterdepletionandlandsubsidence........................................2565.1Cancitiesbeincubatorsofgreeninnovation?....................................................................2795.2Subnationalcatbonds—Dotheywork?.................................................................................281viiiTHRIVING5.3Buildingwater-securecities:Usinginformation,incentives,andinvestments......2905.4Adoptingthepublic-privatedialogueforlocalclimateaction.....................................2965.5Theimportanceofvoiceindecision-making......................................................................3025.6Effectivenessofparticipatoryresponsestoprepareforclimate-relateddisasters.............................................................................................................3035.7Areworkerspreparedforthegreentransition?.................................................................308FiguresO.1Citiesinhigh-andupper-middle-incomecountriesemitthemostCO2andcontributethemosttoglobalurbanCO2emissions.....................................................7O.2GlobalGHGemissionswillremainabovethelevelrequiredtolimitglobalwarmingto1.5°Ciflow-andlower-middle-incomecountriescontinuetofollowtheircurrentpolicies............................................................................................................8O.3Projectedclimatechange–relatedhazardsarestrongestforcitiesinlow-andlower-middle-incomecountries...........................................................................................9O.4Citiesinlow-andlower-middle-incomecountriesarelessresilienttoclimatechange–relatedshocksandstresses..........................................................................11O.5Forcitiesinlower-incomecountries,dry,hot,andwetweatheranomalieshavelargernegativeimpactswhentheyreinforcebaselineclimaticconditions..................................................................................................................................12O.6Withincountries,constructionoftallbuildingsofallheightsisoccurringfastestincitiesthatwillbemostaffectedbyclimatechange.........................................14O.7Accesstoimprovedsanitationandsafelymanageddrinkingwaterisworseincitiesinlow-andlower-middle-incomecountries............................................16O.8Citiesinlow-andmiddle-incomecountrieshaveworseairpollutionthancitiesinhigh-incomecountries.........................................................................................18O.9PM2.5andCO2emissionsarestronglypositivelycorrelatedacrosscitiesglobally..................................................................................................................19O.10Morecompactdevelopmentisassociatedwithloweremissionsfromtheresidentialandtransportationsectors...................................................................21O.11Citiesthatdevelopverticallyaccommodatemorepeople,aremoreprosperous,andconsumelessland..........................................................................................23O.12Averagelevelsofvegetationarelowestforcitiesinupper-middle-incomecountries...............................................................................................................................................25O.13Droughtisassociatedwithfasterbuilt-upareagrowthinlow-andmiddle-incomecountries......................................................................................................27O.14Aframeworkformakingcitiesgreener,moreresilient,andmoreinclusive..............30O.15Rangesofprojectedclimatechange–relatedhazardscoresacrosscitiesforselectedcountries...........................................................................................................37I.1Reportframeworkandstructure.................................................................................................561.1Evolutionofthefrequencyandintensityofextremeheateventsforcitiesglobally,1958–69to2010–20...........................................................................................681.2Evolutionofthefrequencyandintensityofextremecoldeventsforcitiesglobally,1958–69to2010–20...........................................................................................691.3Evolutionofthefrequenciesofextremedryandweteventsforcitiesglobally,1958–69to2010–20.......................................................................................................70ixContents1.4Shareofcities,bydeepestthree-plusyearsofwaterdeficitsandcountryincomegroup,thatfaceddayzero–typeevents,1992–2013...........................711.5Evolutionofthefrequencyoftropicalcyclonesforcitiesglobally,1958–69to2010–20.........................................................................................................................731.6AverageCO2emissionspercapita,bycountryincomegroupandgeographicregion,2015..................................................................................................................761.7ShareofglobalCO2emissionsgeneratedincities,bycountryincomegroupandgeographicregion,2015.............................................................................791.8Averagesharesoflong-cycleCO2emissionssourcesforcities,bycountryincomegroup,2015....................................................................................................801.9DeterminantsofCO2emissionsforcitiesglobally,residentialandtransportationsectors,2015.........................................................................................................811.10RelationshipbetweencitycompactnessandCO2emissionsacrosscitiesglobally,residentialandtransportationsectors,2015...........................................831.11RelationshipbetweenCO2andPM2.5emissionsacrosscitiesglobally,residentialandtransportationsectors,2015.........................................................................841.12DeterminantsofPM2.5emissionsforcitiesglobally,residentialandtransportationsectors,2015.........................................................................................................851.13AverageofPM2.5concentrationsacrosscities,bycountryincomegroup,2000and2015.......................................................................................................861.14Relationshipbetweenaveragegreennessandpopulationsizeacrosscitiesglobally,2014,andaveragegreennessandlevelofdevelopmentacrosscitiesglobally,1990,2000,and2014...........................................................................891.15Estimatedimpactsofhot,cold,wet,anddryanomaliesoncitiesinlow-andlower-middle-incomecountrieswithhot,cold,wet,anddrybaselineclimates,April2012–December2020......................................................................96B1.9.1Socialinclusionindexforurbanandruralareas,bygeographicregion,andrelationshipbetweenurban–ruralgapinsocialinclusionindexandlevelofdevelopmentacrosscountries,circa2020..............................................................................99B1.10.1Socialcohesionindexforurbanandruralareas,bygeographicregion,andrelationshipbetweenurban–ruralgapinsocialcohesionindexandlevelofdevelopmentacrosscountries,circa2020............................................................................1011.16Povertyrates(US$1.90andUS$3.20)forthreetypesofurbanandruralareas,selectedcountries,2015..................................................................................................1021.17Householdaccesstosafelymanageddrinkingwaterandinfantmortalityrates,byurbanandruralareas,selectedregions,circa2015.....................1031.18Relationshipbetweenacity’slevelofdevelopmentanditsaccesstosafelymanageddrinkingwateranditsinfantmortalityrate,selectedregions,circa2015..........................................................................................................................1041.19Reportedlevelsofspousalviolenceexperiencedbycurrentlymarriedwomenandofwomenwhobelievewifebeatingisjustified,bytypeofurbanandruralarea,selectedgeographicregions,circa2015...................................................1051.20Relationshipbetweenincomeinequalityandcitysizein16LatinAmericanandCaribbeancountries,circa2014,andinIndonesia,2017...................1061.21Obesity,bygeographicregion,andhypertensioninIndia,bytypeofurbanandruralarea,circa2015................................................................................................108xTHRIVINGB1.11.1Intakeoffoodgroups,byurbanandruralandmaternaleducation,187countries....................................................................................................................................1092.1Inglobaltypology,numberofcities,byregionandcountryincomegroup.............1302.2Inglobaltypology,distributionofpopulation,bycitysizeandcountryincomegroup....................................................................................................................1312.3Inglobaltypology,distributionoflargecoastalandinlandcities,bylocationandcountryincomegroup,andshareofcoastalcities,bycitysizeandcountryincomegroup..................................................................................................1332.4Inglobaltypology,levelofseverity,bycitysizeandcountryincomegroup:Inclusivenessindicators...............................................................................................................1372.5Inglobaltypology,levelofseverity,bycitysizeandcountryincomegroup:Resilienceindicators....................................................................................................................1382.6Inglobaltypology,levelofseverity,bycitysizeandcountryincomegroup:Greennessindicators......................................................................................................................1392.7Distributionofcitiesinclimatechange–relatedexposuredataset,bycitysizeandcountryincomegroup..................................................................................1452.8Averageweightedoverallclimatechange–relatedhazardexposure,bycitysizeandcountryincomegroup..................................................................................1472.9Averageclimateexposurescoresforsixclimatechange–relatedhazards,bycitysizeandcountryincomegroup..................................................................................1482.10Urbanexposuretocombinedclimatechange–relatedhazards,selectedcountries,bycountryincomegroup.......................................................................1512A.1ComparisonsofadjustedR2acrosscombinationsofcitysizeandcountryincomegroup...................................................................................................................154S1.1Adjustedheadcountratio(M0),incidence(H),andintensity(A)ofmultidimensionalexclusion,bytypeofareaofPeru,2019.............................................162S1.2Incidence(H)versusintensity(A)ofmultidimensionalexclusionacrossPeru’scities..........................................................................................................................163S1.3Multidimensionalexclusionandoutdoorairpollution:Atypologybycity,Peru,2019....................................................................................................1643.1Marginaleffectsofremotelocationwithinacityonexposuretoexcessiveheat,byeducationlevelandconsumptionpercapita,Bamako,Mali.....1923.2PovertyratesatUS$1.90perday,bydegreeofurbanizationclassification,selectedcountries.............................................................................................2043.3Predictedprobabilityofescapingpoverty,bytownpopulationsizeandfloodrisk,ColombiaandChile.........................................................................................2063.4Predictedprobabilityofescapingpoverty,bylocation,Indonesia,1993–2014.....2064.1Evolutionoftallbuildingsandstructuressince1880......................................................2294.2Evolutionoftotalsumoftallbuildingheights:HongKongSAR,China;NewYork;andTokyo,1890–2020...........................................................................................2324.3Estimatedelasticitiesofpopulation,landarea,andnighttimelightintensitywithrespecttototalsumoftallbuildingheights..........................................2354.4Relationshipbetweencarownershipandurbanpopulationdensities,bycitylevelandcountrylevel.........................................................................................................2404.5EstimatedimpactsofsubwaysystemopeningsbetweenAugust2001andJuly2016onairpollutionlevels,58cities.........................................246xi4.6Projectedchangesingloballandareadevotedtocrops,pasture,forests,andurbanusesunderfivescenariosofclimatechangemitigationandadaptation,2010–2100.................................................................................2494.7ProjectedcropandlivestocklossesfromurbanexpansionunderSSPscenariosby2040and2100.......................................................................................................2525.1Shareofgovernmentspendingbylocalgovernments,byselectedregionsandcountries....................................................................................................................295B5.4.1Dimensionsofadaptabilityofthepublic-privatedialogue...........................................296MapsBO.1.1Globaltypologyofcities...................................................................................................................41.1Percentagechangeinaveragegreennessacrosscitiesglobally,1990–2014..............901A.1Frequencyofextremehotmonthsperyear,2011–20.......................................................1111A.2Intensityofextremehotmonths,2011–20...........................................................................1111A.3Frequencyofextremecoldmonthsperyear,2011–20.....................................................1121A.4Intensityofextremecoldmonths,2011–20.........................................................................1121A.5Frequencyofextremedrymonthsperyear,2011–20.......................................................1131A.6Intensityofextremedrymonths,2011–20...........................................................................1131A.7Frequencyofextremewetmonthsperyear,2011–20.......................................................1141A.8Intensityofextremewetmonths,2011–20..........................................................................1142.1Globaltypologyofcities...............................................................................................................1292.2Geographicdistributionofoverallclimatechange–relatedexposure,bycitysizeandcountryincomegroup..................................................................................1463.1UrbanexpansioninNiamey,Niger,byshareofinformalsettlements,1985–2021........................................................................................................................................184B3.6.1Residentsofhotterneighborhoodsfacedaytimetemperaturesthatareupto6.5°Chigher,Johannesburg,SouthAfrica...........................................................1903.2EducationalattainmentandenvironmentalriskwithinIndonesianmultidistrictmetros......................................................................................................................1943.3DisplacementmobilitypatternsfollowingCycloneYaas,eastcoastofIndiaandtheBengalareaofBangladesh,May2021..............................2034.1Examplesofmoreandlesswalkableurbanenvironments............................................239B4.7.1ComparativelandsubsidenceratesacrossIndonesia,2000s,andAsianmegacities,1900–2010.....................................................................................................257Tables1.1Estimatedimpactsofvarioustypesofweatheranomalyontheintensityofacity’snighttimelights,bycountryincomegroup,April2012–December2020...........................................................................................................942.1Categoriesofcitysizeandlevelofdevelopmentusedtodefineglobaltypologyofcities.............................................................................................................................1282.2Summaryofglobaltypologyofcities......................................................................................1312.3Inglobaltypology,severityofchallenges,bycitysizeandcountryincomegroup....................................................................................................................135ContentsxiiTHRIVING2.4Methodologicalreferencetableforprojectedclimatechange–relatedexposurescores,2030–40...........................................................................................................1442.5Intersectionbetweenurbanchallengesandclimateshocksandstresses...............1492.6Urbanexposuretocombinedclimate-relatedhazard,bycountryincomegroup.............................................................................................................1512A.1Criteriaforderivingaglobaltypologyofcities...................................................................1535.1Principlesfordefiningtheroleofnationalandcitygovernmentsinclimatemitigation/adaptation.................................................................................................2985.2Tailoredpolicyoptions,bytypeofcityandinstrument..................................................310xiiiForewordThepast50yearshaveseenbothaquadruplingofglobalurbanpopulationandarapidlychangingclimate,withrisingsurfacetemperaturesandsealevelsandincreasingfrequencyofextremeweatherevents.Fast-growingcities—whichofferarangeofopportunities—cangiverisetoawidevarietyofstresses,especiallyiftheirurbanizationisnotwell-managed.Anunpredictableandfast-changingclimatecompoundstheseunderlyingstresses.Theimpactsofclimatechange–relatedshocksoncitiesmaybesignificant;formanyhouse-holds,theycanbedevastating.Cities,bothsmallandlarge,indevelopingcountriessufferdisproportionatelywhenconfrontedbyextremehotanddryweatherevents,aswellasbytropicalcyclones.Usingdatafromacross10,000citiesglobally,thisreportasksfourimportantquestions:Howgreen,resilient,andinclusivearecitiestoday?Howdoesclimatechangeaffectcitiesandpeopleincities?Howdoesthegrowthofcitiesimpacttheclimateand,moregenerally,theenvironment?Andfinally,whatpolicieswillhelpmakecitiesgreener,moreresilient,andmoreinclusive?Climatechangeisalsoasymptomofalargerproblem—theerosionofnaturalcapital,towhichpoorlymanagedurbanizationcontributes.Thiserosioncontributes,inturn,todangerouslypoorairqualityinmanycities,detrimentalcompetitionforwaterbetweenurbanandruralareas,unnecessarylossoffertileagriculturalland,deforestation,andlossofbiodiversity.Thesetrendsareplayingthemselvesoutagainstabackdropofhighandrisinglevelsofinequalityinmanycitiesgloballyandstalledprogressintheworldwidefightagainstextremepoverty.Thesetrendsbothinteractwithandreinforceclimatechange–relatedstressorstoaffectthegreen-ness,resilience,andinclusivenessofurbandevelopment.Thisreportprovidesacompasstohelppolicymakers—bothlocalandnational—meettheirobjectivestomakecitiesgreener,moreresilient,andmoreinclusive.Itoutlineswhatpolicyinstrumentsareavailable;whowieldstheseinstruments;andhowpolicychoicescouldbetailored,prioritized,andsequencedforeffectiveimplementation.Policymakerscandrawonfivebroadsetsofpolicyinstrumentsthatconstitutethe“fiveI’s”:information,incentives,insurance,integration,andinvestments.Earlywarninginformationcanhelpsavelives,property,andinfrastructure.Accurateinformationthatreflectsriskscanhelpgovernments,individuals,andbusinessesmakebetterdecisions.Incentivesareneededtomotivatepeopleandbusinessestoactontheavailableinformationandtakeaccountoftheimpactsoftheirowndecisionsontheenvironmentandonothers.Insurancecanhelpminimizethefinancialimpactofdisasters,complementingadaptationstrategies.Integrationwithincities,towhichwell-implementedplanningiskey,isgoodforthepoorandgoodforbudgets,helpingminimizeunnecessarysprawlandbringingpeopleclosertojobsandopportu-nities.Integrationacrosscities—throughmeasuresthathelppromotethemovementofpeople,goods,andservices—canhaveadampeningeffectonshocksandstresses.Finally,investmentscanbeusedtoanticipate,prevent,andrespondtoshocks,aswellastoretrofitbuildingsandinfrastructureinresponsetostresses.xivTHRIVINGItiswellknownthatcitiesare“enginesofgrowth”atboththelocalandnationallevels.However,lessisknownabouthowurbandevelopmentandclimatechangeareinteracting,andthisflagshipreportmakesanimportantcontributiontoourcollectiveunderstandingofcitiesandclimatechange.IfirmlybelievethattheinsightsfromthisreportwillprovidevaluableguidancefortheWorldBankGroupaswecollaboratewithpartnersandclientstohelpcitiesnotonlysurvivebutalsothriveinthefaceoftheperilsofclimatechange.Actionnowispossible,actionnowisnecessaryandurgent,andactionnowiswhereweshouldfocusourefforts.JuergenVoegeleVicePresidentforSustainableDevelopmentWorldBankxvAcknowledgmentsThisreportwaspreparedbyateamledbyMarkRobertsandMeghaMukim.ThecoreteamalsoconsistedofPaoloAvner,PaolaMarcelaBallonFernandez,JonathanBower,VladimirChlouba,JoseAntonioCuestaLeiva,MaitreyiDas,ChandanDeuskar,FelipeDizon,BennyIstanto,RemiJedwab,NicholasJones,LuciaMadrigal,ShoheiNakamura,SammyNdayizamvye,JanePark,NeraliPatel,NataliaPecorari,LuisQuintero,GiuseppeRossitti,StevenRubinyi,DmitrySivaev,BenjaminStewart,RuiSu,EigoTateishi,ZoeTrohanis,AishwaryaVenkat,TakahiroYabe,EshaZaveri,andTianyuZhang.AdditionalwritteninputsintothereportwereprovidedbyGautengCity-RegionObservatoryresearchersGraemeGötz,GillianMaree,andLavenNaidoo.TheworkwasconductedunderthegeneralguidanceofJuergenVoegele(VicePresident,SustainableDevelopment),RichardDamania(ChiefEconomist,SustainableDevelopment),SamehWahba(RegionalDirector,EuropeandCentralAsia,SustainableDevelopment),andBerniceVanBronkhorst(GlobalDirector,Urban,DisasterRiskManagement,Resilience,andLand).Theteamwasfortunatetoreceiveexcellentadviceandguidancefromthefollowingpeerreviewersatvariouspointsinthereportpreparationprocess:LouiseCord,MarianneFay,MatthewKahn,MarkLundell,MartinRama,HarrisSelod,AnnaWellenstein,andMingZhang.Thesereviewersarenotresponsibleforanyremainingerrors,omissions,orinterpre-tations.Theteamalsobenefitedgreatlyfromthefeedbackprovidedbythereport’sexternalandinternalpanelsofadvisersatvariouskeypointsduringthereportpreparationprocess.Externaladviserstothereport,whohavenotalreadybeenacknowledged,includedGillesDuranton,RemaHanna,andEstebanRossi-Hansberg.InternaladviserstothereportincludedMadhurGautam,StephaneHallegatte,RuthHill,SomikLall,andKantaKumarRigaud.SomikLallalsoplayedaninstrumentalroleinearlydiscussionsregardingthereport’sscopeandthebackgroundresearchthatshouldfeedintoit.Theteamfurthergratefullyacknowledgestheexceptionalfeedbackprovidedonearlydraftsofthereport’soverviewbyThomasFarole,FrancisGhesquiere,andCatalinaMarulanda.Preparationofthereportalsobenefitedgreatlyfromthefeedbackreceivedfromthediscussantsandotherparticipantsduringanauthors’workshopthatwasheldonFebruary24and28,2022.DiscussantswhohavenotalreadybeenacknowledgedwerePaulaRestrepoCadavid,LucChristiaensen,OliviaD’Aoust,MathildeLebrand,EllenMoscoe,TannerRegan,andForhadShilpi.Theteamisfurthergratefultothenumerousothercolleagues—includingMartinHeger,AngelHsu,GhazalaMansuri,AugustinMaria,JoannaMasic,andCraigMeisner—whocontributedinsightsduringtheteam’sconversationswiththem.SabraLedentwasthesubstantiveeditor,andNoraMarawasthecopyeditor;Voilà:wasresponsibleforthedesign,layout,andvisualizationsofthereport’sstand-aloneOverviewcoverandtext.MaryFisk,JewelMcFadden,andDeborahAppel-BarkeroftheWorldBank’sformalpublishingunitwererespon-sibleforthedesign,typesetting,printing,anddisseminationofboththehardcopyanddigitalversionsofthefullreport,whilealsoprovidinginputsintotheproductionofthestand-aloneOverview.Last,butnotleast,wethankSreypovTepandKaiXinNellieTeoforunfailingadministrativesupport.ThisworkreceivedgenerousfinancialandtechnicalsupportfromtheCityResilienceProgramandtheGlobalFacilityforDisasterReductionandRecovery.xviiMainMessagesBetween1970and2021,thenumberofpeoplelivingincitiesincreasedfrom1.19billionto4.46billion,whiletheEarth’ssurfacetemperatureclimbedby1.19°Caboveitspreindustriallevel.Becauseoftheprosperitytheyhavehelpedgenerate,citieshavebeenamajorcauseofthisclimatechange.Itisalsoincities,however,thatmanyofthesolutionstotheclimatecrisiswillbefound,notleastbecauseby2050almost70percentoftheworld’spopulationwillcallcitieshome.Thisreportcombinesoriginalempiricalanalysisofaglobalsampleofmorethan10,000citieswithinsightsfromsecondaryliteraturetotakestockofhowgreen,howresilient,andhowinclusivecitiesaretoday,andtoexaminethetwo-wayinterplaybetweencitiesandclimatechange.Informedbythisanalysis,thereportprovidesacompassforpolicymakersonhowtohelptheircitiesbecomegreener,moreresilient,andmoreinclusive—inotherwords,onhowtohelptheircitiesthrive—inachangingclimate.Achangingclimate••Climatechangeisexposingcitiestoincreasinglyfrequentextremeweatherevents.Fromthe1970stotheperiod2010–20,thefrequencyofextremeheatanddryeventsincreasedacrosscitiesglobally,andthefrequencyofextremeweteventshasincreasedsincethe1990s.Globalsea-levelriseofabout0.125millimetersperyearisalsoincreasingtheriskoffloodingforcoastalcities.Howgreen,howresilient,andhowinclusivearecities?••Citiesinhigh-andupper-middle-incomecountriesaremajorcontributorstoclimatechange,whereasthecontributionofcitiesinlower-incomecountriesismodest.Globally,about70percentofanthropogenicgreenhousegasemissions,thebulkofwhicharefossilcarbondioxide(CO2)emissions,emanatefromcities.Citiesinlower-incomecountries,however,accountedforonlyabout14percentofallglobalurbanCO2emissionsin2015,andcitiesinlow-incomecitiescontributedlessthan0.20percent.Themitigationchallengeforcitiesinlower-incomecountriesistodevelopwithoutfollowingthehistoricCO2emissionstrajectoriesofcitiesinhigher-incomecountries.••Citiesinlow-andlower-middle-incomecountriesfacethehighestexposuretoprojectedclimatechange–relatedhazards.Projectedexposurefor2030–40forthesecities—basedonacompositeindexthatcombinesprojectionsforsixkeyhazards(floods,heatstress,tropicalcyclones,sea-levelrise,waterstress,andwildfires)—isconsiderablyhigherthanforcitiesinhigher-incomecountries.••Citiesinlow-andlower-middle-incomecountriesarelessresilienttoincreasinglyfrequentclimatechange–relatedshocksandstresses.Thesecitiessufferlargernegativeimpactstotheirlocallevelsofeconomicactivityfromextremehot,dry,andwetweatherevents,aswellasfromtropicalcyclones,thandocitiesinhigher-incomecountries.Theimpactsofextremeweatherforcitiesinlower-incomecountriesareparticularlypronouncedwhentheyreinforceacity’sbaselineclimaticconditions.xviiiTHRIVING••Citiessufferindirectimpactsofclimatechange,especiallyinlow-andlower-middle-incomecountries.Theseindirectimpactsoccurthroughavarietyofchannels.Forexample,whenextremeweathereventshit,peopleinthecountrysideoftenseeksafeharborincities.Extendeddroughtsinruralareasresultinfasterexpansionofurbanareas.Theresultingnewsettlementsareofteninformalandestablishedontheoutskirtsofcities,inurbanfloodplainswithlimitedaccesstoservices.••Constructionincountriesisgravitatingtowardcitiesthatwillbemostaffectedbyclimatechange.Sincethe1960s,constructionincountrieshasincreasinglygravitatedtowardcitiesprojectedtobecomeunbearablyhotbecauseofclimatechange—theoppositeofwhatwouldbeexpectedinthefaceofintensifyingchangesinclimate.••Lackofinclusivenesscontributestothelackofresilienceofcitiesinlow-andlower-middle-incomecountries.Thislackofresiliencecanbeexplained,inpart,bythesecities’higherratesofpovertyandlowerlevelsofaccesstobasicservicessuchashealthcareandeducation;water,electricity,andotherutilities;solidwastemanagement;digitalandfinancialservices;andemergencyrescueservices.••Citiesinlow-andmiddle-incomecountriesarelessgreenintermsofairpollution,andairpollutionfromkeyurbansectorspresentsagreaterchallengeforlargercitiesincountriesatallincomelevels.Onaverage,concentrationsofPM2.5(particulatematterof2.5micronsorlessindiameter)inboth2000and2015werelowerincitiesinhigh-incomecountriesthanincitiesinlower-incomecountries.Andacity’sPM2.5emissionsinitsresidentialandtransportationsectors—sectorsthaturbanplanningandpoliciescanmostdirectlyinfluence—tendtoincreasewithitspopulation.••Policiesthatimproveairqualitycanhelpcitiesbothmitigateandadapttoclimatechange.Manyoftheactivitiesthatcontributetopoorurbanairquality,suchasindustrialactivitiesanddrivinginternalcombustionenginevehicles,alsocontributetoglobalclimatechange.Consistentwiththisfinding,acrosscitiesglobally,fortheresidentialandtransportationsectors,astrongpositivecorrelationexistsbetweenCO2andPM2.5emissions.••Citiesthatdevelopverticallyconsumelessland,accommodatemorepeople,andaremoreprosperous.Acrosscitiesglobally,adoublingofacity’stotalheightleadstoaroughly16percentlong-runincreaseinitspopulationanda19percentlong-runreductioninitslandarearelativetoothercities.Theseresultsareaccompaniedbya4percentlong-runincreaseintheintensityofthecity’snighttimelightspercapita,whichsuggestsincreasedprosperity.••Lackofvegetation,especiallyevidentinlargecitiesandcitiesinupper-middle-incomecountries,canexacerbatetheimpactsofextremeheateventsincities.Itdoessobecausealackofvegetationexacerbatestheurbanheatislandeffect,whichcanleadtourbanlandsurfacetemperaturesthataremorethan10°Chigherthantheequivalentrurallandsurfacetemperatures.ApolicycompasstohelpcitiesthriveAthrivingcityisonethatisgreen,resilient,andinclusiveinthefaceofachangingclimate.Thisreportpresentsgeneralconclusionsrelatedtotherealizationofthisvisionintheformofthreequestionspolicymakersshouldanswer:Whatpolicyinstrumentsareavailable?Whowieldstheseinstruments?Howcanpolicychoicesbasedontheseinstrumentsbeprioritizedandsequencedforeffectiveimplementation?xixMainMessages••WHAT:PolicyoptionstaketheformoffiveI’s:information,incentives,insurance,integration,andinvestments.Inmanyinstances,theinterdependenciesbetweenthesesetsofinstrumentsplayoutincomplementaryways,whereinpoliciesacrossthebundlesstrengthenimpactswhenimplementedtogether.••WHO:Because“traditional”urbanstressesinteractwithclimatechange–relatedstressestodetermineoutcomes,localgovernmentsarewell-placedtodriveclimateaction.Cities,workingwithotherstakeholdersincludingnationalgovernments,theprivatesector,andcivilsociety,haveanimportantpolicywedgeattheirdisposal.••HOW:Toensuretheircitiesthrive,policymakerswillneedtotogglebetween,andsandwichtogether,bundlesofpolicyoptionsdrawnfromthefiveI’s.Thecombinationofinterventions,theirsequencing,andtheprioritizationofoutcomeswillvarydependingonthecharacteristicsofcities,includingtheirlevelofrisk,levelofdevelopment,andsize.xxiAbbreviationsCO2carbondioxideCSOcivilsocietyorganizationEIDemerginginfectiousdiseaseEVelectricvehicleFLLoCAFinancingLocallyLedClimateAction(Kenya)GDPgrossdomesticproductGHGgreenhousegasGHSGlobalHumanSettlementGRIDgreen,resilient,andinclusivedevelopmentIDPinternallydisplacedpersonIVinstrumentalvariableMDEmultidimensionalexclusionμg/m3microgramspercubicmeterMTAMetropolitanTransitAuthority(NewYork)OAPoutdoorairpollutionPM2.5particulatematterof2.5micronsorlessindiameterPPDpublic-privatedialogueSEDLACSocio-EconomicDatabaseforLatinAmericaandtheCaribbeanSSPSharedSocioeconomicPathwayTCIPTurkishCatastropheInsurancePoolWHOWorldHealthOrganizationWUIwildland-urbaninterface1IntroductionForatleast50years,theviewthathumanactivityhasspurredtheworld’swarminghasbeensupportedbyscientificevidence,theweightofwhichisnowbeyonddispute(Benton1970;IPCC2021;MaddenandRamanathan1980).1Globallyduringthistime,thenumberofpeoplelivingincitieshasalmostquadrupled2andtheEarth’ssurfacetemperaturehasclimbedbynearly1.2°Caboveitspreindustriallevels.3Thiswarminghasbeenassociatedwithanincreasedfrequencyofextremehot,dry,andweteventsacrosscitiesworldwide.4Globalsea-levelrisehasalsoincreasedtheriskoffloodingformanycoastalcities.Becauseoftheprosperitytheyhavehelpedgenerate,citieshavebeenanimportantcauseofthisclimatechange(Kahn2010).5Atthesametime,thisprosperityhashelpedmakecitiesmoreresilienttoclimatechange–relatedshocksandstressors.Citieshavealsobecomeincreasinglyvocaladvocatesofclimateaction;6however,intheracebetweenclimatechangeandclimateaction,climatechangeretainsacommandinglead.Citiesinhigh-andupper-middle-incomecountries,whichaccountforthebulkofglobalurbancarbondioxide(CO2)emissions,arenotmovingquicklyenoughtowardnetzero.Similarly,althoughtheircurrentcontributionstoclimatechangemaybesmall,citiesinlower-incomecountriesarenotactingfastenoughtomoderatetheiremissionstrajectories.Thesetrajectories,ifleftunchecked,willeventuallyoffsetanyreductionsinglobalemissionsmadebycitiesinhigher-incomecountries.Poorlymanagedurbanizationalsocontributestoanevenlargerproblem—themoregeneralerosionofnaturalcapital.7Thiserosiontakestheformnotonlyofpollutedskiesbutalsoofcontaminatedwaterbodies,destroyednaturalhabitats,andthelossofbothplantandanimalspecies.8Inadditiontonotactingquicklyenoughtomitigateclimatechange,cities,especiallythoseinlow-andlower-middle-incomecountries,arealsonotadaptingquicklyenoughtoitschallenges.Theresidentsofcitiesinlower-andeveninhigher-incomecountriesmayseeclimatechangeasasecondaryconcern,especiallywhenpittedagainstpoverty,inequality,andalackofaccesstomarketsandservices—problemsthatforsomepeopleandsomecitieshaveworsenedovertime.AsillustratedbyFrance’s“yellowvest”protests,importanttrade-offsundoubtedlyexistbetweensuchproblemsandcertainpoliciesthataimtotackleclimatechange.9Thegoodnews,however,isthatcomplementarypoliciescanhelpeasethesetrade-offs,ascanpoliciesthatmakecitiesmoreinclusivewhilesimultaneouslyhelpingthembecomebothgreenerandmoreresilienttoclimatechange.Inthiscontext,howinclusiveacityistodayisalsoanimportantdeterminantofhowwellitcancopewiththeclimatechange–relatedshocksandstressesofthefuture.Overview2THRIVINGToensurethatcitiesthriveinaworldconfrontedbyclimatechange,policymakersatbothnationalandlocallevelsneedtoworktogethertoimplementboldpoliciestoaddresstheinter-relatedstressesthatarisefromclimatechangeandurbangrowth.Theseincludethestressesarisingfromthepressureofacity’spopulationonitssuppliesofland,housing,andbasicservices;itsstockofinfrastructure;anditsenvironment.10Ifnotwellmanaged,suchstressescangiverisetoslumsandsprawl,deterioratinglevelsandqualityofbasicserviceprovision,streetsgridlockedwithpollutingcarsandmotorcycles,theexcessiveconversionoffertileagriculturallandtourbanuses,chokingairpollution,andheightenedgreenhousegas(GHG)emissions.Drawingonawidevarietyofdatasources,thisreportcombinesoriginalempiricalanalysiswithinsightsfromadiverserangeofsecondaryliteraturetotakestockofhowgreen,howresilient,andhowinclusivecitiesare,andtoshedlightontheinteractionofstressesrelatedtourbangrowthwiththoserelatedtoclimatechange.Toaddressthoseinterrelatedstresses,policymakersneedtoenlisttheuseoffivebroadsetsofpolicyinstruments—information,incentives,insurance,integration,andinvestments—inshort,thefiveI’s.Thereportprovidesacompasstohelpcitiestailortheuseoftheseinstrumentstotheirowncircumstancesandproblems.3Howgreen,howresilient,andhowinclusivearecitiestoday?Totakestockofhowgreen,howresilient,andhowinclusivecitiesaretoday,thisreportdefinesaglobaltypologyofmorethan10,000cities,measuringacity’sgreenness,resilience,andinclusivenessusingavarietyofindicators(boxO.1).Basedontheanalysisofthistypologyandtheindicatorsmoregenerally,aswellasonthereport’sotherglobalanalysis,10keyfindingsemerge.DefiningaglobaltypologyofcitiesThisreportmeasuresacity’sgreenness,resilience,andinclusivenessusingavarietyofindicators.Forgreenness,theseindicatorsincludeabsoluteandpercapitaproduction-basedfossilfuelcarbondioxideemissions,emissionsandconcentrationsofparticulatematterof2.5micronsorlessindiameter,andmeasuresofacity’slevelandextentofgreeneryorvegetation.Forresilience,theyincludeestimatesofthesizeofimpactsofweathereventsonacity’saggregatelevelofeconomicactivity.Indicatorsofinclusivenessincludelevelsofaccesstobasicservicessuchasimprovedsanitationandsafelymanageddrinkingwater,povertyrates,andlevelsofintracityhouseholdincomeinequality.aAlthoughcitiesvarywidelyontheseindicators,somegeneralpatternsareneverthelessevident,withmanyrelatedtobothacity’spopulationsizeanditslevelofdevelopment.Thesepatternsallowdefinitionofaglobaltypologythatdistinguishesbetweenninetypesofcity—small,medium,andlargecitiesinlow-andlower-middle-,upper-middle-,andhigh-incomecountries(mapBO.1.1)—andtherelativeseverityofthegreenness,resilience,andinclusivenesschallengestheyface.Chapter2offersafulldiscussionofthistypologyandtherelativeseverityofchallengesthatdifferenttypesofcitiesface.a.Inadditiontotheseindicators,thereportalsodiscussesarangeofotherdimensionsofthegreenness,resilience,andinclusivenessofcitiesandhowthesedimensionsrelatetoclimatechange(seealsoboxO.2).BoxO.1SmallCitysizeCountryincomelevelMediumLargeHigh-incomeUpper-middle-incomeLow-incomeandlower-middle-incomeSmallCitysizeCountryincomelevelMediumLargeHigh-incomeUpper-middle-incomeLow-incomeandlower-middle-income4THRIVINGMapBO.1.1Globaltypologyofcities5OverviewSource:WorldBankcalculationsbasedondatafromtheEuropeanCommission’sGlobalHumanSettlement(GHS)UrbanCentreDatabaseR2019(https://ghsl.jrc.ec.europa.eu/ghs_stat_ucdb2015mt_r2019a.php).Note:CitiesaredefinedasurbancentersfollowingtheEuropeanCommission’sdegreeofurbanizationmethodology(Dijkstraetal.2021;DijkstraandPoelman2014).Small,medium,andlargecitiesarethosethatin2015hadapopula-tionof50,000–199,999;200,000–1.4999million;and1.5millionormore,respectively.6THRIVINGKEYFINDING1Citiesinhigh-andupper-middle-incomecountriesaretheleastgreengloballyintermsofCO2emissions,whereascitiesinlower-incomecountriesbarelycontributetoglobalemissions.Globally,about70percentofanthropogenicGHGemissions,thebulkofwhicharefossilCO2emissions,emanatefromcities(Hopkinsetal.2016).11Onapercapitabasis,citiesinhigh-andupper-middle-incomecountrieshavethehighestfossilCO2emissions,andthoseinlow-incomecountrieshavethelowest(figureO.1,panela).Indeed,in2015averagepercapitaemissionsincitiesinhigh-incomecountrieswerealmost18timeshigherthanthoseofcitiesinlow-incomecountries,whereasthoseincitiesinupper-middle-incomecountriesweremorethan21timeshigher.Higheraveragepercapitaemissionsincitiesinhigh-andupper-middle-incomecountriesalsotranslateintohighersharesofglobalurbanCO2emissions(figureO.1,panelb).In2015,thesecitiestogetheraccountedfornearly86percentofallglobalurbanCO2emissions.Citiesinlower-middle-incomecountriescontributedalmost13percentandcitiesinlow-incomecountrieslessthan0.2percent.WhenfocusingonjusturbanfossilCO2emissionsfromtheresidentialandtransporta-tionsectors—thesectorsthaturbanplanningandpoliciescanmostdirectlyinfluence—evenstrongerpatternsemerge.In2015,averagepercapitaemissionsfromthesesectorsincitiesinhigh-incomecountriesweremorethan76timesthoseincitiesinlow-incomecountriesandmorethan10timesthoseincitiesinlower-middle-incomecountries(figureO.1,panela).Whereascitiesinhigh-incomecountriesaccountedfor48percentoftotalglobalurbanemissionsfromtheresidentialandtransportationsectors,citiesinlow-incomecountriesaccountedforlessthan0.4percent(figureO.1,panelb).12Thepictureisclear:citiesinhigh-andupper-middle-incomecountriesarethemajordriversofglobalurbanCO2emissionsandthereforeoftheurbancontributiontoglobalclimatechange.Bycontrast,citiesinlow-incomecountriesbarelyregisterintermsoftheircontribution.Thus,fromamitigationperspective,thechallengefacingcitiesinhigh-andupper-middle-incomecountriesishowtoreducetheirhighcurrentlevelsofCO2emissions.Citiesinlow-and,toalesserextent,lower-middle-incomecountriesfaceadifferentchallenge—howtodevelopwithoutfollowingtheCO2emissionstrajectorieshistoricallyfollowedbycitiesinhigher-incomecountries.FigureO.2depictstheimportanceofthechallengeforcitiesinlow-andlower-middle-incomecountries.Itshowsthat—evenif(anditisaverybigif)high-andupper-middle-incomecoun-triescanmakeasuccessfulgreentransitionconsistentwithnetzeroCO2emissionsby2050—globalGHGemissionswillremainabovethelevelrequiredtolimitglobalwarmingto1.5°Caslongaslow-andlower-middle-incomecountriesfollowtheircurrentpolicies.13Thus,acom-parisonofpanelaofthefigurewithpanelbrevealsthattotalglobalGHGemissionsin2050willremain4.2timesthelevelrequiredtokeepwarmingwithin1.5°Cifcurrentpoliciesremainthesameinlow-andlower-middle-incomecountrieswhilehigher-incomecountriesachievenetzero.Eveniflower-middle-incomecountrieswerealsotoachievenetzeroby2050butlow-incomecountriesweretocontinuewiththeircurrentpolicies,GHGemissionswouldremain60percenthigherthanrequiredtolimitglobalwarmingto1.5°C.7OverviewAverageCO2emissionspercapitaandshareofglobalCO2emissionsgeneratedincities,bycountryincomegroup,2015a.AverageCOemissionspercapitab.Shareofglobaltotal(%)COemissionsgeneratedincities020406004812High-incomeAllsources(restrictedsample)ResidentialandtransportationUpper-middle-incomeLower-middle-incomeLow-incomeHigh-incomeUpper-middle-incomeLower-middle-incomeLow-incomeTonnesperyearperpersonFigureO.1Citiesinhigh-andupper-middle-incomecountriesemitthemostCO2andcontributethemosttoglobalurbanCO2emissionsSource:WorldBankanalysisbasedondatafromtheEuropeanCommission’sGlobalHumanSettlement(GHS)UrbanCentreDatabaseR2019(https://ghsl.jrc.ec.europa.eu/ghs_stat_ucdb2015mt_r2019a.php),whichderivesitscarbondioxide(CO2)emissionsdatafromtheEuropeanCommission’sEmissionsDatabaseforGlobalAtmosphericResearch(EDGARv5.0).Fortheresidentialandtransportationsectors,thedatacover10,179cities.Forallsourcesofemissions,thedatacover3,148cities.Note:Inpanela,eachmarkershowstheunweightedaverageoflong-cycle(fossil)CO2emissionspercapita(measuredintonnesperyearperperson)ofcitiesbycountryincomegroup.Inpanelb,eachmarkershowstheshareofglobalurbanlong-cycle(fossil)CO2emissionsgeneratedincitiesclassifiedbycountryincomegroup.8THRIVINGFigureO.2GlobalGHGemissionswillremainabovethelevelrequiredtolimitglobalwarmingto1.5°Ciflow-andlower-middle-incomecountriescontinuetofollowtheircurrentpoliciesSource:WorldBankanalysisbasedonhistoricalemissionsdatafromClimateWatch(2022)andemissionsprojectionsfromtheNetworkofCentralBanksandSupervisorsforGreeningtheFinancialSystem(NGFS)v.2scenariosdata.Note:GHG=greenhousegas;MtCO2e=metrictonsofcarbondioxideequivalent.HistoricalandprojectedaggregateGHGemissionstrajectoriesunderdierentscenarios,bycountryincomegroup,1990–20500510152025303540451990205019902050a.Ifallcountriestransitiontonetzeroby2050b.Iflow-andlower-middle-incomecountriescontinuewithcurrentpolicieswhiletheresttransitiontonetzeroby2050GHGemissions(MtCOe,thousands)1.5CtargetProjectionsProjectionsHigh-income-Upper-middle-income-Lower-middle-income-Low-income-Emissionsreach4.2xthelevelrequired9OverviewKEYFINDING2Citiesinlow-andlower-middle-incomecountriesfacethehighestlevelsofprojectedclimatechange–relatedhazards.Lookingforward,althoughfewcitiesgloballywillescapetheeffectsofclimatechange,citiesinlow-andlower-middle-incomecountriesfacethehighestoveralllevelsofprojectedclimatechange–relatedhazards.Evidenceofthissituationisprovidedbyanindicatorthatcombinesinformationonsixkeyhazards—floods,heatstress,tropicalcyclones,sea-levelrise,waterstress,andwildfires—projectedforwardto2030–40.14Thus,amongtheninetypesofcityidentifiedbythisreport’sglobaltypology(boxO.1),large,medium,andsmallcitiesinlow-andlower-middle-incomecountrieshavethehighestaverageclimatehazardexposurescores(figureO.3).Formediumandlargecities,thesehighaveragescoresaredrivenmainlybyprojectedfloodhazards.Forsmallandmediumcities,theyaredrivenbyprojectedwater,sea-levelrise,andheatstresshazards.Wildfiresalsocontributetohigherprojectedclimatehazardscoresforallsizesofcityinlow-andlower-middle-incomecountries.LargecitiesAverageweightedoverallclimatechange–relatedhazardexposure,bycitysizeandcountryincomegroupMediumcitiesSmallcitiesClimatehazardexposurescoreHigh-incomeUpper-middle-incomeLow-andlower-middle-income404550556065FigureO.3Projectedclimatechange–relatedhazardsarestrongestforcitiesinlow-andlower-middle-incomecountriesSource:WorldBankanalysisbasedondatafromMoody’sESGSolutions,Sub-SovereignPhysicalClimateRiskScores,October2021(https://esg.moodys.io/climate-solutions).Note:Thefigurereportsthemeanprojectedclimatehazardexposurescoresforcitiesthatbelongtoagiventype.Themeanscoresareestimatedbyregressingacity’sscoreonaseriesofdummyvariablesforthedifferenttypesofcity.Small,medium,andlargecitiesarethosethatin2015hadapopulationof50,000–199,999;200,000–1.4999million;and1.5millionormore,respectively.10THRIVINGKEYFINDING3Citiesinlow-andlower-middle-incomecountriesarelessresilienttoincreasinglyfrequentclimatechange–relatedshocksandstresses.Historically,citieshave,withafewexceptions,exhibitedremarkablelong-runresilience15tomanyformsofphysical,andevenhumancapital,destruction.Theirresilienceincludesthattoeventssuchasearthquakes,widespreadflooding,majorfires,pandemics,andevenlarge-scale,includingnuclear,bombing(Glaeser2022).Notwithstandingthisimpressivelong-runresilience,however,itisestimatedthatcitiesinlow-andlower-middle-incomecountriessufferlargernegativeimpactsfromextremehot,dry,andwetweatherevents(or“anomalies”),aswellasfromtropicalcyclones,ontheirlocallevelsofeconomicactivitythandocitiesinhigher-incomecountries.16Thisdisparitycomesatatimewhenclimatechangeisincreasingboththefrequencyandintensityofextremehot,dry,andwetevents.17Thus,althoughextremeweathereventshaverelativelylittleimpactonthelevelsofeconomicactivityofcitiesinhigh-andupper-middle-incomecountries(asproxiedbytheirnighttimelightintensities)inthemonthswhenthoseeventsoccur,theyhavemuchlargernegativeimpactsoncitiesinlow-andlower-middle-incomecountries(figureO.4).18Estimatesalsosuggestthat—forcitiesinlow-andlower-middle-incomecountries—hot,wet,anddryweatheranomalieshavemoreseverenegativeimpactsoneconomicactivitywhentheymirroracity’sbaselineclimate.Thus,hot,wet,anddryanomalieshavelargernegativeimpactsoncitieswithhot,wet,anddrybaselineclimates,respectively(figureO.5).Thiseffectismostevidentfordryanomaliesincitiesinlow-andlower-middle-incomecountrieswithdrybaselineclimates.SuchcitiesareparticularlyprevalentintheMiddleEastandNorthAfricaandpartsofSub-SaharanAfrica.Theincreasingfrequencyofextremedryeventscontributestothegrowingnumberofcitiesgloballyexperiencingnear“dayzero”events,wherebywatersuppliesareonlyweeksordaysfromrunningout.Inaworst-casescenario,estimatessuggestthatawarmingworldcouldmakedayzero–typedroughts100timesmorelikelythantheywereintheearlytwentiethcenturyincertainregions(Pascaleetal.2020).Suchdwindlingwatersuppliescancostacityuptoa12-percentage-pointlossingrossdomesticproductandleadtodamagingcompetitionbetweenurbanandruralareasforwatersuppliesascitiesencroachonsurroundingareastosatisfytheirthirst.Intheabsenceofequitablelegalarrangements,transfersofwaterfromruraltourbanareascanevenbecoercive,suchasinChennai,India(Singhetal.2021;Varadhan2019;Zaverietal.2021).Inallocationdecisions,mostlegalsystemsgivehigherprioritytodrinkingwater,andoftentoindustrialwater,thantoagriculturalwater.Suchprioritizationcanreducethewateravailableforirrigatedurbanandperi-urbanagriculture(Hoekstra,Buurman,andvanGinkel2018).Thefactthat,ingeneral,citiesinlow-andlower-middle-incomecountriesexperiencemoresevereestimatednegativeimpactsofweatheranomaliesthandocitiesinhigher-income11OverviewMediumcities–15.0High-incomeUpper-middle-incomeLow-andlower-middle-incomeLargecitiesEstimatedimpact(%)–10.0–5.00Estimatedimpactofextremeweather(hot,dry,wet,andtropicalcyclone)eventsonacity’slevelofnighttimelightintensity,April2012–December2020FigureO.4Citiesinlow-andlower-middle-incomecountriesarelessresilienttoclimatechange–relatedshocksandstressesSource:DerivedfromParkandRoberts2023.ResultsarebasedonanalysisofmonthlycompositesofnighttimelightsderivedfromVisibleInfraredImagingRadiometerSuite(VIIRS)satellitedata(https://payneinstitute.mines.edu/eog-2/viirs/),monthlyweatherdatafromClimatologyLab,TerraClimate(https://www.climatologylab.org/terraclimate.html),andtropicalcyclonedatafromInternationalBestTrackArchiveforClimateStewardship(https://www.ncdc.noaa.gov/ibtracs/).Note:Thehorizontalaxisshowstheunweightedaverageestimatedimpactofextremehot,dry,andwetweathereventsandtropicalcyclonesonacity’snighttimelightinthemonthwhentheeventoccurred.Foranygivenmonth,anextremehot,dry,orweteventisdefinedasoneinwhichtheweathervariable(temperatureorprecipitation)deviatesbyatleast2standarddeviationsfromacity’sownlong-runhistoricalaverageforthatvariable,withthataveragecalculatedusingmonthlydatafortheperiodJanuary1958–March2012.Atropicalcycloneisdefinedasacategory2orstrongercyclonebasedontheSaffir-Simpsonwindscalethatoccurswithin200kilometersofacity’sgeographiccenter.Mediumandlargecitiescorrespondtothosethatin2015hadapopulationof200,000–1.4999millionand1.5millionormore,respectively.ThecolorsofthemarkersfordifferenttypesofcitiescorrespondtothecolorsinmapBO.1.1.countriesisconsistentwithagreaterlevelofresilienceonthepartofthelatter.Theseestimatespaintapartialpictureofresilience,however,becausetheyconsideronlytheimme-diateimpactsofaweathershockwhiledisregardingthesubsequentpathofrecoveryofeconomicactivity,orthelackthereof.Nevertheless,relatedresearchthatalsouseslightsdatasuggestsaquickerreboundofeconomicactivityforacityinanupper-middle-orhigh-incomecountryinresponsetoafloodeventthanforacityinalow-orlower-middle-incomecountry.Thus,althougheconomicactivityisrestoredtoitspreshocklevelinacityinhigher-incomecountrieswithinonemonth,restorationinacityinalower-incomecountrywilltaketwomonths(Gandhietal.2022;Lalletal.,forthcoming).12THRIVINGFigureO.5Forcitiesinlower-incomecountries,dry,hot,andwetweatheranomalieshavelargernegativeimpactswhentheyreinforcebaselineclimaticconditionsSource:DerivedfromParkandRoberts2023.ResultsarebasedontheanalysisofnighttimelightsmonthlycompositesderivedfromVisibleInfraredImagingRadiometerSuite(VIIRS)satellitedata(https://payneinstitute.mines.edu/eog-2/viirs/)andmonthlyweatherdatafromClimatologyLab,TerraClimate(https://www.climatologylab.org/terraclimate.html).Note:Eachmarkershowstheestimatedimpactonacity’snighttimelightintensityofa1-standard-deviationdepartureoftherelevantweathervariable(eithertemperatureorprecipitation)fromacity’sownmonthlylong-runaverageforthatweathervariable,withthataveragedefinedusingmonthlydatafortheperiodJanuary1958–March2012.Thehorizontallinesindicatetheupperandlowerboundsofthe95percentconfidenceintervalassociatedwiththecorrespondingestimatedimpact.Hotandwetcitiesareinthetophalfoftheglobaldistributionofthelong-runmeanmonthlytemperatureandprecipitation,respectively,andcoldanddrycitiesareinthebottomhalf.Estimatedimpactsofdry,hot,wet,andcoldanomaliesoncitiesinlow-andlower-middle-incomecountrieswithdry,hot,wet,andcoldbaselineclimates,respectively,April2012–December2020–6.0–4.0–2.002.04.0DryHotWetColdEstimatedimpact(%)13OverviewInthefaceofintensifyingclimatechange–relatedhazards,onemightexpectthatconstructionincountrieswouldmoveawayfromcitieswhoseclimaticconditionsareprojectedtodeteriorate(thatis,“futurebadlocations”)andtowardcitieswhoseclimatewillbelessaffectedormayevenimprove.Sincethe1960s,however,theoppositehasoccurred.Constructionincountrieshasincreasinglygravitatedtowardcitiesprojectedtobecomeunbearablyhotbecauseofclimatechange.EvidenceofthisfindingappearsinfigureO.6,which,foranygivenyear,showstheestimatedeffectofafuturebadlocationindexontheaggregatedheightofacity’sbuildings.Highervaluesofthisindexindicatethatacity’saveragemaximumtemperatureduringitshottestseasonisprojectedtopassathresholdof43°Csooner—forexample,anindexvalueof0indi-catesthatacityisnotprojectedtopassthisthresholdduringthecurrentcentury,whereasanindexvalueof5(themaximumvalue)indicatesthatthecitypassedthisthresholdduringtheperiod1995–2014.Ifconstructionweremovingawayfromfuturebadlocations,onewouldexpectthisindextohaveanincreasinglynegativeimpactontheaggregatedheightofacity’sbuildingsastherealityofthetransitiontounbearablyhottemperaturesbecomesmoreevident.FigureO.6showsinsteadanincreasinglypositiveimpactofthefuturebadlocationindexontheaggregatedheightofacity’sbuildings.Moreover,thistrendisevidentforbuild-ingsofallheights,from55meters(roughly15stories)orhigherto195metersorhigher,whichincludestheworld’stallestskyscrapers.Itisalsoevidentforbuilt-uparea,whichincludeslow-riseurbandevelopmentmoregenerally(DesmetandJedwab2022).Thetrendofrisingconstructionincitiesdeemed“futurebadlocations”(evenasthatfutureapproaches)isemergingdespiteincreasingpublicawarenessofclimatechangeanditspotentialimpacts.Consistentwiththisfinding,themajorinternationalconferenceshavehadnodiscernibleimpactonconstructiontrends.Suchconferencesincludethe1985Villachconference19andConferenceofthePartiessessionsheldinDohain2012(COP18)andParisin2015(COP21),whichincreasedglobalawarenessofthethreatposedbyclimatechange.AdoptionoftheKyotoProtocolin1997hadnoeffecteither.Moreover,notonlyhasconstructionincountriesbeenmovingtowardcitiesthatarefuturebadlocations,butalsoovertheperiod1985–2015theglobalgrowthofurbanbuilt-upareainhigh-riskfloodzoneshasoutpacedthatinlow-riskfloodzones.Thistrendhasbeenmostevidentinmiddle-incomecountries,especiallyupper-middle-incomecountries(Rentschleretal.2022).Thesetrendssuggestaspatialmisallocationofinvestmentsinbuildings,withnegativepotentialimpactsonthefuturehealth,safety,andwelfareofpopulations.Thus,becausebuildings,especiallytallbuildings,aredurablestructuresthatdepreciateonlyslowlyoverdecades,theseconstructionpatternsrisklockinginurbandevelopment,andthereforeurbanpopulations,insuboptimallocationsthatwillbemostaffectedbyclimatechange.KEYFINDING4Constructionincountriesisgravitatingtowardcitiesthatwillbemostaffectedbyclimatechange.14THRIVINGFigureO.6Withincountries,constructionoftallbuildingsofallheightsisoccurringfastestincitiesthatwillbemostaffectedbyclimatechange(“futurebadlocations”)Source:BasedonDesmetandJedwab2022,whichusesdataforcitiesfromtheEuropeanCommission’sGlobalHumanSettlement(GHS)UrbanCentreDatabaseR2019(https://ghsl.jrc.ec.europa.eu/ghs_stat_ucdb2015mt_r2019a.php)andonbuildingheightsfromtheEmporisdatabase.Note:Foreachyearshownonthehorizontalaxis,theverticalaxisshowstheestimatedimpact,relativetothebaseyearof1915,ofafuturebadlocationindexontheaggregatedheightofacity’sbuildingsaboveacertainspecifiedheightlevel(55meters,100meters,120meters,140meters,170meters,and195meters).Thefuturebadlocationindexhasavaluebetween0and5,where0,1,2,3,4,and5indicatethatacityisprojectedtosurpassanaveragemaximumtemperatureof43°Cafter2100,during2080–99,during2060–79,during2040–59,during2020–39,andduring1995–2014,respectively.Apositiveestimatedimpactindicatesmovementofconstructiontowardcitiesthatarefuturebadlocations.Theregressionfromwhichimpactsareestimatedincludesbothcityfixedeffectsandcountry-yearfixedeffects.Estimatedeectsof“futurebadlocation”indexonconstructionofbuildingsabovevariousheights,1920–2020Eectof“futurebadlocation”indexonbuildingheights1985:Villachconference1997:KyotoProtocol≥55metersHeightofbuildings:.08.06.04.020≥100meters≥120meters≥140meters≥170meters≥195meters1915193019451960197519902005202015OverviewThemovementofurbandevelopmentincountriestowardlocationsfacingthegreatestclimatechange–relatedhazardsisparticularlyworrisomeforlow-andlower-middle-incomecountriesbecausetheircitieslackresiliencetotheclimatechange–relatedshocksandstresseshigh-lightedinkeyfinding3.Thisrelativelackofresiliencecanbeexplained,inpart,bythesecities’higherratesofpovertyandlowerlevelsofaccesstobasicservicessuchashealthcareandeducation;water,electricity,andotherutilities;solidwastemanagement;digitalandfinancialservices;andemergencyrescueservices.Forexample,householdsincitiesinlow-andlower-middle-incomecountriesinEastAsiaandPacific,LatinAmericaandtheCaribbean,SouthAsia,andSub-SaharanAfricahavelowerlevelsofaccesstoimprovedsanitationandsafelymanageddrinkingwaterthandohouseholdsincitiesintheupper-middle-incomecountriesintheseregions(figureO.7).20Forimprovedsani-tation,smallandmediumcitiesinlow-andlower-middle-incomecountrieshaveparticularlylowlevelsofaccess.Moregenerallywithinincomeclasses,largercitiestendtoprovidebetteraccesstoservicesand,thus,inthissenseatleast,tobemoreinclusive.Thisfindingstemsinpartfromthefactthatlargercitiescanspreadoveragreaterpopulationthefixedcostsofthelarge-scaleinfrastructurethatunderpinstheprovisionofsuchservices.Moregenerally,inclusioncanbedefinedastheabilityandopportunityofallwhoresideinacitytofullyparticipateinmarkets,services(includingdigitalandfinancialservices),andspaces(includingpolitical,physical,cultural,andsocial),therebyenablingthemtoleadtheirliveswithdignity(boxO.2).Theabilityofitsresidentstoparticipateinmarkets,services,andspacescontributestoacity’sresiliencethroughavarietyofchannels.Forexample,par-ticipationinlabormarketsfacilitatesincomegrowth,whichhelpsprovidehouseholdswiththeresourcestoinvestinself-protectionagainstclimatechange–relatedshocks,whilealsoallowingthemtoaccumulatesavingsthatcanactasaformofself-insuranceagainstsuchshocks.Participationinfinancialmarketscansimilarlyassisthouseholdsinbufferingandinsuringagainstshocks.Meanwhile,theirabilitytoparticipateinpolitical,physical,cultural,andsocialspaceshelpsprovideacity’sresidentswithvoice.Thisvoicecan,inturn,leadtopoliciesthataremoreinclusiveofotherwisemarginalizedgroupsinsociety,therebyhelpingbuildtheirresilience.KEYFINDING5Lackofinclusivenesscontributestothelackofresilienceofcitiesinlow-andlower-middle-incomecountries.16THRIVINGFigureO.7Accesstoimprovedsanitationandsafelymanageddrinkingwaterisworseincitiesinlow-andlower-middle-incomecountriesSource:WorldBankcalculationsusingdataonhouseholdaccesstoimprovedsanitationandsafelymanageddrinkingwaterfromHendersonandTurner(2020)anddownloadedfromhttps://doi.org/10.7910/DVN/YZ46FJ.Note:Inbothpanels,theshareofhouseholdsiscalculatedforcitiesin40countries(3intheEastAsiaandPacificregion,5inLatinAmericaandtheCaribbean,3inSouthAsia,and29inSub-SaharanAfrica).Inpanelb,safelymanageddrinkingwaterisdefinedasallimprovedwatersourcesthattakezerominutestocollectorareonthepremises.Improvedwatersourcesincludeallpipedwaterandpackagedwater,protectedwellsorsprings,boreholes,andrainwater.Small,medium,andlargecitiescorrespondtothosethatin2015hadapopulationof50,000–199,999;200,000–1.4999million;and1.5millionormore,respectively.ThecolorsofthemarkersfordifferenttypesofcitiescorrespondtothecolorsinmapBO.1.1.6065707580859095Shareofhouseholdswithaccesstoimprovedsanitationandsafelymanageddrinkingwaterbycitytype,circa2015a.Accesstoimprovedsanitationb.AccesstosafelymanageddrinkingwaterShareofhouseholds(%)LargecitiesMediumcitiesSmallcitiesLargecitiesMediumcitiesSmallcitiesUpper-middle-incomeLow-andlower-middle-income17OverviewHowthisreportdefinesinclusivenessInthisreport,inclusivenessisbroadlyconsideredintermsof(1)abilityandopportu-nityand(2)outcomes.Inclusionisdefinedastheabilityandopportunityofallwhoresideinacitytofullyparticipateinmarkets,services,andspaces(includingpolitical,physical,cultural,andsocial),therebyenablingthemtoleadtheirliveswithdignity(WorldBank2013).Consistentwiththatdefinition,thisreportvariouslydiscusseshowcitiesdiffergloballyintheaccesstheyprovidetobasicurbanservices,financialservices,digitaltechnologies,andlabormarketopportunities.Italsoshinesaspotlightonmultidimensionalexclusionandtouchesonissuesofvoice.Asforoutcomes,thisreportanalyzes,amongotherthings,howcitiesvarygloballyintermsofboththeirratesofpovertyandtheirlevelsofincomeinequality;thelevelsofsocioeconomicmobilitytheyaffordtheirresidents,especiallynewmigrants;gender-differentiatedpatternsofpopulationdisplacementfollowingclimate-relatednaturaldisasters;andthedifferentialimpactsofexposuretoextremeheatonsegmentsofacity’sworkforce,includinginformalversusformal,femaleversusmale,andolderversusyoungerworkers.Despitethisbroadcoverage,however,thereportissilentonsomeimportantdimensionsofinclusion.Forexample,becauseofthelackofadequatedata,itdoesnotdiscusstheimpactsofclimatechangeoncityresidentswholivewithdisabilitiesorwhobelongtoaracialorethnicminority.BoxO.218THRIVINGKEYFINDING6Citiesinlow-andmiddle-income-countriesarelessgreenintermsofairpollution,andairpollutionfromkeyurbansectorspresentsagreaterchallengeforlargercitiesincountriesatallincomelevelsOnaverage,concentrationsofparticulatematterof2.5micronsorlessindiameter(PM2.5)inboth2000and2015werelowerincitiesinhigh-incomecountriesthanincitiesinlower-incomecountries.Acity’saveragePM2.5concentrationalsotendstofirstincreaseandthendecreasewithitslevelofdevelopment,withairpollutionatitsworstforcitiesinlower-middle-incomecountries(figureO.8).21Meanwhile,evidencefromregressionanalysisindicatesthat,controllingforthelevelofdevelopmentofthecountryinwhichacityislocatedandforotherdeterminantsofpollution,acity’slevelofPM2.5emissionsinitsresidentialandtransportationsectorstendstoincreasewithitspopulation.Inotherwords,inthesectorsthaturbanplanningandpoliciescanmostdirectlyinfluence,largercitieshavehigheremissions.Thisfindingisconsistentwithhigherlevelsoftrafficcongestionemanatingfromstrongerurbanstressesinlargercities.FigureO.8Citiesinlow-andmiddle-incomecountrieshaveworseairpollutionthancitiesinhigh-incomecountriesSource:WorldBankanalysisbasedondatafor10,303citiesfromtheEuropeanCommission’sGlobalHumanSettlement(GHS)UrbanCentreDatabaseR2019(https://ghsl.jrc.ec.europa.eu/ghs_stat_ucdb2015mt_r2019a.php),whichderivesitsdataonPM2.5concentrationsfromtheGlobalBurdenofDisease(GBD)2017databaseonambientairpollution.Note:ThegraphshowstheweightedaveragePM2.5concentrationforcitiesineachincomeclassification,withweightsgivenbycitypopulations.μg/m3=microgramspercubicmeter;PM2.5=particulatematterof2.5micronsorlessindiameter.AverageofPM2.5concentrationsacrosscities,bycountryincomegroup,2000and201526.62000201530.547.434.418.742.433.918.5Low-incomeLower-middle-incomeUpper-middle-incomeHigh-incomeAverageconcentrationofPM2.5(µg/m)19OverviewFigureO.9PM2.5andCO2emissionsarestronglypositivelycorrelatedacrosscitiesgloballySource:WorldBankanalysisbasedondatafor10,303citiesfromtheEuropeanCommission’sGlobalHumanSettlement(GHS)UrbanCentreDatabaseR2019(https://ghsl.jrc.ec.europa.eu/ghs_stat_ucdb2015mt_r2019a.php),whichderivesitsdataonPM2.5concentrationsfromtheGlobalBurdenofDisease(GBD)2017databaseonambientairpollution.Note:Thegraphisascatterplotofthelogofthesumofresidentialandtransportationlong-cycleCO2emissionsin2015onthelogofthesumofresidentialandtransportationPM2.5emissionsinthesameyear.CO2=carbondioxide;PM2.5=particulatematterof2.5micronsorlessindiameter.RelationshipbetweenCO2andPM2.5emissionsacrosscitiesglobally,residentialandtransportationsectors,20151050–5LogoffossilCOemissions,2015(tonnes/year)LogoffossilPM2.5emissions,2015(tonnes/year)5101520Fittedline(R2=0.46)Manyoftheactivitiesthatcontributetopoorairqualityincities,includingbothindustrialactivitiesanddrivinginternalcombustionenginevehicles,alsocontributetoglobalclimatechange.Thereason:emissionsoflocalairpollutantssuchasPM2.5tendtoaccompanyCO2emissions,asillustratedinfigureO.9fortheresidentialandtransportationsectors.Moreover,blackcarbon,ashort-livedclimatepollutant,constitutesamajorpartofPM2.5.However,whereaslocalairpollutionis,atleastpartially,alocalnegativeexternalityofactivitiesinacity,climatechangeisaglobalexternality.Asaresult,acity’slocalpolicymakershaveagreaterincentivetoaddresslocalairpollution—aproblemtheyfindmorefixable—thantoaddressglobalclimatechange,whichrequirescollectiveactionacrosscitiesglobally.Thisisespeciallythecaseforcitiesinlow-incomecountries,whosecurrentcollectivecontributiontoglobalurbanCO2emissionsisnegli-gible(figureO.1,panelb).Asaresult,atthelocallevelcitiesmayfindthemostpoliticallyeffectiveapproachtothemitigationofglobalclimatechangetobepoliciesthataimtoimprovelocalairquality,butthatcarryclimatechangeco-benefits.KEYFINDING7Policiesthatimproveairqualitycanhelpcitiesbothmitigateandadapttoclimatechange20THRIVINGKEYFINDING8LesssprawlingdevelopmentisassociatedwithlowerCO2andPM2.5emissionsfromtheresidentialandtransportationsectors.Oneparticularlypromisingsetoflocalpoliciesforimprovinglocalairqualitywithsignifi-cantclimatechangeco-benefitscomprisesthosethataddressurbansprawlandpromotemorecompacturbandevelopment.Acrossthisreport’sglobalsampleofcities,thecompactnessofacity’sdevelopmenthasastrongnegativecorrelationwithitslevelsofbothPM2.5andCO2emis-sionsfromthetransportationandresidentialsectors.FigureO.10illustratesthesenegativeassociationsforboththeresidentialandthetransportationsectorsforPM2.5emissionsandforthetransportationsectoronlyforCO2emissions.ForbothPM2.5andCO2emissions,morecompactdevelopmentisassociatedwithloweremissions(comparingcitieswithincoun-triesatagivenlevelofdevelopmentandholdingbothacity’spopulationanditsbuilt-upareaconstant).Animportantbenefitofmorecompactdevelopmentisthatittendstobeassociatedwithlessdrivingandmoretransit-orienteddevelopment,bothofwhichcontributetolowertransportationsectoremissionsoflocalairpollutantsandCO2.Inthiscontext,itisalsoimportantnottoconfoundcompacturbandevelopmentwithovercrowdedurbandevelopment.Althoughbothtypesofdevelopmentinvolvehighdensitiesofpopulationpersquarekilometerofbuilt-uparea,compacturbandevelopmentinvolvesaccommodatingthishighdensitythroughmoreverticaldevelopment(thatis,theconstruc-tionoftallerbuildings).Ittherebypreserves,orevenincreases,theamountoflivingspaceperperson.Bycontrast,overcrowdeddevelopmentinvolvestheproliferationofslumsandevertighterlivingspacesthatcancontributeto,amongotherthings,thefasterspreadofCOVID-19andotherinfectiousdiseases,nottomentionlessinclusive,andthereforelessresilient,cities.Localpoliciesthatimprovelocalairqualitycannotonlyhelpmitigateclimatechangebutalsocontributetoacity’sadaptationtoclimatechange.Betterairqualitygeneratessignificanthealthandproductivitybenefits(KahnandLi2020).Thesebenefits,inturn,contributetoincomegrowth,whichhelpsmakecitiesmoreresilienttoclimatechange–relatedshocksandstresses.21OverviewFigureO.10MorecompactdevelopmentisassociatedwithloweremissionsfromtheresidentialandtransportationsectorsSource:WorldBankanalysisbasedondatafor2,785citieswitha2015populationofover200,000fromtheEuropeanCommission’sGlobalHumanSettlement(GHS)UrbanCentreDatabaseR2019(https://ghsl.jrc.ec.europa.eu/ghs_stat_ucdb2015mt_r2019a.php),whichderivesitsemissionsdatafromtheEuropeanCommission’sEmissionsDatabaseforGlobalAtmosphericResearch(EDGARv5.0).Note:Panelashowsforeachsectortheestimatedcoefficients,togetherwiththeassociated95percentconfidenceintervals,fromaregressionofacity’slogPM2.5emissionsin2015onthelogofitspopula-tion,thelogoftheGDPpercapitaofthecountryinwhichthecityislocated,thelogofitsbuilt-uparea,andameasureofthecity’scompactness(thePolsby-PopperRatiocompactnessindex).Theregres-sionalsocontrolsforacity’sclimate(precipitation,temperature,biome)andelevation,anditincludesbothadummyvariablethatisequalto1ifacityisinahigh-incomecountry(andto0otherwise)anditsinteractionwithacountry’sloglevelofGDPpercapita(resultsnotshown).PanelbshowsapartialscatterplotoflogCO2emissionsin2015onameasureofcitycompactness(thePolsby-PopperRatio)controllingforthelogofacity’spopulation,thelogofitsbuilt-uparea,acity’sclimate(precipita-tion,temperature,biome)andelevation,andcountryfixedeffects.CO2=carbondioxide;GDP=grossdomesticproduct;PM2.5=particulatematterof2.5micronsorlessindiameter.LogoffossilCO2emissions,2015(tonnes/year)0.1.2.3.4–.3–.2–.1CitycompactnessRelationshipbetweencitycompactnessandPM2.5andCO2emissionsacrosscitiesglobally,2015–4–202–4–202a.PM2.5:estimatedeectofvariouscitycharacteristicsonemissionsResidentialsectorTransportationsectorb.CO2:transportationsectorCitycompactnessLog,countryGDPpercapitaLog,built-upareaLog,citypopulation42–2–4–6Fittedline(R2=0.34)22THRIVINGKEYFINDING9Citiesthatdevelopverticallyconsumelessland,accommodatemorepeople,andaremoreprosperous.Compactcitiesdevelopvertically,aswellasthroughinfilldevelopment,ratherthanhorizontally.But,theoreticallyatleast,moreupwardgrowththroughincreasedmid-andhigh-risedevelopment,asopposedtolow-rise,doesnotnecessarilyleadtolessoutwardgrowthofacity.Byincreasingthesupplyofavailablefloorspace,moreverticaldevelop-mentmakeshousingandcommercialspacewithinacitymoreaffordable.Affordability,inturn,helpsattractsmorepeopletoacity,therebyboostingitspopulation.Ifithasasuffi-cientlylargeinflowofpopulation,acitymayexpandhorizontallyinresponsetoitsoriginalverticalexpansion.Inpractice,however,empiricalanalysisofdataonbuildingheightsforthisreport’sglobalsampleofcitiesrevealsthat,althoughacity’sverticaldevelopmentdoesindeedleadtopopulationgrowth,suchgrowthisinsufficienttoalsoprovokeoutwardexpan-sion.Acity’sverticaldevelopmentleadsittoconsumelesslandoverallthanitwouldother-wise,which,inturn,couldpreservefertileagriculturallandonacity’speriphery—landthatformanycitiesisa“leadingsourceofnutritionallyimportantfreshfruitandvegetables”(Acharyaetal.2021,xviii).Moreover,thetypeofdensityarisingfromverticaldevelopmentgeneratespowerfulagglomerationeconomieswhileavoidingovercrowding,soitisassoci-atedwithgreatereconomicprosperity.Overall,averagingacrossestimatesthatresultfromtheapplicationofdifferentempiricalstrategies,adoublingofacity’stotalheightleadstoaroughly16percentlong-runincreaseinitspopulationanda19percentlong-runreductioninitslandarearelativetoothercities.Theseresultsareaccompaniedbya4percentlong-runincreaseintheintensityofthecity’snighttimelightspercapita,whichsuggestsincreasedprosperity(figureO.11).AlthoughatanygivenlevelofdevelopmentmorecompactcitiestendtohavelowerCO2andPM2.5emissionsinboththeresidentialandtransportationsectors,moreverticaldevelopmentdoesentailadynamictrade-offwhenitcomestothemitigationofclimatechange.Thecon-structionoftallerbuildingstendstorelyonmaterialssuchasconcrete,steel,andglass,whoseproductionentailshighCO2emissions(Pomponietal.2021).Thus,atallbuildingconstructedusingcurrenttechnologiesembedshighup-frontCO2emissions,whichmustbeweighedagainstthefutureflowoflowerCO2emissionsassociatedwithmorecompacturbandevelop-ment.Thisfactorimpliesthat,totransitiontolowerlong-runemissionstrajectoriesusingastrategyofmorecompacturbandevelopment,policymakersincountrieswheresuchdevelop-mentiscurrentlylimited—becauseof,forexample,dysfunctionallandandpropertymarketsandfailuresinplanning—mayhavetotolerateashort-runincreaseinemissions.Policymakersmay,however,beabletosoftenthisdynamictrade-offbycombiningpoliciesthathelpfacili-tatemoreverticaldevelopmentwithcomplementarytransportationinvestmentsandpoliciesthatbothencourageamovetowardless-pollutingmodesoftransportation,includingwalkingandcycling,andfurtherpromotecompactandlivabledevelopment.Technologicalinnovationsthatreducethecarbonembeddedintheproductionofconcrete,steel,andglasswilllikewisesoftenthetrade-off.2223OverviewFigureO.11Citiesthatdevelopverticallyaccommodatemorepeople,aremoreprosperous,andconsumelesslandSource:WorldBankanalysisbasedonresultsfromAhlfeldtandJedwab2022.TheirdatafortallbuildingscomefromEmporis.Note:Figureshowstheestimatedpercentagechangeineachvariableresultingfromadoublingofthetotalsumoftallbuildingheights.Forpopulationandlandarea,theseestimatesarebasedonaveragingresultsfromacrossthreedifferenteconometric(instrumentalvariable)estimationstrategies(fordetails,seechapter4andAhlfeldtandJedwab2022).Forlights,theestimateisbasedontheapplicationofanordinaryleastsquaresestimationstrategy,inwhich“lights”referstotheintensityofnighttimelightspercapitawithinacity’sextent.NighttimelightintensityismeasuredusingradiancecalibrateddataderivedfromDefenseMeteorologicalSatelliteProgramsatellitesensors.PopulationLightsLandareaEstimatedelasticitywithrespecttototalheightofacity’sbuildings(%)Estimatedelasticitiesofpopulation,nighttimelightintensity,andlandareawithrespecttototalsumoftallbuildingheights–20–15–10–505101524THRIVINGKEYFINDING10Lackofvegetation,especiallyevidentinlargecitiesandcitiesinupper-middle-incomecountries,andpoorurbandesigncanexacerbatetheimpactsofextremeheateventsincities.Althoughdiscussionofacity’sverticaldevelopmentconjuresupimagesofconcreteandsteel,fewpeoplewishtoliveinaconcretejungle.Indeed,definingcharacteristicsofmanyoftheworld’smostwell-knownandsuccessfulcities—thinkofSingaporewithitslushvegeta-tionorLondonwithitsmajesticparks—includenotonlytheheightoftheirskylinesbutalsotheirextensiveurbangreenery.Thegreeneryincities—thatis,thetreesandothervegetationinparksandelsewhere—comeswithimportantbenefitsforacity’sresidents.Notonlydoesgreeneryhaveaninherentamenity,butitspresencecanalsoplayanimportantroleinmitigat-ingtheurbanheatislandeffect.Thiseffectcanleadtourbanlandsurfacetemperaturesthataremorethan10°Chigherthantheequivalentrurallandsurfacetemperatures(Deuskar2022).Thedemandforlandthatcomeswiththegrowthofurbanpopulationcanplacedevelopmentpressureongreenspacesinacity,furtherexacerbatingtheimpactsofextremeheatevents,whicharebecomingbothmorefrequentandmoreintensewithclimatechange.Citiesinupper-middle-incomecountriesarenoticeablylessgreen,onaverage,thancitiesinlow-,lower-middle-,andhigh-incomecountries.Atthesametime,withinanyincomeclass,largercitiestend,onaverage,tobelessgreenthansmallercities(figureO.12).Lackofvegetationisnottheonlyculpritbehindtheurbanheatislandeffect.Thedepthofstreetcanyons—thatis,theratiooftheheightofbuildingsalongastreettothewidthofthestreet—canaffectairtemperaturesthroughitsimpactonshadeandventilation.Theorien-tationofstreetsalsoaffectsbothshadeandventilation.Streetswithaneast-westorienta-tionreceivemoreprolongedexposuretothesunthandothosewithotherorientationsandthusexperiencemoreheat,especiallyincitiesclosetotheequator(Laietal.2019).Heatfrommotorizedvehiclesandthewidespreadandexcessiveuseofair-conditioningareadditionalfactorscontributingtotheurbanheatislandeffect.25OverviewFigureO.12Averagelevelsofvegetationarelowestforcitiesinupper-middle-incomecountriesSource:WorldBankanalysisusingdatafromtheEuropeanCommission’sGlobalHumanSettlement(GHS)UrbanCentreDatabaseR2019(https://ghsl.jrc.ec.europa.eu/ghs_stat_ucdb2015mt_r2019a.php).Note:Acity’saveragegreennessismeasuredbytheaveragegreennessofthepixelsinsatelliteimagerythatfallwithinitsurbanextent.Foreachcitytype,themarkershowstheunweightedaverageofthegreennessindexacrosscitiesofthattype.ThecolorsofthemarkersfordifferenttypesofcitiescorrespondtothecolorsinmapBO.1.1.Averagelevelsofgreenness,bycitytype,2014LargecitiesMediumcitiesSmallcitiesAveragegreennessindex0.450.400.350.300.25High-incomeUpper-middle-incomeLow-andlower-middle-income26TheimpactsofclimatechangeoncitiesClimatechange–relatedshocksandstressescanaffectthegreenness,resilience,andinclusivenessofcitiesthroughawidearrayofdirectandindirectchannels.Moreover,theseshocksoftendonotoccurinisolationandcanbecompoundedbyunderlyingurbanchallengesthatarisefromthepressureofgrowingurbanpopulationsonurbaninfrastructure,thesuppliesofbasicservices,landandhousing,andtheenvironment.Climatehazardscanalsocascadeintocitiesfromsurroundingruralareas,aswellasfromareasonwhichacitymightdependforitswatersupply.Inequalitieswithincities,which,especiallyformanycitiesinlow-andlower-middleincomecountries,arealreadylarge,maybefurtherexacerbatedbyclimatechange–relatedshocksandstresses.Andwhilecitieshavetraditionallybeenthoughtofasprovidingescalatorsoutofpoverty(Glaeser2012),climatechangemayslowthespeedoftheseescalators.ClimatehazardscanbecompoundedbyunderlyingurbanchallengesTheclimatechange–relatedshocksandstressorsthataffectgreen,resilient,andinclusivedevelopmentincitiesdonotoccurinisolationbutofteninteractandcompound,bothwitheachotherandwithotherurbanstressors.Tropicalcyclonesandextremeheateventsarerelatedandoftenoccursimultaneously.Poorlymanagedurbandevelopmentpressuresthatleadtotheremovalofurbantreesanddestructionofurbanwetlandscouldcompoundtheeffectsofheatwavesandfloods.Lossesinagriculturalproductionfromheatanddrought,compoundedbothbytheexcessivelossoffertileagriculturallandontheperipheriesofcitiesduetosprawlassociatedwithpoorlymanagedurbanizationandbyheat-inducedreductionsintheproductivityofworkers,couldaffectthefoodsupply.Riskscanspilloveracrosspopulations,places,andsectors,leadingtocascadingimpacts.Ruralmigrantsfleeingdroughteventscansettleinprecariousinformalsettle-mentsinurbanfloodplains,withcascadingrisksforsomegroupsofpeopleandlocations.Wildfiresinagriculturalregionscanincreaseurbanairpollutionwhilealsodisruptingthesupply,andthusprices,ofessentialfoodproducts.Thegeneralinterdependencewithincitiesofcriticalinfrastruc-ture,suchastransportationsystemsandpowergrids,meansthatfailureofoneelementornodecouldresultinacascadeofadverseevents.Thus,stormsurgesandextremeheatcouldleadtopoweroutages.Otherunderlyingstresseswithincities—notnecessarilyrelatedtoclimatechange—canalsoexacerbateitseffects.Forexample,highratesofinformaldumpingofwasteworsenpluvialfloodsbecauseoftheaccumulationofrefuseindrains,waterways,andopenspaces.AccordingtotheIntergovernmentalPanelonClimateChange,multipleclimatehazardswillcontinuetooccursimultaneously,therebycompoundingoverallriskandcausingriskstocascadeacrosssectorsandregions(IPCC2022).Nevertheless,ashighlightedinkeyfinding4,constructionpatternsshowlittlesignofrespondingtothegrowingthreats.Thissituationcouldresult,inpart,frompublicpolicy27Overviewthatencouragessettlementinmorehazard-proneareas.Forexample,intheUnitedStates,subsidizedfloodinsuranceandreliefaidmayhaveincreasedpeople’swillingnesstoliveindisaster-proneareas(Deryugina2014;Gregory2017).Likewise,freefederalfireprotectionmayhaveincreasedconstructioninareaswithhighfirerisk(BaylisandBoomhower2019).ClimatehazardscanalsocascadeintocitiesClimatechange–relatedshocksinruralareasalsoindirectlyaffectcities.Whenextremeweathereventshit,peopleinthecountrysideoftenseeksafeharborincities.Inbackgroundresearchforthisreport,Chlouba,Mukim,andZaveri(2022)showthatperiodsofextendeddroughtintheruralhinterlandsofcitiesresultinfastergrowthoftheurbanbuilt-uparea,presumablybecauseofpushmigration.Therelationshipbetweendroughtandgrowthofurbanbuilt-upareasisparticularlypronouncedinlow-andmiddle-income-countries,whichsuggeststhatclimatechangemaybeoneofthefactorsbehindtherapidurbanizationofmanyofthesecountries(figureO.13).Suchclimate-inducedmigrationaffectsnotjustthepaceofurbanbuilt-upexpansionbutalsothenatureofthatexpansion.Drought-drivenurbanexpansionoftentakestheformofexpandinginformalsettlements,whereservicedeliveryremainsachallenge.Whenclimatemigrantsarriveinurbanareas,theyoftenclusterinperipheralinformalsettlements.Thesesettlementsofferlimitedjobopportunities,lackbasicinfrastructure,andhaveservicedeliverysystemsthatremainintheirinfancy.Becausethesettlementsareinformal,settlersfacetheriskofevictionandforcedreloca-tion,creatingsecondarydisplacementbecausetheynolongerhavetheoptiontoreturntoclimate-decimatedruralareas.Thus,climate-induceddisplacementcanunderminetheinclusivenessofurbandevelopment,leavingsomeofthemostvulnerablemembersofsocietyontheoutskirts,bothfigurativelyandliterally.FigureO.13Droughtisassociatedwithfasterbuilt-upareagrowthinlow-andmiddle-incomecountriesSource:Chlouba,Mukim,andZaveri2022.Note:Thecirclesrepresentthepointestimatesofcoefficientsofinteractionsbetweencountryincomegroupandaspatiallyandtemporallylaggedgridcellmeasureofdrought.Thesolidlinesrunningthroughthecirclesindicatethelowerandupperboundsofthe95percentconfidenceintervalsoftheestimatedcoefficients.Heterogeneousimpactsofdroughtsonbuilt-upareaacrossaglobalsampleofcities,bycountryincomegroup,1985–2014–0.4–0.3–0.2–0.100.10.20.3Low-incomeMiddle-incomeHigh-incomeEstimatedcoecientoninteractionbetweenacity’sincomecategoryanddrought28THRIVINGNevertheless,suchurbanexpansioncanalsocomewithopportunities.Rapidurbanpopulationgrowthinducedbyclimatechangecanstraincitiesbyputtingpressureonland,housing,infra-structure,andservicedelivery;butitcanalsofuelurbaneconomiesbyprovidingasteadyflowofnewlaborerstogrowingsectorssuchasmanufacturing,construction,hotelsandrestau-rants,andtransportation.Urbanenvironmentsalsoholdthepotentialtobreakdowntradi-tionalgenderbarriersbybringingwomenintoprofessionspreviouslyreservedfortheirmalecounterparts(WorldBank2021b).Althoughmostpeopledisplacedormigratingbecauseofclimateimpactsstaywithintheircountriesoforigin,theacceleratingtrendofglobaldisplace-mentcanincreasecross-bordermovementsaswell,particularlywhereclimatechangeinter-actswithconflictandviolence.23Waterstressesalsoaffectcities—evenfromafar.Citieshavealwaysreliedonwaterimportedfromother,sometimesdistant,areas.FromthetimeofancientRometoLosAngelesintheearlytwentiethcenturytocontemporaryinitiativesfromMexicoCitytoKathmandu,Nepal,waterimportshaveofferedapathtourbanwatersecurity.Butthescaleandintensityofthesewatertransfershaveundergonerapidchanges.Nowadays,citiesoftenrelyondozensofwatersourceshundredsofkilometersaway.Thus,afarawaydroughtmayprofoundlyaffectacity.Inadditiontocreatingwatersupplyproblemsforcities,climate-relatedshocksandstressorsaffectruralfood-producingareas.Intheirbackgroundpaperpreparedforthisreport,Venkat,Dizon,andMasters(2022)demonstratethatimpactsspillovertocitiesviahigherurbanfoodprices.Theseimpactsvarysignificantlybythetypesofshocksandstressorsandbythetypeoffood(suchasmorenutrient-denseperishablesversuscalorie-densenonperishables).Bettertransportationnetworkshelpmitigatetheimpactofshocksandstressors,24allowingpoten-tiallymoreresilientfoodsupplychains.ClimatechangecouldalsoexacerbateinequalitiesAroundtheworld,urbanslumsareoftenfoundinprecariouslocations—onsteepslopes,onfloodableland,ornearopendrainsandsewers.ResearchconductedforthisreportbyRossitti(2022)showsthat,foraselectedsampleof18citiesinSouthAsiaandSub-SaharanAfrica,slumsdonotnecessarilyhavemoreexposuretofloodsorexcessiveheat.Whenfocusingonlyontheprobabilityofahigh-riskevent,however,slumsdoappeartohaverelativelyhigherexposurethanformalresidentialareastosuchpotentiallymoredestructiveevents(Rossitti2022).Locationsorting—thetendencyofpeoplewithsimilarcharacteristicstoclustertogetherincertainneighborhoodswithincities—canexplaintheoverexposureofthepoorinsomecities.Thechoicetoliveinhazard-proneareasoftenreflectsadifficulttrade-off—betteraccesstojobsandservicesagainstenvironmentalrisks.Insuchcontexts,alackofaffordablehousingoftenpricespeopleoutofsafer,well-connectedlocations.Itisalsopossiblethatpoorerhouse-holdsarelessabletoacquireinformationontheenvironmentalrisksofspecificlocations,orthatthesehouseholdsconsistofmigrantsfromruralareaswho,asnewcomerstoacity,donotpossessneededinformation.Indeed,prospectivedwellersoftendonothaveaccurateinforma-tionontherisksassociatedwithdifferentlocations(BakkensenandBarrage2022;VotsisandPerrels2016).Becausepovertyinvolvesaseriesofdistractionsthatreduceproductivityaswellasthechancestogatherinformation(BanerjeeandMullainathan2008),poorer,lesseducatedindividualsmayfacehighercostsinacquiringhazard-pronenessinformationthantheirrichercounterparts.25Thepoormayormaynothavegreaterexposuretonaturalhazards,buttheyareoftenthehardest-hitwhendisasterdoesstrike.Inabsoluteterms,therichoftenlosemorebecausetheirassetsareworthmore;inrelativeterms,however,theoppositeholdstrue.Poorerhouseholds29Overviewalsosufferdisproportionallyfromtheindirecteffectsofnaturalhazardsthroughinfrastruc-turedisruption,suchasbeingcutofffromroadsandpublictransportation,whichlimitstheiraccesstojobsandeconomicopportunities,andtheirwatersupply.Notonlydopoorurbandwellerssufferhigherrelativelossesfromclimatechange–relatedshocks,buttheyalsohavelessabilitytoengageinadaptationandmitigation.Ingeneral,poorerhouseholdshavelessaccesstofinancialmarkets(Ermanetal.2019)andinsurancemarkets.Moreover,theyarelesslikelytobenefitfrompublicinvestmentsininfrastructurethatmitigatestheriskofnaturalhazards.Suchinvestmentscantranslateintohigherpropertyprices,therebypricingthepooroutofnow-saferareasofacity(Nakagawa,Saito,andYamaga2007).Finally,manypoorhouse-holdsengageinrecurrentself-financedshort-termmeasures(suchastemporarystructuralimprovements)thatplacealargefinancialburdenonthem(Patankar2015).Climatechange–relatedstresses,suchasexcessivelyhightemperatures,arealsoexpectedtohavenegativedistributionalimpacts.Foronething,theeffectswilllikelybeconcentratedinlow-andmiddle-incomecountries.Citiesinlower-incomecountriesoftendependmoreonsectorsfocusedonoutdoorwork.Moreimportant,theyhavelessabilitytoimplementadapta-tionmeasures.Adaptationmeasuressuchasair-conditioningintheworkplacecoulddecouplehightemperaturesfromworkerproductivity,butsuchmeasurescanberelativelyexpensive.Moreover,climatechange–relatedstresseswillexacerbateinequalitywithincountrieswheninformalworkers,whoalreadyhavelowerincomelevels,seesharperreductionsinthoseincomes.Intheirbackgroundpaperforthisreport,JiangandQuintero(2022)evaluatetheeffectsofbothhighaverageannualtemperaturesandnumberofextremelyhotdaysontheproductivityofworkersinthousandsofcitiesacrosseightcountriesinLatinAmericaandtheCaribbean.26Theyfindthatbothsignificantlyreduceworkerproductivity,asmeasuredbywages.Thus,forexample,adoublingofaverageannualtemperatureisassociatedwithanestimated14.1percentdropinwages.Citieshavetraditionallyservedasescalatorsoutofpoverty;however,morefrequentclimatechange–relatedshocksmayslowdowntheseescalators.Urbanresidentsremainvulnerableorchronicallypoorifdeprivedofaccesstoeconomicopportunities,basicservices,andamenities.Theimpactsofclimatechange–relatedandenvironmentalshocksexacerbatesuchfailures.Forexample,asnotedearlier,thepoorandvulnerablearemorelikelytosufferdisruptionfromfloodingandotherclimatechange–relatedshocks.Poorerhouseholdsalsohavelimitedfinan-cialbufferstocopewithshocks.Intheirbackgroundresearchforthisreport,Abanokovaetal.(2022)estimatetheprobabilitiesofexitfrompovertyforurbanresidentsinfivecountries—Chile,Colombia,Ecuador,Indonesia,andPeru.Consistentwiththehypothesisthatcitiesactasescalatorsoutofpoverty,theyfindthatmanypeoplehaveindeedescapedfrompovertyinurbanareas.Andthatescapeismorelikelyinlargercities.Asfloodrisksriseinlargecities,however,theprobabilityofexitfrompovertyfalls.Forcitieswithlargepopulations,house-holdsinareasathighriskoffloodinghaveasubstantiallylowerprobabilityofescapingfrompovertythandohouseholdsinlow-riskareas.30FigureO.14Aframeworkformakingcitiesgreener,moreresilient,andmoreinclusiveSource:WorldBank.PolicyinstrumentsInformationDevelopmentpathwayGreenResilientInclusiveIncentivesInsuranceIntegrationInvestmentsStressesinducedbyurbangrowthClimatechangeAgglomerationeconomiesPoliciesformakingcitiesgreener,moreresilient,andmoreinclusiveinaworldconfrontedbyclimatechangeAstheprecedingsectionshavemadeclear,stressesassociatedwithurbangrowthinteractwithclimatechange–relatedstressesthroughavarietyofchannels.Indoingso,theyactagainstproductivity-enhancingagglomerationeconomies,adverselyaffectingthegreenness,resilience,andinclusivenessofacity’sdevelopmentpathway(figureO.14).27Itfollowsthat,toimprovedevelopmentoutcomes,policymakersatthenationalandlocallevelsneedtoworktogethertotargettheseinter-relatedstressesinawaythatbestaddressesacity’sspecificgreenness,resilience,andinclusivenesschallenges,wherethesechallengesvarywithbothacity’ssizeanditslevelofdevelopment.Toassistthoseefforts,thissectionlaysoutthegeneralconclusionsinthreequestionsthatpolicymakersshouldconsider:Whatpolicyinstrumentsareavailable?Whowieldstheseinstruments?Howcanpolicychoicesbeprioritized,sequenced,andfinancedforeffectiveimplementation?31OverviewWhatarethechoices?Policymakerscandrawonfivebroadsetsofpolicyinstruments:information,incentives,insurance,integration,andinvestments—thefiveI’s.BoxO.3summarizestheseinstruments,andthesectionsthatfollowprovidemoredetailsoneach.ThefiveI’s:Information,incentives,insurance,integration,andinvestmentsThisreportdistinguishesbetweenthefollowingfivebroadsetsofinstrumentsthatpolicymakerscandrawoninseekingtoimprovethegreenness,resilience,andinclusivenessofacity’sdevelopmentinaworldconfrontedbyclimatechange.InformationPoliciesandmeasurestoimprovethetimelyprovisionofcredibleinformationthathelpspeople,businesses,andlocalgovernmentsbetterunderstandclimatechange–relatedrisksbothacrossandwithincities,and,indoingso,helpspromotebothmitigationandadaptation.IncentivesPolicyinstrumentsthatprovideincentivesforindividualsandbusinessestointernalizenegativeenvironmentalexternalities,aswellasinstitutionalandothertypesofreformthatprovidegovernmentofficialswithincentivestoworkbettertogethertoaddressgreen,resilient,andinclusivedevelopmentchallenges.Incentivesalsoincludetheremovaloffossilfuelsubsidiesandotherincentivesthatencourageactivitieswithnegativeenvironmentalexternalities.InsurancePoliciesandreformsthathelppeople,businesses,andgovernmentseithertoinsurethroughthemarketortoself-insureagainstlossesassociatedwithclimatechangeandotherenvironmentalshocksandstresses.IntegrationPolicyinterventionsandreformsthatpromotemorecompactcitiesandbetterintegrationofcitiesbothwitheachotherandwithruralareasasameansoffacilitatingadaptationthroughmigrationandtrade.InvestmentsInvestmentsbynationalandlocalgovernmentsingreen,resilient,andinclusiveurbaninfrastructure,includingnature-basedsolutions,aswellasmeasurestopromotethecrowding-inofprivatesectorfinanceforsuchinvestments.BoxO.332THRIVINGInformationAnecessarypreconditionforefficientdecision-makingbyhouseholdsandbusinessesthatmaximizestheirexpectedwell-beingandprofits,respectively,iscompleteandaccurateinfor-mationaboutallthepotentialriskstheyface.Inthiscontext,informationcomprisesvariouspolicyinstrumentsrelatingtobothimminentthreats,suchasanimpendingextremeweatherevent,andthelonger-termevolutionofclimatechange–relatedhazards.Effortssurround-inginformationincludedisasterriskandriskmanagementstrategiesorparticipatoryinitiativesthatallowlocal,regional,andnationalauthoritiestoconveyinformationonclimatechange–relatedriskstourbanresidents,aswellasviceversa.Despiteremarkableprogressonthescientificfront,someofthemostvulnerablecommuni-tiesinclimatechange–threatenedcitiesremainpoorlyinformedaboutbothloomingextremeeventsthat,ifnotadequatelyrespondedto,couldspelldisasterandslow-movingchangesthataffecteverydaylife.Availabilityofinformation,however,doesnotautomaticallytranslateintounderstanding,action,orevenacceptance—afterall,manypeopledeny,dismiss,orcastunwarranteddoubtonthescientificconsensusaroundclimatechange.28Climateadaptationinformation,suchastheresultsofclimatemodelingandfutureweatherforecasts,constitutesapublicgood,providingastrongrationaleforgovernmentprovisionorsubsidization.Thediffusionofinformationaboutriskscanresultinmoreefficientdecisionsbybothhouseholdsandbusinesseson,forexample,wheretoliveandinvest,potentiallycounteringthetrendofconstructionin“futurebadlocations”andensuringthathouseholdscanaccuratelyevaluatethetrade-offbetweenalocation’senvironmentalhazardsandtheaccesstojobopportunitiesandservicesitprovides.Suchdiffusioncanalsoprovideacity’sresidentswiththeinforma-tiontheyneedtoholdtheirlocalelectedofficialsaccountableforaddressingenvironmentalhazards.Informationcantaketheformofearlywarninginformationandmonitoringsystems.Earlywarningscanprovidehugebenefitsasapreventivemeasure,savinglivesand,ifissuedsuffi-cientlyinadvance,property.Theearlierthewarning,themoretimeacity’sresidents,busi-nesses,andauthoritieshavetoprepare,includingbyprotectingpropertyandinfrastructureandbypositioningandreinforcingassetsforprotectionandresponse.Often,couplingthemwithotherpreventiveinvestmentscanmaximizethebenefitsofearlywarningsystems.29Evenmodestinvestmentsinsuchsystemscanhavehighreturns.30Finally,informationentailsregularlyupdatedurbanplanningdocuments,buildingcodes,andzoningregulationsthathelpbothguideandcoordinatethedecisionsofhouseholdsandbusi-nesses.Itisalsousefultothinkofinformationinthecontextofkeyurbanmarkets—land,realestate,labor,andcapital.Thepriceoflandistheprimaryelementdrivingefficientlanduseincitiesanddecisionsbyhouseholds,developers,andbusinessesonhowmuchtoinvestandonwhattypeofstructuresinaparticularlocation.Astandardeconomicprescriptionistoprovidetransparentinformationthathelpsensurethatmarketpricesaccuratelyreflectbothriskandsupplyanddemandconditions(inthecaseofdynamicallyefficientvolumetricwaterpricing,forexample).Suchtransparencyhelpsensurethatpricessendthecorrectsignalstoeconomicagentsmakinglocation,consumption,andinvestmentdecisions.Betterinformationthatfacilitatesthemoretransparentandefficientworkingoflandmarketscouldalsocontributetoclimatechangemitigationeffortswithco-benefitsintermsof,forexample,thepreservationoffertileagriculturallandbyloweringthecostsofverticalconstruction.AsresearchforthisreportbyAhlfeldtandJedwab(2022)shows,verticalconstructionresultsinamorecompacturbanform(seealsokeyfinding9).33OverviewIncentivesAlthoughtheprovisionofinformationhelpshouseholdsandbusinessesfactorclimatechange–relatedrisksintotheirowndecisions,informationinandofitselfmaynotbesuffi-cienttomotivatethemtotakeintoaccounttheimpactsoftheirowndecisionsontheenviron-mentandonothers.Citiesthereforeneedincentivestomotivatehouseholdsandbusinessestointernalizeexternalitiessuchaspollution.Incentivescomeinvariousguises,includingpricingpolicies,suchastaxesondirtyfuels,31andcongestioncharging,environmentaltaxes,charges,andsubsidies.Thesepoliciescanalsocarrysignificantclimatechangeco-benefits—forexample,congestionpricingreducesvehiclemilestraveledandmotorvehicleemissions,withasignificantpositiveimpactonhealthintheshorttermandevenlargerlonger-termeffectsasacommunityevolvestoanewlowerpollutionequilibriumlevel.Bycontrast,carbonpricingpolicies(suchasemissionstradingsystemsandcarbontaxes)tendtooccurmoreatthenationallevelandaimataddressingglobalexternalities.Incentivescanalsoinvolveremovingorreducingsubsidiesthat,althoughperhapswellintendedwhenoriginallyintroduced,neverthelesshavetheunintendedconsequenceofinducingoverconsumptionofgoodsandservicesoroverinvestmentinactivitiesthatentailnegativeenvironmentalexternalities.Forexample,theArabRepublicofEgyptinitiatedafuelsubsidyremovalprogramaimedattargetingwastefulconsumptionandbalancingthepublicfiscalburden.ThisprogramresultedintargetedfuelpriceincreasesinNovember2016andJune2017that,dependingonthefuelcategoryandperiod,variedbetween30and80percent.Hegeretal.(2019)estimatethat,byreducingtrafficbelowthelevelsthatwouldhaveother-wiseprevailed,theseincreasesledtoa4percentreductioninGreaterCairo’sconcentrationsofparticulatematterof10micronsorlessindiameter.Thereformalsogavethecountryanopportunitytoestablishanationallydefinedsocialprotectionfloor.Quotacontrolpoliciescanbeeffectiveinreducingpollutingactivities,buttheyneedtobecare-fullydesigned.AnanalysisofSingapore’svehiclequotacontrolbySong,Feng,andDiao(2020)findsthatvehiclequotassubstantiallylimitvehicleownershipandusage.Consumers’desiretomaximizetheirreturnoninvestment,however,yieldshigherusagebyexistingcarowners,par-tiallyoffsettingthemitigationeffectsofthequotas.Globally,countriesaremakingpledgesandcommittingtophaseoutthesale,production,anduseofvehiclesdependentonfossilfuels.Forexample,Rwandaisontracktograduallyphaseouttheuseofpollutingvehiclesby2040,startingwithapilotprograminKigalitoconvertmotorcyclestoelectric.AheadoftheRepublicofKorea’s2030target,Seoul’splantophaseoutdieselvehiclesfromthepublicsectorandmasstransitfleetsby2025alsoseekstopushaheadwithfleetchangeoverinthetaxiindustrywithstrongsubsidies.AsoriginallypointedoutbyNobeleconomistKennethArrow(1963),addressingamarketfailuresuchasinadequateinformationornegativeexternalitiescanoftenleadtomoreefficientandequitablemarketoutcomes.Thisresultoccursbecausesuchmarketfailurestypicallyaffectpoorerhouseholdsmoreheavily.Forexample,thosewhowalktoworkbecausetheycannotaffordeithercarsormotorbikesalsosuffermostfromthepollutionandadditionalheatgener-atedbytheinefficientlyhighlevelsofprivatemotorvehicleuseintheabsenceofcongestionpricingor,evenworse,inthepresenceoffuelsubsidies.Moregenerally,thedesignofpolicypackagesthatinvolveincentivesmustincludecarefulattentiontothedistributionofburdenandfavoracrosseconomicorsocialgroups.And,whenpossible,complementarypolicymeasuresshouldbeintroducedtocompensatethosewhostandtolose.Inthecaseofincreasedfueldutiesorcongestioncharging,forexample,suchmeasurescouldinvolvethehypothecationoftherevenueraisedforinvestmentsinpublictransportation,whichLondondidwhenitintroducedacongestionchargein200334THRIVING(Santos,Button,andNoll2008).Meanwhile,failuretotakeintoaccountthedistributionalimpactsofnewincentivesand,moregenerally,anypolicyintendedtohelpaddressclimatechangeandotherenvironmentalissuesmayresultininsurmountablepoliticaloppositiontothepolicy.Onerecenthigh-profileexampleofthisoppositionistheFrenchgovernment’sfreezingofitscarbonpricingpolicyfollowingthe“yellowvest”protestmovementof2018(RubinandSengupta2018).InsuranceInsuranceaimstominimizethefinancialimpactofdisastersthroughrisksharingandtohelpsecureaccesstopostdisasterfinancingquicklyandefficiently,therebyensuringrapid,cost-effectiveresourcestofinancerecoveryandreconstructionefforts.Governmentscanimplementpolicyandregulatoryreformstodeepeninsurancemarketsandimproveaccessforbusinessesandhouseholdstocomplementtheiradaptationandresiliencestrategies.Marketinsurancethatreflectsdisasterrisksisoftentouted,andwithgoodreason,byeconomistsasthefirstbestoptiontointernalizerisksandminimizedisasterimpacts(WorldBank2017).Atthesametime,marketinsurancecanbedifficulttoimplement,especiallyinlower-incomecountriesthathavelackingorasymmetricinformation,weakregulation,andlimitedriskcapital.Developinginsurancemarketsinthiscontextmayrequiregovernmentinterventionsbeyondsimplepolicyprescriptionsandsubsidymechanisms.Informationabouthazards,vulnerability,andexposureisessentialnotonlyforthesoundpricingofinsurancebutalsoforother(physical)mitigationactions.Governmentactionstogenerateandsharesuchinformationhavebroadco-benefits,andthoseco-benefitsshouldbeatthecoreofadaptationstrategies,includingeffortstodevelopmarketinsurance.Catastropheriskinsurancecanhavehighcapitalandadministrativecosts.Whenitsellsspecificcoverage,aninsurancecompanypromisestocompensateitsclientincaseofadisaster.Itmustsetasidereservesandbuyreinsurancetocoversuchaneventualityandhavesystemsinplacetoassesslossesandpayclaimspromptlyandinfull.Holdingfinancialreserves,buyingreinsurance,andprocessingclaimscanaddsubstantialcosttothatrelatedtotheexpectedloss,particularlywhenthecompanymustdealwithmanysmallclaimsatonce.Inaddition,uncertaintyrelatedtotheunderlyingrisk,whichisgrowingforclimate-relateddisastersbecauseofclimatechange,willbereflectedintheinsurancepremiumthroughuncertaintyloading.Whataresomewaystoreducethehighcosts?Highdeductiblestomanagemoralhazardandavoid“pennyclaims”andmechanismstocoverfirstloss(forexample,throughgovernmentguarantees)canhelpreducethecostofinsurance.Makinginsurancecompulsoryensureswiderdistributionoftheriskandbetterriskpooling.Parametricorindex-basedinstrumentscanhelpaccelerateandreducethecostofclaimprocessingiftheunderlyingindexusedtocalculatethepayouthasahighcorrelationwiththeactualloss.Suchfinancialengineeringcansubstantiallyreducethecostofinsurance,particularlyforthepoorest,whoaregenerallythefirstaffectedbyadisaster.Finally,thedevelopmentofcatastropheinsurancemarketsinlower-incomecountrieshasoftensufferedfromalackofclarityonwhenandwherethegovernmentintervenes.Businessesandindividualstendtoassumethatthegovernmentactsastheinsureroflastresortandthusdonotpurchaseinsurance.Makinginsurancecompulsory,evenifdoingsorequiressomesubsidiesorguaranteestoreducethecosttoaminimum,issometimesmoreefficientifithelpsclarifyresponsibilitiesandencouragesprivateactorstocoveratleastpartoftheirrisk.35OverviewIntegrationWithincities.Between2022and2050,thenumberofpeoplelivingincitiesgloballyisprojectedtogrowfrom4.53billionto6.68billion.32Withthisgrowthwillcomeheightenedurbandemandforhousing,basicservices,infrastructure,andamenities.Thisdemandcouldincreasepressuresonlandandrealestatemarkets,resultingindevelopmentpatternsthatfurtherunderminethegreenness,resilience,andinclusivenessofcities.Reformsthatstrengthenformalinstitutionsfortitlingandpropertytransfer,alongwithflexibleandeffectiveurbanplanningthatisproperlycoordinatedwithinvestmentsininfrastructure,canensurethatcitiesarenotlockedintosuboptimalphysicalformsandinvestments.Onenotableexampleofsuchalock-inisoverlysprawlingcar-andmotorcycle-dependenturbandevelopment.Asshownunderkeyfinding8,suchdevelopmentisassociatedwithhigherlevelsofproduction-basedCO2emissionsinboththetransportationandresidentialsectors,andwithhigherPM2.5emissions.33AccordingtoresearchpreparedforthisreportbyAhlfeldtandJedwab(2022),reformsthatreducethecostsofverticaldevelopmentcanencouragecitiestoexpandupwardratherthanoutward,whilegeneratingeconomicgrowth(seealsokeyfinding9).AhlfeldtandJedwab(2022)calculatethatthewelfarepotentialoftallbuildingsishighestforthelargestcities.Inlow-andmiddle-incomecountries,thesecitiesarepreciselytheoneshithardestbytheconstraintstoverticaldevelopmentposedbydysfunctionallandmarketsandfailures.Beyondsprawlanddependencyoncars,furtherundesirablelock-insincludeenergy-orwater-intensivebuildingtechnologiesandurbansettlementslocatedinvulnerableareas(seealsokeyfinding4).Althoughretrofittinginfrastructuresandbuildingswillbeessentialtogreenerandmoreresilientgrowth,doingsocanbecostly.Earlyeffortsatintegratedplanning,however,couldhelpcities—especiallysmallerbutrapidlygrowingonesinlow-andmiddle-incomecountries—avoidsuchretrofitting.Integrationisgoodforthepoor—andgoodforbudgets.Thepoormayknowthehazardstheyface,buttheydependmorethanthewealthyonpublicservicesthatareofteninade-quate.Inplaceswithoutcoordinatedlanduseandurbaninfrastructureanddevelopmentdecisions,householdsendupdisconnectedfromlabormarketsandhavingtotradesafetyforaccessibility.Localgovernmentsoftenstruggletoprovideessentialurbaninfrastructure;untiltheysucceedindoingso,thepoorwillremainvulnerable.Moresecurelandandpropertyrightswouldencourageinvestmentinpreventionmeasures.Equally,theprovisionoflandandaffordablehousinginsaferareas,withaccessibilitytojobsandessentialservices,wouldgoalongwaytowardloweringtheriskexposureofpoorpeople.Cityleaderscanunderinvestintimespentonlong-termplanningwhendealingwithaseriesofemergencies,becausepressingmatterscrowdoutimportantmatters.Throughinvestmentsundertakenbeforesettlement,citiescanmoreeasilyandcost-effectivelydeliverinterventionsaimedatimprov-ingandincreasingdensity.34Suchprojectsareoftenmoreeffectivewhenjointlydevelopedandimplementedalongsidethecommunitiesthatstandtobenefit,andoftenmustincludeawareness-raisingcomponentstoappropriatelymanagebehaviors.35Betweencities.Climatechangeisforcingindividuals,families,andevenwholecommunitiestoseekmoreviableandlessvulnerableplacestolive.Governmentshaveresortedtospecificlegislation,regulations,andpoliciestodiscourageorrestrictdomesticmigration.Thesepoliciesincluderesidentialregistrationsystems,suchasChina’sHukouandVietnam’sHoKhausystemsthatexplicitlyrestrictinternalmigration(Boskeretal.2012;WorldBank2020).Migrationmayalsobestymiedbyfailuresinplanningandpolicy,suchasafailuretoprovidesecurelandtenureandoverlyrestrictiveregulationsonbuildingheights,thatcontributetoaninsufficiencyofaffordablehousinginthefaceofagrowingurbanpopulation.36THRIVINGInthespecificcaseofclimatechange–inducedmigration,theoveralleconomiceffectonthereceivingcityisambiguousandwilldependonlocalconditionsandthecapacityofthecitytoabsorbalargerlaborforceoflower-skilledworkers.Evenastheprecisepolicymixwillvaryacrosscountries,thefundamentalingredientforeasingsuchmigrationtransi-tionswouldlikelyremainunchanged.Forexample,decision-makerscouldfocusonintegrat-ingmigrantsintothecitytobothlimitimpactsonhostcommunitiesandensureinclusiveopportunitiesfornewmigrants(Zaverietal.2021).Thisemphasiswouldinvolvebuildinghumancapitalandinvestinginworkerproductivitythrougheducationandlabormarketpoliciesthatbuildskillsandprovidetraining.Governmentscouldalsohelpreducethecostsofmigrationby,forexample,fosteringincreasedaccesstofinancialmarketsandtherebyrelaxingthecreditconstraint,loweringthebarrierstoassimilationinreceivingareas,orprovidingbetterinformationonjobsandotheropportunitiesindestinationswithlowerclimatechange–relatedrisks.Finally,theextenttowhichmigrationbetweencitiescanactasaviableadaptationstrategytoclimatechangefurtherdependsontwofactors:thenumberofcitiesandthevariationinthestrengthofclimatechange–relatedhazardsacrossthosecities.ForalargecountrysuchasBrazil,China,ortheUnitedStatesthatboastsmanycitiesacrosswhichthestrengthofprojectedclimatechange–relatedhazardsvariesgreatly,migrationfromcitiesthataremoreexposedtohighlevelsofhazardtothosethatarelessexposedrepresentsanimportantpotentialadaptationmechanism(figureO.15).However,forsmallPacificandCaribbeanIslandnations,forexample,whichhaveonlyoneortwomajorcities,intercitymigrationcarriesmuchlesspotentialasanadaptationmechanism.Thesameholdstrueforlargelow-andmiddle-incomecountriessuchasIndonesiaandVietnamwhosecitiesfaceuniformlystrongprojectedclimatechange–relatedhazards.InvestmentsInvestmentsmaycomelastinthislist,buttheyarenotleast.Investmentsininfrastructure,whenwelldesigned,constructed,andmaintained,canhelpcitiespreventandrespondtourbanandclimatechange–relatedshocksandstressors,therebyreducingtheprobabilityofdisastersandtheassociatedlossoflifeandproperty.Infrastructureinvestmentscanincludeimportantpreventionmeasures,suchasfloodcontrolsystems,constructionofshelters,andprotectionofenvironmentalbuffers.Someinfrastructurecanservemultiplepurposes.Forexample,safeschoolsinBangladeshalsoserveascommunitycycloneshelters.Postdisasterinvestmentscaninvolverehabilitationandreconstruction,includingtherepairandrebuild-ingofpublicandprivatepropertysuchashousingandinfrastructureassets.Reconstructionmayoftenincludedisaster-resistantmeasuresforfutureprevention.Becausethecontinuedeffectivenessofinfrastructurewillalsodependonitsquality,investmentoutlaysmustincludemaintenance,therebyboostingtheresilienceofinfrastructureassetswhilereducingoverallcostsinthelongrun.Investmentsininfrastructurethataffectlanduseandacity’surbanformcanhaveimplicationsfarintothefuture.Forexample,investmentsinroadsthatpromotemotorvehicleoverpublictransportationuse,therebyencouragingsprawl,couldsignificantlyandpermanentlyincreasethecostsofdeliveringbasicservices,suchaswater,sanitation,andelectricity,andbuildingsocialinfrastructure,suchasclinicsandschools.36Indeed,investinginbasicservicesinlow-incomecitiesrepresentsaleaptowardintegration.Itnotonlybuildsresilienceinvulnerablecommunitiesbutalsoenhancesmobilitybyreducingmigrationbarriersbetweenthem.37OverviewFigureO.15Rangesofprojectedclimatechange–relatedhazardscoresacrosscitiesforselectedcountriesSource:WorldBankanalysisbasedondatafromMoody’sESGSolutions,Sub-SovereignPhysicalClimateRiskScores,October2021(https://esg.moodys.io/climate-solutions).Note:Thefigureisorganizedindescendingorderoftheaverageclimatehazardexposurescoresineachincomegroup.Theendpointsofthesolidblacklinesindicatethemaximumandminimumscoresforeachcountry.Thesolidhorizontalcoloredlinestartsatthefirstquartileandendsatthethird.Themarkerindicatesthemedianvalue.Climatehazardexposurescorescombineinformationonsixkeyhazards—floods,heatstress,tropicalcyclones,sea-levelrise,waterstress,andwildfires—projectedto2030–40.Urbanexposuretocombinedclimatechange–relatedhazardscore,selectedcountries,bycountryincomegroupHigh-incomecountriesClimatehazardexposurescoreUpper-middle-incomecountriesLow-andlower-middle-incomecountriesHondurasHaitiMadagascarVietnamIndonesiaBangladeshAlgeriaIndiaJamaicaDominicanRepublicMalaysiaThailandBrazilChinaArgentinaBarbadosTrinidadandTobagoBruneiDarussalamJapanUnitedStatesAustraliaFrance0255075100Investmentscanalsoanticipatetheimpactofshocksandstresses.Oneofthegreatbenefitsofcitiesisthattheycreatethedensityofdemandthatcanjustifylargesunkinvestmentssuchaspublictransportationsystems,including,forsufficientlylargecities,masstransitsystems.Thesepublictransportationsystemsnotonlyhaveimportanceforurbanproductivity,acces-sibility,andlabormarketoutcomesbutalsoactasakeylevertoreducingemissionsofbothCO2andlocalairpollutantssuchasPM2.5.Inastudyofthe58subwaysystemopeningsthat38THRIVINGoccurredincitiesgloballybetweenAugust2001andJuly2016,Gendron-Carrieretal.(2022)findthat,forcitieswithhigherinitialpollutionlevels,subwayopeningsledtoasignificantreductioninpollutionintheareasurroundingacitycenter.Suchsubwayopeningshave,inturn,generatedsignificanthealthbenefitsforcitieswithhigherinitialpollutionlevels.Meanwhile,thisreportdemonstratesthaturbanizationinhigh-riskareas—whethermeasuredbyheatorfloods—seemstobeoutpacingsettlementgrowthinsafeareas(DesmetandJedwab2022;Rentschleretal.2022).Relocatingorretrofittingtheseneighborhoodsisdifficultandcostly,andcanprovepoliticallysensitive.Instead,anticipatingurbangrowthandguidingitspatiallycanbeamuchmoreeffectiveandcost-efficientoption.Layingoutbasicinfrastructurecanactasapowerfulsignalforhouseholdstosettleinareasthatauthoritieshaveidentified,awayfromhighrisks.Theearlystagesofgrowthrequireonlythemostbasicinfrastructure—essentiallyrights-of-wayforroadsandwell-demarcatedlandplots(Angel2012).Scalinguptheinfrastructurecanhappeninasecondphaseoncehouseholdshavesettledin.Michaelsetal.(2021)findthatinTanzaniamodestinfrastructureinvestmentsingreenfieldareaswherepeoplesubsequentlybuilttheirownhouseshelpedfacilitatelong-runneighborhooddevelop-mentintermsoflarger,moreregularlylaid-outbuildingsandbetter-qualityhousing.Finally,investmentscanbeusedtoretrofit.Forexample,retrofittingofresidentialhomesandbuildingscouldhaveimportantimpactsongreeningthroughitseffectsonenergyconsump-tion.Makingimprovementstotheexistinghousingstockcanincreaseresiliencetotropicalcyclones,landslides,floods,andothernaturalhazards.However,retrofittingbuildingstruc-turesandinfrastructureindenselypopulatedareastohelpthemadapttonaturalhazardscanbeacostlysolution,requiringlargeoutlaysoftimeandmoney.Nature-basedinvestmentsmaybemorecost-effectiveinsuchsituations.Narayanetal.(2016)carriedoutasurveyofmultiplenature-basedprojectsaimedatdefendingcoastalhabitats,comparingthemwithinvestmentsin“grayinfrastructure.”37Theyfindthatnature-basedinvestmentscanbehighlycost-effectiveforprotectingcoastalsettlements.Trees,wetlands,greenspaces,andriverscanalleviatetheurbanheatislandeffect(Rajetal.2020;Tan,Lau,andNg2016).Nature-basedsolutionsalsoreducetheimpactofnaturalhazards,suchasflooding,erosion,landslides,anddroughts,incities—oftenbycomplementinggrayinfrastructuresuchasstormdrains,embankments,andretainingwalls.38Whomakesthechoices?Becausecitieswilllikelybearanoutsizedshareofclimateimpacts,cityleadersareprobablythemostmotivatedpoliticalactorstotakeonclimatechange.Cityleadersalsohaveknowl-edgeoflocalcontextandtheabilitytomobilizetheircommunities.Assuch,theycaninfluenceandimplementclimatepoliciesputinplacebyhigherlevelsofgovernment;designandimple-mentcity-specificpoliciesandinitiatives;and,crucially,helpcoordinatecollectiveclimateactionintheircities(DeConincketal.2018).Higherlevelsofgovernmentmayneedtocommittopolicyandinvestmentapproachesthatsupportlocalgovernmentsandgivethemincentivestobetterplanforandinvestinaddress-ingtheimpactsofclimatechange.Nationalgovernmentscanprovidestrategicoversight,facilitateaccesstoclimatefinance,andexercisetheircapacityandauthoritytodriveclimateactionbycreatingasupportiveenablingenvironment.Nationalprogramsonemissionsandcleanenergystandards,carbonpricingmechanisms,appliancestandards,andgreenfinanc-ingaremorelikelytoachieveeconomiesofscalebycreatinglargermarketsforhigh-to39Overviewlow-techcleanertechnologies.Throughpolicyandregulatoryinterventions,governmentscouldspureconomicrestructuringbymanagingtransitionandenablinggreengrowth.Nationalgovernmentsplaytheleadingroleinembeddingsocialprotectionintoclimateplansandshouldfocusonclimateriskswithinsocialpolicies(Costellaetal.2021).Moreover,theyholdthekeytosettingpolicyframeworksforinsuranceandcanprovidecoverforhighlevelsofphysicalandbusinessrisks.Itiswidelyagreedthatclimateactionwillrequiremultilevelinvolvementnotonlybycityandnationalgovernmentsbutalsobynonstateactorssuchasmultilateralinstitutions,largemultinationalcorporations,smallenterprises,andcivilsocietygroups.Privatefinancialflowscancontributeinseveralwaystotacklingclimaterisks—fromportfolioequitytodirectinvestments,tocommercialbanklending,tobondfinance.Atthegrassrootslevel,communitiesoftentaketheleadonclimateaction.Howcanchoicesbemade?Howdopolicymakerschooseamongthedifferentbundlesofpoliciesinawaythatwillproducethegreatestpositiveimpactonthemostpeopleinthemostefficientmanner?TheymusttogglebetweenandsandwichtogetherthebundlesofpolicyinterventionsinthefiveI’stoarriveatgreener,moreresilient,andmoreinclusiveoutcomes.SequencingofthefiveI’sTheorderingofthefiveI’sisdeliberateandrepresentsapotentialsequencingofinstruments.Itgoesfromaddressinginformationfailuresandissuesofexternalitiesandmissingmarketstogovernment-fundedinvestmentsingreen,resilient,andinclusiveinfrastructure.Thereason-ingunderlyingthesequencingmeldsefficiencyandefficacy—thatis,itbalancestheneedtomaximizethedesiredeffectsofapolicywiththeneedtodoitinthemosteconomicalway.——Earlyandeasyinterventionslinkedtotheprovisionofinformationcouldhavelargeknock-oneffectsonimprovingmarketoutcomes.Withaccesstomoreinformation,house-holdsandbusinessescanbetterunderstandthebenefitsandcostsoftheiractions,includinglocationdecisions,inwell-regulatedmarkets.Suchimproveddecision-makingcouldstemtheneedforexpensivegovernmentinterventions,includingpostdisasterrecovery.——Inthesamevein,well-implementedincentivescanscalequickly(forexample,byaffectingbehaviors)atrelativelylowcosts.Economicincentives(suchastaxrebatesorsubsidies)canhavemonetaryimplicationsthatcouldleadtosnowballingcosts,buttheycouldalsotaketheformofdisincentivesthroughmonetaryfines,taxes,andthelike.Removingfossilfuelandothersubsidiesthatdistortbehaviorinawaythatcontributestoclimatechangeandotherenvironmentalillssuchaslocalairpollutioncouldalsofreeupgovernmentexpenditure.——Well-functioningmarketsforinsurancecouldlowerriskstoapointthatwouldminimizetheneedforgovernmentstomakeexpensiveinterventionsandreducetheneedforpostdisasterrelief.Insurancecouldtransferandmitigaterisks,helpingallocateeconomicresourcesmoreefficiently,therebystimulatinggrowth.40THRIVING——Integrationincludesbetterplanningforcities,oftenandideallybeforethebuildingofurbansettlements.Giventhedurabilityofurbaninvestments,well-plannedintegrationcouldhavelong-lastingeffects,notleastforthebalancesheet.Integrationincludespoliciesandreformstoreducebarrierstomigrationbetweencitiesandbetweencitiesandruralareas,which,especiallyforcountrieswithalargenumberofcitiesacrosswhichtheseverityofclimatehazardsvary,canhelpadaptation.——Finally,investmentsininfrastructureofteninvolvelargepublicoutlays,buttheycanhelpshapeacity’sforminawaythatfundamentallyaffectsitsemissionsofbothCO2andlocalairpollutants.Inmanycases,investmentsalsoconstitutetheprimary(andverynecessary)responsefollowingdisasters.Durableinvestmentsinmitigationandadaptationstrategiesarecruciallyimportantbecauseoftheirimplicationsforlonger-termoutcomes.ThepolicyinstrumentsrepresentedbythefiveI’sprovidearelativelysimpleapproachtoorga-nizingthemanypoliciesavailableintodistinctbundles.Nevertheless,manyinterdependenciesexistbetweenthesesetsofinstruments.Insome,theinstrumentsplayoutincomplementaryways,whereinsomepoliciesacrossthebundleshaveastrongerimpactwhenimplementedtogether.Examplesincludehowinformationhelpsfacilitatemigrationdecisionsandthusinte-gration,andhowinformationallowspricestobetterreflectrisks,therebybetterincentivizingbehaviors.Largepublicinvestmentsthemselvessignalbusinessesandhouseholds(informa-tion)aboutthedirectionoffuturedevelopment.Andincentivesviapricesorregulationcandriveinsurancemarketsmoreefficiently,affecting,inturn,investmentdecisionsbyhouseholdsandbusinesses.Thus,decisionsacrossthefiveI’sarenotmadeinisolationofeachotherbutare,infact,interrelatedandcompounding.TailoringofthefiveI’sPolicymakersneedtotailortheapplicationofthefiveI’s,andthespecificpolicyinstrumentsthatfallwithinthem,toacity’sgreenness,resilience,andinclusivenesschallenges,keepinginmindthatthesechallengesvarywithbothacity’ssizeanditslevelofdevelopment.Inthatcontext,buildingonthisreport’sglobaltypologyofcities(boxO.1),annexOAprovidesadetailedmatrixthatmapsthefiveI’spolicyoptionstotheinterrelatedchallengesarisingfromurbangrowthandclimatechangethatconfrontsmall,medium,andlargecitiesinallcoun-tries.Onlywithsuchtailoringcanpolicybeeffective—thatis,onesizedoesnotfitall.However,eventhetailoringpresentedinannexOArepresentsbutacrudeguidetothemostappropriatepoliciesforanygivencity,andpolicymakerswillneedtofurthertailorthemonthebasisofthespecificlocalcircumstancesthatconfrontacityofanygiventype.Challengesrelatedtoclimatechangeandurbanizationareintensifyingacrosscitiesglobally,especiallycitiesinlow-andmiddle-incomecountries.Toachievegreen,resilient,andinclusiveurbandevelopment,policymakersatboththenationalandlocallevelsneedtoworktogethertoaddresstheseinterrelatedchallengeshead-on.TheycandosobydrawingonthefiveI’ssuiteofpolicyinstruments.Byactingnowtoapplytheseinstrumentsinanappropriatelytailoredmanner,policymakerscanensurethattheworld’scitiesnotonlysurvivebutthriveinthefaceoftheperilsofclimatechange.41OverviewTailoredpolicyoptionsbytypeofcityandinstrumentANNEXO.A42THRIVINGINFORMATIONINCENTIVESINSURANCEINTEGRATIONINVESTMENTSpolicyoptionstoaddresschallengesCitysizeSmallMed.LargeSmallMed.LargeSmallMed.LargeResiliencePovertyBasicservicesInequalityVegetationGreenhousegas(GHG)Pollution=Moderate=SevereMAINCHALLENGESLow-andlower-middle-incomecountryUpper-middle-incomecountryHigh-incomecountryCitysizeSmallMed.LargeSmallMed.LargeSmallMed.LargeEarlywarningsystems;hazardmappingandassessmentBuildinstitutionalcapacityDecentralizedlandadministrationservicesParticipatoryriskawarenessJobfairsandlocalforumsGreenhousegas(GHG)emissionsinventoriesPollutionmonitoringBetterzoningofpollutingindustriesUrbanplanningdocumentsUrbandesignguidelinesBuildingcodesDisasterrisk-informedlandvalueDisaster-risklanddevelopmentpenaltyLow-andlower-middle-incomecountryUpper-middle-incomecountryHigh-incomecountryNote:Med.=Medium.Note:Forbrevity,thistableonlycoverspolicyoptionsforseverechallengesforlow-andmiddle-incomecountries.Med.=Medium.43OverviewNote:Forbrevity,thistableonlycoverspolicyoptionsforseverechallengesforlow-andmiddle-incomecountries.Med.=Medium.Note:Forbrevity,thistableonlycoverspolicyoptionsforseverechallengesforlow-andmiddle-incomecountries.Med.=Medium.INFORMATIONINCENTIVESINSURANCEINTEGRATIONINVESTMENTSpolicyoptionstoaddresschallengesINFORMATIONINCENTIVESINSURANCEINTEGRATIONINVESTMENTSpolicyoptionstoaddresschallengesLow-andlower-middle-incomecountryUpper-middle-incomecountryHigh-incomecountryCitysizeSmallMed.LargeSmallMed.LargeSmallMed.LargePhaseoutfossilfuelsubsidiesCashtransfersWorkfareprogramsSubsidizedhousingCongestioncontrolschemes/pricingParkingcharges/reformReformstoreducecostsofverticalconstruction;relaxedheightrestrictionsCarbontaxesInclusionaryzoningDensitybonusExpeditedpermittingFast-trackprojectreviewBuildingretrofitandcleanenergysubsidiesandtaxcreditsElectricvehicletaxcreditPerformancezoningRetrofitincentivesLinkagefeesAirrightsprogramLow-andlower-middle-incomecountryUpper-middle-incomecountryHigh-incomecountryCitysizeSmallMed.LargeSmallMed.LargeSmallMed.LargeSocialprotectionSubsidizedinsurance(low-riskareas)CatastropheinsuranceIncorporateclimateriskconsiderationsinasset(re-)pricing,newinsuranceproductlaunches,andunderwritingprocess44THRIVINGINFORMATIONINCENTIVESINSURANCEINTEGRATIONINVESTMENTSpolicyoptionstoaddresschallengesCitysizeSmallMed.LargeSmallMed.LargeSmallMed.LargeIntegrateclimatechangeadaptationandurbanmanagement;urbanplanningandregulationBasicservices;educationFlexibleurbanplanningConnecttomediumandlargecitiesLowermigrationbarriersLaying-outofstreetnetworkinanticipationoffutureexpansionSecurelandandpropertyrightsIntegratedlanduseandtransportationplanningTransit-orienteddevelopmentLow-andlower-middle-incomecountryUpper-middle-incomecountryHigh-incomecountryINFORMATIONINCENTIVESINSURANCEINTEGRATIONINVESTMENTSpolicyoptionstoaddresschallengesCitysizeSmallMed.LargeSmallMed.LargeSmallMed.LargeWell-locatedaffordablehousingLocalbusservicesLandprovisionImprovebuildingstockClimateadaptationinfrastructureNature-basedsolutionsRenewableenergyBusrapidtransit(BRT)Massrapidtransit(MRT);lightrailtransit(LRT)Energy-efficientretrofitsUrbangreenspaceMobilityLow-andlower-middle-incomecountryUpper-middle-incomecountryHigh-incomecountryNote:Forbrevity,thistableonlycoverspolicyoptionsforseverechallengesforlow-andmiddle-incomecountries.Med.=Medium.Note:Forbrevity,thistableonlycoverspolicyoptionsforseverechallengesforlow-andmiddle-incomecountries.Med.=Medium.Overview4545OverviewNotes1.TheEarth’snatural“greenhouseeffect”wasfirstdescribedbyFrenchphysicistJosephFourierin1824.SwedishchemistSvanteArrheniussubsequentlyconcludedin1869thatindustrialagecoalburningenhancesthenaturalgreenhouseeffect.In1938,theBritishengineerGuyCallendarusedrecordsfrom147weatherstationsaroundtheworldtoshowthattemperatureshadincreasedoverthepreviouscenturyandthat,overthesameperiod,carbondioxide(CO2)emissionshadalsoincreased.Onthebasisofthiscorrelation,hesuggestedthattheincreaseinCO2emissionshadbeenresponsibleforglobalwarming(aneffectreferredtoasthe“Callendareffect”).Fora“briefhistoryofclimatechange,”seeBBCNews(2013).2.Theglobalurbanpopulationclimbedfrom1.19billionin1970to4.53billionin2022.Theurbanpopulationdatacitedherearedrawnfromthefollowingdatabase:UnitedNations,DepartmentofEconomicandSocialAffairs,WorldUrbanizationProspects2018(revision),https://population.un.org/wup/.3.Thesurfacetemperatureisaveragedacrosslandandwater.Thepreindustrialperiodisdefinedas1880–1900.DatacomefromLindseyandDahlman(2021).4.Conversely,extremecoldeventshavedecreasedinfrequencysincethe1970s.5.Citieshelpgenerateprosperityboththroughtheagglomerationeconomiestowhichtheygiveriseandthestructuraltransformationfromagrariantononagrarianactivitiestheyhelpfacilitate.Historically,thisprosperityhas,inturn,helpeddrivedemandforenergyandthereforefossilfuels.Causationhasalsohistoricallyruninthereversedirection,withthesupplyoffossilfuelenergyhelpingspurthegrowthofindustryandcities.6.Thisis,inpart,throughnetworkssuchastheC40CitiesClimateLeadershipGroup,theWorldMayorsCouncilonClimateChange(WMCCC),andtheUrbanClimateChangeResearchNetwork(UCCRN).7.Natureprovidesessentialinputsforhumanlife,health,andprosperity;economiststhereforetreatitasanasset,ornaturalcapital.8.Thewelfareimpactofair,water,andsoilpollutionwasestimatedtohavebeenequivalentto6.2percentofglobaloutputin2015.Ninety-twopercentofpollution-relateddeathsandthehighestburdenofeconomiclosseswereinlow-andmiddle-incomecountries.Globally,poorlymanagedurbanizationhasbeenamongthemajordriversofpollution-relateddeaths(Landriganetal.2017).9.France’s“yellowvest”protestsstartedinNovember2018asaresponsebymotoriststotheannouncementthatagreentaxonfuelwouldgointoeffectonJanuary1,2019,aspartoftheFrenchgovernment’senvironmentalpolicystrategy.Theoriginalprotestors,fromperi-urbanandruralareas,hadtodrivelongdistancesdaily.TheprotestsquicklyspreadtoParis,however,andturnedviolent.TheFrenchgovernmentabandoneditsplannedintroductionofthegreentaxinDecember2018.10.Economistsrefertothesestressesvariouslyascongestionandcrowdingeffects,anddiseconomiesofagglomeration.11.FossilCO2emissionsaccountedfor77percentoftheworld’santhropogenicGHGemissionsin2015.Since2000,theirincreaseconstitutesthemainsourceoftheglobalincreaseinGHGemissions(Crippaetal.2019).12.Additionalregressionanalysis,discussedinchapter1,showsthat,forboththeresidentialandtransportationsectors,astatisticallysignificantpositivecorrelationremainsbetweenacity’sCO2emissionsandtheincomelevelofthecountryinwhichthecityislocatedevenaftercontrollingforacity’sownclimate(temperature,precipitation,biome,andelevation).Thus,therelationshipsarenotdrivensolelybydifferencesindemandforheatingandcoolingassociatedwithtemperature.13.Onthebasisofcurrentpolicies,upper-middle-andhigh-incomecountries,asanaggregate,arenotoncoursetoachievenetzeroemissionsby2050,accordingtoversion2scenariomodelingbytheNetworkofCentralBanksandSupervisorsforGreeningtheFinancialSystem.Accordingtothesescenarios,ifallcountriesmaintaintheircurrentpolicies,globalGHGemissionsin2050willbealmost16percenthigherthantheywerein2020.Evenifupper-middle-andhigh-incomecountriesachievetheirnationallydeterminedcontributionswhilelow-andlower-middle-incomecountriesretaintheircurrentpolicies,globalGHGemissionsin2050willbeonly8.2percentlowerthanin2020.Thislevelfallsfarshortofthatrequiredtolimitglobalwarmingto1.5°C.14.Thedataonprojectedclimatehazardscover2,208citiesgloballyandcomefromMoody’sESGSolutions.Chapter2includesadescriptionofthemethodologyonwhichtheMoody’sclimatehazardscoresisbased.15.Resilienceisdefinedhereastheabilityofcitiestorecovertheirpopulationlevelsandeconomicvitalityafteranadverseshock.16.TropicalcyclonesintheAtlanticarealsoreferredtoas“hurricanes”;inthePacifictheyarealsoreferredtoas“typhoons.”17.Forexample,globally,theaveragenumberofdaysayearacity’stemperaturewasextremelyhotrelativetoitsownhistoricalexperienceincreasedfromjustundertwodaysinthe1970stomorethan41daysovertheperiodfrom2010to2020—astaggering21-foldincreaseinjustfourdecades.Extremeheateventsalsoincreasedinintensityoverthisperiod.Chapter1providesmoredetailsontheevolutionofextremeweatherevents.18.TheresultspresentedinfigureO.4arelimitedtocitieswitha2015populationofatleast200,000.19.The1985Villachconference,organizedbytheUnitedNationsEnvironmentProgram,WorldMeteorologicalOrganization,andInternationalScienceCouncil,wasnotableforitsrecognitionthatclimatechangewasoccurringmuchmorerapidlythanpreviouslythought.Theconferencealsoissuedthefirstcallfromtheworld’sleadingclimatescientistsforcollaborationbetweenscientistsandpolicymakerstoexplorepolicyoptionsforaddressingclimatechange.20.Datausedforthecalculationscitedherecover3countriesinEastAsiaandPacific(Cambodia,Myanmar,andTimor-Leste);5countriesinLatinAmericaandtheCaribbean(Colombia,DominicanRepublic,Guatemala,Haiti,andHonduras);3countriesinSouthAsia(Bangladesh,India,andNepal);and29countriesinSub-SaharanAfrica(EastAfrica:Burundi,Comoros,Ethiopia,Kenya,Malawi,Mozambique,Rwanda,Tanzania,Uganda,Zambia,andZimbabwe;WestAfrica:Benin,BurkinaFaso,Côted’Ivoire,Ghana,Guinea,Liberia,Mali,Nigeria,Senegal,SierraLeone,andTogo;CentralAfrica:Angola,Cameroon,Chad,DemocraticRepublicofCongo,andGabon;southernAfrica:LesothoandNamibia).21.Thisfindingissuggestiveoftheso-calledenvironmentalKuznetscurverelationship.THRIVING4622.ForadescriptionofaproposedtechnologicalinnovationforreducingtheCO2emissionsembeddedinPortlandcement,themostwidelyusedstandardvarietyofcement,seeEllisetal.(2019).23.TheUnitedNationsHighCommissioneronRefugeesestimatesthat89.3millionpeopleworldwidewereforcedtofleetheirhomesin2021becauseofconflicts,violence,fearofpersecution,andhumanrightsviolations.Thisnumberismorethandoublethe42.7millionpeoplewhoremainedforciblydisplacedadecadeagoandthemostsinceWorldWarII(https://www.unhcr.org/en-us/figures-at-a-glance.html).24.However,evengoodtransportationnetworksmaynotbeabletomitigatetheimpactsofsomenonclimatestressors,asillustratedbytheconflictinUkraine.25.Forexample,arichbodyofliteratureindevelopmenteconomicsdocumentshowthepoorpossesslowerfinancialliteracyandhowtheprovisionoffinancialinformationcanhavesubstantialpositivewelfareeffects.See,amongothers,Hastings,Madrian,andSkimmyhorn(2013);Karlan,Ratan,andZinman(2014);LusardiandMitchell(2014).26.JiangandQuintero(2022)defineanextremelyhotdayasoneinwhichthetemperatureduringatleastonehourduringthedayexceeds35°C,eveniftheaveragetemperatureoftheentiredayislower.27.AsfigureO.14indicates,climatechangemayalsopotentiallyinteractwithagglomerationeconomies.Forexample,morefrequentextremeweatherassociatedwithclimatechangecouldmakeitlesslikelythatpeoplecanmingle,therebyreducingthelikelihoodofknowledgespillovers,oritcouldinterruptdenselocalsupplychains.Ingeneral,however,themechanismsthroughwhichclimatechangemayaffectthestrengthofagglomerationeconomiesarelesswellunderstoodandresearchedandthusaredownplayedinthisreport.28.SeeGuardianarticleonwhetherclimatereportingshiftsviewpoints(Harvey2022).29.Forexample,HongKongSAR,China,hasinvestedinhousingimprovementsthatallowforshelteringathomeduringtropicalcyclonesandinearlywarningsystemsthatallowpeopletoreturnsafelytotheirhomesusinganadaptivepublictransportationsystem(RogersandTsirkunov2010).30.The2019GlobalCommissiononAdaptationreportfindsthatearlywarningsystemsprovideda10-foldreturnoninvestment—thegreatestofanyadaptationmeasureincludedinthereport(GCAandWRI2019).31.Inthepresenceofalow-priceelasticity—suchassuggestedbythemeta-analysiscarriedoutbyGalindoetal.(2015)inLatinAmerica—afueltaxwillbeinadequatetocontrolrisingconsumption.32.Theurbanpopulationdatacitedherearedrawnfromthefollowingdatabase:UnitedNations,DepartmentofEconomicandSocialAffairs,WorldUrbanizationProspects2018(revision),https://population.un.org/wup/.33.Seealsoevidencepresentedinchapter1.34.Ithasbeenestimatedthatthecostofproactiveplanning(suchasviasitesandservices)fortheprovisionofaffordableandsafehousingtoaccommodatetheburgeoningpopulationinFreetown,SierraLeone,wouldcostapproximatelyUS$375million(Mukim2018).Bycontrast,theprovisionofapublichousingschemewouldcostalmostninetimesasmuch(US$3.2billion).Overview474835.InJamaica,coastalhazardmappingisunderwaywiththeintenttoupdatelanduseregulation,butintandemlocalauthoritiescontinuetosuccessfullyenforceminimallyintrusiveandlow-costhurricanestraps,whichareconnectors,oftenmadeofgalvanizedorstainlesssteel,usedtostrengthenwood-framedroofsandhomes.36.ForAfricancities,FosterandBriceño-Garmendia(2010)estimatethatdoublingurbandensityreducesthepercapitacostofapackageofinfrastructureimprovementsbyabout25percent.37.Grayinfrastructurereferstobuiltstructuresandengineeringequipment(suchasreservoirs,embankments,andcanals)thatareembeddedinwatershedsorcoastalecosystems.38.SeeWorldBank(2021a)foranoverviewoftheliteratureandexamplesinwhichsuchsolutionscanhelpcitiestargetclimateresilience.ReferencesAbanokova,K.,H-A.Dang,S.Nakamura,S.Takamatsu,C.Pei,andD.Prospere.2022.“IsClimateChangeSlowingtheUrbanEscalatoroutofPoverty?EvidencefromIndonesiaandLAC.”Backgroundpaperpreparedforthisreport,WorldBank,Washington,DC.Acharya,G.,E.Cassou,S.Jaffee,andE.K.Ludher.2021.RICHFood,SmartCity:HowBuildingReliable,Inclusive,Competitive,andHealthyFoodSystemsIsSmartPolicyforUrbanAsia.Washington,DC:WorldBank.Ahlfeldt,G.M.,andR.Jedwab.2022.“TheGlobalEconomicandEnvironmentalEffectsofVerticalUrbanDevelopment.”Backgroundpaperpreparedforthisreport,WorldBank,Washington,DC.Angel,S.2012.PlanetofCities.Hollis,NH:PuritanPress,Inc.Arrow,K.J.1963.“UncertaintyandtheWelfareEconomicsofMedicalCare.”AmericanEconomicReview53(5):941–73.Bakkensen,L.A.,andL.Barrage.2022.“GoingUnderwater?FloodRiskBeliefHeterogeneityandCoastalHomePriceDynamics.”ReviewofFinancialStudies35(8):3666–709.Banerjee,A.V.,andS.Mullainathan.2008.“LimitedAttentionandIncomeDistribution.”AmericanEconomicReview98(2):489–93.Baylis,P.,andJ.Boomhower.2019.“MoralHazard,Wildfires,andtheEconomicIncidenceofNaturalDisasters.”NBERWorkingPaperw26550,NationalBureauofEconomicResearch,Cambridge,MA.https://ssrn.com/abstract=3504434.BBCNews.2013.“ABriefHistoryofClimateChange.”BBCNews,September20,2013.https://www.bbc.com/news/science-environment-15874560.Benton,G.1970.“CarbonDioxideandItsRoleinClimateChange.”ProceedingsoftheNationalAcademyofSciences67:898–91.Bosker,M.,S.Brakman,H.Garretsen,andM.Schramm.2012.“RelaxingHokou:IncreasedLaborMobilityandChina’sEconomicGeography.”JournalofUrbanEconomics72(2):252–66.THRIVING49Chlouba,V.,M.Mukim,andE.Zaveri.2022.“HowDoClimateChange-RelatedStressorsAffectUrbanForm?”Backgroundpaperpreparedforthisreport,WorldBank,Washington,DC.ClimateWatch.2022.“GHGEmissions.”WorldResourcesInstitute,Washington,DC.https://www.climatewatchdata.org/ghg-emissions.Costella,C.,A.McCord,M.vanAalst,R.Holmes,J.Ammoun,andV.Barca.2021.“SocialProtectionandClimateChange:ScalingUpAmbition.”SocialProtectionApproachestoCOVID-19ExpertAdviceService(SPACE),DevelopmentAlternativesIncorporated(DAI)Global,LLC,Bethesda,MD.Crippa,M.,G.Oreggioni,D.Guizzardi,M.Muntean,E.Schaaf,E.LoVullo,E.Solazzo,etal.2019.FossilCO2andGHGEmissionsofAllWorldCountries.EUR29849EN.Luxembourg:PublicationsOfficeoftheEuropeanUnion.DeConinck,H.,A.Revi,M.Babiker,P.Bertoldi,M.Buckeridge,A.Cartwright,W.Dong,etal.2018.“StrengtheningandImplementingtheGlobalResponse.”InGlobalWarmingof1.5°C,editedbyV.Masson-Delmotte,P.Zhai,H.-O.Pörtner,D.Roberts,J.Skea,P.R.Shukla,A.Pirani,etal.Geneva:IntergovernmentalPanelonClimateChange.Deryugina,T.2014.“TheFiscalCostofHurricanes:DisasterAidversusSocialInsurance.”AmericanEconomicJournal:EconomicPolicy9(3):168–98.Desmet,K.,andR.Jedwab.2022.“AreWeOver-buildingin‘Bad’LocationsGlobally?FutureClimateChangeandDurableRealEstate.”Backgroundpreparedpaperforthisreport,WorldBank,Washington,DC.Deuskar,C.2022.“BeatingtheHeat:MeasuringandMitigatingExtremeHeatinEastAsianCities.”Unpublishedmanuscript.Dijkstra,L.,A.Florczyk,S.Freire,T.Kemper,M.Melchiorri,M.Pesaresi,andM.Schiavina.2021.“ApplyingtheDegreeofUrbanisationtotheGlobe:ANewHarmonisedDefinitionRevealsaDifferentPictureofGlobalUrbanisation.”JournalofUrbanEconomics125:103312.Dijkstra,L.,andH.Poelman.2014.“AHarmonisedDefinitionofCitiesandRuralAreas:TheNewDegreeofUrbanization.”RegionalWorkingPaper,Directorate-GeneralforRegionalandUrbanPolicy,EuropeanCommission,Brussels.Ellis,L.D.,A.F.Badel,M.L.Chiang,andY.-M.Chiang.2019.“TowardElectrochemicalSynthesisofCement-AnElectrolyzer-BasedProcessforDecarbonatingCaCO3whileProducingUsefulGasStreams.”PNAS117(23):12584–91.Erman,A.,M.Tariverdi,M.Obolensky,X.Chen,R.C.Vincent,S.Malgioglio,J.Rentschler,etal.2019.“WadingOuttheStorm:TheRoleofPovertyinExposure,VulnerabilityandResiliencetoFloodsinDaresSalaam.”PolicyResearchWorkingPaper8976,WorldBank,Washington,DC.Foster,V.,andC.Briceño-Garmendia.2010.Africa’sInfrastructure:ATimeforTransformation.AfricaDevelopmentForum.Washington,DC:WorldBank.Galindo,L.,J.Samaniego,J.Alatorre,J.FerrerCarbonell,andO.Reyes.2015.“Meta-analysisoftheIncomeandPriceElasticitiesofGasolineDemand:PublicPolicyImplicationsforLatinAmerica.”CEPALReview117:7–25.Gandhi,S.,M.E.Kahn,R.Kochhar,S.V.Lall,andV.Tandel.2022.“AdaptingtoFloodRisk:EvidencefromaPanelofGlobalCities.”WorkingPaper30137,NationalBureauofEconomicResearch,Cambridge,MA.OverviewTHRIVINGTHRIVINGTHRIVINGGCA(GlobalCenteronAdaptation)andWRI(WorldResourceInstitute).2019.AdaptNow:AGlobalCallforLeadershiponClimateResilience.Rotterdam:GCA.Gendron-Carrier,N.,M.Gonzalez-Navarro,S.Polloni,andM.A.Turner.2022.“SubwaysandUrbanAirPollution.”AmericanEconomicJournal:AppliedEconomics14(1):164–96.Glaeser,E.L.2012.TriumphoftheCity:HowOurGreatestInventionMakesUsRicher,Smarter,Greener,Healthier,andHappier.NewYork:PenguinPress.Glaeser,E.L.2022.“UrbanResilience.”UrbanStudies59(1):3–35.Gregory,J.2017.“TheImpactofPost-KatrinaRebuildingGrantsontheResettlementChoicesofNewOrleansHomeowners.”Unpublishedmanuscript.Harvey,F.2022.“TruthfulClimateReportingShiftsViewpoints,butOnlyBriefly,StudyFinds.”Guardian,June20,2022.https://www.theguardian.com/environment/2022/jun/20/truthful-climate-reporting-shifts-viewpoints-but-only-briefly-study-finds.Hastings,J.S.,B.C.Madrian,andW.L.Skimmyhorn.2013.“FinancialLiteracy,FinancialEducation,andEconomicOutcomes.”AnnualReviewofEconomics5(1):347–73.Heger,M.,D.Wheeler,G.Zens,andC.Meisner.2019.MotorVehicleDensityandAirPollutioninGreaterCairo:FuelSubsidyRemovalandMetroLineExtensionandTheirEffectonCongestionandPollution.Washington,DC:WorldBank.Henderson,J.V.,andM.Turner.2020.“UrbanizationintheDevelopingWorld:TooEarlyorTooSlow?”JournalofEconomicPerspectives34(3):150–73.Hoekstra,A.Y.,J.Buurman,andK.C.H.vanGinkel.2018.“UrbanWaterSecurity:AReview.”EnvironmentalResearchLetters13(5):053002.Hopkins,F.,J.Ehleringer,S.Bush,R.Duren,C.Miller,C.Lai,Y.Hsu,etal.2016.“MitigationofMethaneEmissionsinCities:HowNewMeasurementsandPartnershipsCanContributetoEmissionsReductionStrategies.”Earth’sFuture4:408–25.IPCC(IntergovernmentalPanelonClimateChange).2021.ClimateChange2021:ThePhysicalScienceBasis.ContributionofWorkingGroupItotheSixthAssessmentReportoftheIntergovernmentalPanelonClimateChange.Cambridge,UK:CambridgeUniversityPress.IPCC(IntergovernmentalPanelonClimateChange).2022.“SummaryforPolicymakers.”InClimateChange2022:Impacts,Adaptation,andVulnerability,editedbyH.-O.Pörtner,D.C.Roberts,M.Tignor,E.S.Poloczanska,K.Mintenbeck,A.Alegría,M.Craig,etal.Cambridge,UK:CambridgeUniversityPress.Jiang,H.,andL.Quintero.2022.“Can’tStandtheHeat:ClimateStressonLaborProductivityin16LatinAmericanandCaribbeanCountries.”Backgroundpaperpreparedforthisreport,WorldBank,Washington,DC.Kahn,M.2010.“Climatopolis:HowWillClimateChangeImpactUrbanitiesandTheirCities?”VOX-EUBlog,September11,2010,CenterforEconomicPolicyResearch,London.Kahn,M.E.,andP.Li.2020.“AirPollutionLowersHighSkillPublicSectorWorkerProductivityinChina.”EnvironmentalResearchLetters15(8):084003.Karlan,D.,A.L.Ratan,andJ.Zinman.2014.“SavingsbyandforthePoor:AResearchReviewandAgenda.”ReviewofIncomeandWealth60(1):36–78.5050OverviewOverviewLai,D.,W.Liu,T.Gan,K.Liu,andQ.Chen.2019.“AReviewofMitigatingStrategiestoImprovetheThermalEnvironmentandThermalComfortinUrbanOutdoorSpaces.”ScienceoftheTotalEnvironment661(April):337–53.Lall,S.V.,J.Kaw,S.B.Murray,andF.Shilpi.Forthcoming.VibrantCities:PrioritiesforGreen,Resilient,andInclusiveUrbanDevelopment.Washington,DC:WorldBank.Landrigan,J.,R.Fuller,N.J.R.Acosta,O.Adeyi,R.Arnold,N.Basu,A.BibiBaldé,etal.2017.“TheLancetCommissiononPollutionandHealth.”Lancet391(10119):462–512.Lindsey,R.,andL.Dahlman.2021.“ClimateChange:GlobalTemperature.”Climate.gov,March15,2021.NationalOceanicandAtmosphericAdministration,Washington,DC.Lusardi,A.,andO.S.Mitchell.2014.“TheEconomicImportanceofFinancialLiteracy:TheoryandEvidence.”JournalofEconomicLiterature52(1):5–44.Madden,R.A.,andV.Ramanathan.1980.“DetectingClimateChangeduetoIncreasingCarbon-Dioxide.”Science209:763–68.Michaels,G.,D.Nigmatulina,F.Rauch,T.Regan,N.Baruah,andA.Dahlstrand.2021.“PlanningAheadforBetterNeighborhoods:Long-RunEvidencefromTanzania.”JournalofPoliticalEconomy129(7):2112–56.Mukim,M.2018.FreetownUrbanSectorReview:OptionsforGrowthandResilience.Washington,DC:WorldBank.Nakagawa,M.,M.Saito,andH.Yamaga.2007.“EarthquakeRiskandHousingRents:EvidencefromtheTokyoMetropolitanArea.”RegionalScienceandUrbanEconomics37(1):87–99.Narayan,S.,M.W.Beck,B.G.Reguero,I.J.Losada,B.vanWesenbeeck,N.Pontee,J.N.Sanchirico,etal.2016.“TheEffectiveness,CostsandCoastalProtectionBenefitsofNaturalandNature-BasedDefences.”PLoSONE11(5):e0154735.Park,J.,andM.Roberts.2023.“UrbanResilienceandWeatherShocks:AGlobalNighttimeLightsAnalysis.”Backgroundpaperpreparedforthisreport,WorldBank,Washington,DC.Pascale,S.,S.B.Kapnick,T.L.Delworth,andW.F.Cooke.2020.“IncreasingRiskofAnotherCapeTown‘DayZero’Droughtinthe21stCentury.”ProceedingsoftheNationalAcademyofSciences117(47):29495–503.Patankar,A.M.2015.“TheExposure,Vulnerability,andAbilitytoRespondofPoorHouseholdstoRecurrentFloodsinMumbai.”PolicyResearchWorkingPaper7481,WorldBank,Washington,DC.Pomponi,F.,R.Saint,J.H.Arehart,N.Gharavi,andB.D’Amico.2021.“DecouplingDensityfromTallnessinAnalysingtheLifeCycleGreenhouseGasEmissionsofCities.”UrbanSustainability1:33.Raj,S.,S.K.Paul,A.Chakraborty,andJ.Kuttippurath.2020.“AnthropogenicForcingExacerbatingtheUrbanHeatIslandsinIndia.”JournalofEnvironmentalManagement257:110006.Rentschler,J.,P.Avner,M.Marconcini,R.Su,E.Strano,S.Hallegatte,L.Bernard,andC.Riom.2022.“RapidUrbanGrowthinFloodZones:GlobalEvidencesince1985.”PolicyResearchWorkingPaper10014,WorldBank,Washington,DC.Rogers,D.,andV.Tsirkunov.2010.CostsandBenefitsofEarlyWarningSystems.GlobalAssessmentReportonDisasterRiskReduction2011.Washington,DC:WorldBank.5151Rossitti,G.2022.“WhoIsAffected?WhichSegmentsofCityPopulationsAreMostExposedtoClimate-RelatedStresses?”Backgroundpaperpreparedforthisreport,WorldBank,Washington,DC.Rubin,A.J.,andS.Sengupta.2018.“‘YellowVest’ProtestsShakeFrance.Here’stheLessonforClimateChange.”NewYorkTimes,December6,2018.https://www.nytimes.com/2018/12/06/world/europe/france-fuel-carbon-tax.html.Santos,G.,K.Button,andR.G.Noll.2008.“LondonCongestionCharging.”Brookings-WhartonPapersonUrbanAffairs2008:177–234.Singh,C.,G.Jain,V.Sukhwani,andR.Shaw.2021.“LossesandDamagesAssociatedwithSlow-OnsetEvents:UrbanDroughtandWaterInsecurityinAsia.”CurrentOpinioninEnvironmentalSustainability50:72–86.Song,S.,C.-C.Feng,andM.Diao.2020.“VehicleQuotaControl,TransportInfrastructureInvestmentandVehicleTravel:APseudoPanelAnalysis.”UrbanStudies57(12):2527–46.Tan,Z.,K.K.-L.Lau,andE.Ng.2016.“UrbanTreeDesignApproachesforMitigatingDaytimeUrbanHeatIslandEffectsinaHigh-DensityUrbanEnvironment.”EnergyandBuildings114:265–74.Varadhan,S.2019.“VillagersAccuseCityofSeizingWaterasDroughtParches‘India’sDetroit.’”Reuters,July3,2019.Venkat,A.,F.Dizon,andW.Masters.2022.“TheImpactofClimateShocksandStressesonUrbanFoodPrices.”Backgroundpaperpreparedforthisreport,WorldBank,Washington,DC.Votsis,A.,andA.Perrels.2016.“HousingPricesandthePublicDisclosureofFloodRisk:ADifference-in-DifferencesAnalysisinFinland.”JournalofRealEstateFinanceandEconomics53(4):450–71.WorldBank.2013.InclusionMatters:TheFoundationforSharedProsperity.NewFrontiersofSocialPolicy.Washington,DC:WorldBank.WorldBank.2017.SovereignCatastropheRiskPools:WorldBankTechnicalContributiontotheG20.Washington,DC:WorldBank.WorldBank.2020.Vietnam’sUrbanizationataCrossroads:EmbarkingonanEfficient,Inclusive,andResilientPathway.Washington,DC:WorldBank.WorldBank.2021a.ACatalogueofNature-BasedSolutionsforUrbanResilience.Washington,DC:WorldBank.WorldBank.2021b.SomaliaUrbanizationReview:FosteringCitiesasAnchorsofDevelopment.Washington,DC:WorldBank.Zaveri,E.,J.Russ,A.Khan,R.Damania,E.Borgomeo,andA.Jagerskog.2021.EbbandFlowVol.1:Water,MigrationandDevelopment.Washington,DC:WorldBank.THRIVINGTHRIVING525253OverviewForatleast50years,thebalanceofscientificevidencehassupportedtheviewthattheworldiswarming,andthathumanactivityhasspurredthatwarming(Benton1970;MaddenandRamanathan1980).Globallyduringthistime,thenumberofpeoplelivingincitieshasincreasedbyafactorof3.8,from1.19billionin1970to4.46billionin2021,1andtheEarth’ssurfacetemperaturehasclimbedby1.19°Caboveitspreindustriallevels.2Becauseoftheprosperitytheyhavehelpedgenerate,citieshavebeenamajorcauseofthisclimatechange(Kahn2010).Itisalsoincities,however,thatmanyofthesolutionstotheclimatecrisis—intermsofbothadaptationandmitigation—willbefound,notleastbecauseby2050almost70percentoftheworld’spopulationwillcallcitiestheirhome.Assuch,citiesarekeytoarguablythegreatestpublicpolicychallengeofourtimes.Nevertheless,urbaneconomistshave,withafewnotableexceptions,beensurprisinglysilentintermsoftheirresearchonclimatechange.3,4Inpart,thisreportisanattempttorectifythatneglect.Evenmoreimportant,thereportprovidesacompassforpolicymakersonpoliciesthatcanhelpcitiesthriveinthefaceoftheperilsofclimatechange.Climatechange,however,isbutonesymptomofanevenlargerproblem—themoregeneralerosionofnaturalcapitaltowhichpoorlymanagedurbanizationcontributes.Thaterosion,especiallyinmanylow-andmiddle-incomecountries,contributes,inturn,todangerouslypoorairqualityinmanycities,detrimentalcompetitionforwaterbetweenurbanandruralareas,theunnecessarylossoffertileagriculturalland,deforestation,andadamaginglossofbiodiversity.5Atthesametime,thesetrendsplayoutagainstabackdropofhighandrisinglevelsofinequalityinmanycitiesgloballyandstalledprogressintheworldwidefightagainstextremepoverty6—trendsthatthemselvesbothinteractwithandreinforcestressorsrelatedtoclimatechange—toaffectthegreenness,resilience,andinclusivenessofurbandevelopment(boxI.1).IntroductionDefininggreenness,resilience,andinclusivenessIn2021,inresponsetothechallengesposedbyclimatechange,themoregeneraldegradationoftheenvironment,theCOVID-19pandemic,stalledprogressintheglobalfightagainstextremepoverty,andincreasinginequalityinmanycountries,theWorldBanklauncheditsGRID(green,resilient,andinclusivedevelopment)strategy(WorldBank2021).Inkeepingwiththatstrategy,thisreportfocusesonthreebroadbutcloselyinterrelateddimensionsofacity’sdevelopment.••Greennessreferstoacity’simpactsontheenvironment,withagreenercityhavingmorebeneficialorlessnegativeenvironmentalimpacts.Acity’sgreennessisreflectedinitslevelofgreenhousegasemissions,includingitscarbondioxideemissions.Thisreportanalyzesotherdimensionsofacity’sgreennessaswell,includingitslevelofairpollution,theextenttowhichitpollutesbothitsownwaterandthatofsurround-ingareas,andacity’slevelofvegetation.Theseadditionaldimensionsofgreennesshaveimportantinterrelationshipswithclimatechange.Forexample,acity’sPM2.5BoxI.154THRIVINGemissions(thatis,emissionsofparticulatematterof2.5micronsorlessindiameter)arestronglycorrelatedwithitscarbondioxideemissions,whichimpliesthatactiontotacklelocalpollutiontendstocarryclimatechangemitigationco-benefitsandviceversa.Meanwhile,increasedvegetationcanhelpmitigatetheimpactsofextremeurbanheatassociatedwithbothclimatechangeandtheurbanheatislandeffect.Itcanalsohelpmitigateclimatechangeitselfthroughcarbonsequestration.••Resiliencereferstohowwellacitywithstandsclimatechange–relatedshocksasmeasuredby,forexample,thetotallossinoutputsufferedandlossofpopulationexperienced.Resiliencehastwocomponents:thesizeoftheinitiallossesassociatedwithashockandhowquicklyacityreturnstoitspreshockgrowthpathsofoutputandpopulation.Aresilientcityisoneinwhichtheinitialimpactsofaclimatechange–relatedshockaresmallandrecoveryisquick.••Inclusivenessisbroadlyconsideredintermsofboth(1)abilityandopportunityand(2)outcomes.Inclusionisdefinedastheabilityandopportunityforallwhoresideinacitytofullyparticipateinmarkets,services,andspaces(includingpolitical,physical,cultural,andsocial),therebyenablingthemtoleadtheirliveswithdignity(WorldBank2013).Consistentwiththatdefinition,thisreportvariouslydiscusseshowcitiesdiffergloballyintheaccesstheyprovidetobasicurbanservices,financialservices,digitaltechnologies,andlabormarketopportunities.Italsoshinesaspotlightonmultidimensionalexclusionandtouchesonissuesofvoice.Asforoutcomes,thereportanalyzes,amongotherthings,howcitiesvarygloballyintermsoftheirratesofpovertyandlevelsofincomeinequality;thelevelsofsocioeconomicmobilitythattheyaffordtheirresidents,especiallynewmigrants;gender-differentiatedpatternsofpopulationdisplacementfollowingclimate-relatednaturaldisasters;andthedifferentialimpactsofexposuretoextremeurbanheatonsegmentsofacity’sworkforce,includinginformalversusformal,femaleversusmale,andolderversusyoungerworkers.Despitethisbroadcoverage,however,thereportissilentoncertainimportantdimensionsofinclusion.Forexample,becauseofthelackofadequatedata,thereportdoesnotdiscusstheimpactsofclimatechangeoncitydwellerswhoarelivingwithdisabilitiesorwhobelongtoaracialorethnicminority.Althoughneitheracity’sprosperitynoritslivabilityisincludedamongthethreedimen-sionsjustdescribed,bothcriticallydependonthegreenness,resilience,andinclusivenessofacity’sdevelopment.Acitythat,forexample,hascleanerair,moregreenspace,betteraccessforitsresidentstobothservicesandmarkets,andanabilitytobouncebackmorequicklyfromclimatechange–relatedshocksisalsolikelytobeacitythatisbothmoreprosperousandlivable.BoxI.1continuedTheprimaryaudienceforthisreportishigh-levelpolicymakersatbothnationalandcitylevelswhosedecisionscanaffectthegreenness,resilience,andinclusivenessofurbandevelopment.Thereportisalsointendedforthosewhoadvisethesepolicymakers,aswellasfortechni-calstaffacrossrelevantgovernmentministriesandplanningagencies.Withthataudienceinmind,thisreportbeginsinpart1byintertwininganoriginaldescriptiveanalysisofawidevarietyofdatasourceswithareviewofthesecondaryliterature,therebytakingstockofhowgreen,howresilient,andhowinclusivecitiesgloballyaretoday(chapter1).Itthenpresentsaglobaltypologyofcitiesthatidentifiestheseverityofthegreenness,resilience,andinclusive-nesschallengestheyfaceandhowthosechallengesintersectwithfutureprojectedclimatechange–relatedhazards(chapter2).Indoingso,thereporthighlightshowthestressesthaturbaneconomistshavetraditionallyemphasizedasbeingassociatedwithurbangrowth—and55Introductionwhicharisefromthepopulationpressuresoncities’landandpropertymarkets,suppliesofbasicservices,localinfrastructure,andenvironment—interactwithclimatechange–relatedstressorstodeterminehowgreen,howresilient,andhowinclusiveacity’sdevelopmentis.Part2ofthereportthenpresentstheresultsoforiginalempiricalwork,againintertwinedwithinsightsfromthesecondaryliterature.Itexaminesthetwo-wayinterplaybetweenclimatechangeandcitiesandhowthatinterplayaffectsdevelopmentoutcomes(chapters3and4).Finally,part3buildsontheempiricalinsightsofparts1and2toprovidepolicymakerswithacompassforpoliciesthatcanhelpenhancethegreenness,resilience,andinclusivenessofcitiesinaworldconfrontedbyclimatechange(chapter5).Indoingso,thereportintroducesasequencedsetoffivebroadtypesofpolicyinstrument—information,incentives,insurance,integration,andinvestments(inshort,thefiveI’s)—thatpolicymakersatthenationalandlocallevelshaveavailabletothem(boxI.2).Anditprovidesguidancetohelpinformpolicymakers’selectionofvariouscombinationsofthoseinstruments.Inadditiontothefivemainchapters,thereportincludesaspotlightonsocialexclusionandexposuretoairpollutioninPeruviancities.Thatspotlightshowcasesanovelapproachforestimatingmultidimensionalexclusion.ThefiveI’s:Information,incentives,insurance,integration,andinvestmentsThisreportdistinguishesfivebroadsetsofinstrumentsthatpolicymakerscandrawoninseekingtoimprovethegreenness,resilience,andinclusivenessofacity’sdevelopmentinaworldconfrontedbyclimatechange.••Information.Policiesandmeasurestoimprovethetimelyprovisionofcredibleinfor-mationthathelpspeople,businesses,andlocalgovernmentsbetterunderstandclimatechange–relatedrisksbothacrossandwithincities,and,indoingso,helpspromotebothmitigationandadaptation.••Incentives.Policyinstrumentsthatprovideincentivesforindividualsandbusinessestointernalizenegativeenvironmentalexternalities,aswellasinstitutionalandothertypesofreformthatprovidegovernmentofficialswithincentivestoworkbettertogethertoaddressgreen,resilient,andinclusivedevelopmentchallenges.••Insurance.Policiesandreformsthathelppeople,businesses,andgovernmentseithertoinsurethroughthemarketortoself-insureagainstlossesassociatedwithclimatechangeandotherenvironmentalshocksandstresses.••Integration.Policyinterventionsandreformsthatpromotemorecompactcitiesandbetterintegrationofcities,bothwitheachotherandwithruralareas,asameansoffacilitatingadaptationthroughbothmigrationandtrade.••Investments.Investmentsbynationalandlocalgovernmentsingreen,resilient,andinclusiveurbaninfrastructure,includingnature-basedsolutions,aswellasmeasurestopromotethecrowding-inofprivatesectorfinanceforsuchinvestments.Chapter5ofthisreportdiscussesthesefivesetsofinstrumentsinmoredetail,alongwithwhowieldstheseinstruments(nationalorlocalgovernments)andhowpolicychoicesbasedontheirusecanbeprioritizedandsequencedforeffectiveimplementation.BoxI.256THRIVINGAframeworkforassessingthegreenness,resilience,andinclusivenessofacity’sdevelopmentInspiredbyinsightsfromurbaneconomictheoryandtheenvironmentalandclimatechangeliterature,thisreportviewsthegreenness,resilience,andinclusivenessofacity’sdevelop-mentasjointlydeterminedbytheinteractionoftwoopposingsetsofforces(figureI.1).7Ontheonehandarethepositiveagglomerationeconomiestowhichacity’ssizeanddensitygiverise.8Theseagglomerationeconomieshelpgenerateprosperity,therebyfacilitatingresil-ience.Theyalsopromoteinclusivenessbycontributingtopovertyreductionandloweringtheaveragecostsofprovidingthelarge-scaleinfrastructurethatunderpinsnotonlymanyimportantbasicservicessuchaswater,sewerage,andelectricity,butalsomorecomplexservicessuchasadvancedmedicalservicesandmoderndigitalnetworks.Thegreaterprosper-ityassociatedwithagglomerationeconomies,however,alsoleadstohigherconsumptionoffood—forexample,beefandothermeats,aswellasdairyproducts—andgoodswhoseproduc-tionisintensiveingreenhousegasemissions.9And,becauseagglomerationeconomieshelpcitiesactasmagnetsfortalent,theycancontributetohigherincomeinequalityincities.Ontheotherhand,actingagainsttheseagglomerationeconomies,citiessufferfromthenegative,interactingstressesthatarisefromtheirowngrowthandfromclimatechange.Ifnotmanagedproperly,thesestressescanbothdampentheprosperitygainsthatarisefromagglomerationeconomiesandunderminethegreenness,resilience,andinclusivenessofurbandevelopment.Inturn,the“fiveI”policyinstrumentsaffecttheseoutcomesbyinfluencingthestrengthofboththeagglomerationeconomiesandtheinteractingurbanandclimatechange–relatedstresses.Thestressesthatarisefromacity’sowngrowth—congestionforces,crowdingeffects,anddiseconomiesofagglomeration—stemfromthepressureitspopulationexertsonitslandandpropertymarkets,localinfrastructure,basicservices,andenvironment.Therefore,ifacitymakesinsufficientinvestmentininfrastructureandbasicurbanservicesasitgrows,theirqualityandpercapitaavailabilitytendtodeteriorate.Iflandandpropertymarketsfailtorespondflexiblytoacity’sgrowth,housingbecomesincreasinglyunafford-able,givingrisetoovercrowdingandslums,whichcanalsofacilitatethemorerapidspreadofinfectiousdiseasessuchasCOVID-19.And,ifpressureontheenvi-ronmentisnotmanagedappropriately,airqualitycandeteriorate,naturalhabitatscancomeunderthreat,andtheconflictwithruralareasoverwatersourcesmayworsen.Climatechangebothinteractswithandexacerbatesthesestressesinavarietyofways.Forexample,theriseinsealevelreducesthesupplyoflandavailablefordevelopmentinlow-elevationcoastalcities,therebyaddingtothepressureonlandandpropertymarkets(Balboni2021).Droughtsinareassupplyingacity’swaterexacerbatepressuresonitswatersupplyFigureI.1ReportframeworkandstructureStressesinducedbyurbangrowthPART2:Chapters3–4PART3:Chapter5PART1:Chapters1–2PolicyinstrumentsInformationIncentivesInsuranceIntegrationInvestmentsGreenResilientInclusiveDevelopmentpathwayClimatechangeAgglomerationeconomiesSource:WorldBankstaff.57system(Zaverietal.2021).Andrisingtemperaturescompoundtheso-calledurbanheatislandeffect(Deuskar2022).Atthesametime,thestressesthatarisefromacity’sowngrowthinfluencebothacity’scontributiontoclimatechangeandthecapacitiesofitsresidentstoadapttosuchchange.Forexample,sprawlandtrafficcongestionresultingfromfailuresinlandandpropertymarkets,shortcomingsinplanning,andinadequateprovisionofhigh-qualitypublictransportationincreasethenumberofvehiclemilestraveledandthusgreenhousegasemissions.Meanwhile,thewidespreadexistenceofpoorlyconstructedslums,frequentlyinmorehazard-proneareas,increasestheexposureoftheurbanpoorandvulnerabletotheimpactsofclimatechange.Climatechange–inducedmigrationintocitiesalsoaddstothestressesthatcitiesexperiencebutdoesnotnecessarilycontributetoagglomerationeconomies.Althoughmorespeculative,climatechangemayunderminethestrengthofagglomerationeconomies.Forexample,morefrequentextremeweathercouldmakeitlesslikelythatpeoplecanmingle,therebyreducingthelikelihoodofknowledgespillovers,oritcouldinterruptdenselocalsupplychains.Beyonditseffectsonagglomerationeconomiesandotherurbanstresses,climatechangehaspotentialrepercussionsforthesocialoptimalityofacity’slocation.AlthoughnotshowninfigureI.1,whereurbandevelopmentoriginallyoccurswithinacountryislargelydeterminedby“locationalfundamentals,”which,inturn,dependonnature(Brakman,Garretsen,andvanMarrewijk2012).Historically,urbandevelopmenthasbeenmorelikelytooccurwherenatureconfersanadvantageintermsofeitherproduction(suchasanabundanceofnaturalresources),trade(suchastheexistenceofanaturalharbor),oramenities(suchasastableandtemperateclimate).Climatechangeis,however,beginningtochangetherelativeextenttowhichnaturefavorsdifferentlocations.Nevertheless,becauseurbandevelopmenthasastronginertia,ratherthanshiftingtolocationslessaffected,orpossiblyevenfavored,byclimatechange,thatdevelopmentcontinuesinthesameplaces.NotonlydoestheframeworkinfigureI.1helpguidethisreport’sanalysisofthegreenness,resilience,andinclusivenessofacity’sdevelopmentandthepolicyoptionsforachievingbetteroutcomes,butitalsoprovidesthereport’sthree-partstructure.Thus,part1describeswherecitiesgloballystandintermsoftheircurrentlevelsofgreenness,resilience,andinclusiveness.Part2analyzesthetwo-wayinterplaybetweenclimatechangeandcities.Andpart3discussesthe“fiveI”policyinstruments,therebyprovidingpolicymakerswithguidanceonhowtoleverageandapplythem.IntroductionNotes1.Allurbanpopulationdatacitedherearedrawnfromthefollowingdatabase:UnitedNations,DepartmentofEconomicandSocialAffairs,WorldUrbanizationProspects2018(revision),https://population.un.org/wup/.2.Thesurfacetemperatureisaveragedacrosslandandwater.Thepreindustrialperiodisdefinedas1880–1900.DataarefromLindseyandDahlman(2021).3.ThenotableexceptionsincludeEstebanRossi-HansbergoftheUniversityofChicago,MatthewKahnoftheUniversityofSouthernCalifornia,andStephaneHallegatteoftheWorldBank,whoareamongthisreport’sexperttechnicaladvisers.4.Thisis,however,beginningtochangeasevidenced,forexample,bytheconferenceprogramatthe2022NorthAmericanMeetingoftheUrbanEconomicsAssociation,whichincludedtwodedicatedsessionsonclimatechange(seehttps://urbaneconomics.org/meetings/uea2022/program.html).5.Since1900,theaverageabundanceofnativeanimalandplantspecieshasdeclinedbyatleast20percentglobally,andabout1millionmorespeciesarenowthreatenedwithextinction(IPBES2019).6.EvenbeforetheonsetoftheCOVID-19pandemic,progressintheglobalfightagainstextremepovertyhadslowedfollowingthe2008globalfinancialcrisis,andmanycitiesinlow-andmiddle-incomecountrieshadhighinequality(WorldBank2021).7.TheurbanpartofthisframeworkbuildsontheframeworksofEllisandRoberts(2016)andRoberts,GilSander,andTiwari(2019).8.TheconceptofagglomerationeconomiesoriginatedintheworkofMarshall(1890).Agglomerationeconomiesarisefromvarioussources,includingthebettermatchingofworkerswithjobsthatoccursincitiesthaninruralareas,thewidervarietyofintermediateinputsintofinalproductionavailablefromlocalsuppliersinlargeranddensercities,andthespilloversofknowledgethatareacharacteristicfeatureoflargeurbanagglomerations(DurantonandPuga2004;Marshall1890).Furtheragglomerationbenefitsarisefromthesuperiormarketaccessthatlargercitiestendtoenjoy,whichmakesiteasierforbusinessestocovertheirstart-upcosts,thusstimulatingincreasesinprofitsandproductivity(Fujita,Krugman,andVenables1999;Krugman1991a,1991b).9.Theuseofcows,pigs,andotheranimalsforfood,aswellasfortheproductionoflivestockfeed,accountsfor57percentofallfoodproduction–relatedgreenhousegasemissionsglobally,comparedwith29percentfromthecultivationofplant-basedfoods(Xuetal.2021).ReferencesBalboni,C.2021.“InHarm’sWay:InfrastructureInvestmentsandthePersistenceofCoastalCities.”DepartmentofEconomics,MassachusettsInstituteofTechnology,Cambridge,MA.https://economics.mit.edu/sites/default/files/publications/Catastrophe_Risk_and_Settlement_Location.pdf.THRIVING58Benton,G.1970.“CarbonDioxideandItsRoleinClimateChange.”ProceedingsoftheNationalAcademyofSciences67:898–91.Brakman,S.,H.Garretsen,andC.vanMarrewijk.2012.TheNewIntroductiontoGeographicalEconomics,2ndedition.Cambridge,UK:CambridgeUniversityPress.Deuskar,C.2022.“BeatingtheHeat:MeasuringandMitigatingExtremeHeatinEastAsianCities.”Unpublishedmanuscript,WorldBank,Washington,DC.Duranton,G.,andD.Puga.2004.“Micro-FoundationsofUrbanAgglomerationEconomies.”InHandbookofRegionalandUrbanEconomics,Vol.4,CitiesandGeography,editedbyJ.V.HendersonandJ.-F.Thisse,2063–117.Amsterdam:Elsevier.Ellis,P.,andM.Roberts.2016.LeveragingUrbanizationinSouthAsia:ManagingSpatialTransformationforProsperityandLivability.Washington,DC:WorldBank.Fujita,M.,P.Krugman,andA.J.Venables.1999.TheSpatialEconomy:Cities,Regions,andInternationalTrade.Cambridge,MA:MITPress.IPBES(IntergovernmentalSciencePolicyPlatformonBiodiversityandEcosystemServices).2019.GlobalAssessmentReportonBiodiversityandEcosystemServicesoftheIntergovernmentalSciencePolicyPlatformonBiodiversityandEcosystemServices.Bonn:IPBESSecretariat.https://doi.org/10.5281/zenodo.3831673.Kahn,M.2010.“Climatopolis:HowWillClimateChangeImpactUrbanitesandTheirCities?”VOX-EUBlog,September11,2010.https://cepr.org/voxeu/columns/climatopolis-how-will-climate-change-impact-urbanites-and-their-cities.Krugman,P.1991a.GeographyandTrade.Cambridge,MA:MITPress.Krugman,P.1991b.“IncreasingReturnsandEconomicGeography.”JournalofPoliticalEconomy99(3):483–99.Lindsey,R.,andL.Dahlman.2021.“ClimateChange:GlobalTemperature.”Climate.gov,March15,2021.NationalOceanicandAtmosphericAdministration,Washington,DC.Madden,R.A.,andV.Ramanathan.1980.“DetectingClimateChangeDuetoIncreasingCarbon-Dioxide.”Science209:763–68.Marshall,A.1890.PrinciplesofEconomics.London:Macmillan.Roberts,M.,F.GilSander,andS.Tiwari,eds.2019.TimetoACT:RealizingIndonesia’sUrbanPotential.Washington,DC:WorldBank.WorldBank.2013.InclusionMatters:TheFoundationforSharedProsperity.NewFrontiersofSocialPolicy.Washington,DC:WorldBank.WorldBank.2021.“FromCOVID-19CrisisResponsetoResilientRecovery:SavingLivesandLivelihoodswhileSupportingGreen,Resilient,InclusiveDevelopment.”PaperpresentedtoDevelopmentCommitteeat2021WorldBank-IMFSpringMeetings,WorldBank,Washington,DC.Xu,X.,P.Sharma,S.Shu,T.Lin,P.Ciais,F.N.Tubiello,P.Smith,N.Campbell,andA.K.Jain.2021.“GlobalGreenhouseGasEmissionsfromAnimal-BasedFoodsAreTwiceThoseofPlant-BasedFoods.”NatureFood2:724–32.Zaveri,E.,J.Russ,A.Khan,R.Damania,E.Borgomeo,andA.Jagerskog.2021.EbbandFlowVol.1:Water,MigrationandDevelopment.Washington,DC:WorldBank.59IntroductionWhereAreWeNow?PART163TheStylizedRelationshipsCHAPTER1[W]eneedtostartfocusingoursupport,bothfinancialandtechnical,onthegreen,resilient,andinclusivetransformationsthatwillhelpeconomies...braceagainstclimatechange....AxelvanTrotsenburg,ManagingDirectorofOperations,WorldBankGroup,April16,2021••Asclimatechangehasevolvedglobally,manycitieshavebecomesubjecttomorefrequentandmoreintenseextremeweatherevents,mostnotably,extremeheatanddryevents.Bycontrast,extremecoldeventshavebecomelessfrequent.••Confrontedwiththischangingclimate,citiesvarywidelyintermsofhowgreen,howresilient,andhowinclusivetheyare.Nevertheless,somegeneralpatternsemerge,suchastherelationshipofmanyindicatorsofgreenness,resilience,andinclusivenesstobothacity’ssizeanditslevelofdevelopment.••Largercitiestendtobelessgreenintermsoftheircarbondioxideemissions,levelsofairquality,andaveragevegetationlevelsbut,inmanyrespects,aremoreinclusive.Theyhavelowerpovertyrates,providebetteraccesstobasicservices,andhavebetteraverageoutcomesonmanyhealthindicators.••More-developedcitiesalsotendtobelessgreenintermsofcarbondioxideemissions;however,beyondacertainincomelevel,developmentcorrelateswithbothbetterairqualityandahigheraveragelevelofvegetation.Moredevelopedcitiesalsohaveecono-miesthataremoreresilienttounusualweatherevents,andtheytendtobemoreinclu-sivewhenitcomestoaccesstobasicservicesandhealthoutcomes.Theyalsosufferfromlesspovertyandmayhavelowerinterhouseholdincomeinequality.••Thegreenness,resilience,andinclusivenessofcitiesarelinked.Acity’slevelofinclusivenessis,forexample,akeydeterminantofitsresiliencetoboththeshort-andlonger-runstressesassociatedwithclimatechange.MAINFINDINGS64THRIVINGIntroductionItawell-wornbutneverthelessempiricallywell-supportedclichéthatcitiesactas“enginesofgrowth”atboththelocalandnationallevels.Thisexpressionstemsinpartfromthebetterjobsfoundincitiesthaninruralareasandfromtheproductivityandotherwelfare-enhancingbenefitsassociatedwithurbansizeanddensity(DurantonandPuga2020;WorldBank2009).Atthesametime,asthisreport’soverarchinganalyticalframeworkmakesclear,acity’sgrowthcanalsogiverisetoawidevarietyofstresses.Thesestresses—knownascongestionforces,crowdingeffects,anddiseconomiesofagglomeration—arisefromthepressureofacity’spop-ulationonitssuppliesofland,housing,andbasicservices;stockofinfrastructure;andenvi-ronment.1Ifnotwellmanaged,thesestressescanthreatennotonlyacity’sproductivityandgrowthbutalsothequality—and,therefore,sustainability—ofthatgrowth,makingthecitylessgreen,lessresilient,andlessinclusive.Theresultsofthesestressesareevidentintheslumsandsprawl,lackofaccesstobasicservices,gridlockedstreets,chokingairpollution,andcon-taminationofwatersourcesthatcharacterizemanycitiesinlow-andmiddle-incomecoun-triestoday.Inthinkingaboutthesestresses,economistshavetraditionallyassumedthatacity’sclimateisfixedorchangesonlyveryslowly.2Climatechange,however,hasrenderedthisassumptionincreasinglyuntenable.Ithasemergedasanadditionalandgrowingsourceofstressonmanycities,bothaddingtoandinteractingwithmanyofthetraditionallyemphasizedsourcesofstress,therebyfurtherinfluencingthequalityofacity’sgrowth.Ithasalsobecomeclearlyrec-ognizedthatthenatureofacity’sspatialdevelopmenthasthepotentialtoaffectnotonlytheglobalclimatethroughitscontributiontogreenhousegas(GHG)emissionsbutalsoitsownlocalclimatethroughtheurbanheatislandeffect.Inviewoftheseissues,thischapterhasthefollowingmainobjectives:••Totakestockonabroadlevelofhowgreen,howresilient,andhowinclusivecitiesaretodayconsidering,amongotherthings,(1)theircontributiontoclimatechange;(2)theireconomicresiliencetovarioustypesofextremeweatherevents,thefrequencyandintensityofwhichareevolvingwithclimatechange;and(3)thestrengthoftheotherstressestheyfacebecauseofthepressuresofurbangrowthonacity’ssuppliesofland,housing,andbasicservices;stockofinfrastructure;andenvironment.••Tolaythegroundworkforthedevelopmentoftheglobaltypologyofcitiesinchapter2thathighlightstheseverityofurban-andclimatechange–relatedchallengesfacedbydifferenttypesofcities,aswellastheirpossibleintersections.Toachievetheseaims,thischapterfirstdocumentshowdifferenttypesofweatheranomalies—hot,cold,wet,anddryevents,alongwithtropicalcyclones(suchashurricanesintheAtlantic)—havebeenevolvingacrosscitiesgloballyinbothfrequencyandintensity.Itthenpresentsstylizedempiricalrelationshipsbetweenkeyindicatorsofhowgreen,howresilient,andhowinclusivecitiesareandtheirbasiccharacteristicswithanemphasisonacity’ssizeandlevelofdevelopment.Thechapteralsopresentsnewempiricalevidenceontheshort-runresilienceofurbaneconomiestoweatheranomaliesandonhowthelevelofthisresiliencevarieswithacity’ssize,levelofdevelopment,andbaselineclimate.3Inconsideringhowgreen,howresilient,andhowinclusivecitiesaretoday,itisimportanttorecognizetheinterdepen-denciesamongthesethreebroadoutcomes.Forexample,asociety’spovertyrate,akeyindi-catorofitsinclusiveness,isanimportantdeterminantofitslevelofresilience.4Meanwhile,theaveragelevelofvegetationinacity,oneindicatorofitsgreenness,affectsthestrengthofacity’surbanheatislandeffect.Highertemperaturesmay,inturn,influenceacity’slevelofspousalviolence.Anyincreaseinthatviolencefurtherunderminesacity’sinclusiveness.TheStylizedRelationships65Finally,tothegreatestextentpossible,thechapterreliesforitsinsightsonaglobalsampleofmorethan10,000consistentlydefinedcities,5synthesizingboththeoriginalempiricalanalysisofthissampleandtheanalysisofsimilarlydefinedurbanareasbyotherauthors(box1.1).Thissynthesisissupplementedbyinsightsfromsecondaryliteratureandanalysisofotherdatasources,someofwhichrelyonotherdefinitionsofurbanareaswiththecaveatsofcomparabil-itythatthisrelianceimplies.Comparingappleswithapples:HowthisreportdefinescitiesHowshouldacitybedefined?Thisapparentlyfundamentalquestionhas,itturnsout,noeasyanswer.Thus,countriesaroundtheworldvarywidelyinhowtheydefinecities,usingbothdifferentnumbersanddifferentcombinationsofcriteria.Forexample,Senegaldefinescitiesas“agglomerationsof10,000inhabitantsormore,”andAlbaniadefinesthemas“townsandotherindustrialcenterswith400inhabitantsormore.”aMoregen-erally,althoughsomecountriesdefinecitiesbasedonlyonpopulation,othersalsodefinethemusingcriteriasuchaspopulationdensity,thepresenceofcertaintypesofinfra-structureandbasicservices,andtheshareoftheirworkforcethatisemployedoutsideofagriculture.Initsofficialstatisticaldefinition,Indonesiaevenincludesthepresenceofmassageparlorsandcinemasassomeofthefactorsthathelpclassifyaplaceasurban.Alargegroupofcountrieshasnoexplicitlystatedcriteriabywhichtheydefinecities.Rather,theysimplyeitherlisttheircitiesbynameordesignateadministrativeunitsthatconstitutecities(Robertsetal.2017).Thiswidevariationindefinitionsposeshugechallengesforanyglobalanalysisofcities.Toavoidappleswithorangescomparisons,theanalysisreportedherelargelyreliesonagloballyuniformdefinitionofcities—degreeofurbanization—thattheEuropeanCommissiondevisedincollaborationwithseveralotherinternationalorganizations(Dijkstraetal.2021).Thisdefinition,officiallyendorsedinMarch2020bytheUnitedNationsStatisticalCommissionasarecommendedmethodformakinginternationalcomparisonsofurbanareas,identifiesacityasaspatiallycontiguous,denseclusterofpopulationgridcellsinaglobal1squarekilometerpopulationgrid.Thus,acityisacon-tiguoussetofgridcellsforwhicheachcellhasapopulationdensityofatleast1,500peoplepersquarekilometerandtheaggregatepopulationofthesetisatleast50,000.Degreeofurbanizationalsoidentifiesurbanclusters,whichcomprisebothlower-densitysuburbsofcitiesandsmallerurbanareasortowns.Incontrasttocities,theseclustersaredefinedusingapopulationdensitythresholdof300peoplepersquarekilometerandanoverallpopulationthresholdof5,000.Althoughmostofthischapter’sanalysisfocusesoncities,someofitalsoconsidersurbanclusters.OthermajorWorldBankreportshaveusedthedegreeofurbanizationdefinitionofcities(notably,FerreyraandRoberts2018;Lalletal.2021).Thisdefinitionisfastbecomingagloballyacceptedwayofdefiningcities,notonlybytheinternationalpolicycommunitybutalsobyacademicresearcherspublishinganincreasingnumberofglobaldatasetsbasedonthedefinition,someofwhichthisreportuses.Nevertheless,degreeofurbanizationhasitsshortcomings.Forexample,althoughthepopulationdensitythresholdof300peoplepersquarekilometerusedtodesignateurbanBox1.166THRIVINGclustersisreasonableformostcountries,itimpliesimplausiblyhighestimatedurbansharesofthepopulationforcountriesandareassuchasBangladesh,thenortheasternstatesofIndia,andJava(Indonesia)thathaveveryhighaveragepopulationdensities(Bosker,Park,andRoberts2021).Moregenerally,thedegreeofurbanizationisonlyasgoodasthegriddedpopulationdatausedasitsinput,andtheurbanareasdelineatedaresomewhatsensitivetothechoiceofsuchdata(Dijkstraetal.2021;Robertsetal.2017).Althoughseveralalternativemethodsfordelineatingurbanareashaverecentlybeendevelopedandappliedtoindividualcountries(deBellefonetal.2021;Galdo,Li,andRama2021),thesemethodsaremuchmorecomputationallyintensivethanthedegreeofurbanizationandhaveyettobescaledtothegloballevel.Furthermore,toagreaterorlesserextent,globalapplicationsofthesemethodswouldstilldependonthesamegriddedpopulationdatasetsasthedegreeofurbanization,whosequalityreliesfun-damentallyonthequalityofpopulationdataproducedbynationalstatisticalofficesthroughtheirpopulationcensuses.Thisbeingthecase,theglobaldatasetofcitiesusedinthisreportisthebestcurrentlyavailable,especiallyinviewofthenumberofotherglobaldatasetsgeneratedtomatchit.a.“DataSourcesandStatisticalConceptsforEstimatingtheUrbanPopulation”spreadsheet,UnitedNations’WorldUrbanizationProspects,https://population.un.org/wup/Download/.Box1.1continuedAchangingclimateAlthoughweatherandclimatearetemporallydistinctbutphysicallyrelatedconcepts(acity’slong-termclimateconditionisanaggregateoftheshort-termweathereventsitexperiences),climatechangehasincreasinglyaffectedthefrequencyandintensityofabnormalweatherevents(or“anomalies”).Thissectionprovidesanoverviewoftheevolutionofdifferenttypesofweatheranomaly—extremeheat,cold,wet,anddryevents,aswellasstormevents—overthelastsixdecadesforthisreport’sconsistentlydefinedglobalsampleofcities.6Aweatheranomalyisdefinedrelativetoacity’sownbaselineclimateconsideringthehistoricvariabilityofitsweather(box1.2).Extremeheateventshaveincreasedinfrequencysincethe1970s,whereastheoppositeistrueofextremecoldeventsGlobally,theaveragenumberofmonthsayearacity’stemperaturewasextremelyhotrelativetoitsownhistoricalexperienceincreasedfrom0.06inthe1970sto1.35overtheperiod2010–20(figure1.1,panela).Putdifferently,theaveragewentfromjustunder2daysperyeartomorethan41daysperyear—astaggering21-foldincreaseinjustfourdecades.Extremeheateventsalsoincreasedinintensityoverthisperiod.Inthe1970s,theaverageextremeheateventrecordedatemperaturealmost2.3standarddeviationsaboveacity’sownlong-runaverage.Fastforwardto2010–20,andtheintensityoftheaverageextremeheateventhadincreasedtojustover2.5standarddeviations(figure1.1,panelb).Asmap1A.1inannex1Aillustrates,extremeheateventsareparticularlyprevalentinsouthernIndia,SoutheastAsia,upperpartsoftheArabRepublicofEgypt,andpartsofCentralAmericaandCentralAfrica.Whentheydooccur,theseeventsareparticularlyintenseinSouthAsia(seemap1A.2).TheStylizedRelationships67Whatisaweatheranomaly,andhowdoesthischapterdefineit?InJune2021,countriesthroughouteasternEurope,includingBelarus,Estonia,andHungary,occupiedtheheadlinesastheysetall-timehightemperaturerecordsforthemonth.InHungary,thetemperaturereached40°C,whereasthecountry’saverageistyp-icallybetween12.8°Cand23.9°C,dependingontheregion.Although40°Cishotbymoststandards,suchatemperaturewouldbenothingoutoftheordinaryfortheresidentsofKuwaitCity,whereJunedaytimetemperaturesaretypically45°C–46°C.Intheclimateliterature,aweatheranomalyisoftendefinedastheabsolutedeviationofameasurableunitofweather,saydegreesCelsiusormillimetersofrain,overacertainperiodfromalocation’slong-runaverage.aForacross-cityanalysisonaglobalscale,however,suchadefinitionisinadequatetocharacterizeunusualweatherbecauseitdoesnotfullyconsiderthediverseclimatologicalconditionstowhichpeopleindifferentplaceshaveadapted.Givenitsglobalfocus,thischapter’sanalysisdefinesforanygivenmonthacity’sweatheranomalyinstandarddeviationsfromitsownlong-runaverageforthatsamemonth.Denotingxwicmtasthevalueoftheweathervariable(suchastemperatureorprecipitation)incityiincountrycinmonthmofyeart,aweatheranomaly,Anomalywicmt,isdefinedasAnomalywicmt=xwicmt−x_wicmσwicmwherewreferstoeithertemperatureorprecipitation,andx_wicmandσwicmtothecity-specificlong-runaverageandthestandarddeviationformonthm,respectively.Theselong-runvariablesarecalculatedfortheperiodJanuary1958–March2012usinghighspatialreso-lutionmonthlydatafromTerraClimate.bThismeasure,bydefinition,capturesunusualtemperaturesandprecipitationinbothpositive(hotterorwetterthanusual)andnegative(colderordrierthanusual)directionsrelativetoacity’sownclimateoverthelongrun(averageandvariability).Fortempera-ture,forexample,themeasureisequalto+1ifacity’stemperatureinaparticularmonthis1standarddeviationaboveitsownlong-runaverageforthatsamemonth.Whenthemeasuretakesonanabsolutevalueof2ormore,aweatheranomalyisconsideredextreme.Ininterpretingthevaluesofanomalies,however,onemustkeepinmindthat—becausetheimplicationsofananomalyofagivensize(say,–2)varyacrosscitiesaccordingtotheirunderlyingclimates—largevaluesdonotnecessarilytranslateintosevereweathereventssuchasheatorcoldwaves,floods,ordrought.ForKuwaitCity,anygivenJunerecording2standarddeviationsbelowtheaveragereflectsanunusuallycoolmonth,butitmaystillbewarminabsoluteterms.Bycontrast,foracitywithalong-runaveragetem-peratureof5°CinDecember,suchanextremenegativetemperatureanomalywillreflectanunusuallycoldwinterinbothrelativeandabsoluteterms.Source:BasedonParkandRoberts2023.a.NationalOceanicandAtmosphericAdministration,NationalWeatherServiceGlossary,https://forecast.weather.gov/glossary.php?word=ANOMALY.b.ClimatologyLab,TerraClimate,https://www.climatologylab.org/terraclimate.html.Box1.268THRIVINGTheincreasingfrequencyandintensityofextremeheatanomaliesareconsistentwithglobalwarming,wherebyasmallincreaseinaveragetemperatureleadstoasignificantlyincreasednumberofdaysthatareunusuallyhotbyacity’sownstandards(IPCC2021).Theflipsideisthat,asacity’saveragetemperatureincreases,thenumberofunusuallycolddaysthatit,againbyitsownstandards,experiencesdecreases.Itisnotsurprising,then,thattheaveragenumberofmonthsperyearforwhichacity’stemperaturewasextremelycoldbyitsownlong-runstandardsfellfrom0.44inthe1970sto0.10overtheperiod2010–20(figure1.2,panela).Putdifferently,theaveragewentfromjustover13daysperyearinthe1970stoslightlymorethan3daysperyearduring2010–20.However,althoughthefrequencyofextremecoldanomalieshasdropped,theintensityofthoseanomalies,whentheydooccur,hasincreasedsincethe1980s(figure1.2,panelb).Asmap1A.3inannex1Ashows,isolatedgeographic“hotspots”ofmorefrequentextremecoldweather,relativetoacity’sownbaselineclimate,canbefound,forexample,aroundthestateofParaibainBrazil,inEastAfrica,andininlandChina.Meanwhile,whenextremecoldeventsdooccur,theytendtobeveryintense,especiallyinChinaandIndia(map1A.4,annex1A).Thetrendintemperatureanomalies,whichbecauseofclimatechangewillcontinueinthecomingdecades,impliesthatheatwaveswillbecomeanincreasinglycommonproblemformanycities,especiallycitieswithtropicalclimatesinwhichhighbaselinetemperaturesFigure1.1Evolutionofthefrequencyandintensityofextremeheateventsforcitiesglobally,1958–69to2010–20Sources:WorldBankcalculationsbasedonClimatologyLab,TerraClimate(https://www.climatologylab.org/terraclimate.html);EuropeanCommission,GlobalHumanSettlement(GHS)UrbanCentreDatabaseR2019(https://ghsl.jrc.ec.europa.eu/ghs_stat_ucdb2015mt_r2019a.php).Note:Anextremehotmonthisoneinwhichacity’stemperatureforthatmonthisatleast2standarddeviationshigherthanthemonth’scity-specifichistoricalnorm,ascalculatedovertheperiodJanuary1958–March2012.Frequencyiscalculatedasthenumberofextremehotmonthsperyear.Intensityiscalculatedastheaveragesizeoftheanomalyvariable(asdefinedinbox1.2)duringconsecutiveextremehotmonths.PanelapresentstheaveragefrequencyandpanelbtheaverageintensityacrosstheglobalsampleofurbancentersthatexperiencedanextremehotmonthovertheperiodJanuary1958–December2020.00.51.01.51958–691970s1980s1990s2000s2010–202010–20Averageno.ofmonthsperyear2.02.22.42.61958–691970s1980s1990s2000sAveragesizeofanomalies(no.ofstandarddeviations)a.Frequencyb.IntensityTheStylizedRelationships69combinewithhighlevelsofhumiditytocreatethermaldiscomfort.Withoutadequateadap-tation,suchheatwaveswillhavenegativeconsequencesforawiderangeofurbanoutcomes,includinghealth,education,andlearningoutcomes,aswellasworkerproductivity—seeDeuskar(2022)forareviewoftheevidence.Evidencealsoshowsthatincreasedexposuretoveryhottemperaturessignificantlyincreasesurbancrimeandviolence,particularlyinpoorerneigh-borhoodswitholderhousingstocksandlittleair-conditioning(see,amongothers,HeilmannandKahn2019).Consistentwiththesefindings,theglobalannualaverageheat-relatedexcessdeathratioincreasedby0.21percentagepointsbetween2000–2003and2016–19,reaching0.91percentofallglobalexcessdeathsinthelatterperiod(Zhaoetal.2021).Thisfigureisequivalentto489,000excessdeathsperyear,manyconcentratedinlarge,crowded,low-lyingcoastalcitiesineasternandsouthernAsiaandcitiesineasternandwesternEurope.Atthesametime,however,excessdeathsduetocoldfarexceedthoseduetoheat.Indeed,coldaccountedforanannualaverageofalmost4.6millionexcessdeathsgloballyduring2016–19,morethanninetimesthenumberofheat-relatedexcessdeathsduringthesameperiod.And,becauseofthedecreasingfrequencyofextremecoldevents,thetotalnumberofexcessdeathsattributabletononoptimaltemperatures(thatis,eithertoohotortoocold)hasdeclinedoverthelasttwodecades(Zhaoetal.2021).Thisfindingimpliesthat,despiteincreasingconcern00.10.20.30.40.51958–691970s1980s1990s2000s2010–20Averageno.ofmonthsperyear-2.5-2.4-2.3-2.2-2.1-2.01958–691970s1980s1990s2000s2010–20Averagesizeofanomalies(no.ofstandarddeviations)a.Frequencyb.IntensityFigure1.2Evolutionofthefrequencyandintensityofextremecoldeventsforcitiesglobally,1958–69to2010–20Sources:WorldBankcalculationsbasedonClimatologyLab,TerraClimate(https://www.climatologylab.org/terraclimate.html);EuropeanCommission,GlobalHumanSettlement(GHS)UrbanCentreDatabaseR2019(https://ghsl.jrc.ec.europa.eu/ghs_stat_ucdb2015mt_r2019a.php).Note:Anextremecoldmonthisoneinwhichacity’stemperatureforthatmonthisatleast2standarddeviationsbelowthemonth’scity-specifichistoricalnorm,ascalculatedovertheperiodJanuary1958–March2012.Frequencyiscalculatedasthenumberofextremecoldmonthsperyear.Intensityiscalculatedastheaveragesizeoftheanomalyvariable(asdefinedinbox1.2)duringconsecutiveextremecoldmonths.PanelapresentstheaveragefrequencyandpanelbtheaverageintensityacrosstheglobalsampleofurbancentersthatexperiencedanextremecoldmonthovertheperiodJanuary1958–December2020.70THRIVINGamongurbanpolicymakersabouttheimpactsofextremeheatonhumanhealth,unusuallycoldweatherdoesremain,andwillremainfortheforeseeablefuture,thebiggerkiller.Thus,policiestohelpresidentsadapttocoldspellsremainimportantformanycities.Extremedryand,toalesserextent,extremeweteventshavealsoincreasedinfrequencyThisfindingismostobviouslytrueforextremedryevents.Inthe1970s,theaveragenumberofmonthsperyearthatcitiesexperiencedextremelylowrainfallrelativetotheirownhistori-calexperienceswas0.03,oranaverageof0.91daysperyear.By2010–20,however,thisaveragehadmorethantripledtoalmost0.10months(or3.04days)peryear(figure1.3,panela).TheseextremedryeventsareparticularlycommoninCentralAfrica(seemap1A.5).7Theincreasingfrequencyofextremedryeventsgloballyhas,inturn,contributedtotheincreasingnumbersofcitiesexperiencingnear“dayzero”events,wherebywatersuppliesareonlyweeksordaysfromrunningout.WatersupplysystemstypicallyareplannedtomeetadesignstandardfordroughttomaintainsupplieswithoutanyrestrictiononuseFigure1.3Evolutionofthefrequenciesofextremedryandweteventsforcitiesglobally,1958–69to2010–20Sources:WorldBankcalculationsbasedonClimatologyLab,TerraClimate(https://www.climatologylab.org/terraclimate.html);EuropeanCommission,GlobalHumanSettlement(GHS)UrbanCentreDatabaseR2019(https://ghsl.jrc.ec.europa.eu/ghs_stat_ucdb2015mt_r2019a.php).Note:Anextremedry(wet)monthiswhenacity’sprecipitationforthatmonthisatleast2standarddeviationsbelow(above)themonth’scity-specifichistoricalnorm,ascalculatedovertheperiodJanuary1958–March2012.Frequencyiscalculatedasthenumberofextremedry(wet)monthsperyear.Thegraphspresenttheaverageannualfrequencyacrosstheglobalsampleofurbancentersthatexperiencedanextremedry(wet)monthduringJanuary1958–December2020.00.020.040.060.080.101958–691970s1980s1990s2000s2010–20Averageno.ofmonthsperyear0.10.20.30.40.50.61958–691970s1980s1990s2000s2010–20Averageno.ofmonthsperyeara.Extremedryeventsb.ExtremeweteventsTheStylizedRelationships71(Wattsetal.2012;Zaverietal.2021).Althoughthelengthofthedesigndroughtthaturbanwatersystemscanhandlevaries,experienceworldwidehasshownthatlongdroughtslastingthreeormoreyearsaretypicallymoretaxingforurbanwatersupplysystemsandcanleadtoseverewateruserestrictionssuchascutsorcountdowns(Buurman,Mens,andDahm2017;Zaverietal.2021).Usingthisthree-yearperiodasaruleofthumbtocapturebothweathereventsandmanagementresponses,previousanalysishasestimatedthatcumulativerainfalldeficits,measuredasthecumulativedeviationofrainfallfromlong-runaveragesoverthree-yearperiods,canharmoverallcitygrowthwhenthemagnitudefallsbelow–3.5standarddevi-ations.Inotherwords,whenrainfalldeficitsare,onaverage,atleast1.2standarddeviationsbelowthelong-runaverageineachyearoverthethree-yearperiod,thenegativeimpactoncitygrowthbecomessignificant(Zaverietal.2021).Usingasimilarruleofthumbandthreshold,figure1.4showsthatthesenear–dayzeroeventshavebecomeespeciallycommonincitiesinlow-andlower-middle-incomecountries.Asforextremewetevents,theydeclinedsteadilyinaveragefrequencyfromthe1970supto2000–2009,buttheiroccurrencesubsequentlyreboundedduringtheperiod2010–20(figure1.3,panelb).8Increasinglyfrequentextremerainfallsince2010suggestsanincreasedriskofurbanflooding.Asbox1.3discusses,however,theoccurrenceofsuchfloodingalsocriticallydependsonthecharacteristicsofthebuiltenvironment,includingwhetheracityhasadequatedrainageinfrastructure,andonthequalityofitssolidwastemanagement.010203040<=–3.5<=–4.5<=–5.5Shareofcities(%)01020304050Low-andlower-middle-incomeUpper-middle-incomeHigh-incomeShareofcities(%)a.Deepestthree-plusyearsofwaterdeficitsb.CountryincomegroupFigure1.4Shareofcities,bydeepestthree-plusyearsofwaterdeficitsandcountryincomegroup,thatfaceddayzero–typeevents,1992–2013Sources:WorldBankcalculationsbasedonZaverietal.2021.Weatherdata:K.MatsuuraandC.J.Willmott,TerrestrialAirTemperatureandPrecipitation:MonthlyandAnnualTimeSeries(1900–2017),UniversityofDelaware,2018;urbanwatersourcesdata:NatureConservancyandR.McDonald,“CityWaterMap(Version2.2),”KnowledgeNetworkforBiocomplexity,2016(doi:10.5063/).Note:Panelashowstheshareofallcitiesgloballyinthedatasetthatexperiencedthedeepestwaterdeficitfrom1992to2013,theperiodforwhichdataareavailable.Thedeepestwaterdeficit(measuredontheverticalaxis)isdefinedasacumulativez-scoreofrainfallovereachthree-yearperiodforeachwaterpointinthesampleperiod.Dayzeroeventsaredefinedasthosewhenthecumulativez-scoreofrainfallovereachthree-yearperiodforeachwaterpointfrom1992to2013islessthan−3.5standarddeviations.72THRIVINGWhendoesaweteventbecomeaflood?Accra,thecapitalofGhana,experiencesfloodingeveryrainyseason,butthefloodingcausedbyheavyrainfallonJune3,2015,wasunprecedentedinitsmagnitudeandtragicconsequences.Thecity’smarkets,normallycrowded,cametoastandstillasstormwaterseveralfeetdeepdestroyedtraders’goods.Thewaterfloodedhomesandtrappedmanypeopleintheirofficesorcars.Twenty-fivepeoplediedintheflood.Justwhenthingsseemedasthoughtheycouldnotgetanyworse,anotherdisasterstruck.Inoneofthecity’sbusiestareas,dozenshadgatheredatapetrolstationtoshelterfromtherain.Asthefloodwaterrose,itmixedwithfuelfromaleakingpetrolpump.Aflameignitedthemixture,causingamassiveexplosion.Firefightersstruggledtoreachtheblazethroughthefloodwater.Tragically,theinfernokilled150people.Inthemourningperiodthatfollowed,residentsandofficialsrecognizedthat,despitetheheavyrainfall,theunder-lyingcausesofthedisasterweretheinadequatedrainagesystem,thesolidwastethatblockeddrains,andtheunplannedurbangrowththatobstructednaturaldrainagechannels.TheUSFederalEmergencyManagementAgencydefinesurbanfloodingas“theinunda-tionofpropertyinabuiltenvironment,particularlyinmoredenselypopulatedareas,causedbyrainfallingonincreasedamountsofimpervioussurfacesandoverwhelmingthecapacityofdrainagesystems.”aAsthedefinitionmakesclear,floodingincitiesresultsfromacombinationoffactorsbothnaturalandhumaninorigin.Althoughheavyrainfallmaytriggerflooding,characteristicsofthebuiltenvironment—suchastheextentofimpervioussurfacesandthecapacityofdrainageinfrastructure—alsoplayarole.Urbanfloodingmayoccuranywhere,notjustnearwaterbodies.Aschapter3discusses,suchfloodingoften,butnotalways,disproportionatelyaffectslow-incomeresidents,whoarelikeliertoliveinlow-lyingareaswithinsufficientdrainageinfrastructureandwithoutfloodprotectioninsurance.Sources:BBCNews2015;KarimiandLett2015;Weber2019.a.AscitedinHossainandMeng(2020).Box1.3Globalsea-levelrisealsocontributestoanincreasedriskoffloodingforcoastalcitiesGlobalwarmingleadstoariseinsealevelbycausingseawatertoexpandandtheiceoverlandtomelt.Becauseofclimatechange,risesinsealevelhaveacceleratedinrecentdecades.AccordingtotheUSNationalOceanService,9globaltidalrecordsfrom1900to1990revealanestimatedriseintheglobalmeansealevelof101.6–127.0millimeters.Inthe25yearsfrom1990to2015,however,theglobalsealevelroseby76.2millimeters.Currently,sealevelsarerisingabout0.125millimetersperyear.Andby2100theymayhaverisenanother0.3–2.4meters.Historically,coastallocationshavetendedtoenjoyadvantagesintermsofbothtradeandagriculture,makingthemnaturallocationsforcities(Smith1776).Globally,morethan300millionpeople,or5percentoftheworld’spopulation,liveinlow-elevationcoastalzonesbelow5metersabovesealevel(CIESIN2013).However,asanalyzedinmoredetailinchapter2,risingsealevelsposeagrowingthreatoffloodingforcoastalcitiesglobally.TheStylizedRelationships73Tropicalcycloneshavebecomeincreasinglyfrequentsincethe1970sFinally,intermsofweatheranomalies,tropicalcyclones(called“hurricanes”intheAtlanticOceanandsometimes“typhoons”inthePacific)thataffectcitieshavebecomemorefrequentsincethe1970s(figure1.5).Betweenthe1970sand2010–2020,theaveragenumberoftropicalcyclonesexperiencedbytheglobalsampleofcitiesincreasedmorethanfourfold,from0.16to0.71.10Tropicalcyclonesforminalloceanbasins,andcitiesinthePhilippinesareparticu-larlypronetotheirstrongimpacts.Meanwhile,thehurricanesthatforminthehighlyactiveNorthAtlanticbasinaffectcitiesthroughouttheCaribbean,easternMexico,andtheeasternandsouthernUnitedStates.TropicalcyclonesthatformintheeasternPacificcanaffectcitiesinwesternMexico,whereastyphoonsthatforminthewesternPacificarehighlylikelytohitcitiesinsouthernAsia,China,andJapan.TheAsiannationsalongtheIndiansubcontinentarealsopronetotropicalcyclonesfromtheIndianOcean,andtheBayofBengalisahotspotforcyclonicactivities.Finally,atropicalcyclonefromthesouthwesternIndianOceancanaffectcitiesinMadagascarandothercountriesalongtheeasterncoastofAfrica,andonefromthesoutheasternIndianOceancouldaffectAustraliancities.00.20.40.60.81958–691970s1980s1990s2000s2010–20Averageno.ofcyclonesFigure1.5Evolutionofthefrequencyoftropicalcyclonesforcitiesglobally,1958–69to2010–20Sources:WorldBankcalculationsbasedonClimatologyLab,TerraClimate(https://www.climatologylab.org/terraclimate.html);EuropeanCommission,GlobalHumanSettlement(GHS)UrbanCentreDatabaseR2019(https://ghsl.jrc.ec.europa.eu/ghs_stat_ucdb2015mt_r2019a.php).Note:Foranygivenmonth,acityisclassifiedasbeingaffectedbyatropicalcycloneifithadastormwithsustainedwindsofatleast96milesperhour—thatis,thestormwascategory2orstrongeraccordingtotheSaffir-Simpsonwindscaleandwaswithin200kilometersofthecity’sgeographiccenterinthatmonth.Frequencyiscalculatedasthenumberoftropicalcyclonesacityexperiencedovereachdecade.ThegraphpresentstheaveragefrequencyacrosstheglobalsampleofurbancentersthatexperiencedatropicalcycloneatleastonceduringJanuary1958–December2020.74THRIVINGHowgreenarecitiestoday?Towhatextentarecitiesresponsiblefortheclimatechangethathelpsexplaintheevolutioninweatheranomaliestheyhaveexperiencedinrecentdecades?And,moregenerally,how“green”arecitiestoday,notjustintermsoftheircontributionstoclimatechangebutalsointermsof,forexample,theirlevelsofoutdoorairpollutionandvegetation?Toaddressthesequestions,thissectionanalyzesdataoncarbondioxide(CO2)emissions,emissionsofpar-ticulatematterof2.5micronsorlessindiameter(PM2.5),andaveragevegetationlevels(asdetectedinsatelliteimagery)forthisreport’sconsistentlydefinedglobalsampleofmorethan10,000cities.Indoingso,andlikesubsequentsectionsofthischapter,itfocusesonhowtheseoutcomesrelatetobasiccitycharacteristics,including,mostnotably,acity’ssizeanddevelopmentlevel.Althoughtheyhavenottypicallymadethelinktoclimatechange,urbaneconomistshavelongemphasizedthatpollution,andthereforebothCO2andPM2.5emissions,isamongthekeystressesthatcanarisefromacity’sgrowth.FossilCO2emissionsaccountedfor77percentoftheworld’santhropogenicGHGemissionsin2015.Since2000,theirincreasehasbeenthemainsourceoftheglobalincreaseinGHGemissionsdrivingtheEarth’swarming(Crippaetal.2019)and,withthatincrease,theglobaltrendsintheevolutionofweatheranomalies.Meanwhile,asacity’spopulationincreases,onemightexpectdevelopmentpressuresonopenandgreenspacestointensify,leadingtoadecreaseinvegetation.Acity’slevelofvegetationcanalsoaffectitslocalclimatebyinfluencingthestrengthoftheurbanheatislandeffect(box1.4).Hotinthecity:ThecausesandimpactsoftheurbanheatislandeffectUrbanareasareheatislands—thatis,theyarehotterthanthesurroundingruralareas.Severalfactorscontributetothiseffect.Buildingsandpavedsurfaceshavehighthermalinertia,meaningtheyabsorbheatduringthedayandreleaseitatnight,whichpreventscitiesfromcooling.Buildingsalsotrapheatandobstructtheflowofbreezes.Citiestyp-icallyhavelessvegetationthansurroundingareas,sotheybenefitlessfromtheshadeandevaporativecoolingthatgreeneryprovides.Heatfromhumansources(suchasmotorvehicles,factories,andairconditioners)alsoraisesurbantemperatures.Theintensityoftheurbanheatislandeffectvariesbyseason,timeofday,andlocationwithinacity;butitcanresultinlandsurfacetemperaturesthataremorethan10°Chigherthantheequiva-lentrurallandsurfacetemperatures.Exposuretoextremeheatcanresultinhighermortalityandmorbidity,lowereconomicproductivity,poorereducationaloutcomes,andhigherratesofcrimeandviolence,includingsexualviolence.Citiescanmitigateurbanheatbyaddingvegetation,usingreflectivematerialsonstreetsandbuildings,andorientingbuildingstomaximizeshadeandbreeze.Theycanalsoadapttoextremeheateventsbypreparingheatactionplans,raisingawarenessaboutthehealthimpactsofexposuretoextremeheat,trainingpublichealthworkerstotreatheat-relatedillnesses,anderectingcoolingstations.Source:AdaptedfromDeuskar2022.Box1.4TheStylizedRelationships75CitiesassourcesofclimatechangeCitiesaccountforthemajorityoftheworld’sCO2emissionsAsaresultofrapidurbanization,theworld’spopulationincreasinglylivesincities.In2015,48percentoftheworld’spopulationlivedintheconsistentlydefinedglobalsampleofmorethan10,000citiesexaminedhere(Dijkstraetal.2021).Thatpercentageisclosetothenearly52percentoftheglobalpopulationidentifiedaslivingincitiesin2021usingofficialnationaldefinitionsofurbanareas—asharethatisprojectedtoincreasetojustover68percentby2050(UnitedNations2019).11Andanestimated70percentofglobalanthropogenicGHGemissionsemanatefromthesecities(Hopkinsetal.2016).Averagepercapitaemissionsacrosscitiestendtoincreasewithdevelopment,atleastuptoupper-middle-incomestatusUsingdatafromtheGlobalHumanSettlement(GHS)UrbanCentreDatabaseoftheEuropeanCommission,figure1.6showshowaverageproduction-basedfossilCO2emissionsdifferacrosscitiesgloballyonapercapitabasisatdifferentlevelsofdevelopmentandfordifferentregions.12Foremissionsfromtheresidentialandtransportationsectors—thesectorsthaturbanplanningandpoliciescanmostdirectlyinfluence—thedatatellaclearstory:averageemissionsacrosscitiesincreasewiththelevelofdevelopment(figure1.6,panela).13TheaveragepercapitaCO2emissionsfromthesesectorsforcitiesinhigh-incomecountriesarethereforeapproximately4timesthoseofcitiesinupper-middle-incomecountries,10timesthoseofcitiesinlower-middle-incomecountries,and76timesthoseofcitiesinlow-incomecountries.Higherpercapitatransportationandresidentialemissionsinhigher-incomecountrycitiesareconsistentwithhigherlevelsofconsumptionoftransportationandlivingspace.Onaregionalbasis,averagepercapitaemissionsfromtheresidentialandtranspor-tationsectorsforNorthAmericancitiesdwarfthoseforcitiesinanyotherregion;citiesinSub-SaharanAfricahavethelowestaveragepercapitaemissions,followedbycitiesinSouthAsia—seefigure1.6,panelb.However,whenlookingataverageCO2emissionsfromallsectors—theresidentialandtrans-portationsectors,aswellastheenergy,industrial,andagriculturalsectors—thepicturebecomesmorecomplicatedbecausetheGHSUrbanCentreDatabasefailstodistinguishmissingvaluesandzeroesinitsreportingoftheCO2data.Ifthesevalues,whicharecommonintheenergysectordata,aretreatedaszeroes,thenthequalitativepatternsforaverageoverallemissionspercapitaresemblethoseforaverageemissionspercapitafromtheresidentialandtransportationsectors.Thus,averageoverallemissionspercapitaacrosscitiesincreasewiththelevelofdevelopment,withemissionshighestincitiesinhigh-incomecountriesandlowestincitiesinlow-incomecountries(figure1.6,panela).Meanwhile,onaregionalbasisNorthAmericancitiesarethebiggestpercapitaemittersoverall,followedbycitiesintheEastAsiaandPacificandEuropeandCentralAsiaregions.CitiesinSub-SaharanAfricaare,again,thelowestpercapitaemitters(figure1.6,panelb).If,however,suchvaluesaretreatedasmissingandtheassociatedcitiesaredroppedfromthesample,thepicturechanges.Thus,althoughaveragepercapitaoverallemissionsagainincreasewiththelevelofdevelopment,theydosoonlyuptoupper-middle-incomestatus.Citiesinhigh-incomecountrieshavelowerpercapitaoverallemissionsthandocitiesinupper-middle-incomecountries(figure1.6,panela).Onaregionalbasis,EastAsiaandPacificratherthan76THRIVINGNorthAmericancitiesarethelargestpercapitaemitters,whilecitiesinSub-SaharanAfricaremainthelowestpercapitaemittersonaverage(figure1.6,panelb).ThesedifferencesinresultsarelikelyexplainedbythefactthattheCO2emissionsdataintheGHSUrbanCentreDatabasecoveronlyScope1emissions,whicharegenerateddirectlybyactivitywithinacity’sboundaries(box1.5).Crucially,suchemissionsdonotincludeindirectemissionsfromthegenerationofpurchasedenergy,knownasScope2emissions.Theseindirectemissionsareinsteadattributedtothelocationoftheenergy-supplyingpowerplants.Droppingcitieswithmissingorzerovaluesfromthesamplethusimpliesdroppingcitiesthatdonothavepowerplantswithintheirboundaries.Ifpowerplantswithincitiesinupper-middle-incomecountriesproducemoreCO2emissionspercapitathanpowerplantswithincitiesinhigh-incomecountries,thatcouldhelpexplainwhyoverallemissionspercapitaarehigherintheformerthaninthelatter.Figure1.6AverageCO2emissionspercapita,bycountryincomegroupandgeographicregion,2015Source:WorldBankanalysisbasedondatafor10,179citiesfromtheEuropeanCommission’sGlobalHumanSettlement(GHS)UrbanCentreDatabaseR2019(https://ghsl.jrc.ec.europa.eu/ghs_stat_ucdb2015mt_r2019a.php),whichderivesitsCO2emissionsdatafromtheEuropeanCommission’sEmissionsDatabaseforGlobalAtmosphericResearch(EDGARv5.0).Note:Eachbarshowstheunweightedaverageoflong-cycle(fossil)CO2emissionspercapita(measuredintonnesperyearperperson)ofcitiesgroupedbycountryincomegroup(panela)andWorldBankregion(panelb).Fromlefttoright,thefirstbarisbasedonlyonresidentialandtransportationsourcesofCO2emissions.Themiddlebarisbasedonallsources,assumingthatthemissinginformationonCO2emissionsfromtheagriculture,energy,andindustrialsectorsindicatesthatsuchemissionsarezero.Thelastbar,amoreconservativeapproach,isbasedonlyonareducedsampleof3,148citiesforwhichCO2emissionsdataforallsectorsarereported.CO2=carbondioxide.a.Bycountryincomegroupb.BygeographicregionEastAsiaandPacificEuropeandCentralAsiaLatinAmericaandtheCaribbeanMiddleEastandNorthAfricaNorthAmericaSouthAsiaSub-SaharanAfrica0246802468Low-incomeLower-middle-incomeUpper-middle-incomeHigh-incomeAveragefossilCO2emissions,2015(tonnesperyearperperson)AveragefossilCO2emissions,2015(tonnesperyearperperson)ResidentialandtransportationAllsources(fullsample)Allsources(restrictedsample)TheStylizedRelationships77GreenhousegasaccountingandthefourscopesofemissionsGreenhousegasaccountingidentifiesfourscopesofemissions:••Scope1.Emissionsdirectlygeneratedbyactivitywithinacity’sboundaries.••Scope2.Indirectemissionsarisingfromtheuseofenergygeneratedoutsideofacity’sboundaries,includingemissionsfromtheuseofgrid-suppliedelectricitygeneratedbypowerplantslocatedineitherothercitiesorruralareas.••Scope3.Allotherindirectemissionsfromoutsideacity’sboundariesduetoactivitieswithinthecity,includingemissionsthatoccurinvaluechainsassociatedwithfirmslocatedinacity.••Scope4.Avoidedemissionsattributabletotheuseofaproductandthatoccuroutsideofthatproduct’slifecycleorvaluechain,aswellaspolicychoicesorinvestmentdecisionsthathelpavoidemissionsthatwouldhaveotherwiseoccurred.Fuel-savingtiresareanexampleofaproductthatmayhelpavoidemissions;acongestionchargingschemethatsuccessfullyreducesvehiclemilestraveledisanexampleofapolicythathelpsavoidemissions.Informulatinggreenhousegasinventories,mostcountriesandcitieshavetendedtofocusonScope1andScope2emissions(Wiedmannetal.2021);however,thefutureislikelytoseeagreatermovetowardpreparingcompletegreenhousegasinventoriescoveringatleastScopes1,2,and3,butpotentiallyalsotrackingScope4.Box1.5InadditiontoScope1andScope2emissions,GHGaccountingalsoconceptuallydefinesScope3andScope4emissions(box1.5).Animportantrelateddistinctionexistsbetweenproduction-andconsumption-basedemissionsestimates,whichraisestheimportantquestionofwhetheracity’sCO2emissionsshouldbeadjustedforinternationalandinternaltrade(box1.6).Inabsoluteterms,citiesinupper-middle-incomecountriesandinEastAsiacontributethemosttoglobalCO2emissionsMovingfrompercapitatoabsoluteproduction-basedemissions,thedatatellaconsistentstoryregardlessofhowmissingvaluesaretreated.Thus,foraggregateemissionsfromallsectors,citiesinupper-middle-incomecountriescontributemosttoglobalfossilCO2emis-sions,accountingforabout55percentofthetotalemissionsgeneratedincities(figure1.7,panela).Theyarefollowedbycitiesinhigh-incomecountries,whichaccountforabout31percentofglobalemissionsgeneratedincities.Bycontrast,citiesinlower-middle-incomecountriesaccountforonlyabout13percentofemissionsgeneratedincities,andcitiesinlow-incomecountriesforameager0.21percent.Thefactthatcitiesinupper-middle-incomecountriesaccountforagreatershareofemissionsthancitiesinhigh-incomecountriesdespitetheirlowerpercapitaemissions(thecasewhenmissingvaluesaretreatedaszeroes)isaresultoftheircollectivelylargerpopulation.14Meanwhile,onaregionalbasiscitiesintheEastAsiaandPacificregionaccountforabout50percentofglobalCO2emissionsgeneratedincities,78THRIVINGShouldacity’scarbondioxideemissionsbeadjustedforinternationalandinternaltrade?Aconsumption-basedapproachtocarbondioxideemissionsaccounting—onethatassignsemissionstothepointoffinalconsumptionofgoods,services,andenergy—offersaverydifferentpictureofacity’semissionsthandoesonebasedonproduction.Consumption-basedaccountingincludesemissionsassociatedwithgoods,services,andenergyimportedandconsumedlocally,butexcludesemissionsassociatedwithexportedgoods,services,andenergy(orimportedonlytoaddvaluebeforeexport-ingforfinalconsumptionelsewhere).Italsoexcludesemissionsassociatedwithbusi-nessesservingvisitors,suchasemissionsfromhotelsandtouroperatorsservingtourists(Setoetal.2021).Consumption-basedemissionsmay,then,behigherorlowerthanproduction-basedemissions,dependingonthenatureoftrade,includinginternaltradewithbothothercitiesandruralareasinthesamecountry,andtourisminthecity.Mostglobalemissionsdatauseaproduction-basedapproach,includingtheEmissionsDatabaseforGlobalAtmosphericResearch(EDGAR)dataincludedintheGlobalHumanSettlement(GHS)UrbanCentreDatabase.Moranetal.(2018),however,estimateconsumption-basedemissionsforthesamesetofcitiesastheGHSUrbanCentreDatabase(andthereforeasthisreport’sglobalsample).Theirmodelusesurbanversusruralcon-sumptionpatternsandpurchasingpowerasthemainpredictorsofpercapitacarbondioxideemissions,soitisagnosticonthepointoforiginofgoods,services,andenergyconsumed.Onthisbasis,Moranetal.findthatarelativelysmallnumberofcitiesaccountforadisproportionateshareoftheworld’scarbonfootprint.Forexample,thetop100citiesbycarbonfootprintareresponsiblefor18percentofglobalconsumption-basedemissions.Mostofthecitiesinthetop200areinhigh-incomecountrieswithhighnationalcarbonfootprints;however,41ofthetop200arelargecities—forexample,Cairo,Dhaka,andLima—incountriesthatotherwisehavelowcarbonfootprints.Thesecitieshavehightotalcarbonfootprintsbecauseofboththeirsizeandtherelativeaffluenceoftheirres-idents.Mostoftheworld’slargestcitiesandsmallcities,aswellasmostmedium-sizecitiesinhigh-incomecountries,havehigherconsumption-basedthanproduction-basedemissions.Theoppositeistrueformedium-sizecitiesinlow-andmiddle-incomecoun-tries,presumablybecausetheyaremoreorientedtowardmanufacturingforexport,whiletheirownresidentshaverelativelylowconsumptionlevels.Informulatinggreenhousegasinventoriesandthereforetheirapproachestonetzeroemissions,countriesandcitieshavetendedtoadoptaproduction-basedaccountingper-spective.Thisperspectivenaturallyencouragesaninfrastructure-focusedapproachtoclimateaction.Sweden,however,recentlybrokewiththistrend.InApril2022,itspolit-icalpartiesagreedtoincludeconsumption-basedemissionsinthecountry’sclimatetargets—thefirstcountrytodoso.SwedenwillthushavetonotonlylookatitsterritorialemissionsbutalsoaccountforemissionsimportedintoSweden(Morgan2022).OthercountriesandcitiesseekingtobeattheforefrontofclimateactionwilllikelyfollowSweden’slead.C40Cities,aglobalnetworkofmayors,recentlyconductedastudyofitsmembercitiesandfindsthattheyrepresent10percentofglobalgreenhousegasemis-sionswhenaccountingforconsumption-basedemissions(C40CitiesClimateLeadershipGroup,Arup,andUniversityofLeeds2019).Thestudyalsofindsthatactiononconsump-tioninthenetwork’smembercitiescouldcontributetolargereductionsinemissionsby2050forkeyconsumptioncategories,includingclothingandtextiles(66percent),food(60percent),andaviation(55percent).Box1.6TheStylizedRelationships79followedbycitiesintheEuropeandCentralAsiaregionandNorthAmerica,whichtogetheraccountforabout25percent.Sub-SaharanAfricancities,bycontrast,accountforonlyabout1.6percentofoverallemissions(figure1.7,panelb).Fortheresidentialandtransportationsectors,citiesinhigh-incomecountriescontributethemosttoglobalCO2emissionsBycontrast,fortheresidentialandtransportationsectors—again,thesectorsurbanplanningandpoliciescanmostdirectlyinfluence—citiesinhigh-incomecountriesaccountforthelargestshareofglobalCO2emissionsgeneratedincities(figure1.7,panela).AlthoughcitiesintheEastAsiaandPacificregion,collectively,remainthebiggestemitters,theshareofglobalemissionstheygenerateinthesesectorsis,atabout33percent,muchsmallerthantheirshare0204060Low-incomeLower-middle-incomeUpper-middle-incomeHigh-incomeShareofglobalfossilCO2emissions,2015(%)0204060EastAsiaandPacificEuropeandCentralAsiaLatinAmericaandtheCaribbeanMiddleEastandNorthAfricaNorthAmericaSouthAsiaSub-SaharanAfricaShareofglobalfossilCO2emissions,2015(%)a.Bycountryincomegroupb.BygeographicregionResidentialandtransportationAllsources(fullsample)Allsources(restrictedsample)Figure1.7ShareofglobalCO2emissionsgeneratedincities,bycountryincomegroupandgeographicregion,2015Source:WorldBankanalysisbasedondatafor10,179citieswithcompleteinformationonresidentialandtransportationCO2emissionsfromtheEuropeanCommission’sGlobalHumanSettlement(GHS)UrbanCentreDatabaseR2019(https://ghsl.jrc.ec.europa.eu/ghs_stat_ucdb2015mt_r2019a.php),whichderivesitsCO2emissionsdatafromtheEuropeanCommission’sEmissionsDatabaseforGlobalAtmosphericResearch(EDGARv5.0).Note:Eachbarshowstheshareofgloballong-cycle(fossil)CO2emissionsgeneratedincitiesclassifiedbyincomegroupandWorldBankregion.Fromlefttoright,thefirstbarisbasedonlyonresidentialandtransportationsourcesofCO2emissions.Themiddlebarisbasedonallsources,assumingthatthemissinginformationonCO2emissionsfromtheagriculture,energy,andindustrialsectorsindicatesthatsuchemissionsarezero.Thelastbar,amoreconservativeapproach,isbasedonlyonasmallersampleof3,148citiesforwhichCO2emissionsdataforallsectorsarereported.CO2=carbondioxide.80THRIVINGofglobalurbanemissionsfromallsectors.Meanwhile,citiesinNorthAmericaandEuropeandCentralAsiaaccountforabout43percentoftheglobalresidentialandtransportationsectoremissionsgeneratedincities(figure1.7,panelb).Ascitiesdevelop,thesourcesofCO2emissionschangeAscitiesdevelop,notonlydotheirlevelsofCO2emissionstendtochangebutso,too,dothesourcesofthoseemissions(figure1.8).Thus,emissionsfromenergyandindustryaremoreimportantforcitiesinmiddle-incomecountriesthanforthoseinlow-incomecoun-tries.However,althoughthesesourcesremainlargeinabsoluteterms,theyarerelativelylessimportantforcitiesinhigh-incomecountriesthanforthoseinmiddle-incomecountries.Bycontrast,theresidentialandtransportationsectorsaremoreimportantsourcesofemissionsforcitiesinhigh-incomecountriesinbothabsoluteandrelativeterms.Thesepatternsareconsistentwiththepatternofstructuraltransformationthattendstoaccompanydevelopment.Stereotypically,thispatterninvolvestheemergenceofindustryFigure1.8Averagesharesoflong-cycleCO2emissionssourcesforcities,bycountryincomegroup,2015Source:WorldBankanalysisbasedondatafor10,179citieswithcompleteinformationonresidentialandtransportCO2emissionsfromtheEuropeanCommission’sGlobalHumanSettlement(GHS)UrbanCentreDatabaseR2019(https://ghsl.jrc.ec.europa.eu/ghs_stat_ucdb2015mt_r2019a.php),whichderivesitsCO2emissionsdatafromtheEuropeanCommission’sEmissionsDatabaseforGlobalAtmosphericResearch(EDGARv5.0).Note:Eachbarshowstheunweightedaveragepercentagecontributionofeachofthefivesectorstothetotallong-cycle(fossil)CO2emissionsacrosscitiesineachincomeclass.ItisassumedthatmissinginformationonCO2emissionsfromtheagriculture,energy,andindustrialsectorsindicatesthatsuchemissionsarezero.Resultsarequalitativelyidenticalifcitieswithmissingvaluesareinsteaddroppedfromthesample.CO2=carbondioxide.020406080100Low-incomeLower-middle-incomeUpper-middle-incomeHigh-incomeShareofglobalfossilCO2emissionsbyuse,2015(%)AgricultureEnergyIndustryResidentialTransportationTheStylizedRelationships81incitiesatanearlystageofdevelopment,followedbyindustry’ssubsequentmovetotheperipheriesofcitiesandmoreruralareas,aswellasoverseas,asdevelopmentevolves(WorldBank2009).Moregenerally,citiesincreasinglybecomecentersofconsumptionratherthanofproductionasdevelopmentprogresses,againraisingthequestionofwhetheracity’sCO2emissionsshouldbeadjustedforinternationalandinternaltrade(box1.6).LargerandmoredevelopedcitieshavehigherlevelsofresidentialandtransportationCO2emissionsFocusingagainontheresidentialandtransportationsectors,regressionanalysisrevealsthat,intheglobalsample,largerandmoredevelopedcitieshavelargerCO2emissions(figure1.9).Thus,inbothsectorsthesizeofacity’spopulation,built-uparea,andgrossdomesticproduct(GDP)percapitaofthecountryinwhichitislocatedareallsignificantlypositivelyrelatedtoitslevelofCO2emissions.–4.0–3.0–2.0–1.00.01.02.0Log,citypopulationLog,countryGDPpercapitaLog,built-upareaCitycompactnessEstimatedeectonCO2emissionsResidentialTransportationFigure1.9DeterminantsofCO2emissionsforcitiesglobally,residentialandtransportationsectors,2015Source:WorldBankanalysisbasedondatafor2015for2,785citieswithpopulationofover200,000fromtheEuropeanCommission’sGlobalHumanSettlement(GHS)UrbanCentreDatabaseR2019(https://ghsl.jrc.ec.europa.eu/ghs_stat_ucdb2015mt_r2019a.php),whichderivesitsCO2emissionsdatafromtheEuropeanCommission’sEmissionsDatabaseforGlobalAtmosphericResearch(EDGARv5.0).Note:Foreachsector,thegraphpresentsestimatedcoefficients,togetherwiththeassociated95percentconfidenceintervals,fromaregressionofacity’slogCO2emissionsonthelogofitspopulation,thelogoftheGDPpercapitaofthecountryinwhichthecityislocated,thelogofitsbuilt-uparea,andameasureofthecity’scompactness(thePolsby-PopperRatio).Theregressions,whichalsocontrolforacity’sclimate(precipitation,temperature,biome,andelevation),arebasedoncross-sectionaldatafor2015withrobuststandarderrors.CO2=carbondioxide.82THRIVINGMorecompact,lesssprawlingurbandevelopmentmayoffsethigherlevelsofresidentialandtransportationCO2emissionsTheregressionresultsalsorevealthat,controllingforacity’ssizeandlevelofdevelopment,more-compactcitieshavelowerproduction-basedCO2emissionsfromtheresidentialandtransportationsectors(figure1.9andfigure1.10).15Thiseffectisstronginbothsectors.Intheresidentialsector,a1-standard-deviationincreaseinacity’scompactnessisassociatedwitha77percentreductioninemissionsandinthetransportationsectorwithan81percentreduction.TheseresultsareconsistentwiththehypothesisthaturbanplanningandpoliciesthatenablebettermanagementoflandandpropertymarketstressesleadtolowerCO2emissions.Loweremissionscouldstem,forexample,frommoreinfillasopposedtoleapfrogdevelopmentandlessrelianceonpersonalmotorizedvehicles.Thishypothesisisconsis-tent,inturn,withfindingsthatdensercities,whichalsotendtobemorecompact,havelesscommutingandcaruse,aswellaslowerlevelsofdomesticenergyconsumption(AhlfeldtandPietrostefani2019).Theresultsarealsoinlinewiththeideathatencouragingtallbuildingsthrough,forexample,therelaxationofrestrictivefloorarearatios,mayhavebeneficialenvironmentaleffects.Whenconsideringtheenvironmentalimpactsoftallbuildings,however,thelowerCO2emissionsassociatedwithamorecompacturbanformmustbeweighedagainstthehigherembeddedemissionsassociatedwiththeconstructionofthosebuildings(seechapter4).AirpollutionincitiesPoorairqualityandclimatechangearecloselylinkedOfthevariousairpollutants,oneofthemostconcerning,andtheoneonwhichthissectionfocuses,isPM2.5.PM2.5canpenetratedeepintothelungs,increasingthelikelihoodofasthma,lungcancer,severerespiratoryillness,andheartdisease.Furthermore,blackcarbon,ashort-livedclimatepollutant,constitutesamajorpartofPM2.5.Thus,notonlyarePM2.5emissionsstronglycorrelatedwithCO2emissions(figure1.11),butPM2.5emissionsthemselvesalsocontributetoclimatechange.16Thisfindingimpliesthaturbanpoliciesthatmitigateclimatechangealsocouldsignificantlyimproveairquality(andviceversa).Betterairqualitywouldproducenotonlyhealthbenefitsbutalsoproductivitybenefits,becausestudieshaveshownthatpoorairqualityharmsworkerproductivity(KahnandLi2020).LargerandlesscompactcitieshavemorepollutionthandosmallerandmorecompactonesLiketheequivalentCO2emissions,acity’sresidentialandtransportationsectorPM2.5emis-sionsaresignificantlypositivelyrelatedtoitspopulationsize,controllingfor,amongotherthings,theGDPpercapitaofthecountryinwhichthecityislocatedanditsbuilt-uparea(figure1.12).17Inotherwords,foragivenlevelofdevelopment,largercitiestendtohavemoreairpollution.Thisfindingisconsistentwiththoseoftwopreviousstudiesthatfind,usingdifferentdatasources,apositiverelationshipbetweenacity’ssizeanditsannualmeanTheStylizedRelationships83Coe:−1.495;P−value:0.000Coe:−1.724;P−value:0.000LogoffossilCO2emissions,2015(tonnes/year)LogoffossilCO2emissions,2015(tonnes/year)00.10.20.30.4–0.3–0.2–0.1Citycompactnessb.Transportationsectora.Residentialsector42–2–4–600.10.20.30.4–0.3–0.2–0.1Citycompactness42–2–4–6Figure1.10RelationshipbetweencitycompactnessandCO2emissionsacrosscitiesglobally,residentialandtransportationsectors,2015Source:WorldBankanalysisbasedondatafor2,785citieswitha2015populationofover200,000fromtheEuropeanCommission’sGlobalHumanSettlement(GHS)UrbanCentreDatabaseR2019(https://ghsl.jrc.ec.europa.eu/ghs_stat_ucdb2015mt_r2019a.php),whichderivesitsdataonCO2emissionsfromtheEuropeanCommission’sEmissionsDatabaseforGlobalAtmosphericResearch(EDGARv5.0).Note:GraphsarepartialscatterplotsoflogCO2emissionsin2015onameasureofcitycompactness(thePolsby-PopperRatio)controllingforthelogofacity’spopulation,thelogofitsbuilt-uparea,acity’sclimate(precipitation,temperature,biome)andelevation,andcountryfixedeffects.CO2=carbondioxide.84THRIVINGconcentrationofPM2.5foramorelimitedsampleof381citiesinlow-andmiddle-incomecountries(EllisandRoberts2016;FerreyraandRoberts2018).18AlsomirroringthefindingsforCO2emissions,morecompactcitieshavesignificantlylowerPM2.5emissionsfromtheresidentialandtransportationsectors(figure1.12).Thisfindingsuggeststhatmorecompacturbandevelopment,bypromotinglessrelianceoncars,notonlyreducesacity’scontributiontoclimatechangebutalsoreduceslocalairpollution,therebyproducingpositiveeffectsonhealthandwell-being.EvidenceforIndiaandLatinAmericaandtheCaribbeanalsosuggeststhatmorecompacturbandevelopmentmaybeassociatedwithhigherproductivitylevelsandgrowthrates(Duqueetal.2021;TewariandGodfrey2016).Thispossibilitystemsnotonlyfromimprovedairqualitybutalsofromthefactthatmorecompactdevelopmentincreasesthelikelihoodofknowledgespillovers,therebystrengtheningagglomerationeconomies.Figure1.11RelationshipbetweenCO2andPM2.5emissionsacrosscitiesglobally,residentialandtransportationsectors,2015Source:WorldBankanalysisbasedondatafor10,179citieswithcompleteinformationonresidentialandtransportationCO2andPM2.5emissionsfromtheEuropeanCommission’sGlobalHumanSettlement(GHS)UrbanCentreDatabaseR2019(https://ghsl.jrc.ec.europa.eu/ghs_stat_ucdb2015mt_r2019a.php),whichderivesitsemissionsdatafromtheEuropeanCommission’sEmissionsDatabaseforGlobalAtmosphericResearch(EDGARv5.0).Note:Thegraphisascatterplotofthelogofthesumofresidentialandtransportationlong-cycleCO2emissionsin2015onthelogofthesumofresidentialandtransportationPM2.5emissionsinthesameyear.Theslopeofthefittedlineimpliesthatahalvingofacity’sPM2.5emissionsintheresidentialandtransportationsectorsisassociatedwitharoughly43percentreductioninCO2emissions.CO2=carbondioxide;PM2.5=particulatematterof2.5micronsorlessindiameter.0369121518–10–5051015LogoffossilCO2emissions,2015(tonnes/year)LogoffossilPM2.5emissions,2015(tonnes/year)R2=0.46TheStylizedRelationships85Pollutionfirstincreaseswithdevelopment,then(eventually)declinesThus,overallPM2.5concentrationsarehigher,onaverage,forcitiesinlower-middle-incomecountriesthanforcitiesinlow-,upper-middle-,andhigh-incomecountries(figure1.13).Incitiesinhigh-incomecountries,theaverageconcentrationisevenlowerthanthatforcitiesinlow-incomecountries.Thisfindingsuggeststheso-calledenvironmentalKuznetscurverelationship.Manyfactorscouldexplainthetendencyofairpollutionincitiestodeclineascountriesmovefromlower-middle-toupper-middle-andthentohigh-incomestatus.Moreprosperouscoun-triestendtospecializemoreinserviceswhoseproductionislesspollutingthantheproduction–4.0–5.0–3.0–2.0–1.00.01.02.0Log,citypopulationLog,countryGDPpercapitaLog,built-upareaCitycompactnessResidentialTransportationEstimatedeectonPM2.5emissionsFigure1.12DeterminantsofPM2.5emissionsforcitiesglobally,residentialandtransportationsectors,2015Source:WorldBankanalysisbasedondatafor2,785citieswitha2015populationofover200,000fromtheEuropeanCommission’sGlobalHumanSettlement(GHS)UrbanCentreDatabaseR2019(https://ghsl.jrc.ec.europa.eu/ghs_stat_ucdb2015mt_r2019a.php),whichderivesitsdataonPM2.5emissionsfromtheEuropeanCommission’sEmissionsDatabaseforGlobalAtmosphericResearch(EDGARv5.0).Note:Foreachsector,thegraphpresentsestimatedcoefficients,togetherwiththeassociated95percentconfidenceintervals,fromaregressionofacity’slogPM2.5emissionsonthelogofitspopulation;thelogoftheGDPpercapitaofthecountryinwhichthecityislocated,aswellasitsinteractionwithadummyvariablethattakesthevalue1ifthecountryisupper-middle-orhigh-income;thelogofitsbuilt-uparea;andameasureofthecity’scompactness(thePolsby-PopperRatio).Theregressions,whichalsocontrolforacity’sclimate(precipitation,temperature,biome)andelevation,arebasedoncross-sectionaldatafor2015withrobuststandarderrors.PM2.5=particulatematterof2.5micronsorlessindiameter.86THRIVINGofmanufacturedgoods,whichtheyhaveagreatertendencytoimport.Althoughtheconsump-tionofimportedmanufacturedgoodsinmoredevelopedcountriesmaystillbeassociatedwithairpollution,thatpollutionoccursinothercountries(Feng,Hubacek,andYu2019;WiedmannandLenzen2018).Moreprosperouscountriesalsooftenhaveagreatercapacityandresourcestomonitorpollutionandtoenactandenforceregulations(Aweetal.2015;Dasguptaetal.2004).Moreover,cleanerfuelsandtechnologiesaremoreaffordableinrichercountries.Forexample,peopleinpoorercountriesuseinexpensivebuthighlypollutingbiomassfuelssuchaswoodandcoalforhouseholdcookingandheating(Bruce,Perez-Padilla,andAlbalak2000).Meanwhile,residentsofmoredevelopedcountriestendtohavehigherenvironmentalhealthliteracybecausetheyhavehigherlevelsofeducationandbetteraccesstoinformation.Asaresult,theyaremoreawareoftheadverseimpactsofpoorairqualityandthuslikeliertodemandpolicyactioninresponse(Raufmanetal.2020).Wealthierandbetter-educatedpeoplearealsomorewillingtopayforimprovementsinairquality(Mariel,Khan,andMeyerhoff2022).Betterurbanairqualityshouldnot,however,beconsideredanautomaticoutcomeofacoun-try’sevolutionfromlower-middle-tohigher-incomelevels.19AsthehistoryofLondondemon-strates(box1.7),theintroductionofpoliciesthathelpedimproveairqualityinmanyoftoday’smostdevelopedcitieshaveoftenencounteredstiffpoliticalresistancefromvestedinterestgroups.Sometimes,onlyacrisishascatalyzeddecisivepoliticalaction.Figure1.13AverageofPM2.5concentrationsacrosscities,bycountryincomegroup,2000and2015Source:WorldBankanalysisbasedondatafor10,303citiesfromtheEuropeanCommission’sGlobalHumanSettlement(GHS)UrbanCentreDatabaseR2019(https://ghsl.jrc.ec.europa.eu/ghs_stat_ucdb2015mt_r2019a.php),whichderivesitsdataonPM2.5concentrationsfromtheGlobalBurdenofDisease(GBD)2017databaseonambientairpollution.Note:ThegraphshowstheweightedaverageofPM2.5concentrationsacrosscitiesineachincomeclassification,wheretheweightsaregivenbycitypopulation.μg/m3=microgramspercubicmeter;PM2.5=particulatematterof2.5micronsorlessindiameter.1020304050Low-incomeLower-middle-incomeUpper-middle-incomeHigh-incomeAverageconcentrationofPM2.5(µg/m3)20002015TheStylizedRelationships87HowdidLondongetridof(mostof)itssmog?TheproblemofindustrialairpollutioninLondonwasrecognizedasearlyas1661,ifnotbefore,whenagovernmentofficial,JohnEvelyn,presentedKingCharlesIIwithatreatisetitledFumifugium:ortheInconvenienceoftheAerandSmoakeofLondonDissipated.In1819,acommitteeappointedtostudytheproblemconcludedthatthecity’ssmoke-filledairwasbadforresidents’health.However,itwasnotuntilenactmentoftheSmokeNuisanceAbatement(Metropolis)Actsof1853and1856thatLondon’spolicewereempoweredtoactagainstindustrialemittersofsmoke.Theseregulations,enacteddespiteoppo-sitionfromindustrialinterestsclaimingthatsmokehadhealthbenefits,werenotenough.BecausemostofLondon’ssmokecamenotfromindustrialsourcesbutfromthechimneysofprivatehouses,whichusedacoalfireforheating,London’sairqualityhardlyimproved.TheturningpointcameaboutacenturylaterwhentheGreatSmogof1952suffocatedLondon.WeatherconditionscausedunprecedentedlevelsofsmokeandsulfurdioxidetoconcentrateinLondon’sair,leadingtoanestimated3,500–4,000deaths.Althoughthegovernmentinitiallyclaimedthatnonewregulationswereneeded,iteventuallysuc-cumbedtopressureandopenedaninquiryintotheproblemofairpollution.Therec-ommendationsoftheinquiryledtopassageoftheCleanAirActof1956,whichwasextendedin1968.Thisactregulatedpollutionfromresidential,commercial,andindus-trialsourcesandprovidedgrantsforretrofittinghousestoburnsmokelessfuels.Italsocreated“smokelesszones”coveringhalfofGreaterLondon,inwhichonlysmokelessfuelswerepermittedfordomesticheating.Asaresult,London’sairpollutiondeclinedrapidly.Thewidespreadadoptionofcentralheating,alongwithfurtherlegislationinsubsequentdecades,furthercontributedtothisdecline.Despitetheseimprovements,thousandsofLondonersdiedeachyearfromcauseslinkedtoairpollutionexposure—evenasrecentlyas2016.Aseriesofregulationssincethen,includingabanontheentryofpollutingvehiclesintocentralLondon’sUltraLowEmissionsZone,andtechnologicalimprovementstobusandtaxifleetshavehaddramaticresults.Forexample,thenumberofLondonerslivinginareasexceedingthelegallimitfornitrogenoxideconcentrationsfellby94percentbetween2016and2019.Evenafterthisimprovementinairquality,however,nearlyallofLondonstillexceedstheWorldHealthOrganization’sguidelinelimitforparticulatematterof2.5micronsorlessindiameter.London’sexperienceoffersthreelessons.First,improvementsinairqualitydonotauto-maticallygohandinhandwithdevelopment.Rather,theyresultfromdeliberatepolicyactions,oftenenactedinthefaceofstiffoppositionfromvestedinterestgroups.Second,althoughobviouslybetteravoided,adisastersuchastheGreatSmogof1952canpresentanopportunityfordecisiveactiontotackleseriousenvironmentalillsthatmightother-wiselinger.And,third,thebattletoaddressurbanstressesneverends.Eveninthemostdevelopedofcities,policymakersneedtocontinuallyseekpolicyimprovementstokeepatbaythestressesthatarisefromthepressureofacity’spopulationonitslandandhousingmarkets,itssuppliesofbasicservicesandinfrastructure,andtheenvironment.Sources:BasedonGreaterLondonAuthority2002,2020.Box1.788THRIVINGGreencover:Towhatextentarecitiesliterallygreen?Thegreeneryincities—thatis,thetreesandothervegetationinparksandelsewhere—comeswithimportantbenefitsforacity’sresidents.Notonlydoesgreeneryhaveaninherentamenity,butitspresencecanalsoplayanimportantroleinmitigatingtheurbanheatislandeffect(seebox1.4andDeuskar2022).Despitethosebenefits,thedemandforlandthatcomeswiththegrowthofurbanpopulationcanplacedevelopmentpressureongreenspaceswithinacity,addingtotheotherstressesofurbanization.Tocaptureacity’saveragegreenness,thissectiondrawsondataderivedfromsatelliteimageryforthisreport’sglobalsampleofmorethan10,000citiesand,inparticular,onameasureofaveragegreennessbasedonthecolorofthepixelswithineachcity’sarea.20Morepopulouscitieshaveloweraveragegreenness,andaveragegreennessfirstdeclineswithdevelopmentbeforeincreasingRegressionanalysisrevealsanegativerelationshipbetweenacity’saveragegreennessanditspopulationsize,controllingforbothitslevelofdevelopmentandaspectsofitsgeographyandclimatethatmaybecorrelatedwithbothgreennessandsize(figure1.14,panela).LiketheresultsforCO2andPM2.5emissions,thisfindingisconsistentwiththeexistenceofstrongerstressesforlargercities.Bycontrast,mirroringtheresultsforairpollution,acity’saveragegreennessfirsttendstodecreaseasdevelopmentofthecountryinwhichitislocatedadvancesbeforesubsequentlyincreasing(figure1.14,panelb).Thisrelationshipholdsforthethreeyears—1990,2000,and2014—forwhichdataareavailablebuthasshiftedovertime.Althoughin1990citiesinlower-middle-incomecountrieshadthelowestlevelsofaveragegreenness,in2000and2014citiesinupper-middle-incomecountriesdid.Mostcitieshavebecomegreenerinrecentdecades,althoughprobablynotbecauseofdeliberategreeningpoliciesPanelboffigure1.14alsorevealsthatbetween1990and2000averagegreennessincreased,onaverage,acrosscitiesatalllevelsofdevelopment,thenstabilizedbetween2000and2014.Fortheentire1990–2014period,however,mostcitiesgloballyexhibitedanincreaseintheiraveragegreenness(map1.1).ThemostnotableexceptionstothistrendwerethecitiesonChina’seasternseaboard,whichgrewextremelyrapidlyoverthesampleperiod,andthoseinSouthAmerica.Beijinghasrespondedtothesetrendswithdeliberatepoliciesaimedat(re-)greeningthecity(box1.8).Althoughthewidespreadincreasesinaveragegreennessoverthelastthreedecadesmayseemlikeanencouragingdevelopment,itisunlikelythat,withsomeexceptions,adoptionofgreeningpoliciesbycitieshasdriventhetrend.Indeed,onaglobalscalegreennessismorestronglyaffectedbychangesinCO2concentrationsthanbylandusechanges(Schutetal.2015).Thiseffectdoesnotmeanthatclimatechangeisbeneficialintermsofvegetation.Tothecontrary,climatechangeincreasesthestressorsthatundermineplantresilienceanddisruptbothforeststructureandecosystemservices.Risingtemperatures0.40.20–0.2–0.4Averagegreenness,2014–20246Population,2015(ln)a.Averagegreenness,bypopulationsizeb.Averagegreenness,bycountryincomegroup0.250.300.350.400.45Low-incomeLower-middle-incomeUpper-middle-incomeHigh-incomeAveragegreenness199020002014TheStylizedRelationships89alsoincreasethefrequencyofdroughts,wildfires,andinvasivepestoutbreaks,leadingtothelossofplantspecies.21Howresilientarecitiestoday?Nowthatitisclearerhowgreencitiesareglobally,thissectionturnstothequestionoftheirresilience.Toseewhatlessonsmightemergeaboutcities’resiliencetoclimatechange–relatedshocks,itfirstconsidersevidencefromtheurbaneconomicsliteratureonthelong-runresil-ienceofcitiestolarge-scaleshockstotheirstocksofbothphysicalandhumancapital.Itthenpresents,forthisreport’sglobalsampleofcities,newempiricalevidenceontheshort-runFigure1.14Relationshipbetweenaveragegreennessandpopulationsizeacrosscitiesglobally,2014,andaveragegreennessandlevelofdevelopmentacrosscitiesglobally,1990,2000,and2014Source:WorldBankanalysisbasedondatafor8,893citiesfromtheEuropeanCommission’sGlobalHumanSettlement(GHS)UrbanCentreDatabaseR2019(https://ghsl.jrc.ec.europa.eu/ghs_stat_ucdb2015mt_r2019a.php).Note:Panelaisapartialscatterplotofacity’saveragegreennessin2014againstthenaturallogofitspopulationin2015controllingforitsGDPpercapitalevel,latitudeandlongitude,elevation,averagelevelsofrainfallandtemperaturein2014,biomezone,andcountryfixedeffects.Panelbshowstheunweightedmeanofaveragegreennessacrosscitiesbycountryincomegroup.ln=naturallog.90THRIVINGBeijing’safforestationmiracleBeijing—amegacityof22millionpeoplethatsuffersfromairpollution,theurbanheatislandeffect,andahostofotherenvironmentalproblems—mightnotbeexpectedtoofferapositiveexampleofurbangreening.Between2012and2016,however,thecityunder-tookamassiveurbanafforestationprogram,theOneMillionMuPlainAfforestationProject(amuisequivalentto1/15ofahectare).Theprojectinvolvedplanting54milliontreesoveranareaof700squarekilometers.Muchoftheafforestedarealiesbetweenthecity’scentraldistrictsandnewcitycentersontheperiphery.Theurbanforestecosys-temnowincludeslargeforests,ecologicalcorridors,nine“wedges”ofgreenspace,greenbelts,andparksofvarioussizes.Becauseofthisinitiative,treecoverintheBeijingplainincreasedby42percent.IndependentstudiesofBeijing’safforestationinitiativefindthatitsuccessfullyincreasedthecity’sgreencover.Nevertheless,thelocationsinwhichafforestationoccurredwerenotalwayslaidoutasspecifiedintheoriginalplans.Althoughtheinitiativeoriginallyintendedtoavoidafforestingagriculturalland,almost80percentofthenewlyforestedareacoversformercropland.Studiesalsoobservetheneedforgreatercivicparticipation,whichincludesgivingvoicetootherwiseexcludedormarginalizedmembersofsociety,intheinitiative’splanningandimplementation.Sources:FAO2018;Jin,Sheppard,andWang2021;Yao,Xu,andZhang2019.Box1.8Map1.1Percentagechangeinaveragegreennessacrosscitiesglobally,1990–2014Source:WorldBankanalysisbasedondataonthelevelofaveragegreennessfor10,660cities,asdefinedinbox1.1,fromtheEuropeanCommission’sGlobalHumanSettlement(GHS)UrbanCentreDatabaseR2019(https://ghsl.jrc.ec.europa.eu/ghs_stat_ucdb2015mt_r2019a.php).Note:Thepercentagechangeinaveragegreennessbetween1990and2014iscalculatedas[(Avg.GreenBU,2014–Avg.GreenBU.1990)/Avg.GreenBU,1990]×100,whereAvg.GreenBU,xdenotesthelevelofaveragegreennessinacity’sbuilt-upareainyearx.TheStylizedRelationships91resilienceofeconomicactivitytothevarioustypesofweatheranomalies—hot,cold,wet,anddryevents,aswellastropicalcyclones—whoseevolutioninresponsetoclimatechangewasexaminedearlierinthischapter.Indoingso,thissectionalsoconsidershowacity’sresiliencevarieswithitssize,levelofdevelopment,andbaselineclimate.Evidenceonthelong-runresilienceofcitiestothedestructionofmanyformsofphysical,andevenhuman,capitalprovidessomegroundsforoptimismAlargeempiricalliteraturehasemergedoverthelasttwodecadesdocumentingthelong-runresilienceofcities—definedhereastheabilityofcitiestorecovertheirpopulationlevelsandeconomicvitalityfollowinganadverseshock—tomanyformsofextremephysicaldestruction.Forexample,Schencking(2006,833)describesthegreatTokyoearthquakeof1923,whichleftmorethan120,000deadandjustover1.5millionhomeless,as“oneofthemostdevastinganddisruptivenaturaldisastersofthe20thcentury.”YetnotonlydidTokyosurvivetheearthquake,butitalsowentontothrive.By1925,justtwoyearsaftertheearthquake,thepopulationofTokyoprefecturehadreached4.49million,anincreaseof64percentoverthatin1920,afewyearsbeforetheearthquake.By1940,itspopula-tionhadfurtherclimbedto7.35million,almostdoubleitspre-earthquakelevelin1920(Glaeser2022).Similarly,Ager,Eriksson,andLonstrup(2020)findthatthe1906SanFranciscoearthquake,whichdestroyed28,000buildings,killedabout3,000people,andleftmorethan225,000homeless,didnotspellthecity’send.Quitethecontrary—between1900and1910,SanFrancisco’spopulationgrewatafasterratethanbetweenboth1890and1900and1910and1920.Moreover,SanFranciscogrewfasterthancitiessuchasBostonandPhiladelphia,whichwereamongitsmaincompetitorsatthetime.Morerecently,althoughHurricaneKatrinawasassociatedwithalargedropinthepopulationofboththecityandmetropolitanareaofNewOrleansafterithitin2005,thepopulationofthemetropolitanareahassincelargelyrecovered(Glaeser2022).Thesefindingsofcities’long-runresiliencetothephysicaldestructioncausedbyearthquakesandtropicalcyclonesalsocarryovertoothersourcesofmajorphysicaldestruction,includingtheGreatFireofLondonof1666,theBostonFireof1872,andtheChicagoFireof1871(Glaeser2022).Indeed,thelattertwofires,bothofwhichcauseddevastatingshort-termphysicaldamage,seemtohavegeneratedlonger-termeconomicbenefitsbyallowingthecitiestobuildbackbetter(Glaeser2012;HornbeckandKeniston2017).InChicago,aclutchoftalentedarchitectswereattractedtothecitybyitscontinuedpopulationgrowth,whichimpliedahugedemandforfloorspace,andtheemptylandleftwhenphysicalstructuresatthecity’scenterburned.AccordingtotherenownedurbaneconomistEdwardGlaeser(2022,10),“Thisagglom-erationoftalentbothrebuiltthecity,andinventedtheskyscraper,whichwouldreshapecityskylinesworldwide,intheprocess.”Historically,evenlarge-scalebombing,suchasthatexperiencedbyGermanandJapanesecitiesduringWorldWarIIandVietnamesecitiesduringtheVietnamwar,seemstohavedonelittletounderminecities’long-runeconomicprospects(Brakman,Garretsen,andSchramm2004;DavisandWeinstein2002).Indeed,althoughflattenedin1945bytheatomicbombsthatforcedJapan’seventualsurrenderinWorldWarII,NagasakiandHiroshimahadreturnedtotheirprewarpopulationgrowthpathsby1960andthemid-1970s,respectively(DavisandWeinstein2002).92THRIVINGThisevidenceoflong-runresilienceofcitiestoshocksthatdestroylargeportionsoftheirphysicalcapitalismirroredbyevidenceoflong-runresiliencetoshocksthatdestroysignificantportionsoftheirhumancapitalstocks.Thus,inconsideringthepoten-tiallong-runimpactsoftheCOVID-19pandemiconcities,Glaeser(2022,10)notes,“TheBlackDeath,YellowFever,Choleraandthe1918–1919InfluenzaPandemicarethemostdirectantecedentsoftheCOVID-19pandemic.Yetthereislittleevidencethatanyoftheseterriblekillersdidmuchtodeterurbangrowth,atleastinthewest,andatleastsince1200CE.”Moregenerally,Glaeser(2022)arguesthat,inthelongrun,acitywilltendtorecover,intermsofbothpopulationandeconomicvitality,fromeventhemostdevastatingshockstoitsphysicalandhumancapitalstocks,provideditsunderlyingfundamentalsremainsound—inparticular,ifitremainsanattractivelocationforworkers,especiallyskilledworkers,andforbusinesses.Solongasthisisthecase,thedemandforurban(residentialandcommercial)floorspacewillbemorethansufficienttocoverthecostsofrebuilding.Andwithrebuildingcomestheopportunitytobuildbackbetter.Moreover,thedemandforurbanfloorspacewilllikelyremainsufficientlybuoyantinarapidurbanizingcontext,likethatcharacterizingthedevelopingworld.Climatechangewill,however,threatenacity’slong-runfortunesifitprecipitatesamajoreconomicorpoliticalshockThesefindingsoffergroundsforoptimismthat,fromalong-runperspective,citiesinlow-andmiddle-incomecountrieswillbeablenotonlytosurvivebutalsotothriveinthefaceoftheever-harderpunchestheycanexpectfrommothernaturebecauseofclimatechange.Themajorexceptionsmayoccurifclimatechangemakesacity’sclimatesounpleasantoritslocationsohazardousthatlargenumbersofpeoplenolongerwishtolivethere.Thecity’sdeclinewould,then,needtobemanaged.OnehistoricalexampleofthispossibilityisPortRoyalinsoutheasternJamaica,whichwasoncethelargestcityintheCaribbean.Itisnowknowntopostmedievalarchaeologistsasthe“citythatsank”followingadevastat-ingearthquakein1692thatliquifiedthegroundonwhichitwasbuilt(Lanthier2007).Atitsheight,justbeforetheearthquake,PortRoyalhadapopulationrangingfrom6,500to10,000(Buisseret1966;Claypole1972;PawsonandBuisseret1975;Taylor1688).In2011,itspopula-tionstoodatjust884.22Othermajorexceptionsmayoccurifclimatechangeitselfprecipitatesamajoreconomicorpoliticalshocktoacity.23Forexample,climatechangecouldleadtoalossofcompara-tiveadvantageintheproductionofthegoodsorservicesinwhichacityspecializes,oritmightexacerbateinequalitiestosuchanextentthatitleadstostrongpoliticaldiscordandmaybeevenviolence,which,inturn,leadstotheflightofsegmentsofacity’spopulation.Thecitiesthatfaceperhapsthegreatestriskoflossofcomparativeadvantagearethosethatspecializeintheextractionoffossilfuels,suchasthecoalminingtownsofeasternEurope,whichcanexpecttobedirectlyhitbytheenergytransition(seechapter5forfurtherdiscussion).TheStylizedRelationships93Theshort-tomedium-termimpactsofclimatechange–relatedshocksoncitiesmaybesignificantand,formanyhouseholds,devastatingTheobservationthatcitieshavetendedtoexhibitremarkablelong-runresilienceisnottodeny,however,thatnegativeshocksassociatedwithclimatechangecancausepotentiallysevere,aggregateeconomiclossestoacityintheshorttomediumterm.Norisittodenythat,inthefaceofanincreasedfrequencyofshocks,theselosseswillaccumulateovertimeunlessacitybecomesmorefullyadaptedtosuchshocks.Furthermore,althoughacitymaysurviveandeventhriveinresponsetooneormoreclimatechange–relatedshocks,householdsmaysufferdevastatinglossesfromanygivenshock.Afullanalysisofthefactorsthatdetermineresilienceatthehouseholdandneighborhoodlevelisbeyondthescopeofthisreport.Itispossible,however,tooffersomenewempiricalevidenceonhowshort-runaggregateeconomicresiliencevariesacrossthisreport’sglobalsampleofcitiesinthefaceofdifferenttypesofweatheranomaly,thefrequencyandintensityofwhichhavebeenevolving—andwillcontinuetoevolve—withclimatechange.ThisevidencecomesfrombackgroundresearchforthisreportbyParkandRoberts(2023),whoexaminetheimpactsofvarioustypesofweatheranomalyonacity’slevelofeconomicactivity.UsingdatafortheperiodApril2012–December2020,theirresearchanalyzeshowaweatheranomaly(asdefinedinbox1.2)inanygivenmonthaffectsacity’saggregateeconomicactivityinthatsamemonth.Becauseoftheabsenceofhigh-frequencyGDPdataforcities,theresearchfollowsthenowstandardpracticeofusingtheintensityofacity’snighttimelightsasaproxyforitsaggregatelevelofeconomicactivity.24Whenconfrontedbyabnormalweatherevents,bothsmallandlargecitiesinlow-andlower-middle-incomecountriessuffermorethancitiesinupper-middle-andhigh-incomecountriesTable1.1summarizestheestimatesbyParkandRoberts(2023)oftheimpactsoffivedifferenttypesofweatheranomalies—hot,cold,wet,anddryanomalies,andtropicalcyclones—onacity’snighttimelights.Intheiranalysis,ParkandRobertsdistinguishbetweenestimatedimpactsforsmallandlargecitiesinallincomegroups,withasmall(large)citydefinedasonewhosepopulationin2015wasbelow(above)theglobalmedianof371,885.25Inthecasesofhot,cold,wet,anddryanomalies,thetablealsodistinguishesbetweentheestimatedimpactsofanomaliesofanysize(Extreme=—)andthoseofextremeanomalies(Extreme=Y).Fortheanomaliesofanysize,thecellsreporttheestimatedimpactofa1-standard-deviationchangeintherelevantweathervariable(temperatureorprecipitation)fromacity’slong-runaverageforthatvariableonitsnighttimelightintensity.Forextremeanomalies,thetablereportstheestimatedimpactofa2-standard-deviationorgreateranomaly.Finally,fortropicalcyclones,thecellsreporttheestimatedimpactofacategory2orstrongercyclonethatoccurswithin200kilometersofacity’sgeographiccenter.94THRIVINGParkandRobertsestimatethatalltypesofweatheranomalies,exceptforcoldanomalies,willhavenegativeimpactsonacity’lightsinthemonthinwhichtheanomalyoccurs—inbothsmallandlargecitiesinlow-andlower-middle-incomecountries,andregardlessofwhetheralloronlyextremeanomaliesareconsidered.Furthermore,exceptforwetanomaliesinsmallcitiesinthesecountries,allestimatednegativeimpactsarestatisticallysignificant.Theestimatednegativeimpactsoftropicalcyclonesareparticularlylarge—acategory2orstrongercyclonereducesnighttimelightintensityby21percentinasmallcityinlow-andlower-middle-incomecountriesandby30percentinalargecityinthisincomegroup.Extremedryshocksalsohavequiteseverenegativeestimatedimpactsonlocaleconomicactivity—dimmingasmallcity’slightsinalow-orlower-middle-incomecountryby6.5percentandthoseinalargecityinsuchacountryby5.2percent.Bycontrast,bothwetandhotanomalies,evenifextreme,areesti-matedtohavemuchsmaller,althoughstillstatisticallysignificant,negativeimpacts.Inmarkedcontrasttotheresultsforcitiesinlow-andlower-middle-incomecountries,theestimatedimpactsofweatheranomaliesoneconomicactivityforcitiesinupper-middle-andhigh-incomecountriesaregenerallystatisticallyinsignificantand/ornonnegative.ThemajorTable1.1Estimatedimpactsofvarioustypesofweatheranomalyontheintensityofacity’snighttimelights,bycountryincomegroup,April2012–December2020TypeofanomalyExtremeAllcitiesCities,developingcountriesCities,developedcountriesSmallcitiesLargecitiesSmallcitiesLargecitiesSmallcitiesLargecitiesHot—–0.007–0.008–0.009–0.015–0.003–0.003Y–0.005–0.009–0.017–0.0240.0080.004Cold—0.0250.0160.0240.0050.0260.029Y0.0720.0420.0790.0250.0550.061Wet—–0.002–0.004–0.001–0.004–0.005–0.004Y–0.025–0.021–0.024–0.021–0.026–0.020Dry—–0.009–0.004–0.030–0.0150.0130.005Y–0.034–0.053–0.065–0.0520.026–0.051Cyclone—–0.12–0.09–0.21–0.30–0.05–0.000Significantat1%5%10%Sources:DerivedfromParkandRoberts2023.ResultsarebasedontheanalysisofmonthlycompositesofnighttimelightsderivedfromVIIRS(VisibleInfraredImagingRadiometerSuite)satellitedata(https://payneinstitute.mines.edu/eog-2/viirs/),monthlyweatherdatafromClimatologyLab,TerraClimate(https://www.climatologylab.org/terraclimate.html),andtropicalcyclonedatafromInternationalBestTrackArchiveforClimateStewardship(https://www.ncdc.noaa.gov/ibtracs/).Note:Thenumbersinthecellsindicatetheestimatedimpactofa1-standard-deviationchangeineachweathervariablefromitshistoricalnorm;forextremeanomalies(Extreme=Y),thenumbersindicatetheestimatedimpactiftheweathervariabledeviatesby2standarddeviationsormorefromitshistoricalnorm.“Extreme=—”indicateshot,cold,wet,anddryanomaliesofanysize;theimpactsoftropicalcyclonesareestimatedusingadummyvariable,whichtakesthevalueof1ifacityexperiencedacategory2orstrongercyclone(basedontheSaffir-Simpsonwindscale)within200kilometersofitscenter.“Cities,developingcountries”areincountriesclassifiedbytheWorldBankaseitherlow-orlower-middle-income.“Cities,developedcountries”arethoseincountriesclassifiedbytheWorldBankaseitherupper-middle-incomeorhigh-income.Small(large)citiesarethosewitha2015populationbelow(above)thesamplemedian.TheStylizedRelationships95exceptionisextremewetshocks.Ifalargecityinanupper-middle-orhigh-incomecountryexperiencesextremerainfallduringamonth,theintensityofitslightsforthatmonthisreducedbyanestimated2percent.Forasmallcityinthisincomegroup,theestimatedreduc-tionislarger—2.6percent.Theseestimatednegativeimpactsofextremerainfallarecompara-bletothoseforcitiesinlow-andlower-middle-incomecountries.Thefactthat,ingeneral,weatheranomalieshavemuchmoresevereestimatednegativeimpactsforcitiesinlower-thaninhigher-incomecountriesisconsistentwithagreaterlevelofresilienceonthepartofthelatter.Theresultsprovidejustapartialpictureofresilience,however,becausetheyconsideronlytheimmediateimpactsofaweathershockandremainsilentonthesubsequentpathofrecoveryofeconomicactivityorlackthereof.Nevertheless,relatedresearchthatalsouseslightsdatasuggestsaquickerreboundofeconomicactivityforacityinanupper-middle-orhigh-incomecountrythanforacityinalow-orlower-middle-incomecountry.Thus,althoughacityinahigher-incomecountrymayrestoreeconomicactivitytoitspreshocklevelwithinonemonth,acityinalower-incomecountryneedstwomonths(Gandhietal.2022;Lalletal.forthcoming).Inturn,theirgreaterresilienceislikelyduetoinvestmentsinbasicservicesandurbaninfrastructure,nottomentionemergencypreparedness,thatleavecitiesinhigher-incomecountriesbetterplacedtohandleespeciallyextremeanomalies.Thisfindingsuggeststhatbotheconomicdevelopmentandinclusivenessarevitaltobuildingresilience.Interestingly,andincontrasttoallothertypesofweatheranomaly,ParkandRoberts(2023)estimatetheimpactsofcoldanomaliestobepositiveforsmallcitiesinbothlower-andhigher-incomecountries,regardlessofsize.Forcitiesinlow-andlower-middle-incomecountries,theirbaselineclimatemattersInlower-incomecountries,citieswithwarmerbaselineclimatesdrivetheestimatedpositiveimpactofcoldanomaliesonurbaneconomicactivity.Thus,inhotcitiesacoldanomalyleadstoasignificantbrighteningoflightsinthesamemonth,whereassuchananomalyhasnosig-nificantimpactonthelightsofcitieswithcoldbaselinetemperatures.Thisresultisconsis-tentwiththeideathatcoldanomaliesmaketheweatherinhotcitiesunusuallypleasant,anditseemsplausiblethatthisbenefitcouldstimulatemoreeconomicactivity.However,becauseclimatechangehascontributedtoadecliningfrequencyofcoldanomaliesacrosscitiesgloballysincethe1970s,cooler,morepleasanteveningsincitiesinlower-incomecountrieswithhotbaselineclimateshavebecome,andwilllikelycontinuetobecome,increasinglyrare.Moregenerally,andespeciallyforcitiesinlow-andlower-middle-incomecountries,weatheranomaliesareestimatedtohavemoresevereimpactsoneconomicactivitywhentheygointhesamedirectionasacity’sbaselineclimate.Thus,hot,wet,anddryanomalieshavelargernegativeimpactsincitieswithhot,wet,anddrybaselineclimates—withimpactsparticu-larlyevidentfordryanomaliesincitiesinlower-incomecountrieswithdrybaselineclimates(figure1.15).Becausebothdryandhotanomalieshaveincreasedinfrequencyovertime,theseresultsareespeciallyworryingforcitieswithdryorhotbaselineclimatesintheseincomegroups.Unlessthesecitiescanbetteradapttodryandhotanomalies,theywillaccumulategreaterlossesinaggregateeconomicactivityovertime.However,theanalysisbyParkandRoberts(2023)doesprovidesomecautiousreasonsforoptimismonthisfront.Implicitintheconstructionoftheirweatheranomalyvariable,andthereforeintheirresults,istheideathatcitieswilladjustovertimetolong-runchangesintheirclimates.Theirvariabledefinesanomaliesrelativetoacity’sownlong-runaveragetemperature/precipitationlevel,whilealso96THRIVINGconsideringthecity’sownlong-runvariabilityaroundthataverage.Asclimatechangeshiftsthelong-runaverageandlevelofvariability,themeaningofananomalyalsochanges.Akeyquestion,however,ishowfastsuchadaptationoccursandwhetherpolicycandoanythingtospeeditup.Chapter5picksuponthisquestion.Howinclusivearecitiestoday?InclusionfiguresprominentlyintheWorldBank’sGRID(green,resilient,andinclusivedevel-opment)strategylaunchedin2021inresponsetothechallengesposedbyclimatechange,themoregeneraldegradationoftheenvironment,andotherfactors(WorldBank2021).Thissectionfocusesontheinclusiondimensionofacity’sdevelopment,recognizingthat,asmentionedatthechapter’soutset,acity’sgreennessandresiliencearelinkedtoitslevelofinclusivenessandviceversa.Inclusionhasmanyimportant,andfrequentlyoverlapping,dimensions.Theyincludebutarenotlimitedtogender,disabilitystatus,socioeconomicstatus,employmentstatus,religion,age,sexualorientation,residencyorcitizenshipstatus,race,maritalstatus,ethnicity,Sources:DerivedfromParkandRoberts2023.ResultsarebasedontheanalysisofVisibleInfraredImagingRadiometerSuite(VIIRS)nighttimelightsmonthlycomposites(https://payneinstitute.mines.edu/eog-2/viirs/)andmonthlyweatherdatafromClimatologyLab,TerraClimate(https://www.climatologylab.org/terraclimate.html).Note:Eachbarshowstheestimatedimpactofa1-standard-deviationweatheranomaly,whereanomaliesarecalculatedfollowingthemethodologydescribedinbox1.2.Eachverticalblacklineindicatestheupperandlowerboundsofthe95percentconfidenceintervalassociatedwiththecorrespondingestimatedimpact.Hot(cold)citiesareinthetop(bottom)halfoftheglobaldistributionoflong-runmeanmonthlytemperature.Wet(dry)citiesareinthetop(bottom)halfoftheglobaldistributionoflong-runmeanmonthlyprecipitation.Figure1.15Estimatedimpactsofhot,cold,wet,anddryanomaliesoncitiesinlow-andlower-middle-incomecountrieswithhot,cold,wet,anddrybaselineclimates,April2012–December2020-0.08-0.06-0.04-0.020.000.020.04HotColdWetDryEstimatedimpactsTheStylizedRelationships97andlocation.Unfortunately,lackofdatamakesitdifficulttoconsideranddojusticetoallthesedimensionsinthisreport.Thus,thissectionfocusesmainlyonfivemeasuresofinclu-sionforwhichcity-leveldataareavailable:poverty,accesstobasicservices,spousalviolenceasanindicatorof(lackof)genderinclusion,intracityhouseholdincomeinequality,andhealthoutcomes.Itacknowledges,however,thatotherimportantaspectsofinclusionareeithernotwellcoveredornotcoveredatall.Spotlight1(whichappearsafterchapter2)presentsamoredetailedanalysisofinclusioninPeruviancities,showcasinganinnovativemultidimensionalapproachtoassessinginclusion.Chapter3,aspartofitsmoregeneraldis-cussionoftheimpactsofclimatechangeoncities,discussestheresultsofvariousanalysesthatrelatetoinclusion.Theyincludeananalysisofthegender-differentiatedimpactsoftropicalcyclonesondisplacementfromcitiesandhowvariousclimatechange–relatedstress-orsaffectacity’sareaswithslumsversusareaswithoutslumsandneighborhoodswithdiffer-entsocioeconomiccharacteristics.Infocusingonpoverty,accesstobasicservices,spousalviolence,intracityhouseholdincomeinequality,andhealthoutcomes,thissectiondrawsonworkbyavarietyofauthors—Combesetal.(2022);FerreyraandRoberts(2018);HendersonandTurner(2020);andRoberts,GilSanderandTiwari(2019).ExceptforFerreyraandRoberts(2018)andRoberts,GilSander,andTiwari(2019),theseauthorsdefineurbanareasusingthedegreeofurbanizationapproach(seebox1.1).Insteadofconsideringdifferentsizesofurbancenters(cities),however,theiranalysesfocusonthedifferencesbetweencities,lessdenselypopulatedandsmallerurbanclusters(townsandsuburbs),andruralareas.Thesamplesconsideredarealsonotfullyglobal,whichsuggestscautioningeneralizingthederivedinsights.Box1.9presentsamoregloballycomprehensiveanalysisoflevelsofsocialinclusion,buttheanalysisfocusesonlyonaggregateurban–ruraldifferences.Theinclusionmeasuresconsid-eredhererepresentmoretheoutcomesoftheinteractionofagglomerationeconomiesandurbanstresses,andhowwellthesestressesaremanaged,thanthestressesthemselves.Acity’spovertyrateoffersanobviousexample.Aconceptrelatedtobutdistinctfromthatofsocialinclusionissocialcohesion.Box1.10presentsanalysisofaggregatedifferencesinsocialcohesion(whichmayalsoconditiontheimpactsofclimatechange)betweenurbanandruralareasacrosscountries.LargerandmoredevelopedurbanareashavelowerpovertyratesAsrevealedinfigure1.16,allsevenlow-andlower-middle-incomecountries—Angola,Bangladesh,theArabRepublicofEgypt,Ethiopia,Ghana,Tanzania,andVietnam—analyzedbyCombesetal.(2022)havelowerpovertyratesinurbanthaninruralareasbasedonthesamegloballyconsistentdefinitionofcitiesusedelsewhereinthisreport.Thefigureshowsthat,atboththeextremeglobalpovertylineofUS$1.90adayandthemiddle-incomeglobalpovertylineofUS$3.20aday,povertyratesarelowerinmoredenselypopulatedlargercities(urbancenters)thaninlessdenselypopulatedsmallertownsandsuburbs(urbanclusters).Inturn,townsandsuburbshavelowerpovertyratesthanruralareas.TheoneexceptiontothispatternisEgypt,wheretheUS$1.90-a-daypovertyrateisslightlyhigherforcitiesthanfortownsandsuburbs.26Althoughtheyusenationaldefinitionsofbothurbanareasandpovertyratesratherthangloballyconsistentdefinitions,Ferre,Ferreira,andLanjouw(2012)alsofindclearevidencethatpovertyratestendtobelowerinlargerurbansettlementsinafurthereightlow-andmiddle-incomecountries:Albania,Brazil,Kazakhstan,Kenya,Mexico,Morocco,SriLanka,andThailand.Meanwhile,Roberts,GilSander,andTiwari(2019)provideevidencethatin98THRIVINGComparinglevelsofsocialinclusioninurbanversusruralareasTheanalysisinthisreportrevealsthatmorepopulousurbanareastendtobemoreinclu-siveonseveralindicators:povertyrate,accesstovarioustypesofbasicservices,andbasichealthoutcomessuchasinfantmortality.Theanalysis,however,isbasedonlyonvarioussubsamplesofcountriesoronselectedregions.Foramorecomprehensiveglobalanalysis,thisboxpresentstheresultsfromtheconstructionofanewindexofsocialinclusionthatcaptures,forasampleof79countries,accesstomarkets;servicessuchasfinancialones,whichmayaffectaperson’sabilityto(self-)insureagainstclimatechange–relatedshocksandstresses;anddigitalspaces,whichcould,forexample,beimportantinhelpingspreadinformationonclimatechange–relatedhazards.ThisindexleveragestheSocialSustainabilityGlobalDatabaserecentlydevelopedbytheWorldBank.aAlthoughthedatadonotpermittheconstructionoftheindexforindividualcitiesorsizeclassesofcities,ordisaggregationforspecificvulnerableindividualsandgroups,theydoallowanalysisofoverallurban–ruraldifferencesacrosscountriesforcirca2020andhowthesedifferencesvarywithacountry’slevelofdevelopment.Moreprecisely,foranygivencountrythesocialinclusionindextakesavalueofbetween0and1,withhighervaluessignifyinghigheraveragelevelsofaccesstomarkets,services,anddigitalspaces.Theindexisconstructedseparatelyforeachcountry’surbanandruralpopulations,andisbasedontheequalweightingofsixindicators:laborforceparticipationrate,shareofthepopulationwithabankaccount,shareofhouseholdswithaccesstoadequatesani-tation,shareofhouseholdswithaccesstoelectricity,secondaryschoolenrollmentrate,andshareofhouseholdswithaccesstotheinternet.bAsfigureB1.9.1,panela,shows,theindexindicateshighersocialinclusioninurbanthaninruralareasinallWorldBankregions.Morespecifically,accesstomarkets,services,anddigitalspacesis,onaverage,higherinurbanareasthaninruralareas.Itisinthatsensethaturbanareasarefoundtobemoreinclusive.ThisurbanadvantageislargestinSub-SaharanAfricaandSouthAsia,andsmallestinEuropeandCentralAsiaandtheMiddleEastandNorthAfrica.Consistentwiththisfinding,theurbanadvantagetendstodeclinewithacountry’sdevelopmentlevel(figureB1.9.1,panelb)because,ascountriesdevelop,theprovisionofbasicservicestendstospreadincreasinglyfromurbantoruralareas,leadingtocatch-upevenwhileprovisioncontinuestoimproveincities(WorldBank2009).Whatdrivestheoverallgapbetweenurbanandruralareasglobally?Itturnsoutthathigheraccesstointernet,electricity,andadequatesanitationactsasthemaindriver.Differencesinfinancialinclusionand(secondary)educationbetweencitiesandruralareasarelessmarked.Ruralpopulationsreporthigheraccesstolabormarketsascapturedbylabormarketparticipationrates.Thefactthathigheraccesstointernet,elec-tricity,andsanitationdrivestheoverallgapisconsistentwiththeloweraveragecostsofsupplyinginfrastructurenetworksindenselypopulatedareas,whichcanspreadoveragreaternumberofusersthelargeup-frontfixedcostsofconstructingsuchnetworks.Thespreadingoftheselargeup-frontfixedcostsisasourceofagglomerationeconomiesthatbenefitsnotonlyproductivitybutalsosocialinclusion.Box1.9TheStylizedRelationships99Box1.9continuedSource:WorldBankcalculationsbasedondatafromWorldBank,SocialSustainabilityGlobalDatabase;WorldBank2022a.AlsoseeWorldBank2022b.Note:Theverticalaxisinpanelaisthevalueofthesocialinclusionindex;inpanelbitisthedifferenceinthevalueoftheindexinacountry’surbanandruralareas.Thesocialinclusionindexiscalculatedastheaverageofsixindicators:laborforceparticipationrate,shareofpopulationwithabankaccount,shareofhouseholdswithaccesstoadequatesanitation,shareofhouseholdswithaccesstoelectricity,secondaryschoolenrollmentrate,andshareofhouseholdswithaccesstotheinternet.Theindexisconstructedwiththelatestvalueforeachindicatorbetween2016and2020.TheGDPpercapitavaluesinpanelbarefor2020.Thesolidanddottedlinesinpanelbrespectivelyindicatethefittedlineand95percentconfidenceintervalsbasedonallcountriesincludedinthecalculation.PPP=purchasingpowerparity.a.WorldBank,SocialSustainabilityGlobalDatabase2022a.b.Resultsarerobusttothenumberofindicatorsusedandtoalternativechoicesoftheindicators.SeeCuesta,Madrigal,andPecorari(forthcoming)foradetaileddiscussionoftheserobustnesschecks.FigureB1.9.1Socialinclusionindexforurbanandruralareas,bygeographicregion,andrelationshipbetweenurban–ruralgapinsocialinclusionindexandlevelofdevelopmentacrosscountries,circa202000.20.40.60.8EuropeandCentralAsiaLatinAmericaandtheCaribbeanEastAsiaandPacificGlobalMiddleEastandNorthAfricaSub-SaharanAfricaSouthAsiaUrbanRural00.10.20.378910GDPpercapita,PPP(log)Low-incomeLower-middle-incomeUpper-middle-incomea.Socialinclusionindex,bygeographicregionb.Relationshipbetweenurban–ruralsocialinclusionindexgapandGDPpercapitaSocialinclusionindexSocialinclusionindexgap100THRIVINGCitiestendtohaveslightlylowerlevelsofsocialcohesionthanruralareasAlthoughthisreportfocusesonissuesofinclusion,adistinctbutrelatedconceptisthatofsocialcohesion—thatis,asenseofsharedpurpose,trust,andwillingnesstocooperateamongmembersofagroup,betweenmembersofdifferentgroups,andbetweenpeopleandgovernment.Morecohesivesocietiesareexpectedtobebetterpreparedtowithstandclimatechange–relatedshocks,avoidconflict,redistributeincomeandwealthtowardvulnerableandmarginalizedpopulations,andleavenoonebehind(Chatterjee,Gassier,andMyint,forthcoming).Asocialcohesionindexwasconstructedtocomparethesocialcohesionofurbanandruralpopulationsacrossaglobalsampleofcountries.Thisindexusestheequalweightingoffiveindicatorsthatcapturetheextenttowhichresidentstrusteachother,trusttheirgovernment,tolerateminorities,havenotbeenthevictimofacrime,andactivelypartic-ipateinpoliticalandcivicspaces.aTheindextakesavaluebetween0and1,withhighervaluesindicatingmorecohesivesocieties.Theresultsshowthat,inareversalofthepatternseenforsocialinclusion(seebox1.9),ruralareasareslightlymorecohesivethanurbanareasbothgloballyandinallWorldBankregions(figureB1.10.1,panela).SocialcohesioninruralareasexceedsthatincitiesbythelargestproportionintheEastAsiaandPacificregionandSub-SaharanAfrica,andbythesmallestproportionintheMiddleEastandNorthAfrica.Ruralareashave,onaverage,higherconfidenceingovernmentsandresidentswhoaremorelikelytoparticipateinelections,tobemembersofanorganization,andtoreportlowervictimizationrates.Bycontrast,residentsinurbanareashaveslightlyhighertrustinpeopleandtoleranceof(racial)minorities.Furthermore,associetiesbecomericherandmoreurbanized,theruralsocialcohesionedgediminishes(figureB1.10.1,panelb).Thisfindingsuggeststhat,asacountrydevelops,itscitiesshouldalsobecomeatleastascohesiveasitsruralareas.Box1.10Indonesiapovertyratesarelowerinthecoresoflargemetropolitanareasthanintheirurban-izedperipheries,whoseratesare,inturn,lowerthanthoseinsmaller,lessdenselypopulatednonmetrourbanareas.Thetendencyforpovertyratestobelowerinlargerandmoredevelopedurbanareasoflower-incomecountriesisconsistentwiththetendencyoftheseareastoprovidebothmoreformalandhigher-payingjobs.Inturn,largerandmoredevelopedurbanareastendtoprovidebetterjobsbecause,comparedwithsmallerandlessdevelopedurbanareas,theytendtobehometohigher-valued-addedindustriesandbecauseagglomerationeconomiesmakeworkersintheseareasmoreproductive(Roberts,GilSander,andTiwari2019;WorldBank2009).Thuscitiesprovide“escalatorsoutofpoverty”forin-migrants(Glaeser2012).Chapter4assesseswhetherclimatechange–relatedshocksandstressesareslowingdowntheseescalators.Althoughthecross-countrydataonurbanpovertyratesarenotsufficienttoallowforamean-ingfulanalysisoftheircorrelationwithnationallevelsofGDPpercapita,nationalratesofoverallpovertytendtodeclinewithacountry’slevelofdevelopmentatboththeUS$1.90andUS$3.20globalpovertylines.Indeed,globallya1percentincreaseinacountry’slevelofurban-ization,whichhasastrongpositivecorrelationwithitsGDPpercapita,isassociatedwithaTheStylizedRelationships101Box1.10continuedSource:WorldBankcalculationsbasedondatafromWorldBank,SocialSustainabilityGlobalDatabase;WorldBank2022a.AlsoseeWorldBank2022b.Note:Theverticalaxisinpanelaisthevalueofthesocialcohesionindex;inpanelbitisthedifferenceinthevalueoftheindexinacountry’surbanandruralareas.Thesocialcohesionindexiscalculatedastheaverageoffiveindicators:shareofpopulationthat(1)saysmostpeoplecanbetrusted,(2)hasconfidenceingovernment,(3)votedinthelastnationalelection,(4)isanactivememberofanyorganization,and(5)hasnotbeenavictimofacrime.Theindexisconstructedwiththelatestvalueforeachindicatorbetween2016and2020.TheGDPpercapitavaluesinpanelbarefor2020.Thesolidanddottedlinesinpanelbrespectivelyindicatethefittedlineand95percentconfidenceintervalsbasedonallcountriesincludedinthecalculation.PPP=purchasingpowerparity.a.DataontheseindicatorscomefromWorldBank,SocialSustainabilityGlobalDatabase2022,https://worldbankgroup.sharepoint.com/:u:/r/teams/SSI-DataandAnalytics-WBGroup/Shared%20Documents/Global%20Dataset/Global%20database%20versions/Base%2003222022/final_24.dta?csf=1&web=1&e=Ym2tDU.AlsoseeWorldBank2022.FigureB1.10.1Socialcohesionindexforurbanandruralareas,bygeographicregion,andrelationshipbetweenurban–ruralgapinsocialcohesionindexandlevelofdevelopmentacrosscountries,circa2020UrbanRuralLow-incomeLower-middle-incomeUpper-middle-incomeHigh-incomea.Socialcohesionindex,bygeographicregionb.Relationshipbetweentheurban-ruralsocialcohesionindexgapandGDPpercapita00.20.40.6EastAsiaandPacificSouthAsiaGlobalEuropeandCentralAsiaMiddleEastandNorthAfricaSub-SaharanAfricaLatinAmericaandtheCaribbeanSocialcohesionindex–0.15–0.10–0.0500.057.58.59.510.5SocialcohesionindexgapGDPpercapita,PPP(log)102THRIVING1percentdeclineintheshareofitspopulationlivingonlessthanUS$3.20perday(Roberts,GilSander,andTiwari2019).Itseemslikely,therefore,thaturbanpovertyratesatbothlinesalsofallascountriesdevelop.ResidentsofcitiestendtohavebetteraccesstobasicservicesandbetterhealthoutcomesthandoresidentsoftownsandsuburbsNotonlydopovertyratestendtobelowerincitiesthanintownsandsuburbs—whosepovertyrates,inturn,tendtobelowerthanthoseinruralareas—butaccesstobasicservicesalsotendstobebetter,asdomanyhealthoutcomes.Forexample,asshowninfigure1.17,panela,householdsincitiesinLatinAmericaandtheCaribbean,SouthAsia,SoutheastAsia,andSub-SaharanAfricahavebetteraccesstosafelymanageddrinkingwaterthandohouseholdsintownsandsuburbs,which,inturn,havebetteraccessthanruralhouseholds(HendersonandTurner2020).27Meanwhile,infantmortalityratesinthefourregionstendtofollowareversepattern—lowerincitiesthanintownsandsuburbs,andlowerintownsandsuburbsthaninruralareas.ExceptionsareLatinAmericaandtheCaribbeanandSoutheastAsia,wherecitiesSource:Combesetal.2022,basedonWorldPopgriddedpopulationdatafor2015andhouseholdbudgetsurveydataforcirca2015.Note:Urbancenters(cities),urbanclusters(townsandsuburbs),andruralareasaredefinedusingthedegreeofurbanizationmethodologyoutlinedinbox1.1.BoththeUS$1.90andUS$3.20globalpovertylinesaredefinedin2011constantinternationalpricesatpurchasingpowerparityexchangeratesusingcountry-specificspatialpricedeflators.Figure1.16Povertyrates(US$1.90andUS$3.20)forthreetypesofurbanandruralareas,selectedcountries,2015020406080AngolaBangladeshEgypt,ArabRep.EthiopiaGhanaTanzaniaVietnamPovertyrate(%)UrbancenterUrbanclusterRuralUrbancenterUrbanclusterRural020406080100AngolaBangladeshEgypt,ArabRep.EthiopiaGhanaTanzaniaVietnamPovertyrate(%)a.US$1.90-a-dayglobalpovertylineb.US$3.20-a-dayglobalpovertylineTheStylizedRelationships103haveslightlyhigherinfantmortalityratesthandotownsandsuburbs.Thesamepatternsholdforaccesstoawiderangeofotherbasicservices,includingschooling,electricity,andimprovedsanitation,aswellasforchildhealthoutcomes,suchastheshareofchildrenforwhomdiarrheaisreported.Althoughtheyuseadifferentdefinitionofurbanareas,Roberts,GilSander,andTiwari(2019)likewisereportpatternsofbetteraccesstobasicservicesandhealthoutcomesinlargercitiesthaninsmalleronesinIndonesia.Suchpatternshavealsobeenreportedfornumerousotherlow-andmiddle-incomecountriesintheWorldBank’sseriesofUrbanizationReviews.28Usingoncemoretheexamplesofaccesstosafelymanageddrinkingwaterandinfantmortal-ity,figure1.18showsthataccesstobasicservicesandmanybasichealthoutcomesincitiestendtoimproveasdevelopmentoccurs.Thus,althoughforLatinAmericaandtheCaribbean,SouthAsia,SoutheastAsia,andSub-SaharanAfricatheshareofhouseholdsinacountry’scitieswithaccesstosafelymanageddrinkingwaterispositivelycorrelatedwithitslevelofGDPpercapita,therateofinfantmortalityincitiesisnegativelycorrelated.Source:WorldBankcalculationsusingdatafromHendersonandTurner2020anddownloadedfromhttps://doi.org/10.7910/DVN/YZ46FJ.TheunderlyingdatafromtheDemographicandHealthSurveyscoverdifferentsurveyyearsacrosscountriesfrom2010to2016.Note:Urbancenters(cities),urbanclusters(townsandsuburbs),andruralareasaredefinedusingthedegreeofurbanizationmethodologyoutlinedinbox1.1.Inpanela,safelymanageddrinkingwaterisdefinedasallimprovedwatersourcesthattakezerominutestocollectorareonthepremises.Improvedwatersourcesincludeallpipedwaterandpackagedwater,inadditiontoprotectedwellsorsprings,boreholes,andrainwater.Inpanelb,infantmortalityisdefinedaschilddeathwithinthreemonthsofbirthamongthosebornfromthreemonthstothreeyearsbeforethesurveywasadministered.Figure1.17Householdaccesstosafelymanageddrinkingwaterandinfantmortalityrates,byurbanandruralareas,selectedregions,circa2015RuralareasTowns/SuburbsCities020406080100Shareofhouseholds(%)01234Shareofchildren(%)Sub-SaharanAfricaSouthAsiaLatinAmericaandtheCaribbeanSoutheastAsiaSub-SaharanAfricaSouthAsiaLatinAmericaandtheCaribbeanSoutheastAsiaa.Householdswithaccesstosafelymanageddrinkingwaterb.Infantmortalityrates104THRIVINGCitiesinsomeregionstendtohavelowerprevalenceofspousalviolenceInSouthAsiaandSoutheastAsia,theshareofcurrentlymarriedwomenwhoreporthavingeverexperiencedspousalviolenceislowerincitiesthanintownsandsuburbs(figure1.19,panela).Inbothregions,alowershareofmarriedwomenintownsandsuburbsthaninruralareasalsoreportshavingeverexperiencedspousalviolence.Evenincities,however,reportedratesofspousalviolenceremainveryhigh.ThepatternoflessspousalviolenceinlargeranddenserareasreversesforLatinAmericaandtheCaribbean—thatis,theshareofmarriedwomenwhoreporteverhavingexperiencedspousalviolenceishigherincitiesthanintownsandsuburbs.Thatshareis,inturn,higherthanthatinruralareas.InSub-SaharanAfrica,theshareofmarriedwomenwhoreporteverhavingexperiencedspousalviolenceisroughlythesameincitiesasinruralareas,butmarriedwomenintownsandsuburbsreportlessspousalviolence.Source:WorldBankcalculationsusingdataonhouseholdaccesstosafelymanageddrinkingwaterandinfantmortalityfromHendersonandTurner2020anddownloadedfromhttps://doi.org/10.7910/DVN/YZ46FJ.TheunderlyingdatafromtheDemographicandHealthSurveyscoverdifferentsurveyyearsacrosscountriesfrom2010to2016;dataongrossdomesticproduct(GDP)andpopulationarefromtheEuropeanCommission’sGlobalHumanSettlement(GHS)UrbanCentreDatabaseR2019(https://ghsl.jrc.ec.europa.eu/ghs_stat_ucdb2015mt_r2019a.php).Note:Inbothpanels,GDPpercapitaandthepercentagevaluesarecalculatedforcitiesin40countries(3inSoutheastAsia,5fromLatinAmericaandtheCaribbean,3fromSouthAsia,and29fromSub-SaharanAfrica).PanelaexcludesCambodiabecauseoflimiteddata.Thetermcitiesreferstourbancentersasdefinedinbox1.1.Inpanela,safelymanageddrinkingwaterisdefinedasallimprovedwatersourcesthattakezerominutestocollectorareonthepremises.Improvedwatersourcesincludeallpipedwaterandpackagedwater,inadditiontoprotectedwellsorsprings,boreholes,andrainwater.Inpanelb,infantmortalityisdefinedaschilddeathwithinthreemonthsofbirthamongthosebornthreemonthstothreeyearsbeforethesurveywasadministered.ln=naturallog.Figure1.18Relationshipbetweenacity’slevelofdevelopmentanditsaccesstosafelymanageddrinkingwateranditsinfantmortalityrate,selectedregions,circa2015y=19.14x–89.07R2=0.380204060801005.56.57.58.59.5Shareofhouseholds(%)GDPpercapita(ln,2015)y=–0.53x+7.34R²=0.13024685.56.57.58.59.5Shareofchildren(%)GDPpercapita(ln,2015)b.Infantmortalityratesa.HouseholdswithaccesstosafelymanageddrinkingwaterTheStylizedRelationships105Ingeneral,theshareofwomenwhobelievewifebeatingisjustifiedforatleastonereasonislowerincitiesthanintownsandsuburbs,andlowerintownsandsuburbsthaninruralareas.AlthoughSoutheastAsiahasthelowestlevelofreportedspousalviolenceamongcurrentlymarriedwomen,ithasthehighestshareofwomenwhosaytheybelievewifebeatingisjusti-fied.Moregenerally,evenincitiestheshareofwomenwhobelievewifebeatingisjustifiedforatleastonereasonisshockinglyhigh,especiallyoutsideofLatinAmericaandtheCaribbean,rangingfrommorethan3in10womeninSub-SaharanAfricatomorethan5in10womeninSoutheastAsia.UsingdataonSouthAfrica,Bruederle,Peters,andRoberts(2017)estimatethata1-standard-deviationincreaseintemperatureiscorrelatedwithan8.6percentincreaseinsexualcrimes.Meanwhile,forMadrid,Spain,Sanz-Barberoetal.(2018)reportapositivecorrelationbetweenheatwaves(temperaturesover34°C)andtheincidenceofintimatepartnerviolence.Largercities,especiallyinlower-incomecountries,tendtohavehigherintracityhouseholdincomeinequalityInlow-andmiddle-incomecountries,incomeequalitybetweenhouseholdswithinacitytendstoincreasewithacity’ssize—seefigure1.20for16countriesinLatinAmericaandtheCaribbean(panela)andforIndonesia(panelb).SuchapositiverelationshipisconsistentwithagreaterconcentrationofhouseholdswithskilledworkersinlargercitiesthaninsmallerSource:WorldBankcalculationsusingdatafromHendersonandTurner2020anddownloadedfromhttps://doi.org/10.7910/DVN/YZ46FJ.TheunderlyingdatafromtheDemographicandHealthSurveyscoverdifferentsurveyyearsacrosscountriesfrom2010to2016.Note:Urbancenters(cities),urbanclusters(townsandsuburbs),andruralareasaredefinedusingthedegreeofurbanizationmethodologyoutlinedinbox1.1.Figure1.19Reportedlevelsofspousalviolenceexperiencedbycurrentlymarriedwomenandofwomenwhobelievewifebeatingisjustified,bytypeofurbanandruralarea,selectedgeographicregions,circa20150102030Sub-SaharanAfricaSouthAsiaLatinAmericaandtheCaribbeanSoutheastAsiaSub-SaharanAfricaSouthAsiaLatinAmericaandtheCaribbeanSoutheastAsiaShareofwomen(%)RuralareasTowns/SuburbsCities0204060Shareofwomen(%)a.Currentlymarriedwomenwhohaveeverexperiencedphysicalspousalviolenceb.Womenwhobelievewifebeatingisjustifiedforatleastonereason106THRIVINGcitiesbecauseofskill-selectivemigrationandbetterlearningoutcomesinlargercities.Thus,forbothIndonesiaandLatinAmericaandtheCaribbean,alargeportionofthehigherincomeinequalitythatexistsinlargercitiescanbeexplainedbytheincomegapbetweenskilledandunskilledhouseholds.Householdsinsmallcitiesconsistmainlyofrelativelyunskilledworkers;therefore,relativelylowincomesgenerallyprevail.Bycontrast,largecitieshaveamixofhouseholdswithbothskilledandunskilledworkers,leadingtohigherincomeinequal-ity.Moreover,theevidenceforIndonesiaandLatinAmericaandtheCaribbeansuggeststhatskilledworkersbenefitmorethanunskilledworkersintermsofbothproductivityandwagesfromagglomeration,whichfurtherreinforcesthedividebetweenthem(FerreyraandRoberts2018;Roberts,GilSanderandTiwari2019).Largecities,then,willlikelyhavegreaterincomeinequalitywhentheircountriesprovideunevenaccesstohigh-qualityeducationandwhenthesecitiespresenttheirresidentswithuneveneducationandlearningopportunities.Unequalaccesstobasicserviceswillalsolikelyfeedintounevenhumancapitalaccumulationincities,therebycontributingtohigherSources:Panela:FerreyraandRoberts2018,withcalculationsbasedondatafromSocio-EconomicDatabaseforLatinAmericaandtheCaribbean(SEDLAC)forcountriesotherthanBrazilandIntegratedPublicUseMicrodataSeries(IPUMS)forBrazil;panelb:Roberts,GilSander,andTiwari2019,withcalculationsbasedondatafromthe2017roundofIndonesia’sNationalSocio-EconomicSurvey(SUSENAS).Note:Panelashows,foreachsizeclassofcity,theweightedaverageGinicoefficientacrosscities,wheretheweightsareequaltocitysizes.Theclassificationofcitysizefollowsthecountry-specificthresholdsdefinedinFerreyraandRoberts(2018).Inpanelb,theunitofanalysisisdistricts.Formultidistrictmetroareas,theGinicoefficientandpopulationarecalculatedoveralldistrictsinthemetroarea.Figure1.20Relationshipbetweenincomeinequalityandcitysizein16LatinAmericanandCaribbeancountries,circa2014,andinIndonesia,20170.30.40.50.6BrazilColombiaParaguayBoliviaHondurasEcuadorGuatemalaCostaRicaArgentinaChilePeruNicaraguaElSalvadorUruguayMexicoDominicanRepublicGinicoecientSmallMediumLargey=0.0063x+0.25R2=0.0190.200.250.300.350.400.451012141618GinicoecientPopulation,2017(log)NonmetroareasMetroareasb.Indonesiaa.LatinAmericaandtheCaribbeanTheStylizedRelationships107incomeinequality.IntheUnitedStates,however,despitehigherincomeinequalityinlargercitiesthaninsmallercities(BehrensandRobert-Nicoud2015),therelationshipbetweeninequalityandcitysizeisnotasstrongasforeitherIndonesiaorcountriesinLatinAmericaandtheCaribbean(FerreyraandRoberts2018).Thisfindingstems,atleastinpart,fromthebettergeneralaccesstoeducationintheUnitedStatesandthehigherlevelsofaccesstobasicservicesthatfeedintohumancapitalaccumulationincities.Unevenaccesstobasicservicesandlearningopportunitiesnotonlycontributestopresent-dayinequalitywithincitiesbutalsocanundermineintergenerationalmobilityTwotypesofintergenerationalmobilitycanbedistinguished:absolutemobility,whichreferstotheextenttowhichchildrentendtobericherormoreeducatedthantheirparents,andrelativemobility,whichreferstotheextenttowhichchildrentendtooccupyahigherrungontheincomeorsocioeconomicladderthantheirparents.29Evidenceonintergenerationalmobilityissparseforlower-incomecountriesandgenerallyrestrictedtoeducationlevels.Buttheevidencethatdoesexistsuggeststhat,althoughtheshareofadultsinthosecountrieswhoaremoreeducatedthantheirparentshasincreasedovertime,absoluteintergenerationalmobilityremainshigherinhigh-incomecountries.Furthermore,althoughalmost62percentofadultsbornin1960inlower-incomecountrieshavemoreeducationthantheirparents,lessthan57percentofadultsbornin1980do.Thus,inlower-incomecountries,relativeinter-generationalmobilityhasdeclinedovertime.Asforwhethertheurbanareasofthosecoun-triesprovideforhigherlevelsofmobilitythantheruralareas,theevidenceismixed.InLatinAmericaandtheCaribbean,mobilitytendstobehigherinlargercities,whereasIndonesiahasnodifferenceinmobilitybetweenurbanandruralareas.Meanwhile,inIndiamobilityinurbanareasisslightlylowerthaninruralareas,althoughurbanareassawalargerimprovementinmobilityinthe1980sand1990s(Lalletal.forthcoming).Residentsinlargercitiessuffermorefromillnessesassociatedwithmodernlifestyles,andshortcomingsofurbanfoodsystemsmayexacerbatethisproblemFinally,inadditiontohavinghigherinequality,citiestendtohavegreaterprevalenceofillnessesassociatedwithmodernlifestylesthandotownsandsuburbs.Theseillnessesare,inturn,moreprevalentintownsandsuburbsthaninruralareas—seefigure1.21forevidenceonobesityandhypertension(HendersonandTurner2020).Thesepatternslikelyarisebecauseofthehigherincomesthattendtoprevailinlargerandmoredenselypopulatedurbanareas:higherincomesareoftenassociatedwithlesshealthydiets,includingtheconsumptionofmoremeatandmoreprocessedfoodproductswithhigherfatandsugarcontents.Higherconsumptionofmeat(especiallybeef)alsocontributestohighergreenhousegasemissions.Accordingtoonerecentstudy,theproductionof1kilogramofbeefresultsingreenhousegasemissionsthatare28timeshigherthanthoseassociatedwiththeproductionof1kilogramofwheat.Moregenerally,theuseofcows,pigs,andotheranimalsforfood,aswellaslive-stockfeed,accountsfor57percentofallfoodproduction–relatedgreenhousegasemissionsglobally,comparedwith29percentfromthecultivationofplant-basedfoods(Xuetal.2021).Shortcomingsinurbanfoodsystemsfurtherexacerbatetheprevalenceoflesshealthydietsincities,particularlyamongthepoor(box1.11).108THRIVINGSource:WorldBankcalculationsusingdatafromHendersonandTurner2020anddownloadedfromhttps://doi.org/10.7910/DVN/YZ46FJ.TheunderlyingdatafromtheDemographicandHealthSurveyscoverdifferentsurveyyearsacrosscountriesfrom2010to2016.Note:Urbancenters(cities),urbanclusters(townsandsuburbs),andruralareasaredefinedusingthedegreeofurbanizationmethodologyoutlinedinbox1.1.Figure1.21Obesity,bygeographicregion,andhypertensioninIndia,bytypeofurbanandruralarea,circa2015RuralareasTowns/SuburbsCities0102030b.Householdmembersage25andupwithhighbloodpressure,Indiaa.Obeserespondents,bygeographicregionSub-SaharanAfricaSouthAsiaLatinAmericaandtheCaribbeanSoutheastAsiaShareofobeserespondents(%)0102030RuralareasTowns/SuburbsCitiesShareofhouseholdmembers(%)UnhealthydietsandpoorlygovernedfoodsystemsforurbanconsumersThedemographicandsocioeconomicchangesassociatedwithurbanizationarecausinglargeshiftsindietaryandactivitypatterns.Urbanconsumerseatmorefruitsandvegetablesthanruralconsumersdo(figureB1.11.1),buttheyalsoconsumemorepro-cessedmeatsandsugarydrinks.Urbanconsumerswithlowersocioeconomicstatus(lowmaternaleducation)havethepoorestdiets.Theyhavelowlevelsofvegetableconsumption—almostaslowasthoseforruralconsumerswithlowersocioeconomicstatus.Atthesametime,theyhavehighlevelsofsodaconsumption—almostashighasthatforurbanconsumerswithhighersocioeconomicstatus.Globally,thecostofhealthydietscouldbeprohibitive—ahealthydietcostsUS$3.75perday,makingitoutofreachforabout3billionpeople(Herforthetal.2020).Atthesametime,unhealthyfoodsarelargelyaffordable.Thepricepercalorieoffatsandoils,sugar,softdrinks,andsaltysnacksis0.67,0.83,5.26,and2.54times,respectively,thepricepercalorieofabasketofstaples.Bycontrast,thepricepercalorieoffoodsrichinvitaminABox1.11TheStylizedRelationships109anddarkgreenleafyvegetablesis7.74and16.12times,respectively,thepricepercalorieofabasketofstaples(HeadeyandAlderman2019).Atthecountrylevel,urbanizationandimprovedconnectivitybetweenruralandurbanareaslowerthecostofnutrient-adequatediets(Baietal.2021).Withincountries,althoughhealthierdietscouldbemoreexpensiveinurbanareas,theyaretypicallymoreaffordabletourbanconsumerswithhigherincomesincountriessuchasBangladesh,India,andMyanmar(Dizon,Wang,andMulmi2021;Herforthetal.2020).Poorurbandietslargelyresultfrompoorgovernanceofurbanfoodsystems(Acharyaetal.2021).Suchsystemshavecomplexgovernance,involvingdisparateactors(suchastheministriesofagricultureandhealth,foodsafetyauthorities,andmarketdevelopmentauthorities)andabroadrangeofdecisions(on,forexample,landusezoning,waterandsanitation,wastemanagement,urbantransportandlogistics,andpublicinfrastructure).Giventheircomplexity,governancestructuresofurbanfoodsystemsareoftenweak.Asurveyof170citiesinSouthandEastAsiafindsthatmanyofthemexhibitreactive,fragmented,andexclusionaryfoodpolicies,whereasonlyafewhavefoodpoliciesthatareproactive,forward-looking,integrative,andinclusive(Acharyaetal.2021).Box1.11continuedSource:WorldBankstaffcalculationsbasedondatafromthe2018GlobalDietaryDatabase(GDD)(https://www.globaldietarydatabase.org/).Note:TheGDDprovidesintakesfor11foodcategoriesfor187countries.Itsdatacomefrommorethan300dietarynationallyrepresentativesurveysof1.75millionindividualsrepresenting89percentoftheglobaladultpopulation.g/d=gramsperday;WHO=WorldHealthOrganization.FigureB1.11.1Intakeoffoodgroups,byurbanandruralandmaternaleducation,187countries050100150200400FruitsVegetablesBeansandlegumesNutsandseedsWholegrainMilkFishUnprocessedmeatProcessedmeatSugarsweetnedbeveragesFruitjuicesGramsperdayperadultRural,loweducationUrban,loweducationRural,high/middleeducationUrban,high/middleeducationWHOrecommendationforsugarintake(<50g/d)WHOrecommendationforfruitsandvegetablesintake(≥400g/d)110THRIVINGSummaryandconclusionsThischapterhastakenstockofhowgreen,howresilient,andhowinclusivecitiesgloballyaretoday,revealingthatmanyindicatorsrelatetobothacity’ssizeanditslevelofdevelopment.AlthoughlessgreenintermsoftheirCO2emissions,levelsofairquality,andaveragevegeta-tionlevels,morepopulouscitiesareinmanyrespectsmoreinclusive,asevidencedbytheirlowerpovertyrates,betterlevelsofaccesstobasicservices,andbetteraverageoutcomesonmanyhealthindicators.Beyondacertainincomelevel,developmentalsocorrelateswithbothbetterairqualityandahigheraveragelevelofvegetation.Furthermore,moredevelopedcitiesaremoreresilienttoextremeweatherevents,thefrequencyandintensityofwhichare,exceptforextremecoldevents,increasingwithclimatechange.Theseresultssuggestthatacity’sresiliencetoclimatechange–relatedshocksandstressesanditseconomicdevelopmentgotogether,andthateconomicdevelopmentis,atleastbeyondsomelevel,alsoassociatedwithmorepositiveoutcomesonmanydimensionsofgreennessandinclusiveness.Anexceptionisacity’slevelofCO2emissionsandthereforeitscontributiontoclimatechange.Thus,forcitiesinlow-andlower-middle-incomecountries,thekeychallenge,fromaclimatechangeperspective,ishowtodeveloponalowerCO2emissionstrajectorythanthathistoricallyfollowedbytoday’scitiesinupper-middleandhigh-incomecountries.Althoughputtingthisresponsibilityonlower-incomecountriesmayseemunfair,thefactthat,forexample,lowerCO2emissionsarestronglycorrelatedwithbetterlocalairquality,which,inturn,carriespositivehealthandproductivitybenefits,suggeststhatthischallengecomeswithanopportunity.Byimplementingpoliciesthathelpimprovelocalairquality,citiesinlower-incomecountriescannotonlyavoidcalamitiessuchasLondon’sGreatSmogbutalsoperhapsfollowamoreclimate-friendly—andfaster—growthpath.Theresultsalsosuggestthatthegreenness,resilience,andinclusivenesschallengesthatcitiesfacevarywiththeirsizesandevolveascitiesdevelop.Thus,therecanbenoblanketpolicyprescriptions.Instead,policiesneedtobetailoredaccordingtobothacity’ssizeanditslevelofdevelopment,aswellastheclimatechange–relatedhazardsitfaces.TheStylizedRelationships111Annex1A:SpatialdistributionsofextremeweatheranomaliesSource:WorldBankcalculationsbasedonClimatologyLab,TerraClimate(https://www.climatologylab.org/terraclimate.html);EuropeanCommission,GlobalHumanSettlement(GHS)UrbanCentreDatabaseR2019(https://ghsl.jrc.ec.europa.eu/ghs_stat_ucdb2015mt_r2019a.php).Note:Anextremehotmonthisoneinwhichacity’stemperatureanomalyvariable(asdefinedinbox1.2)recordsavalueofatleast2.Frequencyisdefinedasthenumberofextremehotmonthsperyear.ThemappresentstheaverageannualfrequencyfromJanuary2011toDecember2020foreachcityintheglobalsample.Map1A.1Frequencyofextremehotmonthsperyear,2011–20Sources:WorldBankcalculationsbasedonClimatologyLab,TerraClimate(https://www.climatologylab.org/terraclimate.html);EuropeanCommission,GlobalHumanSettlement(GHS)UrbanCentreDatabaseR2019(https://ghsl.jrc.ec.europa.eu/ghs_stat_ucdb2015mt_r2019a.php).Note:Anextremehotmonthiswhenacity’stemperatureanomalyvalue(asdefinedinbox1.2)recordsavalueofatleast2.Intensityiscalculatedastheaveragesizeoftheanomalyvariableduringconsecutiveextremehotmonths.ThemapshowstheaverageintensityfromJanuary2011toDecember2020foreachcityintheglobalsample.Map1A.2Intensityofextremehotmonths,2011–20112THRIVINGSources:WorldBankcalculationsbasedonClimatologyLab,TerraClimate(https://www.climatologylab.org/terraclimate.html);EuropeanCommission,GlobalHumanSettlement(GHS)UrbanCentreDatabaseR2019(https://ghsl.jrc.ec.europa.eu/ghs_stat_ucdb2015mt_r2019a.php).Note:Anextremecoldmonthiswhenacity’stemperatureanomalyvariable(asdefinedinbox1.2)recordsavalueof–2orlower.Intensityiscalculatedastheaveragesizeoftheanomalyvariableduringconsecutiveextremecoldmonths.ThemapshowstheaverageintensityfromJanuary2011toDecember2020foreachcityintheglobalsample.Map1A.4Intensityofextremecoldmonths,2011–20Sources:WorldBankcalculationsbasedonClimatologyLab,TerraClimate(https://www.climatologylab.org/terraclimate.html);EuropeanCommission,GlobalHumanSettlement(GHS)UrbanCentreDatabaseR2019(https://ghsl.jrc.ec.europa.eu/ghs_stat_ucdb2015mt_r2019a.php).Note:Anextremecoldmonthiswhenacity’stemperatureanomalyvariable(asdefinedinbox1.2)recordsavalueof–2orlower.Frequencyisdefinedasthenumberofextremecoldmonthsperyear.ThemappresentstheaverageannualfrequencyfromJanuary2011toDecember2020foreachcityintheglobalsample.Map1A.3Frequencyofextremecoldmonthsperyear,2011–20TheStylizedRelationships113Sources:WorldBankcalculationsbasedonClimatologyLab,TerraClimate(https://www.climatologylab.org/terraclimate.html);EuropeanCommission,GlobalHumanSettlement(GHS)UrbanCentreDatabaseR2019(https://ghsl.jrc.ec.europa.eu/ghs_stat_ucdb2015mt_r2019a.php).Note:AnextremedrymonthiswhenthevalueoftheStandardizedPrecipitation-EvapotranspirationIndex(SPEI)is–1.5orbelow.Frequencyisdefinedasthenumberofextremedrymonthsperyear.ThemappresentstheaverageannualfrequencyfromJanuary2011toDecember2020foreachcityintheglobalsample.Map1A.5Frequencyofextremedrymonthsperyear,2011–20Sources:WorldBankcalculationsbasedonClimatologyLab,TerraClimatehttps://www.climatologylab.org/terraclimate.html;EuropeanCommission,GlobalHumanSettlement(GHS)UrbanCentreDatabaseR2019(https://ghsl.jrc.ec.europa.eu/ghs_stat_ucdb2015mt_r2019a.php).Note:AnextremedrymonthiswhenthevalueoftheStandardizedPrecipitation-EvapotranspirationIndex(SPEI)is–1.5orbelow.IntensityiscalculatedastheaveragevalueofSPEIduringconsecutiveextremedrymonths.ThemapshowstheaverageintensityfromJanuary2011toDecember2020foreachcityintheglobalsample.Map1A.6Intensityofextremedrymonths,2011–20114THRIVINGSources:WorldBankcalculationsbasedonClimatologyLab,TerraClimate(https://www.climatologylab.org/terraclimate.html);EuropeanCommission,GlobalHumanSettlement(GHS)UrbanCentreDatabaseR2019(https://ghsl.jrc.ec.europa.eu/ghs_stat_ucdb2015mt_r2019a.php).Note:AnextremewetmonthiswhenthevalueoftheStandardizedPrecipitation-EvapotranspirationIndex(SPEI)is+1.5orabove.Frequencyisdefinedasthenumberofextremewetmonthsperyear.ThemappresentstheaverageannualfrequencyfromJanuary2011toDecember2020foreachcityintheglobalsample.Map1A.7Frequencyofextremewetmonthsperyear,2011–20Sources:WorldBankcalculationsbasedonClimatologyLab,TerraClimate(https://www.climatologylab.org/terraclimate.html);EuropeanCommission,GlobalHumanSettlement(GHS)UrbanCentreDatabaseR2019(https://ghsl.jrc.ec.europa.eu/ghs_stat_ucdb2015mt_r2019a.php).Note:AnextremewetmonthiswhenthevalueoftheStandardizedPrecipitation-EvapotranspirationIndex(SPEI)is+1.5orabove.IntensityiscalculatedastheaveragevalueofSPEIduringconsecutiveextremewetmonths.ThemapshowstheaverageintensityfromJanuary2011toDecember2020foreachcityintheglobalsample.Map1A.8Intensityofextremewetmonths,2011–20Notes1.Urbanstressestraditionallyemphasizedbyeconomistsalsoincludenegativeexternalitiesassociatedwithurbandensity,suchasaheightenedriskofcrimeandmorerapidspreadofinfectiousdiseases(see,forexample,Glaeser2012).ThefactthaturbandensitymaygiverisetothemorerapidspreadofinfectiousdiseaseshasreceivedmoreattentionrecentlybecauseoftheCOVID-19pandemic.Despitethisattention,evidenceontheroleofdensityindrivingthepandemicismixed(FangandWahba2020;LallandWahba2020).2.Urbanandspatialeconomistsdistinguishbetweenthe“first-nature”and“second-nature”advantagesoflocations.First-natureadvantagesareassociatedwithalocation’sphysicalgeographyandclimate,andsecond-natureadvantagesarisefromagglomerationeconomies(Brakman,Garretsen,andvanMarrewijk2012;Krugman1991).First-natureadvantagestendtobetakenasexogenouslyfixed,independentoftheprocessesofurbangrowth,whereassecond-natureadvantagesaremodeledasbothacauseandaconsequenceofurbangrowth.3.Thisnewempiricalevidence,whichleveragesfinespatialresolutionandhighfrequencysatelliteimagery,isbasedonbackgroundresearchforthisreportbyParkandRoberts(2023).4.SeealsoHallegatteetal.(2017),amongothers.Theinterdependenciesbetweenhowgreen,howresilient,andhowinclusivesocietiesareisakeythemeoftheoriginalWorldBank(2021)paperthatintroducedtheGRID(green,resilient,andinclusivedevelopment)concept.5.TheEuropeanCommission’sUrbanCentreDatabaseR2019,onwhichthischapterheavilyrelies,providesdatafor10,303citiesafterdiscardingbothfalsepositiveanduncertaindetectionsofcities—formoredetails,seeFlorczyketal.(2019).Theexactsamplesizeusedinboththischapterandthereportmoregenerallyvariesaccordingtothevariablebeinganalyzed.6.Chapter2includesalookforwardathowclimateandtheassociatedhazardsthatcitiesfaceareprojectedtoevolveinthecomingdecades.7.Map1A.6showstheglobalspatialdistributionoftheintensityofextremedryeventswhentheydooccurfortheperiod2011–20.8.Maps1A.7and1A.8inannex1Ashow,respectively,theglobalspatialdistributionsofthefrequencyofextremeweteventsandtheintensityofextremeweteventswhentheydooccurfortheperiod2011–20.9.FromtheUSNationalOceanServicewebpage,“HowIsSeaLevelRiseRelatedtoClimateChange?”https://oceanservice.noaa.gov/facts/sealevelclimate.html.10.Noevidenceexiststhattropicalcyclonespersehavebeenincreasinginfrequency.Infact,agrowingliteraturesuggeststhattheoppositemightbemorelikelythecase(Chandetal.2022;Knutsonetal.2020).Evidencedoessuggest,however,thattheirtrackshavebeenmigratingtowardcoastsaswellaspoleward(WangandToumi2021).Althoughfurtherinvestigationisneeded,thismigrationpresumablyexplainstheincreasedexposureofcitiesgloballytotropicalcyclonesovertimeshowninfigure1.5.11.CustomdatawereacquiredviatheUnitedNations’WorldUrbanisationProspectis:2018Revisionwebsite,https://population.un.org/wup/.TheStylizedRelationships115THRIVING12.TheGHSUrbanCentreDatabasederivesitsCO2emissionsdatafromtheEuropeanCommission’sEmissionsDatabaseforGlobalAtmosphericResearch(EDGAR).Thedatacover“Allhumanactivitiesleadingtoclimaterelevantemissions...,exceptbiomass/biofuelcombustion(short-cyclecarbon)inthepower,industry,buildings,transportandagriculturalsectors,large-scalebiomassburningandlanduse,land-usechangeandforestry”(Crippaetal.2019,24).13.FollowingEDGAR,theGHSUrbanCentreDatabasedistinguishesfivesectors—agriculture,energy,industry,residential,andtransportation—fromwhichCO2emissionsoriginate.Residentialsectoremissionsincludeemissionsassociatedwithenergyforbuildingsandwaste.Allsectorsaredefinedusingthe1996codesoftheIntergovernmentalPanelonClimateChange(Florczyketal.2019).14.Citiesinupper-middle-incomecountriesintheglobalsamplehadanaggregatepopulationofapproximately1.3billionin2015,whereascitiesinhigh-incomecountrieshadanaggregatepopulationofabout0.57billion.15.Acity’scompactnessismeasuredusingthePolsby-PopperRatio,whichrangesfrom0to1,with1indicatingthemaximallevelofcompactness.ThePolsby-PopperRatioisequaltotheratioofacity’sactualareatothatofacirclewithcircumferenceequaltotheperimeterofthearea(PolsbyandPopper1991).16.FromtheUnitedNationsEconomicCommissionforEurope’swebpage,“ImprovingAirQualityWhileFightingClimateChange,”https://unece.org/unece-and-sdgs/improving-air-quality-while-fighting-climate-change.17.Resultssimilartothosereportedinthissectionholdforbothacity’spopulationandGDPpercapitaofthecountryinwhichacityislocated,ifoverallPM2.5concentrationsasopposedtoPM2.5emissionsfromtheresidentialandtransportationsectorsareexamined.18.EllisandRoberts(2016)revealthatthisrelationshipisstrongerforSouthAsiancitiesthanforothercitiesinlow-andmiddle-incomecountries.Theyattributethisfindingtotheparticularly“messy”natureofurbanizationinSouthAsia,whichhasseveresprawlandveryhighlevelsoftrafficcongestion.19.Equally,worseurbanairqualityshouldnotbeconsideredanautomaticoutcomeofacountry’sevolutionfromlow-tolower-middle-incomestatus.Whetherurbanairqualitydeterioratesorimprovesasanygivencountrydevelopswilldepend,atleasttoacertainextent,onpolicychoicesatthenationalandlocallevelswithinthecountry.20.Formoredetailsonthismeasureofaveragegreenness,seeCorbaneetal.(2020)andFlorczyketal.(2019).Inadditiontoaveragegreenness,chapter2alsoanalyzestheshareofacity’sareathatis“highgreen”—thatis,coveredbydensevegetationcorrespondingtogardens,parks,andurbanforests.21.FromtheUSNationalParkServicewebpage,“PlantsandClimateChange,”https://www.nps.gov/articles/000/plants-climateimpact.htm.22.Basedoncensusdatafromcitypopulation.de.23.In1431,Angkor,thecapitaloftheKhmerempire,beganarapiddecline.ThedeclinefollowedthesackingandlootingofthecitybyAyutthayainvaders,butitalsocoincidedwithaseriesofnaturaldisastersandatimeofdrasticclimatechange(suchasmonsoonrainsfollowedbydrought).Althoughthesackingandlootingofthecityseemunassociatedwiththedisasters116THRIVINGandclimatechangeofthetime,thisexampledoesdemonstratethepossibleriskstoacity’slong-runfortunesshouldclimatechangeitselfsparkamajoreconomicorpoliticalshock.24.Theuseofnighttimelightstoproxyforacity’slevelofeconomicactivityhasbecomewidespreadsincetheseminalarticlesbyHenderson,Storeygard,andWeil(2011,2012),demonstratingthatmovementsinnationalGDParestronglycorrelatedwithmovementsinnighttimelightintensity.25.TheanalysisbyParkandRoberts(2023)isrestrictedtocitiesthathada2015populationofatleast200,000.26.When,insteadofdegreeofurbanization,Combesetal.(2022)adoptanalternative“dartboard”methodologyforthegloballyconsistentdefinitionofurbanareas,theyalsofindevidenceoflowerpovertyratesincitiesthanintownsandsuburbs,whoseratesare,inturn,lowerthanthoseinruralareas.Theirresultsarealsorobustwhendifferentgriddedpopulationdatasetsserveasinputinthedefinitionofurbanareas.27.Datausedforthecalculationsinthissectioncover3countriesinSoutheastAsia(Cambodia,Myanmar,Timor-Leste);5inLatinAmericaandtheCaribbean(Colombia,DominicanRepublic,Guatemala,Haiti,Honduras);3inSouthAsia(Bangladesh,India,Nepal);and29inSub-SaharanAfrica(EastAfrica:Burundi,Comoros,Ethiopia,Kenya,Malawi,Mozambique,Rwanda,Tanzania,Uganda,Zambia,Zimbabwe;WestAfrica:Benin,BurkinaFaso,Côted’Ivoire,Ghana,Guinea,Liberia,Mali,Nigeria,Senegal,SierraLeone,Togo;CentralAfrica:Angola,Cameroon,Chad,DemocraticRepublicoftheCongo,Gabon;southernAfrica:Lesotho,Namibia).28.WorldBank,“UrbanizationReviews”(dashboard),https://www.worldbank.org/en/topic/urbandevelopment/publication/urbanization-reviews.29.ThediscussioninthisparagraphislargelybasedonthatcontainedinanewWorldBankreport,VibrantCities:PrioritiesforGreen,Resilient,andInclusiveUrbanDevelopment(Lalletal.,forthcoming).ReferencesAcharya,G.,E.Cassou,S.Jaffee,andE.K.Ludher.2021.RICHFood,SmartCity:HowBuildingReliable,Inclusive,Competitive,andHealthyFoodSystemsIsSmartPolicyforUrbanAsia.Washington,DC:WorldBank.Ager,P.,K.Eriksson,andL.Lonstrup.2020.“Howthe1906SanFranciscoEarthquakeShapedEconomicActivityintheAmericanWest.”ExplorationsinEconomicHistory77:101342.Ahlfeldt,G.,andE.Pietrostefani.2019.“TheEconomicEffectsofDensity:ASynthesis.”JournalofUrbanEconomics111:93–107.Awe,Y.,J.Nygard,S.Larssen,H.Lee,H.Dulal,andR.Kanakia.2015.“CleanAirandHealthyLungs:EnhancingtheWorldBank’sApproachtoAirQualityManagement.”EnvironmentandNaturalResourcesGlobalPracticeDiscussionPaperNo.3,WorldBank,Washington,DC.Bai,Y.,R.Alemu,S.Block,D.Headey,andW.Masters.2021.“CostandAffordabilityofNutritiousDietsatRetailPrices:Evidencefrom177Countries.”FoodPolicy99(February):101983.TheStylizedRelationships117BBCNews.2015.“GhanaPetrolStationInfernoKillsabout150inAccra.”BBCNews,June5,2015.https://www.bbc.com/news/world-africa-33003673.Behrens,K.,andF.Robert-Nicoud.2015.“AgglomerationTheorywithHeterogeneousAgents.”InHandbookofRegionalandUrbanEconomics,Volume5:CitiesandGeography,editedbyG.Duranton,J.V.Henderson,andW.Strange,171–245.Amsterdam:Elsevier.Bosker,M.,J.Park,andM.Roberts.2021.“DefinitionMatters.MetropolitanAreasandAgglomerationEconomiesinaLarge-DevelopingCountry.”JournalofUrbanEconomics125:103275.Brakman,S.,H.Garretsen,andM.Schramm.2004.“TheStrategicBombingofGermanCitiesduringWorldWarIIandItsImpactonCityGrowth.”JournalofEconomicGeography4(2):201–18.Brakman,S.,H.Garretsen,andC.vanMarrewijk.2012.TheNewIntroductiontoGeographicalEconomics.Cambridge,UK:CambridgeUniversityPress.BruceN.,R.Perez-Padilla,andR.Albalak.2000.“IndoorAirPollutioninDevelopingCountries:AMajorEnvironmentalandPublicHealthChallenge.”BulletinoftheWorldHealthOrganization78(9):1078–92.Bruederle,A.,J.Peters,andG.Roberts.2017.“WeatherandCrimeinSouthAfrica.”RuhrEconomicPapera739.RWI–LeibnizInstitutfürWirtschaftsforschung,Ruhr-UniversityBochum,TUDortmundUniversity,UniversityofDuisburg-Essen.Buisseret,D.J.1966.“PortRoyal1655–1725.”JamaicanHistoricalReview6:21–28.Buurman,J.,M.J.Mens,andR.J.Dahm.2017.“StrategiesforUrbanDroughtRiskManagement:AComparisonof10LargeCities.”InternationalJournalofWaterResourcesDevelopment33(1):31–50.C40CitiesClimateLeadershipGroup,Arup,andUniversityofLeeds.2019.TheFutureofUrbanConsumptionina1.5°CWorld.C40CitiesHeadlineReport.https://www.c40knowledgehub.org/s/article/The-future-of-urban-consumption-in-a-1-5-C-world?language=en_US.Chand,S.,K.Walsh,S.Camargo,J.Kossin,K.Tory,M.Wehner,J.Chan,etal.2022.“DecliningTropicalCycloneFrequencyunderGlobalWarming.”NatureClimateChange12:655–61.Chatterjee,S.,M.Gassier,andN.Myint.Forthcoming.“LeveragingSocialCohesionforDevelopmentOutcomes.”AFramingPaper,SocialSustainabilityandInclusionGlobalPractice,WorldBank,Washington,DC.CIESIN(CenterforInternationalEarthScienceInformationNetwork).2013.“LowElevationCoastalZoneUrban-RuralPopulationandLandAreaEstimates,Version2.”Technicalreport,CIESIN,ColumbiaUniversity;NASASocioeconomicDataandApplicationsCenter,Palisades,NY.Claypole,W.A.1972.“TheMerchantsofPortRoyal,1655to1700.”Unpublishedmaster’sthesis,UniversityoftheWestIndies,Kingston,Jamaica.Combes,P.P.,S.Nakamura,M.Roberts,andB.Stewart.2022.“EstimatingUrbanPovertyConsistentlyacrossCountries.”WorldBankPovertyandEquityNotes,WorldBank,Washington,DC.Corbane,C.,P.Martino,P.Panagiotis,F.Aneta,M.Michele,F.Sergio,S.Marcello,etal.2020.“TheGrey-GreenDivide:Multi-TemporalAnalysisofGreennessacross10,000UrbanCentresTHRIVING118DerivedfromtheGlobalHumanSettlementLayer(GHSL).”InternationalJournalofDigitalEarth13(1):101–18.Crippa,M.,G.Oreggioni,D.Guizzardi,M.Muntean,E.Schaaf,E.LoVullo,E.Solazzo,etal.2019.FossilCO2andGHGEmissionsofAllWorldCountries.EUR29849EN.Luxembourg:PublicationsOfficeoftheEuropeanUnion.Cuesta,J.,L.Madrigal,andN.Pecorari.Forthcoming.“SocialSustainability,PovertyandIncome:AnEmpiricalExploration.”PolicyResearchWorkingPaper,WorldBank,Washington,DC.Dasgupta,S.,K.Hamilton,K.Pandey,andD.Wheeler.2004.“AirPollutionduringGrowth:AccountingforGovernanceandVulnerability.”PolicyResearchWorkingPaper3383,WorldBank,Washington,DC.Davis,D.,andD.Weinstein.2002.“Bones,Bombs,andBreakPoints:TheGeographyofEconomicActivity.”AmericanEconomicReview92(5):1269–89.deBellefon,M.P.,P.P.Combes,G.Duranton,L.Gobillon,andC.Gorin.2021.“DelineatingUrbanAreasUsingBuildingDensity.”JournalofUrbanEconomics125:103226.Deuskar.2022.“BeatingtheHeat:MeasuringandMitigatingExtremeHeatinEastAsianCities.”TechnicalWorkingPaper1:LiteratureReview,WorldBank,Washington,DC.Dijkstra,L.,A.Florczyk,S.Freire,T.Kemper,M.Melchiorri,M.Pesaresi,andM.Schiavina.2021.“ApplyingtheDegreeofUrbanisationtotheGlobe:ANewHarmonisedDefinitionRevealsaDifferentPictureofGlobalUrbanisation.”JournalofUrbanEconomics125:103312.Dizon,F.,Z.Wang,andP.Mulmi.2021.“TheCostofaNutritiousDietinBangladesh,Bhutan,India,andNepal.”PolicyResearchWorkingPaper9578,WorldBank,Washington,DC.Duque,J.C.,N.Lozano-Gracia,J.E.Patino,andP.Restrepo.2021.“UrbanFormandProductivity:WhatShapesAreLatin-AmericanCities?”EnvironmentandPlanningB:UrbanAnalyticsandCityScience49(1):131–50.Duranton,G.,andD.Puga.2020.“TheEconomicsofUrbanDensity.”JournalofEconomicPerspectives34(3):3–26.Ellis,P.,andM.Roberts.2016.LeveragingUrbanizationinSouthAsia:ManagingSpatialTransformationforProsperityandLivability.Washington,DC:WorldBank.Fang,W.,andS.Wahba.2020.“UrbanDensityIsNotanEnemyintheCoronavirusFight:EvidencefromChina.SustainableCities.”SustainableCities(blog),April20,2020.https://blogs.worldbank.org/sustainablecities/urban-density-not-enemy-coronavirus-fight-evidence-china.FAO(FoodandAgricultureOrganizationoftheUnitedNations).2018.ForestsandSustainableCities:InspiringStoriesfromAroundtheWorld.FAOForestryDepartment.Rome:FAO.Feng,K.,K.Hubacek,andY.Yu.2019.LocalConsumptionandGlobalEnvironmentalImpacts:Accounting,Trade-OffsandSustainability.London:Routledge.Ferre,C.,F.Ferreira,andP.Lanjouw.2012.“IsThereaMetropolitanBias?TheRelationshipbetweenPovertyandCitySizeinaSelectionofDevelopingCountries.”WorldBankEconomicReview26(3):351–82.Ferreyra,M.M.,andM.Roberts,eds.2018.RaisingtheBarforProductiveCitiesinLatinAmericaandtheCaribbean.Washington,DC:WorldBank.TheStylizedRelationships119Florczyk,A.,J.M.Melchiorri,C.Corbane,M.Schiavina,M.Maffenini,M.Pesaresi,P.Politis,etal.2019.“DescriptionoftheGHSUrbanCentreDatabase2015,PublicRelease2019,Version1.0.”PublicationsOfficeoftheEuropeanUnion,Luxembourg.Galdo,V.,Y.Li,andM.Rama.2021.“IdentifyingUrbanAreasbyCombiningHumanJudgmentandMachineLearning:AnApplicationtoIndia.”JournalofUrbanEconomics125:103229.Gandhi,S.,M.E.Kahn,R.Kochhar,S.V.Lall,andV.Tandel.2022.“AdaptingtoFloodRisk:EvidencefromaPanelofGlobalCities.”WorkingPaper30137,NationalBureauofEconomicResearch,Cambridge,MA.Glaeser,E.L.2012.TriumphoftheCity:HowOurGreatestInventionMadeUsRicher,Smarter,Greener,HealthierandHappier.NewYork:PenguinPress.Glaeser,E.L.2022.“UrbanResilience.”UrbanStudies59(1):3–35.GreaterLondonAuthority.2002.50YearsOn:TheStruggleforAirQualityinLondonsincetheGreatSmogofDecember1952.London:GreaterLondonAuthority.GreaterLondonAuthority.2020.AirQualityinLondon,2016–2020.LondonEnvironmentStrategy:AirQualityImpactEvaluation.London:GreaterLondonAuthority.Hallegatte,S.,A.Vogt-Schilb,M.Bangalore,andJ.Rozenberg.2017.Unbreakable:BuildingtheResilienceofthePoorintheFaceofNaturalDisasters.Washington,DC:WorldBank.Headey,D.,andH.Alderman.2019.“TheRelativeCaloricPricesofHealthyandUnhealthyFoodsDifferSystematicallyacrossIncomeLevelsandContinents.”JournalofNutrition149:2020–33.Heilmann,K.,andM.E.Kahn.2019.“TheUrbanCrimeandHeatGradientinHighandLowPovertyAreas.”WorkingPaper25961,NationalBureauofEconomicResearch,Cambridge,MA.Henderson,J.V.,A.Storeygard,andD.N.Weil.2011.“ABrightIdeaforMeasuringEconomicGrowth.”AmericanEconomicReview101(3):194–99.Henderson,J.V.,A.Storeygard,andD.N.Weil.2012.“MeasuringEconomicGrowthfromOuterSpace.”AmericanEconomicReview102(2):994–1028.Henderson,J.V.,andM.Turner.2020.“UrbanizationintheDevelopingWorld:TooEarlyorTooSlow?”JournalofEconomicPerspectives34(3):150–73.Herforth,A.,Y.Bai,A.Venkat,K.Mahrt,A.Ebel,andW.A.Masters.2020.“CostandAffordabilityofHealthyDietsacrossandwithinCountries.”BackgroundpaperpreparedforTheStateofFoodSecurityandNutritionintheWorld2020.FAOAgriculturalDevelopmentEconomicsTechnicalStudyNo.9.Rome:FoodandAgricultureOrganizationoftheUnitedNations.Hopkins,F.,J.Ehleringer,S.Bush,R.Duren,C.Miller,C.Lai,Y.Hsu,etal.2016.“MitigationofMethaneEmissionsinCities:HowNewMeasurementsandPartnershipsCanContributetoEmissionsReductionStrategies.”Earth’sFuture4:408–25.Hornbeck,R.,andD.Keniston.2017.“CreativeDestruction:BarrierstoUrbanGrowthandtheGreatBostonFireof1872.”AmericanEconomicReview107(6):1365–98.Hossain,M.K.,andQ.Meng.2020.“AFine-ScaleSpatialAnalysisoftheAssessmentandMappingofBuildingsandPopulationatDifferentRiskLevelsofUrbanFlood.”LandUsePolicy99(December):104829.THRIVING120IPCC(IntergovernmentalPanelonClimateChange).2021.ClimateChange2021:ThePhysicalScienceBasis.ContributionofWorkingGroupItotheSixthAssessmentReportoftheIntergovernmentalPanelonClimateChange.Cambridge,UK:CambridgeUniversityPress.Jin,J.,S.Sheppard,andC.Wang.2021.“PlanningtoPractice:ImpactsofLarge-ScaleandRapidUrbanAfforestationonGreenspacePatternsintheBeijingPlainArea.”Forests12(3):316.Kahn,M.E.,andP.Li.2020.“AirPollutionLowersHighSkillPublicSectorWorkerProductivityinChina.”EnvironmentalResearchLetters15(8):084003.Karimi,F.,andC.Lett.2015.“GhanaExplosion:PeopleFleeingHeavyRainAreamong150KilledinAccra.”CNN.com,June5,2015.https://www.cnn.com/2015/06/05/africa/ghana-explosion-floods/index.html.Knutson,T.,S.Camargo,J.Chan,K.Emanuel,C.-H.Ho,J.Kossin,M.Mohapatra,etal.2020.“TropicalCyclonesandClimateChangeAssessment.PartII:ProjectedResponsetoAnthropogenicWarming.”BulletinoftheAmericanMeteorologicalSociety101(3):E303–22.Krugman,P.1991.“IncreasingReturnsandEconomicGeography.”JournalofPoliticalEconomy99(3):483–99.Lall,S.V.,J.Kaw,S.B.Murray,andF.Shilpi.Forthcoming.VibrantCities:PrioritiesforGreen,Resilient,andInclusiveUrbanDevelopment.Washington,DC:WorldBank.Lall,S.V.M.Lebrand,H.Park,D.M.Sturm,andA.J.Venables.2021.PancakestoPyramids:CityFormtoPromoteSustainableGrowth.Washington,DC:WorldBank.Lall,S.V.,andS.Wahba.2020.NoUrbanMyth:BuildingInclusiveandSustainableCitiesinthePandemicRecovery.SustainableDevelopmentSeries.Washington,DC:WorldBank.https://www.worldbank.org/en/news/immersive-story/2020/06/18/no-urban-myth-building-inclusive-and-sustainable-cities-in-the-pandemic-recovery.Lanthier,N.2007.“TalkTellsStoryofJamaican‘UnderwaterCity.’”VancouverSun,March24,2007.Mariel,P.M.,A.Khan,andJ.Meyerhoff.2022.“ValuingIndividuals’PreferencesforAirQualityImprovement:EvidencefromaDiscreteChoiceExperimentinSouthDelhi.”EconomicAnalysisandPolicy74:432–47.Moran,D.,R.Wood,E.Hertwich,K.Mattson,J.Rodriguez,K.Schanes,andJ.Barrett.2018.“CarbonFootprintsof13,000Cities.”EnvironmentalResearchLetters13(2018).Morgan,S.2022.“SwedenSettoBeWorld’sFirstCountrytoTargetConsumption-BasedEmissionCuts.”ClimateHomeNews,April8,2022.Park,J.,andM.Roberts.2023.“UrbanResilienceandWeatherShocks:AGlobalNighttimeLightsAnalysis.”Backgroundpaperpreparedforthisreport,WorldBank,Washington,DC.Pawson,M.,andD.J.Buisseret.1975.PortRoyal,Jamaica.Oxford,UK:ClarendonPress.Polsby,D.D.,andR.D.Popper.1991.“TheThirdCriterion:CompactnessasaProceduralSafeguardagainstPartisanGerrymandering.”YaleLawandPolicyReview9(2):301–53.Raufman,J.,D.Blansky,D.W.Lounsbury,E.W.Mwangi,Q.Lan,J.Olloquequi,andH.D.Hosgood.2020.“EnvironmentalHealthLiteracyandHouseholdAirPollution-AssociatedSymptomsinKenya:ACross-SectionalStudy.”EnvironmentalHealth19(1):89.TheStylizedRelationships121THRIVINGRoberts,M.,B.Blankespoor,C.Deuskar,andB.Stewart.2017.“UrbanizationandDevelopment.IsLatinAmericaandtheCaribbeanDifferentfromtheRestoftheWorld?”PolicyResearchWorkingPaper8019,WorldBank,Washington,DC.Roberts,M.,F.GilSander,andS.Tiwari,eds.2019.TimetoACT:RealizingIndonesia’sUrbanPotential.Washington,DC:WorldBank.Sanz-Barbero,B.,C.Linares,C.Vives-Cases,J.L.González,J.J.López-Ossorio,andJ.Díaz.2018.“HeatWaveandtheRiskofIntimatePartnerViolence.”ScienceoftheTotalEnvironment644(December):413–19.https://doi.org/10.1016/j.scitotenv.2018.06.368.Schencking,J.C.2006.“Catastrophe,Opportunism,Contestation:TheFracturedPoliticsofReconstructingTokyofollowingtheGreatKantôEarthquakeof1923.”ModernAsianStudies40(4):833–73.Schut,A.G.,T.E.Ivits,J.G.Conijin,B.tenBrink,andR.Fensholt.2015.“TrendsinGlobalVegetationActivityandClimateDriversIndicateaDecoupledResponsetoClimateChange.”PLoSONE10(10):e0138013.Seto,K.,G.Churkina,A.Hsu,M.Keller,P.Newman,B.Qin,andA.Ramaswami.2021.“FromLow-toNet-ZeroCarbonCities:TheNextGlobalAgenda.”AnnualReviewofEnvironmentandResources46:377–415.Smith,A.1776.TheWealthofNations.London:W.StrahanandT.Cadell.Taylor,J.1688.“SecondPartoftheHistorieofHisLifeandTravelsinAmerica.”Manuscriptonfile,InstituteofJamaica,Kingston.Tewari,M.,andN.Godfrey.2016.BetterCities,BetterGrowth:India’sUrbanOpportunity.TheNewClimateEconomy,GlobalCommissionontheEconomyandClimate.Washington,DC:WorldResourcesInstitute;London:OverseasDevelopmentInstitute;NewDelhi:IndianCouncilforResearchonInternationalEconomicRelations.UnitedNations.2019.WorldUrbanizationProspects:The2018Revision.NewYork:DepartmentofEconomicandSocialAffairs,PopulationDivision,UnitedNations.https://population.un.org/wup/publications/Files/WUP2018-Report.pdf.Wang,S.,andR.Toumi.2021.“RecentMigrationofTropicalCyclonestowardCoasts.”Science371:514–17.Watts,G.,B.vonChristierson,J.Hannaford,andK.Lonsdale.2012.“TestingtheResilienceofWaterSupplySystemstoLongDroughts.”JournalofHydrology414:255–67.Weber,A.2019.“WhatIsUrbanFlooding?”NaturalResourcesDefenseCouncilExpertBlog,January15,2019.https://www.nrdc.org/experts/anna-weber/what-urban-flooding.Wiedmann,T.,andM.Lenzen.2018.“EnvironmentalandSocialFootprintsofInternationalTrade.”NatureGeoscience11:314–21.Wiedmann,T.,G.Chen,A.Owen,M.Lenzen,M.Doust,J.Barrett,andK.Steele.2021.“Three-ScopeCarbonEmissionsInventoriesofGlobalCities.”JournalofIndustrialEcology25:735–50.WorldBank.2009.WorldDevelopmentReport2009:ReshapingEconomicGeography.Washington,DC:WorldBank.WorldBank.2021.Green,Resilient,andInclusiveDevelopment.Washington,DC:WorldBank.122WorldBank.2022a.SocialSustainabilityGlobalDatabase2022.Washington,DC:WorldBank.WorldBank.2022b.SocialSustainabilityGlobalDatabase2022Codebook.Washington,DC:WorldBank.Xu,X.,P.Sharma,S.Shu,T.Lin,P.Ciais,F.N.Tubiello,P.Smith,etal.2021.“GlobalGreenhouseGasEmissionsfromAnimal-BasedFoodsAreTwiceThoseofPlant-BasedFoods.”NatureFood2:724–32.Yao,L.,Y.Xu,andB.Zhang.2019.“EffectofUrbanFunctionandLandscapeStructureontheUrbanHeatIslandPhenomenoninBeijing,China.”LandscapeandEcologicalEngineering15:379–90.Zaveri,E.,J.Russ,A.Khan,R.Damania,E.Borgomeo,andA.Jagerskog.2021.EbbandFlowVol.1:Water,MigrationandDevelopment.Washington,DC:WorldBank.Zhao,Q.,Y.Guo,T.Ye,A.Gasparrini,S.Tong,A.Overcenco,A.Urban,etal.2021.“Global,Regional,andNationalBurdenofMortalityAssociatedwithNon-optimalAmbientTemperaturesfrom2000to2019:AThree-StageModellingStudy.”LancetPlanetaryHealth5(7):E415–25.TheStylizedRelationships123125IntroductionAschapter1revealed,citiesacrosstheglobevarywidelyintermsofhowgreen,howresilient,andhowinclusivetheyarecurrently.Theydiffernotonlyinphysicalgeographyandclimatebutalsointheircapacitiestomanagethestressesarisingfromthepressuretheirpopulationsputontheirland,housing,basicservices,infrastructure,andenvironment.Thus,thechallengesfacedbyawell-plannedwealthycitywithahighlyproductiveserviceeconomydiffersignificantlyfromthoseconfrontingafast-growinglower-incomecitywithalargemanufacturingsector.AGlobalTypologyofCitiesCHAPTER2Climatechange...doesnotrespectwhoyouare—richandpoor,smallandbig.BanKi-Moon,Secretary-GeneraloftheUnitedNationsRemarks,“MomentumforChange”Initiative,December2011••Citiescanbeusefullyclassifiedusingatypologyoftwodimensions,populationsizeandlevelofdevelopmentofthecountryinwhichtheyarelocated.Suchatypologydistinguishesbetweenninetypesofcities:small,medium,andlargecitiesinlow-,middle-,andhigh-incomecountries.••Acrosstheseninetypesofcities,themixandseverityofgreenness,resilience,andinclusivenesschallengesvarywidely.Likewise,citiesvaryinboththemixandtheseverityoftheclimatechange–relatedhazardstheywillfaceinthenexttwodecades.••Lookingforwardto2030–40,citiesinlow-incomecountriesandlargecitiesworldwidewillmorelikelyhavegreaterexposuretoclimatechange–relatedhazards.Thus,althoughcitiesinmiddle-andhigh-incomecountriesareresponsibleforthebulkoftheworld’surbancarbondioxideemissions,citiesinlow-incomecountries—ifnotwellprepared—willlikelybearthebruntofclimatechange.••SmallislanddevelopingstatesintheCaribbeanandthePacificareprojectedtobeworstpositionedtowithstandclimatechange–relatedhazards,inpartbecausetheycannotadaptthroughinternalmigration.••Althoughthequalityandbreadthofclimatedatahaveimproveddramaticallyinrecentdecades,highlevelsofuncertaintyandcoveragegapsonmorelocalscaleshighlighttheimportanceoffurtherinvestmentsininformationontheclimatechange–relatedrisksthatcitiesgloballyface.Suchinformationisanindispensableinputintobetterpolicymaking.MAINFINDINGS126THRIVINGAlthoughthesedifferencesmakeitdifficulttoidentifythechallengesthatconfronteachcityglobally,atypologyofcitiescanprovideusefulinformationtocitiesthatsharesimilarcharac-teristics.Indeed,multipleinitiativeshaverecentlyemergedfromglobalcitynetworkswiththeaimofimprovingcities’capacityforclimatechangemitigationandadaptationthroughcol-laborationandknowledgesharing.1Althoughauniversalupscalingofeffectivemeasuresandinterventionsisdifficult,understandingsomegeneralpatternsemergingfromsimilarcitieselsewherecanhelpinidentifyingcommonpolicyneeds,whichiscrucialtotransferringgoodorbestpracticessuccessfully(Sterzeletal.2020).Inviewoftheseissues,thischapterhasthefollowingmainobjectives:••Defineaglobaltypologyofcitiesthatprovidesahigh-leveloverviewofthemixandseverityofgreenness,resilience,andinclusivenesschallengesthatatypicalcityofagiventypefaces.••Presenttheheterogeneitiesinthemixandseverityofclimate-relatedhazardstowhichdifferenttypesofcitiesareprojectedtobeexposedin2030–40.Toachievetheseobjectives,thechapterbuildsontheinsightsfromchapter1thatvariousindicatorsofgreenness,resilience,andinclusivenessarerelatedtobothacity’ssizeandthelevelofdevelopmentofthecountryinwhichitislocated.Suchatypologyisundeniablycrude,butitdoesgiveabroadsenseofhowurbanchallengesvaryacrosscitiesgloballyandhowthesechallengesmayevolvebothascitiesgrow—orinsomecasesshrink—andascountriesmovefromonelevelofdevelopmentuptoanother.Topresenttheheterogeneitiesinclimateexposure,thischapteroverlaysthistypologywithcity-specificinformationonthelevelsofexposuretoclimate-relatedhazardsprojectedforaglobalsubsampleofmorethan2,200citiesin2030–40.Thisexercisenotonlyrevealshowthemixandseverityofthesehazardsvaryacrosstypesofcitiesbutalsohighlightsuniquechallengesthatcitiesinsmallislandnationsfaceintermsoflimitedadaptationstrategies.Togetherwiththeanalyticalinsightsgeneratedinpart2ofthisreportonhowclimatechange–relatedstressesaffectcities(chapter3)andhowcitiesaffecttheenvironment(chapter4),thisanalysisprovidescrucialinputintothehigh-levelpolicyguidanceprovidedinpart3(chapter5)ofthereport.DefiningaglobaltypologyofcitiesThegoalofdefiningaglobaltypologyofcitiesistogrouptogethercitiesthatresembleoneanotherintheircurrentchallengesrelatedtogreenness,resilience,andinclusiveness,whileensuringbroadapplicability.Fulfillingthisgoalrequiresconfrontingthefollowingfundamen-talproblems.••Thecurseofdimensionality.Chapter1exploredmultipleindicatorsforassessinghowgreen,howresilient,andhowinclusivecitiesarecurrently.However,groupingcitiesbasedonmultipleindicatorstoformatypologypresentsamultidimensionalandhighlycomplexclus-teringproblem.••Missingdata.Thisproblemappliesespeciallytoindicatorsthatcaptureaspectsoftheinclu-sivenessofcities.Forthatreason,thestylizedrelationshipspresentedinchapter1arebasedoneithercombininginsightsfromthesecondaryliteraturederivedfromsamplesofcitiesoranalyzingdichotomousurban–ruraldifferencesinoutcomesacrosscountriesatdifferentlevelsofdevelopment.Thecurseofdimensionalitycanbeovercomebytakingadvantageofthefactthat,asshowninchapter1,theintensityofmanyofthegreenness,resilience,andinclusivenesschallengesAGlobalTypologyofCities127confrontingcitiesvarywithacity’ssizeandthelevelofdevelopmentofacity’sparentcountry.Thecursecanthenbeaddressedbydefiningaglobaltypologyalongthosetwodimensions.Althoughthisapproachmayraiseconcernsthatitmissesimportantdimensionsofacity’scharacterization,theexistingliterature,whichattemptstoquantifytheintensityofvariousurbanchallenges,suggeststhatpopulation(eithersizeordensity)isaprimarydeterminantofthestrengthofchallengesacrossdevelopmentstages.2Themissingdataproblemcanbeovercomebyprimarilyfocusing,atleastindefiningthecitysizeanddevelopmentcategoriesusedinthetypology,onthegreennessindicators,whichareavailableforthisreport’sfullglobalsampleofmorethan10,000cities.Theemergingtypologyisthenalsoappliedtoassesstheseverityofresilienceandinclusivenesschallengestothelargestextentpossible(box2.1).DefiningandapplyingtheglobaltypologyofcitiesTheapproachusesthreestepstodefinetheglobaltypologyofcitiesandsubsequentlytoassesstheassociatedseverityofcurrentgreenness,resilience,andinclusivenesschallenges.Annex2Aprovidesamoredetaileddescription.Step1.Defineeightpossibletypologies,eachofwhichcombinesthethreecategoriesofcitysizeandthethreecategoriesofdevelopmentlevel.Therearemultipleplausiblewaysofdefiningwhetheracityissmall,medium,orlargeand,similarly,whetherthecountryinwhichacityislocatedshouldbeclassifiedaslow-,middle-orhigh-income.Toincorporatethesemultipledefinitions,eightalterna-tivetypologieswereinitiallydefined.Theyarebasedonfourcompetingdefinitionsofthethreecitysizecategories(small,medium,andlarge)andtwocompetingdefinitionsofthethreedevelopmentlevelcategories(low-,middle-,andhigh-income).Thefirstandseconddefinitionsofcitysizecategoriesarebasedonthetercilesoftheglobalandthewithin-countrydistributionsofcitypopulations,respectively,whereasthethirdandfourtharebasedontheOrganisationforEconomicCo-operationandDevelopment’surbanareaclassification.ThetwocompetingdefinitionsofdevelopmentlevelcategoriesarederivedfromtheWorldBank’scountryincomeclassificationsforfiscalyear2021/22.Step2.Identifythetypologythatbestaccountsforthevariationingreennessindicatorsacrosstheglobalsampleofcities.Thisstepidentifies,amongtheeighttypologiesdefinedinstep1,thetypologythatbestaccountsforthevariationinthegreennessindicators.Thisstepiscarriedoutbyapplyinganalysisofcovariancetechniquestothesixgreennessindicators:(1)totalfossilcarbondioxideemissions;(2)percapitafossilcarbondioxideemissions;(3)totalemissionsofparticulatematterof2.5micronsorlessindiameter;(4)totalconcentrationsofparticu-latematterof2.5micronsorlessindiameter;(5)acity’saveragelevelofvegetation;and(6)theshareofacity’sareathatis“highgreen”—thatis,coveredbydensevegetationcorrespondingtogardens,parks,andurbanforests.Thetypologybestabletoaccountforthestatisticalvariationacrosscitiesinthesesixindicatorsisselectedasthe“winning”typology.Box2.1128THRIVINGStep3.Assesseachtypeofcityintermsoftherelativeseverityofthecurrenturbanchallenges.Thefinalstepistodistinguishtherelativelevelsofseverityassociatedwithgreenness,resilience,andinclusivenesschallengesfacingthetypesofcities,onaverage.Althoughstatisticalmethodsandunderlyingdatasetsvaryacrosstypesofchallenges,dependingonthedatastructureandavailability,theapproachtoclassifyingtherelativelevelofseverityessentiallycomprisesfoursubsteps.First,estimatethemeanabsolutelevelofeachindicatorforeachtypeofcity(suchasthemeanfossilcarbondioxideemissions).Second,assigneachtypeofcityalevelofseverity(high,medium,orlow)basedontheorderofthemeans.Iftwotypesofcitieshavemeanabsolutelevelsofanindicatorthatarenotstatisticallydifferentinpairwisecomparisons,theseverityofthechallengeisclassifiedasbeingthesame.Third,becauseeachtypeofchallengeisrepresentedbymultipleindicators(forexample,sixindicatorsforthegreennesschallenge),calculatetheunweightedmeanlevelofseverityacrossallindicatorsastheoveralllevelofseverityforeachtypeofchallenge.Fourth,filltheremaininggapsinclassifyingthelevelsofseveritybymakinginferencesbasedonwhatisknownaboutsuchchallengesfromthesecondaryliterature.Box2.1continuedNinetypesofcitiesgloballyemergeApplyingsteps1and2ofthemethodologydescribedinbox2.1resultsinapreferredglobaltypologyinwhichsmall,medium,andlargecitiescorrespondtotheOrganisationforEconomicCo-operationandDevelopment’scategoriesofsmallcities(populationof50,000–199,999),metropolitanareasandmediumcities(populationof200,000–1.499million),andlargemetropolitanareas(populationofatleast1.5million),respectively.Meanwhile,low-incomecountriescorrespondtocountriesthattheWorldBankclassifiesaseitherlow-orlower-middle-income,andmiddle-incomecountriescorrespondtothosethattheWorldBankclassifiesasupper-middle-income.High-incomecountriesremainasclassifiedbytheWorldBank(table2.1).Onaverage,thistypologyaccountsfor32percentofthevariationacrossindicators.Map2.1showstheglobaldistributionofthetypesofcities.Table2.1CategoriesofcitysizeandlevelofdevelopmentusedtodefineglobaltypologyofcitiesDevelopmentlevelGNIpercapitarangeCitysizePopulationrangeLow-income<US$4,096Small50,000–199,999Middle-incomeUS$4,096–US$12,695Medium200,000–1.499millionHigh-income>US$12,695Large1.5millionormoreSources:WorldBankanalysisbasedonpopulationdatafromtheEuropeanCommission’sGlobalHumanSettlement(GHS)UrbanCentreDatabaseR2019(https://ghsl.jrc.ec.europa.eu/ghs_stat_ucdb2015mt_r2019a.php).PercapitaGNIdatacomefromtheWorldBank’sWorldDevelopmentIndicatorsdatabase(http://wdi.worldbank.org).Note:Citiesaredefinedasurbancentersfollowingthedegreeofurbanizationmethodology(seebox1.1,chapter1).Thepopulationrangesarebasedonthepopulationsoftheseurbancentersin2015.GNIpercapitaismeasuredincurrent2020USdollarsandcalculatedusingtheWorldBankAtlasmethod.GNI=grossnationalincome.AGlobalTypologyofCities129EachtypecoversabroadspectrumofcitiesLow-incomecountriesThisgroupcomprises5,344citiesfrom76countries,ormorethanhalfthecitiescoveredbytheglobalsample.Ofthese,42percentareinSouthAsia,withIndiaaloneaccountingfor36percent.Sub-SaharanAfricafollows,with7.3percentand4.4percentoflow-incomecountrycitiesinNigeriaandEthiopia,respectively.Bycontrast,EuropeandCentralAsiaandLatinAmericaandtheCaribbeanaccountforonly4percentofcitiesinlow-incomecountries(figure2.1).Thelow-incomegroupconsistsofthefollowingcategories.••Low-incomesmallcities(type1)accountforalmostathird(about26percent)ofthetotalpopulationofcitiesinlow-incomecountries(figure2.2).Citiesofthistypeincludeagricul-turalmarkettowns(suchasPirganj,Bangladesh),naturalresourceindustrycenters(suchasLuwuk,Indonesia),touristdestinations(suchasBattambang,Cambodia),transportationhubs(suchasGalle,SriLanka),satellitecitiesadjacenttoalargercitybutfallingunderaseparateadministrativeunit(suchasDemak,Indonesia),andcapitalsofislandstates(suchasHoniara,SolomonIslands).Nevertheless,theirmediangrossdomesticproductpercapitalevelsimplythatcitiesofthistypearetheleastproductive(table2.2).Source:WorldBankcalculationsbasedondatafromtheEuropeanCommission’sGlobalHumanSettlement(GHS)UrbanCentreDatabaseR2019(https://ghsl.jrc.ec.europa.eu/ghs_stat_ucdb2015mt_r2019a.php).Note:Citiesaredefinedasurbancentersfollowingthedegreeofurbanizationmethodology.Typesofcitiesaredefinedusingthemethodologydescribedinbox2.1andannex2A.Map2.1Globaltypologyofcities130THRIVINGSource:WorldBankanalysisbasedondatadrawnfromtheEuropeanCommission’sGlobalHumanSettlement(GHS)UrbanCentreDatabaseR2019(https://ghsl.jrc.ec.europa.eu/ghs_stat_ucdb2015mt_r2019a.php).Note:Citiesaredefinedasurbancentersfollowingthedegreeofurbanizationmethodology(seebox1.1,chapter1).TheincomeclassesarebasedontheWorldBankcountryclassificationsforfiscalyear2021/22.Low-incomecountriescorrespondtolow-andlower-middle-incomecountries,andmiddle-incomecountriestoupper-middle-incomecountries.Figure2.1Inglobaltypology,numberofcities,byregionandcountryincomegroup05001,0001,5002,0002,500Low-incomeMiddle-incomeHigh-incomeNo.ofcitiesSouthAsiaSub-SaharanAfricaMiddleEastandNorthAfricaLatinAmericaandtheCaribbeanEastAsiaandPacificEuropeandCentralAsiaNorthAmerica••Low-incomemediumcities(type2)arehometoalmost550millionpeopleinabout1,300cities,accountingfor35.4percentofthepopulationofcitiesinlow-incomecountries(figure2.2).Asexpectedfromthewiderangeofpopulationsize(200,000–1.499million),thistypeofcityalsocoversabroadspectrumoffunctionality,dependingonthesizeofthehostingcountry.Forexample,nationalcapitalsofmanyAfricancountries,includingAbuja,Nigeria,andMonrovia,Liberia,andthoseofsmallercountries,suchasBishkek,KyrgyzRepublic,andUlaanbaatar,Mongolia,belongtothisgroup.Subnationalcapitals(suchasRaipur,India)andregionaleconomiccenters(suchasMarrakesh,Morocco;Odessa,Ukraine;Shiraz,IslamicRepublicofIran;andSylhet,Bangladesh)belongtothistypeaswell.Thisgroupalsoincludescities—suchasMuzaffarpur,India—whoseeconomicbaseisstillmainlyprovidedbythelarge-scaleprimarysector.••Low-incomelargecities(type3)includemetropolitanareas—suchasDelhi,Dhaka,Jakarta,andMumbai—inemergingeconomies.These134citiesarehometojustover600millionpeople,ornearly40percentofthecitypopulationinlow-incomecountries(figure2.2).Inadditiontonationalprimatecities,thisgroupincludessecondarycitieswithfunctionalspecialty,suchasgatewaycitiesthataccountforasignificantportionofacountry’sinternationaltrade(amongthem,Chittagong,Bangladesh),majorinformationandcommunicationtechnologyservicecenters(notably,BangaloreandHyderabad,India),andinternationaltourismdes-tinationssuchasDenpasar,Indonesia.Despitetheirsizeandimportancetotheirnationaleconomies,however,themediangrossdomesticproductpercapitaacrosstype3citiesfallsfarbehindthatofevensmallcitiesinmiddle-incomecountries.Atthesametime,almosthalfofthesecitiesarelocatedwithin100kilometersofthenearestcoastline,implyingpossiblevulnerabilityto,forexample,tropicalcyclonesandhighersealevels(figure2.3).AGlobalTypologyofCities131Source:WorldBankanalysisbasedonpopulationdatafromtheEuropeanCommission’sGlobalHumanSettlement(GHS)UrbanCentreDatabaseR2019(https://ghsl.jrc.ec.europa.eu/ghs_stat_ucdb2015mt_r2019a.php).Note:Citiesaredefinedasurbancentersfollowingthedegreeofurbanizationmethodology(seebox1.1,chapter1).Acityiscategorizedassmall,medium,orlargewhenitspopulationisbetween50,000and199,999;between200,000and1.499million;oratleast1.5million,respectively.TheincomeclassesarebasedontheWorldBankcountryclassificationsforfiscalyear2021/22.Low-incomecountriescorrespondtolow-andlower-middle-incomecountriesandmiddle-incomecountriestoupper-middle-incomecountries.Figure2.2Inglobaltypology,distributionofpopulation,bycitysizeandcountryincomegroup0100200300400500600700Low-incomeMiddle-incomeHigh-incomeLow-incomeMiddle-incomeHigh-incomeNo.ofpersons(millions)25.523.716.635.435.628.739.040.854.6020406080100Shareofpopulation(%)a.Numberofpersonsb.ShareofpopulationSmallMediumLargeTable2.2SummaryofglobaltypologyofcitiesIncomegroupCitysizeNumberofcitiesMediancitysizeMedianGDPpercapitaFivemostpopulouscitiesLow-incomeSmall3,913(38.5%)91,216US$2,426Pirganj(Bangladesh);Pleiku(Vietnam);Baraut(India);Turbat(Pakistan);Siguiri(Guinea)Medium1,297(12.8%)327,342US$3,022Abuja(Nigeria);Muzaffarpur(India);Raipur(India);Kharkiv(Ukraine);Monrovia(Liberia)Large134(1.3%)2,700,292US$3,879Jakarta(Indonesia);Delhi(India);Dhaka(Bangladesh);Mumbai(India);QuezonCity–Manila(Philippines)(Continued)132THRIVINGTable2.2continuedIncomegroupCitysizeNumberofcitiesMediancitysizeMedianGDPpercapitaFivemostpopulouscitiesMiddle-incomeSmall2,538(25.0%)85,346US$7,087Wenshan(China);Shanghai(China);Gangyu(China);Xinchang(China);CaboFrio(Brazil)Medium820(8.1%)357,205US$8,475Basra(Iraq);Huizhou(China);Pretoria(SouthAfrica);Liuzhou(China);Yinchuan(China)Large94(0.9%)3,087,354US$11,118MexicoCity(Mexico);SãoPaulo(Brazil);Beijing(China);Bangkok(Thailand);Istanbul(Türkiye)High-incomeSmall980(9.6%)82,763US$19,723Arnhem(TheNetherlands);SantaClarita(UnitedStates);Appleton(UnitedStates);MorenoValley(UnitedStates);Newcastle(Australia)Medium327(3.2%)359,117US$23,729SanAntonio(UnitedStates);Gwangju(RepublicofKorea);Valencia(Spain);Brussels(Belgium);Baltimore(UnitedStates)Large66(0.6%)3,118,764US$28,438Tokyo(Japan);Seoul(RepublicofKorea);NewYork(UnitedStates);Osaka(Japan);LosAngeles(UnitedStates)Overall10,169111,958US$5,152Jakarta,Tokyo,Delhi,Dhaka,MumbaiSource:WorldBankanalysisbasedondatafromtheEuropeanCommission’sGlobalHumanSettlement(GHS)UrbanCentreDatabaseR2019(https://ghsl.jrc.ec.europa.eu/ghs_stat_ucdb2015mt_r2019a.php).Note:Citiesaredefinedasurbancentersfollowingthedegreeofurbanizationmethodology(seebox1.1,chapter1).Typesofcitiesaredefinedusingthemethodologydescribedinbox2.1andannex2A.Percentagesaretheshareofthetotalnumberofcitiesusedtoderivethetypologyofcities(10,169).GDPpercapitaismeasuredinconstant2007USdollarsatpurchasingpowerparityexchangerates.Middle-incomecountriesChinahosts52percentofthe3,452citiesinthiscategory.ApartfromChina,citiesinLatinAmericaandtheCaribbeanaccountfor25percentofthisgroup(901cities),followedbycitiesinEuropeandCentralAsia,whichaccountforabout14percentofthisgroup(figure2.1).ExceptforIraqandSouthAfrica,countriesintheMiddleEastandNorthAfricaandinSub-SaharanAfricamakefewcontributionstothisgroup.Themiddle-incomecountrygroupconsistsofthefollowingcategories:••Middle-incomesmallcities(type4)arehometoover245millionpeoplein2,538cities.Amongthosepeople,132million(54percent)arespreadover1,343Chinesecities,whichcanberoughlycharacterizedasthelowest-tiercitiesandtownswithagriculturalbasesoraculturalorecologicalconnotation,orwithanintegratedfunctionofindustrialproduction,tourism,andresidence(suchasWenshanandXinchang).BecauseoftheChinesegovernment’sefforttomakesmallcitiesmoreattractiveforruralmigrantsby,forexample,eliminatingthehukou(householdregistration)system,thesecitiescouldexpandrapidlyinthecomingyears(Zhang2019).OutsideofChina,type4citiesincluderegionalmarkettowns(suchasItabuna,Brazil),industrialcitiesinformerSovietstates(suchasBabruysk,Belarus;KaragandaandOskemen,Kazakhstan;andKomsomolsk-on-Amur,RussianFederation),AGlobalTypologyofCities133tourismdestinationsintheEastAsiaandPacificandLatinAmericaandtheCaribbeanregions(suchasTaiping,Malaysia,andCaboFrio,Brazil),andsmallcitiesspecializinginprimaryproductionandprocessing(suchasRioVerde,Brazil,andTulu,Colombia).••Middle-incomemediumcities(type5)representabout12percentofthepopulationintheglobalsample.Aswithsmallcities,Chinesecitiesmakeupbyfarthelargestportionofthisgroup,accountingfornearly50percentoftheaggregatepopulation.NotableexamplesareAnshan,Liuzhou,andYantai,allofwhichserveasregionalindustrialcenters.Similarly,majorindus-trialcitiesinothercountries,includingnationalcapitalswithstrongmanufacturingbases,fallintothisgroup.SuchcitiesincludeBucaramanga,Colombia;Campinas,Brazil;Córdoba,Argentina;Gaziantep,Türkiye;Novosibirsk,Russia;Toluca,Mexico;andthecapitalcitiesNur-Sultan,Kazakhstan,andYerevan,Armenia.••Middle-incomelargecities(type6)house422millionpeople,oraround41percentofthepopulationofcitiesinmiddle-incomecountries,inonly94cities(seetable2.2andfigure2.2).3Althoughsometype6cities,suchasBeijing,havealreadyreachedapostindustrialphasewitheconomiesdominatedbythetertiarysector,mostremaininthetransitionstageoraskeyplayersinmanufacturingandheavyindustry.Likelargecitiesinlow-incomecoun-tries,justoverhalfoflargecitiesinmiddle-incomecountriesarenearcoastlines(figure2.3).Sources:WorldBankanalysisbasedondatafromtheEuropeanCommission’sGlobalHumanSettlement(GHS)UrbanCentreDatabaseR2019(https://ghsl.jrc.ec.europa.eu/ghs_stat_ucdb2015mt_r2019a.php)andWorldCoastlinepolylinedata(https://datacatalog.worldbank.org/int/search/dataset/0038272/World-Bank-Official-Boundaries).Note:Acoastalcityisdefinedasonewhosegeographiccenteriswithin100kilometersofthenearestcoastline.Thepercentageofcoastalcitiesiscalculatedfromthetotalnumberofcitiesbytype.Citiesaredefinedasurbancentersfollowingthedegreeofurbanizationmethodology(seebox1.1,chapter1).Typesofcitiesaredefinedusingthemethodologydescribedinbox2.1andannex2A.Figure2.3Inglobaltypology,distributionoflargecoastalandinlandcities,bylocationandcountryincomegroup,andshareofcoastalcities,bycitysizeandcountryincomegroup134THRIVINGHigh-incomecountriesThisgroupconsistsof1,373citiesin53countries.Although76percentofthesecitiesareinWesternEuropeandNorthAmerica,citiesfromoil-richcountriesinLatinAmericaandtheCaribbeanandtheMiddleEastandNorthAfricaalsomakeuppartofthisgroup(figure2.1).Citiesinhigh-incomecountriesexhibittwodistinctivefeaturesrelativetolow-andmiddle-incomecountrycities.First,morepeopleareconcentratedinlargecities:55percentofthepopulationinhigh-incomecountrieslivesinlargecitiesasopposedtoabout40percentinlower-incomecountries(figure2.2).Second,morecitiesarelocatednearcoastlines.Whereasthemajorityoflow-andmiddle-incomecitiesaremorethan100kilometersawayfromcoastlines,theoppositeistrueforhigh-incomecities(figure2.3).Thehigh-incomecountrygroupconsistsofthefollowingcategories:••High-incomesmallcities(type7)arehometoover93millionpeople,accountingforabout17percentofthepopulationofcitiesinhigh-incomecountries(figure2.2).MostofthesecitiesareinWesternEuropeandNorthAmerica(77percent)andcanbebroadlycharacterizedasprovincialorregionalcapitalsthatarepartofmidsizemetropolitanareaswithaggregatepopu-lationsoflessthanamillion.NotableexamplesareArnhem,theNetherlands;Appleton,UnitedStates;Clermont-Ferrand,France;Heidelberg,Germany;Messina,Italy;andRegina,Canada.Outsideofthesetworegions,thefollowingcitieshavesimilarcharacteristics:Abha,SaudiArabia;Beersheba,Israel;andNewcastle,Australia.••High-incomemediumcities(type8),whichhostsome162millionpeoplein327citiesglobally,arealsomostlylocatedinWesternEuropeandNorthAmerica(seetable2.2andfigure2.2,panela).Thesecitiescanbecharacterizedasthecorecityandneighboringmunicipalitiesofmetropolitanareas.Dependingonthesizeofthecountry,thecorecitycanbethenationalcapital,aprovincialcapital,oramajorregionalcity.ExamplesarethecapitalregionsofnorthwesternEuropeancountries(exceptFrance,Germany,andtheUnitedKingdom)and,outsideofEurope,Auckland,NewZealand;Manama,Bahrain;andMontevideo,Uruguay.UScitiessuchasBaltimore,Orlando,SanAntonio,andsomecitiesintheMidwestarepartofthiscategoryaswell.AmongthemostpopulouscitiesinthiscategoryareGwangju,RepublicofKorea;Valencia,Spain;Mecca,SaudiArabia;Liverpool,UnitedKingdom;Perth,Australia;Hiroshima,Japan;andLyon,France.••High-incomelargecities(type9)host307millionpeoplein66citiesglobally,largelyinEastAsiaandPacific,NorthAmerica,andWesternEurope(seefigure2.1andfigure2.2,panela).Thecitiesrangefromnationalorregionaleconomiccenterswithpopulationsofjustover1.5million(forexample,Cologne,Germany;Doha,Qatar;andKatowice,Poland)toglobalmegacitieshosting10millionormorepeople,includingLondon,LosAngeles,NewYork,Osaka,Paris,Seoul,andTokyo.Thesecitiesappeartobeathighriskofcoastalhazardsbecausetheyarelocatedclosetocoastlines(figure2.3).HowcurrentchallengesvaryacrossdifferenttypesofcitiesUsingthemethodologyoutlinedinstep3ofbox2.1,table2.3providesanoverviewofhowtheseverityofcurrenturbanchallengesrelatedtogreenness,resilience,4andinclusivenessvaryacrosstheninetypesofcities.Theupperpartofthetableemploysatrafficlightsystem(red,yellow,andgreen)todistinguishthelevelofrelativeseverity(high,moderate,andlow)ofeachtypeofchallenge.Thelowerpartofthetableliststhechallengesidentifiedassevereandmoderateforatypicalcityofeachtype.Forexample,inahigh-incomecountrytakeatypicallargecitywithapopulationofatleast1.5millionsuchasAtlantaintheUnitedStatesorTokyoinJapan.Relativetoothertypesofcitiesglobally,suchacityfacesseverechallengesAGlobalTypologyofCities135Table2.3Inglobaltypology,severityofchallenges,bycitysizeandcountryincomegroupIncomegroupLow-incomeMiddle-incomeHigh-incomeCitysizeSmallMediumLargeSmallMediumLargeSmallMediumLargeGCarbonPollutionVegetationRResilienceIPovertyInequalityServicesSeverechallengesResiliencePovertyServicesResiliencePovertyServicesPollutionResilienceServicesPovertyInequalityVegetationInequalityCarbonPollutionVegetationInequalityCarbonInequalityModeratechallengesPollutionInequalityPollutionCarbonVegetationPovertyPollutionVegetationResilienceServicesCarbonPollutionResiliencePovertyServicesResiliencePovertyServicesCarbonResilienceInequalityCarbonVegetationResilienceInequalityPollutionVegetationSeverity:HighModerateLowSources:WorldBankanalysisbasedonthefollowingsources:(1)carbondioxideemissions,pollution,andvegetation:EuropeanCommission,GlobalHumanSettlement(GHS)UrbanCentreDatabaseR2019(https://ghsl.jrc.ec.europa.eu/ghs_stat_ucdb2015mt_r2019a.php);(2)resilienceindicator:VisibleInfraredImagingRadiometerSuite(VIIRS)nighttimelightssatellitedata(https://payneinstitute.mines.edu/eog-2/viirs/);InternationalBestTrackArchiveforClimateStewardshiptropicalcyclonedata(https://www.ncdc.noaa.gov/ibtracs/);(3)inequalityindicator:compiledfromRoberts,GilSander,andTiwari2019forIndonesia;BehrensandRobert-Nicoud2015fortheUnitedStates;FerreyraandRoberts2018for16countriesinLatinAmericaandtheCaribbean;and(4)basicservicesindicators:HendersonandTurner2020andhttps://doi.org/10.7910/DVN/YZ46FJ.Note:Citiesaredefinedasurbancentersfollowingthedegreeofurbanizationmethodology(seebox1.1,chapter1).Typesofcitiesaredefinedusingthemethodologydescribedinbox2.1andannex2A.G=greenness;I=inclusiveness;R=resilience.136THRIVINGinreducingthesizeofitsoverallcarbonfootprintandintracityinequality,5andmoderatechallengesinreducingairpollutionandexpandingitsgreencover.Bycontrast,inalow-incomecountryatypicalsmallcitywithapopulationofbetween50,000and199,999facesseverechallengesinreducingpoverty,increasingaccesstobasicservices,andstrengthen-ingeconomicresiliencetoclimatechange–relatedshocksandstresses,aswellasmoderatechallengesintacklingairpollutionandinequality.Forsmallandmediumcitiesinlow-incomecountries,povertyandaccesstobasicservicesareseverechallenges,dampeningresilienceagainstclimatechange–relatedstressesaswellAsdiscussedinchapter1,urbanareastendtoprovidebetterstandardsoflivingandaccesstobasicservicesthanruralareasdo.Yet,forcitiesinlow-incomecountries,thechallengesasso-ciatedwithlimitedaccesstoservicesremainsignificant,especiallyinsmallandmediumcities.Forcountrieswithadequatedata,evidencealsosuggeststhatpovertyinlow-incomecountriesisdeeperandmoreprevalentinsmallandmediumcitiesthaninlargecities.Povertyisnormallycompoundedbyseriousdeprivationinaccesstobasicservices.Analysisofcitiesin41low-andmiddle-incomecountriesrevealsthataccesstoimprovedsanitationismorepressingforsmallandmediumcitiesinlow-incomecountries,whereasaccesstosafelymanageddrinkingwaterisacommonseverechallengeforcitiesofdifferentsizesinlow-incomecountries.Thus,20percentofhouseholdsinsmallandmediumcitiesstilllackaccesstoimprovedsanitation,andsafelymanageddrinkingwaterisunavailableforabout30percentofhouseholds(figure2.4).Partlybecauseofinadequatebasicservicesandinfrastructuremoregenerally,smallandmediumcitiesinlow-incomecountriessuffermoreseverelythanothertypesofcitieswhenconfrontedbyweathervariations,evenroutinechangesinweather.Thus,whenatypicalmediumcityinalow-incomecountryishitbyanyweatherthatishotter,wetter,ordrierthanusual,thecity’soveralllevelofeconomicactivity,asmeasuredbynighttimelightsintensity,fallsbynearly1.5percent(figure2.5,panela).Theaveragesizeofimpactreachesnearly26percentwhenatropicalcyclonehitsthecountry.6Becauseoftheirlimitedresourcesandcapacitytoadapttoemergingchallenges,smallcitiesinlow-incomecountriesarealsolikelytofindresiliencetoclimatechange-relatedstressesaseverechallenge.7Foratypicalsmallormediumcityinalow-incomecountry(whichismostlikelyinanearlystageofindustrializationorstillheavilydependentontheprimarysector),greenhousegasemissions,measuredbyproduction-basedfossilcarbondioxide(CO2)emissions(figure2.6,panelsaandb),andlackofgreenspace(figure2.6,panelseandf)arelessurgentchallengesthanforlargercitiesorthoseinhigher-incomecountries.LowerCO2emissionsandmoregreenspacedonotnecessarilytranslateintocleanair,however,inpartbecauseoftheunaffordabil-ityofcleanerfuelsforcooking,heating,andlighting,aswellaslessstringentandlesswell-enforcedregulationsonairqualityandvehicleemissions.Thus,pollutionposesamoderatechallengeforsmallandmediumcitiesinlow-incomecountries.TheiraverageannualPM2.5(particulatematterof2.5micronsorlessindiameter)concentrationsofmorethan40micro-gramspercubicmeter(μg/m3)ofair(figure2.6,paneld)farexceedthesafeairguidelinesoftheWorldHealthOrganization(WHO).8Finally,analysisof4,574citiesfromIndonesia,theUnitedStates,and16countriesinLatinAmericaandtheCaribbean9identifiesintracityinequality,measuredbytheGinicoefficientofincomeorconsumption,asamoderatechallengeforsmallcitiesinlow-incomecountriesAGlobalTypologyofCities13760708090100SmallMediumLargeSmallMediumLargeLow-incomeMiddle-incomeShareofhouseholds(%)60708090100SmallMediumLargeSmallMediumLargeLow-incomeMiddle-incomeShareofhouseholds(%)a.Accesstoimprovedsanitationb.Accesstosafelymanageddrinkingwater0.30.40.50.6SmallMediumLargeSmallMediumLargeSmallMediumLargeLow-incomeMiddle-incomeHigh-incomeGinicoecientc.InequalityFigure2.4Inglobaltypology,levelofseverity,bycitysizeandcountryincomegroup:InclusivenessindicatorsSources:WorldBankanalysisbasedonthefollowingsources:(1)inequalityindicator:compiledfromRoberts,GilSander,andTiwari(2019)forIndonesia;BehrensandRobert-Nicoud(2015)fortheUnitedStates;FerreyraandRoberts(2018)for16countriesinLatinAmericaandtheCaribbean;(2)servicesindicator:HendersonandTurner(2020)andhttps://doi.org/10.7910/DVN/YZ46FJ.Note:Fortheserviceindicators(accesstoimprovedsanitationandaccesstosafelymanageddrinkingwater),citiesaredefinedasurbancentersfollowingthedegreeofurbanizationmethodology(seebox1.1,chapter1).Forinequality,citiesaredefinedas(1)metropolitanstatisticalareasfortheUnitedStates;(2)thetypesofdistricts(level-2administrativeunits)derivedfromRoberts,GilSander,andTiwari(2019)forIndonesia;and(3)urbanclustersfollowingtheclusteralgorithm(DijkstraandPoelman2014).Typesofcitiesaredefinedusingthemethodologydescribedinbox2.1andannex2A.Themeasureofinequality,theGinicoefficient,iscalculatedonthebasisofpercapitaexpenditureforIndonesiaandofincomeelsewhere.Thegreen,lightgreen,andbluebarsindicatelow,moderate,andhighlevelsofseverity,respectively.138THRIVINGSources:WorldBankanalysisbasedondataderivedfrommonthlycompositesofVisibleInfraredImagingRadiometerSuite(VIIRS)nighttimelightssatellitedata(https://payneinstitute.mines.edu/eog-2/viirs/),monthlyweatherdatafromClimatologyLab,TerraClimate(https://www.climatologylab.org/terraclimate.html),andtropicalcyclonedatafromInternationalBestTrackArchiveforClimateStewardshiptropicalcyclonedata(https://www.ncdc.noaa.gov/ibtracs/).Note:Theelasticityforeachtypeofcityiscalculatedastheunweightedaverageelasticityacrossthehot,wet,anddryanomaliesinpanela,andacrosstheextremehot,extremewet,andextremedryanomaliesaswellastropicalcyclonesinpanelb.Becausethepurposeofthissectionistoassesstheseverityofurbanchallenges,thecalculationsexcludecoldanomaliesandextremecoldshocks,theelasticitiesofwhicharepositiveforalltypesofcities.Citiesaredefinedasurbancentersfollowingthedegreeofurbanizationmethodology(seebox1.1,chapter1).Typesofcitiesaredefinedusingthemethodologydescribedinbox2.1andannex2A.Thegreen,lightgreen,andbluebarsindicatelow,moderate,andhighlevelsofseverity,respectively.Figure2.5Inglobaltypology,levelofseverity,bycitysizeandcountryincomegroup:Resilienceindicators–0.015–0.010–0.00500.005MediumLargeMediumLargeMediumLargeLow-incomeMiddle-incomeHigh-incomeElasticity–0.12–0.08–0.0400.04MediumLargeMediumLargeMediumLargeLow-incomeMiddle-incomeHigh-incomeElasticityb.Elasticitytoextremeweathershocksa.Elasticitytoweatheranomalies(figure2.4,panelc).10Suchinequalitycanpotentiallyhaveadverseconsequencesincludinglowlocaleconomicgrowth,highcrime,andsocialunrest(FerreiraandSchoch2020;Glaeser,Resseger,andTobio2009).Thisfinding,albeitbasedondatawithlimitedgeographiccoverage,isdrivenbysmallcitiesinLatinAmericaandtheCaribbean,aregionwell-knownforitshighlevelsofincomeinequality(BussoandMessina2020;FerreyraandRoberts2018).Largecitiesinlow-incomecountriesfaceseverechallengesinreducingpollution,strengtheningresilience,andimprovingaccesstoservicesLargecitiesinlow-incomecountriesstrugglewithpoorairquality,alackofresilience,andinadequateaccesstobasicservices.Thus,ofalltypesofcities,theyrecordthesecond-highestlevelofseverityintermsofbothPM2.5emissionsand(lackof)airquality,asmeasuredbytotalPM2.5concentration(figure2.6,panelscandd).Whenhitbyextremeweatherevents,thistypeofcityalsoappearsevenmorevulnerablethansmallercities(figure2.5,panelb).And,despiteAGlobalTypologyofCities1390102030SmallMediumLargeSmallMediumLargeSmallMediumLargeLow-incomeMiddle-incomeHigh-incomeMilliontonnesperyear0246SmallMediumLargeSmallMediumLargeSmallMediumLargeLow-incomeMiddle-incomeHigh-incomeTonnespercapitab.PercapitafossilCO2emissionsa.TotalfossilCO2emissions048121620SmallMediumLargeSmallMediumLargeSmallMediumLargeLow-incomeMiddle-incomeHigh-incomeThousandtonnesperyear01020304050SmallMediumLargeSmallMediumLargeSmallMediumLargeLow-incomeMiddle-incomeHigh-incomeµg/m3d.TotalPM2.5concentrationsc.TotalPM2.5emissionsFigure2.6Inglobaltypology,levelofseverity,bycitysizeandcountryincomegroup:Greennessindicators(Continued)140THRIVINGFigure2.6continued0.300.350.400.45SmallMediumLargeSmallMediumLargeSmallMediumLargeLow-incomeMiddle-incomeHigh-incomeIndexvalues0.00.20.40.6SmallMediumLargeSmallMediumLargeSmallMediumLargeLow-incomeMiddle-incomeHigh-incomeShareofhighgreenareaf.Shareofhighgreenareae.AveragegreennessindexSource:WorldBankanalysisusingdatafromtheEuropeanCommission’sGlobalHumanSettlement(GHS)UrbanCentreDatabaseR2019(https://ghsl.jrc.ec.europa.eu/ghs_stat_ucdb2015mt_r2019a.php).Note:Citiesaredefinedasurbancentersfollowingthedegreeofurbanizationmethodology(seebox1.1,chapter1).Typesofcitiesaredefinedusingthemethodologydescribedinbox2.1andannex2A.Thegreen,lightgreen,andbluebarsindicatelow,moderate,andhighlevelsofseverity,respectively.CO2=carbondioxide;µg/m3=microgrampercubicmeter;PM2.5=particulatematterof2.5micronsorlessindiameter.betteraccesstoimprovedsanitationinlargecities,accesstosafelymanageddrinkingwaterisalmostaspoorasinsmallcitiesinlow-incomecountries(figure2.4,panelb).Basedonthediscussioninchapter1,itcanalsobeassumedthat,althoughextremepovertymaybelessprevalentinlargercitiesthaninsmallercities,significantpocketsneverthelesscontinuetoexistinlargecitiesinlow-incomecountries.Unlikesmallercitiesinlow-incomecountries,largecitiesinthesecountriesalsomakequitesignificantcontributionstoglobalwarming.Thus,eventhoughtheirlevelsofpercapitafossilCO2emissionsremainfarlower,theiraggregateemissionsexceedthoseofsmallandmediumcitiesinhigher-incomecountries(figure2.6,panela).Becauseofpoorlymanageddevelopmentpressures,largecitiesinlow-incomecountriesexperiencesignificantlygreaterstressesongreenspacethandosmallercitiesinthosecountries(figure2.6,panelseandf).Smallcitiesinmiddle-incomecountriesexperiencesevereinequality,andpoverty(presumably)persistsSevereincomeinequalityisanacuteissueforsmallcitiesinmiddle-incomecountries.AnalysisofcitiesinLatinAmericaandtheCaribbeanrevealsthattheaverageGinicoeffi-cientofincomeforthistypeofcityis0.46,whichishigherthanthatofcitiesinhigh-incomeAGlobalTypologyofCities141countries(figure2.4,panelc).CitiesinBrazil,Colombia,andCostaRicastandoutinthisregard.Also,citiesofthistypelikelyfaceseverechallengesdealingwithpovertybecausetheypredominatelyconsistofthelowest-tierChinesecitiesandtowns—whichrepresentlessdevelopedpartsofChina—aswellassmallcitiesinLatinAmericaandtheCaribbeanandinformerSovietstates(Fay2005;Slay2009;Xuetal.2021).Pollution,lackofvegetation,andinadequateaccesstobasicservices,meanwhile,presentmoderatechallengesforthistypeofcity.Althoughairpollutionisnotassevereasincitiesinlow-incomecountries,theoveralllevelofPM2.5concentratedintheair(≈32μg/m3)isstillcon-sideredunsafeaccordingtoWHO’sguideline.Likewise,asboththeaveragelevelofgreennessandtheshareofhighgreenareaindicate,greenspaceismuchlessprevalentthaninsmallcitiesinlow-incomecountries(figure2.6,panelseandf).Thisfindingisconsistentwithashrink-ageofsuchspaceincitiesascountriesmovefromlow-tomiddle-incomestatus.Intermsofbasicservices,about20percentofhouseholdsstilldonothaveaccesstosafedrinkingwater,accordingtoanalysisdrawingonasubsampleofcitiesfrom41low-andmiddle-incomecoun-tries(figure2.4,panelb).Judgingbytheresultsformediumcitiesinmiddle-incomecoun-tries(figure2.5,panelb),smallcitiesinthosecountriesalsolikelyfacemoderatechallengesinstrengtheningeconomicresiliencetoclimatechange–relatedshocksandstresses.Mediumandlargecitiesinmiddle-incomecountriesfaceseveregreennesschallengesAsdiscussedearlier,mediumandlargecitiesinmiddle-incomecountriesmainlyrelyonthesecondarysector.Thus,thesetypesofcitieshaverelativelyhighlevelsofcarbonemissionsandpollution,togetherwithlessgreenspace.Largecitieshaveamuchhigherlevelofseverity,whichisnotsurprisingbecausethesecitiesincludekeyplayersinmanufacturingandheavyindustrydomesticallyaswellasglobally.Evenwithouttwoextremeoutliers(GuangzhouandShanghai),largecitiesinmiddle-incomecountriesemit,onaverage,abouttwo-thirdsasmuchfossilCO2as,andfarmorePM2.5than,citiesofequivalentsizeinhigh-incomecountries(figure2.6,panelsaandc).Meanwhile,mediumcitiesinmiddle-incomecountrieshavepercapitafossilCO2emissionsapproachingthoseoflargecitiesinthosecountries,althoughtheirrelativelysmallsizesallowtotalemissionstoobscurethisfinding.Airqualityisbynomeanssatisfactoryineithermediumorlargecitiesinmiddle-incomecountries.Moreover,witheconomiesgenerallyspecializinginland-intensiveactivities,thesetypesofcitiesfaceaveryseriouslackofgreenspace,whichisparticularlypronouncedinlargecities(figure2.6,panelseandf).Atthesametime,largecitiesinmiddle-incomecountriesarethemostunequalamongalltypesofcities,followedbymediumcitiesinthosecountries.TheyhaveaverageGinicoefficientsof0.52and0.49,respectively,bothhigherthanthe0.46forlargecitiesinhigh-incomecountries(figure2.4,panelc).Likewise,Chen,Liu,andLu(2017)findsevereincomeinequalityinlargerChinesecitiesmainlybecausethesecitiestendtohavemorelow-skilled(low-income)migrants.Thesefindingssuggestthatmediumandlargecitiesinmiddle-incomecountriesmuststilladdresspocketsofpoverty.Meanwhile,thesetypesofcitiesfacemoderatechallengesinprovidinguniversalbasicservices(figure2.4,panelsaandb).142THRIVINGSmallandmediumcitiesinhigh-incomecountriesfarebestoverall,butstillfacemoderatechallengesrelatedtocarbonemissions,resilience,andinequalityAsdiscussedearlier,smallandmediumcitiesinhigh-incomecountriesincludesecond-arycitiesinlargerdevelopedcountriesandthenationalcapitalsofsmallercountries.Thesecitiestendtoberelativelywellgovernedandhaveaccesstoadequateresourcesandtechni-calcapacity,makingthemlesssusceptiblethancitiesinlower-incomecountriestogreenness,resilience,andinclusivenesschallenges.Nevertheless,theyhavemoderatelevelsofCO2emis-sionsperperson,mainlyfromtransportation-relatedactivitiesandelectricityconsumption(figure2.6,panelb;seethesourcesofemissionsdiscussedinchapter1aswell).Moreover,mediumcitiesinhigh-incomecountriesareeconomicallylessresilienttoweathershocksthantheirlargercounterparts.Unlikeforlargecities,theaverageestimatedimpactofweatheranomaliesoneconomicactivityissignificantlynegativeformediumcitiesregard-lessofwhetherthoseanomaliesareextreme(figure2.5).Althoughfurtherinvestigationoftheunderlyingmechanismsbehindtherelationshipbetweencitysizeandeconomicresilienceisbeyondthescopeofthisreport,previousstudiespointoutthatresilienceplanningandactionareoftenbuiltaroundtheneedsandinterestsoflargemetropolitanareas,andtendtooverlooksmallercities.Thus,eveninhigh-incomecountries,smallandmediumcitieslikelyhavelimitedaccesstothenecessaryfinancialandhumanresources,aswellasweakergovernance,con-strainingtheirpreparednessandadaptivecapacityforclimateshocksandstresses.11Finally,intracityincomeinequalitypresentsacommonmoderatechallengeforsmallandmediumcitiesinhigh-incomecountries.Becauseoftheirlevelofdevelopment,thesecitieslikelyhostpopulationswithmoreuniformlevelsofeducationandskillrelativetothoseincitiesofsimilarsizeinmiddle-incomecountries.Nevertheless,smallcitiesinhigh-incomecountriesremainquiteunequal,asreflectedintheaverageGinicoefficientof0.41calcu-latedusingthedatafor363USmetropolitanstatisticalareas.Mediumcitiesexhibithigherlevelsofincomeinequality—nearlyashighasthoseoflargercitiesinthesample(figure2.4,panelc).Largecitiesinhigh-incomecountriesgrapplewithsevereinequality,alongsidepersistentlyhighCO2emissionsLargecitiesinhigh-incomecountrieshavebyfarthehighestlevelsoffossilCO2emissionsbothinabsolutetermsandonapercapitabasis(figure2.6,panelsaandb).Althoughtheireconomicactivitieshavemostlyshiftedtohumancapital–intensivetertiaryindustry,theystillemitthegreatestquantitiesofCO2,largelygeneratedbytransportation-relatedactivities,aswellasbyenergyconsumptionbyresidentialandcommercialbuildings(seethediscussioninchapter1).Inaddition,largecitiesinhigh-incomecountrieshaveveryhighlevelsofintracityinequality.AlthoughnotashighasinmediumandlargecitiesinLatinAmericaandtheCaribbean,theaverageGinicoefficientacrosslargecitiesinhigh-incomecountries,representedbylargeUSmetropolitanareas,is0.46(figure2.4,panelc).Furthermore,amonghigh-incomecountries,ahighlevelofincomeinequalityinlargecitiesisnotuniquetotheUnitedStates.Boulant,Brezzi,andVeneri(2016)andCastells-Quintana,Royuela,andVeneri(2020)findevidencethatincomeinequalitytendstoincreasewithcitysizeforamoregeneralsampleofOrganisationforEconomicCo-operationandDevelopmentcountriesaswell.12AGlobalTypologyofCities143Comparedwithcitiesofcomparablesizeinmiddle-incomecountries,however,largecitiesinhigh-incomecountriesfindpollutionandlackofvegetationonlymoderatelychallenging.AlthoughtheaverageamountofPM2.5theyemitisnottrivial,itisroughlyathirdoftheaverageamountemittedbylargecitiesinmiddle-incomecountries.Airquality(measuredbyPM2.5concentration)appearstobesignificantlybetterthanthatinlower-incomecities,althoughitstilldoesnotmeetWHO’ssafetystandard.Consistentwiththisfinding,thesecitieshavemoregreenspacethancomparablysizedcitiesinmiddle-incomecountries,althoughtheyremainlessgreenintermsofvegetationthansmallcitiesinhigh-incomecountries(figure2.6,panelseandf).Howclimatechange–relatedhazardsvaryacrosscitiesgloballyFollowingonthepreviousoverviewofatypicalcityofeachtypeandtheseverityofitscurrenturbanchallengesrelatedtogreenness,resilience,andinclusiveness,thissectionusesthesameninetypesofcitiestolookateachtype’sexposuretosixclimatechange–relatedhazards—floods,heatstress,tropicalcyclones,sealevelrise,waterstress,andwildfires—aswellasitsweightedexposuretocombinedclimatechange–relatedhazards.Thedata,providedbyMoody’sESGSolutions,13includecity-levelscoresonascaleofzeroto100,fromlesstomoreexposed(table2.4).Thedatasetcoversmorethan2,200citiesfromthefullglobalsampleofmorethan10,000citiesexaminedinthisreport.Figure2.7showsthedistributionofcities,bytype,intheMoody’sdataset.Exposurescoresforeachcityarecalculatedbybringingtogetherspatiallyexplicitphysicalhazardandsocioeconomicdataprojectedto2030–40.Forphysicalhazarddata,estimatesarebasedonthehighestemissionpathway—RepresentativeConcentrationPathway8.5.Emissionspathwaysdonotvarysignificantlywithinthetimeframeselectedbecauseofthedelayedimplicationsofpotentiallydivergentpolicyresponses,makingtheselectionappropri-atefornear-termplanningpurposes.Forsocioeconomicdata,estimatesofpotentialexposurearederivedfromspatiallyexplicitdataofacity’simportantassets,includingpopulation,grossdomesticproduct(measuredusingexchangeratesadjustedforpurchasingpowerparity),andagriculturalarea.SharedSocioeconomicPathway2(SSP2),amiddle-of-the-roadSSP,isusedfor2040socioeconomicprojections.SSP2wasselectedasthemostrealisticdepictionoffuturegrowthpatternsbecause,oftheSSPs,itmakesthefewestoverallassumptions.14Althoughthequalityandbreadthofclimateandsocioeconomicdatahaveimproveddramati-callyinrecentdecades,highlevelsofuncertaintyandcoveragegapsonamorelocalizedscaleremain,limitingtheabilityofcitystakeholderstomakeinformedpolicyandinvestmentdeci-sions.Furthermore,althoughthescoresprovideabaselineunderstandingofsocioeconomicexposuretoclimatechange–relatedhazards,theydonotaccountforvulnerability—conditionsthatincreaseone’ssusceptibilitytotheimpactofahazard—andsoshouldnotbeconflatedwithclimatechange–relatedrisk.Vulnerabilitytoclimatechange–relatedhazardscanbeinformedbyseveralfactors,suchassensitivitytoharmandcapacitytocopeandadapt,thattogetherconsti-tuteanimportantcomponentofclimaterisk(IPCC2022b).Thus,highexposurescoresmaynotnecessarilytranslateintoasignificantimpactoncitiesandtheirpopulations.Thisinformationgaphighlightstheimportanceofongoinginvestmentindatacollectionthatwouldfacilitateabetterunderstandingofclimatechange–relatedvulnerabilityandriskatthecitylevel.Forthisanalysis,sevenindicators—thesixidentifiedclimatechange–relatedhazardsplustheweightedoverallhazard—areregressedontheninetypesofcities.Toensurethattheanalysisincludesonlythecities’relevanthazards,itexcludeshazardsidentifiedasnotaffectingacity.144THRIVINGTable2.4Methodologicalreferencetableforprojectedclimatechange–relatedexposurescores,2030–40NoexposureLowexposureMediumexposureHighexposureRedflagCombinedclimatechange–relatedhazardAverageofallexposurescoresn.a.Notsignificantlyexposedtohistoricaland/orprojectedhazardsExposedtosomehistoricaland/orprojectedhazardsExposedtohighhistoricaland/orprojectedhazardsExposedtoextremelyhighhistoricaland/orprojectedhazardsFloodsOne-in-100-yearflood(rainfall-andriverine-based)Wetdays(>10millimeters)Verywetdays(>95thpercentile)Rainfallintensityn.a.Populationminimallyexposedtoflooding;futurerainfallintensificationPopulationmoderatelyexposedtoflooding;futurerainfallintensificationPopulationhighlyexposedtoflooding;futurerainfallintensificationPopulationextremelyhighlyexposedtoflooding;futurerainfallintensificationHeatstressEnergydemandExtremeheatdaysExtremetemperaturen.a.ChangesinheatextremesarerelativelylesssevereChangesinheatextremesarewithinrangeofglobalaverageChangesinheatextremesarerelativelymoresevereChangesinheatextremesaremuchmoresevereTropicalcyclonesCumulativewindspeedNoknownhistoricaloccurrencePossibletropicalcycloneactivitybutinfrequentand/orlesssevereInfrequentbutpossiblyseveretropicalcycloneactivityFrequentandpossiblyseveretropicalcycloneactivityFrequentseveretropicalcycloneactivitySea-levelriseOne-in-100-yearflood(coastal-based)NotcoastalornearcoastalwaterwaysPopulationminimallyexposedtocoastalfloodingPopulationmoderatelyexposedtocoastalfloodingPopulationhighlyexposedtocoastalfloodingPopulationextremelyhighlyexposedtocoastalfloodingWaterstressCurrentbaselinewaterstressCurrentinterannualvariabilityFuturewatersupplyanddemandWatersupplyanddemandchangen.a.Watersupplyand/ordemandchangesrelativelysmallWatersupplyand/ordemandchangeslikelytoincreaseCurrentwaterstresslikelyalreadyhigh,andsuppliesdiminishingCompetitionforwaterresourcesextreme,andfuturewatersupplyfailurepossibleWildfiresTotalandchangeindayswithhighwildfirepotentialTotalandchangeinmaximumwildfirepotentialNotburnablebasedonlandtypeLowwildfirepotential,littlechangeinseverity,relativelyfewerhigh-hazarddaysModeratewildfirepotential,withsomedegreeofchangeinfutureseverityHighwildfirepotentialwithsizableincreasesinfutureseverityVeryhighwildfirepotentialandatleastseveraladditionalweeksofhigh-hazarddaysSource:AdaptedfromMoody’sESGSolutions,Sub-SovereignPhysicalClimateRiskScoresMethodology,October2021,https://esg.moodys.io/climate-solutions.Note:n.a.=notapplicable.Wedonotpresentaspecificscorerangeassociatedwitheachlevelofexposure,becausetherangevariesbythetypeofhazard.AGlobalTypologyofCities145Theanalysisprovidespredictedestimatesofscoresforeachoftheninetypesofcities,conditionalonexposuretoanexistinghazard.Thisexerciseallowsexaminationofwhetheragiventypeofcityisexpectedtohavemoreorlessexposurethanothertypesofcitiestothevarioushazards.Underlyinggeographicfactors,suchasacity’sclimatezoneorproximitytocoastline,provideimportantcontextforunderstandingpotentialexposuretocertainclimatechange–relatedhazards(Lietal.2018).Althoughtherelationshipbetweencitytypologyandgeolocationisnotanalyzed,many“redflag”citiesareclusteredincoastalandtropicalareasandaremorelikelytobeseverelyexposedorhighlyexposedtooneormoreclimatechange–relatedhazards(map2.2).AnexampleofaredflagcityisSylhet,Bangladesh,whichishighlyexposedtobothheatstressandfloods.Climateshocksandstressescanacttogethertoexacerbatetheoverallimpact,eveninthemostresilientcities(Simpsonetal.2021;Zscheischleretal.2018).Overall,citiesinlow-incomecountriesaremoreexposedthanothercitiesofalltypestoclimatechange–relatedhazardsOverall,citiesinlow-incomecountries—types1,2,and3—aremoreexposedtoclimatechange–relatedhazards.Thesethreetypesofcitiesshowthehighestaveragescoresacrossallincomegroupsforsuchhazards(figure2.8).Whensegmentedbyhazard,floods,heatstress,sea-levelrise,andwildfiresdemonstrateasimilarexposurebiastowardcitiesinlow-incomecountries.Notableexceptionsarewaterstress,whichdoesnotshowanoticeabletrend,andtropicalcyclones,whichshowaninversetrend(figure2.9).Theresultshighlighttheadditionalchal-lengesthatmanylow-incomecitiesfaceinadaptingtoachangingclimatebecauseofapreex-istingexposurebiastoclimatechange–relatedhazards.Source:WorldBankanalysisbasedondatafromMoody’sESGSolutions,Sub-SovereignPhysicalClimateRiskScores,October2021(https://esg.moodys.io/climate-solutions).Note:Thepercentagesinparenthesesarecalculatedbyincomegroup.Citiesaredefinedasurbancentersfollowingthedegreeofurbanizationmethodology(seebox1.1,chapter1).Typesofcitiesaredefinedusingthemethodologydescribedinbox2.1andannex2A.Low-incomenumbersaddupto99ratherthan100duetorounding.Figure2.7Distributionofcitiesinclimatechange–relatedexposuredataset,bycitysizeandcountryincomegroup152(24%)361(58%)108(17%)433(45%)434(46%)85(9%)354(56%)222(35%)59(9%)02004006008001,000Low-incomeMiddle-incomeHigh-incomeNo.ofcitiesSmallMediumLarge146THRIVINGProportionally,largercitiesaremoreexposedtooverallclimatechange–relatedhazardsAteachincomelevel,atrendalsoshowsproportionallyhigheroverallclimatechange–relatedexposureforlargercities.Thistrendismostdefinedforcitiesinmiddle-incomecountries—types4,5,and6—withtheaveragescoremovingfrom51to56to61ascitysizeincreasesfromsmalltomediumtolarge(figure2.8).Whensegmentedbythetypeofhazard,thetrendsoftendiverge.Forfloods,resultsindicatethatinlow-incomecountrieslargecitieshavehigherlevelsofexposurethansmallcities,whereasinhigh-incomecountriessmallcitieshavehigherlevelsofexposurethanlargecities.Theresultshighlightthevastchallengesthatlargecitiesinlow-incomecountries—type3—faceintermsofmanagingfloodrisk(figure2.9,panela).Thesechallengescanbeexacerbatedwhenpairedwithrapidurbanizationandhighlighttheneedtosupportlargecitiesinlow-incomecountries,whichcanactasenginesofsustainablegrowthforcountriesandregionsintheareasofresilienturbanplanninganddevelopment.Sources:WorldBankanalysisbasedondatafromMoody’sESGSolutions,Sub-SovereignPhysicalClimateRiskScores,October2021(https://esg.moodys.io/climate-solutions),andtheEuropeanCommission’sGlobalHumanSettlement(GHS)UrbanCentreDatabaseR2019(https://ghsl.jrc.ec.europa.eu/ghs_stat_ucdb2015mt_r2019a.php).Note:Theexposureratingforeachcityisdeterminedusingthemethodologyoutlinedintable2.4.Citiesaredefinedasurbancentersfollowingthedegreeofurbanizationmethodology(seebox1.1,chapter1).Typesofcitiesaredefinedusingthemethodologydescribedinbox2.1andannex2A.Map2.2Geographicdistributionofoverallclimatechange–relatedexposure,bycitysizeandcountryincomegroupAGlobalTypologyofCities147Forwaterstress,citiesinmiddle-andhigh-incomeclassesfaceincreasinglyhigherlevelsofexposureastheybecomelarger,whereasthedistributionofexposureinlow-incomeclasscitiesisrelativelyuniformacrosscitysize(figure2.9,panelb).Thisfindinghighlightsthechallengesthatmanycitiesfaceinmanagingwaterresourcesforlargepopulationswhenshocks,suchasprolongeddrought,affectasystem.Watercrises,suchastheoneexperiencedbyCapeTownin2017and2018,maybecomeincreasinglycommoninthefaceofachangingclimate.Toadapt,largecities,particularlythosewithmoreresourcesinmiddle-andhigh-incomecountries,muststrengthentheresilienceoftheirwatersupplysystemsbyenhancingtheirrobustness—suchasbypromotingwaterreuseandsourcediversification—andstrengthentheirabilitiestocope.IntersectionwithurbanchallengesexacerbatestheimpactofclimateshocksandstressesWhenintersectedwithurbanchallenges,suchasthoseidentifiedintable2.3,citieswithhighexposurescoresaremorelikelytosustainsignificantdamageorimpactsfromclimatechange–relatedshocksandstresses.Thisanalysisfindsproportionallyhigherlevelsofexposureforlow-incomeandlargecitytypologies.Unfortunately,thesamecitytypologiesalsofacehigherlevelsofseverityformanyurbanchallenges,exacerbatingthechallengestheyfaceinadaptingtoachangingclimate.Table2.5providesexamplesofhowtherelativeseverityofacity’surbanchallengescaneitherbluntorexacerbatetheimpactsofclimateshocksandstresses.Source:WorldBankanalysisbasedondatafromMoody’sESGSolutions,Sub-SovereignPhysicalClimateRiskScores,October2021(https://esg.moodys.io/climate-solutions).Note:Thevaluesreportedinthefigurearethemeanclimateexposurescoresforcitiesthatbelongtoagiventype.Themeanscoresareestimatedbyregressingacity’sclimateexposurescoreonaseriesofdummyvariablesforthedifferenttypesofcities.Citiesaredefinedasurbancentersfollowingthedegreeofurbanizationmethodology(seebox1.1,chapter1).Typesofcitiesaredefinedusingthemethodologydescribedinbox2.1andannex2A.Figure2.8Averageweightedoverallclimatechange–relatedhazardexposure,bycitysizeandcountryincomegroup40455055606570SmallMediumLargeSmallMediumLargeSmallMediumLargeLow-incomeMiddle-incomeHigh-incomeClimateexposurescore148THRIVINGFigure2.9Averageclimateexposurescoresforsixclimatechange–relatedhazards,bycitysizeandcountryincomegroup2535455565SmallMediumLargeSmallMediumLargeSmallMediumLargeLow-incomeMiddle-incomeHigh-incomeLow-incomeMiddle-incomeHigh-incomeExposurescore3040506070SmallMediumLargeSmallMediumLargeSmallMediumLargeExposurescoreLow-incomeMiddle-incomeHigh-incomeLow-incomeMiddle-incomeHigh-income5060708090SmallMediumLargeSmallMediumLargeSmallMediumLargeExposurescore3040506070SmallMediumLargeSmallMediumLargeSmallMediumLargeExposurescorea.Floodsb.Waterstressc.Tropicalcyclonesd.Sea-levelrise(Continued)AGlobalTypologyofCities149Table2.5IntersectionbetweenurbanchallengesandclimateshocksandstressesUrbanchallengesIntersectionwithclimateshocksandstressesGreennessCarbonHighemissionsinglobalcitieswillincreasetheseverityofcertainclimatechange–relatedhazardsfeltatthelocalscale(suchassea-levelrise),whereaslow-carboninvestmentsofferopportunitiesforadaptationco-benefits.Forexample,treecovercanreduceheatstresswhilealsoloweringtheenergydemandforcooling(IPCC2022a;Seddonetal.2020).PollutionHighpollutionlevelslowerthehealthofpopulations,makingthemmorevulnerabletohazardssuchasheatstress(O’Lenicketal.2019).Highpollutionlevelsalsoundermineworkerproductivity(Deuskar2022),therebydamagingtheresilienceofcitiestoclimatechange–relatedshocksandstresses.VegetationVegetationandnature-basedsolutionscanblunttheimpactofstormsurgesfromtropicalcyclones,provideretentionforfloods,andreduceambientheat(Debeleetal.2019;Deuskar2022;WorldBank2021).ResilienceResilienceResilientcitiesexperiencelowereconomiclossesfromclimatechange–relatedshocksandstressesbecauseoffewerdisruptionsofbasicservicesandquickerrecovery(Hallegatte,Rentschler,andRozenberg2019;Qiang,Huang,andXu2020),andasshownbythisreport’sanalysisofnighttimelightsdatabasedontheworkofParkandRoberts(2023).(Continued)Source:WorldBankanalysisbasedondatafromMoody’sESGSolutions,Sub-SovereignPhysicalClimateRiskScores,October2021(https://esg.moodys.io/climate-solutions).Note:Ineachfigure,thereportedvalues(verticalaxis)arethemeanclimateexposurescoresforcitiesthatbelongtoagiventype.Themeansareestimatedbyregressingacity’sexposurescoreassociatedwitheachtypeofclimatehazardonaseriesofdummyvariablesforthedifferenttypesofcities.Citiesaredefinedasurbancentersfollowingthedegreeofurbanizationmethodology(seebox1.1,chapter1).Typesofcitiesaredefinedusingthemethodologydescribedinbox2.1andannex2A.Low-incomeMiddle-incomeHigh-incomeLow-incomeMiddle-incomeHigh-income3040506070SmallMediumLargeSmallMediumLargeSmallMediumLargeExposurescore4050607080SmallMediumLargeSmallMediumLargeSmallMediumLargeExposurescoree.Heatstressf.WildfiresFigure2.9continued150THRIVINGSmallandisolatedislandstateshavefeweroptionsforinternalclimatemigrationBuildingontheanalysisofcitytypologies,cityscoreswithineachcountrywereaveragedtobetterunderstandacountry’soverallurbanexposureandcapacitytoadapttolong-termclimateexposurethroughinternalmigration.TheresultsoftheanalysishighlighttheuniquestrugglesfacedbysmallerislandandcoastalnationssuchasthoseintheCaribbeanandthePacific(figure2.10).Inaddition,inseverallargerislandandcoastalcountries,suchasIndonesiaandVietnam,exposuretoclimatechange–relatedhazardsispervasiveacrossallcitiesbeinganalyzed.Bycontrast,intheUnitedStatesacitysuchasMiamihasahighscoreforoverallexposuretoclimatechange–relatedhazards,butothercitieswithlowerexposuresuchasMinneapolisorPittsburghcounterthatscore.Whenscoresareaveraged,lower-incomecountriesfacehigherlevelsofexposuretoclimatechange–relatedrisk(table2.6).SummaryandconclusionsBuildingontheinsightsinchapter1,thischapterclassifiestheglobalsampleofcitiesanalyzedinthisreportintoninetypesbasedonthepopulationsizeofthecities(small,medium,andlarge)andthelevelofdevelopmentofthecountriesinwhichtheyarelocated(low-,middle-,andhigh-income).Theresultingtypologyhighlightstheheterogeneityacrossdifferenttypesofcitiesinthemixandseverityoftheircurrenturbanchallenges,aswellasinthelevelofexposuretoclimatechange–relatedhazardsprojectedforwardto2030–40.Althoughcitiesinhigh-incomecountriesfacemoderatetoseverechallengesinreducingCO2emissionsandintracityinequality,thoseinlow-andmiddle-incomecountriesareconfrontedwithmoderatetoseverechallengesindealingwithairpollution,poverty,anduniversalaccesstobasicservices,aswellaseconomicresiliencetoclimateshocksandstresses.Consistentwiththatlackofaccessandloweconomicresilience,theanalysisofdatafromMoody’sESGSolutionspredictsthat,overthenexttwodecades,theloweracity’slevelofdevelopmentthehigheritsexposurewillbetotheoveralllevelofclimatechange–relatedhazards.Table2.5continuedUrbanchallengesIntersectionwithclimateshocksandstressesInclusivenessPovertyPoorpopulationsmayfaceanexposurebiastoclimatechange–relatedhazardssuchasfloodsby,forexample,livinginslumsinfloodplainsandtoheatstressbyhavinglessaccesstoair-conditioningandlessgreenspaceinpoorareas,amongotherthings(Hsuetal.2021;Winsemiusetal.2015—seealsothediscussioninchapter3).InequalityPoorpopulationsmayincurdisproportionatelyhigherlosseswhenaffectedbyshocksandstresses(Hallegatteetal.2017).ServicesLackofbasicservicesisacross-cuttingchallengethatcanexacerbatetheimpactsofshockssuchasfloods(lackofdrainage),aswellaswaterstress(lackofdrinkingwatersupply)andheatstress(lackofenergyforair-conditioning)—seeBaker2012.Source:WorldBank.AGlobalTypologyofCities151Figure2.10Urbanexposuretocombinedclimatechange–relatedhazards,selectedcountries,bycountryincomegroupSource:WorldBankanalysisbasedondatafromMoody’sESGSolutions,Sub-SovereignPhysicalClimateRiskScores,October2021(https://esg.moodys.io/climate-solutions).Note:Thefigureisorganizedindescendingorderofthemeanclimateexposurescoresineachincomegroup.Theupperandlowerbarsindicatethemaximumandtheminimumscores,respectively,foreachcountry.Thebottomofthebox,theborderoftwocolors,andthetopofthebox,respectively,depictthefirst,second(median),andthirdquartilesofthescoresineachcountry.ThehorizontalbarsindicatetheclimateexposurescoresforcountrieswithonlyonecitybeinganalyzedexceptforHonduras,whichhastwocitieswiththesamescore.Citiesaredefinedasurbancentersfollowingthedegreeofurbanizationmethodology(seebox1.1,chapter1).020406080100HondurasHaitiMadagascarVietnamIndonesiaBangladeshAlgeriaIndiaJamaicaDominicanRepublicMalaysiaThailandBrazilChinaArgentinaBarbadosTrinidadandTobagoBruneiDarussalamJapanUnitedStatesAustraliaFranceLow-incomeMiddle-incomeHigh-incomeClimateexposurescoreTable2.6Urbanexposuretocombinedclimate-relatedhazard,bycountryincomegroupIncomegroupAveragecountryscoreFivemosthighlyexposedcountriesLow-income64.0Honduras,Liberia,SriLanka,Haiti,andMadagascarMiddle-income53.7DominicanRepublic,Jamaica,Guyana,Cuba,andSurinameHigh-income48.4Bahamas,Barbados,TrinidadandTobago,Brunei,andTaiwan,ChinaSource:WorldBankanalysisbasedondatafromMoody’sESGSolutions,Sub-SovereignPhysicalClimateRiskScores,October2021(https://esg.moodys.io/climate-solutions).Note:Averagecountryscoreiscalculatedastheunweightedaveragescoreacrossallcitieswithineachincomegroup,withcitiesdefinedasurbancentersfollowingthedegreeofurbanizationmethodology(seebox1.1,chapter1).152THRIVINGMoreover,theoveralllevelofclimatechange–relatedhazards,alongsideCO2emissions,increasesascitiesbecomemorepopulated.Thus,largercitiesatthesamelevelofdevelopmentareexpectedtobemoreafflictedwithclimateshocksandstressesoverthecomingdecades.Thistrendisparticularlypronouncedwhenitcomestowaterstressandgeographicexposuretotropicalcyclones,ontopofwhichexposuretofloodsisprojectedtoplaceanadditionalburdenonlargercitiesinlow-incomecountries.Waterstressandthreatstolong-termhabitabilityaremajordriversofinternalclimatemigra-tion(Clementetal.2021).Unfortunately,thesmallislandstatesintheCaribbeanandthePacific,havingfewoptionsforinternalmigration,aretheworst-positionedtowithstandfutureclimatehazards.Inthedevelopingworld,suchlimitedadaptationoptions,compoundedbyinadequatefinancialandsocialcapacities,mayreducetheopportunitiesforhouseholdstomoveoutofpovertyandbuildresilientlivelihoods.Thechaptersthatfollowdelveintohowclimatechange–relatedshocksandstressesaffectcities(chapter3)andviceversa(chapter4).Chapter5thenelaboratesonthetailoredpolicyguidancefortheuniquechallengesfacingeachtypeofcity.Annex2A:MethodologyfordefiningaglobaltypologyofcitiesTheglobaltypologyofcitiesdescribedinthischapterisdefinedaccordingtotwodimen-sions:thesizeofacityandthelevelofdevelopmentrepresentedbytheincomegroupofthecountryinwhichthecityislocated.Thedimensionsareidentifiedinchapter1asimportantincharacterizingtheintensityofcurrenturbanchallengesrelatedtogreenness,resilience,andinclusiveness.Restrictingthedimensionsattheoutsetsimplifiedacomplexclusteringproblem.Basedonthetwodimensions,thestudyteamsplittheglobalsampleofcitiesintoninetypessothat(1)eachtypehasalowvarianceofseverity—thatis,citieswithinthesametypearerelativelyhomogeneous;and(2)eachtypehasadistinctivemeanlevelofseverity.Thederiva-tionofthetypologyandassessmentoftheseverityofcurrenturbanchallengesproceededinthefollowingsteps.Step1.Generateeighttypologiesaccordingtothecombinationofcitysizeandincomeclass.Theteamgeneratedeighttypologiesofcitiesbycombiningfouroptionsforthreecitysizes(small,medium,andlarge)andtwoforthethreeincomegroups(low,middle,andhigh).Table2A.1describesthecriteriaforthecategorization,aswellasthepopulationrangesandpercapitagrossnationalincomeforeachgroup.Step2.Identifythetypologythataccountsforthelargestvariationinurbanchallengeindicators.Amongtheeighttypologiesgeneratedinstep1,theteamchoseastheglobaltypologyofcitiestheonethatproducesthemosteffectivewayofdistinguishingcitiesintermsofseverityofurbanchallenge.Tomakeanobjectivejudgmentaboutthebestcombinationsofgrouping,theAGlobalTypologyofCities153teamemployedananalysisofcovariance.Thus,itfirstseparatelyregressedeachoftheurbanchallengeindicatorsoneachoftheeightcategoricalvariablesoftypologywhilecontrollingfortheeffectsofotherrelevantvariables.ItthenidentifiedthetypologythatproducesthehighestadjustedR2acrosstheestimatedmodels.Inderivingthetypology,theteamfocusedexclusivelyonsixindicatorsthatcharacterizegreennesschallenges:aggregatefossilCO2emissionsandpercapitafossilCO2emissionsforcarbonfootprint,aggregatePM2.5emissionsandtotalPM2.5concentrationforpollution,andaveragegreenness(index)andtheshareofhighgreenareaforvegetation.15Thisfocusstemmedprimarilyfromthefactthatthedataonthosegreennessindicatorscoverthefullsampleofcities,unlikethedataontheresilienceandinclusivenessindicators.Previousstudiessuggestthatacity’sclimateandgeographyarecorrelatedwiththegreennessindicators.Thus,intheregressionsofcarbonemissionsandpollution,theteamcontrolledforthepotentialeffectsoftemperatureandprecipitation(bothinannualmean),aswellasthemeanelevationandthemajortypeofbiomewithinthespatialdomainofeachcity.Forvegeta-tion,themeanelevation,longitude,latitude,andtheirsquareswerecontrolledalongsidethosecontrolledintheregressionsofcarbonfootprintandpollution.Table2A.1CriteriaforderivingaglobaltypologyofcitiesDimensionCriteriaGroupPopulationandincomerangesCitysize1TercilesofglobalpopulationdistributionThreegroupswithequalnumberofcitiesSmallMediumLarge50,007–76,56076,562–142,264142,276–36,312,5402TercilesofnationalpopulationdistributionThreegroupswithequalnumberofcitiesSmallMediumLarge50,007–164,79555,126–354,77878,040–36,312,5403BasedonOECDclassificationofurbanareasMetrosandlargemetroscombinedaslargecitiesSmallMediumLarge50,000–199,999200,000–499,999>=500,0004BasedonOECDclassificationofurbanareasMetrosandmediumcitiescombinedasmediumcitiesSmallMediumLarge50,000–199,999200,000–1,499,999>=1,500,000Incomegroup1WorldBankcountryclassificationforfiscalyear2021/22:upper-middle-andlower-middle-incomecombinedasmiddle-incomeLowMiddleHigh<1,0461,046–12,695>12,6952WorldBankcountryclassificationforfiscalyear2021/22:low-andlower-middle-incomecombinedaslow-incomeLowMiddleHigh<4,0964,096–12,695>12,695Source:PopulationdatafromtheEuropeanCommission’sGlobalHumanSettlement(GHS)UrbanCentreDatabaseR2019(https://ghsl.jrc.ec.europa.eu/ghs_stat_ucdb2015mt_r2019a.php).Note:Citiesaredefinedasurbancentersfollowingthedegreeofurbanizationmethodology(seebox1.1,chapter1).Populationisasof2015.IncomerangesarebasedongrossnationalincomepercapitaincurrentUSdollarscalculatedusingtheWorldBankAtlasmethodforexchangeratesin2020.Forcountrieswithfewerthanthreecitiesintheglobalsample,thecategoriesbasedonthenationaldistributionarereplacedwiththeonesbasedontheglobaldistribution.TheOrganisationforEconomicCo-operationandDevelopment(OECD)classificationofurbanareasisavailableathttps://data.oecd.org/popregion/urban-population-by-city-size.htm.154THRIVINGFigure2A.1comparestheR2acrosstheestimatedmodels,whichledtothechoiceoftypologycombiningcitysizeoption4(groupingsmallurbanareasassmall,metrosandmediumcitiesasmedium,andlargemetrosaslarge,drawingontheOrganisationforEconomicCo-operationandDevelopment’sclassification)andincomeclassoption2(combininglow-andlower-middle-incomeaslow-income,whiledesignatingupper-middle-incomeasmiddle-incomeandhigh-incomeashigh-income).Step3.Assesseachtypeofcityintermsoftheseverityofcurrenturbanchallenges.Theassessmentofeachindicatorbroadlyproceededinthefollowingsteps:3.1.Obtainthemeansacrosstheninetypesofcities.3.2.Basedonpairwisecomparisons,groupthepairsofcitytypesthatarenotstatisticallydifferentatthesignificancelevelof0.1.Source:WorldBankanalysisbasedondatadrawnfromtheEuropeanCommission’sGlobalHumanSettlement(GHS)UrbanCentreDatabaseR2019(https://ghsl.jrc.ec.europa.eu/ghs_stat_ucdb2015mt_r2019a.php).Note:Citiesaredefinedasurbancentersfollowingthedegreeofurbanizationmethodology(seebox1.1,chapter1).Income1andIncome2refertothegroupingsofincomeclassbasedontheWorldBankclassification.InIncome1,upper-middle-andlower-middle-incomeclassesarecombinedasmiddle-income.Inincome2,lower-middle-andlow-incomeclassesarecombinedaslow-income.GlobalandNationalrefertothegroupingsofcitysizebasedonglobalandnationalterciles,respectively.OECD1andOECD2refertothegroupingsbasedontheOrganisationforEconomicCo-operationandDevelopment’sclassificationofurbanareas.InOECD1,largemetrosandmetroareasarecombinedaslargecities.InOECD2,largemetrosbythemselvesconstitutelargecities,whereasmetroareasandmediumurbanareasarecombinedasmediumcities.CO2=carbondioxide;PM2.5=particulatematterof2.5micronsorlessindiameter.Figure2A.1ComparisonsofadjustedR2acrosscombinationsofcitysizeandcountryincomegroup0.00.10.20.30.40.5TotalfossilCO2emissionsPercapitafossilCO2emissionsTotalPM2.5emissionsTotalPM2.5concentrationAveragegreennessindexShareofhighgreenareaOverallAdjustedR2Income1×GlobalIncome2×GlobalIncome1×NationalIncome2×NationalIncome1×OECD1Income2×OECD1Income1×OECD2Income2×OECD2AGlobalTypologyofCities1553.3.Assignoneofthreelevelsofseverity—high(3),medium(2),andlow(1)—consideringtherankingofthemeansobtainedinstep3.1andthegroupingderivedinstep3.2.Thus,thesamelevelofseverityisassignedtothepairswhosemeansarestatisticallyindistinguishable.3.4.Calculatetheaggregatelevelofseverityforeachtypeofchallengebytakingtheunweightedmeanlevelofseverityacrossmultipleindicators—forexample,thecarbonfootprintbytotalandpercapitafossilCO2emissions.Table2.3inthemaintextpresentstheoutcomeofstep3.Essentially,thesamesetofstepswasappliedinthecontextofanalysisofcovariancetoallbuttheresilienceindicatorsforwhichthedataareinthepanelstructure.16Althoughtheassessmentofgreennessindicatorscoversthefullsampleofcities,thesameglobalcoveragewasnotfeasibleforotherdimensions.Dataonresilienceindicators,forexample,addressabout2,700mediumandlargecitiesforwhichtheproxymeasureofacity’seconomicactivity—nighttimelightsintensity—isavailable.Similarly,theseverityoflackofaccesstobasicserviceswasdrawnfromtheanalysisof1,970citiesin41low-andmiddle-incomecountries.17Forinequality,limiteddataonincomeorconsump-tionatthecityleveldidnotallowconstructionofameasure,especiallyforcitiesoutsidehigh-incomecountries.Thus,datausedforpreviousstudieswereconsolidatedtocover4,574citiesinIndonesia,theUnitedStates,and16countriesinLatinAmericaandtheCaribbean.18Acaveathereisthatcitiesaredelineateddifferentlyacrosstheregions,allofwhichdifferfromurbancentersidentifiedbythedegreeofurbanizationmethodology.19Finally,theremaininggapsinempiricalevidence,includingthemissingresultsforpoverty,areaddressedbyconsul-tationswithWorldBankexperts,aswellasbyreviewsoftheexistingliterature,includingtheWorldBank’sseriesofUrbanizationReviews.156THRIVINGTHRIVING156Notes1.SuchinitiativesincludetheC40CitiesClimateLeadershipGroup,theWorldMayorsCouncilonClimateChange,andtheUrbanClimateChangeResearchNetwork.2.Formoreontherelationshipsbetweenvariousgreennessindicatorsandacity’ssize,seechapter1,aswellasCorbaneetal.(2020);Huangetal.(2014);Kennedyetal.(2009);andMarcotullioetal.(2013).3.TheanalysisthatdefinestheglobaltypologyexcludesGuangzhouandShanghai,thetwolargesttype6cities,becausetheyareextremeoutliersintermsofbothCO2andPM2.5emissions.Assuch,theirinclusionwouldconsiderablyskewtheaveragepicture.4.Themeasureofresilienceusedinthischapterisrathernarrowinthatitdoesnotexplicitlymeasureacity’scapacitytoabsorb,recover,andprepareforfutureshocks.Nevertheless,thischapterdoesnotdistinguishbetweenthetermsvulnerabilityandresiliencebecausethebaselineanalysis(ParkandRoberts2023)implicitlycapturesacity’sabilitytoabsorban(extreme)weathershockinthemonththatitoccurs,which,inturn,isalsocloselyrelatedtohowpreparedthecityisforsuchaclimatechange–relatedstress.5.Themeasureofinequality,theGinicoefficient,iscalculatedonthebasisofpercapitaexpenditureforIndonesiaandworkerincomefortheUnitedStatesand16countriesinLatinAmericaandtheCaribbean.6.Panelboffigure2.5presentstheaverageelasticitycalculatedacrossdifferenttypesofextremeweathershocks(hot,wet,anddry),aswellastropicalcyclones.Thus,thefiguredoesnotexplicityshowtheestimatedelasticityreferredtohere.Seechapter1foracompletesetofindividualresultsforfivetypesofextremeweathershocks,includingtropicalcyclones.7.Throughoutthissection,thediscussionsaroundresiliencechallengesarebasedontheresultsforsome2,700mediumandlargecities.Empiricalevidenceforsmallcitiesislackingbecausethedataonnighttimelightsintensity—theproxymeasureforacity’seconomicactivity—areavailableonlyforcitieswithpopulationsofover200,000.8.TheWHOguidelinelevelforsafeairisanannualmeanPM2.5concentrationof5μg/m3ora24-hourmeanof15μg/m3.9.The16countriesincludedintheanalysisareArgentina,Bolivia,Brazil,Chile,Colombia,CostaRica,DominicanRepublic,Ecuador,ElSalvador,Guatemala,Honduras,Mexico,Nicaragua,Peru,Paraguay,andUruguay.10.Specifically,countriesinLatinAmericaandtheCaribbeanclassifiedaslow-incomeinthesampleareBolivia,ElSalvador,Honduras,andNicaragua.SmallcitiesinallbutElSalvadorexhibitaverageGinicoefficientsofincomehigherthan0.4.11.ForEurope,see,forexample,therelevantdiscussionsinHäußlerandHaupt(2021);Ottoetal.(2021);andReckienetal.(2015).FortheUnitedStates,seeClimateResilienceConsulting(2018).12.Thesestudiesanalyzeincomedatafrom18countries,ofwhichMexicoistheonlynon-high-incomecountry.13.Moody’sESGSolutions,https://esg.moodys.io/climate.AGlobalTypologyofCities157157AGlobalTypologyofCities14.Seebox4.5inchapter4foradescriptionthefivedifferentSSPs.15.Thehighgreenareacorrespondstoadenselyvegetatedarea,suchasaforestorgarden(Florczyketal.2019).16.Fordetailsoftheestimationprocess,seechapter1andthebackgroundpaperbyParkandRoberts(2023).17.ThedatacomefromHendersonandTurner(2020),https://doi.org/10.7910/DVN/YZ46FJ,andcovertheEastAsiaandPacific,LatinAmericaandtheCaribbean,SouthAsia,andSub-SaharanAfricaregions.18.DataforIndonesiacomefromRoberts,GilSander,andTiwari(2019);fortheUnitedStatesfromBehrensandRobert-Nicoud(2015);andfor16countriesinLatinAmericaandtheCaribbeanfromFerreyraandRoberts(2018).19.Specifically,level-2administrativeunitsareusedforIndonesia,metropolitanstatisticalareasfortheUnitedStates,andurbanclustersidentifiedbytheclusteringalgorithm(DijkstraandPoelman2014)forcountriesinLatinAmericaandtheCaribbean.ReferencesBaker,J.L.2012.ClimateChange,DisasterRisk,andtheUrbanPoor:CitiesBuildingResilienceforaChangingWorld.Washington,DC:WorldBank.Behrens,K.,andF.Robert-Nicoud.2015.“AgglomerationTheorywithHeterogeneousAgents.”InHandbookofRegionalandUrbanEconomics,Volume5:CitiesandGeography,editedbyG.Duranton,J.V.Henderson,andW.Strange,171–245.Amsterdam:Elsevier.Boulant,J.,M.Brezzi,andP.Veneri.2016.“IncomeLevelsandInequalityinMetropolitanAreas:AComparativeApproachinOECDCountries.”RegionalDevelopmentWorkingPaper2016/06,OrganisationforEconomicCo-operationandDevelopment,Paris.Busso,M.,andJ.Messina,eds.2020.TheInequalityCrisis:LatinAmericaandtheCaribbeanattheCrossroads.Washington,DC:Inter-AmericanDevelopmentBank.Castells-Quintana,D.,V.Royuela,andP.Veneri.2020.“InequalityandCitySize:AnAnalysisforOECDFunctionalUrbanAreas.”PapersinRegionalScience99:1045–64.Chen,B.,D.Liu,andM.Lu.2017.“CitySize,Migration,andUrbanInequalityinthePeople’sRepublicofChina.”WorkingPaperSeriesNo.723,AsiaDevelopmentBankInstitute,Mandaluyong,Philippines.Clement,V.,K.K.Rigaud,A.deSherbinin,B.Jones,S.Adamo,J.Schewe,N.Sadiq,etal.2021.GroundswellPart2:ActingonInternalClimateMigration.Washington,DC:WorldBank.ClimateResilienceConsulting.2018.“DoesClimateResilienceMatterforSmall-TownAmerica?”ClimateResilienceConsultingBlog,October1,2018.https://www.climateresilienceconsulting.com/blog/does-climate-resilience-matter-for-small-town-america.Corbane,C.,P.Martino,P.Panagiotis,F.Aneta,M.Michele,F.Sergio,S.Marcello,etal.2020.“TheGrey-GreenDivide:Multi-TemporalAnalysisofGreennessacross10,000UrbanCentres158THRIVINGTHRIVING158DerivedfromtheGlobalHumanSettlementLayer(GHSL).”InternationalJournalofDigitalEarth13(1):101–18.Debele,S.E.,P.Kumar,J.Sahani,B.Marti-Cardona,S.B.Mickovski,L.S.Leo,F.Porcu,etal.2019.“Nature-BasedSolutionsforHydrometeorologicalHazards:RevisedConcepts,ClassificationSchemesandDatabases.”EnvironmentalResearch179(B):108799.Deuskar,C.2022.“BeatingtheHeat:MeasuringandMitigatingExtremeHeatinEastAsianCities.”TechnicalWorkingPaper1:LiteratureReview,WorldBank,Washington,DC.Dijkstra,L.,andH.Poelman.2014.“AHarmonisedDefinitionofCitiesandRuralAreas:TheNewDegreeofUrbanization.”RegionalWorkingPaper,Directorate-GeneralforRegionalandUrbanPolicy,EuropeanCommission,Brussels.Fay,M.2005.TheUrbanPoorinLatinAmerica.DirectionsinDevelopmentSeries.Washington,DC:WorldBank.Ferreira,F.,andM.Schoch.2020.“InequalityandSocialUnrestinLatinAmerica:TheTocquevilleParadoxRevisited.”WorldBankBlogs,February24,2020.https://blogs.worldbank.org/developmenttalk/inequality-and-social-unrest-latin-america-tocqueville-paradox-revisited.Ferreyra,M.M.,andM.Roberts,eds.2018.RaisingtheBarforProductiveCitiesinLatinAmericaandtheCaribbean.Washington,DC:WorldBank.Florczyk,A.J.,M.Melchiorri,C.Corbane,M.Schiavina,M.Maffenini,M.Pesaresi,P.Politis,etal.2019.“DescriptionoftheGHSUrbanCentreDatabase2015,PublicRelease2019,Version1.0.”PublicationsOfficeoftheEuropeanUnion,Luxembourg.Glaeser,E.L.,M.Resseger,andK.Tobio.2009.“InequalityinCities.”JournalofRegionalScience49(4):617–46.Hallegatte,S.,J.Rentschler,andJ.Rozenberg.2019.Lifelines:TheResilientInfrastructureOpportunity.Washington,DC:WorldBank.Hallegatte,S.,A.Vogt-Schilb,M.Bangalore,andJ.Rozenberg.2017.Unbreakable:BuildingtheResilienceofthePoorintheFaceofNaturalDisasters.Washington,DC:WorldBank.Häußler,S.,andW.Haupt.2021.“ClimateChangeAdaptationNetworksforSmallandMedium-SizedCities.”SNSocialSciences1:262.Henderson,J.V.,andM.Turner.2020.“UrbanizationintheDevelopingWorld:TooEarlyorTooSlow?”JournalofEconomicPerspectives34(3):150–73.Hsu,A.,G.Sheriff,T.Chakraborty,andD.Manya.2021.“DisproportionateExposuretoUrbanHeatIslandIntensityacrossMajorUSCities.”NatureCommunications12:2721.Huang,Y.,H.Shen,H.Chen,R.Wang,Y.Zhang,S.Su,Y.Chen,etal.2014.“QuantificationofGlobalPrimaryEmissionsofPM2.5,PM10,andTSPfromCombustionandIndustrialProcessSources.”EnvironmentalScienceandTechnology48(23):13834–43.IPCC(IntergovernmentalPanelonClimateChange).2022a.“Chapter6:Cities,SettlementsandKeyInfrastructure.”InClimateChange2022:Impacts,Adaptation,andVulnerability.ContributionofWorkingGroupIItotheSixthAssessmentReportoftheIntergovernmentalPanelonClimateChange,editedbyH.-O.Pörtner,D.C.Roberts,M.Tignor,E.S.Poloczanska,K.Mintenbeck,A.Alegría,M.Craig,S.Langsdorf,etal.Cambridge,UK:CambridgeUniversity.AGlobalTypologyofCities159159AGlobalTypologyofCitiesIPCC(IntergovernmentalPanelonClimateChange).2022b.“SummaryforPolicymakers.”InClimateChange2022:Impacts,Adaptation,andVulnerability.ContributionofWorkingGroupIItotheSixthAssessmentReportoftheIntergovernmentalPanelonClimateChange,editedbyH.-O.Pörtner,D.C.Roberts,M.Tignor,E.S.Poloczanska,K.Mintenbeck,A.Alegría,M.Craig,etal.Cambridge,UK:CambridgeUniversityPress.Kennedy,C.,J.Steinberger,B.Gason,Y.Hansen,T.Hillman,M.Havranck,D.Pataki,etal.2009.“GreenhouseGasEmissionsfromGlobalCities.”EnvironmentalScienceandTechnology43(19):7297–302.Li,D.,S.Wu,L.Liu,Y.Zhang,andS.Li.2018.“VulnerabilityoftheGlobalTerrestrialEcosystemstoClimateChange.”GlobalChangeBiology24(9):4095–106.Marcotullio,P.J.,A.Sarzynski,J.Albrecht,N.Schulz,andJ.Garcia.2013.“TheGeographyofGlobalUrbanGreenhouseGasEmissions:AnExploratoryAnalysis.”ClimateChange121:621–34.O’Lenick,C.R.,O.V.Wilhelmi,R.Michael,M.H.Hayden,A.Baniassadi,C.Wiedinmyer,A.J.Monagham,etal.2019.“UrbanHeatandAirPollution:AFrameworkforIntegratingPopulationVulnerabilityandIndoorExposureinHealthRiskAnalyses.”ScienceoftheTotalEnvironment660:715–23.Otto,A.,K.Kern,W.Haupt,P.Eckersley,andA.H.Thieken.2021.“RankingLocalClimatePolicy:AssessingtheMitigationandAdaptationActivitiesof104GermanCities.”ClimateChange167:5.Park,J.,andM.Roberts.2022.“UrbanResilienceandWeatherShocks:AGlobalNighttimeLightsAnalysis.”Backgroundpaperpreparedforthisreport,WorldBank,Washington,DC.Qiang,Y.,Q.Huang,andJ.Xu.2020.“ObservingCommunityResiliencefromSpace:UsingNighttimeLightstoModelEconomicDisturbanceandRecoveryPatterninNaturalDisasters.”SustainableCitiesandSociety57:102115.Reckien,D.,J.Flacke,M.Olazabal,andO.Heidrich.2015.“TheInfluenceofDriversandBarriersonUrbanAdaptationandMitigationPlans—AnEmpiricalAnalysisofEuropeanCities.”PLoSOne10(8):e0135597.Roberts,M.,F.GilSander,andS.Tiwari,eds.2019.TimetoACT:RealizingIndonesia’sUrbanPotential.Washington,DC:WorldBank.Seddon,N.,A.Chausson,P.Berry,C.A.J.Girardin,A.Smith,andB.Turner.2020.“UnderstandingtheValueandLimitsofNature-BasedSolutionstoClimateChangeandOtherGlobalChallenges.”PhilosophicalTransactionsoftheRoyalSocietyB375(1794):20190120.Simpson,N.P.,K.J.Mach,A.Constable,J.Hess,R.Hogarth,M.Howden,J.Lawrence,etal.2021.“AFrameworkforComplexClimateChangeRiskAssessment.”OneEarth4(4):489–501.Slay,B.2009.Poverty,Inequality,andSocialPolicyReformintheFormerSovietUnion.NewYork:UnitedNationsDevelopmentProgramme.Sterzel,T.,M.K.B.Ludeke,C.Walther,M.T.Kok,D.Sietz,andP.L.Lucas.2020.“TypologyofCoastalUrbanVulnerabilityunderRapidUrbanization.”PLoSOne15(1):e0220936.Winsemius,H.,B.Jongman,T.Veldkamp,S.Hallegatte,M.Bangalore,andP.Ward.2015.“DisasterRisk,ClimateChange,andPoverty:AssessingtheGlobalExposureofPoorPeopletoFloodsandDroughts.”PolicyResearchWorkingPaper7480,WorldBank,Washington,DC.160THRIVINGTHRIVING160WorldBank.2021.ACatalogueofNature-BasedSolutionsforUrbanResilience.Washington,DC:WorldBank.Xu,X.,C.Wang,S.Ma,andW.Zhang.2021.“China’sSpecialPoorAreasandTheirGeographicalConditions.”Sustainability13:8636.Zhang,Z.Y.2019.“ChinaIsRelaxingHukouRestrictionsinSmallandMedium-SizedCities.”China-Briefing.com,April17,2019.https://www.china-briefing.com/news/china-relaxing-hukou-restrictions-small-medium-sized-cities.Zscheischler,J.,S.Westra,B.vandenHurk,S.Seneviratne,P.Ward,A.Pitman,A.Aghakouchak,etal.2018.“FutureClimateRiskfromCompoundEvents.”NatureClimateChange8:469–77.161161SPOTLIGHTMultidimensionalExclusionandExposuretoAirPollutioninPeruvianCitiesIntroductionAsustainablecityisbuiltoninteractingeconomic,environmental,andsocialpillars.1Ifwellbalanced,thesepillarscontributetowell-functioningcitiesinwhichtheintertwinedstressesassociatedwithurbangrowthandclimatechangearewellmanagedandpeopleleadfulfillinglives.Bycontrast,imbalancesamongthepillarscantranslateintoexclusion,therebyunder-miningthequalityoflifethatacityoffers.Perupresentsagoodcasestudyofthispossibleimbalance.Inrecentdecades,theoutdoorairqualityinPeru’scities,especiallyinLima-Callao,seemstohaveimproved(Sanchez-Triana2017).By2021,however,Perustillfacedahigheconomiccost,equivalenttosomeUS$1.5billionayear(IEc2021),fromoutdoorandindoorairpollu-tion.Theimpactofpoor-qualityaironthehealthofPeru’surbanitesiscompoundedbytheirexclusionfromlaborandfinancialmarkets,basicservices,andopportunitiestoaccumu-latehumancapital,aswellasfromcivilparticipation,equalityoflaw,andfreedomofspeech.OthermanifestationsofexclusionarelackofgovernmentaccountabilityandthedecliningresilienceofPeruviancitizenstohumanandenvironmentalshocks.ThisspotlightdescribesmultidimensionalexclusioninPeru’scitiesanditsoverlapwithexposuretoairpollution.Multidimensionalexclusionismeasuredalongthreedimensions:(1)economicinclusion,(2)resilienceandsocialcohesion,and(3)empowerment,voice,andaccountability.2Airpollutionismeasuredusingdataonground-levelfineparticulatematterof2.5micronsorlessindiameter(PM2.5).3MethodologyThecountingapproachusedtomeasuremultidimensionalpoverty(AlkireandFoster2011;Alkireetal.2015)isadaptedheretoanalyzethejointdistributionofexclusionin2019acrossthethreedimensionsjustnoted.4Indoingso,theanalysisheremeasuresmultidimensionalexclusionusingthenumberofindicatorsfromwhichapersonissimultaneouslyexcluded.Thismeasurementofmultidimensionalexclusionisachievedbyapplyingacross-dimensionalthresholdthatdefinesthenumberof(weighted)indicatorsofsimultaneousexclusionapersonmustexperiencetobeclassifiedasmultidimensionallyexcluded.5Themethodologyfocusesontheadjustedheadcountratio(M0).Thisindexencompassesboththeincidence(H)andintensity(A)ofmultidimensionalexclusion.Incidencereferstotheshareofthepopulationthatfallsabovethecross-dimensionalthreshold,whereasintensityreferstothedegreetowhichpeopleareexcluded.6M0isadditivelydecomposablebysubgroup,facilitat-inganalysisofthemultidimensionalexclusionprofilesbylocation.THRIVING162MeasurementAppraisalofeconomicinclusionconsidersfiveindicatorsthatcaptureaccesstolabormarkets,financialservices,basicservices,andhumancapital(educationandhealth)services.Resilienceandsocialcohesionaremeasuredbyfourindicatorscapturingaperson’snumberofincomesources,confidenceingovernmentinstitutions,experiencewithdiscrimination,andexperi-enceasavictimofcrime.Finally,empowerment,voice,andaccountabilityarecapturedbyfourindicatorsthatassesscivilparticipation,freedomofspeech,governmenteffectiveness,andequalityofthelaw.Allindicatorsareequallyweightedwithinadimension.7ProfilesofmultidimensionalexclusionProfilingofmultidimensionalexclusionisbasedonacross-dimensionalthresholdof33percent8—thatis,apersonismultidimensionallyexcludedifthatpersonexperiencesexclusiononathirdofthe13indicators.Accordingtothisthreshold,84percentofPeruviansaremultidimensionallyexcluded(H=84percent)and,onaverage,theyexperienceexclusionon7ofthe13indicators(intensityrate,A)—seefigureS1.1.Theincidenceofmultidimensionalexclusionishigherinruralareas(91percent)thaninurbanareas(82percent).Anditismoreprevalentandintenseincoastalareasthaninrainforestareasandthehighlands.Theprofilingofexperiencesofexclusionisalsoquiteheterogenousacrosscitieswithinagiventypeofnaturalregion.Thus,experienceswithexclusionvarymoreamongpeoplelivingincoastalcities,withincidenceratesrangingfrom68to88percent.Bycontrast,incidenceSource:WorldBankcalculationsbasedondatafromPeru’sNationalHouseholdSurvey.FigureS1.1Adjustedheadcountratio(M0),incidence(H),andintensity(A)ofmultidimensionalexclusion,bytypeofareaofPeru,20190.20.40.60.81.020406080100NationalRuralUrbanCoastalHighlandsRainforestAdjustedheadcountratio(M0)Incidence,H,intensity,A(%)Incidence(H)Intensity(A)Adjustedheadcountratio(M0)THRIVING163163ratesacrosscitiesinrainforestareasvarymuchless—between72and78percent.Citiesinthehighlandsexhibitthelowestdegreeofmultidimensionalexclusion.WhichcitiesinPeruarethemostexcluding?Lima-CallaoandTacnaincoastalareas,PuertoMaldonadoinarainforestarea,andArequipainthehighlandshavethehighestlevelsofmultidimensionalexclusion(figureS1.2).Lima-Callao,Peru’scapitalandmostpopulatedcity,hasthecountry’shighestratesofbothincidenceandintensity.PuertoMaldonado,amuchsmallercity,hasanintensityrateabovethenationalaveragebutanincidencesimilartothenationalaverage.Clearly,theextenttowhichpeopleexperiencemultidimensionalexclusionisverydiverseacrossPeru’scities.ExperiencewithmultidimensionalexclusionandexposuretooutdoorairpollutionHowdoesexperiencewithmultidimensionalexclusionoverlapwithexposuretooutdoorairpollutionacrossPeruviancities?ThisspotlightfocusesonairpollutionbecauseofthestrongcorrelationbetweenPM2.5emissionsandcarbondioxide(CO2)emissions(highlightedinchapter1).FigureS1.3presentsacategorizationbylevelsofmultidimensionalexclusion(MDE)andoutdoorairpollution(OAP).Low(orhigh)levelsaredefinedasratesofexclusionFigureS1.2Incidence(H)versusintensity(A)ofmultidimensionalexclusionacrossPeru’scitiesSource:WorldBankcalculationsbasedondatafromPeru’sNationalHouseholdSurvey.Note:Bubblesizesareproportionaltocitypopulations.Nationalratesofintensityandincidenceareindicatedbytheintersectionoftheverticalandhorizontaldashedlinesat84percentand53percent,respectively.TacnaLima-CallaoArequipaPuertoMaldonado4045505560556065707580859095Intensity(%)Incidence(%)CoastalHighlandsRainforestSpotlight:MultidimensionalExclusionandExposuretoAirPollutioninPeruvianCitiesTHRIVING164andpollutionthatarebelow(orabove)themedianoftherespectivedistributions.Thisdefi-nitionresultsinaclassificationoffourgroups:highMDE–highOAP,lowMDE–lowOAP,andmismatches.ThehighMDE–highOAPgroupconsistsofArequipa,Lima-Callao,Moyobamba,PuertoMaldonado,Pucallpa,Tacna,andTrujillo,drawnfromallthreeofPeru’snaturalregions.Bycontrast,thelowMDE–lowOAPgroupincludescitiesfromthecoastalandhighlandareas,butnotfromtherainforestareas.CitiesshowinglowlevelsofexclusionbuthighpollutionareChiclayoandIcaincoastalareas,ChachapoyasandIquitosinrainforestareas,andAbancayinthehighlands.CitieswithlowpollutionbuthighexclusionareTumbesandMoqueguaincoastalareasandCerrodePasco,Huánuco,andPunoinhighlandsareas.Policiesaimedataddressingexclusionshouldaccountforthelinkswithpollutionandclimatechange.Notes1.ThisspotlightwaspreparedbyPaolaBallonandJoseCuesta.2.ThesethreedimensionsareanchoredintheWorldBank’sSocialSustainabilityStrategy2021(WorldBank2021).3.ThesedatacomefromHammeretal.(2022)Source:WorldBankcalculationsbasedondatafromPeru’sNationalHouseholdSurveyandNationalAeronauticsandSpaceAdministrationSocioeconomicDataandApplicationsCenter.Note:“Lowlevel”referstocitieswithratesbelowthemedian,and“highlevel”referstocitieswithratesabovethemedian.FigureS1.3Multidimensionalexclusionandoutdoorairpollution:Atypologybycity,Peru,2019HighlevelofairpollutionLowincidenceofmultimensionalexclusionLowlevelofairpollutionHighincidenceofmultidimensionalexclusionChiclayaIcaPiuraTumbesMoqueguaHuánucaCerrodePascoPunoLima-CallaoTacnaTrujilloMoyobambaPucallpaPuertoMaldonadoArequipaChachapoyasIquitosAyacuchoCajamarcaCuscoHuancavelicaHuancayoHuarazAbancayCoastalRainforestHighlandsCoastalRainforestHighlandsCoastalRainforestHighlandsCoastalRainforestHighlandsTHRIVING1651654.TheassessmentiscarriedoutusingdatafromPeru’sNationalHouseholdSurvey.5.Twotypesofthresholdareusedtoidentifythemultidimensionallyexcluded:asetofindicator-specificthresholdsandathresholdacrossdimensions.Theindicator-specificthresholdsidentifyapersonasexcludedonaparticularindicatorifthatpersonfallsbelowtherelevantthreshold.Thecross-dimensionalthresholdspecifieshowwidelyexcludedapersonmustbetobeidentifiedasmultidimensionallyexcluded.6.Thisincidencerateindicatesonhowmanyindicators,onaverage,thosewhoareabovethethreshold(themultidimensionallyexcluded)experienceexclusion.7.Amoredetaileddescriptionoftheindicatorsisavailableuponrequest.Theindicator-specificthresholdsusedtoidentifyexclusionfollowthedefinitionsusedintheUnitedNations’SustainableDevelopmentGoals.8.Resultsarerobusttothealternativeuseofthresholdsof50and75percent.ReferencesAlkire,S.,andJ.Foster.2011.“CountingandMultidimensionalPovertyMeasurement.”JournalofPublicEconomics95(7–8):476–87.Alkire,S.,J.M.Roche,P.Ballon,J.Foster,JM.E.Santos,andS.Seth.2015.MultidimensionalPovertyMeasurementandAnalysis.Oxford,UK:OxfordUniversityPress.IEc(IndustrialEconomicsIncorporated).2021.“AssessingtheMortalityBurdenofAirPollutioninLima-Callao.”FinalreportpreparedfortheUSEnvironmentalProtectionAgency.IEc,Cambridge,MA.Hammer,M.S.,A.vanDonkelaar,C.Li,A.Lyapustin,A.M.Sayer,N.C.Hsu,R.C.Levy,M.J.Garay,etal.2022.“GlobalAnnualPM2.5GridsfromMODIS,MISRandSeaWiFSAerosolOpticalDepth(AOD),1998-2019,V4.GL.03.”NationalAeronauticsandSpaceAdministrationSocioeconomicDataandApplicationsCenter,Palisades,NY.https://doi.org/10.7927/fx80-4n39.Sanchez-Triana,E.2017.“AccesstoQualityInformationIsCrucialtoTacklePeru’sEnvironmentalProblems.”WorldBankBlogs,January31,2017.WorldBank,Washington,DC.WorldBank.2021.TimetoChampionSocialSustainabilityandInclusion.SocialSustainabilityandInclusionGlobalPractice,Washington,DC:WorldBank.Spotlight:MultidimensionalExclusionandExposuretoAirPollutioninPeruvianCitiesWhatDoWeKnow?PART2169IntroductionThetrendsareclear:Allcountrieswillurbanize;theymaydoitintheirownwaysandattheirownpace,buturbanizetheywill.Andclimatewillalsochange,bringingwithitvariouscomplex,compounding,andcascadingimpactsonplacesandpeople.ThelatestreportbytheIntergovernmentalPanelonClimateChangepredictsthattheworseningimpactsofclimatechange,especiallyheatstress,floods,andhighersealevels,willbefeltacutelyincities(IPCC2022).Thischapterreviewsthesecondaryliteratureonhoweachoftheclimatechange–relatedstressesdescribedandanalyzedinchapters1and2hasaffectedurbandevelopment.AnalysisundertakeninpreparationforthisreportcomplementsthatreviewtofillinsomeoftheTheImpactsofClimateandEnvironmentalChangeonCitiesCHAPTER3Thefutureisalwaysuncertain,yetitseemsclearthatcitieswillgrow,andthatclimatewillchange,althoughdisparately.NaturalHazards,UnNaturalDisasters:TheEconomicsofEffectivePrevention,WorldBank(2010)••Fromthegrowingexposuretoclimatehazards,toshortagesoffoodandnaturalresources,tofatalities,theeffectsofclimatechangearevastyetvariable.Citiesmust,then,planaccordingly.••Becausecitiesconcentratepeople,activities,andinfrastructuregeographically,theyareparticularlyvulnerabletothedirecteffectsofclimatechangeevents—bothrapid-onsetshocksandslower-onsetstressors.••Ruralandurbanareasarelinkedeconomically,socially,andenvironmentally.Moreover,theeffectsofclimateeventsoccurringoutsideofcitiescanspillovertothem.••Spillovercouldoccurbywayofmigrationofpeople,affectingthepaceandlevelofurbanization(andurbanform)andofeconomicsectorswithincities,andbywayofthemovementofgoodsandservicesviatrade.••Climateeventsaffectgroupsofpeopledifferently,andthosewhoarealreadymorevul-nerablehavelessabilitytocopewithormitigatetheeffects.MAINFINDINGS170THRIVINGexistinggaps.Specifically,thechapterlooksmorecloselyathowclimatechange–relatedstressesaffectcitiesdirectly,throughtheirimpactsonurbanareas,andindirectly,throughtheirimpactsoutsideurbanareas.Italsodescribestheheterogeneousimpactsondifferentpopulationsincities.Chapter4thenassesseshowcitiesaffecttheclimate(and,moregenerally,theenvironment)throughtheirurbanformandinvestmentsin,forexample,publictransportation,aswellastherepercussionsofurbanexpansionforagriculturalsystemsandcompetitionforwater.Althoughclimatechangeisexpectedtomakecitiesmorevulnerableovertime,importantuncertaintiesremainaboutthemagnitudeofitseffectsandtheirpersistenceovertime,theheterogeneityoftheeffectsacrosssocioeconomicgroups,theconditionsunderwhichtheycausemoreorlessharm,andtheeffectivenessofinterventionsindealingwiththem.Thischaptertacklestheeffectsandconditions,leavingadiscussionoftheinterventionstochapter5.Thischapterhasthefollowingobjectives:••Reviewtheevidenceonhowclimatechange–relatedstressesofvarioustypesalreadyaffectdifferenttypesofcitiesgloballyandhowtheseeffectsfurtheraffectthesustainabilityofexistingandnewurbanconstruction.••Investigatehowclimatechangestressesaffectcitiesthroughtheirimpactsonnonurbanlocations.••Examinehowdifferentgroupsofpopulationsincitiesareaffectedsimilarlyordifferentiallybytheseimpacts.••Setthestageforthenextchapter,whichwilltacklethesecondpartofthetwo-wayrelation-shipbetweencitiesandclimate—thatis,howurbandevelopmentaffectsclimatechange.Howdoesclimatechangeaffectcitiesdirectly?Inallcitiesandurbanareas,therisksfacedbypeopleandassetsfromthehazardsassociatedwithclimatechangehaveincreasedAscitiesgrow,theirurbanlandscapeswillincreasinglyfeeltheeffectsofacuteclimateshocksandchronicclimatestressors(box3.1).Seventypercentofcitiesgloballyarealreadydealingwithclimatechange,andnearlyallareatrisk.Thecultural,demographic,andeconomicchar-acteristicsofurbanresidents,citygovernments,builtenvironment,ecosystemservices,andhuman-inducedstressessuchasoverexploitationofresourcesandenvironmentaldegradationdefinethevulnerabilityofcitiestoclimate-relateddisasters.Meanwhile,climatechangewillcontinuetoamplifytheexistingsocialandenvironmentalrisksandcreatenovelrisksforcitiesattheintersectionofclimatehazardsandthevulnerability,exposure,andresilienceofurbansocial-ecologicalsystemsandpopulations(Genceretal.2018).Climateshocks—sudden-onseteventsExtremeheat.About250cities(hometo200millionpeople)currentlyexperienceextremeheatconditionsarisingfromthree-monthaveragemaximumtemperaturesofabout35°C.By2050,950cities(1.6billionpeople)willlikelybeexposedtoextremesummertemperatures(UCCRN2018).Urbancentersandcitiesareoftenseveraldegreeswarmerthansurroundingareasbecauseofthepresenceofheat-absorbingmaterials,reducedevaporativecoolingcausedTheImpactsofClimateandEnvironmentalChangeonCities171bylackofvegetation,andwasteheatproduction.Manycitiesinthesouthernhemispherearealreadyfamiliarwithextremelyhightemperaturesforsustainedperiods,butcitiesinhigherlatitudesarenot,andwillforthefirsttimefaceheatextremes.Inurbanareas,exposuretoheatislandeffectsisuneven,withsomepopulations—includinglow-incomecommunities,children,theelderly,thedisabled,andethnicminorities—moreatriskthanothers.Forexample,socio-economicallydisadvantagedpopulationsaremorelikelytoliveinthehotterpartsofcitiesassociatedwithhigher-densityresidentiallanduseandindwellingsbuiltofpoorerorolderconstructionmaterialssuchasinsulation(Inostroza,Palme,anddelaBarrera2016).Extremeheataffectsthebuiltenvironmentandcityinfrastructurethroughmaterialexpansionandcor-rosion,therebydamagingroads,buildings,andinfrastructure.Extremecold.Citiesalsofeeltheeffectsofextremecoldevents.Theevents,whoseimpactsmaybemoredelayedthanthoseofextremeheat,aremorelikelytoaffectcitieslocatedinwarmerclimates.SmithandSheridan(2019)findastrongcorrelationbetweenlatitudeandextremecoldeventmortalityforcitiesintheUnitedStates,aswellashighermortalityratesforeventsearlyinthecoldseason.Eventhoughheat-relatedmortalityisexpectedtoincreasewithclimatechange,extremecoldeventswillcontinuetoposeapublichealthproblem(Broadbent,Krayenhoff,andGeorgescu2020).Heavyprecipitationandfloods.Rainfallisspatiallyheterogeneous,muchmoresothantemperature.Thus,fullyunderstandingtheeffectsofprecipitationwillrequirespatiallydisag-gregatedanalyses.Between1985and2015,settlementsworldwidegrewby85percent,toover1.28millionsquarekilometers.Duringthesameperiod,settlementsexposedtothehighestfloodhazardlevelincreasedby122percent,and36,500squarekilometersofsettlementswerebuiltintheworld’shighest-riskzones,82percentoftheminlow-andmiddle-incomecountries(Rentschleretal.2022).Fullyunderstandingtheeffectsofprecipitation,however,willrequirespatiallydisaggregatedanalyses.Heavyprecipitationandtheconsequentfloodsdisrupturbansettlements,stagnatetradeandcommerce,interrupttransportationnetworks,andplaceimmensepressureonurbaninfrastructure.Howdoclimateshocksandstressorsdiffer?Aclimateshockisanunpredictableweathereventthatdamagesthesustainabilityofacommunity.Examplesareheatwaves,floods,cyclones,andwildfires.Climatestress-ors,orslow-onsetevents,unfoldgraduallyovertime.Examplesarehighersealevels,climbingtemperatures,oceanacidification,glacialretreatandrelatedimpacts,saliniza-tion,landandforestdegradation,biodiversityloss,anddesertification.Somemeaningfulrelationshipsexistbetweenrapid-onsetandslow-onsetevents.Drought,forexample,isanextremeweathereventbutisalsocloselylinkedtoslow-onset,incrementalclimaticchange.Interactionsamongrapid-onsetandslow-onseteventsmayresultinthecrossingofthresholds.Forexample,astudyinFloridafindsthat,oncetheseareachedacriticallevel,thetransitionfromalandscapecharacterizedbyuplandforestsandfreshwaterwetlandstoonedominatedbymangrovesoccurredsuddenlyfollowingasinglestormsurgeevent(Rossetal.2009).Changesinclimateparameters—forexample,temperature,precipitation,andtheirassociatedimpactssuchaswateravailabilityandcropproductiv-itychanges—occuroverlongperiods.Box3.1172THRIVINGCoastalflooding.Almost11percentoftheglobalpopulation—approximately896millionpeople—livesincoastalcitiesandsettlements,andthatnumberisexpectedtoexceed1billionby2050(IPCC2022).Coastalflooding(oftencoupledwithinundationanderosion)arisesfromseverestorm-inducedsurges,waveovertopping,andrainfallrunoff.Underahigh-emissionsscenario,by2050anestimated340millionpeoplewilllivebelowtheprojectedhightideline(KulpandStrauss2019).Coastalcitiesareoftenimportantnodesforglobaltradesupplychains,andtheyplayanessen-tialroleinnationaleconomies.Withchangingclimate,coastalcitiesandtheirassociatedassetsandinfrastructurewillcontinuetofaceescalatingriskslinkedtohighersealevelsandcoastalflooding.Infact,coastalfloodingnotonlystrainscriticalurbaninfrastructurebutalsohassignificanteffectsonperi-urbancoastalculturalheritagesandresources(Choyetal.2016).Moreover,populationandassetgrowth,climatechange,andsubsidencewillalllikelycontrib-utetoadrasticincreaseinglobalaveragefloodlosses,fromUS$6billionperyearin2005tooverUS$60billionperyearin2050,assumingallcitiesundertakeproactiveadaptationactions(Hallegatteetal.2013).Highersealevelswillalsoallowstormsurgestoreachfurtherinland(IPCC2018).Wildfires.Hotter,drierregionstendtoexperiencewildfiresaffectingcitiesatthewildland-urbaninterface(WUI)—thatis,thetransitionzonewherewoodlandanddensevegetationmeethumandevelopment(Bento-GonçalvesandVieira2020).IntheUnitedStates,theWUIgrewrapidlybetween1990and2010(Radeloffetal.2018).Infact,itwasthefastest-growinglandusetypeintheUnitedStatesafterurbanizationandurbansprawltrends.LanddefinedasWUIinCaliforniagrewbetween1990and2010andnowencom-passes6.4percentofthestate’stotallandarea.WUIgrowthoftenresultsinmorewildfireignitions,puttingmorelivesandsettlementsatrisk.DuringCalifornia’s2020fireseason,nearly11,500structures—morethan6,000residencesand700commercialstructures—weredestroyed.Apartfromtherarepossibilityofwildfiresannihilatingcities,thedirectimpactofwildfiresoncitiesisthesevereburdenofsmokeonpublichealth.TheBayAreaalonesawathree-monthaverageAirQualityIndexabove100in2020.InSouthAfrica,regularlyoccur-ringwildfireshavecausedsignificantdamagetocitiesandtownsalongthecountry’sGardenRoute,whichdependprimarilyonagricultureandtourism.In2017,wildfiressweptthroughtheSouthernCape,destroyingcriticalinfrastructure,displacingmorethan1,500house-holds,anddirectlyaffecting134businessesWind-relateddisasters(hurricanes,cyclones,andtyphoons).Winddisasterscontrib-utetosignificanteconomiclossesandhumancasualtiesincitiesbydestroyingbuildingsandinfrastructure,damagingvegetation,andinducingrespiratorydisease(He,Thies,etal.2021).Alsoincities,extremewindeventshavehaddevastatingeffectsonpoorbuildingstock,includ-inglow-incomehousesinAfricancities(Okunola2019).InTimor-Leste,forexample,CycloneSerojadestroyedover25,000homes(morethan50percentofallhouseholds)inthecapital,Dili,in2021.ThecostofrepairingandreplacingthehousingassetsaffectedwasestimatedatUS$70million(WorldBank2021b).Climatestressors—slow-onseteventsDrought.Anestimated350.0millionadditionalpeopleinurbanareaswillbeexposedtowaterscarcityfromseveredroughtsifwarmingof1.5°Coccurs—andthatestimateincreasesto410.7millionifwarmingof2°Coccurs(Dodmanetal.2022).Anotherestimatepredictsthatmorethan2billionpeoplelivinginurbanareas(including284cities)willfacewaterscarcityTheImpactsofClimateandEnvironmentalChangeonCities173by2050(He,Wu,etal.2021).Theimpactsofdroughtonurbanareasemergegraduallyandlessvisibly.Waterscarcitycanbedrivenbydroughtandheightenedbycompetingwaterusers.Thehigherconcentrationofpeopleinnaturallyaridcitiesresultsinwaterscarcity,especiallyduringadrought.Asurbanizationskyrockets,demandandcompetitionforlimitedwaterresourceswillincrease,anddroughtwillexacerbatethesepressures.Airpollution.Exposuretoairpollutionishigherindensercities.TheanalysisbyAnenbergetal.(2019)oftheworld’s250mostpopulouscitiesestimatesthatconcentrationsofpartic-ulatematterof2.5micronsorlessindiameter(PM2.5)arehighestincitiesinAfrica,EastAsia,SouthAsia,andtheMiddleEast.AirpollutioninmanycitiesintheMiddleEastandNorthAfricaresultsmainlyfromwindblowndust,whereasthatinEastAsiaandSouthAsiaismainlyanthropogenicinorigin.Airpollutantsemittedbyonecitycanbemovedbyairflowandtracedinneighboringareas(Abasetal.2019).Undoubtedly,airpollutionharmshumanhealthandaffectsglobalwarming.Exposuretoambientairpollutionincreasesmorbidity1andmortality,andcontributestotheglobaldiseaseburden(Cohenetal.2017).From2000to2017,residentsofEuropeancitieswereexposedtoPM2.5andozonelevelsthatsurpassedtheWorldHealthOrganizationlimits,despiteamarkedreductioninemissions(Sicardetal.2021).AccordingtoCroitoru,Chang,andAkpokodje(2020),thecostofmortalityandmorbidityinLagosduetoairpollutionfromPM2.5exposureisanestimatedUS$2.1billion,or0.5percentofNigeria’sgrossdomesticproductin2018.BeyondPM2.5,urbanpopulationsareexposedtoground-levelozone,nitrogendioxide,andothercombustion-relatedairpollutants.Landdegradation.Urbanizationaffectsnaturalecosystemsbyremovingforestsandcreatinganewbuiltenvironment.Changesinlandcovercontinuetoaffectcity-scaleclimatebyalteringtheflowofenergy,water,andgreenhousegasesbetweenthelandandtheatmo-sphere.TheIntergovernmentalPanelonClimateChangedefineslanddegradationasanegativetrendinlandconditionstemmingfromhuman-inducedprocesses(IPCC2019a).Theprocessesleadtoalong-termreductioninorlossoftheland’sbiologicalproductivity,ecologi-calintegrity,andvaluetohumans.Changesinlanduse,suchasdevelopinginfloodplains,alsoconstituteastressor.Forexample,pressuretobuildinflood-proneareasmayarisefromlowerrealestatevaluesandlimitedlandforexpansion.Intheirstudyof55Africancountries,Lietal.(2016)revealacorrelationbetweenurbanexpansion,flooddisasters,loss,anddamage.Thesealterationscandirectlyreinforceandcontributetothenegativeimpactsofclimatechange.Ofthemanyexamples,anincreaseinimpervioussurfacesincreasesfloodriskthroughstormwaterrunoff(Hamilton,Coops,andLokman2021);toomanydarksurfaces(suchaspavementandtarroofs)increasetheurbanheatislandeffect(SantamourisandYun2020);cuttingdowntreescontributestogreaterfloodingandtheurbanheatislandeffect(Qin2020);destruc-tionofnaturalmangroveforestsreducesanaturalbuffertocoastalstorms(Dasguptaetal.2019);andsubsidencefromoverpumpingofgroundwaterleadstoincreasedflooding(Erban,Gorelick,andZebker2014).Risingsealevels.Thefutureriseintheglobalmeansealevelstemmingfromthermalexpan-sion,meltingofglaciersandicesheets,andchangesinlandwaterstoragewilldependstronglyonwhichRepresentativeConcentrationPathwayemissionscenarioisfollowed(IPCC2019b).Risingsealevelsthreatenpopulations,criticalinfrastructure,andvaluableassetsincoastalfloodplains,coastalcities,urbanatollislands,arcticcommunities,anddenselypopulateddeltas.Forexample,buildingsandcommunitiesintheurbanatollsofMajurointheMarshallIslandsareathighriskofinundationfromariseinsealevel.Withoutanyformofadaptation,37percentofMajuro’sbuildingstock,mainlyconcentratedinDelap-Uliga-Djarrit,isatriskofpermanentinundationfroma1-meterriseinsealevel(WorldBank2021a).Risingsealevels174THRIVINGalsoincreasethesalinityofsurfaceandgroundwatersourcesandcoastalaquifersthroughsaltwaterintrusion,therebyaffectingurbandrinkingwatersupplies.Increasesinsealevelandintropicalcyclonestormsurgeandrainfallintensitywillraisetheprobabilityofcoastalcityflooding,withmorethanabillionpeoplelocatedinlow-lyingcitiesandsettlementsexpectedtobeatriskfromcoastal-specificclimatehazardsby2050(IPCC2022).Thecompounding,cascading,andindirecteffectsofclimatechangeTheclimate-relatedshocksandstressorsthataffectcitiesdonotoccurinisolationbutofteninteractwithandintensifyeachother,makingtheirpossibleimpactsoncitiesevenmoreuncertainandpotentiallydisastrous.Forthatreason,thecompoundrisksofclimaticshocksarereceivinggreaterscrutiny.Forexample,cyclonesandheatwavescanoccurconcurrently,addingcomplexitytotheireffectsonlocalpopulationsandassets(Fordetal.2018).Removalofurbantreescouldcompoundtheeffectsofheatwavesandfloods.Foodsupplieswillbeaffectedbyproductionlossesfromheatanddrought,compoundedbyheat-inducedfalloffsintheproductivityofworkers(asdiscussedlaterinthischapter).Riskscanalsobetransmit-tedacrosspopulations,places,andsectors,leadingtocascadingimpacts.Forexample,ruralmigrantsfleeingadroughtcansettleinprecariousinformalsettlementsinurbanfloodplains,withcascadingrisksforothergroupsofpeopleandlocations.Wildfiresinagriculturalregionscanincreaseurbanairpollution,whiledisruptingthesupply,andthusprices,ofessentialfooditems(Zscheischleretal.2018).Meanwhile,incities,criticalinfrastructure,suchastranspor-tationsystemsandpowergrids,isgenerallyinterdependent,sothefailureofoneelementornodecouldresultinacascadeofadverseevents.Thus,stormsurgesandextremeheatcouldleadtopoweroutages.Indeed,theimpactsofclimatechangearepervasive.Forexample,urbanresidentswillsufferfinanciallythroughhigherdemandforelectricityforair-conditioning,medicalcosts,andmissedwork(seebox3.8laterinthischapter).Transportationnetworksinurbanareasareespeciallyvulnerabletoweather-relatedhazards.Nearlyalltransportationmodesareatriskfromextremeeventssuchasextensivefloods(Reballyetal.2021),whichcanreducethecapacityofaroadtransportationnetworkbyrenderingitimpassable,therebycreatingseveretrafficcongestion.Thetangibleimpactsofthephysicaldestructionoftransportationinfra-structuresuchasroadsandbridgescanbequantifiedinmonetaryterms.2Somelow-incomecountriesclearlyshowtherelationshipbetweenrainfallandgrowthofgrossdomesticproduct.Intheirstudyoftheimpactsoflarge-scaleurbanfloodsbetween2003and2008on1,868citiesin40mainlylow-incomecountries,Kocornik-Minaetal.(2019)findthatlow-elevationurbanareas,whichtendtoconcentratemoreeconomicactivitypersquarekilo-meter,experiencefloodingmorefrequently.Onaverage,floodingreducesacity’seconomicactivityby2–8percentinthefloodyear.Thiseffectmasksthemedium-tolonger-termdamagesthatcouldincludethecostsofrestoringdamagedinfrastructure,theopportunitycostofadultspulledawayfromproductivework,andthelossofhumancapitalaschildrenarewith-drawnfromschools.MostcitieswiththehighestrelativecoastalfloodlossesareinSouthandSoutheastAsia,andtheyareprojectedtosufferUS$52billioninfloodlossesperyearby2050(Hallegatteetal.2016).Low-lyingcitiessuchasHoChiMinhCity,Vietnam,orLagos,Nigeria,areespeciallyvulnerable.Desmetetal.(2021)evaluatethecostsofcoastalfloodingusingaspatiallydisaggregateddynamicmodelandfindthatrealglobalgrossdomesticproductcouldfallby4.5percentwithoutadynamicmigrationandinvestmentresponse.Althoughwildfiresmaybreakoutinperi-urbanorevenruralareas,citiesbeartheireconomicburdens.ThetotaldirectandindirecteconomicimpactsofwildfiresoncitiesincludetheTheImpactsofClimateandEnvironmentalChangeonCities175costofdamages,healthcosts,andindirectlossesfrompowershutoffs,businessclosures,travelcancellations,andsupplychaindisruptions.TheeconomiclossesofCapeTown’s2021wildfireswereestimatedatUS$1.5million,affectingbusinessesandagriculturalyield.ThewildfiresacrossCaliforniain2018producedUS$7.8billioninestimatedhealthcostsintheBayArea,upticksinhospitaladmissions,andashiftinthehousingmarket,withrentsjumpingmorethan40percent(Bellisario,Cowan,andRaisz2021).BertinelliandStrobl(2013)findthathurricanestrikesreduceincomegrowthby1.5percent,onaverage,atthelocallevel.Indaco,Ortega,andTaspinar(2021)studiedtheeffectsofHurricaneSandyonfirmsinNewYork,findingthatfloodingledtoreductionsinemploy-mentofabout4percentandaveragewagesofabout2percentamongaffectedbusinesses.Theeffectswereheterogeneousacrossboroughs,reflectingtheseverityofflooding,buildingtypes,andindustrycomposition.Adaptationalsoinvolvedmigration,withsomefirmsrelo-catingtootherneighborhoods.Shughrue,Werner,andSeto(2020)findthatcitiesarevul-nerabletoeconomicharmeveniftheyaregeographicallydistantfromthedirectimpactsofcyclones,andthatvulnerabilitytosecondaryimpactssuchasmaterialshortagesandpricespikingishighestincitiesthatdependstronglyontheglobaltradenetworkbuthaverela-tivelyfewersuppliers.Changesinwateravailabilityandintheintensityandfrequencyofdroughtwillhavedirecon-sequencesforcitiesintermsofwaterresourcesandwatermanagementsystems.Oneinfourcitiesglobally,withatotalofUS$4.2trillionineconomicactivity,isclassifiedaswaterstressed(McDonaldetal.2014).Adroughtaffectsacity’spowerinfrastructure,reducingoverallactivity.In2016,18thermalpowerplantsacrossIndiancitiesexperiencedsignificantoperationaldisruptionsandshutdownscausedbywatershortages,costingIndia1percentofitsannualpoweroutput(WRI2017).Adroughtcanalsoleadtohigherconcentrationsofpollutants,lackofadequatewaterflowforsewerage,andflood-relateddamagetophysicalassets.Deficitsintheurbanwatersupplywillgreatlyaffectthefutureavailabilityandcostofwaterincitiesandperi-urbanareas.Duringdroughts,healthconditionsworsenand,followingtheabsenceofadequatesanitation,citieswithhigherpopulationdensitiessuffertherapidspreadofdisease,reducingtheprobabilityofemployment(DesbureauxandRodella2019).Droughts,morethanfloods,reduceworkerproductivityandlaborincomeandmayhavealonger-lasting,severerimpactonworkersandfirms(Damaniaetal.2017).Underlyingnon-climate-relatedstressesinurbanareascanalsoexacerbatetheeffectsofclimatechange.Forexample,highratesofinformaldumpingofwasteexacerbatepluvialfloodsasrefusepilesupindrains,waterways,andopenspaces(Jha,Bloch,andLamond2012).Studieshavelookedathowariseinsealevelandheavyrainfallmayaffecturbandrainagesystems(Grip,Haghighatafshar,andAspegren2021)andtheeffectsthatacombi-nationofdaytimeandnighttimeheatextremesmayhaveonurbanareas(Wangetal.2021).Airpollutionalsocanactasastressoronurbanresidents,especiallywhencombinedwiththeheat-relatedeffectsofclimatechange(HarlanandRuddell2011).Otherstressors,suchasglobalpandemics,willpushcitiestomanagecompoundrisks(Phillipsetal.2020)andforcecitiestounderstandtheirlinks,suchashowpandemicsmayaffectsharedtranspor-tation(Moracietal.2020)andairpollution(andviceversa).Forexample,somecitiesexpe-riencedreducedairpollutionduringtheCOVID-19pandemic,whereascitieswithmorepollutionhadhighercaseloadsofCOVID-19(ChingandKajino2020).AccordingtotheIntergovernmentalPanelonClimateChange,multipleclimatehazardswillcontinuetooccursimultaneously,withinteractionacrossmultipleclimateandnonclimaterisks,therebycompoundingtheoverallriskandcausingriskstocascadeacrosssectorsandregions(IPCC2022).Thus,predictingthemedium-tolong-termeffectsofclimatechangeoncitiesischal-lenging,tosaytheleast.3176THRIVINGMeasurementofeffectsisnotalwaysstraightforwardGlobalwarming4isaprotractedglobalphenomenonwithheterogeneouslocaleffects,makingtheassessmentofsucheffectscomplex(seebox3.2forabriefoverviewofthesechallenges).Thedifficultyinmappingthephysicalimpactsofextremeweathereventsiscompoundedbythenonlinearitiesintheclimatesystem,acrossbothspace5andtime.6Economicgeog-raphershavethusintroducedspatialdynamicassessmentmodelstotakeintoaccountthetemporalandspatialdimensionsoftheeconomicimpactsofclimatechange(seeBalboni2021;DesmetandRossi-Hansberg2011,2012,2015;Desmetetal.2021).Somemodelsalsoemphasizetheroleofeconomicadaptationthroughmigration,trade,andinnovation.CruzandRossi-Hansberg(2021)allowchangesinlocaltemperaturestoinfluencethreecharacter-istics:localproductivity,localamenities,andthedifferencesbetweenbirthanddeathrates.Theyfindthatthespatialheterogeneityoftheimpactofglobalwarmingisstark—welfarelossesgloballyof5percent—butthattheworld’spoorestregionslosesubstantiallymore—asmuchas15percent.Theirestimatessuggestthatanincreaseof1°Cinlocaltemperaturesimpliesadeclineinamenitiesofabout2.5percentintheworld’shottestareasandacommen-surateincreaseintheworld’scoldestareas.Theeffectsoftemperatureonproductivityarelargerandasymmetric:anincreaseof1°Cinlocaltemperaturesleadstoa15percentdeclineMeasurementissuesBecauseofthebroaduncertaintiesabouthowtheeffectsofclimatechangemayplayoutovertime,thebestapproachtopredictionsmaybetounderstandhowthecurrenteffectsmightchangeunderdifferentscenarios.Thosescenarioswillneedtoincludetheeffectsofclimatechangeacrossdifferentfuturewarmingestimates—thatis,1.5°C,2.0°C,or3.0°C.Inaddition,scenarioplanningwillhavetoconsiderurbanization,landusechanges(AvashiaandGarg2020),anduncertainty.Thesescenarioswouldalsobeaffectedbyassumptionsaboutotherkeydrivingforces,suchastherateoftechnologicalchangeandprices.Evenwhendirectimpactscanbeestimatedwithsomelevelofconfidence,theindirecteffectsaremorecomplextoassess.Directeffectsfromclimaticshocksorexposuretoprolongedstressorsincludelossoflife,ecosystemdegradation,andeconomiclosses.Withnewadvancesinattributionscience,itisnowpossibletoassessthelikelihoodthatclimatechangecaused,orintensified,anevent.Thispossibilityappliestoseveraltypesofshocks:flooding,tropicalstorms(Reedetal.2020),drought(Philipetal.2018a,2018b),heatwaves(Vautardetal.2020),andfires(vanOldenborghetal.2021).Despitetheseadvances,attributingtoclimatechangelivesloststilldoesnotoccurcommonly,nordoesassessingtheenvironmentalconse-quences.Severalstudies,however,haveplaceddollarvaluesontheeconomiclossesfrombothfloodinganddrought(Frame,Rosier,etal.2020;Frame,Wehner,etal.2020).Anassessmentofindirecteffectsisparticularlyimportantinurbanareasbecausetheseareasconcentratesomuchactivity,functioningasintegratedsystemswithvarioussectorsandinfrastructurescloselyinterlinked.Theseindirecteffectshaveimpactsoncitiesdowntotheindividualswholiveinthemandinadjacentareas.Attimes,indirecteffectscanhaveglobalimpacts,suchasonmentalhealthandconflicts(Evans2019),andcancausetradeinterruptionsandreductionsingrossdomesticproduct(Botzen,Deschenes,andSanders2019).Box3.2TheImpactsofClimateandEnvironmentalChangeonCities177inproductivityinthewarmestregionsanda10percentincreaseinthecoldestregions.Tradeandmigrationcanactassubstitutesforadaptationmechanisms.Conteetal.(2021),usingaspatialdynamicmodeloftheworldeconomy,extendthespatialgrowththeoryofDesmetetal.(2021)toincludemultiplesectors,withsectoralproductivitydependingontempera-turesandwithproductionleadingtoemissionsthatfeedbackintotheatmosphericstockofcarbon.Theyfindthatagriculturebecomesmorespatiallyconcentrated;however,regionsthatloseagriculturedonotreplaceitwithproductivenonagricultural(manufacturingorservices)sectors.Urbanindustrialhubsfaceadditionalchallengesbecausetheyrelyonkeyinputssuchaswater,timber,orenergyfortheproductionprocess,orbecausetheymaysufferfromdis-ruptionsinsupplychains.GlobaltrademeansthatatsunamiinJapanorfloodsinThailandcanresultinsignificantlossesforautomobilemanufacturersinOntario,Canada,becauseofshortagesofpartsanddeliverydelays(KovacsandThistlethwaite2014).IntheUnitedStates,in2021freezingtemperaturesinTexasledtoanenergyblackout,whichthenforcedmajorsemiconductorplantstoclose,exacerbatingthesemiconductorshortagetriggeredbytheglobalpandemicandfurtherslowingtheproductionofmicrochip-dependentcarsinseveralcountries(Fitch2021).Thus,weatherandclimateeventsincreasinglydisruptsupplychainsthatrelyonspecializedcommoditiesandkeyinfrastructure.Ofcourse,citiesnotonlyhostmanufacturersandstorageanddistributionfacilities,butalsoserveasprincipaldemandpoints.AsGomezetal.(2021)find,extremeweathereventstransmitfoodproductionordis-tributionlossesthroughsupplychainstootherlocations,usuallywithcitiesbearingthebruntoftheshocks.Changingclimaterisksareexpectedtointeractwiththealreadyintensecompetitionforlandandhousinginurbanareas.Thedesiretoliveclosetojobsandamenitiesmeansthatevenareaswithhighfloodorlandsliderisksoftenhostlargecommunities.Becausehousingpricesoftenreflecttherisks,thepoorestoftenliveinthemosthazardousareas(asubjectdiscussedinmoredetaillaterinthechapter).Thosesettledinvulnerablelocationsfacerepeateddamagetotheirhomesandmayneedtobearthebruntofretrofittingtoenhanceresilience.Homeownersalsofacerisinginsuranceratesastheinsuranceindustrybetterunderstandstherisksofhazards.Becausethemarketforhousingisnotstatic,householdscanmigrateinresponsetorepeateddisastereventsandwiththeadventofimprovedhazardinformation.Forexample,inCali,Colombia,poorerresidentsdecidedtomoveawayfrominner-cityneighborhoodstoinformalsettlementsontheperiphery(seethedetailedanalysisbyCamposGarciaetal.2011).Overtime,housingmarketsmayalsoreflectheterogeneityinbeliefsaboutlong-runclimaterisks.Baldauf,Garlappi,andYannelis(2020)findthathousingpricesintheUnitedStatesexhibitdifferentelasticitiestoclimateriskandtomeasuresofbeliefsaboutclimatechange.Ineffect,inzonesincreasinglyatriskoffloods,housesinneighborhoodswithhighlevelsofbeliefinclimatechangeriskssellatadiscountcomparedwithhousesinneighborhoodswithhighlevelsofdenial.ClimatechangecouldalsoaffectcitiesbyunderminingthefiscalhealthoflocalgovernmentsClimateimpactscouldhavefiscalconsequencesforlocalgovernments,primarilythroughdeclinesincertainsourcesofrevenuesandgrowingexpendituresrelatedtoclimatehazards(seebox3.3forabriefoverview).Gilmore,Kousky,andSt.Clair(2022)demonstratethepathwaysbywhichclimateshocksandstressesdrivelocalgovernmentfiscalstressbyaffectingspecificcategoriesofrevenueandexpenditures.Declinesinrevenuecanresult,178THRIVINGforexample,notonlyfromthedestructionofpropertybutalsofromadropinpropertyvalues.AnanalysisofrealestatetransactionsacrosstheUnitedStatesshowsthatfrequenttidalfloodingcausedbytheriseinsealevelresultedinaUS$15.9billionlossinhomevalueappreciationinjustoveradecade(seeBernstein,Gustafson,andLewis2019).Householdsandfirmsrespondtophysicalimpactsbyrelocating(eithertemporarilyorpermanently)orbychangingtheirconsumptionofandpaymentforbasicservices,furtheraffectingrevenuesources.Ontheexpenditureside,localgovernmentsareincreasinglyattheforefrontofresponsestoclimateshocksviafundingemergencyservices,investinginclimateadaptation,ordealingwiththechronicorcumulativeimpactsonpeopleandassets.InthePhilippines,localgovernmentunitsaretheprimaryfirstrespondersduringadisaster,andthenationalgovernmentprovidesfinancingandtechnicalassistancewhenlocalcapacityislimitedoroverwhelmed.However,asdemonstratedbystudiesonpublicexpenditurereviews(WorldBank2020b)andparametriccatastrophicriskinsuranceprograms(WorldBank2021e),lackofcoordinationandlowcapacitybylocalagencieshaveresultedinslowresponses.Althoughthenationalgovernmentcontributedfrom66percentto100percentoftotalpostdisasterexpendituresinthePhilippinesbetween2015and2018,thatcontributiondidnotaccountforallspendingbythelocalgovernments,includingforpreparednessandriskreduction.Jerch,Kahn,andLin(2021)provideevidencethatexposuretohurricanesjeop-ardizeslocalprovisionofpublicgoodsinthelongerterm.Theirlookat2,000localgov-ernmentsinUSAtlanticandGulfstateshitbyhurricanesfindsreductionsintaxrevenuesandexpendituresandincreasesinthecostofdebtinthedecadefollowingexposure.Theyalsofindthatmunicipalitieswithahigherproportionofracialminoritiessufferedhigherexpenditurelosses(twofold)andhigherdebtdefaultrisks(eightfold).Theresultsimplythathurricane-induceddeclinesincurrentfinancialresourcescantranslateintolowerfutureinvestments.CitybudgetsbearthebruntofclimateeventsInlow-andmiddle-incomecountries,localgovernmentsareoftenforcedtoredirectfinancesfromotherservicedeliveryaccountsinordertofinancerecoveryactivities,therebyunderminingnotjustthefiscalpositionoflocalauthoritiesbutalsothedeliveryofbasicservices.Fundsprovidedbynationalorprovincialgovernmentsinthewakeofdisasters,includingprogramsforadaptation,canprovideanimportantsourceofbudgetaryassistanceforlocalgovernments.Inmanycountries,however,asubstantialfundingshortfallexists,evenforthemosturgentofneeds.TheWorldBank’sSouthAfricaDisasterRiskFinancingDiagnostic(forthcoming)collatesfindingsfromcasestudiesoffloodsineThekwini(previously,Durban)anddroughtinCapeTown.Itfindsthatbothmetropolitangovernmentswereforcedtoreprioritizespendingawayfromcoreservicedeliverytowarddisasterresponse.Ninety-eightpercentoftheestimateddamagefromthe2019floodsineThekwiniwasunfunded,andthebudgetwasfoundfromgrantspreviouslyallocatedforeducationinfrastructure,maintenanceofprovincialroads,anddevelopmentofhumansettlements.Inaddition,althoughmetropolitancitieshaveaccesstofinancinginstrumentstofunddisasterresponses,theapplicationprocessistooonerousandthefundscomewithhighlevelsofoversight,discouragingtheiruse.Box3.3TheImpactsofClimateandEnvironmentalChangeonCities179Climatechangeistransformingthesustainabilityofurbanconstruction—fortheworseFutureclimatechangemayrendermanyoftheworld’scitiesunsustainableorevenunliv-able.Ifpeoplecanmigrateandifcapitaldepreciatesquicklyenough,riskcouldbetransferredeasilyacrosslocations.Asthissectiondescribes,themigrationofindividualsinresponsetoclimateshocksandstressorsisalreadyanestablishedphenomenon.Muchofthismigrationtakesplacefromruraltourbanareasandwithinnationalborders.However,themigrationofcapital—specifically,capitalthatisembeddedindurablestructuresandthusdepreciatesveryslowly—isawholedifferentmatter.Aspartofthebackgroundresearchconductedforthisreport,DesmetandJedwab(2022)investigatedtowhatextentthemostexpensive(andverydurable)realestatestructures—skyscrapers—arenotbeingbuiltinlocationsthatwouldbecomeincreasinglyunlivablebecauseofextremeheat.Theyfocusontallbuildingsfortworeasons.First,urbanplannersandpolicymakersareencouragedtoincreasedensity,includingbybuildinghigher,tomeetthegrowingdemandforspacebyhouseholdsandfirms.Second,tallbuildingsarebuilttolast,makingthemverydurableovertimeandthuslesslikelytoreallocatecapitalinresponsetochangingclimaterisks.TheauthorsusetheEmporisdataset,aninventoryofalltheworld’sbuildingsexceeding55metersinheight,whichincludesinformationontheiryearofconstructionandheight.7Thisstudyencompasses12,877urbanagglomerationsin182countries,whichtogetheraccountfor90percentoftheworld’scurrenturbanpopulation.DesmetandJedwabhypothesizethat,astheinformationaboutthegrowingrisksofclimatechangeincreases,overtimetheconstructionofdurablestructures(thatis,thosewithembeddedcapitalthatcannotbeeasilyreallocatedoverspace)shouldslowin“futurebadlocations.”8Ineconomicterms,ifasetoflocationsbecomesunlivableinthenot-too-distantfuture,theassumptionisthatagovernment,actingasadynamicsocialplanner,9woulddiscouragetheconstructionofexpensiveandverydurablestructuresintheselocations.Suchanapproachshould,attheveryleast,lowertherateofconstructionofsuchbuildingsinfuturebadlocations.DesmetandJedwabfindtheopposite,however:countriesarenotnecessarilyavoidingconstruc-tionofexpensiveanddurablestructuresinlocationsthatinthefuturearepotentiallyunsustain-ableorevenunlivable,implyingthattheworldcouldseeagrowingdynamicspatialmisalloca-tionofcapitalovertime.Despitethefactthatclimatechangeconferences—Kyoto(1997),Doha(2012),andParis(2015)—putthespotlightontherisksassociatedwithfutureclimatechange,thisinformationdoesnotseemtobeassociatedwithanychangesinbuildingpatterns.Evidencealsoexiststhatlessdemocraticcountriesaremorelikelytobuilddurableskyscrapersinless-livablelocations.Publicpolicycouldencouragesettlementinmorehazard-proneareasMoralhazard—atermappliedtoengagementinrecklessbehaviorbecauseonedoesnothavetofullybearthecostofthatbehavior—iswidelyacceptedasoneoftheleadingcausesofthe2008globalfinancialcrisis.Bankmanagerspushedforrecklesslendingtoincreaseprofits(andcon-sequentlytheirbonuses)undertheimplicitguaranteethatifthingswentwrongthegovern-ment(taxpayers)wouldbailthemouttoavoidasystemiceffectontheeconomy.ThatexampleperfectlycapturestheissueaspresentedintheseminalpaperbyKydlandandPrescott(1977),whoexplainthattheoptimaltimetoapplyapolicyisofteninconsistent.Thatis,governmentwouldwanttodiscouragerecklessbehaviorbyavoidinganycommitmenttobailoutlarge180THRIVINGbanks,butitwouldwanttorescuethemoncetheyarebankrupt.10KydlandandPrescott(1977)focusonmonetarypolicy(whichismoreeffectivewhenitisasurprise)butprovideanexampleoftimeinconsistencyinvolvingfloodinsurance.Verymuchlikebankbailouts,governmentsubsidiesoffloodinsurancewouldencouragebuildinginrisk-proneareas;afteradisasteragovernmentnormallyofferssomemonetaryrelief,therebyactingasadefactoinsurerandpro-vidinganimplicitguaranteethatmayencouragebuildinginflood-proneareas.AlargeUS-basedliteraturehasstudiedtheeffectsofsubsidizedfloodinsuranceandreliefaid(Deryugina2014;Gregory2017;Kousky,Luttmer,andZeckhauser2006).Deryugina(2014)pointsoutthatfloodinsurancebenefitsrisk-averseindividualsandthosewhoarecredit-constrained,butitdiscouragestheprovisionofprivateinsuranceandincreasesthewilling-nessofpeopletoliveindisaster-proneareas.Moreover,shepointsoutanothertypeofmoralhazard:becauseunemploymentinsurancepremiumsdonottakeintoaccountdisasterrisk,theysubsidizeactivityinriskierareas..Thesamepatternemergesforothertypesofdisasters.BaylisandBoomhower(2019)estimatethatfireinsuranceintheformoffreefederalfireprotectionmayhaveincreasedconstructioninareasathighriskoffireinthewesternUnitedStatesby2.5percentoverall,andbyanevenhigherpercentageinthehighest-riskareas.Despitetheincreaseinriskybehavior,however,insuranceandpostdisasteraidpoliciesmaystillbeoptimalinwelfareterms.EvaluatingalargeUSdisasterreliefprogram(LouisianaRoadHome)inlocationsaffectedbyHurricaneKatrina,Gregory(2017)findsthatthedistortioncausedbyanexantepromiseofpostdisasteraidisnotlargeenoughtojustifycommittingtolessgenerousrelief.Eventhemostconservativeestimatefindsthattheexcessburdenduetoriskylocationchoicesismorethanoneorderofmagnitudesmallerthanthewelfaregainsfromguaranteedpostdisastertransfers.Thedeadweightlossfromactuariallyunfair(notaccount-ingforrelativerisk)disasterreliefisestimatedtobeonly4percentofthetotalyearlyreliefexpenditure.Twosolutionstothemoralhazardlocationissuearetoforbidconstructioninriskierareas(viazoning)andtocreateactuariallyfairreliefinsurance.However,bothsolutionsfacepolitical,social,andtechnicalchallengesbecauseurbanlandisofteninshortsupply,andfloodriskishardtodetermineprecisely(especiallyincountrieswithpooradministrativecapacities).AvnerandHallegatte(2019)explainthat,whenmostofacityisflood-prone,subsidizedinsuranceimproveswelfare,andthesameappliestoareaswithalowriskofflooding.Zoningbecomesfeasibleonlyinthepresenceofgeographicallylimitedareasatveryhighrisk.Howdoesclimatechangeaffectcitiesindirectly?ClimateshocksinruralareasleadtofasterurbanizationWhenextremeclimateeventshit,peopleinthecountrysidecanseekshelterincities,whichhavestrongerinfrastructureandmorehospitalsandotheressentialservices.InSub-SaharanAfrica,forexample,manufacturingtownshavegrownafterdroughtsintheagriculturalhinterlands(Henderson,Storeygard,andDeichmann2017).Burzynskietal.(2019)predictthatclimatechangewillinducebothvoluntaryandforceddisplacementof200million–300millionpeopleoverthecourseofthetwenty-firstcentury,althoughonly20percentofthatmovementwillinvolvecross-bordermigration.Inanotherrecentstudy,Benveniste,Oppenheimer,andFleurbaey(2020)findsubstantiallysmallernumbers,estimatingexcessclimate-inducedcross-bordermigrationflowsin2100at75,000.Mostpeopledisplacedbynaturaldisas-terseventuallyreturntotheirplaceoforigin,butnotalldo.Somesettleinthesuburbsorontheoutskirtsofcities,placingfurtherstressonurbaninfrastructure,services,andresources.TheImpactsofClimateandEnvironmentalChangeonCities181About40percent(14million)oftheresidentsofBangladesh’scapital,Dhaka,liveininformalsettlements,and70percentofthosewereforcedtoleavetheirhomesbecauseofphenomenarelatedtoclimatechange,includingcyclonesandcoastalandriverbankerosion(LombrañaandDodge2021).Understandingthesepatternsremainsparamountbecausepoorercountriestendtourbanizefasterandearlierthandevelopingstatesofthepast.Withthatinmind,investigatorsareturningtheirattentiontothepushfactors(Barrios,Bertinelli,andStrobl2006;MaurelandTuccio2016),includingthechangingclimate(KaczanandOrgill-Meyer2020).AbackgroundglobalanalysisconductedforthisreportbyChlouba,Mukim,andZaveri(2022)showsthatfrom1985to2014periodsofextendeddroughtresultedinfasterurbangrowth,providingfurtherevidenceoftheimportanceofpushfactors.Thefindingsindicateanunam-biguouspositiverelationshipbetweendroughtinareassurroundingcitiesandurbanexpan-sion.ThestudyfocusesontheshareofthecalendaryearthatanareaexperienceddroughtandthesizeofterritoriesclassifiedasurbanlandintheWorldSettlementFootprintdatabase.Theresultsindicatethat,ifanareaspentthepreviousyearunderdroughtconditions,nearbycitieswouldexperiencea15percentincreaseinurbanlanduse.Althoughdroughtrarelyextendstoanentireyear,therelationshipbetweendroughtandurbangrowthremainssubstantial.Thisresultechoespatternssuggestedbyrecentstudies(suchasCastells-Quintana,Krause,andMcDermott2021),buttheapproachtakenbyChlouba,Mukim,andZaveri(2022)offersseveralkeyimprovements.Mostcrucially,theirapproachcapturestheeffectsofdroughtspecificallyoncropsthatmattermostforlocaleconomiesratherthanstudyingtheeffectsofmerechangesintemperatureorprecipitation.Similarly,theyemployanovelandhighlydisaggregatedmeasureofbuilt-upareasusingremotelysenseddatafromtheWorldSettlementFootprint.Thisapproachallowsforaglobalscopeofanalysisrarelyattemptedintheexistingliterature.Althoughlocalmigrationflowsaredifficulttomeasureaccurately,evidencesuggeststhatmigrationtocitiesisoneofthekeycopingmechanismsindrought-strickenregions(Castells-Quintana,Lopez-Uribe,andMcDermott2018).Forexample,migrationcouldoccurgraduallyasaresponsetocumulativechangesinwatervariability.Recentworkhasshownthatmigrationisasignificantresponsetopersistentdrought,particularlyforhouseholdsstronglydependentonagriculture.Evidencealsosuggeststhatclimatemigrantsareoftenfromthelowerendoftheskilldistribution(SedovaandKalkuhl2020;Zaverietal.2021).Indeed,estimatessuggestthatthoseescapingdroughtoftenmigratetonearbycitiesandarefrequentlylessadvantagedthantypicalmigrants,raisingimportantimplicationsforthemigrantsthemselvesandthereceivingurbanareas(KleemansandMagruder2018).Chlouba,Mukim,andZaveri(2022)donotoffermuchinsightintotheexactdecision-makingcalculusofmigratingfamiliesandindividuals;nevertheless,theirresultslineupconvincinglywithevidencegatheredinspecificcontexts.Forexample,cropconditionsemergeasaleadingfactorpushingSomalisoutoftheirhomes.TheSomaliaUrbanizationReviewmeasuredcropconditionsusingthenormalizeddifferencevegetationindex—aremotelysensedmeasureoflandgreenness—toinvestigatehowSomalisreacttodrought(WorldBank2021c).Thereportfindsthat,whencropssufferfromlackofwater,peoplemove,manyofthemtocities.Followingadroughtin2017,manySomalisdidpreciselythis,increasingthenumberofinter-nallydisplacedpersons(IDPs)inBaidoaandMogadishu.Thesignificanceofdroughtdecreaseswithtime,suggestingthatthedecisiontomigrateisoftenreachedratherquickly,atleastintheSomalicontext.AccordingtoChlouba,Mukim,andZaveri(2022),therelationshipbetweendroughtandurbangrowthisparticularlypronouncedinlow-andmiddle-incomecountries,hintingatalikelyexplanationfortherapidurbanizationinthesecountries(Glaeser2014).Ofthe182THRIVINGnoticeableheterogeneity,theeffectsarestrongestinpoorercountries—thatis,countriesthatdependmoreonagriculturefornationalincome.Between1985and2014,industrial-izedcountriessawanaverageeffectthatwasninetimessmallerthantheeffectofdroughtonurbangrowthobservedinnonindustrializedcountries.11Theeffectsarealsostrongestforsmallerandlessdevelopedurbanareas—thatis,thosecharacterizedbylowerbaselinesharesofurbanlandandasmeasuredbynighttimelightintensity,respectively.Althoughcitiesinhigh-incomecountriesaresomewhatimmunetodrought-inducedurbangrowth(likelybecauseofboththeirlocationandagreaterabilitytocopewithharshclimaticcondi-tions),thephenomenonishardtomissinmiddle-incomeand,aboveall,low-incomecoun-tries.Notsurprisingly,themostaffectedregionsaretheMiddleEastandNorthAfricaandSub-SaharanAfrica.SouthoftheSahara,asingleyearspentunderdroughtconditionsispredictedtoincreasetheshareofurbanlandwithineach55×55kilometercellby5percent.InSub-SaharanAfrica,low-incomecountries—whererain-fedagriculturecontinuestobeacrucialsourceofsubsistenceformuchofthepopulation—aremoreaffectedbydroughtthanmiddle-incomecountries.TheeffectofdroughtontheshareofurbanlandamongthepoorestcountriesinSub-SaharanAfricaisaboutthreetimesaslargeastheaverageeffectestimatedfortheentireAfricancontinent.Again,thisfindingsupportstheimpressionthattheimpactsofclimatechangearestrongestinregionsdependentonsubsistenceagricul-ture(Cattaneoetal.2019).Thus,itisnowonderthatwhentheclimatechangesso,too,dopeople’slivelihoods.Thedrought–urbangrowthnexusisclearestincontextswheredroughtremainsapersistentproblem.Oneofthecommonformsofadaptationtochangingclimateiscropmigration—thatis,whenconditionsforagriculturebecometoopoorinoneregion,farmersmovetheircropstoanotherone.Inregionswheredroughthasbeenapersistentissue,however,thismechanismisnotalwaysavailablebecausefewerpotentiallyproductiveareasremain.Chlouba,Mukim,andZaveri(2022)showthisproblemempirically.Theeffectofdroughtsisstrongerinlocationsbelowtheglobalprecipitationmedian.Inotherwords,areasthatare,onaverage,morehumid(andwet)thantheglobalmediandonotexperienceurbangrowthatnearlyasfastapaceasareasthatnormallyaregenerallydry.Inthesecontexts,epitomizedbytheSahelregion,someofthefastest-growinginformalsettlementsattheoutskirtsoflargecitieshavebeenobservedoverthelastfewdecades.Theresearchalsosuggeststhatinrapidlyurbanizingenvironmentsdryspellsmoststronglyaffecturbangrowth.Manycountriesinthedevelopingworld,mostnotablyinAfrica,remainpredominantlyruralcountriesbuthaveprecipitousratesofurbanization.InBurundi,forexample,87percentofthepopulationcontinuestoresideinruralareas,makingthecountryoneoftheleasturbanizedglobally.Atthesametime,Burundihasoneofthefastesturbangrowthratesinitsregion—5.7percentbetween2000and2019(Mukim2021).IncountrieslikeBurundi,theeffectofdroughtonurbanizationisparticularlynoticeable.Thebackgroundanalysesconductedforthisreportindicatethattheeffectofdroughtonurbanlandinareasthatwerebelowtheglobalmedianintermsofshareofurbanlandin1985ismorethan10timesbiggerwhencomparedwithareasabovethemedianin1985.Inotherwords,droughtappearstobedrivingmigrationtocities,especiallyinlocationsthathadrelativelylittleurbanlandonlyfourdecadesago.TheSomalicityofGalkayoisacaseinpoint.Thoughttohostmorethan100,000IDPsacrossdozensofsettlements,Galkayohastripleditsspatialfootprintsince1985.The2017droughtinGalkayo’ssurroundingareasbroughtintensofthousandsofnewarrivals(WorldBank2021c).Thechallengeofclimate-drivendisplacementiscompoundedbythefactthatmanymigrantsremain“invisible”inofficialstatistics.Onereasonforthatinvisibilityisthedifficultyinuntan-glingthenexusofclimatechange,conflict,andeconomicmigration;andnationalgovernmentsTheImpactsofClimateandEnvironmentalChangeonCities183(nottomention,hostcities)arenotwellequippedtocountandcategorizedisplacedpop-ulations.PoliciesaimedatassistingIDPsoftenpaintthemwithabroadbrush,neglectingtodifferentiatebetweenmigrantswhoseektointegratethemselvesintolocaleconomiesandstayforthelongtermanddisplacedindividualswhomerelyseekshorter-termprotec-tionfromtemporarynaturaldisastersintheirlocationsoforigin.UrbanIDPswhostaywiththeirfamiliesandrelativesescapeofficialstatistics,eventhoughtheirarrivalputsadditionalpressureonresource-strickenservicedeliverysystems.Finally,authoritiesincountriessuchasBurundidonotalwaysrecognizedomesticpoliticalreasonsasasufficientmotivefordisplace-mentandsodonotregistercertainIDPs.Climate-inducedmigrationisassociatedwithgrowingurbansprawlDrought-drivenurbangrowthoftentakestheformofexpandinginformalsettlementswhereservicedeliveryremainsachallenge.Whenclimatemigrantsarriveinurbanareas,theyoftenclusterinperipheralareasthatareunderpreparedforhostinglargenumbersofpeople.Ratherthanseamlesslyintegratingintothehostcommunitiesandeconomies,thesemigrantsoftenjointhecommunityofthedisplacedalreadythere.Suchcommunitiesareinvariablyfoundinurbanperipheries,whichhavelimitedjobopportunities,lackbasicinfra-structure,andhaveservicedeliverysystemsthatremainintheirinfancy.TheWorldBank’sBurundiUrbanizationReviewdescribestheclusteringofIDPsaroundtheBujumburaairport,hardlyanideallocationforthegrowingcity’snewinhabitants(Mukim2021).InLiberia,thePeaceIslandCampestablishedontheoutskirtsofMonroviaduringthecountry’scivilwarsoffersanexampleofarefugeecampthatevolvedintoapermanentsettlement(WorldBank2020a).Informalsettlementsestablishedinsuboptimallocationsincreasethelikeli-hoodofevictionsandforcedrelocation,creatingaphenomenonknownassecondarydis-placement.Insuchcircumstances,newarrivalsruntheriskoflong-termdisplacementwhenintegrationinhostcitiesremainsdifficultandreturntoclimate-decimatedruralareasisnolongeraviableoption.Map3.1illustratesthedegreetowhichurbangrowthinthelastfourdecadeshastakentheformofexpandinginformalsettlements,onlysomeofwhichsubse-quentlyformalize.ThemapshowshowNiamey,thecapitalofNiger,hasexpandedsince1985.Mostnewneighborhoodsattheoutskirtsofthecitycontainasubstantialshareofinformalhousing.Climate-induceddisplacementtorapidlygrowingcitiescanseverelyunderminetheinclu-sivityofurbandevelopment,leavingsomeofthemostvulnerablemembersofsocietyattheoutskirts,bothfigurativelyandliterally.Evidencefromvarioussourcessuggeststhatwomenandchildrenareoverrepresentedinthepopulationofindividualsdisplacedtocitiesbyclimateshocks.Forexample,womenandchildrenmakeupnearlyathirdofalldisplacedindividualsinBurundi.Asalarmingasthisproportionmaybe,thesesortsofdemographicdisparitiesareevenmorepronouncedinhotconflictzones.IncountriessuchasAfghanistan,theDemocraticRepublicofCongo,andSomalia,womenandchildrenaccountfornearlytwo-thirdsofthosedisplaced(WorldBank2020a).ThedisproportionaterepresentationofwomenandchildrendisplacedtocitiesinconflictzonescompoundsthepressureplacedonservicedeliverybecausetheseIDPsoftenhavelessmobilityoncetheyarriveyetrequiregreateraccesstothehealthcareandeducationsectors.Peopledisplacedtourbanlocationsmaybelessabletorelyonthecopingmechanismsavailabletorural-to-ruralmigrants,suchassupportfromtheirextendedfamiliesorfromcustomaryinstitutions.Thus,theburdenofclimate-drivendisplacementtogrowingcitiesoftenfallshardestonvulnerablepopulations.184THRIVINGRapidurbangrowthinducedbyclimatechangecanstrainservicedelivery,butitcanalsofuelurbaneconomiesbyprovidingasteadyflowofnewlaborerstogrowingsectorssuchasmanufacturing,construction,hospitality,communication,andtransportation.Urbanenvi-ronmentsalsocouldbreakdowntraditionalgenderbarriersbybringingwomenintoprofes-sionspreviouslyreservedfortheirmalecounterparts.TheSomaliaUrbanizationReviewfindsthatnearlyhalfofSomalifirmsinthemanufacturing,retail,andservicesectorsinBosasoandMogadishutriedtohirenewemployeesintheprecedingtwoyears(WorldBank2021c).Forexample,45.6percentofmanufacturingfirmssurveyedinBosasoandMogadishureportednewhiresintheprecedingfiscalyear.Despitepersistentsecuritychallenges,thefirmsthatoperateinSomalicitiesexperiencedrapidgrowthinsales,evenwhencomparedwithregionalpeerssuchastheDemocraticRepublicofCongoandRwanda.Theopportunitiesenjoyedbyurban-specificeconomicsectorsnotwithstanding,notablebottleneckslimittheabilityofclimatemigrantstogetnewjobs.Foronething,jobseekersoftenlacktheabilitiesrequiredbyemployers,includinginterpersonal,problem-solving,computer,andmanagerialskills.Incitiesacrossthedevelopingworld,ethnicandsectariancleavagesalsooftencontinuetolimitjobseekers’abilitytolookforjobsforwhichtheyarequalified.Furthermore,governmentandeducationalinstitutionsdonotalwaysprovidesufficientjobassistancetolinkprospectiveemployeeswithsuitableemployers.Asforexpandingfirms,limitedfinancialintermediation,Map3.1UrbanexpansioninNiamey,Niger,byshareofinformalsettlements,1985–2021Source:Chlouba,Mukim,andZaveri2022.TheImpactsofClimateandEnvironmentalChangeonCities185thehighcostandunavailabilityofland,unfavorabletaxrates,andthepricevolatilityofimportedproductsrepresentsomeofthemostimportantbottlenecks(WorldBank2021c).Usingcountry-leveldataonclimate,conflict,emigration,andimmigrationin126countriesfor1960–2000,Bosetti,Cattaneo,andPeri(2021)lookintowhetherclimate-inducedmigrationcouldactasapossibledriverofsocialunrestintheareasofdestination.Theydonotfindanyevidencethatinflowsofclimatemigrantshadasignificanteffectonconflictsinthereceivingregions.Evenfollowingmigrationawayfromclimate-stressedregionstourbanareas,evidencesuggeststhatruralmigrantsmayhavereplacedonesetofriskswithanother.IntheircasestudyofruralmigrantsarrivingintheBholasluminBangladesh’scapital,Dhaka,McNamara,Olson,andRahman(2016)findthattheerstwhileruralmigrantswerenowdealingwithurbanfloods,outbreaksofcholera,firehazards,andlackofaccesstobasicservices,includingpotablewater.Althoughmostpeopledisplacedormigratingbecauseofclimateimpactsstayintheircoun-triesoforigin,theacceleratingtrendofglobaldisplacementrelatedtoclimateimpactscanincreasecross-bordermovements,particularlywhereclimatechangeinteractswithconflictandviolence(seebox3.4formoredetails).Waterstressesalsoaffectcities—evenfromafarCitieshavealwaysreliedonimportedwater(alsoseechapter4).FromancientRomeandLosAngelesintheearlytwentiethcenturytomodern-dayinitiativesinMexicoCityandKathmandu,Nepal,waterimportshaveofferedapathtourbanwatersecurity(Garricketal.2019;Hoekstra,Buurman,andVanGinkel2018).Butthescaleandintensityofneedhaveundergonerapidchange,withcitiesoftodayoftenrelyingondozensofwatersourceslocatedhundredsofkilometersoutsidetheirmunicipalboundaries(McDonaldetal.2014).Accordingtosomeestimates,theurbanwaterinfrastructureoflargecitiesnowmovesapproximately500billionlitersofwaterdailyacrossnearly27,000kilometers.Moredisruptiveiswhenthereachofurbanwaterinfrastructureextendsoverlargerareasandacrossbasins,suchaswhenittransportswaterfromruralregions.AsystematicreviewoftheglobalexperiencebyGarricketal.(2019)findsthatsuchtransfersserve69citiesthatin2015hadatotalpopulationofalmost400millionandthatthetransfersinvolvereallocatingapproximately16billioncubicmetersofwatereachyear.Itisthereforecriticaltoaccountforthislocalizednatureofwateravailabilitywhenstudyingtheimpactsofwaterstressonurbanareas(seebox3.5).Forexample,aclimaticshocksuchasadroughtfarawayfromacitycanhaveprofoundimpacts,whereasadroughtjustoutsideofthecity,orevenwithinthecityitself,maybebenign(Zaverietal.2021).Consistentwiththisexample,recentresearchhasfoundthatdroughtimpingingonwatersourcesinoftendistantlocationsaffectscitygrowthbecauseofthepressuressuchasituationexertsonacity’swatersupplysystem.Estimatessuggestthatdwindlingurbanwatersuppliesduetoprolongeddroughtinlocationsfromwhichacity’swatersupplyoriginatescanreduceacity’sgrossdomesticproductbyupto12percentagepoints(Zaverietal.2021).12Upstreamcontributingareascanalsoaffecttherawwaterqualityandquantityofsurfacewatersources(McDonaldetal.2014).Acrosslocalwatershedsandeventhousandsofkilo-metersaway,forestsalterthemovement,quality,andavailabilityofwaterbyregulatingflow,absorbingwaterwhenitisplentiful,andreleasingitwhenitisscarce(Damaniaetal.2017;Zaverietal.2021).Althoughlargecitiesoccupyonly1percentoftheEarth’slandsurface,theirsourcewatershedscover41percentofthatsurface.Asaresult,therawwaterqualityoflarge186THRIVINGDoesclimatechangeleadtoviolenceandmigration?Anemergingliteratureoverthelastfewyearshasattemptedtoexploretherelationshipamongclimatechange,conflict,andmigration.Thenexusofthesethreeforceshaslongbeenanecdotallyinvokedasastylizedfactseekingtoexplain,forexample,phenom-enasuchastheArabSpringandsubsequentwavesofbothconflictandmigration.Yetthescholarshiphasyieldeddivergentfindings(Machetal.2020);contrarytocommonbelief,theevidencelinkingclimaticriskstoconflictandforceddisplacement,especiallyinregionssuchastheMiddleEastandNorthAfrica,isnotunequivocal(Borgomeoetal.2021).Focusingontherelationshipbetweenclimateandviolencebroadlyconstrued,Burke,Hsiang,andMiguel(2015)attempttosummarizetheexistingfindingsrelatedtobothinterpersonalconflict,suchasdomesticviolence,andintergroupstrife,suchasriots,civilwars,andcoups.Theauthorssurveyed55econometricstudiesinameta-analysis,findingthat,foreverystandarddeviationriseintemperature,thefrequencyofinterpersonalconflictincreasedby2.4percent.Somestudiesintheirreviewalsosuggestedthatdevi-ationsineitherdirectionincreasedthelikelihoodofconflict,extremeheat,orextremecold.Extremerainfallseemstobeassociatedwiththerisingprobabilityofintergroupconflicts.HarariandLaFerrara(2018)suggestthatclimate-inducedviolenceisparticularlypro-nouncedinregionstraditionallydependentonrain-fedagriculture,suchasSub-SaharanAfrica.Asacaseinpoint,McGuirkandNunn(2021)arguethatdroughtcanfatallydisruptthearrangementbetweentranshumantpastoralists,whoneedlandtograzetheiranimals,andsedentaryagriculturalists,whoneedarablelandforcropfarmingduringthewetseason.Persistentdroughtforcespastoraliststomigratebeforetheharvest,leadingtoconflictwithsedentaryfarmers.Asmallerstrandofstudiespointstoclimate-inducedconflictasadriverofsubsequentmigration.Poorwatermanagementtogetherwithincreasingdroughtandmultiyearcropfailureshavebeenshowntoacceleratemassmigrationofruralhouseholdstourbanareas(Kelleyetal.2015).Abeletal.(2019)exploitdataonasylumapplicationsfrom157countriesbetween2006and2015toshowthatlocalclimate,throughitsimpactonbothdroughtseverityandarmedconflict,playedapivotalroleinexplainingthevolumeofasylumapplicationsbetween2011and2015.SeveraloftheWorldBank’sUrbanizationReviewshaveprovidedevidenceoftheclimate-conflict-migrationnexusatthesubnationallevelsinsomeoftheworld’spoorestcountries.EvidencefromtheBurundiUrbanizationReviewshowsthat,eventhoughmostclimatemigrantstendtoremainintheirhomeprovinces,theyflockdisproportionatelytolocalurbancenters(Mukim2021).Oncetheyarriveingrowingcities,internallydisplacedpopulationsdonotautomaticallyintegrateintolocaleconomies,therebycontributingtothegrowthofinformalsettlementsandputtingextrapressureonlocalservicedelivery.InSomalia,three-quartersofinternallydisplacedpersonsarethoughttoliveinurbanareas,mostofthemdisplacedbyacombinationofpersistentconflictanddifficultclimatecondi-tions(WorldBank2021c).Box3.4TheImpactsofClimateandEnvironmentalChangeonCities187citiesalsodependsonthelanduseinthismuchlargerarea(McDonaldetal.2014).For33oftheworld’s105largestcities,nearbyprotectedforestlandsareaprimaryfactorindrinkingwateravailabilityandquality(DudleyandStolton2003;Zaverietal.2021).13ClimatestressesandshocksinfluenceurbanfoodpricesHighfoodpricesnegativelyaffectthedietsandnutritionofurbanconsumers,particularlyinpoorercountriesandfortheurbanpoor,whospendalargershareoftheirincomesonfood.Foodexpendituresinlow-andlower-middle-incomecountriesrepresentabout40percentofhouseholdincome,whereasthoseinupper-middle-andhigh-incomecountriesare29and15percent,respectively(GrafvonLuckner,Holston,andReinhart2022).Urbanresidents,especiallythemoredisadvantaged,alreadyhavelesshealthydiets(seechapter1).Higherfoodpricesincreasepovertyandvulnerabilityandleadtopoorerdietsandnutrition,withpotentiallonger-termandlargelyirreversibleconsequencesforthepoor,particularlyformothersandchildren(Baezetal.2017;Greenetal.2013;Ivanic,Martin,andZaman2012).Inalreadyvulnerablecontexts,climatechange–relatedshocksandstressorsaffectruralfood-producingareas.Theseimpactsspillovertourbanareasviahigherurbanfoodprices(Venkat,Dizon,andMasters2022).Accordingtoanalysisthatcombinesfoodpricesandclimatechange–relatedshocksandstressorsforasubsetofvulnerablecountries,varioustypesofshocksinruralareasincreasefoodpricesincities(Baietal.2021;Florczyketal.2019;Pesaresietal.2019).14Theseimpactsvarybythetypeofshockorstressorandbythetypeoffood(suchasmorenutrient-denseperishablesversuscalorie-densenonperishables).HowwatersuppliesfromproximatenaturalsourcesarestrugglingtokeepupwiththeneedsofnearbycitiesandtownsThebackgroundpaperpreparedforthisreportbytheGautengCity-RegionObservatorycitesSouthAfricancitiesasgoodexamplesofmajoragglomerationsthathaveeffectivelyrunoutofwater(GCRO2022).SouthAfricaisawater-scarcecountryandreceivesonlyhalfoftheaverageglobalrainfall—495millimeterscomparedwiththeworldaverageof1,033millimeters.AmongothercitiesexperiencingwatersupplyextremeslikethoseinCapeTownandtheNelsonMandelaBayMetropolitanMunicipality,Johannesburgfacedacriticalwaterdeficitduringaseveredryspellin2015–16.Farfromanymajorwaterbodyandbuiltongold-richgrasslands,GreaterJohannesburgandits16millionresidentsfaceinevitablecrisesbecauseofthecity’sunusualgeographicpositionatthepinnacleofacontinen-talwatershed.ThecityfirstemergedinanareacalledWitwatersrand(“ridgeofwhitewaters”),suggestingaplentifulwatersupplyfromlocalspringsandstreamsflowingfromtheridge.Presently,however,allwaterflowsoutandawayfromthecoreoftheregiontowardtheIndianOceanononesideandtheAtlanticOceanontheother.Infact,muchofthewaterconsumedinthecityissuppliedbytheLesothoHighlandsWaterSchemeacrosstheLesotho–SouthAfricaborder.Withoutincreasingdamcapacityandinviewofthegrowingenvironmentaluncertaintiesarisingfromclimatechange,thiswatermaynotbesufficienttomeetthefutureneedsofresidentsandindustryintheregion.Box3.5188THRIVINGBettertransportationnetworkshelpmitigatetheimpactofshocksandstressors,allowingforpotentiallymoreresilientfoodsupplychains(Venkat,Dizon,andMasters2022).15Roadsandroadqualityreducepovertyandincreaseconsumption,reducepricevolatility,andhelphouseholdscopewithshocks(Derconetal.2009;Moctar,d’HôtelElodie,andTristan2015;Nakamura,Bundervoet,andNuru2019;ShivelyandThapa2016).Roadsareimportantforfoodsecurityintimesofdisaster.Economiclossesfromtransportationdisruptionsincreaselinearlywiththedurationofdisruptions,whichcallsforquickrepairsbutalsoflexibleprocurementstrategies(Colon,Hallegate,andRozenberg2021).Howdoesclimatechangeaffectdifferentpopulationsincities?Citiesarelargeecosystemscomposedofmanydifferentgroupsofpeople.Becauseincomeandeducationlevels,serviceprovision,andamenitiesoftenvarydramaticallywithinthebound-ariesofacity,thesameshouldbeexpectedofexposuretonaturalhazards.Ifdiversetypesofinequalitiesarecorrelated,vulnerabilitiesarecompoundedbecausethosewhofacehigherriskwillalsobemorevulnerableandlessabletoadoptmitigationstrategies.Moreover,byincreas-ingthelikelihoodandfrequencyofadverseevents,climatechangehasthepowertoexacerbateinequality(Hallegatteetal.2016).Itisimportanttounderstandtherelativeexposuredynamicsincitiesbecausethosedynamicsmaydifferconsiderablyfromcountry-levelones.RecentWorldBankstudieshaveusedglobalhazardmapstobuildageneralunderstandingofexposuretoclimatechangebycountry(Hallegatte,Bangalore,andVogt-Schilb2016a,2016b;Parketal.2018;RentschlerandSalhab2020;Winsemiusetal.2018).Findingsbasedonlargersubnationalunits,suchastheDemographicandHealthSurveys,donotalwaystranslateintowhathappensatthecitylevel.Forexample,inVietnamthepoorarenotsystematicallyoverexposedtofloods(NarlochandBangalore2016);however,Bangalore,Smith,andVeldkamp(2019)findthattheslumsinHoChiMinhCityarerelativelymoreexposedthantherestofthecity.Thepoorarenotuniversallymoreexposedthanthewell-offtoclimaterisksPreciseinformationonindividualorhouseholdincomeisoftennotavailableforcitiesinthedevelopingworld.Informationonthelocationofslums,however,canhelpidentifypocketsofdeprivationincities.Matchingdataonslumswithenvironmentalriskmapscanshedlightonwhichslumdwellersaremorehighlyexposedtostressors.Urbanslumsareoftenfoundinunsafelocations—thatis,onsteepslopes,onfloodableland,ornearopendrainsandsewers(Fay2005;Hallegatteetal.2016).Bangalore,Smith,andVeldkamp(2019)findthat64percentofHoChiMinhCityfacesayearlyfloodriskof1percent,butthatfigurerisesto80percentforslums.Elsewhere,slumsinMedellín,Colombia(RestrepoCadavid2011),aswellasinCaracas,RepúblicaBoliarianadeVenezuela,andRiodeJaneiro,Brazil(Fay2005;LallandDeichmann2012),areinperipheralcitylocationsonsteepslopesandclosetowaterbodies,whereasslumsinDhaka,Mumbai,andPunearemostlyexposedtofloodsandpollutedlocations(Hallegatteetal.2016).Althoughthisbodyofevidenceestablishestheexposureofpoorpeopleincities,itdoesnotfullyspeaktorelativeexposurebecausemostofthestudiesdidnotsetouttomakeanempiricallyrobustcomparisonwithformalresidentialareas.Accordingly,thecalcula-tionsbyPatankar(2015)offloodexposureforeachwardinMumbaidonotshowclearlythatinformalresidentialareasaremoreexposedthanformalones.TheImpactsofClimateandEnvironmentalChangeonCities189Thepoor,however,arenotuniversallymoreexposedthanthewell-offtoclimaterisks,despitebeingmoreexposedinsomecities(seebox3.6).InBogotá,Colombia,residentsinthebottom20percentoftheincomedistributionaremorelikelythanricherresidentstoliveinareasathigherriskofearthquake(LallandDeichmann2012).Similarly,Ermanetal.(2019),IntensifyingurbanheatinJohannesburgwilldisproportionatelyaffectthepoorJohannesburg,knownforahistoricallymildclimate,facesasharpincreaseintemperaturesovercomingdecadesthatwilldisproportionatelyaffectpoorerneighborhoodswhosephysicalformhasastrongertendencytoabsorbandstoreheat.TheHighveldregionhasalreadyseena1.2°Cincreaseinaveragetemperaturesoverpreindustriallevelsandwillwarmbyafurther1.2–1.7°Cby2050,shiftingfromatemperatetoahotanddryclimatezone.Detailedheatmaps,producedthroughacommunitymonitoringcampaign,showthatJohannesburgalreadyfacesastrongurbanheatislandeffect:mostneighborhoodsare3–4°Cwarmeratnightthannearbyruralareas;andneighborhoodslikeAlexandra,Katlehong,Soweto,andTembisa,aswellastheCentralBusinessDistrict,areupto6.5°Chotter(mapB3.6.1).Thehighestheatintensitiesoccurinhistoricallymarginalizedneighborhoods,whicharehometononwhiteandlow-incomecommunities,andwhichhavehighbuildingandpopulationdensityaswellasverylowlevelsofvegetationandtreecover.Theseneighborhoodsoftencontainmanylow-costdwellingsthattendtooverheat,andtheirinhabitantshavelimitedcapacitytoadapttoexcessiveheat,suchasthroughairconditioning.Historicalfactorsincludingapartheid-eralandusepracticeshavecontributedtosharpdisparitiesinheat,whichmayintensifyincomingdecades.Urbanclimatemodelingsuggeststhatby2050thenumberofhotnightsperyear(nightswhenthetemperatureremainsabove20°C)willincreasefrom10to40formuchofthecitybutfrom40to100forthehottestneighborhoods—primarilypoorerandmajorityblacktownshipareas.Theheatexposureresultsfromphysicalcharacteristicsoftheseneighborhoodsthatabsorbsolarradiationandreemititatnight,includingdensehousing,lackofgreenfeatures,anduseofheat-trappingmaterialssuchascorrugatedironroofs.Extremeheatwillaffectthehealthandlivelihoodsofpoorercommunities.Modelingsuggestsanincreaseintheheat-relatedexcessmortalityrateby2050,resultinginseveralhundredadditionaldeathsperyear.Healthrisksarehigherfortheelderly,peoplewithtuberculosisorHIV/AIDS,andpeoplewhosehomesabsorbandretainheat.IndoorheatmeasurementsconductedbyvolunteersinFebruary2022confirmedthatindoortemperaturesinwood-frame,corru-gatedirondwellingscanbe15°Chigherthaninnearbybrickandconcretehomes.Laborproductivityforoutdoorandinformalsectorworkerswillalsolikelydiminishasoutdoorheatstressreacheshighorveryhighcategoriesformorehoursperday.ThecitiesofEkurhuleniandJohannesburghavealreadyintegratedcoolingmeasuresintotheirclimateactionplans.Oneaspectaddressesthephenomenonof“greenapartheid,”wherebylow-incomeareashaveavegetationcoverrateof0–15percentcomparedwith30–60percentforwell-offareas.Forthe2010WorldCup,Johannesburgplantedsome200,000trees.Furtherbuildingontheseefforts,interventionsforcoolerplaces—suchastargetinggreeningandimplementingcoolbuildingdesignsforhomes,schools,andclinics—andinterventionstoprotectvulnerablepeople—suchasprovisionofheatwaveearlywarningalertsandcommunitycoolingcenters—offercost-effectivewaystoadapttorisingtemperatures.Box3.6190THRIVINGBox3.6continuedSource:Souverijnsetal.2022.Note:Mapshowsthepresent-dayurbanheatislandintensity(thatis,averagedaytimetemperatureascomparedwithnearbyruralareas)correctedfortopographical(surfaceheight)effects.Insets1and2showTembisaandAlexandra,respectively.MapB3.6.1Residentsofhotterneighborhoodsfacedaytimetemperaturesthatareupto6.5°Chigher,Johannesburg,SouthAfricausingresponsestoahouseholdsurveyontheApril2018floods,findthattheshareofdirectlyaffectedhouseholdsinDarEsSalaam,Tanzania,generallydecreasesasincomequartilegoesup(from29percentinthefirstquartileto15percentinthefourthquartile).Bycontrast,asurveyconductedinGreaterAccra,Ghana,onthe2015floodsfindsnodifferenceinexposurebetweenricherandpoorerhouseholds(Ermanetal.2020).Inabackgroundpaperforthisreport,Rossitti(2022)providesresultsofanovelanalysisreveal-ingthatslumsindevelopingworldcitiesare,onaverage,notmoreexposedtofloodsorexces-siveheatbutfaceahigherprobabilityofahigh-riskfloodevent.Rossittiemploysgloballandusedatafor18citiesintheSouthAsiaandSub-SaharanAfricaregions16toidentifypovertyclustersanddeterminetheirrelativeexposuretoclimatehazards,andderivesinformationonslums’locationsfromaerialimagery.17Theauthorfindsthat,ingeneral,theaverageresultsconceallargedifferencesamongcities.Regressionanalysissuggeststhat,in8ofthe18citiesTheImpactsofClimateandEnvironmentalChangeonCities191inthesample,slumsaremoreexposedtofloods,whereashalfshowanexposurebiasintermsofexcessiveheat.Adifferentsampleofcitiesmaythereforeyielddifferentaverageresults.18Rossitticoncludesthat,althoughinmanycitiesthepoorappeartobemoreexposedtoclimatehazards,thisfindingcannotbegeneralizedbecausethepresenceofexposurebiasoftendependsonindividualcitycharacteristics.LocationsortingmayofferoneexplanationLocationsorting19canexplainwhyinsomecitiesthepoorareoverexposedtoclimaterisks:theychoosetoliveinhazard-proneareasbecausethey,likehouseholdsinslumsaffectedbyrecurrentfloodsinMumbai(Patankar2015)andthoselivinginareaswithhigherearthquakeriskinBogotá(LallandDeichmann2012),prefertotradesafetyforbetteraccesstojobs,services,andeconomicactivity.Financialconstraintsmayalsoexplainsortingintohigh-riskareas.Becausepoorhouseholdsoftenhavelessleverageandfinditdifficulttoborrowmoneyinthefinancialmarketplace,theymayborrowagainsttheirlocationassetbychoosingtoliveinaworselocationwithlowhousingcostsbutrelativelypoorworkandeducationopportunities(BilalandRossi-Hansberg2021).Heblich,Trew,andZylberberg(2021)provideclearevidenceofthismechanismusingahistoricaltimeseriesofneighborhood-leveldataonemploymentandpollutioninBritishcities.Thecurrentincomedistributiondepends,inpart,onhowpoorerresidentssortedduringtheindustrialrevolutionintheeasternareasofthecitiesbecausethoseareashadmoreexposuretoindustrialcoalpollution.Adifferentbutnotmutuallyexclusivemechanismforlocationsortingisinformationasym-metries,whichalsomayplayaroleintheresidentiallocationchoicesofthepoorincities.Itispossiblethatpoorerhouseholdsarelessabletoacquireinformationontheenvironmen-talrisksofaspecificlocation,orthatmigrantsfromruralareasdonothaveaccesstosuchinformationbeforesettlingonalocation.20Itisalsopossiblethatrisk-averselow-incomehouseholdsthatwouldvaluesafetyoveraccessibilitystillsortinriskylocationsbecausetheylackinformationonenvironmentalhazards.Evenifprospectivedwellersdohaveinforma-tionontherisksassociatedwithdifferentlocations,thatinformationmaynotbeaccurate(BakkensenandBarrage2021;VotsisandPerrels2016).Becausepovertyinvolvesaseriesofdistractionsthatreduceproductivityaswellasthechancestoacquireinformation(BanerjeeandMullainathan2008),poorer,lesseducatedindividualsmayfaceahighercostofacquiringhazard-pronenessinformationthandotheirrichercounterparts.21HouseholdsurveydatafromBamakoalsosuggestthatthepoorfaceatrade-offbetweenaccessibilityandenvironmentalrisk.Rossitti(2022)usesinformationonmigrationstatusfromMali’s2018householdincomeandexpendituresurveyinBamako22toprovideevidencesupportingthelocation-sortingmechanisms.Althoughpoorhouseholdsdonotappeartoberelativelymoreexposedtoeitherfloodsorexcessiveheat,pooranduneducatedmigrantsface1.24moredaysofexcessiveheatthandopooranduneducatednonmigrants.Atthesametime,nonmigrantsandhighlyeducatedmigrantsappeartohavenodifferenceinexposure.Thesefindingscouldbeexplainedbyacombinationofsortingandinformationmechanisms.Moreover,thedatahighlightthetrade-offbetweenbeinginacentrallocationandtheenvi-ronmentalriskfacedbythemostdeprivedhouseholds,asshowninfigure3.1.Low-incomeorpoorlyeducatedhouseholdsthatliveclosetothecitycenterfaceahigherriskthanthoselivingfartheraway.Forthewealthierormoreeducated,bycontrast,distanceisaweakerpre-dictorofexposure.192THRIVINGThepoorarethehardest-hitbydisastersThepooroftenlosemorewhenhitbyadisaster.Althoughthepresenceofexposurebiasdependsonthespecificcityconsidered,moreevidencesupportsthepresenceofvulnerabilitybiasincities—thatis,whenhitbyadisaster,poorurbandwellerslosemorethantheirrichercounterparts.Inabsoluteterms,therichoftenlosemorebecausetheirassetsareworthmore;however,inrelativeterms,theoppositeholdstrue(Hallegatte,Bangalore,andVogt-Schilb2016b).Patankar(2015)reportsthat,duringtheMumbaifloodsin2005,householdsbelowthepovertylinesufferedlossesequaltosixtimestheirmonthlyincome,higherinrelativetermsthanthelossesbyallotherincomegroups.23InHoChiMinhCity,Vietnam,disastershaverelativelylargernegativeimpactsonthepoorintermsofhealth,employment,andincome(Hallegatte,Bangalore,andVogt-Schilb2016b).24Ermanetal.(2020)findsimilarevidenceinAccra,Ghana.Poorerhouseholdsalsosufferdispro-portionallyfromtheindirecteffectsofnaturalhazardswheninfrastructuredisruptions,forexample,cuthouseholdsofffromroads,theirwatersupply(Ermanetal.2019),orpublictransit(He,Liu,etal.2021).Notonlydopoorurbandwellerssufferhigherlossesfromenvironmentalhazards,buttheyalsohaveimpairedabilitytodealwiththosehazards.Ingeneral,poorerhouseholdshavelessaccesstofinancialmarkets(Ermanetal.2019)andinsurancemarkets(whicharealsolessdevelopedFigure3.1Marginaleffectsofremotelocationwithinacityonexposuretoexcessiveheat,byeducationlevelandconsumptionpercapita,Bamako,MaliSource:Rossitti2022.Note:ThegraphshowsthemarginaleffectsofthedistancefromthenotionalcitycenterofBamako(proxiedbylocationofthetownhall)ontheaverageyearlynumberofdaysofexcessheatmeasuredforeachhousehold(seebox3.7onmeasuringenvironmentalrisk).Theregressionincludesaveragehouseholdeducationlevels(ofmembersolderthan15)andlogconsumptionpercapita(bothvariablesnormalizedona0–1scale);thedistancefromthetownhallinkilometers;andtherespectiveinteractions,migrationstatus,andaltitudeatthelocationwherethehouseholdlives.Averageeducation(nonstudents,>15years)Logtotalconsumptionpercapita–1.0–0.8–0.6–0.4–0.200.20.40.60.81.0MarginaleectofdistancefromtownhallNormalizededucationandconsumption(0–1scale)TheImpactsofClimateandEnvironmentalChangeonCities193inlower-incomecountries).Moreover,withoutsubsidiesandtenantprotectionlaws,poorerhouseholdsarelesslikelytobenefitfrompropertyowners’investmentsinmitigationstruc-turesbecausesuchinvestmentswilllikelyleadtohigherpropertyprices(Nakagawa,Saito,andYamaga2007),whichwill,inturn,forcepoorhouseholdsoutofnowsaferareasofthecity.Finally,manypoorhouseholdsengageinrecurrentself-financedshort-termmeasures(suchastemporarystructuralimprovements)thatimposealargefinancialburdenonthepooresturbandwellerswiththelowest-qualityhousing(Patankar2015).Beyonditsphysicalandfinancialimpacts,climatechangecanalsoadverselyaffectmentalhealth(Evans2019).Becausethebruntofclimatechangeshocks,suchasflooding,dispro-portionatelyaffectsthepoor(Hallegatteetal.2016),aclimateshockmayseverelylowertheirincomeandtheirabilitytoearnalivingwage,likelyincreasingtheincidenceofstress.Areviewofstudieslinkingheatandmentalhealthfindsahighersuicideraterelatedtoheatin15ofthe35studiesreviewed(Thompsonetal.2018).Studieshavealsofoundalinkbetweenflooding,especiallyrepeatedflooding,andmentalhealthdisorderssuchaspost-traumaticstressdisorder(Evans2019).EvidencesuggeststhatsomesocialgroupsareaffectedmorethanothersAlthoughthissectionhasconcentratedontheexposurebiasofthepoor,environmentalrisksincitiesmayalsodisproportionatelyaffectvulnerablepopulationssuchaswomenandthelesseducated.Ermanetal.(2019)findthatfloodscompoundwomen’schallengesbecause,accordingtogendernorms,womenareresponsibleforcleaningupintheaftermathandfortakingcareofchildrenduringschoolclosures.FollowingthefloodsinDarEsSalaam,Tanzania,in2018,ofallthosewhohadtostayhomeandmisswork,themajority(60percent)werewomen.TheWorldBankreportHealthyCities(forthcominga)outlinesthewaysinwhichclimatestressesinurbanareasaffecthealthdirectlyandindirectly.Stormsandfloodingcauseinjuries,exposuretotoxins,anddiseaseoutbreaks.Theincreasedfrequencyandintensityofextremeheatcancauseheatstroke,worsenchronicconditions,and,asnotedearlier,affectmentalhealth.Climatechange–causedshiftsinwildlifehabitatshavealsoledtogreateropportunitiesfortheemergenceofzoonoticdiseasessuchasCOVID-19.Meanwhile,droughtcanhavelonger-termimpactsonnutritionandhealth,andwildfirescanaffectlong-termheartandlungissues.TheHealthyCitiesreportalsolooksathowsomeofthesehealthimpactsdisproportionatelyaffectmarginalizedpopulationgroups.Forexample,urbanheatislandeffectsarefarworseformorevulnerablegroupssuchasmigrants,racialminori-ties,children,theelderly,thosewithchronicillnesses,andthosewhoselivelihoodsrequirethemtoworkoutdoors(Madriganoetal.2015;Mashhoodi2021;Shahmohamadietal.2011;Voelkeletal.2018).Rossitti(2022)suggestsastrongrelationshipbetweenenvironmentalhazardsandeducationallevelinthelargestIndonesianmetropolitanareas(map3.2).Hebaseshisfindingonalargedatasetof5,821village(desa)-levelobservationsfromtheeightlargestmulti-metroagglom-erationsinIndonesiaonaverageeducationalattainment.25Intermsofenvironmentalhazards,hisanalysisconsidersfluvialandpluvialflood,andexcessiveheat,measuredasdescribedinbox3.7.Regressionanalysisconfirmsthat,withincities,moredeprivedareas(wheredepriva-tionisproxiedbyalowereducationlevel)aremoreexposed.26194THRIVINGMap3.2EducationalattainmentandenvironmentalriskwithinIndonesianmultidistrictmetrosContinuedMap3.2continuedTheImpactsofClimateandEnvironmentalChangeonCities195Source:Rossitti2022.Note:Foreachmultidistrictmetro,mapsshowfromlefttorighttheshareofindividualswithatertiaryeducationdegree(adarkercolorequalsamorehighlyeducatedpopulation),fluvial(riverine)floodrisk,pluvial(surfacewater)floodrisk,andthetotalnumberofdayseachyearwithexcessiveheat(adarkercolorindicateshigherexposuretotheenvironmentalhazardconsidered).Allvariablesareaveragedatthevillage(desa)level.Seebox3.7foradetaileddescriptionoftheenvironmentalriskvariables.MetroareasaredefinedasinRoberts,GilSander,andTiwari(2019).196THRIVINGMeasuringenvironmentalriskincitiesManyoftheclimatehazardanddisasterdatathathaveinformedunderstandingofthespatialvariationofenvironmentalriskcannotbeusedtostudyrelativeexposurewithincities.Eventdataaremostlycataloguedforlargeareas(cities,regions),someasuringrelativeexposurewithincitiesrequiressmaller-scaledatathatreflectwithin-cityvariation.Forthisreason,Rossitti(2022)focusesonfloodsandheat—forwhichhigh-resolutiongriddeddataareavailable—tostudyexposurewithincities.FloodriskdataareobtainedfromthenationalgridsofFathomFloodHazardDataandMaps(Sampsonetal.2015).Fathomdatareportbothfluvial(riverine)andpluvial(surfacewater)maximumflooddepthat3-arcsecond(~90-meter)resolutionforseveralreturnperiods.aAreturnperiodindicatestheprobabilityassociatedwiththeevent.Forexample,aFathomrecordof0.4witha1-in-100-yearsreturnindicatesayearly1percentprobabilityofa40-centimeterfloodevent.Forconsistency,theanalysisusesthe1-in-100-yearsreturn(1percentprobabilityperyear)becauseitistheonlyavailablereturnperiodforsomeofthecitiesinthegloballandusesample.bForagivenreturnperiod,itisalsopossibletocategorizefloodeventsbylevelofrisk.AlthoughFATHOMclassifiesveryminorevents,theseeventsmaynotcarryanyenvironmentalrisk.RentschlerandSalhab(2020)classifyeventswithflooddepthunder15centimetersasminorandthoseover0.5meterashighrisk.Afloodwithadepthfrom0.5meterto1.5meterscarriesaveryhighriskandislikelytocreateextensivedamage.Followingtheircategorization,partoftheanalysisfocusesonlyontheexistenceofhighrisk,testingthehypothesisthat,althoughthepoorarenotnecessarilymoreexposed,onaverage,withincities,theymaybemorelikelytofaceahigh-riskevent.Exposuretotemperature-relatedhazardsismeasuredasthenumberofdaysofextremeheateachyear(usinga2000–16averagetocontrolforpotentialoutlieryears).ThedailytemperatureisobtainedfromaCHELSAglobalhigh-resolution(250metersto1kilometer)landsurfacetemperaturegrid(Kargeretal.2021).Adayofextremeheatisadayinwhichthetemperatureexceeds35°C.Sources:Kargeretal.2021;Sampsonetal.2015.a.Fluvialflooddataarefurtherbrokendownintodefendedandundefended(datathatdonotaccountformodeledriverbarriers).Forconsistency,theanalysisusesundefendedfluvialfloodriskdatabecausedefendeddataareunavailableforsomecountries.b.ThelastiterationofthenationalgridsforBangladeshandIndia(version1,May2016)doesnotincludedataforotherreturnperiodsthan1in100yearsandfluvialdefendedflooddepth.Alltheothercitiesintheanalysisuseversion2data(June2020).Box3.7TheImpactsofClimateandEnvironmentalChangeonCities197ClimateshocksaffectdifferenttypesofworkersandindifferentwaysAwarmingworldcouldaffectlivelihoodsandsocioeconomicoutcomesthroughitsimpactsonlaborproductivity(seebox3.8describingnewresearchcommissionedforthisreport).Climatestressorscoulddirectlyaffectworkerproductivityby,forexample,circumscribingthetimeavailableforworkor,conversely,throughthehealthimpactslinkedtolaboringinhightem-peratures(Shayegh,Manoussi,andDasgupta2021).Extremetemperaturesreduceperformanceduringworkinghoursbecauseworkersslowdown(Parsons2014),takemorebreakstorehy-drateandcooloff,andmakemoremistakes(Dasguptaetal.2019,2021;Parketal.2020).Thiseffectishigherinjobsexposedmoredirectlytoweatherconditions(Orlovetal.2020),suchasagricultureandconstruction.Theleastperceptiblebutmostwidespreadeffectsarise,however,fromslow-onseteventsthatmaynotbeenoughtokeeppeoplefromworkbutthatcouldaffecttheirperformance(Kahnetal.2021).TheeconomicandsocietalconsequencesofurbanheatarepervasiveAbackgroundpaperpreparedforthisreportbyJiangandQuintero(2022)evaluatestheeffectsofhightemperaturesonlaborproductivityacrossthousandsofcitiesineightcountriesinLatinAmericaandtheCaribbean.JiangandQuinterousesuccessiveroundsofhouseholdsurveymicrodatathat,apartfromBrazil,comefromtheSocio-EconomicDatabaseforLatinAmericaandtheCaribbean(SEDLAC).aBecausecomparabilitypresentsanimportantobstacleinassessinglaborproductivityacrossalargesetofcoun-tries,oneofthegreatadvantagesofSEDLACisthatitprovidesharmonizedmicrodatacollectedfromover300householdsurveys.Itensuresstrongcomparabilityofthedataacrosstimeandcountriesbyusingsimilardefinitionsofvariablesineachcountryandyear,andbyapplyingconsistentmethodsofprocessingthedata(CEDLASandWorldBank2014).ThepaperbyJiangandQuintero(2022)buildsonQuinteroandRoberts(2022)tocleanthedataandbuildwagevariables.Ineachcountry,thelocationanalyzedchangesslightlyaccordingtothelowestpossibleadministrativedivisionthatcanbeidentifiedinthesurveys.Thepaperalsoincludescitycharacteristicsfromageospatialdatabasecon-structedbytheUniversityofSouthampton’sGeoDatacenter.bThenumberofhotdayscintheyearandaveragetemperaturefromMODISTerraLandSurfaceTemperatureareproxiesforclimatestressors.Theanalysisfocusesonabroadsampleofworkerscoveringboththeformalandinformalsectorsifwagesarereported.Aworker’swageisassumedtobethenominalhourlywageearnedintheprimaryoccupation.JiangandQuintero(2022)findthatbothhotdaysandhightemperaturesreducethelaborproductivityofworkers,asmeasuredbywages.Theresultsarestatisticallysignif-icantformostcountries.Estimatedcoefficientsimply,forexample,that,ifthenumberofhotdaysinatemperatecitylikeMedellínincreasesby1percentayear,averagewageswouldfallby0.8percent.Onaverage,anannualtemperatureincreaseof1°Cisasso-ciatedwitha1percentdeclineinwages.TheeffectofhotdaysisparticularlystrongincountrieswithwarmerclimatessuchasColombia,Mexico,Panama,andPeru.Bycontrast,resultsareweakerandthepointestimatesmuchsmallerincountrieswithmoredistinctseasons,whichcouldindicatebetteradaptation.Itcouldalsobeexplainedbythedistributionofsectoralactivityincountries.Box3.8198THRIVINGClimatestressorscanfurtheraffectlaborproductivityindirectlythroughtheireffectoncapitalinvestment(Somanathanetal.2021),populationloss,macroeconomicconditions(Cashin,Mohaddes,andRaissi2017),andtheproductivityofotherproductionfactors(LettaandTol2019).Becauseadaptationtakestime,deviationsfromhistoricaltemperaturescouldbemoreharmfulthanhighertemperaturesperse(Kahnetal.2021).Evidencealsoexiststhattheseeffectswillhavenegativedistributionalimpacts.Foronething,theeffectsarelikelytobeconcentratedinlow-andmiddle-incomecountries.Lower-incomecountriesoftendependmoreonagricultureandsectorsfocusedonoutdoorwork.Moreimportant,theyarelessabletoimplementadaptationmeasures.Adaptationmeasuressuchasair-conditioningintheworkplacecoulddecouplehightemperaturesfromworkproductivity,butsuchmeasurescanberelativelyexpensive.Moreover,sharperreductionsintheincomesofinformalandruralworkers,whoalreadyhavelowerlevelsofincome,willexacerbateinequalitywithincountries.Tounderstandtheimpactsacrossdifferentgroupsofworkers,JiangandQuinteropooldataacrosscountriesandstudyworkersbydemographiccharacteristicsincluding,age,gender,education,andsector(privateorpublic,formalorinformal).dItturnsoutthattheimpactofclimatestressorsonproductivityarenothomogeneousacrossthepop-ulation.Youngerworkersaremoreaffectedthanolderworkers,probablybecauseofdisparitiesinworkingconditionsandsectorofoccupation.Similarly,informalworkers,andthosewithlowereducationalattainment,arealsomorestronglyaffectedbyhighertemperatures.Theseresultsareconsistentwiththelowerimplementationofadapta-tionmeasuresininformalsectors.Thoseintheprivatesectorarealsomoreaffected.Thus,JiangandQuinterofindthatclimatestressesnegativelyaffectlaborproductivityincitiesacrosslargegroupsofcountries.Furthermore,theeffectsareheterogeneous,withmorevulnerableworkersworseaffectedandwithfurtherevidenceofexacerbatinginequalities.a.ThisdatabasewasjointlyconstructedbytheCenterforDistributive,LaborandSocialStudies(CEDLAS)attheUniversidadNationaldeLaPlataandtheWorldBank.b.ThisdatabasewasconstructedforthisreporttoaligntheresearchwiththeidentifiersforsubnationalareasinSEDLAC.c.Hotdaysareconstructedastheaveragenumberofhotdaysperyearmeasuredduringtheday,weightedbythepixelpopulationusingLandScan2012population.Adayisconsideredhotiftheaveragedailytemperatureexceeds35°C.Thisthresholdforhotdaysissupportedbytheliteratureonthetemperatureabovewhichworkabilityisaffected(Andrewsetal.2018).d.SEDLACprovidestwoindicatorsforwhetheraworkerisconsideredinformal(CEDLASandWorldBank2014).Thefirst,basedonaproductivedefinitionofinformality,identifiesaworkerasinformaliftheworkerbelongstoanyofthefollowingcategories:unskilledself-employed,salariedworkerinasmallprivatefirm,orzero-incomeworker.Thesecond,basedonalegalisticorsocialprotectionnotionofinformality,identifiesasalariedworkerasinformalifthatworkerdoesnothavetherighttoapensionlinkedtoemploymentwhenretired.JiangandQuinterousethefirstindicatorbecausesurveysmorefrequentlyprovidethisinformation.Thesamplealreadyexcludesself-employedworkers,anddoingsoequatesinformalemploymentwithemploymentbyverysmallprivatefirms(fiveorfeweremployees).Box3.8continuedTheImpactsofClimateandEnvironmentalChangeonCities199WaterscarcityhasadisproportionateimpactonvulnerablegroupsGlobally,althoughtheprovisionofwaterandsanitationserviceshasgreatlyimproved,urbanareasstillfaceseveralchallenges.Thesechallengesariseinlargepartbecauseurbanpopula-tiongrowthhascontinuedtooutpacetheabilitytoadequatelymeetthewaterandsanitationneedsofgrowingperi-urbanandmarginalizedareas.27Andmarginalizedgroupsoftencarrythedisproportionateburdenoftheimpactsofwaterscarcity.Eventhoughpipedutilitywateristheleastexpensiveoptionformosthouseholds,manystilllackaccess(Mitlinetal.2019).Poorerhouseholds,oftenlocatedinareasnotservedbyutilities,aremostaffectedbecausetheymustbuypoor-qualitywaterfromwatervendorsatpricesmuchhigherthanthosepaidbyusersconnectedtopipedwatersupplies(Borgomeoetal.2021).AcrossAfrica,waterdeficitsinperi-urban,slum,andinformalurbanareasarewelldocumented(Keener,Luengo,andBanerjee2010).Theseareasmustpayahighercostforwaterscarcitybecausetheyaremorelikelytorelyonoff-sitestandpostwaterandunregulatedmarketsofinformalwaterresellersoralternativeserviceproviderswhochargehigherprices.Forexample,vendor-deliveredwaterinNairobicosts10timesmorethanhouseholdpipedwater—andnearly30timesasmuchinDaresSalaam(Azunreetal.2022).InurbanEthiopia,whichhassubstantiallyexpandedcoverageofpipedwater,richerhouseholdsarealmostfourtimesmorelikelythanpooreronestohavepipedwater.InsmallerEthiopiantowns,weakerinfrastructuremakesitdifficulttoconnecthouseholds;inlargercities,poorerhouseholdssimplylackthemeanstopayforservices(Das2020;WorldBank2017).InIndia,only38percentofhouseholdsamongthepoorestfifthofthecountry’surbanpopulationhaveaccesstoindoorpipedwater,comparedwith62percentoftherichestfifth(Frumkinetal.2020).InformalareasinLima,Peru,alsoexemplifytheimpactofurbangrowthandinequality.Manyofthe1.5millionwhoareunderserved(Howson2015)liveingrowingperi-urbansettlements,relyoninformalwatertankers,andmustpayuptoUS$6percubicmeter,comparedwithUS$0.40percubicmeterfromtheformalwatersystem(Ritter2018).Becausewaterfrominformalsourcesisoftenuntreated,italsoincreasesthecostofhouseholdtreatmentandposeshealthrisks.Meanwhile,indigenouspeoplemigratingtourbanareasoftenfacesevereshortfallsofaccesstobasicwaterservices,alongwithmarginalization(UN2018).InLatinAmerica,indigenousmigrantsinurbanareasaretwiceaslikelyasnonindigenousmigrantstoenduplivingininformalsettlementswithlimitedaccesstobasicservices,includingwater(WorldBank2015;Zaverietal.2021).ClimatechangehasexacerbatedtheseimpactsCitiesaffectedbydisasterorfragilityhaveadditionalchallenges.Theyaremorelikelytorelyontheresaleofinformalwater,therebycompoundingtheirvulnerability.Haiti,pronetobothdisasterandconflict,sawadropintheprovisionanduptakeofsafedrinkingwaterbetween1990and2015.Thedeclinewasparticularlynoticeableinurbanareas,where,amongotherreasons,residentsdidnottrustthequalityofpubliclyprovidedwater(Das2020;WorldBank2017).Waterscarcityandclimate-relateddroughtfurtherexacerbatethesestructuralgapsandarefeltmostkeenlyinunderservedinformalareas.Forexample,inUgandaresidentsaremorelikelytopayuserfeesduringadroughtandspend13percentmoretimefetchingwater—anincreaseof1.9hoursaweek—thaninanondroughtyear(Kamei2020).Womenandgirlsdisproportion-atelybearthisburden,andsohavelesstimetodevotetoeducationandproductiveendeavors.200THRIVINGAlthoughwomenincitiesspendlesstime,onaverage,collectingwaterthandowomeninruralareas,inmanyplacestheystillhaveresponsibilityforthattask.IntheurbanareasofAfrica,forexample,thedifferentialbetweenmenandwomenwhospendaminimumof30minutesadaycollectingwaterislargeinmanycountries.In18outof24Africancountriesstudied,womenhaveresponsibilityformostofthewatercollectioninurbanareas;and,inmostcountrieswherewomendonothavetheprimaryresponsibility,itfallsonchildreninamajorityofhouseholds(Graham,Hirai,andKim2016).Climatechangehasnegativeknock-oneffectsonalreadyvulnerablepopulationsTheneedtospendmoretimefetchingwatercanleadtomoregender-basedviolenceandlesstimetoengageinotheractivitiessuchaseducationorincome-generatingactivities.OnestudycalculatesthatinKhayelitsha,anurbantownshipofCapeTown,SouthAfrica,doublingthenumberoftoiletstoreducethedistancefromthehouseholdcoulddecreasetheincidenceofsexualassaultby30percent(Gonzalves,Kaplan,andPaltiel2015).Otherstudiessuggestthatreducingthetimeneededtocollectwaterfreesuptimeforleisureandchild-rearingactivi-tiesandreducesstresslevelsandintrahouseholdconflict(Devotoetal.2012).Wherewomenhaveahigherlikelihoodoffetchingwater,children’sparticipationinschoolisaffectedbecausetheyhelpwithdomesticchores.Thus,improvementsinhouseholdaccesstopipedwatercanalsoleadtobetterschoolingoutcomesforchildren,althoughtheeffectdependsonthecountry-specificcontext(Das2017;KoolwalandVandeWalle2013).Forexample,astudyfindsthathalvingthetimeittakestocarrywaterinGhanawouldincreaseenrollmentratesbyabout7percentagepointsforgirls,withsimilareffectsforboys(NaugesandStrand2017).IntheSundarbansinBangladeshandWestBengal,India,salinewaterandlackofaccesstoimprovedwatersourcesincreasethechancesofgirlsdroppingoutofschoolandbecomingresponsibleforwatercollection.Forboys,however,thereislittledifferenceamongthehouse-holdswithdifferentwaterqualityandwatersources(Das2017;KomatsuandJoseph2016).ReducingthedistancetoaccesssafelymanagedwaterbecameevenmoreimportantduringtheCOVID-19pandemicbecausetheneedforbetterhygieneincreasedtheburdenthatwomenandchildrenfacedinpoorerregions.Thisburdenand,conversely,thepotentialbenefitsofreducingthegaphaveunfoldedamongexistinginequalities,sothatgroupsalreadyfacingadeficitinaccesstocleanandsafedrinkingwaterexperiencedthemostacuteimpacts.Surveysoffemalerefugeesanddisplacedpersonsin15Africancountriesduringthepandemichigh-lightedharassmentandsexualviolenceencounteredonthewaytoandatwatercollectionpoints(AbwolaandMichelis2020).Althoughtheybearahigherburdeninsecuringwaterfortheirhouseholds,womenandothermarginalizedgroupsaresubstantiallyunderrepresentedintheworkforceanddecision-makingrolesinthewatersector(Das2014;WorldBank2019).28Asaresult,theirneedsandopinionsarenotconsideredwhenplanningfortheprovisionofwaterservicestowater-scarceinformalurbanareas,inresilienceplanning,orindevisingrecoverymeasures.A2013surveyof65countriesrevealsthatonly15percentofcountrieshadagenderpolicyintheirwaterministry,andonly35percentofcountriesincludedgenderconsiderationsintheirwaterpoliciesandprograms.Furthermore,only22percentofsurveyedwaterministrieshadgenderfocalpoints(Fauconnieretal.2018),andonly16percentofnationalwaterplansmentionwomenaskeystakehold-ersorprimaryparticipantsinclimateadaptation(UNESCO2015).Urbanwaterandsanita-tionutilitiesandwaterresourcesmanagementinstitutionsarealsostaffedpredominantlybymen(WorldBank2019).Asurveyofabout64waterutilitiesin28low-andmiddle-incomeTheImpactsofClimateandEnvironmentalChangeonCities201countriesfindsthatwomenmakeuplessthan20percentofemployeesandaresignifi-cantlyunderrepresentedintechnicalandmanagerialroles(WorldBank2019).29Oneinthreesurveyedutilitiesdidnothaveasinglefemaleengineer,andoneinsixhadnofemalemanagers(WorldBank2019).Suchegregiousdisparitiesreflectabroadertrendoffemaleunderrepresen-tationinthelaborforce(Jayachandran2021;Klasen2019).Thislackofvoicepresentsnotonlyaproblemofequitybutalsoabarriertodevisingsustain-ablesolutionstowaterscarcity.Becausewomenarethekeyclientsofwaterandsanitationutilities,theirpresenceinamoregender-diverseworkforcecanhelputilitiesbetterunder-standandrespondtotheconcernsandneedsoftheirfemaleclients.Womenalsooftenteachchildrenaboutwateruseandwaterconservation,andcountry-levelstudieshavepointedtothehighervaluewomenplaceonprotectingwaterquality;watermanagementissuesareapolicypriorityraisedbywomen(ChattopadhyayandDuflo2004;Chaturvedi,Das,andMahajan2021).ResearchfromIndiahasshownthat,whenwomenoccupyleadershipposi-tions,theytendtogiveprioritytoissuesvaluedmorehighlybywomenrelativetomen,suchasaccesstotoilets,cleandrinkingwater,andwatercontrolandharvesting.Women,therefore,remainimportantchangeagentsincommunitieswhenitcomestoimprovingwaterresourcemanagement.Thegender-differentiatedcausesandimpactsofdisplacementTheInternalDisplacementMonitoringCenterrecorded18.8millionnewdisplacementsasso-ciatedwithdisastersin2017;asoftheendofthatyear,nearly40millionpeopleinmorethan50countrieswerelivingininternaldisplacementbecauseofconflictorviolence(IDMC2018b).In2020,thetotalglobaleconomiclossduetointernaldisplacementwasUS$20.5billion(IDMC2021).Becauseofthegrowingfrequencyandintensityofnaturalhazards,thesocialandeconomicimpactsofdisaster-drivendisplacementareexpectedtocontinuerisingglobally.Inadditiontoitseconomiccosts,displacementmayhavemanynegativesocioeconomiceffects,suchasseparatingthedisplacedculturallyandsociallyfromtheiroriginalcommunities,lesssecurity,worsesanitaryconditionswithdegradationofhousingquality,interruptededucationformanychildren,andlessaccesstohealthcare(IDMC2018a).Moreover,postdisasterdisplacementisagenderedprocess.Multiplesocio-culturalfactorssuchasincome,education,health,andaccesstonaturalresourcesaffectwomen’sadaptivecapacities,whichinturnaffecttheirlikelihoodofdisplacement(Chindarkar2012).StudiesinBangladeshfindthatculturalnormscouldpreventwomenfromleavinghomeduringemergencies(Nelsonetal.2002),whereasinhigh-incomecountriessuchastheUnitedStateswomenevacuatetofulfillfamilyobligationsandcaregivingduties(BatemanandEdwards2002).Inlow-incomecountrieslikeNepal,womentendtohavealowcapacitytoadapttodisplacement,andtheirlownutritionalstatuscouldexacerbatepostdisasterrecovery(Cannon2002).Women’sloweducationlevelscouldpushthemintolabor-intensiveandlow-payingjobs(Kakissis2010),andwomenaremorelikelytobepermanentlydisplacedbecausetheylackhousingownership(Willinger2008)andemploymentopportunities.Moreover,womenfacegreaterrisksofgender-basedviolenceanddistressbecauseofsocialdisintegration(Mitchell,Tanner,andLussier2007).Toachievemoreinclusivedisasterriskmanagementpolicies,governmentsshouldtakeintoaccountgenderdifferencesintheeffectsofhazards.Thestudiesjustnotedpointoutgender-differentiatedimpactsinvariousdisastercontexts.Becauseoflimitationsinconven-tionaldatasuchassurveys,however,thesestudieslackregional-scaleandtemporallygranularanalysis.Somequestionsremainunanswered:Whatfactorsdeterminegenderdifferencesin202THRIVINGinitialandlong-termdisplacementratesandtraveldistances?Domenandwomenchoosemigrationdestinationsdifferently,and,ifso,how?Novelmobilitydataprovidevaluableoppor-tunitiestoshedlightonsuchquestions.MobilitydatafordisasterdisplacementanalysisEffortstoquantitativelyunderstandthemovementoflargepopulationswithinandacrosscitiesbefore,during,andafterdisastereventsuselarge-scaleGPSlocationdatasets(mobilitydata)collectedfromsmartphonesandmobilephones.Thesesourcespresentanadvantagebecausetheyprovidereal-timedataanddosomorecost-efficientlythansurveys.Forpolicyinterventions,mobilitydatacollectedwithhighspatialandtemporalgranular-ityofferbettermonitoring,forecasting,andunderstandingofmasspopulationmovementsacrossalongertimehorizon.Withcarefulhandlingofthedataandadequatesafeguardstoprotectusers’privacy,mobilitydataholdimmensepotentialforhelpingdisasterreliefagenciesandpolicymakersbuildinclusive,sustainable,andresilientcities(WorldBank2021f).Intheaftermathofadisaster,suchanalysesofmobilitydatacouldprovidesupportforrapidlydirectingresourcestothehardest-hitareas.Intheirseminalpaper,Lu,Bengtsson,andHolme(2012)usedcalldetailrecorddatacollectedfrommobilephonestostudythepredictabilityofdisplacementmobilitypatternsafterthe2010Haitiearthquake.Basedondatacollectedfrom1.9millionmobilephoneusersovertheperiodfrom42daysbeforetheshockto341daysaftertheshock,thestudyestimatedthat23percentofthepopulationinPort-au-Princehadbeendisplacedbytheearthquake.Despitethesubstantialdisplacement,theyalsofoundthatthedestinationsofthedisplacedwerehighlycorrelatedwiththeirpre-earthquakemobilitypatterns.Thisfindingshowsthepossibilityofpredictingpostdisastermobilitypatternsandhassignificantimplicationsforreliefoperations,includingthepredisasterpositioningofdis-tributioncentersandevacuationshelters.Anotherdisaster,theGorkhaearthquake—whichmeasured7.8onthemomentmagnitudescalewhenitstruckNepalin2015—alsohighlightstheuseofmobilephonelocationdata.Wilsonetal.(2016)rapidlyanalyzedthedisplacementmovementsof12millionde-identifiedmobilephoneuserswithinninedaysoftheearthquake.Duringthatperiod,anestimated390,000peoplelefttheKathmanduValley.Onalongertimescale,Luetal.(2016)usedcalldetailrecorddataforthreemonthsduringandtwoyearsfollowingCycloneMahasentoquantifythemagnitude,direction,duration,andseasonalityofmigrationinBangladesh.30Towardgender-differentialdisplacementanalysisDespitetheincreasinguseoflarge-scalemobilitydataforanalyzingdisaster-inducedmovementpatterns,gender-differentiatedimpactshavebeendifficulttoassessbecauseofthelackofgender-labeleddata.Toovercomethisdatagap,theDataforGoodteamatFacebook(nowMeta),incollaborationwiththeInternalDisplacementMonitoringCenter,producedDataforGoodDisplacementmaps,whichprovidegender-aggregatedandanonymizedesti-matesofdisplacementbasedoninformationfromFacebookuserswhohaveenabledlocationhistory.Thegoalofthesemapsistohelphumanitarianorganizationsunderstandtheoriginsanddestinationsofdisplacedpeople,andwhenthosepeoplecanreturntotheirhomes.ThemapsuseinverseprobabilityweightingtocorrectpotentialbiasesintheFacebookusergroupandtoimprovethepopulationdataset’srepresentativenessofthetrueunderlyingsociodemo-graphicandeconomiccharacteristics.31Map3.3DisplacementmobilitypatternsfollowingCycloneYaas,eastcoastofIndiaandtheBengalareaofBangladesh,May2021Source:WorldBankanalysisusingdatafromkepler.gl.Note:Comparisonofmapsrevealsthatdisplacedmalesmovefartherawaythandisplacedfemales.TheImpactsofClimateandEnvironmentalChangeonCities203UsingMeta’sdisplacementdata,Yabeetal.(2022),inabackgroundpaperpreparedforthisreport,analyzethedisplacementbehavioraldifferencesacrossgenderbothduringandafterCycloneYaas,whichstrucktheeastcoastofIndiainMay2021.Thedatausedintheirstudyprovideestimatesofthenumberofmalesandfemalesdisplacedfromonecitytoanother,coveringmorethan100citiesintheregion.Map3.3showstheoriginsanddestinationsofdisplacedpeopleafterthecyclone.In55–60percentofthecitiesaffectedbythecyclone,morementhanwomenweredisplaced,bothovertheshortterm(14days)andlongterm(50daysafterlandfallofthecyclone),andtheytraveledfarther.Kolkatawasthelargestmajorcityaffectedbythecyclone,with4.0percentofmalesand2.1percentoffemalesdisplacedinthetwoweeksfollowinglandfall.Furtherquantitativeanalysisofdisplacementdestinationsusingregressionmodelsrevealsthatdisplacedmalesandfemalesdiffersignificantlyinhowtheychoosetheirdestinations.Distancetodestination,distancefromcoastline,andinequalityofwealthinthedestinationcitywereallsignificantfactorsinthechoiceofdestinationformaledisplacedpopulations,whereasthesocialconnectednessindex(computedbythedensityoffriendtiesonFacebookbetweentwocities)wasmostsignificantinthechoicesofdisplacedfemales.Thisfindingisconsistentwithpreviousfindingsthatwomensuffermorethanmenfromdisintegrationofsocialties(Mitchell,Tanner,andLussier2007)andthatsocialties(friendsandfamily)canprovideamajorsourceofsupportforwomenbecauseoftheirlackofeconomicindependence(Kaya2018).204THRIVINGSuchinsightscouldhelpbetterdesignprogramsaimedatsupportingandempoweringdis-placedwomeninundertakingeconomicindependenceandrecovery(UNOCHA2019).Moreover,ifdataagreementsandanalyticalpipelinesarepreparedbeforehand,nearreal-timeanalysisofgender-differentiateddisplacementpatternscouldaiddecision-makersineffec-tivelydeliveringdisasterreliefaidandfinancialsupport.Animportantfuturestepwouldbetoexpandthestudytomultipledisastereventsworldwidetoevaluategender-differentiatedeffectsinregionswithdifferentculturalbackgroundsandunderdifferentlevelsofdisruptionscausedbyahazard.ClimatechangemaybeslowingtheurbanescalatoroutofpovertyByprovidingeconomicopportunitiesandamenities,urbanareascansupportpovertyreduction.Theurban–ruralgapsinpovertyandlivingstandardsremainstark.Asshowninfigure3.2,whenbothpovertyandurbanareasaremeasuredinagloballyconsistentmanner,povertyratesinurbanareas—particularlycitieswithahigherpopulationdensity—tendtobelowerthaninruralareas(Combesetal.2022).Nonmonetarylivingstandardsalsotendtobebetterindenserareas(Gollin,Kirchberger,andLagakos2021).32Beyondsuchstaticcompar-isons,urbanareashaveproventofacilitatepovertyreductioniftheurbanpoorhaveescapedfrompovertyovertime,ifrural-to-urbanmigrantshaveescapedfrompoverty,andiftheruralpoorbenefitfromurbanareasthrough,forexample,ahigherdemandforagriculturalproductsandhigherremittancesfromurbanareas.Indeed,despitetheprevalenceofchronicpovertyinSub-SaharanAfrica,upwardmobilityishigherinurbanareasthaninruralareas(DangandDabalen2019).Figure3.2PovertyratesatUS$1.90perday,bydegreeofurbanizationclassification,selectedcountriesSource:Combesetal.2022.Note:PovertyismeasuredusingtheinternationalpovertylineofUS$1.90perday,2011,purchasingpowerparity–adjusted.Becauseofthemodificationsinconsumptionaggregatesandspatialpricedeflators,thenational-levelpovertyratesdonotmatchtheofficialinternationalpovertyrates.Urbancenters,urbanclusters,andruralareasareclassifiedusingthedegreeofurbanizationdefinitionappliedtoWorldPopdata.SeeCombesetal.(2022)fordetails.020406080AngolaBangladeshEgypt,ArabRep.EthiopiaGhanaTanzaniaVietnamPovertyrate(%)NationalUrbanareasUrbancentersUrbanclustersRuralareasTheImpactsofClimateandEnvironmentalChangeonCities205Althoughpoverty,measuredbymonetaryincomeorconsumption,constitutesanimport-antdimensionofinclusiveness,itprovidesonlyastaticpictureofthedistributionofwelfareoutcomesacrosssegmentsofacity’spopulationataparticulartime.Inthiscontext,arguablyevenmoreimportanttoinclusivenessistheextenttowhichacitycanfacilitatetheupwardmovementofhouseholdsintheincomedistributionovertime,especiallytheextenttowhichthecity’senvironmentprovidestheconditionstofacilitateahousehold’sescapefrompovertyand,fromthere,upintothemiddleclass.AccordingtotheprominenturbaneconomistEdwardGlaeser(2012),existingempiricalevidencesuggeststhatcitiesprovidean“escalatoroutofpoverty.”Moreover,thatescalatoroperatesfasterinlargercities,eventhoughthosecitiestendtohavehigherlevelsofincomeinequality.33Astheincidenceofclimaticshockscontinuestorise,however,citiesmayfailtofulfilltheirpotentialastheescalatoroutofpoverty.Urbanresidentsremainvulnerableorchronicallypoorifdeprivedofaccesstoeconomicopportunitiesandamenitiesin,forexample,informalsettlements(Marx,Stoker,andSuri2013).Itisalsopossiblethathighmigrationcostspreventallbutalimitedshareofpeoplefrommigratingtourbanareas(Lagakos2020).Moreover,thelackofspaceandaffordablehousinginurbandestinationslimitsthenumberofmigrantsthatcanbeaccommodatedorforcesthemtoliveinperipheralandpossiblyenvironmentallyhaz-ardousareas,leavingthemtrappedinpovertyandsqualor.Meanwhile,urbaneconomiesmayhavenospillovereffectstoruraleconomiesiftheyarenotintegratedbyadequateconnectiveinfrastructure.Theimpactsofclimateandenvironmentalshocksexacerbatesuchfailures.Forexample,floodingcouldworsenvulnerabilityandchronicpovertyinurbanareasbecausethepoorhavealimitedfinancialbuffertocopewithshocks.Inaddition,asdescribedinearliersections,poorpeopleincitiestendtoliveinareaspronetoflooding,sotheirassetsaremoreexposedtoshocks.Thus,climatechange–relatedandotherenvironmentalstresses,especiallyininteractionwiththeotherstressesassociatedwithurbanpopulationgrowth,couldslowdown,orevenhaltcompletely,acity’spovertyescapeescalator.ChileandColombiaaretwoexamplesofcountriesinwhichclimaticrisksappeartohaveweakenedupwardmobilityinurbanareas.AbackgroundpaperpreparedforthisreportbyAbanokovaetal.(2022)analyzesthetransitionsinhouseholds’povertystatusbyapplyingasyntheticpanelanalysisasusedinDangetal.(2014).34Theiranalysisrevealsthatinthetwocountriesmanypeoplehaveescapedfrompovertyinurbanareas.InColombia,13percentoftheurbanpoorescapedfrompovertybetween2008and2010,and,inChile,almost65percentbetween2011and2015.Inbothcountries,peopleinlargercitiesweremorelikelytoescapefrompoverty(figure3.3);however,asfloodrisksroseinlargecities,thetransitionfrompoortononpooressentiallyhalted.Incitieswithlargepopulations,householdsinhighflood-riskareashaveasubstantiallylowerpredictedprobabilityofescapingfrompovertythanthoseinlow-riskareas.AsimilarpatternisobservedinIndonesia,whereclimateshockshavelessenedtheabilityofurbanareastoreducepoverty.AnempiricalanalysisconductedforthisreportusingpaneldatafromtheIndonesiaFamilyandLifeSurvey,andbuildingonastudybySetiawan,Tiwari,andRizal(2018),confirmsthatfrom1993to2014peoplemovingtolargemetroareashadahigherchanceofescapingpoverty(figure3.4).Inthoseareas,however,floodrisksappeartohavereducedsuchupwardmobility;thepredictedprobabilityofescapingfrompovertywaslowinlargemetroareasexperiencingextremerainfalls.Thus,climateshocksinurbanareasmaymakeanescapefrompovertylesslikely.Inotherwords,citiesmoreexposedtoandlessabletomanagethesestresseswillbelessinclusiveintermsoftheopportunitiestheyprovideformovingupthewelfareladder.206THRIVINGFigure3.3Predictedprobabilityofescapingpoverty,bytownpopulationsizeandfloodrisk,ColombiaandChileSource:Abanokovaetal.2022.Note:Theverticalaxisindicatestheprobabilityofachangefrompoortononpoorpredictedforeachhouseholdbythesyntheticpanelanalysis.Povertyismeasuredbytheupper-middle-incomepovertylineofUS$5.50perday,2011,purchasingpowerparity–adjusted.Thehorizontalaxisindicatesthelogoftownpopulations.Thefloodriskisclassifiedhighformunicipalities(Colombia)andcomuna(Chile),withina100-yearfloodreturnperiod,withthetop25percentflooddepthineachcountry.a.Colombia,2008–10b.Chile,2011–150.130.140.150.160.170.1878910111213141516PredictedprobabilityPredictedprobabilityLogofpopulation,2015Logofpopulation,2015LowfloodriskHighfloodrisk0.700.720.740.760.78678910111213141516Figure3.4Predictedprobabilityofescapingpoverty,bylocation,Indonesia,1993–2014Source:Abanokovaetal.2022.Note:Theprobabilityofbeingnonpoorispredictedforeachhouseholdbasedonthetwo-wayfixed-effectsregressionmodelusingthefiveroundsoftheIndonesiaFamilyandLifeSurveypaneldata.Locationsareclassifiedbycore(multidistrictmetrocores),peripheryurban(urbannoncoredistricts),otherurban(single-districtmetrosornonmetrourbandistricts),andruraldistricts.SeeRoberts,GilSander,andTiwari(2019)fordetails.TherainyyeardistrictisidentifiedwithaStandardizedPrecipitation-EvapotranspirationIndex(SPEI)scoreof2orgreater.0.70.80.91.00.70.80.91.00.70.80.91.00.70.80.91.0PredictedprobabilityDryNormalRainya.CoreDryNormalRainyb.PeripheryurbanDryNormalRainyc.OtherurbanDryNormalRainyd.RuralTheImpactsofClimateandEnvironmentalChangeonCities207SummaryandconclusionsTheeffectsofachangingclimateoncitiesareheterogeneousacrossplaces,sectors,andpeople.Becausecitiesconcentratepeople,activities,andinfrastructure,theycanexperienceintensedirecteffectsofclimateshocksandstressors.Thesedirecteffectsincluderapid-onsetevents—suchasextremeheatandcold,floods,andwildfires—andslow-onsetevents—suchasdrought,airpollution,andlanddegradation.Buttheseimpactsdonotoccurinisolation;theyinteractandcompound,makingtheeffectsmoreuncertainandpotentiallymoredisastrous.Yetinvest-mentsindurableconstruction(representingcapitalthatwouldbeverydifficulttoreallocate)continueinlocationsthatmaybecomeunlivableovertime.Thepatternsofdisplacementinresponsetosuddenshocksillustrategender-differentiatedcharacteristics,suggestingtheexis-tenceofmanypoorlyunderstoodfacetsofresponsethatrequireattention.Citiesfunctionaspartofalargerenvironment.Climateshocksinruralareasleadtofasterurbanizationbecauselocal-levelmigrationrepresentsanimportantcopingmechanism,par-ticularlyinpoorercountries.Suchclimate-inducedmigrationisalsoassociatedwithgrowingurbansprawl,oftenexpandinginformalsettlementswhereservicedeliveryremainsachal-lenge.Droughtsinruralareasaffectcitiesnotonlyviamigrationbutalsobydirectlyaffectingtheurbanwatersupply.Citiesrelyonruralareasformanynaturalresources—waterispara-mount,butfoodisalsoimportant.Climateshocksaffectagriculturalproductionandproduc-tivity,withimpactsspillingovertourbanareasthroughhigherandmorevolatilefoodprices.Climatechangeaffectspopulationgroupsincitiesdifferently.Inmanycities,thepoor,thoughnotallofthem,aremoreexposedtoclimatehazards.Thisincreasedexposurecouldresultfromsortingandinformationasymmetriesandbecausetheurbanpoortradesafetyforbetteraccesstojobsorservices.Evenifnotthemostexposed,thepoorarehardest-hitbyclimateshocksandarethemostvulnerablebecausetheyareleastabletomitigatetheeffectsofhazards,suchasthroughaccesstofinancialorinsurancemarkets.Inaddition,thosewithlowerlevelsofedu-cationorworkingininformaljobswilllikelyfeelmoresignificanteffectsthroughsharperfallsintheirproductivity.Forthesereasons,climatechangemaybeslowingtheurbanescalatoroutofpoverty.208THRIVINGTHRIVING208Notes1.Thecostofmorbidityincludesresourcecosts(thefinancialcostsofavoidingortreatingpollution-associatedillnesses),opportunitycosts(theindirectcostsoflossoftimeforworkandleisure),anddisutilitycosts(thecostofpain,suffering,ordiscomfort)—seeWorldBank(2020c).2.Bycontrast,theintangibleimpacts,suchasdisruptionsofcommuters’travelroutes,aredifficulttoquantify.AccordingtoHe,Liu,etal.(2021),clustersoflow-incomeresidentsincuralargeshareoftheeffectsofdisruptions.TheirstudyrevealsthattheregularfloodsinKinshasa,DemocraticRepublicofCongo,affectedtransportationservices(increasingpublictransitheadways,forcingtransitrerouting,anddecreasingtravelspeeds)andjobaccessibility(causingtraveldelaysresultinginlossesinaccessibility).3.Thiscomplexityisequally,ifnotmore,relevanttohowresponsestoclimateeventsareframed.Forexample,watermanagementwouldhavetorespondsimultaneouslytointensifyingdroughtsandfloods—asubjectaddressedinmoredetailinchapter5.4.Globalwarmingisashorthandterm.Climatemodelspredictthatsomeplacesmaybecomewarmer,somecooler,somewetter,somedrier,andsomestormierorlessstormy.5.Becauseofdifferencesinoceandynamicsandtectonics,sealeveldoesnotriseuniformlyacrossspace.A1°Criseinglobaltemperaturescouldresultinmorethana2°Criseinsomenorthernlatitudes,butonlya0.5°Criseinsomeequatorialregions.6.Warmingtrendsarenonlinear.Warmingacceleratedovermostofthetwentiethcentury,butthetrendhasbeenmuchstrongersince1980(Franzke2014).7.Emporisdatacover693,855completedbuildings,alongwiththeirexactgeographicalcoordinates,yearofconstruction,anddateofdemolition.Thedatasetmostlycapturesbuildingsover55meterstall.Skyscrapersaredefinedasbuildingsmorethan100meterstall.8.Thestudyusesthemeanofthemax(daytime)temperaturesduringthehottestseasonasaproxyforwhetheracityisa“futurebadlocation.”9.Ifacountryhasonlybadlocationsinthefuture,oneconsiderationiswhetherthegovernmentmaywanttohedgeitsrealestateportfoliobyencouragingitseconomicagents(thosewhoinfluencecapitalmarketsandtheeconomyatlarge)toinvestinrealestateabroad.10.TheUSCongressenactedtheDodd-FrankActinJuly2010totacklethisspecificissue.SimilarlegislationwasthenenactedintheEuropeanUnionandtheUnitedKingdom.11.Forsimplicity’ssake,industrializedcountriesaredefinedasthoseinwhichagriculture’sshareofnationalincomein1985wasgreaterthan30percent.12.Thisfindingisbasedonanalysisofnighttimelightsdata.13.AnearlyexampleisaninnovativedecisionbyNewYorkCitytoacquirelandinthenearbyCatskillsregiontofilterandstorewaterinnaturalecosystems,savingthecityUS$6billionincapitalcostsrelatedtobuildingawaterfiltrationplant(Damaniaetal.2017).14.Thesestudiesarebasedonspatialanalysisoftheimpactofruralshocksonurbanfoodprices,controllingforinternationalcommodityprices(basedonthecommoditypriceindexoftheTheImpactsofClimateandEnvironmentalChangeonCities209209TheImpactsofClimateandEnvironmentalChangeonCitiesFoodandAgricultureOrganizationoftheUnitedNations)anddifferentiatingbytypeoffoodgroup—thatis,perishableandnonperishablefooditems.15.Assessmentofthemitigationoftheimpactofshocksandstressorsisbasedontraveltimetothenearestcity,whichdependsonthetypeandqualityoftheroadnetwork(Meijeretal.2018;Nelsonetal.2019).16.SouthAsia:JalalabadandKabul(Afghanistan);Chittagong(Bangladesh);Auraiya,Amritsar,andKanpur(India);Karachi(Pakistan);ColomboandKandy(SriLanka).Sub-SaharanAfrica:Bujumbura(Burundi);Ougadougou(BurkinaFaso);Bangui(CentralAfricanRepublic);Bamako(Mali);Niamey(Niger);N’djamena(Chad);Bulawayo,Harare,andMutare(Zimbwabwe).17.TheSouthAsiandatawereproducedinWorldBankprojectsin2014and2015fromsurveyinghigh-resolutionimagerytodeterminelandusecategoriesforcityparcels.ThedataforAfricancitiesareatahigherresolutionandwereassembledforthisreport.Formalresidentialareasarepredictedusinganunsupervisedmachinelearningalgorithm(Chlouba,Mukim,andZaveri2022).18.Accordingly,whenthesampleissplitbyregion,slumsinSub-SaharanAfricaare,onaverage,lessexposedtobothhazards,whereastheoppositeistrueinSouthAsia.19.Thisreferstothesortingofdifferenttypesofworkersacrosslocationsduetoemploymentopportunitiesordifferencesinamenities(Coutureetal.2019;Diamond2016;Moretti2013).20.Theliteratureonrural-urbanmigrationconfirmstheexistenceofsuchinformationalgapsbutpointstoanoppositeeffect:thelackofinformationonhigherreturnsinurbanareasdiscouragesadvantageousmigration(Aker,Clemens,andKsoll2011;Baseler2021;Bryan,Chowdhury,andMobarak2014).21.Forexample,arichbodyofliteratureindevelopmenteconomicsdocumentshowthepoorpossesslowerfinancialliteracy;thus,theiracquisitionoffinancialinformationcanhavesubstantialpositivewelfareeffects,amongothers(Hastings,Madrian,andSkimmyhorn2013;Karlan,Ratan,andZinman2014;LusardiandMitchell2014).22.TheHarmonizedSurveyonHouseholdLivingConditions(EnquêteHarmoniséesurlesConditionsdeViedesMénages)isconductedinWestAfricanEconomicandMonetaryUnionmemberstatestoprovidegovernmentswithreliablestatisticstounderpintheirpovertyreductionefforts.Datacollectiontookplaceintworounds:October17–December31,2018,andApril1–June30,2019.TheresultsarerepresentativeatthesubnationallevelandacrossBamakoinbothurbanandruralareas(WorldBank2021d).23.Thosebelowthepovertylinehadlossesfourtofivetimeshigherthanthemonthlyincomeoflow-incomehouseholds,threetimeshigherthoseofmiddle-incomehouseholds,andtwotimeshigherthanthoseofthehigh-incomegroup.PatankarandPatwardhan(2016)alsofindthatpoorpeoplehavelost60percentmoreoftheirestimatedwealthrelativetononpoorpeople.24.OriginallyfromWorldBankandAustralianAID(2014).ThesurveyconductedinHoChiMinhCityrevealsthat86percentofsurveyedpoorhouseholdshadmorehealthproblemsstemmingfrompollutedfloodwater,comparedwith64percentofnonpoorones.Sixty-ninepercentofthepoorwereaffectedintermsofemploymentand67percentintermsofincome,comparedwith56percentand40percent,respectively,ofthenonpoor.25.TheagglomerationsareDenpasarinBali;Jakarta(Jabodetabek),Pasuruan,Probolinggo,Semarang,Surabaya,andYogyakartainJava;andMedaninSumatra.Educationalattainment210THRIVINGTHRIVING210referstothehighestlevelofschoolingorqualificationthataworkerhasacquired,rangingfromincompleteprimaryschooltotertiarydegrees.(Roberts,GilSander,andTiwari2019).26.Theempiricalanalysisconsistsofregressingtherelevantmeasureofenvironmentalrisk(eitherpluvialorfluvialfloodriskorexcessiveheat)ontheaveragelevelofeducationinthevillageandcityfixedeffects,therebycapturingthevariationwithincities.27.Almostone-fourthoftheworld’surbanpopulation,overabillionpeople,livesininformalsettlementsorslumswith80percentinEastandSoutheastAsia,Sub-SaharanAfrica,andCentralandSouthAsia,https://unstats.un.org/sdgs/report/2019/goal-11/.28.Thisincludestheirunderrepresentationin(1)watersupplyandsanitationutilitiesandministries;(2)agenciesallocatingemergencyorpostdisasterrecoveryassistance;(3)firmsandbusinessesthatstandtobenefitfromemergencyrelief;and(4)increasinglyimportantupstreamdecisionsonwaterallocationsandtechnicaloptionsformoresustainablewatermanagement(UNICEF2016).29.Thesesurveysconstitutethefirstindustrywideglobaldatabaseongenderdiversityinwaterinstitutions.AdditionalsurveyswillexpandthedatabasethroughtheEqualAquaplatform,acollaborativeinitiativelaunchedbytheWorldBanktodeepenthedialogueongenderdiversityandinclusioninwatersectorjobs.SeeWorldBankWaterDataforsummariesofpastandongoingsurveys,https://wbwaterdata.org/breakingbarriers/home/.30.Foracomprehensivereviewonthescientificliteratureandtheusageofmobilitydatafordisasterresponseandrecovery,seeYabeetal.(2022).31.FacebookDataforGood,DisplacementMaps:Methodology,2021,https://dataforgood.facebook.com/dfg/docs/methodology-displacement-maps.32.Examplesofthenonmonetarydimensionsofurbanpovertyincludesqualidlivingconditions,risksfromthedis-amenitiesofurbanization(poorsanitation,airpollution,crimeandviolence,trafficaccidents),andincreasingdisasterrisks.33.ForevidenceofthisforIndonesiancities,seechapter4inRoberts,GilSander,andTiwari(2019).34.Thesyntheticpanelanalysiscreatespseudo-paneldatafromarepeatedcross-sectionalhouseholdsurveydataset,bywhichthechangesinpovertystatusamongthesamplehouseholdsfortwotimepointscanbeanalyzed.ReferencesAbanokova,K.,H.-A.Dang,S.Nakamura,S.Takamatsu,C.Pei,andD.Prospere.2022.“IsClimateChangeSlowingtheUrbanEscalatoroutofPoverty?EvidencefromIndonesiaandLAC.”Backgroundpaperpreparedforthisreport,WorldBank,Washington,DC.Abas,N.,M.S.Saleem,E.Kalair,andN.Khan.2019.“CooperativeControlofRegionalTransboundaryAirPollutants.”EnvironmentalSystemsResearch8(1):1–14.Abel,G.J.,M.Brottrager,J.C.Cuaresma,andR.Muttarak.2019.“Climate,ConflictandForceMigration.”GlobalEnvironmentalChange54:239–49.TheImpactsofClimateandEnvironmentalChangeonCities211211TheImpactsofClimateandEnvironmentalChangeonCitiesAbwola,N.,andI.Michelis.2020.WhatHappened?HowtheHumanitarianResponsetoCOVID-19FailedtoProtectWomenandGirls.InternationalRescueCommittee,NewYork.Aker,J.C.,M.A.Clemens,andC.Ksoll.2011.“MobilesandMobility:TheEffectofMobilePhonesonMigrationinNiger.”ProceedingsoftheGermanDevelopmentEconomicsConference,Berlin.Andrews,O.,C.LeQuéré,T.Kjellstrom,B.Lemke,andA.Haines.2018.“ImplicationsforWorkabilityandSurvivabilityinPopulationsExposedtoExtremeHeatunderClimateChange:AModellingStudy.”LancetPlanetaryHealth2(12):e540–47.Anenberg,S.C.,P.Achakulwisut,M.Brauer,D.Moran,J.S.Apte,andD.K.Henze.2019.“ParticulateMatter–AttributableMortalityandRelationshipswithCarbonDioxidein250UrbanAreasWorldwide.”ScientificReports9(1):1–6.Avashia,V.,andA.Garg.2020.“ImplicationsofLandUseTransitionsandClimateChangeonLocalFloodinginUrbanAreas:AnAssessmentof42IndianCities.”LandUsePolicy95:104571.Avner,P.,andS.Hallegatte.2019.“MoralHazardvs.LandScarcity:FloodManagementPoliciesfortheRealWorld.”PolicyResearchWorkingPaper9012,WorldBank,Washington,DC.Azunre,G.A.,O.Amponsah,S.A.Takyi,H.Mensah,andI.Braimah.2022.“UrbanInformalitiesinSub-SaharanAfrica(SSA):ASolutionfororBarrieragainstSustainableCityDevelopment.”WorldDevelopment152:105782.Baez,J.E.,L.Lucchetti,M.E.Genoni,andM.Salazar.2017.“GonewiththeStorm:RainfallShocksandHouseholdWell-BeinginGuatemala.”JournalofDevelopmentStudies53(8):1253–71.Bai,Y.,L.Costlow,A.Ebel,S.Laves,Y.Ueda,N.Volin,M.Zamek,A.Herforth,andW.Masters.2021.“Review:RetailConsumerPriceDataRevealsGapsandOpportunitiestoMonitorFoodSystemsforNutrition.”FoodPolicy104:1021–48.Bakkensen,L.A.,andL.Barrage.2021.“GoingUnderwater?FloodRiskBeliefHeterogeneityandCoastalHomePriceDynamics.”ReviewofFinancialStudies35(8):3666–709.Balboni,C.2021.“InHarm’sWay?InfrastructureInvestmentsandthePersistenceofCoastalCities.”DepartmentofEconomics,MassachusettsInstituteofTechnology,Cambridge,MA.Baldauf,M.,L.Garlappi,andC.Yannelis.2020.“DoesClimateChangeAffectRealEstatePrices?OnlyifYouBelieveinIt.”ReviewofFinancialStudies33(3):1256–95.Banerjee,A.V.,andS.Mullainathan.2008.“LimitedAttentionandIncomeDistribution.”AmericanEconomicReview98(2):489–93.Bangalore,M.,A.Smith,andT.Veldkamp.2019.“ExposuretoFloods,ClimateChange,andPovertyinVietnam.”EconomicsofDisastersandClimateChange3(1):79–99.Barrios,S.,L.Bertinelli,andE.Strobl.2006.“ClimateChangeandRural-UrbanMigration:TheCaseofSub-SaharanAfrica.”JournalofUrbanEconomics26:656–73.Baseler,T.2021.“HiddenIncomeandthePerceivedReturnstoMigration.”http://dx.doi.org/10.2139/ssrn.3534715.Bateman,J.M.,andB.Edwards.2002.“GenderandEvacuation:ACloserLookatWhyWomenAreMoreLikelytoEvacuateforHurricanes.”NaturalHazardsReview3(3):107–17.212THRIVINGTHRIVING212Baylis,P.,andJ.Boomhower.2019.“MoralHazard,Wildfires,andtheEconomicIncidenceofNaturalDisasters.”NBERWorkingPaper26550,NationalBureauofEconomicResearch,Cambridge,MA.https://ssrn.com/abstract=3504434.Bellisario,J.,G.Cowan,andA.Raisz.2021.TheTrueCostofWildfires:AnalyzingtheImpactofWildfiresontheCaliforniaEconomy.Berkeley,CA:BayAreaCouncilEconomicInstitute.Bento-Gonçalves,A.,andA.Vieira.2020.“WildfiresintheWildland-UrbanInterface:KeyConceptsandEvaluationMethodologies.”ScienceoftheTotalEnvironment707:135592.Benveniste,H.,M.Oppenheimer,andM.Fleurbaey.2020.“EffectofBorderPolicyonExposureandVulnerabilitytoClimateChange.”ProceedingsoftheNationalAcademyofSciences117(43):26692–702.Bernstein,A.,M.T.Gustafson,andR.Lewis.2019.“DisasterontheHorizon:ThePriceEffectofSeaLevelRise.”JournalofFinancialEconomics134(2):253–72.Bertinelli,L.,andE.Strobl.2013.“QuantifyingtheLocalEconomicGrowthImpactofHurricaneStrikes.”JournalofAppliedMeteorologyandClimatology52(8):1688–97.Bilal,A.,andE.Rossi-Hansberg.2021.“LocationasanAsset.”Econometrica89(5):2459–95.Borgomeo,E.,A.Jägerskog,E.Zaveri,J.Russ,A.Khan,andR.Damania.2021.EbbandFlow,Volume2:WaterintheShadowofConflictintheMiddleEastandNorthAfrica.Washington,DC:WorldBank.Bosetti,V.,C.Cattaneo,andG.Peri.2021.“ShouldTheyStayorShouldTheyGo?ClimateMigrantsandLocalConflicts.”JournalofEconomicGeography21(4):619–51.Botzen,W.W.,O.Deschenes,andM.Sanders.2019.“TheEconomicImpactsofNaturalDisasters:AReviewofModelsandEmpiricalStudies.”ReviewofEnvironmentalEconomicsandPolicy13(2):167–88.Broadbent,A.M.,E.S.Krayenhoff,andM.Georgescu.2020.“TheMotleyDriversofHeatandColdExposurein21stCenturyUSCities.”ProceedingsoftheNationalAcademyofSciences117(35):21108–17.Bryan,G.,S.Chowdhury,andA.M.Mobarak.2014.“UnderinvestmentinaProfitableTechnology:TheCaseofSeasonalMigrationinBangladesh.”Econometrica82(5):1671–748.Burke,M.,S.M.Hsiang,andE.Miguel.2015.“ClimateandConflict.”AnnualReviewofEconomics7:577–617.Burzynski,M.,C.Deuster,F.Docquier,andJ.deMelo.2019.“ClimateChange,Inequality,andHumanMigration.”DiscussionPaperSeries13997,CentreforEconomicPolicyResearch,London.CamposGarcia,A.,N.Holm-Nielsen,C.DiazG.,D.M.RubianoV.,C.R.CostaP.,F.RamirezC.,andE.Dickson.2011.AnalysisofDisasterRiskManagementinColombia:AContributiontotheCreationofPublicPolicies.Washington,DC:WorldBank.Cannon,T.2002.“GenderandClimateHazardsinBangladesh.”GenderandDevelopment10(2):45–50.Cashin,P.,K.Mohaddes,andM.Raissi.2017.“FairWeatherorFoul?TheMacroeconomicEffectsofElNiño.”JournalofInternationalEconomics106:37–54.TheImpactsofClimateandEnvironmentalChangeonCities213213TheImpactsofClimateandEnvironmentalChangeonCitiesCastells-Quintana,D.,M.Krause,andT.K.J.McDermott.2021.“TheUrbanisingForceofGlobalWarming:TheRoleofClimateChangeintheSpatialDistributionofPopulation.”JournalofEconomicGeography21:531–56.Castells-Quintana,D.,M.d.P.Lopez-Uribe,andT.K.J.McDermott.2018.“AdaptationtoClimateChange:AReviewthroughaDevelopmentEconomicsLens.”WorldDevelopment104:183–96.Cattaneo,C.,M.Beine,C.Fröhlich,D.Kniveton,I.Martínez-Zarzoso,M.Mastrorillo,K.Millock,etal.2019.“HumanMigrationintheEraofClimateChange.”ReviewofEnvironmentalEconomicsandPolicy13:189–206.CEDLAS(CenterforDistributive,LaborandSocialStudies)andWorldBank.2014.AGuidetoSEDLAC—Socio-EconomicDatabaseforLatinAmericaandtheCaribbean.Washington,DC:WorldBank.Chattopadhyay,R.,andE.Duflo.2004.“WomenasPolicyMakers:EvidencefromaRandomizedPolicyExperimentinIndia.”Econometrica72(5):1409–43.Chaturvedi,S.,S.Das,andK.Mahajan.2021.“TheImportanceofBeingEarnest:WhatExplainstheGenderQuotaEffectinPolitics?”WorkingPaperNo.52,DepartmentofEconomics,AshokaUniversity,Haryana,India.Chindarkar,N.2012.“GenderandClimateChange–InducedMigration:ProposingaFrameworkforAnalysis.”EnvironmentalResearchLetters7(2):025601.Ching,J.,andM.Kajino.2020.“RethinkingAirQualityandClimateChangeafterCOVID-19.”InternationalJournalofEnvironmentalResearchandPublicHealth17(14):5167.Chlouba,V.,M.Mukim,andE.Zaveri.2022.“HowDoClimateChange-RelatedStressorsAffectUrbanForm?”Backgroundpaperpreparedforthisreport,WorldBank,Washington,DC.Choy,D.L.,P.Clarke,S.Serrao-Neumann,R.Hales,O.Koschade,andD.Jones.2016.“CoastalUrbanandPeri-UrbanIndigenousPeople’sAdaptiveCapacitytoClimateChange.”InBalancedUrbanDevelopment:OptionsandStrategiesforLiveableCities,vol.72,editedbyB.Maheshwari,B.Thoradeniya,andV.P.Singh,441–61.CapitalRegionofDenmark:ChamSpringer.Cohen,A.J.,M.Brauer,R.Burnett,H.R.Anderson,J.Frostad,K.Estep,K.Balakrishnanetal.2017.“Estimatesand25-YearTrendsoftheGlobalBurdenofDiseaseAttributabletoAmbientAirPollution:AnAnalysisofDatafromtheGlobalBurdenofDiseasesStudy2015.”Lancet389:1907–18.Colon,C.,S.Hallegate,andJ.Rozenberg.2021.“CriticalityAnalysisofaCountry’sTransportNetworkviaanAgent-BasedSupplyChainModel.”NatureSustainability4:209–15.Combes,P-P.,S.Nakamura,M.Roberts,andB.Stewart.2022.“EstimatingUrbanPovertyConsistentlyacrossCountries.”WorldBankPovertyandEquityNotes,WorldBank,Washington,DC.Conte,B.,K.Desmet,D.Nagy,andE.Rossi-Hansberg.2021.“LocalSectoralSpecializationinaWarmingWorld.”JournalofEconomicGeography21(4):493–530.Couture,V.,C.Gaubert,J.Handbury,andE.Hurst.2019.“IncomeGrowthandtheDistributionalEffectsofUrbanSpatialSorting.”NBERWorkingPaper26142,NationalBureauofEconomicResearch,Cambridge,MA.214THRIVINGTHRIVING214Croitoru,L.,J.C.Chang,andJ.Akpokodje.2020.“TheHealthCostofAmbientAirPollutioninLagos.”JournalofEnvironmentalProtection11:753–65.Cruz,J-L.,andE.Rossi-Hansberg.2021.“TheEconomicGeographyofGlobalWarming.”NBERWorkingPaper28466,NationalBureauofEconomicResearch,Cambridge,MA.Damania,R.,S.Desbureaux,M.Hyland,A.Islam,S.Moore,A.-S.Rodella,J.Russ,andE.Zaveri.2017.UnchartedWaters:TheNewEconomicsofWaterScarcityandVariability.Washington,DC:WorldBank.Dang,H.-A.H.,andA.L.Dabalen.2019.“IsPovertyinAfricaMostlyChronicorTransient?EvidencefromSyntheticPanelData.”JournalofDevelopmentStudies55(7):1527–47.Dang,H.-A.H.,P.Lanjouw,J.Luoto,andD.McKenzie.2014.“UsingRepeatedCross-SectionstoExploreMovementsintoandoutofPoverty.”JournalofDevelopmentEconomics107:112–28.Das,M.B.2017.TheRisingTide:ANewLookatWaterandGender.Washington,DC:WorldBank.Das,M.B.2020.“DryCitiesCan’tBeHealthywithoutReducingInequalities.”BMJOpinion(blog),November16,2020.https://blogs.bmj.com/bmj/2020/11/16/dry-cities-cant-be-healthy-without-reducing-inequalities/.Das.,P.2014.“Women’sParticipationinCommunity-LevelWaterGovernanceinUrbanIndia:TheGapbetweenMotivationandAbility.”WorldDevelopment64:206–18.Dasgupta,S.,M.S.Islam,M.Huq,Z.HuqueKhan,andM.R.Hasib.2019.“QuantifyingtheProtectiveCapacityofMangrovesfromStormSurgesinCoastalBangladesh.”PLoSOne14(3):E0214079.Dasgupta,S.,N.vanMaanen,S.N.Gosling,F.Piontek,C.Otto,andC-F.Schleussner.2021.“EffectsofClimateChangeonCombinedLabourProductivityandSupply:AnEmpirical,Multi-ModelStudy.”LancetPlanetaryHealth5(7):E455–65.Dercon,S.,D.O.Gilligan,J.Hoddinott,andT.Woldehanna.2009.“TheImpactofAgriculturalExtensionandRoadsonPovertyandConsumptionGrowthinFifteenEthiopianVillages.”AmericanJournalofAgriculturalEconomics91(4):1007–21.Deryugina,T.2014.“TheFiscalCostofHurricanes:DisasterAidversusSocialInsurance.”AmericanEconomicJournal:EconomicPolicy9(3):168–98.Desbureaux,S.,andA.-S.Rodella.2019.“DroughtintheCity:TheEconomicImpactofWaterScarcityinLatinAmericanMetropolitanAreas.”WorldDevelopment114:13–27.Desmet,K.,andR.Jedwab.2022.“AreWeOver-buildingin‘Bad’LocationsGlobally?FutureClimateChangeandDurableRealEstate.”Backgroundpreparedpaperforthisreport,WorldBank,Washington,DC.Desmet,K.,andE.Rossi-Hansberg.2011.“SpatialDevelopment.”NBERWorkingPaper15349,NationalBureauofEconomicResearch,Cambridge,MA.Desmet,K.,andE.Rossi-Hansberg.2012.“OntheSpatialEconomicImpactofGlobalWarming.”NBERWorkingPaper18546,NationalBureauofEconomicResearch,Cambridge,MA.Desmet,K.,andE.Rossi-Hansberg.2015.“TheGeographyofDevelopment:EvaluatingMigrationRestrictionsandCoastalFlooding.”NBERWorkingPaper21087,NationalBureauofEconomicResearch,Cambridge,MA.TheImpactsofClimateandEnvironmentalChangeonCities215215TheImpactsofClimateandEnvironmentalChangeonCitiesDesmet,K.,R.E.Kopp,S.A.Kulp,D.K.Nagy,M.Oppenheimer,E.Rossi-Hansberg,andB.H.Strauss.2021.“EvaluatingtheEconomicCostofCoastalFlooding.”AmericanEconomicJournal:Macroeconomics13(2):444–86.Devoto,F.,E.Duflo,P.Dupas,W.Parienté,andV.Pons.2012.“HappinessonTap:PipedWaterAdoptioninUrbanMorocco.”AmericanEconomicJournal:EconomicPolicy4(4):68–99.Diamond,R.2016.“TheDeterminantsandWelfareImplicationsofUSWorkers’DivergingLocationChoicesbySkill:1980-2000.”AmericanEconomicReview106(3):479–524.Dodman,D.,B.Hayward,M.Pelling,V.CastanBroto,W.Chow,E.Chu,R.Dawson,etal.2022.“Cities,SettlementsandKeyInfrastructure.”InClimateChange2022:Impacts,Adaptation,andVulnerability,ContributionofWorkingGroupIItotheSixthAssessmentReportoftheIntergovernmentalPanelonClimateChange.Cambridge,UK:CambridgeUniversityPress.Dudley,N.,andS.Stolton.2003.RunningPure:TheImportanceofForestProtectedAreastoDrinkingWater.Washington,DC:WorldBank/WWFAllianceforForestConservationandSustainableUse.Erban,L.E.,S.M.Gorelick,andH.A.Zebker.2014.“GroundwaterExtraction,LandSubsidence,andSea-LevelRiseintheMekongDelta,Vietnam.”EnvironmentalResearchLetters9(8):084010.Erman,A.,M.Tariverdi,M.Obolensky,X.Chen,R.C.Vincent,S.Malgioglio,J.Rentschler,etal.2019.“WadingOuttheStorm:TheRoleofPovertyinExposure,VulnerabilityandResiliencetoFloodsinDaresSalaam.”PolicyResearchWorkingPaper8976,WorldBank,Washington,DC.Erman,A.,E.Motte,R.Goyal,A.Asare,S.Takamatsu,X.Chen,S.Malgioglio,A.Skinner,N.Yoshida,andS.Hallegatte.2020.“TheRoadtoRecovery:TheRoleofPovertyintheExposure,VulnerabilityandResiliencetoFloodsinAccra.”EconomicsofDisastersandClimateChange4(1):171–93.Evans,G.W.2019.“ProjectedBehavioralImpactsofGlobalClimateChange.”AnnualReviewofPsychology70:449–74.Fauconnier,I.,A.Jenniskens,P.Perry,S.Fanaian,S.Sen,V.Sinha,andL.Witmer.2018.WomenasChange-MakersintheGovernanceofSharedWaters.Gland,Switzerland:InternationalUnionforConservationofNature.Fay,M.2005.TheUrbanPoorinLatinAmerica.Washington,DC:WorldBank.Fitch,A.2021.“TexasWinterStormStrikesChipMakers,CompoundingSupplyWoes.”WallStreetJournal,February17,2021.https://www.wsj.com/articles/texas-winter-storm-strikes-chip-makers-compounding-supply-woes-11613588617.Florczyk,A.J.,M.Melchiorri,C.Corbane,M.Schiavina,M.Maffenini,M.Pesaresi,P.Politis,etal.2019.“DescriptionoftheGHSUrbanCentreDatabase2015.”PublicRelease2019,Version1.0,PublicationsOfficeoftheEuropeanUnion,Luxembourg.Ford,J.D.,T.Pearce,G.McDowell,L.Berrang-Ford,J.S.Sayles,andE.Belfer.2018.“VulnerabilityandItsDiscontents:ThePast,Present,andFutureofClimateChangeVulnerabilityResearch.”ClimaticChange151(2):189–203.Frame,D.J.,S.M.Rosier,I.Noy,L.J.Harrington,T.Carey-Smith,S.N.Sparrow,D.A.Stone,andS.M.Dean.2020.“ClimateChangeAttributionandtheEconomicCostsofExtremeWeatherEvents:AStudyonDamagesfromExtremeRainfallandDrought.”ClimaticChange162(2):781–97.216THRIVINGTHRIVING216Frame,D.J.,M.F.Wehner,I.Noy,andS.M.Rosier.2020.“TheEconomicCostsofHurricaneHarveyAttributabletoClimateChange.”ClimaticChange160(2):271–81.Franzke,C.L.E.2014.“NonlinearClimateChange.”NatureClimateChange4:423–24.Frumkin,H.,M.B.Das,M.Negev,B.C.Rogers,R.Bertollini,C.Dora,andS.Desai.2020.“ProtectingHealthinDryCities:ConsiderationsforPolicyMakers.”BMJ2020:371.Garrick,D.,L.DeStefano,L.Turley,I.Jorgensen,I.Aguilar-Barajas,B.Schreiner,R.deSouzaLeão,etal.2019.DividingtheWater,SharingtheBenefits:LessonsfromRural-to-UrbanWaterReallocation.Washington,DC:WorldBank.GCRO(GautengCity-RegionObservatory).2022.“NotEnoughintheRightPlace,TooMuchintheWrongPlace:UnderstandingandManagingtheGautengCity-Region’sCoupledLong-TermWaterChallenges.”Backgroundpaperpreparedforthisreport,WorldBank,Washington,DC.Gencer,E.,R.Folorunsho,M.Linkin,X.Wang,C.E.Natenzon,S.Wajih,N.Mani,etal.2018.“DisastersandRiskinCities.”InClimateChangeandCities:SecondAssessmentReportoftheUrbanClimateChangeResearchNetwork,editedbyC.Rosenzweig,W.Solecki,P.Romero-Lankao,S.Mehrotra,S.Dhakal,andS.AliIbrahim,61–98.NewYork:CambridgeUniversityPress.Gilmore,E.A.,C.Kousky,andT.St.Clair.2022“ClimateChangeWillIncreaseLocalGovernmentFiscalStressintheUnitedStates.”NatureClimateChange12:210–18.Glaeser,E.L.2012.TriumphoftheCity:HowOurGreatestInventionMadeUsRicher,Smarter,Greener,HealthierandHappier.NewYork:PenguinPress.Glaeser,E.L.2014.“AWorldofCities:TheCausesandConsequencesofUrbanizationinPoorerCountries.”JournaloftheEuropeanEconomicAssociation12:1154–99.Gollin,D.,M.Kirchberger,andD.Lagakos.2021.“DoUrbanWagePremiaReflectLowerAmenities?EvidencefromAfrica.”JournalofUrbanEconomics121:103301.Gomez,M.,A.Mejia,B.L.Ruddell,andR.R.Rushforth.2021.“SupplyChainDiversityBuffersCitiesagainstFoodShocks.”Nature595:250–54.GonzalvesG.S.,E.H.Kaplan,andA.D.Paltiel.2015.“ReducingSexualViolencebyIncreasingtheSupplyofToiletsinKhayelitsha,SouthAfrica:AMathematicalModel.”PLoSOne10(4):e0122244.GrafvonLuckner,C.,K.HolstonandC.Reinhart(2022,April15).“IsAnotherFoodCrisisUnfolding?Let’sTalkDevelopment.”https://blogs.worldbank.org/developmenttalk/another-food-crisis-unfoldingGraham,J.P.,M.Hirai,andS.S.Kim.2016.“AnAnalysisofWaterCollectionLaboramongWomenandChildrenin24Sub-SaharanAfricanCountries.”PLoSOne11(6):e0155981.Green,R.,L.Cornelsen,A.D.Dangour,R.Turner,B.Shankar,M.Mazzocchi,andR.D.Smith.2013.“TheEffectofRisingFoodPricesonFoodConsumption:SystematicReviewwithMeta-Regression.”BMJ346:f3703.Gregory,J.2017.“TheImpactofPost-KatrinaRebuildingGrantsontheResettlementChoicesofNewOrleansHomeowners.”Unpublishedmanuscript.TheImpactsofClimateandEnvironmentalChangeonCities217217TheImpactsofClimateandEnvironmentalChangeonCitiesGrip,I.L.,S.Haghighatafshar,andH.Aspegren.2021.“AMethodologyfortheAssessmentofCompoundSeaLevelandRainfallImpactonUrbanDrainageNetworksinaCoastalCityunderClimateChange.”CityandEnvironmentInteractions12:100074.Hallegatte,S.,M.Bangalore,L.Bonzanigo,M.Fay,T.Kane,U.Narloch,J.Rozenberg,etal.2016.“ShockWaves:ManagingtheImpactsofClimateChangeonPoverty.”Washington,DC:WorldBank.Hallegatte,S.,M.Bangalore,andA.Vogt-Schilb.2016a.“AssessingSocioeconomicResiliencetoFloodsin90Countries.”PolicyResearchWorkingPaper7663,WorldBank,Washington,DC.Hallegatte,S.,M.Bangalore,andA.Vogt-Schilb.2016b.“SocioeconomicResilience:Multi-HazardEstimatesin117Countries.”PolicyResearchWorkingPaper7886,WorldBank,Washington,DC.Hallegatte,S.,C.Green,R.J.Nicholls,andJ.Corfee-Morlot.2013.“FutureFloodLossesinMajorCoastalCities.”NatureClimateChange3:802–06.Hamilton,B.,N.C.Coops,andK.Lokman.2021.“TimeSeriesMonitoringofImperviousSurfacesandRunoffImpactsinMetroVancouver.”ScienceoftheTotalEnvironment760:143873.Harari,M.,andE.LaFerrara.2018.“Conflict,Climate,andCells:ADisaggregatedAnalysis.”ReviewofEconomicsandStatistics100(4):594–608.Harlan,S.L.,andD.M.Ruddell.2011.“ClimateChangeandHealthinCities:ImpactsofHeatandAirPollutionandPotentialCo-benefitsfromMitigationandAdaptation.”CurrentOpinioninEnvironmentalSustainability3(3):126–34.Hastings,J.S.,B.C.Madrian,andW.L.Skimmyhorn.2013.“FinancialLiteracy,FinancialEducation,andEconomicOutcomes.”AnnualReviewofEconomics5(1):347–73.He,C.,Z.Liu,J.Wu,X.Pan,Z.Fang,J.Li,andB.A.Bryan.2021.“FutureGlobalUrbanWaterScarcityandPotentialSolutions.”NatureCommunications12(1):1–11.He,Y.,S.Thies,P.Avner,andJ.Rentschler.2021.“FloodImpactsonUrbanTransitandAccessibility—ACaseStudyofKinshasa.”TransportationResearchPartD:TransportandEnvironment96:102889.He,Y.,B.Wu,P.He,W.Gu,andB.Liu.2021.“WindDisastersAdaptationinCitiesinaChangingClimate:ASystematicReview.”PLoSOne16(3):e0248503.Heblich,S.,A.Trew,andY.Zylberberg.2021.“East-SideStory:HistoricalPollutionandPersistentNeighborhoodSorting.”JournalofPoliticalEconomy129(5):1508–52.Henderson,J.V.,A.Storeygard,andU.Deichmann.2017.“HasClimateChangeDrivenUrbanizationinAfrica?”JournalofDevelopmentEconomics124:60–82.Hoekstra,A.Y.,J.Buurman,andK.C.VanGinkel.2018.“UrbanWaterSecurity:AReview.”EnvironmentalResearchLetters13(5):053002.Howson,A.2015.“ThirstyLimaUsesRobustPlanningtoAddressItsFutureWaterNeeds.”WorldBankNews,October5,2015.IDMC(InternalDisplacementMonitoringCentre).2018a.TheRippleEffect:EconomicImpactsofInternalDisplacement;MultidimensionalImpactsofInternalDisplacement.Geneva:IDMC.218THRIVINGTHRIVING218IDMC(InternalDisplacementMonitoringCentre).2018b.TheRippleEffect:EconomicImpactsofInternalDisplacement;ResearchAgendaandCallforPartners.Geneva:IDMC.IDMC(InternalDisplacementMonitoringCentre).2021.TheRippleEffect:EconomicImpactsofInternalDisplacement;UnveilingtheCostofInternalDisplacement.Geneva:IDMC.Indaco,A.,F.Ortega,andS.Taspinar.2021.“Hurricanes,FloodRiskandtheEconomicAdaptationofBusinesses.”JournalofEconomicGeography21(4):557–91.Inostroza,L.,M.Palme,andF.delaBarrera.2016.“AHeatVulnerabilityIndex:SpatialPatternsofExposure,SensitivityandAdaptiveCapacityforSantiagodeChile.”PLoSOne11(9):e0162464.IPCC(IntergovernmentalPanelonClimateChange).2018.“SummaryforPolicymakers.”InGlobalWarmingof1.5°C:AnIPCCSpecialReportontheImpactsofGlobalWarmingof1.5°CabovePre-IndustrialLevelsandRelatedGlobalGreenhouseGasEmissionPathways,intheContextofStrengtheningtheGlobalResponsetotheThreatofClimateChange,SustainableDevelopment,andEffortstoEradicatePoverty.Geneva,Switzerland:WorldMeteorologicalOrganization.IPCC(IntergovernmentalPanelonClimateChange).2019a.“SummaryforPolicymakers.”InClimateChangeandLand:AnIPCCSpecialReportonClimateChange,Desertification,LandDegradation,SustainableLandManagement,FoodSecurity,andGreenhouseGasFluxesinTerrestrialEcosystems.Geneva,Switzerland:WorldMeteorologicalOrganization.IPCC(IntergovernmentalPanelonClimateChange).2019b.“SummaryforPolicymakers.”InIPCCSpecialReportontheOceanandCryosphereinaChangingClimate.Cambridge,UKandNewYork:CambridgeUniversityPress.IPCC(IntergovernmentalPanelonClimateChange).2022.“SummaryforPolicymakers.”InClimateChange2022:Impacts,AdaptationandVulnerability.ContributionofWorkingGroupIItotheSixthAssessmentReportoftheIntergovernmentalPanelonClimateChange.Cambridge,UK:CambridgeUniversityPress.Ivanic,M.,W.Martin,andH.Zaman.2012.“EstimatingtheShort-RunPovertyImpactsofthe2010–11SurgeinFoodPrices.”WorldDevelopment40:2302–17.Jayachandran,S.2021.“SocialNormsasaBarriertoWomen’sEmploymentinDevelopingCountries.”IMFEconomicReview69(3):576–95.Jerch,R.,M.E.Kahn,andG.C.Lin.2021.“LocalPublicFinanceDynamicsandHurricaneShocks.”NBERWorkingPaper28050,NationalBureauofEconomicResearch,Cambridge,MA.Jha,A.K.,R.Bloch,andJ.Lamond.2012.CitiesandFlooding:AGuidetoIntegratedUrbanFloodRiskManagementforthe21stCentury.Washington,DC:WorldBank.Jiang,H.,andL.Quintero.2022.“Can’tStandtheHeat:ClimateStressonLaborProductivityin16LatinAmericanandCaribbeanCountries.”Backgroundpaperpreparedforthisreport,WorldBank,Washington,DC.Kaczan,D.J.,andJ.Orgill-Meyer.2020.“TheImpactofClimateChangeonMigration:ASynthesisofRecentEmpiricalInsights.”ClimaticChange158:281–300.Kahn,M.E.,K.Mohaddes,R.N.C.Ng,M.H.Pesaran,M.Raissi,andJ.-C.Yang.2021.“Long-TermMacroeconomicEffectsofClimateChange:ACross-CountryAnalysis.”EnergyEconomics104:105624.Kakissis,J.2010.“EnvironmentalRefugeesUnabletoReturnHome.”NewYorkTimes,January3,2010.http://www.nytimes.com/2010/01/04/world/asia/04migrants.html.TheImpactsofClimateandEnvironmentalChangeonCities219219TheImpactsofClimateandEnvironmentalChangeonCitiesKamei,A.2020.“WhoWalksforWater?WaterConsumptionandLaborSupplyResponsetoRainfallScarcityinUganda.”http://dx.doi.org/10.2139/ssrn.4013213.Karger,D.N.,S.Lange,C.Hari,C.P.O.Reyer,andN.E.Zimmermann.2021.CHELSA-W5E5v1.1:W5E5v1.0downscaledwithCHELSAv2.0.ISIMIPRepository.https://doi.org/10.48364/ISIMIP.836809.1.Karlan,D.,A.L.Ratan,andJ.Zinman.2014.“SavingsbyandforthePoor:AResearchReviewandAgenda.”ReviewofIncomeandWealth60(1):36–78.Kaya,Z.2018.“ResiliencePolicyandInternallyDisplacedWomeninIraq:AnUnintentionallyFlawedApproach.”WorkingPaperSeries13/2018,LSEWomen,PeaceandSecurity.Keener,S.,M.Luengo,andS.G.Banerjee.2010.“ProvisionofWatertothePoorinAfrica:ExperiencewithWaterStandpostsandtheInformalWaterSector.”PolicyResearchWorkingPaper5387,WorldBank,Washington,DC.Kelley,C.P.,S.Mohtadi,M.A.Cane,R.Seager,andY.Kushnir.2015.“ClimateChangeintheFertileCrescentandImplicationsoftheRecentSyrianDrought.”ProceedingsoftheNationalAcademyofSciences112(11):3241–46.Klasen,S.2019.“WhatExplainsUnevenFemaleLaborForceParticipationLevelsandTrendsinDevelopingCountries?”WorldBankResearchObserver34(2):161–97.Kleemans,M.,andJ.Magruder.2018.“LabourMarketResponsestoImmigration:EvidencefromInternalMigrationDrivenbyWeatherShocks.”EconomicJournal128(613):2032–65.Kocornik-Mina,A.,T.McDermott,G.Michaels,andF.Rauch.2019.“FloodedCities.”AmericanEconomicJournal:AppliedEconomics12(2):35–66.Komatsu,H.,andG.Joseph.2016.DrinkingWaterSalinity,BurdenofWaterCollectionandSchoolAttendanceofGirls:EvidencefromBangladeshandWestBengal.Washington,DC:WorldBank.Koolwal,G.,andD.VandeWalle.2013.“AccesstoWater,Women’sWorkandChildOutcomes.”EconomicDevelopmentandCulturalChange61(2):369–405.Kousky,C.,E.Luttmer,andR.Zeckhauser.2006.“PrivateInvestmentandGovernmentProtection.”JournalofRiskandUncertainty33(1):73–100.Kovacs,P.,andJ.Thistlethwaite.2014.“Industry.”InCanadainaChangingClimate:SectorPerspectivesonImpactsandAdaptation,editedbyF.J.WarrenandD.S.Lemmen,135–58.Ottawa,ON:GovernmentofCanada.Kulp,S.A.,andB.H.Strauss.2019.“NewElevationDataTripleEstimatesofGlobalVulnerabilitytoSea-LevelRiseandCoastalFlooding.”NatureCommunications10(1):1–12.Kydland,F.E.,andE.C.Prescott.1977.“RulesRatherthanDiscretion:TheInconsistencyofOptimalPlans.”JournalofPoliticalEconomy85(3):473–92.Lagakos,D.2020.“Urban-RuralGapsintheDevelopingWorld:DoesInternalMigrationOfferOpportunities?”JournalofEconomicPerspectives34(3):174–92.Lall,S.V.,andU.Deichmann.2012.“DensityandDisasters:EconomicsofUrbanHazardRisk.”WorldBankResearchObserver27(1):74–105.Letta,M.,andR.S.J.Tol.2019.“Weather,ClimateandTotalFactorProductivity.”EnvironmentalandResourceEconomics73(1):283–305.220THRIVINGTHRIVING220Li.,C.-J.,Y-Q.Chai,L.-S.Yang,andH.-R.Li.2016.“Spatio-TemporalDistributionofFloodDisastersandAnalysisofInfluencingFactorsinAfrica.”NaturalHazards82:721–31.Lombraña,L.M.,andS.Dodge.2021.“WhateverClimateChangeDoestotheWorld,CitiesWillBeHitHardest.”Bloomberg,April19,2021.https://www.bloomberg.com/graphics/2021-cities-climate-victims/.Lu,X.,L.Bengtsson,andP.Holme.2012.“PredictabilityofPopulationDisplacementafterthe2010HaitiEarthquake.”ProceedingsoftheNationalAcademyofSciences109(29):11576–81.Lu,X.,D.J.Wrathall,P.R.Sundsøy,M.Nadiruzzaman,E.Wetter,A.Iqbal,andL.Bengtsson.2016.“UnveilingHiddenMigrationandMobilityPatternsinClimateStressedRegions:ALongitudinalStudyofSixMillionAnonymousMobilePhoneUsersinBangladesh.”GlobalEnvironmentalChange38:1–7.Lusardi,A.,andO.S.Mitchell.2014.“TheEconomicImportanceofFinancialLiteracy:TheoryandEvidence.”JournalofEconomicLiterature52(1):5–44.Mach,K.J.,W.N.Adger,H.Buhaug,M.Burke,J.D.Fearon,C.B.Field,C.S.Hendrix,etal.2020.“DirectionsforResearchonClimateandConflict.”Earth’sFuture8(7):e2020EF001532.Madrigano,J.,K.Ito,S.Johnson,P.L.Kinney,andT.Matte.2015.“ACase-OnlyStudyofVulnerabilitytoHeatWave–RelatedMortalityinNewYorkCity(2000–2011).”EnvironmentalHealthPerspectives123(7):672–78.Marx,B.,T.Stoker,andT.Suri.2013.“TheEconomicsofSlumsintheDevelopingWorld.”JournalofEconomicPerspectives27(4):187–210.Mashhoodi,B.2021.“EnvironmentalJusticeandSurfaceTemperature:Income,Ethnic,Gender,andAgeInequalities.”SustainableCitiesandSociety68(February):102810.Maurel,M.,andM.Tuccio,2016.“ClimateInstability,UrbanisationandInternationalMigration.”JournalofDevelopmentStudies52:735–52.McDonald,R.I.,K.Weber,J.Padowski,M.Flörke,C.Schneider,P.A.Green,T.Gleeson,etal.2014.“WateronanUrbanPlanet:UrbanizationandtheReachofUrbanWaterInfrastructure.”GlobalEnvironmentalChange:HumanandPolicyDimensions27(July):96–105.McGuirk,E.F.,andN.Nunn.2021.“TranshumantPastoralism,ClimateChange,andConflictinAfrica.”NBERWorkingPaper28243,NationalBureauofEconomicResearch,Cambridge,MA.McNamara,K.E.,L.L.Olson,andM.A.Rahman.2016.“InsecureHope:TheChallengesFacedbyUrbanSlumDwellersinBholaSlum,Bangladesh.”MigrationandDevelopment5(1):1–15.Meijer,J.R.,M.A.J.Huijbregts,C.G.J.Schotten,andA.M.Schipper.2018.“GlobalPatternsofCurrentandFutureRoadInfrastructure.”EnvironmentalResearchLetters13:064006.Mitchell,T.,T.Tanner,andK.Lussier.2007.WeKnowWhatWeNeed:SouthAsianWomenSpeakOutonClimateChangeAdaptation.Brighton,UK:ActionAidandInstituteofDevelopmentStudiesattheUniversityofSussex.Mitlin,D.,V.A.Beard,D.Satterthwaite,andJ.Du.2019.“UnaffordableandUndrinkable:RethinkingUrbanWaterAccessintheGlobalSouth.”WorldResourcesInstitute,Washington,DC.Moctar,N.,M.d’HôtelElodie,andL.C.Tristan.2015.“MaizePriceVolatility:DoesMarketRemotenessMatter?”PolicyResearchWorkingPaper7202,WorldBank,Washington,DC.TheImpactsofClimateandEnvironmentalChangeonCities221221TheImpactsofClimateandEnvironmentalChangeonCitiesMoraci,F.,M.F.Errigo,C.Fazia,T.Campisi,andF.Castelli.2020.“CitiesunderPressure:StrategiesandToolstoFaceClimateChangeandPandemic.”Sustainability12(18):7743.Moretti,E.2013.“RealWageInequality.”AmericanEconomicJournal:AppliedEconomics5(1):65–103.Mukim,M.2021.BurundiUrbanizationReview.Washington,DC:WorldBank.Nakagawa,M.,M.Saito,andH.Yamaga.2007.“EarthquakeRiskandHousingRents:EvidencefromtheTokyoMetropolitanArea.”RegionalScienceandUrbanEconomics37(1):87–99.Nakamura,S.,T.Bundervoet,andM.Nuru.2019.“RuralRoads,Poverty,andResilienceEvidencefromEthiopia.”PolicyResearchWorkingPaper8800,WorldBank,Washington,DC.Narloch,U.G.,andM.Bangalore.2016.“EnvironmentalRisksandPoverty:AnalyzingGeo-spatialandHouseholdDatafromVietnam.”PolicyResearchWorkingPaper7763,WorldBank,Washington,DC.Nauges,C.,andJ.Strand.2017.“WaterHaulingandGirls’SchoolAttendance:SomeNewEvidencefromGhana.”EnvironmentalandResourceEconomics66(1):65–88.Nelson,A.,D.J.Weiss,J.vanEtten,A.Cattaneo,T.S.McMenomy,andJ.Koo.2019.“ASuiteofGlobalAccessibilityIndicators.”ScientificData6(1):1–9.Nelson,V.,K.Meadows,T.Cannon,J.Morton,andA.Martin.2002.“UncertainPredictions,InvisibleImpacts,andtheNeedtoMainstreamGenderinClimateChangeAdaptations.”GenderandDevelopment10(2):51–59.Okunola,O.2019.“SpatialAnalysisofDisasterStatisticsinSelectedCitiesofNigeria.”InternationalJournalofEmergencyManagement15(4):299–315.Orlov,A.,J.Sillmann,K.Aunan,T.Kjellstrom,andA.Aaheim.2020.“EconomicCostsofHeat-InducedReductionsinWorkerProductivityduetoGlobalWarming.”GlobalEnvironmentalChange63:102087.Park,J.,M.Bangalore,S.Hallegatte,andE.Sandhoefner.2018.“HouseholdsandHeatStress:EstimatingtheDistributionalConsequencesofClimateChange.”EnvironmentandDevelopmentEconomics23(3):349–68.Park,R.J.,J.Goodman,M.Hurwitz,andJ.Smith.2020.“HeatandLearning.”AmericanEconomicJournal:EconomicPolicy12(2):306–39.Parsons,K.2014.HumanThermalEnvironments.TheEffectsofHot,Moderate,andColdEnvironmentonHumanHealth,Comfort,andPerformance,3ded.BocaRaton,FL:CRCPress.Patankar,A.,andA.Patwardhan.2016.“EstimatingtheUninsuredLossesduetoExtremeWeatherEventsandImplicationsforInformalSectorVulnerability:ACaseStudyofMumbai,India.”NaturalHazards80(1):285–310.Patankar,A.M.2015.“TheExposure,Vulnerability,andAbilitytoRespondofPoorHouseholdstoRecurrentFloodsinMumbai.”PolicyResearchWorkingPaper7481,WorldBank,Washington,DC.Pesaresi,M.,A.Florczyk,M.Schiavina,M.Melchiorri,andL.Maffenini.2019.“GHSSettlementGrid,UpdatedandRefinedREGIOModel2014inApplicationtoGHS-BUILTR2018AandGHS-POPR2019A,Multitemporal(1975-1990-2000-2015),R2019A.”JointResearchCentre,EuropeanCommission,Brussels.222THRIVINGTHRIVING222Philip,S.,S.F.Kew,G.J.vanOldenborgh,E.Aalbers,R.Vautard,F.Otto,K.Haustein,etal.2018a.“ValidationofaRapidAttributionoftheMay/June2016Flood-InducingPrecipitationinFrancetoClimateChange.”JournalofHydrometeorology19(11):1881–98.Philip,S.,S.F.Kew,G.J.vanOldenborgh,F.Otto,S.O’Keefe,K.Haustein,A.King,etal.2018b.“AttributionAnalysisoftheEthiopianDroughtof2015.”JournalofClimate31(6):2465–86.Phillips,C.A.,A.Caldas,R.Cleetus,K.A.Dahl,J.Declet-Barreto,R.Licker,L.D.Merner,etal.2020.“CompoundClimateRisksintheCOVID-19Pandemic.”NatureClimateChange10:586–88.Qin,Y.2020.“UrbanFloodingMitigationTechniques:ASystematicReviewandFutureStudies.”Water12(12):3579.Quintero,L.,andM.Roberts.2022.“CitiesandProductivity:Evidencefrom16LatinAmericaandCaribbeanCountries.”Unpublishedmanuscript.Radeloff,V.C.,D.P.Helmers,H.AnuKramer,M.H.Mockrin,P.M.Alexandre,A.Bar-Massada,V.Butsic,etal.2018.“RapidGrowthoftheUSWildland-UrbanInterfaceRaisesWildfireRisk.”ProceedingsoftheNationalAcademyofSciences115(13):3314–19.Rebally,A.,C.Valeo,J.He,andS.Saidi.2021.“FloodImpactAssessmentsonTransportationNetworks:AReviewofMethodsandAssociatedTemporalandSpatialScales.”FrontiersinSustainableCities3:10.3389/frsc.2021.732181.Reed,K.A.,A.M.Stansfield,M.F.Wehner,andC.M.Zarzycki.2020.“ForecastedAttributionoftheHumanInfluenceonHurricaneFlorence.”ScienceAdvances6(1):eaaw9253.Rentschler,J.,P.Avner,M.Marconcini,R.Su,E.Strano,S.Hallegatte,L.Bernard,andC.Riom.2022.“RapidUrbanGrowthinFloodZones:GlobalEvidencesince1985.”PolicyResearchWorkingPaper10014,WorldBank,Washington,DC.Rentschler,J.,andM.Salhab.2020.“PeopleinHarm’sWay:FloodExposureandPovertyin189Countries.”PolicyResearchWorkingPaper9447,WorldBank,Washington,DC.RestrepoCadavid,P.2011.“TheImpactsofSlumPoliciesonHouseholds’Welfare:TheCaseofMedellin(Colombia)andMumbai(India).”EconomicsandFinance.ÉcoleNationaleSupérieuredesMines,Paris.Ritter,K.2018.“WaterAccessinLimaComplicatedbyInequalityandClimateUncertainty.”CircleofBlue(blog),June28,2018.https://www.circleofblue.org/2018/south-america/water-access-in-lima-complicated-by-inequality-and-climate-uncertainty/.Roberts,M.,F.GilSander,andS.Tiwari,eds.2019.TimetoACT:RealizingIndonesia’sUrbanPotential.Washington,DC:WorldBank.Ross,M.R.,J.J.O’Brien,R.G.Ford,K.Zhang,andA.Morkill.2009.“DisturbanceandtheRisingTide:TheChallengeofBiodiversityManagementonLow-IslandEcosystems.”FrontiersinEcologyandtheEnvironment7(9):471–78.Rossitti,G.2022.“WhoIsAffected?WhichSegmentsofCityPopulationsAreMostExposedtoClimate-RelatedStresses?”Backgroundpaperpreparedforthisreport,WorldBank,Washington,DC.Sampson,C.C.,A.M.Smith,P.D.Bates,J.C.Neal,L.Alfieri,andJ.E.Freer.2015.“AHigh-ResolutionGlobalFloodHazardModel.”WaterResourcesResearch51(9):7358–81.TheImpactsofClimateandEnvironmentalChangeonCities223223TheImpactsofClimateandEnvironmentalChangeonCitiesSantamouris,M.,andG.Y.Yun.2020.“RecentDevelopmentandResearchPrioritiesonCoolandSuperCoolMaterialstoMitigateUrbanHeatIsland.”RenewableEnergy161:792–807.Sedova,B.,andM.Kalkuhl.2020.“WhoAretheClimateMigrantsandWhereDoTheyGo?EvidencefromRuralIndia.”WorldDevelopment129(C):104848.Setiawan,I.,S.Tiwari,andH.Rizal.2018.“EconomicandSocialMobilityinUrbanizingIndonesia.”BackgroundpaperpreparedforTimetoACT:RealizingIndonesia’sUrbanPotential,editedbyM.Roberts,F.GilSander,andS.Tiwari.WorldBank,Washington,DC.Shahmohamadi,P.,A.I.Che-Ani,I.Etessam,K.N.A.Maulud,andN.M.Tawil.2011.“HealthyEnvironment:TheNeedtoMitigateUrbanHeatIslandEffectsonHumanHealth.”ProcediaEngineering20(January):61–70.Shayegh,S.,V.Manoussi,andS.Dasgupta.2021.“ClimateChangeandDevelopmentinSouthAfrica:TheImpactofRisingTemperaturesonEconomicProductivityandLabourAvailability.”ClimateandDevelopment13(8):725–35.Shively,G.,andG.Thapa.2016.“Markets,TransportationInfrastructure,andFoodPricesinNepal.”AmericanJournalofAgriculturalEconomics99:660–82.Shughrue,C.,B.Werner,andK.C.Seto.2020.“GlobalSpreadofLocalCycloneDamagesthroughUrbanTradeNetworks.”NatureSustainability3:606–13.Sicard,P.,E.Agathokleous,A.DeMarco,E.Paoletti,andV.Calatayud.2021.“UrbanPopulationExposuretoAirPollutioninEuropeovertheLastDecades.”EnvironmentalSciencesEurope33(1):1–12.Smith,E.T.,andS.C.Sheridan.2019.“TheInfluenceofExtremeColdEventsonMortalityintheUnitedStates.”ScienceoftheTotalEnvironment647:342–51.Somanathan,E.,R.Somanathan,A.Sudarshan,andM.Tewari.2021.“TheImpactofTemperatureonProductivityandLaborSupply:EvidencefromIndianManufacturing.”JournalofPoliticalEconomy129(6):1797–1827.Souverijns,N.,K.DeRidder,N.Veldeman,F.Lefebre,F.,Kusambiza-Kiingi,W.Memela,andN.Jones.2022.“UrbanHeatinJohannesburgandEkurhuleni,South-Africa:AMeter-ScaleAssessmentandVulnerabilityAssessment.”SubmittedtoUrbanClimate.Thomson,R.,R.Hornigold,L.Page,andT.Waite.2018.“AssociationsbetweenHighAmbientTemperaturesandHeatWaveswithMentalHealthOutcomes:ASystematicReview.”PublicHealth161:171–91.UCCRN(UrbanClimateChangeResearchNetwork).2018.TheFutureWeDon’tWant:HowClimateChangeCouldImpacttheWorld’sGreatestCities.UrbanClimateChangeResearchNetworkTechnicalReport.NewYork:ColumbiaUniversity.UN(UnitedNations).2018.“IndigenousPeoplesandEthnicMinorities:MarginalizationIstheNorm.”InPromotingInclusionthroughSocialProtection:ReportontheWorldSocialSituation2018.NewYork:UnitedNations.UNOCHA(UnitedNationsOfficefortheCoordinationofHumanitarianAffairs).2019.“Gender-BasedViolence:FinancialIndependenceandEconomicEmpowermentKeytoSurvivors’Recovery.”UNOCHANewsandUpdates,May17,2019.https://www.unocha.org/story/gender-based-violence-financial-independence-and-economic-empowerment-key-survivors-recovery.224THRIVINGTHRIVING224UNESCO(UnitedNationsEducational,ScientificandCulturalOrganization).2015.GenderandWaterData:ProjectforGenderSensitiveWaterMonitoring,AssessmentandReporting.Perugia:UnitedNationsWorldWaterAssessmentProgramme.UNICEF(UnitedNationsChildren’sFund).2016.“CollectingWaterIsOftenaColossalWasteofTimeforWomenandGirls.”Pressrelease,August29,2016.https://www.unicef.org/press-releases/unicef-collecting-water-often-colossal-waste-time-women-and-girls.vanOldenborgh,G.J.,F.Krikken,S.Lewis,N.J.Leach,F.Lehner,K.R.Saunders,M.vanWeele,etal.2021.“AttributionoftheAustralianBushfireRisktoAnthropogenicClimateChange.”NaturalHazardsEarthSystemScience21(3):941–60.Vautard,R.,M.vanAalst,O.Boucher,A.Drouin,K.Haustein,F.Kreienkamp,G.J.vanOldenborgh,etal.2020.“HumanContributiontotheRecord-BreakingJuneandJuly2019HeatwavesinWesternEurope.”EnvironmentalResearchLetters15(9):094077.Venkat,A.,F.Dizon,andW.Masters.2022.“TheImpactofClimateShocksandStressesonUrbanFoodPrices.”Backgroundpaperpreparedforthisreport,WorldBank,Washington,DC.Voelkel,J.,D.Hellman,R.Sakuma,andV.Shandas.2018.“AssessingVulnerabilitytoUrbanHeat:AStudyofDisproportionateHeatExposureandAccesstoRefugebySocio-DemographicStatusinPortland,Oregon.”InternationalJournalofEnvironmentalResearchandPublicHealth15(4):640.Votsis,A.,andA.Perrels.2016.“HousingPricesandthePublicDisclosureofFloodRisk:ADifference-in-DifferencesAnalysisinFinland.”JournalofRealEstateFinanceandEconomics53(4):450–71.Wang,J.,Y.Chen,W.Liao,G.He,S.F.B.Tett,Z.Yan,P.Zhai,etal.2021.“AnthropogenicEmissionsandUrbanizationIncreaseRiskofCompoundHotExtremesinCities.”NatureClimateChange11(12):1084–89.Willinger,B.A.,ed.2008.KatrinaandtheWomenofNewOrleans.NewOrleans,LA:NewcombCollegeCenterforResearchonWomen,TulaneUniversity.Wilson,R.,E.Z.Erbach-Schoenberg,M.Albert,D.Power,S.Tudge,M.Gonzalez,S.Guthrie,etal.2016.“RapidandNearReal-TimeAssessmentsofPopulationDisplacementUsingMobilePhoneDataFollowingDisasters:The2015NepalEarthquake.”PloSCurrents,February24,2016.https://pubmed.ncbi.nlm.nih.gov/26981327/.Winsemius,H.C.,B.Jongman,T.I.Veldkamp,S.Hallegatte,M.Bangalore,andP.J.Ward.2018.“DisasterRisk,ClimateChange,andPoverty:AssessingtheGlobalExposureofPoorPeopletoFloodsandDroughts.”EnvironmentandDevelopmentEconomics23(3):328–48.WorldBank.2010.NaturalHazards,UnNaturalDisasters:TheEconomicsofEffectivePrevention.Washington,DC:WorldBank.WorldBank.2015.IndigenousLatinAmericaintheTwenty-FirstCentury.Washington,DC:WorldBank.WorldBank.2017.MaintainingtheMomentumwhileAddressingServiceQualityandEquity:ADiagnosticofWaterSupply,Sanitation,Hygiene,andPovertyinEthiopia.WASHPovertyDiagnostic.Washington,DC:WorldBank.TheImpactsofClimateandEnvironmentalChangeonCities225225TheImpactsofClimateandEnvironmentalChangeonCitiesWorldBank.2019.WomeninWaterUtilities:BreakingBarriers.Washington,DC:WorldBank.WorldBank.2020a.ForcedDisplacement:AnAgendaforCitiesandTowns.Washington,DC:WorldBank.WorldBank.2020b.PublicExpenditureReview:DisasterResponseandRehabilitationinthePhilippines.Washington,DC:WorldBank.WorldBank.2020c.TheGlobalHealthCostofAmbientPM2.5AirPollution.Washington,DC:WorldBank.WorldBank.2021a.“AdaptingtoSeaLevelRiseintheMarshallIslands.”StoryMaps,October22,2021.https://storymaps.arcgis.com/stories/8c715dcc5781421ebff46f35ef34a04d.WorldBank.2021b.LearningfromTropicalCycloneSeroja:BuildingDisasterandClimateResilienceinTimor-Leste.Washington,DC:WorldBank.WorldBank.2021c.SomaliaUrbanizationReview:FosteringCitiesasAnchorsofDevelopment.Washington,DC:WorldBank.WorldBank.2021d.StrengtheningtheLinkbetweenEconomicGrowthandPovertyReductioninMali.APovertyAssessment.Washington,DC:WorldBank.WorldBank.2021e.ThePhilippinesParametricCatastropheRiskInsuranceProgramPilot.Washington,DC:WorldBank.WorldBank.2021f.WorldDevelopmentReport2021:DataforBetterLives.Washington,DC:WorldBank.WorldBank.Forthcominga.HealthyCities.Washington,DC:WorldBank.WorldBank.Forthcomingb.SouthAfricaDisasterRiskFinancingDiagnostic.Washington,DC:WorldBank.WorldBankandAustralianAID(AgencyforInternationalDevelopment).2014.“WhereAreWeDuringFlooding?AQuantitativeAssessmentofPovertyandSocialImpactsofFloodinginSelectedNeighborhoodsofHCMC.”WorldBank,Washington,DC.WRI(WorldResourcesInstitute).2017.“DroughtsandBlackouts:HowWaterShortagesCostIndiaEnoughEnergytoPowerSriLanka.”WorldResourcesInstituteBlog,July26,2017.https://www.wri.org/insights/droughts-and-blackouts-how-water-shortages-cost-india-enough-energy-power-sri-lanka.Yabe,T.,N.K.Jones,P.S.C.Rao,M.C.Gonzalez,andS.V.Ukkusuri.2022.“MobilePhoneLocationDataforDisasters:AReviewfromNaturalHazardsandEpidemics.”Computers,EnvironmentandUrbanSystems94:101777.Zaveri,E.,J.Russ,A.Khan,R.Damania,E.Borgomeo,andA.Jagerskog.2021.EbbandFlowVol.1:Water,MigrationandDevelopment.Washington,DC:WorldBank.Zscheischler,J.,S.Westra,B.J.J.M.VanDenHurk,S.I.Seneviratne,P.J.Ward,A.Pitman,A.AghaKouchak,etal.2018.“FutureClimateRiskfromCompoundEvents.”NatureClimateChange8(6):469–77.227TheImpactofCitiesonClimateandtheEnvironmentCHAPTER4Citygrowthhascausedclimatechange,butthatgrowthisalsowhat’sgoingtogetusoutofit.MatthewE.Kahn,ProvostProfessorofEconomicsandSpatialSciencesUniversityofSouthernCaliforniaTheRoutledgeHandbookonSpacesofUrbanPolitics(Wardetal.2018)••Citiesgrowalongthreemargins:horizontalexpansion,infilldevelopment,andverticallayering.Therelativeimportanceofeachvarieswithbothacity’slevelofincomeandthecostsofverticaldevelopment.••Whentoday’sadvancedcountriesbegantodevelopandurbanize,theyfacedprohibi-tivelyhightechnologicalcostsofverticaldevelopment.Theelevatorandothermodernbuildingtechnologieshaveloweredsuchcostsforcitiesintoday’slower-incomecoun-tries.Yetcitiesinthesecountriestendtoexpandoutwardratherthanupwardbecauseofthecapital-intensivenatureofverticalconstruction,failingsinlandandpropertymarkets,andregulationsthatconstraindevelopersfrombuildingupward.••Previousworkhaspointedtoacity’sproductivitylevelasthemaindriverofwhetheritgrowsoutwardorupward,andtheorizedthatcitiesthatbuildupwardalsoexpandfasteroutward.Newresearchundertakenforthisreportfindsinsteadthatmoreverticaldevelopmentresultsinbothamorecompact,environmentallyfriendlyurbanformandhigherincomes;however,lowercarbondioxideemissionsassociatedwithamorecompacturbanformmustbeweighedagainstthehigheremissionsembeddedintheconstructionoftallerbuildings.••Inadditiontotheirindirectimpactsthroughacity’surbanform,incentivesandinvestmentsthatencouragetheuseofpublictransportationandreducedependenceonprivatevehiclescanalsodirectlycontributetobetterairquality.Theeffectsare,however,highlyheterogeneousacrosscitiesglobally,andpoorlydesignedpublictrans-portationsystemsmaycontributetomoretrafficcongestion,leadingtolower-qualityairandhighergreenhousegasemissions.••Ifnotwellmanaged,urbanexpansionthreatensfertileagriculturalland—withnegativerepercussionsforurbanfoodsystemsaswellasforwaterpollutiondownstreamofcities.Measurestoenhanceagriculturalproductivityandreducefoodlossandwastecanhelpoffsettheadverseimpactsofurbanexpansiononfoodproduction,whilecontributingtoreducedgreenhousegasemissions.MAINFINDINGS228THRIVINGIntroductionAshighlightedinchapter1,citiesareresponsiblefor70percentofglobalanthropogenicgreen-housegasemissions(Hopkinsetal.2016).In2019,about86percentoftheworld’surban-iteslivedincitieswithdangerouslyhighairpollutionlevels,whichcontributedto1.8millionexcessdeathsworldwide(Southerlandetal.2022).InAsia,urbanexpansionhasbeenacentraldriverofcroplandlossand,accordingtoonerecentreport,nowposes“anexistentialthreattoperi-urbanagriculture—aleadingsourceofnutritionallyimportantfreshfruitandvegetablesformanycities”(Acharyaetal.2021,xviii).Moreover,citiesfrequentlycompetewithagricul-turallandsforwater(Garricketal.2019),andfarmlanddownstreamofcitiesoftenmustrelyoncontaminatedurbanwastewaterforirrigation(Damaniaetal.2019).Althoughitisdifficulttodenythenegativeenvironmentalimpactsofmanycities,itisalsoimportanttobeawareoftheirenvironmentalbenefits.Cities,afterall,concentratemorethan50percentoftheworld’spopulationintolessthan1percentofitslandarea.1Econometricestimatesindicatethathigherurbandensityisassociatedwithlesscommutingandcaruseandlowerlevelsofdomesticenergyconsumption(AhlfeldtandPietrostefani2019).Chapter1ofthisreportprovidedevidenceofastrikingnegativecorrelationbetweenthecompactnessofacity’sdevelopmentanditslevelsofcarbondioxide(CO2)emissionsandemissionsofparticu-latematterof2.5micronsorlessindiameter(PM2.5)inboththetransportationandresidentialsectors.Manyfactorscontributetodeterminingtheimpactofacity’sdevelopmentontheenviron-ment,including,forexample,practicesofsolidwastemanagement,theenergyefficiencyofitsbuildings,andthelocaluptakeofrenewableformsofenergy.Althoughactioninalltheseareasisundoubtedlyimportanttothemitigationofclimatechange,thischaptermainlyfocusesonanareainwhichitcanprovidenewevidenceandinsights:theroleplayedbytheformthatacity’sdevelopmenttakesand,inparticular,whetheracityaccommodatespopulationgrowthbyexpandingoutwardlikeapancakeorupwardlikeapyramid(Lalletal.2021).Thetendencytogrowoutwardorupwarddependsonhowwellurbanpolicymakerscanmanagethestressesassociatedwithacity’sgrowth.Thesestressesputpressureonlandandpropertypricesincentralareas,whilealsounderminingthelivabilityofthoseareas,therebypushingbothpeopleandfirmsoutwardtoacity’speriphery(EllisandRoberts2016).Thischapterthereforehasthefollowingmainaims:••Investigatehowtheformofacity’sdevelopment—especiallyhowverticalthatdevelopmentis—affectsbothitsconsumptionofland(andthusthecompactnessofitsform)anditslevelofproductivityandthereforeprosperity.••Reviewtheevidenceontheimpactsonairpollutionofincentivesandinvestmentsthataffecttheuseofdifferentmodesofurbantransportationandthatalsoplayakeyroleinshapingtheevolutionofacity’sform.••Examineprojectionsofhowurbanexpansionwillaffectagriculturallandandproductionundervariouspolicyscenarios.••Discusshowurbanexpansionaffectswaterpollutiondownstreamofcitiesandcompetitionforwaterbetweenurbanandruralareas.Toachievethefirstoftheseaims,thechapterpresentstheresultsofempiricalanalysisunder-takenforthisreportonthecausalimpactsofverticaldevelopmentonboththeareasandpro-ductivitylevelsofcitiesusingthereport’sglobalsampleofmorethan10,000cities(AhlfeldtandJedwab2022).Itcombinesthiswork,whichdrawsonauniqueglobaldatabaseofalmost700,000tallbuildings,withinsightsfromsecondaryliteraturetoderivethechapter’smaininsights.TheImpactofCitiesonClimateandtheEnvironment229Pancakesorpyramids—Howdocitiesaroundtheworldevolveinform?ModerncitiescandevelopalongthreemarginsAmodern-daycitycandevelopalongthreemargins:horizontalexpansion,infilldevelopment,andverticallayering(Lalletal.2021).Horizontalexpansionoftenmanifestsitselfaslow-densitysprawl;infilldevelopmenttakestheformofinwardadditionsofbuilt-upareainthegapsleftbetweenexistingstructures;andverticallayeringconsistsofbuildingtallerresidentialandcommercialproperties.Althoughnofirmheightdefinitionsexist,thesetallerpropertiesmayconsistofmidrisebuildings(6–14stories),high-risebuildings(15–24stories),orskyscrapers(25storiesormore).Historically,citiescoulddevelopalongonlytwoofthesemargins—horizontalexpansionandinfilldevelopment.Thislimitationwasdue,inlargepart,totheabsenceofasafe,fast,andconvenientmeansofmovingbetweenthefloorsofabuilding,whichmeantthattheheightofbuildingswasconstrainedbythewillingnessandabilityofpeopletowalkupflightsofstairs.Until1884,whentheheightofUlmMinsterinGermanywassurpassedbythatoftheWashingtonMonument,thetalleststructuresintheworldwerecathedrals(figure4.1).Withtheintroductionofthepassengerelevator,however,thepossibilityofsignificantverticaldevelopmentbeyondafewstoriesbecameareality.2Evenso,theelevatorgotofftoaslowstart.Thefirstpassengerelevator,installedatthefive-storyHaughwoutDepartmentStoreinFigure4.1Evolutionoftallbuildingsandstructuressince1880Source:WorldBankelaborationbasedondatafromEmporisandWorldEconomicForum(https://www.weforum.org/agenda/2019/09/tallest-historical-structures).0100200300400500600700800188018901900191019201930194019501960197019801990200020102020Height(meters)TowersandbroadcastmastsChurchesBuildingsPetronasTowersUlmMinsterTaipei101PhiladelphiaCityHallWashingtonMonumentEielTowerEmpireStateBuildingBurjKhalifaWillisTower230THRIVINGNewYorkin1857,waspoweredbyasteamengineand,comparedtotoday’sfastestelevatorsthatcanascendatspeedsupwardof40feetpersecond,traveledatasnail’spaceof40feetperminute.Afterjustthreeyears,itceasedoperationsbecausethestore’scustomersrefusedtouseit.Onlyafterpassageofanotherdecadeandcompletionofthe130-foot,eight-storyHomeInsuranceBuildingindowntownManhattanwastheelevatorincorporatedintothedesignofanofficebuilding.However,itwasnotuntiltheopeningin1885oftheworld’s“firstskyscraper”—the10-storyHomeInsuranceBuildinginChicago—thattheelevatorbecameamainstayofarchitecturalofficebuildingdesign(Prisco2019).3Seebox4.1forabriefhistoryoftheelevator.Abriefhistoryoftheelevatoror“verticalrailway”Althoughanunremarkablefeatureoftoday’smanytallbuildings,elevatorsare,infact,arevolutionarymeansofmasstransportationthatallowedtheemergenceofmidrises,high-rises,andskyscrapers(Gottmann1966).Theinventionofthemodernelevatoropenedupanewfrontierofurbandevelopment,allowingcitiestogrowverticallybeyondthelimitsoftheirscarceusableland.Thefirstpatentforaso-calledverticalrailwaywasfiledin1859bytheengineerOtisTuft.Thewell-knownOtisElevatorCompany,however,takesitsnamefromAmericanindustrialistElishaOtis.Heheldspectaculardemonstrationsatthe1853World’sFairinNewYorkatwhichhecuttheelevatorropetoshowcasehisinvention,thesafetybrake,whichwouldstoptheelevatorcarfromcrashingtotheground.ThesteamengineandOtis’ssafetybrakecreatedtheconditionsforthetakeoffofelevatorsattheendofthenineteenthcentury.Becausethefirstmodelswereextremelyslowandexpen-sive,however,theywerenotwidelyadopted.Amongthefirstuserswereluxuryhotels,whichfeaturedrichlydecoratedcarsthatofferedanexperienceratherthanatransportationservice.Theevolutionfromsteamengine–poweredmachinestohydraulicones,andlatertoelectricengines,increasedthespeedofelevatorsandsoledtomorewidespreadadoption.Manhattan’seight-storyEquitableLifeBuilding,completedin1870,wasthefirstcom-mercialbuildingdesignedwithanelevator.Chicago’sHomeInsuranceBuildinghadfourelevators.Inthe1910sand1920s,astoweraftertowerbegantospringupintheNewYorkskyline,manywerefittedwithOtis’selevators.TheuptakeinEurope,however,wasslower,inpartbecauseofbuildingheightrestrictions.The1920salsobroughtthefirstpenthousesandthebeginningofaculturalswitch.Beforeelevators,theupperfloorsofbuildingsoftencamewiththelowestrentsbecauseoftheirinaccessibility;afterelevators,upperfloorscarriedlargepricepremiumsbecauseoftheirgreaterprivacyandbetterviews(Wongetal.2011).Elevatorscontinuetoevolvetoday.Intheworld’stallestbuildings,theyareanintegralpartoftheadvanceddesigns.Formidrises,elevatorsaddsignificantlytothecostofbuilding,buttheirmarginalcostisclosetozeroabovetheseventhfloor.aThenearfuturemayseetheevolutionfromelectricity-poweredelevatorstomagneticones,whichtakeuplessspaceandwillfurtherimprovethelanduseefficiencyoftallbuildings.Sources:Bernard2014;Glaeser2011;Prisco2019.a.Theworld’stallestskyscrapers,however,requiremultiplecostlysetsofelevatorstocovertheirheight.InSaudiArabia,the167-storyJeddahTower,whoseconstructioniscurrentlyonhold,requiresverycomplexengineeringsolutionsandcostliermaterials.Box4.1TheImpactofCitiesonClimateandtheEnvironment231Bydramaticallyreducingthecostsofverticaltransportationwithinabuilding,thedevelopmentofthemodernpassengerelevator,aidedbyotheradvancesintallbuildingconstruction(mostnotably,mass-producedsteelandimprovedtechniquesformeasuringandanalyzingstructuralloadsandstresses),hasallowedcitiestosprawlnotjustoutwardbutalsoupward.Tallbuildings4constructedaroundtheworldsince1885haveacombinedheightofmorethan16,000kilome-ters,equivalenttothecombinedheightofalmost43,000EmpireStateBuildings(AhlfeldtandJedwab2022).Although,atleastfortheverytallestofthesebuildings,themainconstructionmaterialwassteel,sinceabout1964,ithasbeenconcrete(Jedwab,Barr,andBrueckner2022).Despitethelowercostsofverticaltransportationandconstruction,citiesinlow-incomecountriesstillmainlyexpandhorizontallyUsingthesameglobalsampleofmorethan10,000citiesthatunderpinsmuchofthisreport’sanalysis,Lalletal.(2021)addresstheextenttowhichmoderncitiesdevelopalongthethreemarginsopentothem—horizontaldevelopment,infilldevelopment,andverticallayering.Onthebasisoftheiranalysisoftheperiod2000–15,Lalletal.drawastrikingcontrastbetweenthedevelopmentpatternsofcitiesinlow-incomecountriesandthoseofcitiesinupper-middle-andhigh-incomecountries.5Incitiesinlow-incomecountries,91percentofthebuilt-upareagrowthexperiencedoverthisperiodwashorizontalexpansionandonly9percentinfilldevel-opment.Incitiesinupper-middle-andhigh-incomecountries,infilldevelopmentaccountedfor35percentofbuilt-upareagrowthandhorizontalexpansionfor65percent.Atthesametime,buildingheightsinhigher-incomecitiesbecametallerandmorepeaked.Inviewoftheirfindings,Lalletal.describecitiesinlow-incomecountriesasgrowinglike“pancakes,”whereascitiesinmoredevelopedcountriesgrowlike“pyramids”becauseverticalgrowthaccompaniestheiroutwardgrowth.Thepresenceofbothverticalandhorizontalgrowthisconsistentwithpredictionsofthestandardopencitymodelofurbaneconomicsinwhichacityfacesacompletelyelasticsupplyoflabor.Therefore,anyincreaseinthesupplyoffloorspaceduetoverticallayeringthatreducesrentsresultsinaverylargepopulationinflowtothecity,whichalsopushesitsdevelopmentoutward(AhlfeldtandBarr2022).Incomegrowthappearstobethemaindriverofverticallayering,butthatmaynotbethewholestoryAccordingtoLalletal.(2021),thisstrikingdifferenceingrowthpatternsismainlyexplainedbythefactthatcitiesinhigher-incomecountriesaremoreproductivethancitiesinlower-incomecountries.Becausebuildingtalliscapital-intensive,itrequiressufficienteconomicdemandfornewfloorspace.Atlowproductivitylevels,thisdemandisinsufficient,anddevelopersthereforefailtobuildupward.6Policymakersinlow-incomecountriesshouldthereforeaimtocreateanenablingenvironmentthatwillallowverticallayeringtooccuronceacitybecomessufficientlyproductive.Doingsowillfacilitatethecity’sdevelopmentasapyramidratherthanasapancake.Beyondproductivity,thedevelopmentexperiencesofcitiesintoday’shigh-incomecountriessuggestanextradimensiontothestory.WhenChicago’sHomeInsuranceBuildingopenedin1885,theUSgrossdomesticproduct(GDP)percapitawasUS$6,424(in2011constantinternationaldollars),lessthanAngola’sGDPpercapita(US$7,771)in2018.Bythetimethe232THRIVINGEmpireStateBuildingwascompletedin1931,USGDPpercapitawasUS$8,381,roughlyequaltoMorocco’sGDPpercapita(US$8,451)in2018.7Whereasthe102-storyEmpireStateBuildingis1,454feettall,Morocco’stallestcompletedbuilding—theHassanIIMosqueinCasablanca,8completedin1993—is,at690feet,lessthanhalftheheightoftheEmpireStateBuilding.In1970,whenitsGDPpercapitawasUS$9,250,HongKongSAR,China,embarkedonaneraofexponentialgrowthinitsstockoftallbuildings(figure4.2).Overtheperiod1975–85,thecityaddedmoreheighttoitsskylinethandidNewYorkovertheentirecenturyfrom1885to1985.Indoingso,itbecametheundisputedgloballeaderintallbuildings.TheevolutionoftheskylineinHongKongSAR,China,alsoappearstohaveencouragedothercountriesinEastandSoutheastAsiatobecomemorepro-development.Asaresult,citiessuchasBangkok,KualaLumpur,andManilaadoptedtallbuildingsmuchearlierthandidnon-Asiancountrieswithsimilareconomicconditions(AhlfeldtandJedwab2022).ThefactthatmanyEastandSoutheastAsiancities,nottomentioncitiesintheUnitedStates,begantobuildtallatarelativelyearlystageofdevelopmentsuggestsconsiderableuntappeddemandforverticallayeringinmanydeveloping,especiallymiddle-income,countries.Figure4.2Evolutionoftotalsumoftallbuildingheights:HongKongSAR,China;NewYork;andTokyo,1890–2020Source:AhlfeldtandJedwab2022,basedondatafortallbuildingsfromEmporis.Note:Thefigureshowsthetotalsumofheightsoftallbuildings.Tallbuildingsaredefinedashavingaheightofatleast55meters(approximately15stories).Tallbuildingsincludehigh-rises(15–24stories)andskyscrapers(morethan25stories).km=kilometers.1975010020030040050060070018901900191019201930194019501960197019801990200020102020Totalheightoftallbuildings(km)TokyoNewYorkHongKong,SAR,ChinaTheImpactofCitiesonClimateandtheEnvironment233Consistentwiththissuggestion,Jedwab,Barr,andBrueckner(2022)estimate,onthebasisofaglobalanalysisoftallbuildingswithaheightofatleast80meters,thatthetotalheightoftallbuildingsislessthanpredictedinseveralmiddle-incomecountries,includingArmenia,EquatorialGuinea,India,Lesotho,Mauritius,SriLanka,andUzbekistan.9Forexample,theyestimatethatinSriLankaa261percentincreaseinaggregatedurbanbuildingheightswouldberequiredtomakethecountry’sbuildingheightstockequaltothetotalheightpredictedforacountryatitslevelofurbandevelopmentandwithitslevelofagriculturalrents.Finally,althoughtheexperienceofEastandSoutheastAsiancitiesthatmadeanearlystartonbuildingtallisconsistentwiththetheorythatincomegrowthisanecessarypreconditionforverticallayering,thefactthatthetwowenttogetherallowsareversepossibility—inadditiontobeingaconsequenceofeconomicdevelopment,buildingtallmayalso,atleastinpart,havecausedit.Theimpactsofverticaldevelopmentonlandconsumption,theenvironment,andproductivityTheideathatverticaldevelopment(developmentabovelow-riselevels)shouldresultinmorecompactcitiesthatconsumelesslandisanintuitivelyplausibleone.Asmentionedearlier,however,thestandardopencitymodelofurbaneconomicspredictsthatverticallayeringwillalsoleadtohorizontalexpansionbecauseofitsassumptionofperfectlabormobility.Inthisstandardmodel,thereductioninrentsthatresultsfromtallbuildingconstructiondrawsinpeoplefromthecountryside,therebybiddingrentsbackup,untiltheirutility(or“happiness”)hasreturnedtoitsoriginallevel.Theresultisaverylargewaveofin-migrationwiththeimpli-cationthatverticalexpansionalsocauseshorizontalexpansionthatneutralizesanywelfaregains(AhlfeldtandBarr2022).Migrationisfarfromperfectintherealworld,andresearchundertakenforthisreportdemon-stratesthatthepredictionsofthestandardopencitymodelchangeoncethisimperfectionistakenintoaccount(AhlfeldtandJedwab2022).Thus,underacanonicalparameterizationofthemodel,areductioninthecostofheightleadstoanincreaseinthesumofheightsacrossallbuildingsinthecity.Asinthestandardmodel,thisoutcomestimulatesnetmigrationintothecity,butthatmigrationisaccommodatedbythetallerbuildings.Moreover,inthemodelthecity’sverticalexpansionispartiallyoffsetbyahorizontalcontraction.10Withthepositiveneteffectonhousingsupply,theaveragerentonfloorspacefalls.Inthenewequilibriumrentsarelowerandwageshigherbecauseofagglomerationeconomies,sothecityalsoseesanoverallwelfaregain.Asthisdescriptionimplies,themodelpredictsthatareductioninthecostofheightshouldnotonlyleadtoanincreaseinacity’spopulationwithinamorecompacturbanformbutalsostimu-lateincomegrowthandresultinresidentsbeingbetter-offbecauseoftheirhigherwelfarelevels.Byleadingtomorecompacturbanformsandhigherproductivity,lowercostsofverticaldevelopmentcouldsavefertileagriculturallandandreducepollutionHowdothepredictionsofthemodelholdupinpractice?Toanswerthisquestion,AhlfeldtandJedwab(2022)drawonEmporis,aglobaldatabaseof693,855tallbuildings.Foreachcityin234THRIVINGthisreport’sglobalsample,AhlfeldtandJedwabconstructameasureofitstotalheight(thatis,thesumofheightsofalltallbuildings)forallyearssince1884,theyearinwhichconstructionoftheHomeInsuranceBuildinginChicagostarted.Indoingso,theyconfinetheirinvestiga-tiontobuildingswithaheightofatleast55meters(150feet)orapproximately15floors—thatis,high-risesandskyscrapers—judgingthedataaslikelyunreliableforbuildingsunderthisheight.Theirmainfindings,however,arerobusttothemeasurementerrorthatresultsfromignoringlow-andmidrisebuildings.Usingthesedata,AhlfeldtandJedwabestimatetheimpactofacity’stotalheightonitspopu-lation,area,compactness,andproductivity,ascapturedbytheintensityofitsnighttimelightspercapita.Tocontrolfortheexpectationthatcausationwillruninbothdirections,includingfromacity’sproductivitytoitsextentofverticallayering,AhlfeldtandJedwabemploythreeinstrumentalvariable(IV)strategies.11ThefirststrategyisbasedonthedemonstrationeffectoftheearlystartinbuildingtallinHongKongSAR,China,whichappearstohavestimulatedtheadoptionofmorepro-developmentpoliciesinotherEastandSoutheastAsiancountries.Thesecondstrategyexploitsthevariationacrosscitiesinbedrockdepth,whichaffectsthecostsofverticalconstruction.Andthethirdstrategyexploitsvariationinearthquakeriskacrosstheglobalcitysample,whichsimilarlyaffectsverticalconstructioncosts.Overall,theempiricalresultsalignwiththetheory.Dependingontheestimationstrategyadopted,adoublingofacity’stotalheightisassociatedwithalong-runincreaseinitspopula-tionofbetween5percent(ordinaryleastsquaresestimation)and23–24percent(earthquakestrategy),andalong-rundecreaseinitsareaofbetween9percent(ordinaryleastsquaresestimation)and22–25percent(bedrockstrategy)—seefigure4.3.Averagingacrosstheesti-matesthatcontrolforreversecausation,thesefindingsimplythatadoublingofacity’stotalheightleadstoaroughly16percentincreaseinitspopulationanda19percentreductioninitslandarearelativetoothercities.Atthesametime,adoublingofheightisassociatedwithanestimated4percentincreaseintheintensityofacity’snighttimelightspercapita(figure4.3).Althoughthisincreasemayseemsmall,itisaboutthesameorderofmagnitudeastheesti-matesofthestrengthofagglomerationeconomiesreportedintheliterature(AhlfeldtandPietrostefani2019;RosenthalandStrange2004).12Thefactthatanincreaseinacity’soverallheightisassociatedwithareduction,relativetoothercities,initsareaimpliesthatloweringthecostsofverticalconstructioncouldresultinvaluablesavingsinfertileagriculturalland(discussedinmoredetaillaterinthischapter).Furthermore,amorecompacturbanformisalsocorrelatedwithloweremissionsofPM2.5and,therefore,cleaner,morebreathableair,whichcanhavebeneficialimpactsonbothhealthandproductivity.Despitetheseresults,“buildingtall”doesnotimplyapolicyrecommendationtoconstructskyscrapers.Indeed,AhlfeldtandJedwab’smeasureofacity’sheight—thesumofheightsacrossalltallbuildingsinacity—doesnotdistinguishbetween,forexample,construc-tionofone60-storyskyscraperandfour15-storyhigh-risebuildings.Moregenerally,tallerbuildingheightsarebestseenastheresultofamarketresponsetolowercostsofverticalconstruction.Governmentdecisionstobecomedirectlyinvolvedindevelopingtallbuildingsmayormaynotbegoodforurbaninclusion(box4.2).Jedwab,Barr,andBrueckner(2022)provideevidencesuggestingthatseveralcountries—mostnotably,theDemocraticPeople’sRepublicofKoreaandseveralGulfstates—mayhaveengagedinexcessiveconstructionoftallbuildings.TheImpactofCitiesonClimateandtheEnvironment235Figure4.3Estimatedelasticitiesofpopulation,landarea,andnighttimelightintensitywithrespecttototalsumoftallbuildingheightsSource:WorldBankbasedonresultsfromAhlfeldtandJedwab2022,whosedataontallbuildingsarebasedondatafromEmporis.Note:Figureshowstheestimatedpercentagechangeineachvariableresultingfromadoublingofthetotalsumoftallbuildingheightsbasedonfoureconometricestimationstrategies:ordinaryleastsquaresandthreeinstrumentalvariablestrategies(“HongKongSAR,China,”“bedrock,”and“earthquake”),afulldescriptionofwhichcanbefoundinAhlfeldtandJedwab(2022).“Lights”referstotheintensityofnighttimelightspercapitawithinacity’sextent,wherenighttimelightintensityismeasuredusingradiancecalibrateddataderivedfromDefenseMeteorologicalSatelliteProgramsatellitesensors.–30–20–100102030OrdinaryleastsquaresHongKongSAR,ChinaBedrockEarthquakeEstimatedelasticitywithrespecttobuildingheights(%)EstimationstrategyLandareaLightsPopulationWhatdohigh-risebuildingsmeanforurbaninclusion?Thequestionofwhetherhigh-riseconstructionisgoodforurbaninclusionhasnostraightforwardanswer.Verticaldevelopmentprovidesameansofachievinghigherpopulationdensities,which,ifwelllocated,welldesigned,andcoordinatedwithpublictransportation,canhavearangeofbenefitsforurbaninclusion.Dense,transit-orienteddevelopmentmakesitmoreaffordableforresidentstoaccessemployment,education,andamenities.Denserconstructionincreaseshousingsupplyindesirablelocations,which,allelsebeingequal,reduceshousingprices.Ewingetal.(2016)findthatintheUnitedStateslow-densitysprawlisassociatedwithlowerlevelsofupwardincomemobility(measuredasthelikelihoodthatachildbornintothebottomfifthofthenationalincomedistributionreachesthetopfifthbyage30),therebysuggestingthatmorecompactverticaldevelopmentmaybegoodforsuchmobility.Box4.2236THRIVINGDespitetheirpotentialbenefits,high-riseresidentialbuildings,whetherbuiltbytheprivateorthepublicsector,havenotalwayssucceededinprovidingdecenthousingforthepoor.Intheabsenceofthekindsofreformsoflandandpropertymarketsadvocatedinthisreport,high-riseconstructioninmostlow-andmiddle-incomecountriesremainsexpensive,andthusunaffordableformuchofthepopulation.IntheUnitedKingdom,UnitedStates,andWesternEurope,high-risepublichousingforlow-incomeresidents,onceseenasasolutiontotheirhousingproblems,isnowwidelyperceivedtobeafailure(Hess,Tammaru,andvanHam2018;Hunt2018;WhiteandSerin2021).Thisstyleofpublichousingwaseventuallyseenassimplyconcentratingpovertyandcrime,andthecommunallifethatoccursatstreetlevelrarelyflourishesinhigh-rises.Theoftenpoorconstructionandmaintenanceofsuchbuildingshassometimesresultedinhigh-profiledisasterssuchastheGrenfellTowerfireinLondonin2017,whichkilled72people.Manycitieshavedemolishedtheirhigh-risepublichousing.Notallcities,however,havehadbadexperienceswithpubliclybuilthigh-risehousing.ViennaandespeciallySingaporehavehadrelativesuccessinbuildingaffordablehigh-rises,oftendesignedtoprovidespaceforsocialinteractionatmultiplelevels(FalkandRudlin2018;SamantandHsi-En2017).Theirexperiencessuggestthatthedesign,con-structionquality,andmaintenanceofbuildings—ratherthantheirheight—determinetheirsuitabilityasaffordablehousing.Box4.2continuedDoesverticaldevelopmentalsoreduceacity’soverallcontributiontoclimatechange?Althoughmorecompactcities,atanygivenlevelofdevelopment,tendtohavelowerCO2emissionsinboththeresidentialandtransportationsectors,constructionoftallerbuildingstendstorelyonmaterialssuchasconcrete,steel,andglass,whoseproductionentailshighCO2emissions(Pomponietal.2021).Thus,atallbuildingconstructedusingcurrenttechnologiesembedshighup-frontCO2emissions,whichmustbeweighedagainstthefuturelowerflowofCO2emissionsassociatedwithmorecompacturbandevelopment.Relativetoascenariooflower-risedevelopment,itisnotclearwhichofthesetwoopposingeffectsdominatesinthelongrunorunderwhatconditions.13Becauseremarkablylittleresearchhasbeendevotedtothisissue,itisnotpossibletosaywhetherverticaldevelopmentislikelytoleadtoanetlong-runreductionoranincreaseinacity’scontributiontoclimatechange.Thisisnotwith-standingthebeneficialimpactsofmoreverticaldevelopmentintermsofsavingfertileagricul-turallandandinreducinglocallevelsofairpollutionandraisingincomes.Despitethatuncertainty,itdoesseemclearthatthenetimpactofmoreverticaldevelopmentonacity’slong-runcontributiontoclimatechangewillmorelikelybebeneficialifcombinedwithcomplementarytransportationinvestmentsandpoliciesthatbothencourageamovetowardless-pollutingmodesoftransportation,includingwalkingandcycling,andfurtherpromotecompactandlivabledevelopment.Asdiscussedfurthershortly,citieswillmorelikelyseebeneficialenvironmentalimpactsiftheirverticaldevelopmentresultsinamorearchitec-turallyinterestingurbanlandscape(seethesubsectiononurbandesignandlandusediversity).TheImpactofCitiesonClimateandtheEnvironment237Verticalandinfilldevelopmentincitiesinlow-andmiddle-incomecountriesisconstrainedbydysfunctionalurbanlandmarkets,zoning,andrestrictivebuildingregulationsGiventheirapparentbenefits,whatfactorspreventmorecitiesinlow-andmiddle-incomecountriesfromfollowingtheleadsetbycountriesinEastandSoutheastAsiainthe1970sand1980sincreatingamoreenablingenvironmentforverticalandinfilldevelopment?Theanswerliesinacombinationofthedysfunctionalityofurbanlandmarketsinmanylow-andmiddle-incomecountriesandfailuresinplanning.Weakformalinstitutionsfortitlingandpropertytransferfrequentlydeterinvestorsfromputtingcapitalinformalstructures,con-tributingtothepersistenceofslums(Lalletal.2021).Meanwhile,currenturbanplansandplanninginstitutionstendtolackeffectiveness,failingtobothcoordinatemarket-driveninvestmentinstructuresandmanagethespatialformofcitiesinwaysthatpromotetheirgreen,resilient,andinclusivedevelopment(EllisandRoberts2016;Lalletal.2021).Insomecases,citiesenactplanningregulationsinanexplicitattempttodiscourageverticaldevelopment.Indiaisaclassicexample.MajorcitiessuchasBangaloreandMumbaihavehis-toricallyusedrestrictivefloorarearatiostodeliberatelylimitverticaldevelopment(BertaudandBrueckner2004;EllisandRoberts2016).Inpart,policymakersfearedthatallowingmoreofsuchdevelopmentwouldleadtoanunmanageableinflowofpopulationthatwouldover-whelmalreadyoverstretchedurbaninfrastructure.Instead,suchregulationsendeduppro-motinga“messy”patternofurbanization,characterizedbysprawlandslums,underminingboththeprosperityandthelivabilityofmajorIndiancitieswhilealsodrivingupthenegativeenvironmentalimpactsofurbanization.ManySouthAsiancitiesarealsocharacterizedbylarge,fragmentedpubliclandholdings,ofteninprimelocations,thatfurthercomplicatelandassemblybyprivatedevelopers(EllisandRoberts2016).Thissituationfurtherraisesthecostofverticaldevelopment.Intheirglobalanalysisoftallbuildings,Jedwab,Barr,andBrueckner(2022)findthatcountrieswithalowermaximumfloorarearatioandmorestringentlanduseregulationshavehigherbuildingheightpercentagechangegaps.Thisfindingindicatesthat,relativetoasetofbenchmarkcountries,thetotalheightofacountry’sbuildingsislowerthanexpectedgivenitslevelsofurbanincomeandagriculturalrents.Addressingconstraintscouldleadtopro-poorwelfaregains,especiallyinlargercitiesToassessthewelfareimplicationsofconstrainedverticaldevelopment,AhlfeldtandJedwab(2022),intheirbackgroundresearchforthisreport,conductedathoughtexperimentusingtheircalibratedtheoreticalmodel.Whatifallcitieslimitedverticaldevelopmenttoamaximumof15floors?Howmuchlowerwouldwelfarebethaninahypotheticalworldinwhichverticaldevelopmentwascompletelyunconstrainedbyregulationandabletorespondfreelytomarketdemand?Theyfindthat,inthecompletelyunconstrainedworld,globalwelfarewouldbeabout1.5percenthigherandworkerwelfareabout3.3percenthigher.Thesefigures,then,provideestimatesoftheoverallwelfarepotentialoftallbuildings.AhlfeldtandJedwabcalculatethatcitiescurrentlyrealizeonlyaboutone-thirdofthispotential,presumablybecauseofthedysfunctionalityofurbanlandmarketsinmanylow-andmiddle-incomecountriesandthefailuresinplanningdiscussedearlier.238THRIVINGAhlfeldtandJedwabalsocalculatethatthewelfarepotentialoftallbuildingsishighestforthelargestcities,preciselythecitiesinlow-andmiddle-incomecountrieswhereonemightexpecttheconstraintstoverticaldevelopmentposedbydysfunctionallandmarketsandfailuresinplanningtohavethebiggesteffects.Thus,foracitywithapopulationof3million,constrain-ingverticaldevelopmentto15floorsresultsina10percentlossinworkerwelfareandanincreaseinaggregatelandrentofmorethan15percent.Thelandrentincreasealsoimpliesawelfaretransferfromworkerstolandowners.Thisfindingsuggestsnotonlythatartificiallyconstrainedverticaldevelopmentresultsinanaggregatenetwelfarelossbutalsothatthepoorestwilllikelysuffermostfromthiswelfarelossbecauseofthehigherthannecessaryrentstheymustpay.Aglobalaggregateunrealizedwelfarepotentialof0.5percent(1.0percentforworkers)mayseemsmall,butitisimportanttonotethatthesewelfarecalculationsdonotincludetheenvironmentalbenefitsintermsof,forexample,thelowerairpollutiongenerallyassoci-atedwithmorecompacturbandevelopment.Afullaccountingthatincludesthesebenefitswouldundoubtedlyyieldmuchhigherestimatesofthewelfarecostsimposedbythelandandpropertymarketandplanningfailuresthatinflateverticaldevelopmentcostsinmanylow-andmiddle-incomecountries.Urbandesignandlandusediversityalsomatterforthewalkability,associatedcarbon“carprints,”andurbanheatislandeffectofcitiesUsingdetailedhouseholdsurveydataforFrenchcities,BlaudindeThé,Carantino,andLafourcade(2021)pointtothebeneficialcausalimpactsnotonlyofurbancompactnessbutalsoofbothurbandesignandthediversityofacity’seconomicactivitiesonhouseholdfuelconsumptionforcaruse,whichtranslatesintosignificantCO2emissionssavings.Urbandesigninthiscaserefersto,amongotherthings,accessibilitytopublictransportationnetworksandacity’swalkability,basedonthefractaldimensionofthelocalbuilt-uparea.14WhereasahighfractaldimensionistypicalofmorehistoricalFrenchcitycentersdevelopedthroughaquasi-randomdemolitionandreconstructionprocess,amuchlowerfractaldimensiondescribesthemorehomogeneoussuburbsofthesecities,whicharecharacterizedbyregularlyshapedhousingcompounds(map4.1).Ahigherfractaldimensionencourageswalkabilitybyprovidingbothgreaterlegibility(whichmakesaplacecomprehensibletoanobservermovingthroughit)andgreaterdiversityofvisualexperience.Thelessonemergingfromthisanalysisisthatpoliciesthatreducethecostsofbuildingtallmaymorelikelybenefittheenvironmentifcombinedwithurbanandarchitecturaldesignthatmakestheresultingbuiltenvironmentmoreinterestingtopedestrians,therebyencouragingwalkingovercaruse.Conversely,ifbuildingtallleadstoamorehomogeneous—andthereforemoreboring,lesspedestrian-friendly,andevenpotentiallylesssafebuiltenvironment—thenitwillmorelikelyhavenegativenetenvironmentalimpacts.Increasedwalkability,whichalsocriticallydependsonthequalityofsidewalksandstreetlighting,canalsohelppromotebetterhealthoutcomes,ascanthereductionsinairpollutionthattendtogowithreductionsinCO2emissions.Goodurbandesigncanalsohelpcombattheurbanheatislandeffect(seebox1.4,chapter1,foramoredetaileddescriptionofthiseffect).Thedepthofstreetcanyons—thatis,theratiooftheheightofbuildingsalongastreettothewidthofthestreet—canaffectairtemperaturesthroughitsimpactonshadeandventilation.TheorientationofstreetsalsoaffectsbothshadeTheImpactofCitiesonClimateandtheEnvironment239andventilation:streetswithaneast-westorientationreceivemoreprolongedexposuretothesunthanthosewithotherorientationsandthusexperiencemoreheat,especiallyincitiesclosetotheequator(Laietal.2019).Theoverallshapeofacity’sbuilt-upfootprintcanalsoaffecttheurbanheatislandeffect.ResearchonEuropeancitiessuggeststhatlargercities,morecompactcities,andcitieswithmorecircularbuilt-upfootprintsexperienceamoreintenseheatislandeffect(PiererandCreutzig2019;Zhou,Rybski,andKropp2017).Bycontrast,star-shapedcities,withtranspor-tationcorridorsradiatingfromahigh-densitycore,canretainthetransportationbenefitsofcompactnesswhilereducingheateffects.Thedesignofgreenspacesalsoinfluencesurbantemperatures,althoughnotonlythequantitybutalsotheconfigurationofvegetationmatter.Pocketparks—smallparksofabout1,000squaremeters—placedatregularintervalsaremoreeffectivethanasinglelargepark,especiallyinhigh-rise,high-densitytropicalregions(GiridharanandEmmanuel2018).Whatroledoes(horizontal)transportationplayinshapingcitiesandtheirenvironmentalimpacts?Alongwithincomeandotherfactors—technologies,institutions,andpolicies—thataffectverticaldevelopmentcosts,amajordeterminantofacity’surbanformis(horizontal)trans-portationtechnologiesandsystems—andtheincentivesthataffectthemodesoftransporta-tionthatacity’sresidentschoosetouse.Moreover,investmentsintransportationsystemsandincentivesthatinfluencechoicesoftransportationmodeaffecttheenvironment.Theydosonotonlyindirectlybyhelpingshapeacity’sformbutalsodirectlythroughtheemissionsofbothCO2andairpollutantssuchasPM2.5towhichdifferentmodesgiverise.Acloseandmutuallyreinforcingrelationshipalsoexistsbetweentransportationandlanduse.Forexample,theviabilityofmasstransitsystemsdependsheavilyonlanduseanddensity.Map4.1ExamplesofmoreandlesswalkableurbanenvironmentsSource:BlaudindeThé,Carantino,andLafourcade2021.240THRIVINGOnthesupplyside,thecostsofprovidingmasspublictransitarehigherinlow-densitysettingsbecauseoftheneedtoservemoredistantlocations.Onthedemandside,compacturbandevel-opmentintheformofinfillandverticaldevelopmentrequiresmasstransitsystemstosupporttheresultantdensity.Consistentwiththisfinding,inameta-analysisofstudiesontheeffectsofdensity,AhlfeldtandPietrostefani(2019)findthatmetroraildensityispositivelyassociatedwithpopulationdensity,whereascaruseisnegativelyassociatedwithpopulationdensity.Policiestopromotemasstransitarethereforecomplementarytopoliciesthatpromoteverticalandinfilldevelopment.CitiesthatrelymoreoncarsandlessonmasstransitsystemsaremoresprawlingInthesamevein,Ostermeijeretal.(2022)show,foraglobalsampleof123cities,astrongnegativerelationshipbetweenacity’spopulationandemploymentdensitiesanditsrateofcarownership.Asimilarlystrongnegativerelationshipexistsacrosscountriesbetweenaverageurbanpopulationdensityandaveragecarownership(figure4.4).ToaddressconcernsofFigure4.4Relationshipbetweencarownershipandurbanpopulationdensities,bycitylevelandcountrylevelSource:Ostermeijeretal.2022usingdataforvariousyearsbetween1960and2012withmultipledatapointsforsomecities.Note:Greenmarkersindicatecitiesandcountrieswithadomesticcarmanufacturerin1920.Cityandcountrylabelsarebasedonminimum,median,andmaximumpopulationdensitiesforeachbinof10carsper100people.Thesolidlinerepresentstheestimatedbestfitlinearregressionline.ln=naturallog.HoChiMinhCityJakartaKocaeliNewDelhiTaipeiBudapestBarcelonaLisbonBrusselsAbuDhabiAthensFrankfurtBrisbanePerthPhoenixFrankfurtPhoenixSydneyLosAngelesStrasbourgBrisbaneDenverRomeRomeCalgaryAtlanta67891011020406080Carownershipper100peopleVietnam(1995)Denmark(1960)Germany(1960)RussianFederation(1995)UnitedKingdom(1960)Korea,Rep.(1995)SouthAfrica(2012)Australia(1960)Taiwan,China(2012)UnitedKingdom(1980)Canada(1960)UnitedKingdom(1995)RussianFederation(2012)Morocco(2012)UnitedStates(1960)Spain(1995)Germany(1995)Australia(1980)Canada(1980)UnitedArabEmirates(2012)Canada(1995)Italy(1995)UnitedStates(1995)Italy(2012)67891011020406080Averagecarownershipper100peopleb.Countrylevela.CitylevelPopulationdensity(ln)Averageurbanpopulationdensity(ln)TheImpactofCitiesonClimateandtheEnvironment241reversecausality—thatlowerdensitydevelopmentalsoencouragespeopletobuymorecars—Ostermeijeretal.(2022)useanIVstrategy.Thisstrategyexploitsthefactthatcitiesthathadacommercialcarmanufacturerin1920,beforetheconstructionofmodernhighwaysintoday’shigh-incomecountriesandtheriseofmasscarownership,morelikelyhavehighercarowner-shiptoday.Acity’spopulationdensitytoday,however,cannotexplainthepresenceofacom-mercialcarmanufacturerin1920,norwasacity’spopulationdensityin1920relatedtothepresenceofacarmanufactureratthetime.Usingthisstrategy,Ostermeijeretal.estimatethatanincreaseof1standarddeviationincarownership(equivalenttoabout20carsper100urbaninhabitants)causesalong-runreductioninacity’spopulationdensityofabout35percent.Thiseffectismainlydrivenbyanexpansioninthesizeofacity’sextentratherthanalossofpopulation,suggestingthatcarsallowlower-densityexpansionintoacity’speriphery.ThisevidenceontheroleofcarsindrivingsprawlisconsistentwithevidencethatshowsthattheUSinterstatehighwaysystemcontributedtosuburbanizationofthecitiesthroughwhichitpasses(Baum-Snow2007).HighwaysalsocontributedtothesuburbanizationofcitiesinChina(Baum-Snowetal.2016).Partofthereasonthepresenceofacarmanufacturerin1920predictscarownershipincitiesgloballytodayisthat,fromthe1950stothelatetwentiethcentury,largecommercialcarman-ufacturerslobbiedstrongly,especiallyintheirdomesticmarkets,forincentivesandinvest-mentsaimedatpromotingcarownership,oftenattheexpenseofothertransportationmodes.Lobbyistssoughttolimitvehicletaxes,toincreaseroadconstructionandparkingincities,andtoquashinvestmentinpublictransportation(Ostermeijeretal.2022).15Thehistoriclobbyingofmajorcarmanufacturersagainstinvestmentsinurbanpublictrans-portationandmasstransitsuggeststheybelievedsuchinvestmentswouldlimitcarownership.EvidencethatthisisindeedthecaseisprovidedforGreaterCairobyHegeretal.(2019).Usingmachinelearningtechniquestoestimate,fromhigh-resolutionsatelliteimagery,thedailynumberofcarsonCairo’sroadsandinputtingtheseestimatesintoaneconometricmodel,Hegeretal.findthatthetwo-phasedopeningofCairo’sMetroLine3inFebruary2012andJuly2014significantlyreducedthenumberofcarsonthecity’sroads.Investmentsinmasstransitsystemsmaycontributedirectlytoreducingairpollution,withsignificanthealthbenefitsInastudyofthe58subwaysystemopeningsthatoccurredincitiesgloballybetweenAugust2001andJuly2016,Gendron-Carrieretal.(2022)findthat,forcitieswithhigherinitialpollutionlevels,subwayopeningsreducedparticulatesby4percentintheareasurroundingacitycenter.Theseeffectspersistforatleastfouryears,thelongesttimehorizonoverwhichtheauthors’datapermitthemtoestimateimpacts.Furthermore,onthebasisofobservedsubwayridershiplevels,Gendron-Carrieretal.findthattheseimpactscanbeplausiblyexplained“onlyifsubwaysdiverttripsthatwouldotherwisehaveoccurredinparticularlydirtyvehiclesoratparticularlycongestedtimes”(Gendron-Carrieretal.2022,166).Theygoontostateonthesamepage,“Thisisconsistentwithevidencethatpublictransitservesthepoorandthatsubwaysaremuchmoreheavilyusedatpeaktimesforvehicletraffic.”Investmentsinpublictransitarethereforenotonlygoodfortheenvironmentbutalsopro-poor,therebyhelpingcontributetomoreinclusivecities.16Substantialhealthbenefitsareassociatedwiththereductioninparticulatelevelsstemmingfromthelaunchofnewsubwaysystemsinmorepollutedcities.Gendron-Carrieretal.calcu-latethat,onaverageinsuchacity,asubwaysystemopeningprevents22.5infantdeathsand242THRIVING500totaldeathsperyear.Usingstandardincome-adjustedlifevalues,thisavertedmortalitycanbevaluedatroughlyUS$43millionandUS$1billionperyear,respectively.AsGendron-Carrieretal.note,theirestimatesdonottakeintoaccounttheeffectsofreducedairpollu-tiononmorbidityoronproductivity.Moreover,theestimatesdonotconsiderthepotentialexternalenvironmentalandhealthbenefitsthatcouldresultfromlonger-runlandusechangesthatmasspublictransithelpsinduce.Thesechangesmaycontributetoamorecompacturbanform,whichiscorrelatedwithlowerCO2andPM2.5emissions(seetheearlierdiscussioninthischapterandinchapter1).LookingatGreaterCairo,Hegeretal.(2019)providefurtherevidencethatsubwayshelpreduceairpollutionwithsubstantialhealthbenefitsforhighlypollutedcities.Asmentionedearlier,Hegeretal.estimatethatCairo’stwo-phasedopeningofitsMetroLine3inFebruary2012andJuly2014significantlyreducedthenumberofcarsonitsroads.Theyfurtherestimatethatthisreductionintraffichelpedreducebyabout3percentthecity’sconcentrationsofparticu-latematterwithadiameterof10micronsorless,generatingestimatedavertedmortalitygainsvaluedataboutUS$98.7million.Removaloffuelsubsidiescanhelpreducedependenceoncars,withsignificantenvironmentalandhealthbenefitsInadditiontoestimatingtheimpactsoftheopeningofMetroLine3oncaruseandpollu-tioninGreaterCairo,Hegeretal.(2019)studiedtheimpactsoftheArabRepublicofEgypt’sfuelsubsidyremovalprogram.ThisprogramresultedintargetedfuelpriceincreasesinNovember2016andJune2017that,dependingonthefuelcategoryandperiod,variedbetween30and80percent.Hegeretal.estimatethat,byreducingtrafficbelowlevelsthatotherwisewouldhaveprevailed,theseincreasesledtoa4percentreductioninconcentrationsinGreaterCairoofparticulatematterwithadiameterof10micronsorless,ontopofthe3percentreduc-tionassociatedwiththeopeningofthemetroline.Inturn,this4percentreductiongeneratedestimatedavertedmortalitygainsvaluedatroughlyUS$110.4million.Thisfindingsuggeststhat,inadditiontoinvestmentsinpublictransportationsystems,theremovalofdistortionaryincentivesthatskewmodechoicesawayfrompublictransportationandtowardprivatecarusecanhavesignificantenvironmentalbenefits.Thesameisalsotrueofpoliciesthatgobeyondjustremovingsubsidiesthatartificiallyencour-agecarownershipandusebytryingtoaddressthenegativecongestionandenvironmentalexternalitiesthatarisefromprivatevehicleuse.Themostobvioussuchpolicyiscongestionchargingasappliedin,forexample,DurhamandLondonintheUnitedKingdom,GothenburgandStockholminSweden,MilaninItaly,andRigainLatvia.Otherpoliciesincludethosetoencouragehigheroccupancyofvehicles,suchasJakarta’s3-in-1policyinplacebetweenMarch1992andMarch2016(box4.3),andthosetodiscouragecitycenterparkingand,therefore,driving.Theuseofdemand-sidepoliciesthathelpcorrectforthecongestionandenvironmentalexter-nalitiesarisingfromprivatevehicleuseisimportanttostaveofftheriskoftrafficcongestionreboundinginthelongruninresponsetoanypublictransportationinvestment.Intheabsenceofdemand-sidepolicies,suchalong-runresponseistobeexpectedbecause,byreducingcongestionintheshortrun,publictransportationinvestmentslowerdrivingcosts,which,inturn,stimulatesmoredriving.17Atthesametime,policiestodiscourageprivatecarusemaymatterless,fromanenvironmentalperspective,inthelongrungiventherapidgrowthofelectricvehicles.Box4.4discussesthepossibilitythatsuchvehiclescoulddramaticallyreducetransportation-relatedCO2emissionsincities.TheImpactofCitiesonClimateandtheEnvironment243High-occupancyvehiclepolicyinJakartaJakarta’s3-in-1high-occupancyvehiclerestriction,implementedinMarch1992,requiredcarsridinginbothdirectionsonthemajorcorridorsinJakartaatspecifictimesofdaytohaveatleastthreepassengers,includingthedriver.Thepenaltyfornotcomplyingwasamaximumfineof500,000Indonesianrupiah(aboutUS$37.50)ortwomonthsinprison.BecausemanyaffluentcaruserscommutingintoJakartadidnotmeettheoccupancyrestriction,aninformalmarketforprofessionalpassengers,or“jockeys,”emerged.Jockeyswouldridealongfrom3-in-1accesspointsforaboutUS$1.20.Thepracticecontributedtotheterminationofthe3-in-1policy,temporarilyinMarch2016andthenpermanentlyinMay2016.TheterminationsetbackJakarta’sattempttocurbcongestion.Despitethepolicy’sproblems,trafficworsenedsignificantlywhenitwaseliminated(Hanna,Kreindler,andOlken2017).Delaysincreasedby39–45percentduringthemorningpeakperiodandby69–85percentduringtheeveningpeakperiod.Theaveragespeedfellfrom28kilome-tersperhourto20inthemorningandfrom21kilometersperhourto12intheevening.Inaddition,negativeknock-oneffectsoccurredatothertimesofthedayandonroadsoutsidetheregulatedareas,mainlybecauseoftheincreasednumberofcarsontheroadfromeliminatingthe3-in-1policy.Thisepisoderevealsseveralfacts.First,driverswerewillingtopaytoaccessthecitycenterbycar—enoughtosustaintheinformaljockeymarket.Second,theauthorities’failuretoanticipatedriverwillingnesstopaycontributedtothepolicy’slimitedimpactoncongestion.Third,facedwiththisevidence,theauthoritiesrespondedbyscrappingthepolicyratherthantryingtoimproveit.The3-in-1experimentshowedthat,intheJakartaarea,sometypeofcongestionpricingwouldbetteraddresstheproblem.Proceedscouldthenbeusedtofundurbanmasstransitormobilityforlower-incomepeople.Source:AdaptedfromRoberts,GilSander,andTiwari2019andbasedonHanna,Kreindler,andOlken2017.Box4.3Willelectriccarssavetheday?AutomobileswithinternalcombustionengineswereonceacclaimedastheecologicalsaviorsthatwouldsolvethegreathorsemanurecrisisafflictingLondon,NewYork,andotherlargecitiesattheturnofthetwentiethcentury.Instead,automobilesbroughttoxicpollutantsthat,althoughlessvisiblethanmanure,werenocleaner.Globally,transporta-tionemissions,mainlygeneratedbypersonalautomobiles,representedaboutaquarterofallurbancarbondioxide(CO2)emissionsin2015.aThelastdecade,however,hasseenthetremendousgrowthofelectricvehicles(EVs)withglobalsalessoaringfrom130,000in2012to6.6millionin2021.China’smarket,thelargest,accountedformorethanhalfofglobalEVsalesin2021.Meanwhile,inrelativetermsEuropeancountriesleadtheway—in2021,EVsaccountedforalmost75percentofvehiclesalesinNorway,and45percentinSweden.bBox4.4244THRIVINGBycomparison,intheUnitedStates,only4.5percentofvehiclessoldin2021wereEVs.Nevertheless,eventhissharerepresentsadoublingoverthatof2020.OutsideofChina,Europe,andtheUnitedStates,EVscurrentlyhaveonlymeagermarketshares,includ-ingincountriessuchasBrazil,India,andIndonesia,althoughBrazilhasannualsalesincreasesofover200percent.HigherpricespresentoneimportanthurdletothewidespreadadoptionofEVsinlower-incomecountries.Thisproblemwilllikelychangeaspricesfallwithasharpandcontinuingdeclineinbatteryproductioncosts—fromUS$1,000perkilowatt-hourin2007toUS$410perkilowatt-hourin2014.And,justasforsmartphones,whichhadmuchfasteradoptionthanforecast,lower-incomecountriesarelikelytocloselyfollowricherones.Forexample,Cherif,Hasanov,andPande(2017)predictthat,inaslow-adoptionscenario,EVswillaccountforabout36percentofallcarsontheworld’sroadsbytheearly2040s.Inafast-adoptionscenario,almost90percentoftheworld’stotalvehiclestockwillbeelectricbythe2040s.Nevertheless,althoughEVsarewidelyseenasbeinggreenerthanconventionalvehicles,theexactsizeoftheirenvironmentalbenefitsisnotclear.Thus,studiescometodiffer-entconclusionsaboutthereductionofCO2emissionsassociatedwithEVadoption,dependingonthestudies’assumptionsabouthowdirtytheelectricitygenerationmixis,cwhethertheyuseaverageormarginalemissionsintheircalculations,andtheirassump-tionsaboutdrivingpatternsandtheweather.DespitesimilaremissionsassociatedwiththemanufacturingofinternalcombustionenginecarsandEVs,differencesinlife-cycleemissionsbetweenthetwotypesofcardependonthefuelcycle,batteryproduction,andtailpipeemissions.BecauseEVslackthelatter,resultstendtoshowthattheyfarebetterthancombustionenginevehicles,butthemarginsarefinerwhencomparingEVswithhybrids.AlthoughtheenvironmentaladvantagesarisingfromEVsaremarginalundersomeassumptions,therearereasonstobeoptimisticaboutEVs’potentialcontributiontoreducingurbanCO2emissions.Manycountriesaresteadilydecarbonizingtheirenergysupplies,sothelifetimeemissionsofEVshavefallensharplyandareprojectedtokeepfalling.Moreover,becausepowerplantsburnfuelmoreefficientlythancombustionenginesdo,anEVfleetpoweredbya100percentcarbonenergymixwouldstillgeneratelessfuelcycleemissionsthananequivalentfleetofinternalcombustionenginevehicles.Thebenefitsappearlowerinlow-andmiddle-incomecountriesbecauserenewablesourcesrepresentasmallershareoftheenergymix.China,India,andcountriesinSub-SaharanAfrica,however,havepickedupthepaceoftheirenergytransition.Insomeofthesecountries,theunreliabilityofthegrid,whichwillnecessitatesubstantialinvestmentstosustainalargenetworkofEVs,representsaseriousobstacletoadoption.TheneedtobuildasufficientlydensenetworkofEVchargingstationspresentsanotherchallenge.Thewidespreaduseofmotorcyclesmaybebeneficialbecausetheirelectricalcounterparts,whosemarketisalsogrowingveryrapidlyinAsia,havesmallerbatteries.Box4.4continuedTheImpactofCitiesonClimateandtheEnvironment245Effectively“computersonwheels,”EVsofferverydifferentcapabilitiesfromtheirprede-cessors,andcouldfundamentallyshiftthewaypeopleusevehicles.Althoughmostpro-jectionsassumeaone-to-onereplacementofmotorvehicleswithEVs,thesimultaneousdevelopmentofautonomousdrivingcouldmovetheworld’scitiestowardasystemofsharedvehicles.Sources:Cherif,Hasanov,andPande2017;HallandLutsey2018;Hausfather2019;PaoliandGül2022.a.WorldBankanalysisbasedonCO2emissionsdatafromtheEuropeanCommission’sGlobalHumanSettlement(GHS)UrbanCentreDatabaseR2019,https://ghsl.jrc.ec.europa.eu/ghs_stat_ucdb2015mt_r2019a.php,whichderivesitsCO2emissionsdatafromtheEuropeanCommission’sEmissionsDatabaseforGlobalAtmosphericResearch(EDGARv5.0).b.PassageofnationallegislationsubsidizingownershipofEVsandsettingstrictemissionsstandardshasencouragedandacceleratedthissurgeofEVsalesinEurope.Somelargecitieshaveharddeadlinesforcompletelyphasingoutpetrolanddieselcars.c.Incountrieswhereenergyproductiondoesnotinvolveemissions—suchasFrance(nuclear)andNorway(hydroelectric)—thefuelcycleemissionsassociatedwithEVsarealmostzero(HallandLutsey2018).FuelcycleemissionsreferstothedieselorpetrolthatpowersaninternalcombustionenginevehicleortotheenergymixthatgeneratestheelectricityforcharginganEV’sbattery.Box4.4continuedMasstransitsystemsarenotasilverbulletforreducingcongestionandpollution—goodplanninganddesignarecrucialtosuccessAlthoughGendron-Carrieretal.(2022)reportthatsubwayopeningshaveasignificantnegativeaverageimpactonairpollutioninthemorepollutedcities,thisresulthidesconsid-erableheterogeneityofestimatedimpactsacrosscities.For9ofthe28citiesintheirsamplethathadlevelsofairpollutionabovethesamplemedian,theopeningofasubwaysystemhadastatisticallysignificantpositiveimpactonairpollution—thatis,airpollutionincreased(figure4.5).18Onepotentialexplanationforhigherpollutionlevelsinthesecitiescouldbehighertrafficcongestioninthevicinityofnewsubwaystations.Forexample,privatecarsandmotorbikes,taxis,andride-sharingvehiclesstoptodropoffandpickuppassengersusingthesubway,par-ticularlyaroundpeakhours.Consistentwiththisidea,Rao(2016)findsthattheopeningofLondon’s“nighttube”serviceinAugust2016,whichextendedtheweekendoperatinghoursoftheCentralandVictorialinesonthecity’smetrosystemtoalsocoverthehours12:30a.m.to5:30a.m.,hadastatisticallysignificantpositiveimpactonridesharejourneysintheneighbor-hoodsinwhichnighttubemetrostationsarelocated.Meanwhile,forJakarta,Gaduha,Gračner,andRothenberg(2022)estimatethattheopeningofTransJakarta,oneoftheworld’slargestbusrapidtransitsystems,hadtheunintendedeffectofincreasingtrafficcongestionalongservicecorridors.Theyattributethisoutcometofailingsintheimplementationofthesystem.Gaduha,Gračner,andRothenbergrelyoncounterfactualsimulationsperformedusingaquan-titativespatialgeneralequilibriummodeltoshowhowimprovementsindesigncouldyieldsignificantwelfareimprovementswithonlymodestcosts.Theseexamplessuggestthatgooddesignandplanningarekeytorealizingtheenvironmentalandotherbenefitsofmasspublictransportationsystems.246THRIVINGFigure4.5EstimatedimpactsofsubwaysystemopeningsbetweenAugust2001andJuly2016onairpollutionlevels,58citiesSource:Gendron-Carrieretal.2022.Note:Graphshowstheestimated58city-specificsubwayeffectsasmarkers.Bluemarkersindicateestimatesthatarestatisticallysignificantatthe5percentlevel;greenmarkersindicateestimatesthatarestatisticallyinsignificantatthe5percentlevel.ThehorizontalaxisindicatestheAerosolOpticalDepth(AOD)level,ameasureofairpollution,in2000,andtheverticalaxisindicatesthecoefficientestimates.Thesolidlineshowsalinearfitofthemarkers,andtheverticaldashedlinethepartitionofthesampleintothosebelowandabovemedianinitialAODlevel.Standarderrorsareclusteredatthecitylevel.–0.15–0.10–0.0500.050.100.1500.20.40.60.81.0CoecientestimatesBaselinepollutionlevel(AODin2000)Howwillurbanexpansionaffectagriculturallandandproduction?HowclimatechangeisaddressedwillhaveimportantimplicationsforlanduseThegrowthofcities,andurbanizationmoregenerally,isintertwinedwithfoodsystemsandopportunitiesfortheirtransformation.Becauseitgivesrisetoincomegrowth,urbanizationisassociatedwitharisingandchangingdemandforfood.Italsoleadstochangesinlandusethataffectagriculture,andcitieschannelimportantfinance,inputs,information,services,andoff-farmemploymentopportunitiestoruralareas(AbuHatabetal.2019;deBruin,Dengerink,andvanVliet2021;WorldBank2009).Moreover,theinteractionsamongclimatechange,urbanexpansion,landuse,andfoodproductionexacerbatechallengesforcities.TheImpactofCitiesonClimateandtheEnvironment247Theworld’sapproachtoclimatechangeadaptationandmitigationis,toanextent,determinedbyeconomicanddemographicdrivers,includingpopulationgrowth,theurbanshareofthepopulation,andincomepercapita.Inthiscontext,thefiveSharedSocioeconomicPathways(SSPs)representscenariosofhowtheworldmightrespondtoclimatechangemovingforwardaccordingtodifferentprojectionsofthesedrivers(box4.5).TheSSPstracetheprojectedtra-jectoriesofvariousoutcomes,includingenergyuse,landuse,greenhousegas(GHG)emis-sions,andclimatechange.OfthefiveSSPs,SSP1isalargelypositivescenarioreferredtoasa“sustainabilitypathway,”whereasSSP3isalargelynegativescenariocalleda“regionalrivalrypathway”(Riahietal.2017).Alookattheunderlyingeconomicanddemographicdriversrevealsthefollowing.First,totalpopulationwillgrowfasterandtheilliterateproportionofthepopulationwilldeclinemoreslowlyunderSSP3thanunderSSP1.Populationwillpeakaround2050underSSP1,whereasitwillcontinuetogrowthrough2100underSSP3.AlsounderSSP3,theilliteracyratewillbegintogrowby2040.Second,theurbanshareofthepopulationwillincreaseunderbothscenarios,butitwilldosomuchfasterunderSSP1thanunderSSP3becauseurbanizationisassumedtolinktobetterdevelopmentoutcomes.Third,GDPpercapitagrowsexponentiallyunderSSP1,whereasitremainsstagnantunderSSP3.IncomeinequalitydeclinesmuchfasterunderSSP1thanunderSSP3.FiveSharedSocioeconomicPathwaysThefiveSharedSocioeconomicPathways(SSPs)arepartofanewscenarioframeworkestablishedbytheclimatechangeresearchcommunitytofacilitateintegratedanalysisoffutureclimateimpacts,vulnerabilities,adaptation,andmitigation(Riahietal.2017).Theyinvolvethreemainsetsofscenariodrivers:(1)populationandeducation;(2)urban-ization,definedastheurbanshareofthepopulation(JiangandO’Neill2017);and(3)grossdomesticproductpercapitaandinterpersonalincomeinequality.ThefiveSSPsarethefollowing:••SSP1:Sustainability—takingthegreenroad(lowchallengestoadaptationandmitigation)••SSP2:Middleoftheroad(mediumchallengestomitigationandadaptation)••SSP3:Regionalrivalry(highchallengestomitigationandadaptation)••SSP4:Inequality(lowchallengestomitigation,highchallengestoadaptation)••SSP5:Fossil-fueleddevelopment(highchallengestomitigation,lowchallengestoadaptation)ThenarrativeforSSP1,forexample,saysthat“theworldshiftsgradually,butperva-sively,towardamoresustainablepath,emphasizingmoreinclusivedevelopmentthatrespectsperceivedenvironmentalboundaries.Managementoftheglobalcommonsslowlyimproves,educationalandhealthinvestmentsacceleratethedemographictran-sition,andtheemphasisoneconomicgrowthshiftstowardabroaderemphasisonhumanwell-being.Drivenbyanincreasingcommitmenttoachievingdevelopmentgoals,inequalityisreducedbothacrossandwithincountries.Consumptionisorientedtowardlowmaterialgrowthandlowerresourceandenergyintensity.”(Riahietal.2017,157).Box4.5248THRIVINGFoodproductionhassignificantnegativeimpactsonclimate.In2015,GHGemissionsfromtheglobalfoodsystemamountedto18gigatonnesofCO2equivalentorone-thirdoftotalGHGemissions.Thelargestcontribution,71percent,camefromagricultureandlanduseorlandusechangeactivities,withtheremainderfromsupplychainactivities,includingretail,trans-portation,consumption,fuelproduction,wastemanagement,industrialprocesses,andpack-aging(Crippaetal.2021).Emissionsassociatedwithanimal-basedfoodsaretwiceaslargeasthoseassociatedwithplant-basedfoods.OftheglobalGHGemissionsfromfoodproduction,57percentcomefromanimal-basedfoods(includinglivestockfeed),29percentfromplant-basedfoods,and14percentfromotheruses(Xuetal.2021).Thetwolargestcontributingplant-andanimal-basedcommoditiesarerice(12percent)andbeef(25percent).Underascenarioofsustainability(SSP1),croplandwouldbarelyexpandbetween2010and2100,whereaspasture,suitableforlivestockgrazing,woulddecrease(figure4.6,panelsaandb).Meanwhile,forestsandothernaturallandareprojectedtoexpand(figure4.6,panelc).Bycontrast,underaregionalrivalrypathway(SSP3),croplandandpasturewouldincrease,whereasforestsandothernaturallandwoulddeclinedramatically(Riahietal.2017).UnderSSP1,therefore,sustainablefoodproductionimpliestheuseoflessland-intensiveandmoreproductivemethodstofeedagrowingpopulation.UnderSSP3,thecombinationofpopula-tiongrowth,deforestation,anduncheckedlivestockproductionwillresultinsignificantlyincreasedrisks,includingaheightenedriskforpandemics(box4.6).Nomatterthescenario,however,urbanlandisprojectedtoexpand—andmoreso,atleastini-tially,underascenarioofsustainabilitythanunderoneofregionalrivalry.Theassumedunder-lyinggrowthtrendsintotalpopulationandintheurbanpopulationsharewillthusleadtoanincreaseinthedemandforurbanlandupto2100acrossallSSPpathways(Chenetal.2020)—seefigure4.6,paneld.UnderSSP1,thedemandforurbanlandwillincreasebutthendeclineafter2070,giventheassumedeventualdeclineinpopulationafter2050underthisscenario.UnderSSP3,thedemandforurbanlandwillinitiallyincreasemoreslowlythanunderSSP1,giventheassumedslowgrowthintheurbanshareofthepopulationunderSSP3.Becauseofthecontinuedpopulationgrowthto2100underSSP3,however,urbanlanddemandwillcontinuetoexpandto2100andwillapproachthesamelevelasSSP1by2100.Theseprojections,however,donotaccountforthepossibilityofverticalandinfillurbandevelopment,which,asdiscussedearlierinthischapter,canhelpcitiesaccommodatethegrowingdemandforurbanlandwithouthavingtoexpandasrapidlyoutward.TotheextentthatincomepercapitaisprojectedtogrowfasterunderSSP1thanunderSSP3,verticallayeringcan,alongwithinfilldevelopment,beexpectedtooccurmorequicklyundertheformerthanunderthelatterscenario.Policyandinstitutionalreforms,whichimprovethefunctioningofurbanlandandpropertymarkets,therebyreducingverticaldevelopmentcosts,canlikewisebeexpectedtoslowthehorizontalexpansionofcities,ascantherelaxationofplanningrestrictionsthatconstraintallbuildingdevelopmentandfostercardependence.Mosturbanexpansionwillbeintocropland,followedbyforestsAcrossallpathways,itisprojectedthatmosturbanexpansionwillentailtheconversionofparticularlyproductivecropland,followedbyforestsandthengrassland.By2100,underSSP1,55percentofnewlyexpandedurbanland(63percentunderSSP3)isexpectedtohavedis-placedcurrentcropland,27percent(21percent)forests,and12percent(6percent)grassland.Theconversionofcroplandandgrasslandintourbanlandhasimportantimplicationsforcropandlivestockproduction,andtheconversionofforestswillhavenegativeimplicationsforbio-diversityandcarbonsequestration(Chenetal.2020).UrbanexpansionwillthereforeresultinTheImpactofCitiesonClimateandtheEnvironment249Sources:AdaptedbasedondatafromRiahietal.2017(panelsa,b,andc)andChenetal.2020(paneld).Note:Panelsa,b,andcdepictchangesincropland,pasture,andforestforthefiveSharedSocioeconomicPathways(SSPs)andtheirranges.Changesareshownrelativetothebaselineyear2010=0.PanelddepictsprojectionsofurbanlanddemandundertheSSPscenarios,withtheshadedarearepresenting95percentconfidenceintervalsofprojectedurbanlanddemand.km2=squarekilometers.Figure4.6Projectedchangesingloballandareadevotedtocrops,pasture,forests,andurbanusesunderfivescenariosofclimatechangemitigationandadaptation,2010–2100d.Urbanusesc.Forestb.Pasturea.Cropland-600-400-2000200400600800200020202040206020802100200020202040206020802100Hectares(millions)-800-600-400-2000200400600800Hectares(millions)-600-400-2000200400200020202040206020802100Hectares(millions)4060801001201401601802002010202020302040205020602070208020902100Hectares(millions)SSP1SSP2SSP3SSP4SSP5250THRIVINGPandemicrisksattheinterfaceofhumans,animals,andtheenvironmentThecontinuedperiodicsurgesofCOVID-19highlighttheurgentneedtomitigatetheriskofanotherpandemic,whichmayresultinevenmoredeathsandgreaterdisruptionsofeconomicactivity.Itisimpossibletoanticipatewherethenextpandemicwillemergebecauseeveryregionhasexperiencedanincreaseinoutbreaksofemerginginfectiousdiseases(EIDs),withagrowingprevalenceofzoonoticandviraldiseases.Despitetheglobalnatureofpandemicrisksandvulnerabilities,theyarenotdistrib-utedevenly.MosthotspotsforEIDsareinlow-andmiddle-incomecountries,wherethebiodiversityandabundanceofanimalhostsarehigh,andthebehaviorsandoccu-pationsthatbringpeopleandanimalsintocontactarewidespread.Diseasescanspreadrapidlyinlargecities,whichcanthenserveasincubatorsforepidemicsandpandemics.Thelow-qualityhousingandpoorsanitationarisingfrompoorlymanagedurbanizationprovidebreedinggroundsforvariousdiseases.Poorlymanagedandunplannedurban-izationcanalsoleadtocloseencounterswithwildlifeasurbandevelopmentintrudesonpreviouslyuntouchedecosystems.ThebeststrategyinthefaceofuncertaintiesaboutEIDsandpandemicrisksistoadopta“OneHealthapproach.”Thiscross-sectoralapproachcentersaroundtheconnectionsamongpeople,animals,plants,andtheirsharedenvironment.ItfostersmitigationofpandemicriskbypreventingEIDsatthesource.ApplyingaOneHealthapproachentailslearningfrompastzoonoticoutbreaksandfocusingonhotspotswherelandusechangesandfoodsystemcharacteristicsinvolvecertainhigh-riskpractices.••Themostimportantreservoirsofpathogenswithpandemicpotentialaremammals,particularlybats,rodents,andprimates,aswellassomebirdsandlivestocksuchaspigs,poultry,andcamels.••High-riskpracticesinlandusechangeincludedeforestation,miningpractices,ecotour-ismpractices,andthewildlifetrade.••High-riskpracticesinfoodsystemsincludelowbiosecurityinlivestocksystems,uncontrolledintensificationofanimalproduction,livestocktrade,slaughterhouses,andliveanimalmarkets.••About70percentofEIDs(suchasEbola,Zika,andNipahencephalitis)andalmostallknownpandemics(suchasinfluenza,HIV/AIDS,COVID-19)arezoonoses,meaningtheyarisefrommicrobesofanimalorigin.Thesemicrobesspillovertopeopleaftercontactwithwildlifeandlivestock.EstimatedcostsoffinancingaOneHealthinvestmentframeworkrangefromUS$22billiontoUS$31billion,orroughly1percentoftheglobaleconomiclossin2020fromCOVID-19.Suchaframeworkwouldincludeinvestmentstopreservekeywildlifehabitatsandresourcestoavoidwildlife-humanconflict;toimprovehousingcondi-tionstoavoidwildlifeintrusions;todevelopeffectivewastemanagementsystems;toimplementcommunityengagement,regulations,andenforcementtocurbillegalorBox4.6TheImpactofCitiesonClimateandtheEnvironment251unsafewildlifetrade;andtoenhancemarkets(suchasimproveddesignoffacilities,pestcontrol,oroff-siteslaughter).Additionally,investmentsinbiosecuritymeasuresandtechnologyforlivestockfacilitiesarecrucialtopreventthespreadofpathogensbetweenlivestockandfromlivestocktohumans.Suchinvestmentsmaybeledbydiffer-ententities/bodiesdependingonthenatureoftheinterventionsandtheresponsibilitiesofthedifferententities(suchaswastemanagement,urbanplanning,andmarketsuper-vision).Theywould,therefore,bebestdesignedandsupervisedwiththehelpofaOneHealthcoordinatingmechanism,suchasacross-sectorplatformthatbringstogetherthenecessarytechnicalexpertise.Suchpreventiveinvestmentreducestheneedforhealthsector–basedpreventionandpreparednessmeasures,andgeneratesenvironmen-talandhealthco-benefitsfrompreventingdeforestationandstrengtheninglivestockproduction.Source:AdaptedfromWorldBank2022.Box4.6continueda1.8–2.4percentreductioninglobalcroplandby2030,and80percentofthislosswilllikelytakeplaceinAfricaandAsia.Moreover,expansionisexpectedtotakeplaceoncroplandthatis1.77timesmoreproductivethantheglobalaverage(d’Amouretal.2017).Thislossofcropandotherlandduetourbanexpansionistiedtoaprojecteddeclineincerealandvegetableproductionacrossallpathways.Fortheworld’smaincereals—rice,wheat,andmaize—productiondeclinesof2–3percent,1–3percent,and1–4percent,respectively,areprojected.Projectedproductionreductionsforvegetablesare2–4percentandforpotatoes1–3percent.Thesereductionscorrespondtothefoodneedsofbetween122millionand1.39billionpeople(Chenetal.2020).Otherglobalestimatessimilarlyprojectatotalcropproductionlossofabout3.7percent,risingtolossesof5.6percentforAsiaand8.9percentforAfrica(d’Amouretal.2017).Poorlymanagedurbanexpansionisalsolinkedtoaprojecteddeclineinotherkeycropsandanimal-sourcefoodsimportantfordietsandnutrition.UnderSSP1,by2100,thelargestprojecteddeclinesamongcropswilloccurforfruits,vegetables,andtubers/roots,andthesmallestdeclinesforoils,sugar,andpulses(relativetocurrentlevelsofproduction)—seefigure4.7.Thesedeclinescouldpotentiallyleadtolessdiversediets.Althoughfoodproduc-tioninitiallydeclinesfasterunderSSP1thanunderSSP3to2040,itthenslowssothattheprojecteddecreaseinfoodproductionby2100somewhatalignsforboththeSSP1andSSP3scenarios.UnderSSP1,by2100,thenumberoflivestock,importanttotheproductionofanimal-sourcefoodsrichinprotein,alsofacesaprojecteddecline.Thelargestdeclinesareforpoultry(chickensandducks)andpigs,withsmallerdeclinesforruminants(cattle,sheep,andgoats).Thisprojectionalignswiththenotionthattheproductionofpoultryandpigsismoreperi-urbanthanthatofruminants.ThedeclinesforcropandlivestockproductionarelargestundertheSSP5scenario.Finally,theseprojectedimpactsshowimportantregionaldifferences.UnderSSP1by2100,thelargestdeclineforcerealsisinEurope,forvegetablesinNorthAmerica,andforpoultryinAfrica.252THRIVINGFigure4.7ProjectedcropandlivestocklossesfromurbanexpansionunderSSPscenariosby2040and2100Source:WorldBankStaffcalculationsbasedonChenetal.2020;Gilbertetal.2018;InternationalFoodPolicyResearchInstitute,“GlobalSpatially-DisaggregatedCropProductionStatisticsDatafor2010Versi2.0,”HarvardDataverse,V4,2019(https://doi.org/10.7910/DVN/PRFF8V).Note:SSP=SharedSocioeconomicPathway.CerealsVegetablesSugarFruitsOilsRootsPulses01234SSP1SSP2SSP3SSP4SSP501234SSP1SSP2SSP3SSP4SSP5b.Projectedcroplossesby2100e.Projectedlivestocklossesby210001234SSP1SSP2SSP3SSP4SSP5CerealsVegetablesSugarFruitsOilsRootsPulses01234SSP1SSP2SSP3SSP4SSP5ChickenHorseDuckGoatPigSheepBualoCattleChickenHorseDuckGoatPigSheepBualoCattlea.Projectedcroplossesby2040c.Projectedlivestocklossesby2040Projectedlosses(%)Projectedlosses(%)Projectedlosses(%)Projectedlosses(%)TheImpactofCitiesonClimateandtheEnvironment253Thedeclineinfoodproductionprojectedtoresultfromurbanexpansionreinforcestheimpor-tanceofpolicymeasuresandinstitutionalreformstopromotemorecompacturbandevelop-ment.Becauseurbanlandexpansionisprojectedtobefasterandremainhigherby2100underasustainablepathwaythanunderaregionalrivalrypathway,theprojecteddeclineinfoodproductionisslightlylargerunderthesustainablepathway.Ifanything,suchafindingplacesevenmoreimportanceonmeasurestoreducethecostsofverticaldevelopmentandreducecardependenceunderasustainablepathwaythanunderaregionalrivalrypathway.Alongwithencouragingmorecompactdevelopment,enhancingagriculturalproductivityandreducingfoodlossandwastealsocanoffsettheimpactsofhorizontalurbanexpansion.Sustainablyimprovingagriculturalproductivityrequiresacceleratedinnovation,asopposedtoincreasingtheuseofland,water,andotherproductioninputs.Innovationthroughtheinven-tion,adaptation,anddisseminationoflocallyadaptednewtechnologieswillbekey(Fuglieetal.2019).Inaddition,reducingfoodlossandwaste(forexample,viaagro-logistics,coldchains,improvedinfrastructure,easieraccesstomarkets,enhancedconsumerawareness,andimprovedurbanwasteandlandfillpolicies)notonlyreducesthecarbonfootprintandenviron-mentalstressesofthefoodsystembutalsoimprovesfoodsecuritybymakingmorefoodavail-ableinthesupplychainandloweringprices(WorldBank2020).Measurestoenhanceagriculturalproductivityareimportantalsobecause,inadditiontolandconstraints,agriculturefacesclimatechange–relatedproductivitylossesandurbanization-relatedlaborconstraints.Ontheonehand,foodsystemshavesignificantimpactsonclimatelargelythroughagricultureandlandusechanges.Ontheotherhand,climatechangehasslowedtheglobalgrowthinagriculturalproductivity.Anthropogenicclimatechangehasreducedglobalagriculturaltotalfactorproductivitybyanestimated21percentsince1961,equivalenttolosingthelastsevenyearsofproductivitygrowth(Ortiz-Bobeaetal.2021).ThesenegativeimpactsarelargestinAfricaandLatinAmericaandtheCaribbean.Inadditiontoclimatechange–relatedproductivitylosses,animportantcorollarytoincreasedurbanization(andstructuraltransformationmorebroadly)isthattheshrinkingshareoflaboremployedinfoodproductionmustbecomemuchmoreproductivetomeetthegrowingandchangingfooddemandofamuchlargernonagriculturalpopulation.19Otherindicationssuggest,however,thatthemovementofpeoplefromruraltourbanareasmayrelaxagriculturallandconstraints.Suchmovementwouldreleaserurallandsforagricultureandmakefarmslessfragmentedandpotentiallymoreefficient(Wangetal.2021).Finally,therisingneedforfoodcouldbeaddressedthroughinternationaltrade,combinedwithproductivitygainsincountriesthathavesuchcomparativeadvantages.Doingsomustbebalanced,however,bymitigatingtherisksassociatedwithfoodproductionbecomingconcentratedinanarrowersetofcountries.Howwillurbangrowthaffectcompetitionforwaterbetweencitiesandagriculturallands?CompetitionbetweencitiesandagricultureforwatersupplieswilllikelygrowSince1960,theglobalurbanpopulationhasquadrupled,pushingupthedemandforwaterbyurbanareas.Inmanycities,populationsaregrowingfasterthantheirwaterservices.RecentheadlinesfromCapeTown,SouthAfrica;Chennai,TamilNadu,India;and254THRIVINGSãoPaulo,Brazilrevealthatsomeoftheworld’smegacitieshavebeguntoface“dayzero”events,wherebywatersuppliesbecomeseverelylow(Zaverietal.2021).Althoughtheseeventshavegrabbedinternationalattention,theyarebynomeansunique.Scoresofsmallcitiesthroughouttheworldfacesimilarwatershortages(Zaverietal.2021).Themanysmallurbancenterscloselytiedtoruralareasaresurroundedbyagriculturallandexperiencingveryhighlevelsofwaterstressordroughtfrequency(FAO2020).Estimatessuggestthat150millionpeopleliveincitieswithperennialwatershortages,definedashavinglessthan100litersperpersonperdayofsustainablesurfacewaterorgroundwater.Manymorepeople—885million,roughlyequivalenttofourtimestheentirepopulationofBrazil—liveincitieswithaseasonalwatershortage(thatis,withmonthlywateravailabilityoflessthan100litersperpersonperday)andinsufficientflowsoccurringinatleastonemonthoftheyear(McDonaldetal.2011).Theexpectedadditionof2.5billionurbandwellersby2050willacceleratethistrend,increas-ingtheurbanwaterdemandbyaprojected50–80percent(Flörke,Schneider,andMcDonald2018;Garricketal.2019;WorldBank2018).Thisincreasewillbefuelednotonlybythegrowingnumbersofurbandwellersbutalsobylifestylesandconsumptionpatternsthataremorewater-intensive.Alargepartofthisincreasewillbedrivenbytheconsumptionofmoremeatanddairyproducts,althoughsuchproductscanhavevastlydifferentwaterfoot-prints,dependingonhowtheyareproduced(FAO2020).Suchadietaryshiftwillthereforeplayastrongroleinshapingwaterdemandinanagriculturesectorthatwillbecalledontomeetthegreaterdemandforfood,which,inturn,couldfuelcompetitionforwaterresources.Associetiesurbanize,demandfromindustry,energy,andotherservicescouldalsoincrease.EvidencefromSingaporesuggeststhatindustrialpoliciessubstantiallyincreasedtheshareofnondomesticwaterconsumptionto55percent(Hoekstra,Buurman,andvanGinkel2018).Overall,by2050almost1billionurbandwellersgloballywillliveinwater-stressedcities(Damaniaetal.2017).Growingdemandsnotwithstanding,thealterationoftheglobalhydrologiccyclestemmingfromclimatechangeisalsoleadingtoanincreaseinthenumberofextremewatershortageepisodes,makingwatersupplieslesspredictable(seealsochapter1).Withclimatechange,watershortageswillproliferatetootherpartsoftheworld,potentiallyaffectingevenmorecitydwellers.Inaworst-casescenario,estimatessuggestthatawarmingworldcouldmakeeventslikethedayzerodrought100timesmorelikelythantheywereintheearlytwentieth-centuryworldincertainregions(Pascaleetal.2020).Watershortages,andtherestrictionsputinplacetodealwiththem,havehighcostsforbothurbanresidentsandbusinesses.Dwindlingwatersuppliescancostacityuptoa12-percentage-pointlossinGDP(Zaverietal.2021).Meanwhile,unreliablewatersuppliesandwatershortagesadverselyaffectproductivitybyreducingworkers’incomes,inducinglowersalesforfirms,andworseninghealthoutcomes(DesbureauxandRodella2019;IslamandHyland2019),therebyreinforcingdeepinequalitiesandtrappingvulnerablehouseholdsincyclesofpoverty(seechapter3).Thesefindingshighlightthecriticalimportanceofinvestinginpoliciesandinfrastructurethatcanenhanceurbanwaterresilience.Ascitiesgrow,theyoftenencroachonsurroundingareastosatisfytheirthirstTheencroachmentofdevelopmentintothefloodbanksofrivers,intowetlands,anduponnaturalinfrastructureoftenoccursinresponsetothegrowingdemandforscarcelandinTheImpactofCitiesonClimateandtheEnvironment255cities.Suchencroachmentdestroysthenaturalstorageandspongesthatregulatecities’watersupplies,largelybecausehouseholdsanddeveloperslacktheincentivetopreservenaturalinfrastructureastheprivatebenefitsderivedfromdevelopinglanddonotfullyreflectthebenefitsoftheseecosystemstosociety(TaylorandDruckenmiller2022).Whenrenewablelocalwaterresourcesareinsufficienttomeetincreasingdemandandpercapitaconsump-tion,theresultmaybeoverexploitationofsurfacewaterandgroundwaterresources(includingtheconsumptionoffossilgroundwater)ortheuseofadditionalexternalwaterresources(Hoekstra,Buurman,andvanGinkel2018;alsoseebox4.7).Thesepressuresareespeciallyexacerbatedinareasofunplannedgrowthontheperipheriesofcities,whichsufferfromalackofmunicipalwaterconnectionsandinadequatepublicservicedelivery.Overall,thesetrendscreatepressureandincreasecompetitionforwateramongcities,aswellasbetweencitiesandruralareas.Waterreallocationfromruraltourbanareasisfastbecomingacommonstrategytomeetfresh-waterneedsinever-thirstiercities(Garricketal.2019;Zaverietal.2021).Globally,69citieswithapopulationof383millionreceivewaterthroughreallocationprojectsthattypicallytransferwaterfromsurroundingruralareas(Garricketal.2019).Globalestimatessuggestthatmovinganestimated16billioncubicmetersofwateracrossnearly13,000kilometersiscostly(Garricketal.2019).Reallocationcanoccurinotherwaysaswell.Whenareasurbanize,reallocatingwaterfromagriculturaltourbanusescanoccurorganicallyandsomewhatincon-spicuouslythroughlandusechange;however,othermethodsofreallocationcanbemorecon-spicuous.Insomeplaces,thefailureoflocalutilitiesortheinadequacyofurbanwatersupplyinfrastructurepromptsinformalwatervendorstofillthegapbypumpingwaterfromsur-roundingagriculturaltubewells.Elsewhere,reallocationoccursthroughexpropriationofwaterrights,watergrabbingbycities,orsingle-sourcewaterresourcedevelopment.Forallthesereasons,reallocationispoliticallycomplexbecauseitmayentaildeprivingfarmersofacruciallivelihoodresourceandinput.Intheabsenceofequitablelegalarrange-ments,transfersofwaterfromruralareascanbecoercive.20Mostlegalsystemsprioritizedrinkingwater,andoftenindustrialwater,aboveagriculturalwaterinallocationdecisions.Suchdecisionscanreducethewateravailableforirrigatedurbanandperi-urbanagriculture(Hoekstra,Buurman,andvanGinkel2018).Manycountriesalsohavepoorlydevelopedwaterrightssystems,andadaptiveallocationsystemsarestillfewandfarbetween.Reallocationsareoftencrisis-driven,adhoc,andthereforeverypoorlydesigned.Thesetrendssuggesttheneedforcapacitybuildingtocreatethelegalandinstitutionalpreconditionsformoreequitablearrangements.Variousotherfactorscanalsoinfluencethesetransfers.Investmentsinefficientuseofirriga-tionwaterandreuseofwastewaterinirrigationcanenableagriculturalareastogrowmorewithlesswater.A10percentincreaseintheefficientuseofirrigationwatercouldfreeupenoughwatertoreduceurbanwaterdeficitsby2.7billioncubicmetersby2050,benefitingalmost240millionpeople(Flörke,Schneider,andMcDonald2018).Sucheffectswilllikelybemostvisibleandrelevantforcitiesinwater-stressedbasinshighlydependentonirrigation.Atthesametime,itwillbeequallyimportanttoimplementpoliciesthatencouragecitiestouselesswater.Indeed,urbandemandmanagementwillbecritical.Dynamicallyefficientvol-umetricwaterpricing,forexample,canadjustthepriceofwatertobettermatchthescarcitycitiesface.Whencitiesallowutilitiestocarefullyadjustthepriceofwateraccordingtoitsscarcity,theutilitiescanavoidtheneedtoinvestinwater-augmentingtechnologiesandthussavemoney,reducewaterfootprints,andkeepwatercostslowerinthelongrun.Othertech-nologies,suchassmartwatermetersandwater-savingandwater-reusingappliances,canhelphouseholdsreducetheirwaterfootprintwithlittlesacrifice(Zaverietal.2021).256THRIVINGSinkingcities:GroundwaterdepletionandlandsubsidenceUrbanwatersystemsinteractwithmanyothersystemsandareaffectedinindirectways.Aclearexampleofcomplexinteractionsonanurbanscaleiswhenthelandbeginstoslowlysink,aphenomenonknownaslandsubsidence.Landsubsidenceoftenresultsfromunsustainablegroundwaterextractiontomeetincreasingdemandorfromdrainagetomakewetlandssuitableforurbanexpansion(Hoekstra,Buurman,andvanGinkel2018).HighratesofgroundwaterextractionaroundmajorcitiesintheAmericas,Asia,Europe,andtheMiddleEasthaveledtolocalizeddepletionthat,inmanycases,hasresultedinlandsubsidence.Intheseenvironments,groundwaterdepletionstemsfromhighpumpingratesandthereductioninrechargeasthecoverageofimpervioussurfaceareasincreases.Withcontinuedurbanization,espe-ciallyincoastalcitiesandcitiesindeltas,thetrendsofgroundwaterdepletionandlandsubsidencewilllikelycontinue,withpotentiallyhighersusceptibilitytofloodingandgroundwatersalinization(Lall,Josset,andRusso2020).Newestimatessuggestthat22percentoftheworld’smajorcitiesareinpotentialsubsidenceareas,with57percentofthesealsolocatedinflood-proneareas.(Herrera-Garcíaetal.2021;Lall,Josset,andRusso2020).Landsubsidencealsocompromisesthestructuralintegrityofexistingbuildings.Therisksposedbylandsubsidencecanleadtolowerpropertyvaluesandthusaffecttheincomestreamsofpropertydevelopersandtaxrevenuesoflocalgovernments(WorldBank2021;YooandPerrings2017).Intheabsenceofwell-functioningmarketsandinstitutions,theseexternalitiesareoftennotinternalized.OnestarkexampleoflandsubsidenceisJakarta,Indonesia,theworld’sfastest-sinkingcity.Onaverage,ithassubsidedmorethan3.5meterssincethe1980sandcontinuestosinkatratesofupto20centimetersayear.Theabsenceofaccesstoreliablepipedwater,whichforcesuserstoresorttounregulatedgroundwaterabstraction,isamajorcauseofgroundwateroverexploitation.AcrossIndonesia,government-ownedwaterenterprisesprovideonly9percentoftotaldomesticwaterdemand,andalmosthalfofallhouseholdsandmostcommercialandindustrialpremisesrelyonsuppliesfromonsitegroundwater(WorldBank2021).MapB4.7.1showsthecomparativelandsubsidenceratesacrossIndonesiaandAsianmegacities(WorldBank2021).Jakartaalreadyliessignificantlybelowsealevel,andlandsubsidenceincreasinglyexposesittohighcoastalandinlandfloodrisks,evenwithoutconsideringsea-levelrise.Partlyinresponsetothegrowingpressures,theIndonesiangovernment,inadramaticmoveinJanuary2022,passedalawtoofficiallymovethecapitalfromJakartatoanundevelopedjungletractinEastKalimantan,Borneo.Box4.7TheImpactofCitiesonClimateandtheEnvironment257Box4.7continuedSources:WorldBank2021.LandsubsidenceratesacrossIndonesiafromAndreasetal.2018.LandsubsidenceratesinAsianmegacities(insetgraphatbottomright)fromKanekoandToyota2011andTakagietal.2016.Note:RatesforIndonesiaincentimetersperyear(cm/yr).RatesforAsianmegacitiesinmeters(m).MapB4.7.1ComparativelandsubsidenceratesacrossIndonesia,2000s,andAsianmegacities,1900–2010CitiesalsooftenusetheirsurroundingregionsassinksforwasteFarmlanddownstreamfromcitiesoftenreliesoncontaminatedurbanwastewaterforirrigation.Globally,65percentofallirrigatedcroplandwithin40kilometersdownstreamofcitiesreliesheavilyonwastewaterflows(Theboetal.2017).Insomeways,wastewaterirrigationprovidesatriplewin:itreducestheamountofwaterthatneedstobeextracted;itoffersasolutiontodischargingurbanwastewater,oftenwithouttreatment;andsomedissolvednutrientscanactasafertilizertoboostagriculturalyields,which,inturn,canhelpmeetthegrowingurbanfooddemand(Damaniaetal.2019).Nevertheless,wastewaterreuseforirrigationisnotalwaysapanacea.Urbanwastewatercanhavehighconcentrationsofheavymetals,particularlywastewaterfromcitiesinlow-andmiddle-incomecountrieswhereheavyindustryismorelikelytobepresent(Damaniaetal.2019).Ifnotcarefullymanaged,wastewaterirrigationcancauseenvi-ronmentaldamage,harmcropquality,and,inturn,potentiallyexposesome885millionurbanresidentswhoconsumetheseagriculturalproductstoserioushealthrisks.TheEuropeanCommission,theUSEnvironmentalProtectionAgency,andtheWorldHealth258THRIVINGOrganizationallhaveminimumwaterqualityrequirementsandguidelinesforwastewaterthatshouldbeproperlyenforcedtofosterbestpracticesandimplementation(Damaniaetal.2019).Adoptingfitforpurposetreatmentforwastewater—thatis,treatingwaste-wateraccordingtotheendusers’needs—makesthisresourceawin-winandrealisticoptionfornonconventionalsourcesofwaterandnutrientsforagriculture(Hoekstra,Buurman,andvanGinkel2018).Becauseasignificantshareofurbanwaterisusedbutnotconsumed,itsreuseinagricultureaftertreatmenthasgreatpotential,particularlyinwater-scarcecountries(Hoekstra,Buurman,andvanGinkel2018).Theseexampleshighlightthatwaterreallocationplaysanimportantroleinurbandevelop-ment.Questionsremain,however,abouthowtrulyeffective,equitable,andsustainablewaterreallocationis.Aproperunderstandingoftherangeofinteractionsbetweenurbanareasandtheirlocalhinterlandsiscriticaltofullyunderstandingthelong-runwatersecurityofcities.SummaryandconclusionsThischapterhasanalyzedsomeofthewaysthedevelopmentofcitiescanaffecttheclimateand,moregenerally,theenvironment.Indoingso,ithaspaidparticularattentiontotheinterrelatedrolesofurbanform(thatis,whetheracitydevelopshorizontally,asistypicalofmanylower-incomecountrycities,orverticallythroughbuildingtaller)andchoicesofurbantransportationmode.Policiesandinstitutionalreformsthatimprovetheworkingoflandandpropertymarkets,enhanceurbanplanning,andencourageamoveawayfromprivateinternalcombustionenginevehiclestowardpublictransportationcouldhelpcitiesfollowamoreverticalandcompactdevelopmenttrajectory.Inadditiontopromotingproductivitygrowth,suchatrajectorycancontributetoreductionsinacity’sCO2emissionsandairpollutantssuchasPM2.5,whilehelpingtobothpreventdeforestationandconservefertileagriculturallandontheperipheriesofcitiesthatisamajorsourceoffood.Suchpoliciesaremorelikelytobesuccessfulwhenverticaldevelopmentiscombinedwithgoodurbanandarchitecturaldesign,aswellaswithinvestmentsinstreetpavingandlightingthatpromotewalkability.And,wheninvestmentsinmasstransitsystemsarecombinedwithdemand-sidepoliciesthatreducetheincentivestodrive,theevidencesuggeststhatsuchinvestmentscan,ifwelldesignedandwellimplemented,alsohelpdirectlyreducepollutioninhighlycongestedandpollutedcities.Theseinvestmentsareparticularlyrelevanttolargecitiesthatcurrentlylackmasstransitsystems.Bycontrast,noevidenceexiststosuggestthatsuchinvestmentshaveanydirectimpactonpollutioninlesscongestedandpollutedcities;however,iftheseinvestmentshelppromotemorecompactdevelopmentinrapidlygrowingcities,theymayhavelonger-termindirectbenefitsintermsofhelpingreducepollution.Thatsaid,amoreverticalandcompactdevelopmenttrajectoryisnotnecessarilyapanaceafortheenvironmentalstressestowhichpoorlymanagedurbangrowthcangiverise.Buildingtallerimpliestheuseofmorebuildingmaterialswhoseproductionisitselfhighlycarbon-intensive.Thus,amorevertical,compactcitymayembedmorecarboninitsbuildingsthanamorehorizontal,lesscompactcity.Compactdevelopment,therefore,involvesatrade-offbetweenthelargeup-frontCO2emissionsinvolvedinbuildingtallerandthelong-termloweremissionsassociatedwiththemorecompacturbanform.Currentresearchisunclearabouttheconditionsunderwhichmorecompactdevelopmentofacitywillhavenetbenefi-ciallong-termimpactsonitscontributiontoclimatechange.Whatiscleararethebeneficialeffectsofmorecompactdevelopmentonacity’sproductivityandresilience,theconservationofagriculturalandforestedland,andlocalairpollutionlevels.TheImpactofCitiesonClimateandtheEnvironment259Moreover,thecalculusofthenetlong-termimpactsofmorecompactdevelopmentonacity’scontributiontoclimatechangewilllikelychangeovertime,dependingontechnologi-calprogressintheconstructionandtransportationsectors.Aselectricvehiclesbecomemoreprevalent,someoftheenvironmentalbenefitsofmorecompactdevelopment—atleastintermsofCO2emissionsandlocalairpollution—arelikelytowane.Bycontrast,anyreductioninthecarboncontentofconcrete,steel,andglassresultingfromimprovedtechnologiesanddecar-bonizationofacountry’senergysupplywillincreasethenetenvironmentalbenefitsofmorecompactdevelopment.Inthinking,then,aboutthefuturedevelopmentoftheircities,urbanpolicymakersneedtobecognizantoftechnologicaltrends,especiallybecauseurbandevelop-mentisahighlypath-dependentprocesswherebydecisionsthataffectthebuiltenvironmentandurbanformtodaycanreverberatefordecades,evencenturies,tocome.Finally,thischapterhashighlightedthestressesthatpoorlymanagedurbanizationcanplaceonacity’swatersupplyandtheresultingpotentiallyharmfulcompetitionforwaterbetweenurbanandruralareas.Bettermanagementofurbanwaterdemandthroughscarcitypricingandbehavioralaswellastechnologicalsolutionsthatreducewaterusewithinhomesandbybusinessesarecriticaltohelpingreducesuchconflict.Carefullymanagedreuseofurbanwastewaterforagricultureaftertreatmentalsohasgreatpotential,particularlyinwater-scarcecountries.THRIVING260Notes1.Usingthesamedefinitionofcitiesasthisreport,Lalletal.(2021)estimatethatin2015thetotalbuilt-upareaofcitiesgloballywas294,550squarekilometers,or0.20percentofthegloballandarea.Otherdefinitionsofcitiesgeneratehigherestimatedsharesoflandareaforcitiesglobally,buttheyare,nevertheless,stillsmall.Forexample,anighttimelights–baseddefinitionofcitiesderivedbytheCenterforInternationalEarthScienceInformationNetworkimpliesthatcitiesoccupied2.44percentoftheworld’slandareain2010(CIESIN2013).2.Otherimportanttechnologicaladvancesinthelatenineteenthcenturythatenabledtheconstructionoftallbuildingsweretheabilitytomass-producesteelandthedevelopmentofbettertechniquesformeasuringandanalyzingstructuralloadsandstresses.Duringthe1920sand1930s,theinventionofelectricarcweldingandfluorescentlightbulbs(whosebrightlightallowedpeopletoworkfartherfromwindowsandgeneratedlessheatthanincandescentbulbs)furtherfacilitatedtallbuildingconstruction(fromtheHowProductsAreMadewebpage,“Skyscraper,”http://www.madehow.com/Volume-6/Skyscraper.html#ixzz7Re2detju).3.Bymodern-daystandards,itwouldbemoreaccuratetodescribetheHomeInsuranceBuildingasamidrisebuilding.4.Atallbuildingisdefinedasabuildingwithaheightofatleast55meters.5.TheinsightsbyLalletal.(2021)onverticallayeringarederivedfromarestrictedglobalsampleof400citiesforwhichbuildingheightdataareavailable.Thissampleisrepresentativeofthefullglobalsampleofmorethan10,000cities.TheirbuildingheightdatacomefromtheGermanAerospaceCenter’sWorldSettlementFootprint3Dproduct,whichprovidesestimatesofbuildingheightatanaggregated90-meterresolutionderivedfromsatelliteimagery.6.Citiesinhigher-incomecountriesalsohaveastrongertendencytogrowverticallybecausehigheragriculturalproductivitymakeslandatacity’sedgemoreexpensive.Additionally,becausewagesarehigher,theopportunitycostofcommutingtimeisalsohigherinrichercountries,whichprovidesafurtherincentiveforvertical,asopposedtohorizontal,development(Jedwab,Barr,andBrueckner2022).7.TheGDPpercapitadataquotedinthisparagraphcomefromtheMaddisonProjectDatabase2020,https://www.rug.nl/ggdc/historicaldevelopment/maddison/releases/maddison-project-database-2020.8.Oncecompleted,theMohamedVITower—currentlyunderconstructioninRabat,Morocco—willbethetallestinAfrica.Formoreinformation,seetheBESIXwebpageabouttheproject,https://www.besix.com/en/projects/mohammed-vi-tower.At820feet(250meters)and55stories,thetowerwillstillbelessthan60percentoftheheightoftheEmpireStateBuilding.9.Bycontrast,fortheDemocraticPeople’sRepublicofKoreaandseveralGulfstates,resultsfromJedwab,Barr,andBrueckner(2022)indicateexcessivetallbuildingconstructiongiventhesecountries’levelsofurbandevelopmentandagriculturalrents.10.Becauseurbanlandisrarelyconvertedbackintoruralland,suchacontractionshouldbethoughtofasbeingrelativetowhatwouldhaveotherwiseoccurred—thatis,horizontalexpansiontakesplaceataslowerpacethanwouldhaveotherwiseoccurred,resultinginacitythatissmallerinequilibriumrelativetothecounterfactualscenario.261261TheImpactofCitiesonClimateandtheEnvironment11.AnIVstrategyisastandardeconometrictechniquedesignedtocontrolforvariouspotentialsourcesofendogeneity,includingreversecausation.12.Incontrasttotheresultsforpopulationandarea,theresultsfortheimpactsofacity’stotalheightonitsnighttimelightsarebasedonordinaryleastsquaresestimationonly,whichimpliesapotentialupwardbiasdueto,forexample,possiblereversecausation.Intheresultsforpopulationandarea,however,usingIVstrategiestodealwithreversecausationleadstohigherestimatedimpactsoftallbuildingsintermsofbothexpandingpopulationandconservingarea.Thisresultmayoccurbecausethesestrategiesalsohelpcontrolforclassicalmeasurementerror,whichbiasesestimatedcoefficientstowardzero(Wooldridge2013).13.Usingdatafor25addressesinsevenUKcities,Pomponietal.(2021)conductasimulationexercisethatsuggeststhatahigh-density,low-risepatternofurbandevelopmentisassociatedwithmuchlowerlife-cycleGHGemissionsthanahigh-density,high-risepattern.InadditiontobeingbasedonasmallnumberofUKaddresses,however,theiranalysisdoesnotaccountforthelowerCO2emissionsthattendtobeassociatedwithmorecompactdevelopment.Furthermore,ahigh-density,low-risepatternofurbandevelopmentisexactlywhatcharacterizesmanycitiesindevelopingcountriestoday.Thispatternmanifestsitselfintheformofslumsandovercrowding,whichareassociatedwithpoverty.14.Thefractaldimensionisameasureofthe“degreeofinhomogeneity”ofageometricobject(Mandelbrot1982).Itcanbethoughtofasameasureofthecomplexityofageometricobject’sshape.15.Asanexample,Ostermeijeretal.(2022)citetheStreetcarConspiracyof1949,whenGeneralMotorsandothercarmanufacturerswereconvictedofmonopolizingthesaleofbusesandaccusedofcontrollingthetransitsystemtodismantleexistingstreetcarnetworks.16.IntheUnitedStates,theconstructionofhighwaysand,beforethat,railwaysthroughcitieshascontributedtogreatersegregationwithincitiesbycreatingphysicalseparationbetweenneighborhoods,oftenreinforcingexistingpatternsofracialsegregation(Ananat2011;BadgerandCameron2015).17.The“ironlawofcongestion”assertsthatanyinvestmentininfrastructurethatreducescongestionandthereforethecostofdrivingintheshortrunwillinducemoredriving,withtheresultthatcongestioneventuallyreboundstoitsoriginallevel.EvidenceofthisoutcomeisprovidedbyChenandKlaiber(2020);DurantonandTurner(2011);Garcia-López,Pasidis,andViladecans-Marsal(2020);andHsuaandZhang(2014)forChinese,US,European,andJapanesecities,respectively.18.Gendron-Carrieretal.(2022)alsofindnoevidencethatsubwaysystemexpansionsaffect(positivelyornegatively)levelsofairpollution.19.Historically,improvedagriculturalproductivityhasbeenthoughtofasadriverofurbanization(thatis,labor-savingtechnologicalprogressinagriculturereleaseslaborfromtheland,therebyfacilitatingindustrializationandurbanization).20.Whenconfrontedbyintermittentwatersupply,IndiancitiessuchasChennaidrawonwaterresourcesfromneighboringruraldistricts,oftenattheexpenseoffarmingcommunities,therebyfuelingurban-ruraltensions(Singhetal.2021;Varadhan2019;Zaverietal.2021).THRIVING262ReferencesAbuHatab,A.,M.Cavinato,A.Lindermer,andC.Lagerkvist.2019.“UrbanSprawl,FoodSecurityandAgricultureSystemsinDevelopingCountries:ASystematicReviewofLiterature.”Cities94:129–42.Acharya,G.,E.Cassou,S.Jaffee,andE.K.Ludher.2021.RICHFood,SmartCity:HowBuildingReliable,Inclusive,Competitive,andHealthyFoodSystemsisSmartPolicyforUrbanAsia.Washington,DC:WorldBank.Ahlfeldt,G.M.,andJ.Barr.2022.“TheEconomicsofSkyscrapers:ASynthesis.”JournalofUrbanEconomics129:103419.Ahlfeldt,G.M.,andR.Jedwab.2022.“TheGlobalEconomicandEnvironmentalEffectsofVerticalUrbanDevelopment.”Backgroundpaperpreparedforthisreport,WorldBank,Washington,DC.Ahlfeldt,G.M.,andE.Pietrostefani.2019.“TheEconomicEffectsofDensity:ASynthesis.”JournalofUrbanEconomics111:93–107.Ananat,E.O.2011.“TheWrongSide(s)oftheTracks:TheCausalEffectsofRacialSegregationonUrbanPovertyandInequality.”AmericanEconomicJournal:AppliedEconomics3(2):34–66.Andreas,H.,H.Abidin,I.Gumilar,T.P.Sidiq,D.A.Sarsito,andD.Pradipta.2018.“InsightintotheCorrelationbetweenLandSubsidenceandtheFloodsinRegionsofIndonesia.”InNaturalHazards—RiskAssessmentandVulnerabilityReduction,editedbyJoséSimãoAntunesDoCarmo.IntechOpen.Badger,E.,andD.Cameron.2015.“HowRailroads,HighwaysandOtherMan-MadeLinesRaciallyDivideAmerica’sCities.”WashingtonPost,July15,2015.Baum-Snow,N.2007.“DidHighwaysCauseSuburbanization?”QuarterlyJournalofEconomics122(2):775–805.Baum-Snow,N.,L.Brandt,V.Henderson,M.Turner,andQ.Zhang.2016.“Highways,MarketAccess,andUrbanGrowthinChina.”IGCWorkingPaperC-89114-CHN-1,InternationalGrowthCentre,London.Bernard,A.2014.Lifted:ACulturalHistoryoftheElevator.NewYork:NewYorkUniversityPress.Bertaud,A.,andJ.K.Brueckner.2004.“AnalyzingBuildingHeightRestrictions:PredictedImpacts,WelfareCosts,andaCaseStudyofBangalore,India.”PolicyResearchWorkingPaper3290,WorldBank,Washington,DC.BlaudindeThé,C.,B.Carantino,andM.Lafourcade.2021.“TheCarbon‘Carprint’ofUrbanization:NewEvidencefromFrenchCities.”RegionalScienceandUrbanEconomics89(C).Chen,G.,X.Li,X.Liu,Y.Chen,X.Liang,J.Leng,X.Xu,etal.2020.“GlobalProjectionsofFutureUrbanLandExpansionunderSharedSocioeconomicPathways.”NatureCommunications11:537.Chen,W.,andH.A.Klaiber.2020.“DoesRoadExpansionInduceTraffic?AnEvaluationofVehicle-KilometersTraveledinChina.”JournalofEnvironmentalEconomicsandManagement104:102387.263263TheImpactofCitiesonClimateandtheEnvironmentCherif,R.,F.Hasanov,andA.Pande.2017.“RidingtheEnergyTransition:OilBeyond2040.”IMFWorkingPaperWP/17/120,InternationalMonetaryFund,Washington,DC.CIESIN(CenterforInternationalEarthScienceInformationNetwork).2013.“LowElevationCoastalZoneUrban-RuralPopulationandLandAreaEstimates,Version2.”Technicalreport,CIESIN,ColumbiaUniversity,Palisades,NY.Crippa,M.,E.Solazzo,D.Guizzardi,F.Monforti-Ferrario,F.N.Tubiella,andA.Leip.2021.“FoodSystemsAreResponsibleforaThirdofGlobalAnthropogenicGHGEmissions.”NatureFood2:198–209.Damania,R.,S.Desbureaux,M.Hyland,A.Islam,S.Moore,A.-S.Rodella,J.Russ,etal.2017.UnchartedWaters:TheNewEconomicsofWaterScarcityandVariability.Washington,DC:WorldBank.Damania,R.,S.Desbureaux,A.S.Rodella,J.Russ,andE.Zaveri.2019.QualityUnknown:TheInvisibleWaterCrisis.Washington,DC:WorldBank.d’Amour,C.B.,F.Reitsma,G.Baiocchi,S.Barthel,B.Güneralp,K.Erb,H.Haberl,etal.2017.“FutureUrbanLandExpansionandImplicationsforGlobalCroplands.”ProceedingsoftheNationalAcademyofSciences114(34):8939–44.deBruin,S.,J.Dengerink,andJ.vanVliet.2021.“UrbanizationasDriverofFoodSystemsTransformationandOpportunitiesforRuralLivelihoods.”FoodSecurity13:781–98.Desbureaux,S.,andA.Rodella.2019.“DroughtintheCity:TheEconomicImpactofWaterScarcityinLatinAmericanMetropolitanAreas.”WorldDevelopment114:13–27.Duranton,G.,andM.Turner.2011.“TheFundamentalLawofRoadCongestion:EvidencefromUSCities.”AmericaEconomicReview101(6):2616–52.Ellis,P.,andM.Roberts.2016.LeveragingUrbanizationinSouthAsia:ManagingSpatialTransformationforProsperityandLivability.Washington,DC:WorldBank.Ewing,R.,S.Hamidi,J.B.Grace,andY.D.Wei.2016.“DoesUrbanSprawlHoldDownUpwardMobility?”LandscapeandUrbanPlanning148(April):80–88.Falk,N.,andJ.Rudlin.2018.“LearningfromInternationalExamplesofAffordableHousing.”Shelter/TheURBEDTrust.FAO(FoodandAgricultureOrganizationoftheUnitedNations).2020.TheStateofFoodandAgriculture2020.OvercomingWaterChallengesinAgriculture.Rome:FAO.Flörke,M.,C.Schneider,andR.McDonald.2018.“WaterCompetitionbetweenCitiesandAgricultureDrivenbyClimateChangeandUrbanGrowth.”NatureSustainability1:51–58.Fuglie,K.,M.Gautam,A.Goyal,andW.Maloney.2019.HarvestingProsperity:TechnologyandProductivityGrowthinAgriculture.Washington,DC:WorldBank.Gaduha,A.,T.Gračner,andA.D.Rothenberg.2022.“LifeintheSlowLane:UnintendedConsequencesofPublicTransitinJakarta.”JournalofUrbanEconomics128:103411.Garcia-López,M.À.,I.Pasidis,andE.Viladecans-Marsal.2020.“CongestioninHighwayswhenTollsandRailroadsMatter:EvidencefromEuropeanCities.”WorkingPapers2020/11,Institutd’EconomiadeBarcelona(IEB).THRIVING264Garrick,D.,L.DeStefano,W.Yu,I.Jorgensen,E.O’Donnell,L.Turley,I.Aguilar-Barajas,etal.2019.“RuralWaterforThirstyCities:ASystematicReviewofWaterReallocationfromRuraltoUrbanRegions.”EnvironmentalResearchLetters14(4):043003.Gendron-Carrier,N.,M.Gonzalez-Navarro,S.Polloni,andM.A.Turner.2022.“SubwaysandUrbanAirPollution.”AmericanEconomicJournal:AppliedEconomics14(1):164–96.Gilbert,M.,G.Nicolas,G.Cinardi,T.P.VanBoeckel,S.O.Vanwambeke,G.R.WilliamWint,andT.P.Robinson.2018.“GlobalDistributionDataforCattle,Buffaloes,Horses,Sheep,Goats,Pigs,ChickensandDucksin2010.”ScientificData5(1):180227.https://doi.org/10.1038/sdata.2018.227.Giridharan,R.,andR.Emmanuel.2018.“TheImpactofUrbanCompactness,ComfortStrategiesandEnergyConsumptiononTropicalUrbanHeatIslandIntensity:AReview.”SustainableCitiesandSociety40(July):677–87.Glaeser,E.2011.“HowSkyscrapersCanSavetheCity.”Atlantic,March2011.https://www.theatlantic.com/magazine/archive/2011/03/how-skyscrapers-can-save-the-city/308387/.Gottmann,J.1966.“WhytheSkyscraper?”GeographicalReview56:190–212.Hall,D.,andN.Lutsey.2018.“EffectsofBatteryManufacturingonElectricVehicleLife-CycleGreenhouseGasEmissions.”InternationalCouncilonCleanTransportation,Washington,DC.Hanna,R.,G.Kreindler,andB.A.Olken.2017.“CitywideEffectsofHigh-OccupancyVehicleRestrictions:Evidencefrom“Three-in-One”inJakarta.”Science357(6346):89–93.Hausfather,Z.2019.“Factcheck:HowElectricVehiclesHelptoTackleClimateChange.”Eco-Business,May14,2019.https://www.eco-business.com/news/factcheck-how-electric-vehicles-help-to-tackle-climate-change/.Heger,M.,D.Wheeler,G.Zens,andC.Meisner.2019.MotorVehicleDensityandAirPollutioninGreaterCairo:FuelSubsidyRemovalandMetroLineExtensionandTheirEffectonCongestionandPollution.Washington,DC:WorldBank.Herrera-García,G.,P.Ezquerro,R.Tomás,M.Béjar-Pizarro,J.López-Vinielles,M.Rossi,R.M.Mateos,etal.2021.“MappingtheGlobalThreatofLandSubsidence.”Science371(6524):34–36.Hess,D.B.,T.Tammaru,andM.vanHam.2018.“LessonsLearnedfromaPan-EuropeanStudyofLargeHousingEstates:Origin,TrajectoriesofChangeandFutureProspects.”InHousingEstatesinEurope:Poverty,EthnicSegregationandPolicyChallenges,editedbyD.B.Hess,T.Tammaru,andM.vanHam,3–31.Cham,Switzerland:SpringerInternationalPublishing.Hoekstra,A.Y.,J.Buurman,andK.C.H.vanGinkel.2018.“UrbanWaterSecurity:AReview.”EnvironmentalResearchLetters13(5):053002.Hopkins,F.,J.Ehleringer,S.Bush,R.Duren,C.Miller,C.Lai,Y.Hsu,etal.2016.“MitigationofMethaneEmissionsinCities:HowNewMeasurementsandPartnershipsCanContributetoEmissionsReductionStrategies.”Earth’sFuture4:408–25.Hsua,W.,andH.Zhang.2014.“TheFundamentalLawofHighwayCongestionRevisited:EvidencefromNationalExpresswaysinJapan.”JournalofUrbanEconomics81:65–76.Hunt,D.B.2018.“PublicHousinginUrbanAmerica.”OxfordResearchEncyclopediaofAmericanHistory.Oxford,UK:OxfordUniversityPress.265265TheImpactofCitiesonClimateandtheEnvironmentIslam,A.,andM.Hyland.2019.“TheDriversandImpactsofWaterInfrastructureReliability—AGlobalAnalysisofManufacturingFirms.”EcologicalEconomics163(C):143–157.Jedwab,R.,J.Barr,andJ.Brueckner.2022.“CitiesWithoutSkylines:WorldwideBuilding-HeightGapsandtheirPossibleDeterminantsandImplications.”https://www.remijedwab.com/_files/ugd/ea9b22_1310bef406ba40b3be2ba5e98ecf441d.pdf.Jiang,L.,andB.O’Neill.2017.“GlobalUrbanizationProjectionsfortheSharedSocioeconomicPathways.”GlobalEnvironmentalChange42:193–99.Kaneko,S.,andT.Toyota.2011.“Long-TermUrbanizationandLandSubsidenceinAsianMegacities:AnIndicatorsSystemApproach.”InGroundwaterandSubsurfaceEnvironments:HumanImpactsinAsianCoastalCities,editedbyMakotaTaniguchi.SpringerScience&BusinessMedia.Lai,D.,W.Liu,T.Gan,K.Liu,andQ.Chen.2019.“AReviewofMitigatingStrategiestoImprovetheThermalEnvironmentandThermalComfortinUrbanOutdoorSpaces.”ScienceoftheTotalEnvironment661(April):337–53.Lall,U.,L.Josset,andT.Russo.2020.“ASnapshotoftheWorld’sGroundwaterChallenges.”AnnualReviewofEnvironmentandResources45:171–94.Lall,S.,M.Lebrand,H.Park,D.Sturm,andA.Venables.2021.PancakestoPyramids:CityFormtoPromoteSustainableGrowth.Washington,DC:WorldBank.Mandelbrot,B.1982.TheFractalGeometryofNature.SanFrancisco:W.H.FreemanandCo.McDonald,R.I.,I.Douglas,C.Revenga,R.Hale,N.Grimm,J.Grönwall,andB.Fekete.2011.“GlobalUrbanGrowthandtheGeographyofWaterAvailability,Quality,andDelivery.”Ambio40(5):437–46.Ortiz-Bobea,A.,T.R.Ault,C.M.Carrillo,R.G.Chambers,andD.B.Lobell.2021.“AnthropogenicClimateChangeHasSlowedGlobalAgriculturalProductivityGrowth.”NatureClimateChange11:306–12.Ostermeijer,F.,H.R.A.Koster,J.vanOmmeren,andV.M.Nielsen.2022.“AutomobilesandUrbanDensity.”JournalofEconomicGeography22(5):1073–95.Paoli,L.,andT.Gül.2022.“ElectricCarsFendOffSupplyChallengestoMorethanDoubleGlobalSales.”InternationalEnergyAgency,Paris.Pascale,S.,S.B.Kapnick,T.L.Delworth,andW.F.Cooke.2020.“IncreasingRiskofAnotherCapeTown‘DayZero’Droughtinthe21stCentury.”ProceedingsoftheNationalAcademyofSciences117(47):29495–503.Pierer,C.,andF.Creutzig.2019.“Star-ShapedCitiesAlleviateTrade-OffbetweenClimateChangeMitigationandAdaptation.”EnvironmentalResearchLetters14(8):085011.Pomponi,F.,R.Saint,J.H.Arehart,N.Gharavi,andB.D’Amico.2021.“DecouplingDensityfromTallnessinAnalysingtheLifeCycleGreenhouseGasEmissionsofCities.”UrbanSustainability1:33.Prisco,J.2019.“AShortHistoryoftheElevator.”CNNStyle,February2,2019.https://www.cnn.com/style/article/short-history-of-the-elevator/index.html.THRIVING266Rao,S.2016.“London’sNewLateNightAlternative:TheNightTube+Uber.”UberUndertheHood,October7,2016.https://medium.com/uber-under-the-hood/londons-new-late-night-alternative-the-night-tube-uber-8f38e56de983.Riahi,K.,D.P.vanVuuren,E.Kriegler,J.Edmonds,B.C.O’Neill,S.Fujimori,N.Bauer,etal.2017.“TheSharedSocioeconomicPathwaysandTheirEnergy,LandUse,andGreenhouseGasEmissionsImplications:AnOverview.”GlobalEnvironmentalChange42:153–68.Roberts,M.,F.GilSander,andS.Tiwari,eds.2019.TimetoACT:RealizingIndonesia’sUrbanPotential.Washington,DC:WorldBank.Rosenthal,S.S.,andW.C.Strange.2004.“EvidenceontheNatureandSourcesofAgglomerationEconomies.”InHandbookofUrbanandRegionalEconomics,vol.4,editedbyJ.V.HendersonandJ.Thisse,2119–71.NewYork:North-Holland.Samant,S.,andN.Hsi-En.2017.“ATaleofTwoSingaporeSkyGardens.”CouncilonTallBuildingsandUrbanHabitat(CTBUH)Journal3:26–31.Singh,C.,G.Jain,V.Sukhwani,andR.Shaw.2021.“LossesandDamagesAssociatedwithSlow-OnsetEvents:UrbanDroughtandWaterInsecurityinAsia.”CurrentOpinioninEnvironmentalSustainability50:72–86.Southerland,V.A.,M.Brauer,A.Mohegh,M.S.Hammer,A.vanDonkelaar,R.V.Martin,J.S.Apte,etal.2022.“GlobalUrbanTemporalTrendsinFineParticulateMatter(PM2.5)andAttributableHealthBurdens:EstimatesfromGlobalDatasets.”LancetPlanetaryHealth6(2):E139–146.Takagi,H.,M.Esteban,T.MikamiandD.Fujii.2016.“ProjectionofCoastalFloodsin2050Jakarta.”UrbanClimate17:135–45.Taylor,C.A.,andH.Druckenmiller.2022.“Wetlands,Flooding,andtheCleanWaterAct.”AmericanEconomicReview112(4):1334–63.Thebo,A.L.,P.Drechsel,E.F.Lambin,andK.L.Nelson.2017.“AGlobal,Spatially-ExplicitAssessmentofIrrigatedCroplandsInfluencedbyUrbanWastewaterFlows.”EnvironmentalResearchLetters12(7):074008.Varadhan,S.2019.“VillagersAccuseCityofSeizingWaterasDroughtParches‘India’sDetroit.’”Reuters,July2,2019.https://www.reuters.com/article/us-india-water-chennai/villagers-accuse-city-of-seizing-water-as-drought-parches-indias-detroit-idUSKCN1TX1BF.Wang,S.,X.Bai,X.Zhang,S.Reis,D.Chen,J.Xu,andB.Gu.2021.“UrbanizationCanBenefitAgriculturalProductionwithLarge-ScaleFarminginChina.”NatureFood2(3):183–91.Ward,K.,A.E.G.Jonas,B.Miller,andD.Wilson,eds.2018.TheRoutledgeHandbookonSpacesofUrbanPolitics.Abingdon,UK:Routledge.White,J.,andB.Serin.2021.“High-RiseResidentialDevelopment:AnInternationalEvidenceReview.”UKCollaborativeCentreforHousingEvidence,Glasgow,UK.Wong,S.K.,K.W.Chau,Y.Yau,andA.K.C.Cheung.2011.“PropertyPriceGradients:TheVerticalDimension.”JournalofHousingandtheBuiltEnvironment26(1):33–45.Wooldridge,J.M.2013.IntroductoryEconometrics:AModernApproach.Mason,OH:South-Western.WorldBank.2009.WorldDevelopmentReport2009:ReshapingEconomicGeography.Washington,DC:WorldBank.267267TheImpactofCitiesonClimateandtheEnvironmentWorldBank.2018.WaterScarceCities:ThrivinginaFiniteWorld.Washington,DC:WorldBank.WorldBank.2020.AddressingFoodLossandWaste:AGlobalProblemwithLocalSolutions.Washington,DC:WorldBank.WorldBank.2021.IndonesiaVision2045:TowardWaterSecurity.Washington,DC:WorldBank.WorldBank.2022.“PuttingPandemicsbehindUs:InvestinginOneHealthtoReduceRisksofEmergingInfectiousDiseases.”OneHealthreport,October,WorldBank,Washington,DC.Xu,X.,P.Sharma,S.Shu,T.Lin,P.Ciais,F.N.Tubiello,P.Smith,etal.2021.“GlobalGreenhouseGasEmissionsfromAnimal-BasedfoodsAreTwiceThoseofPlant-BasedFoods.”NatureFood2:724–32.Yoo,J.,andC.Perrings.2017.“ModelingtheShort-RunCostsofChangesinWaterAvailabilityinaDesertCity:AModifiedInput-OutputApproach.”InternationalReviewofAppliedEconomics31(4):549–64.Zaveri,E.,J.Russ,A.Khan,R.Damania,E.Borgomeo,andA.Jagerskog.2021.EbbandFlowVol.1:Water,MigrationandDevelopment.Washington,DC:WorldBank.Zhou,B.,D.Rybski,andJ.P.Kropp.2017.“TheRoleofCitySizeandUrbanFormintheSurfaceUrbanHeatIsland.”ScientificReports7:4791.HowDoWeGetItDone?PART3271PoliciesforPromotingGreen,Resilient,andInclusiveUrbanDevelopmentCHAPTER5Nomatterwhereyouareinlife,thereisalwaysalongjourneyahead.NelsonMandela••TheWorldBank’sGRID(green,resilient,andinclusivedevelopment)approachaddsasustainabilitylenstoitspursuitofthetwingoalsofendingextremepovertyandcreatingsharedprosperity.ThisreportalignswiththeGRIDapproach—initsdescrip-tionofhowgreen,howresilient,andhowinclusivecitiesareandintheanalyticsofhowclimatechangeaffectscitiesandviceversa.Thisprescriptivechaptermirrorsthatalignment.••Becauseofthecomplexityofthesubject,thischapterpresentsthegeneralconclusionsintheformofthreequestionsthatpolicymakersshouldanswer:Whatpolicyinstru-mentsareavailable?Whowieldstheseinstruments?Howcanpolicychoicesbasedontheuseoftheseinstrumentsbeprioritizedandsequencedforeffectiveimplementation?••WHAT:PolicyoptionstakeformoffiveI’s:information,incentives,insurance,integra-tion,andinvestments.Theirorderrepresentsasequencingofbundlesofinterventions.Inmanyinstances,theinterdependenciesbetweenthesesetsofinstrumentsplayoutincomplementaryways,whereinpoliciesacrossthebundlesstrengthenimpactwhenimplementedtogether.••WHO:Becausetraditionalurbanstressesinteractwithclimatechange–relatedstressestodetermineoutcomes,localgovernmentsarewellplacedtodriveclimateaction.Cities,workingwithotherstakeholders—includingnationalgovernments,theprivatesector,andcivilsociety—havealargepolicywedgeattheirdisposal.••HOW:PolicymakerswillneedtotogglebetweenandsandwichtogetherthebundlesofpolicyoptionsdescribedinthischaptertoarriveattheGRIDoutcomes.Thecombi-nationofinterventions,theirsequencing,andtheprioritizationofoutcomeswillvarydependingonthecharacteristicsofcities,includingprimarilytheirlevelofrisk,levelofdevelopment,andsize.MAINFINDINGS272THRIVINGIntroductionEveninancienttimes,changingclimateconditionshadcatastrophiceffectsonsomecities.ThepreindustrialcityofAngkor,theKhmerEmpire’scenterofpoliticalandeconomicpowerandhometomorethan1millioninhabitantsatitspeak,wasknownforitselaboratetemples,modernwatersystem,advancedtaxadministration,andambitiousbuildingprojects.Between1300and1400CE,thecityexperiencedextremeweatheranomalies,withseverefloodsthatcausedthereservoirsandcanalstocollapseandprolongeddroughtsthatstrainedfoodproduction.Theseeventsmayhavecontributedtothecollapseofthecity’scivilizationthatfolloweditssackingandlootingbyAyutthayainvaders(Buckleyetal.2010).Otherancientcitieswerealsovictimsofcatastrophicclimatechangeeffects.ThedeclineoftheIndusValley—knownforitsurbansettlements,agriculturalproduction,andtradeprowess—wasprecipitatedbyextensiveperiodsofdrought,whichintersectedwithincreasinglevelsofinterpersonalviolenceanddisease(Schugetal.2013).Fastforwardtotoday.Usingthe2021assessmentsoftheIntergovernmentalPanelonClimateChange,ClimateCentralprovidesmapsofcitiesexpectedtofindthemselvesunderwaterbecauseofrisingsealevels.1ThecitiesincludeBangkok,Thailand;Georgetown,Guyana;Kolkata,India;andNewOrleans,UnitedStates.Thepanel’sSixthAssessmentReportprojectsthatmorethan1billionpeoplelocatedinlow-lyingcitiesandsettlementswillbeatriskfromcoastal-specificclimatehazards(IPCC2022).Citieswill,ofcourse,facemorethanjustsea-levelrise;theywillalsohavetocontendwithamultitudeofsudden-onsetshocksandslower-onsetstressors.Withthesechallengesinmind,theWorldBank’sGRID(green,resilient,andinclusivedevelopment)approachaddsasustainabilitylenstoitspursuitofthetwingoalsofeliminatingextremepovertyandcreatingsharedprosperity(WorldBank2021a).2TheGRIDapproachsetsoutaframeworkfortheurgentinvestmentsneededinhuman,physical,natural,andsocialcapital—thatis,inkeystonesystems:energy,foodandlanduse,transportation,andurbansystems.Suchinvestmentsneedtobebalancedwithpoliciesforlaborandcapitalmarketstosafeguardthevulnerable,compensatethosewhomaylosefromthesepolicies,anddeliverajusttransition.UrbanizationoffersacriticalopportunitytoadvanceGRIDinresponsetothepressuresofclimatechange–relatedstresses.ThisprescriptivechaptermirrorsthealignmentoftheWorldBank’sgoalswiththeGRIDapproach.Buildingontheanalysisofpreviouschapters,thischapterlaysoutasystematicapproachtopolicyformulationthatfocusesonwhatcanbedone,bywhom,andhow.Indoingso,thechapterdemonstrateshowstridescanbemadebeyondtimid,temporizingpolicies.Giventhecomplexityofthesubject,thechapterlaysoutitsgeneralconclusionsinthreequestionsthatpolicymakersshouldconsider:Whatpolicyinstrumentsareavailable?Whowieldstheseinstruments?Howcanpolicychoicesbeprioritized,sequenced,andfinancedforeffectiveimplementation?Theexperiencesofvariouscountriesandcitiesqualifysomeofthegeneralconclusions.Therefore,thechapterinterpretstheuniquenessofexperiencesacrossdifferentcontextsandprovidesaframeworkforthinkingthroughhowpolicymaytacklethechallengescurrentlyfacingdifferenttypesofcities,buildingontheglobaltypologyofcitieslaidoutinchapter2.Whatarethechoices?Totacklethemyriadchallenges,andinsomecasesopportunities,associatedwithclimatechange,thepolicymakersofcitieswillhavetomakeinformed,astutechoicesacrossanPoliciesforPromotingGreen,Resilient,andInclusiveUrbanDevelopment273arrayofpossiblepolicies.Thissectiondescribesthelargevarietyofpolicyinstrumentsatthedisposalofpolicymakers,coupledwiththetoolstheyneedinordertothinkthroughthechoicesandtheirsequencing.Theseinstrumentsfallintofivebroadgroups,whichtogethermakeupasequencedset:1.Information2.Incentives3.Insurance4.Integration(withinandbetweencities)5.InvestmentsInformationInformationhelpspeopleandfirmsbetterunderstand,andthereforebetteradaptto,climaterisksbothacrossandwithincitiesInformationcantakemanyformsinvolvingvariouspolicyinstruments.Forexample,riskmanagementstrategiesorparticipatoryinitiativescouldallowordinarycitizenstorelaylocalinformationtolocal,regional,andnationalauthoritiesandpolicymakers.Orinformationcouldtaketheformofresponsivemonitoringandearlywarningmechanisms.Otherexamplesincluderegularlyupdatedurbanplanningdocuments,buildingcodes,andzoningregulations.Informationguidesdecisionsbybothhouseholdsandbusinessesonprevention,mitigation,andadaptation;yetaspectsofthecollectionandsharingofinformationcontinuetosufferfromshortcomings.Theclimatevectorembodiesambiguousrisks.Theprobabilityofextremeeventshasincreased.Fortunately,climatescientistsaremakingsignificantprogresstowardrigorouslymodelingclimate-relatedeffects,anditisbecomingcheapertospreadlocallyspecificinformationaboutimpendingrisks.Evenso,policymakers,especiallyatthesubnationallevel,mayfinditdifficulttodeterminepreciselythepotentialimpactsortodefineinvestmentprioritiesonthebasisoftheseprojections.Meanwhile,needisgrowingforunderstandable,reliableinformationpairedwithno-regretspolicymeasuresthatinvolvekeystakeholders,includingbothlocalcommunitiesandfirms.Despiteremarkablescientificprogress,someofthemostvulnerablecommunitiesinclimate-threatenedcitiesremainpoorlyinformedaboutbothloomingdisastersandslow-movingchangesthataffecteverydaylife.Someofthereasonsforthisshortcoming—suchasinsufficientstatecapacityorlackoftrustingovernmentinstitutionsinconflict-affectedareas—havebeenaroundforalongtime.Theproblemisoftennotjustthedearthofinformationbutalsotheasymmetryofinformationacrossnationalandlocalactors,oracrosssectoragencies.3Nevertheless,theopportunitiesforgettingaccurateandtime-sensitiveinformationtothosewhoneeditmostarefarfromexhausted.Investmentsinresearchandanalyticalcapacitieswithinrelevantinstitutions,forexample,remainintheirinfancy.Informationcriticaltoclimateadaptation,suchastheresultsofclimatemodelingandweatherforecasts,isapublicgood,providingastrongrationaleforgovernmentprovisionorsubsi-dization.Thediffusionofaccurateriskinformationfacilitatesbetterdecision-makingbyhouseholds,firms,andlocalgovernments.Italsoprovidesthefactsthatcitizensneedinordertoholdtheirlocallyelectedofficialsaccountable.Intheageofmachinelearningandartifi-cialintelligence,costlysurveystoidentifychangingrisksacrosslocationscouldgivewaytoalgorithmsthatprovidecheap,up-to-datedatawithhighdegreesofaccuracy.Suchtechniques274THRIVINGhavebeenused,forexample,toassesswildfirerisksinBulgaria,4seasonalpredictionsofdroughtacrossseveralregions,5andevenriskfinancingforhostcommunitiesinUganda.6Earlywarningsystems7cansavelivesand,ifactivatedsufficientlyinadvance,property.Accesstoearlywarningsystemsis,however,loweramongpoorerpopulations,whoalsousuallyhavetheleastaccesstoothertoolstomitigateortransferrisk(Hallegatteetal.2017).Often,thebenefitsofearlywarningsystemscanbemaximizedwhencoupledwithotherpreventativeinvestments.Forexample,HongKongSAR,China,hasinvestedinhousingimprovementsthatallowshelteringathomeduringtropicalcyclones,andearlywarningallowspeopletoreturnsafelytotheirhomesusingstorm-adaptivepublictransportation(RogersandTsirkunov2010).Evenmodestinvestmentsinsuchsystemscanhavehighreturns.AreportfromtheGlobalCenteronAdaptationandWorldResourcesInstitute(2019)findsthatearlywarningsystemsprovidea10-foldreturnoninvestment—thehighestofanyadaptationmeasureconsidered.Whenitcomestoemergingclimaterisks,perceptionversusrealityisimportantItisnotonlyaccesstoinformationthatmatters,butalsohowthatinformationinfluencesbehaviorandpolicymaking.Evenifwellinformed,peopleandbusinessesmayengageinmyopicbehaviorsthatcancreatemarketfailures.Forexample,althoughmoststudiesfindthatinformationonfloodrisksreduceslandandpropertyprices,8othersfindeithermixedorweakevidenceofpropertyfloodriskdiscounting(Beltrán,Maddison,andElliott2018;HinoandBurke2021).Likewise,severalstudiesfindevidencethatlandandhousingmarketsadjustintheaftermathofnaturaldisastersbut“forget”abouttheriskseveralyearsdowntheroad.BinandLandry(2013),forexample,findstrongevidenceofpropertiesinNorthCarolinabeingsoldatadiscountintheaftermathofHurricanesFran(1996)andFloyd(1999),butdocumentaswellthat,consistentwithmyopicbehavior,thediscountslowlyvanishesovertime.Anotherimportantdimensionistheambiguouseffectoflandpricesthatreflectinforma-tionondisasterrisks.Ontheonehand,thisinformationsignalshighercoststoretrofitstruc-turesintheaftermathofdisastersortobuildtohigherandmorecostlystandards.Thiseffecttendstodepressthedemandfortheriskylocationsandreduceconstructiondensity,which,inturn,shouldreducedamagesfromdisasters.Ontheotherhand,lowerlandpricesincreaselocationalaffordabilityforthepoorest,allowingthemtosettleinriskybutotherwiseadvanta-geousareas.Rentschleretal.(2022)documentthatthegrowthofsettlementsinflood-proneareasisoutpacinggrowthinlow-riskareas,withmuchoftheincreaseinriskconsumptionhappeninginlow-andmiddle-incomecountries.Thestudyhypothesizesthatasurbanareasgrowso,too,doeslandscarcity,pushingpeopletosettleinareaspreviouslyavoided.Damagesincurredbythedestructionofdwellingsoccupiedbythepoorinsuchareasmaybelowinmonetaryterms—becausethepoorhavefewerassetstolose—butwillstilldramaticallyreducetheirmedium-tolong-termeconomicprospects.Somewillbeforcedtoresorttoshort-termcopingmechanismsthatwilljeopardizetheirfuturewell-being.Ermanetal.(2020)findthatasignificantshareofhouseholdsaffectedbythe2015floodinAccra,Ghana,hadnotrecoveredtwoyearsaftertheevent.Informationalandotherprevailingmarketfailurescanalsoslowdownthetransitiontowardacarbon-neutraleconomy.Suchfailurescontributetothemispricingofclimaterisksinfinancialmarkets.Forexample,environmentalratingsforgreenbondsandotherfinancialinstrumentsareofteninconsistent,incomparable,and,attimes,unreliable,reflect-ingtheabsenceofgranulardataintheconstructionofsuchindexes.Theoutcomemaybe“greenwashing,”whereingreenbondsdonotnecessarilytranslateintocomparativelyloworPoliciesforPromotingGreen,Resilient,andInclusiveUrbanDevelopment275fallingcarbonemissionsatthefirmlevel(Ehlers,Mojon,andPacker2020).Demandsfromthemarket,rangingfromsupervisoryguidancetostrongdisclosurestandardsforstakeholders,canhelpidentifygapsandvulnerabilitiesandimprovethequalityofclimate-relatedfinancedisclosures.Thisproblemhasrelevanceforcitiesbecauseanincreasingnumberofthemusegreenbondstofinancelow-carbonprojects.9Itisalsousefultoconsidercities’landandrealestate,labor,andcapitalmarkets.Becausethepriceoflandistheprimaryelementdrivingefficientlanduse,bothapprais-ersandthepublicneedreliableinformationontherisksaffectinglocationsandthevalueofland.Astandardeconomicprescriptionistoprovidetransparentinformationthathelpsensurethatmarketpricesaccuratelyreflectrisk.Forexample,intheUnitedStatesseveralrealestateplatformsnowprovideneighborhood-levelclimatechange–relatedriskinformationforprospectivehomebuyers.Betterinformationonlandmarketscouldalsocontributetoclimatemitigationeffortsbyloweringthecostsofverticalconstruction,therebyhelpingpromotemorecompacturbanforms(seechapter4).Forlandandpropertypricestoaccuratelyconveyinformationinturnrequiresstrongpropertyrightandlandadministrationsystems.Migrationrespondstopullfactors,suchasaccesstobetterlabormarketoutcomesindestinationlocations,andpushfactors,suchasadverseshockstolivelihoodsinoriginlocations.Throughitsinfluenceonthesefactors,climatechangeactsasanimportantdriverofmigration(seechapter3).Informationalnudgescanactasapowerfulimpetustomigration(Wilson2021).Betterinformationonclimateriskscouldencouragepeopletoincreaseinvest-ments,eitherforinsituadaptationtorisksortofundmigrationtoareaswithlowerrisks.Incountriesthathavelargevariationinclimaterisksacrosscities,thisinvestmentcouldhelptheoverallresilienceoftheurbansystem(seechapter2).AsharedunderstandingisimportantforcrediblepoliciesThemorecrediblepoliciesare,themoreeffectivetheyare.Thus,planningregulationsthatlimitnewdevelopmentsinhigh-riskareaswillworkiftheinformationonrisksiscredibleandifthereisacrediblecommitmenttoenforcement.Onlywithasharedtheoryofhowprograminputsrelatetooutputscantheinputsbecredible,especiallywhencommitmentsaremadeoveralongperiodtowhatmaybepainfulandunpopularpolicies.Forexample,EuropeancountriesarenowseeingsteeprisesinenergypricesbecauseofthesupplyshortagesarisingfromthewarinUkraine.Timewilltell,then,whetherclimatechangecommitmentstodiversificationawayfromfossilfuelenergysourceswillholdsteadyinthecomingmonthsandyears.Policiesarealsomorelikelytobecredibleandacceptedifpolicy-makingprocessesaretransparentandinvolvecitizens.Forexample,so-calledproceduraljusticecanbeusedtointegratefairnessandequityintodecarbonizationapproaches.Proceduraljustice,whichentailsformallyinvolvingcitizensintheprocessofdesigningpolicies,is,atitscore,aboutinclusiveandtransparentprocessestobuildtrustandbetterpolicies(WorldBank,forthcominga).Finally,evenwheninformationisavailable,thepoliticalwilltoembraceitmaybelacking.Forexample,accordingtothereportNaturalHazards,UnNaturalDisasters:TheEconomicsofEffectivePrevention,evenwhentheUSFederalEmergencyManagementAgencyregularlyupdatedcoastalfloodmaps,householdsandfirmsinmorethan100USGulfcommunitiesdidnotacceptthemapsastrustedinformation,therebypreventingthelocaladjustmentofpropertyprices(WorldBank2010).276THRIVINGIncentivesIncentiveswillencouragepeopleandfirmstoproperlyfactorinclimatechangerisks,andgovernmentofficialstoaddressthechallengesofpursuinggreen,resilient,andinclusivedevelopmentAlthoughinformationhelpshouseholdsandbusinessestofactorclimatechange–relatedrisksintotheirdecisions,information,inandofitself,maynotbesufficienttomotivatethemtotakeintoaccounttheimpactsoftheirowndecisions,goodorbad,ontheenvironmentandothers.Motivatinghouseholdsandbusinessestoconsidertheenvironmentaleffectsoftheiractionsrequiresincentives.Incentivescomeinvariousguises,suchastaxes,charges,subsidies,andtradablepermits.Whatevertheincentive,attentionmustbepaidtoequitybecause,althoughincentivepoliciesraiseoverallwelfare,foranygivenpolicysomepeopleorfirmsmaygainmorethanothers.Asaresult,someeconomicorsocialgroupscouldloseinabsoluteterms.Failuretotakeintoaccountthedistributionalimpactsofnewincentives,andmoregenerallyofanypolicyintendedtohelpaddressclimatechangeandotherenvironmentalissues,mayresultininsurmountablepoliticaloppositiontothepolicy.Onerecenthigh-profileexamplewastheFrenchgovernment’sfreezingofitscarbonpricingpolicyfollowingtheYellowVestsprotestmovementof2018(RubinandSengupta2018).Becausetheatmosphereisaglobalcommon,nocountry(orcity)hasanindividualinteresttodomuchaboutgreenhousegas(GHG)emissions.Howthendoesoneincentivizeloweremissions?BecauseaunitofGHGemissionshasthesameeffectregardlessofwhereitoccurs,market-orientedapproachesarewellsuitedtocontrollingemissions.Theadditionalflexibilityprovidedthroughmarketapproachestotheemitterindetermininghowtoreduceemissionscouldalsoincentivizeefficiencyandinnova-tioninmeetingtheregulatoryrequirement.Specificmarket-orientedapproachesforGHGreductionsincludetradablepermits(suchascap-and-tradeprograms)thatsetaspecificcapontotalemissionsandallocateorauctionthenumberofpollutionpermitsorallowancesneededtomeetthatgoal.Anotherapproach,emissiontaxes,providesincentivesforpolluterstofindcost-effectivesolutionstoemissionscontrol.Subnationalregionshaveoftenbeenattheforefrontofsuchprograms.Forexample,in2010Tokyosetupthefirsturbancap-and-tradeprogramthatfocusedonurbanbuildings.In2013,Chinalauncheditsownpilotinseveralcitiesaimedatreducingtheemissionsintensityofcoal-andgas-firedenergyplants(Nogrady2021).CitingJapanesecities,Domonetal.(2022)findthatcongestiontollscouldbehighlyeffective(farmoresothanemissionstaxes)inreducingurbancarbondioxide(CO2)emissionsfromcommutingandhousingenergy—withreductionsof22percentand3percent,respectively.Moregenerally,however,regulationsandstandardsprovidemorecertaintyaboutemissionlevelsandcouldbepreferablewheninformationorotherbarrierspreventproducersandconsumersfromrespondingtopricesignals.Indeed,someregulations,suchasthoseforvehicleemissions,usuallyatthenationallevel,haveprovenimpressiveinachievingreductions.TheUnitedStatesandcountriesintheEuropeanUnionhavebeenregulatingemissionssincethe1970s,andwide-spreadevidencesuggeststhattheyhavereducedvehicleemissionsandimprovedairqualitysig-nificantlyinmostlargecities—seeWinkleretal.(2018)foranoverview.Quotacontrolpoliciescanalsobeeffectivebutmustbecarefullydesigned.TheanalysisbySong,Feng,andDiao(2020)ofSingapore’svehiclequotacontrolfindsthatvehiclequotashavesubstantiallylimitedvehicleownershipandusage.Thedesiretomaximizereturnoninvestment,however,canpartiallyoffsetmitigationeffectsbecauseofhigherusagebyexistingcarowners.PoliciesforPromotingGreen,Resilient,andInclusiveUrbanDevelopment277Bansareoftenalow-hangingfruit—severalAsiancountries,includingBangladesh,thePhilippines,andSriLanka,havenationalregulationsorbansonsingle-useplastic;and,wherenonationalpoliciesexist,localgovernmentshavetakenituponthemselvestoregulateplasticwaste.Throughphasedimplementationandstrictenforcement,thecityofSanFernandointhePhilippinesmaintains98percentcompliancewithsingle-useplasticbans.In2021,FrancepassedaClimateandResilienceBilltotackleclimatechange.Thebillencompassesbans,incentives,andquotasontransportation,housing,andconsumptionintendedtolowercarbonemissions,improveenergyefficiency,andcutwaste.Thebansrangefrompreventinglandlordsfromrentingpoorlyinsulatedbuildings,toa2025targettocurbsingle-usefoodpackaging,toablanketbanonshort-haulflightsthatcouldbereplacedbyafour-hourtraintrip.Althoughclearerrulesanddefinitionsareuseful,theyarenotlikelytobesufficient.Tocorrectformissingmarketsandtosetinmotionaprocesswherebyfirmsandhouseholdsreducetheirenergyuseandmakeapermanentshifttocleanerenergysourceswillrequirecarbonpricing,cap-and-tradeprograms,andothermarket-basedpolicies.10InVietnam,theWorldBankissupportingtheestablishmentofadomesticcarbonfinancemarketforcitiesbylinkingthemtointernationalmarkets.Thisarrangementallowslargecitiestoaggregatemanysmallactions—suchasimprovingtheenergyefficiencyofbuildings,upgradingtolight-emittingdiodestreetlights,installingrooftopsolarpanels,andupgradingtoe-motorbikes—thatamounttoenoughemissionreductionstotransactonthemarket.Suchanapproachwouldenableacitytoupgradeitsownassetstoincreaseefficiencyandrenewableenergygenerationcapacity,whilecatalyzingactionbytheownersofprivateassetsthroughtheprovisionoftechnicalsupportandincentivepayments.Landuseregulation,atypicalurbanpolicy,canaffectenergydemandandcongestionexternalitiesbyinducingchangesinurbanformandthespatialdistributionofresidents,therebychangingthetotalcommutingdistanceandcongestionlevels.Inaddition,whenfloorspaceperhouseholdchanges,householdenergyconsumptionoflighting,cooling,andheatingalsochanges.Aschapter3discusses,landuseandotherpolicies(includingbuildingheightandfloorarearatioregulations)thataffectthecostsofverticalconstruction,thewalkabilityofcities,and,moregenerally,thechoiceoftransportationmodecanalterbothcitydensityandthedegreeofsprawl.Suchchanges,inturn,haveimplicationsforemissionsofbothCO2andlocalairpollutantssuchasparticulatematter2.5micronsorlessindiameter(PM2.5).AccordingtothelatestIntergovernmentalPanelonClimateChangereport,in2019buildingsaccountedfor31percentofglobalenergydemand,18percentofelectricitydemand,and30–40percentofglobalCO2emissions(IPCC2022).Operationsofbuildings(thatis,notincludingthecarbonembodiedinmaterialsandconstruction)accountedforalargeproportionofannualglobalemissions.ExistingbuildingsareespeciallyimportantforeffectiveGHGmitigationsimplybecausetheyarelong-livedassets.Incentivestoretrofithousingtoincreaseenergyefficiencycouldthereforehavelong-termbeneficialimpactsonemissions.Weiss,Dunkelberg,andVogelpohl(2012)studiedtheimpactsofvariousincentivestomotivatehomeownersinGermancitiestocarryoutenergy-efficientrefurbishments.Theyfoundthat,althoughfinancialincentives(suchasgrantsandtaxrebates)ledtobetterresultsthanregulatorystandardsalone,randomauditsalsohelpedensurecompliance.Bettercommunica-tionofthebenefitsofmakingtheenergy-savingchangesalsopromotedhighertake-up.Finally,fiscalincentivestohouseholdstoretrofitandupgradetheirpropertiescanalsobeeffectiveinreducingtheriskstheyfacefromclimatechange–relatedandothernaturalhazards.Forexample,thegovernmentofTürkiyeprovidesfamilieswithfinancialsupporttoupgradeorretrofitresidentialunitsatriskofdamageorcollapseintheeventofanearthquakeorotherhazard.Thissupportcomesintheformofeitheraninterestratebuy-downforaloanfromacommercialbankorrentalsupportforaperiodofupto18months.Theinterestratebuy-down278THRIVINGismorefavorableforthosehouseholdsthatincludehigherstandardsofenergyefficiencyinthereconstructionprocess.Suchapproachescanbothbringdownbuildings-relatedemissionsandincreaseresiliencetonaturalhazards.LocalgovernmentsrespondtoincentivesInternationalexperiencehasshownthatlocalgovernmentsrespondtoincentivesandthatnationalauthoritiescanshapetheseincentivestoachieveresults.Incentivesforlocalgovern-mentstoactonclimatemighttaketheformofofferingthemgrantsconditionalondevelopingaclimatemitigationoradaptationplan,orofgrantingcitieswithsufficientcapacity(suchassufficientlystaffedplanningandutilitymanagementteams)greaterauthorityinareasrelevanttolocalclimateaction.Somegovernmentshavebeguntoexperimentwithsuchincentives.Forexample,BritishColumbia,Canada,introducedaClimateActionRevenueIncentiveProgram,aconditionalgrantschemethatrequireslocalgovernmentstosignontotheprovince’sClimateActionChartertoaccessadditionalfunding.11Theseeffortsarestillyoung,butexperimenta-tion,adoption,andknowledgeexchangeswilleventuallyprovideevidenceontheeffectivenessofsuchmechanismsforlocalgovernment–ledclimateaction.SomeincentivesaimatadaptationCitiesmustassess,andadjustasnecessary,theirbuildingsiting,design,andconstructionpracticestoavoidorwithstandtheincreasedforceandfrequencyofhydro-meteorologicalhazardssuchasextremewind,flood,stormsurge,andsea-levelrise.Theyshouldbaseassessmentsandadjustmentsonhazarddataandthecalculationofexpectedhazardloadsonstructures.Althoughitisnotpossibletobuildacompletelystorm-orearthquake-proofhouse,constructionofresilienthousingisbothtechnicallyandeconomicallyfeasible(WorldBank2019b).Financialincentivescanhelphomeownersanddevelopersmovetowardintegratingadaptationmeasuresintohousingconstruction.Forexample,inCanadathecityofTorontolaunchedin2009theEco-RoofIncentiveProgramthatprovidesgrantstopropertyownersfortheconstructionofbothgreenroofsthatsupportvegetationandcoolroofsthatreflectthesun’sthermalenergyandmanagewaterrunoff.Agreenroofreducestheamountofheattransferredtothebuildingbelow,keepingthebuildingcoolerandatamoreconstanttemperature.12Insurancepremiumsforriskcoveragecanalsoincentivizehomeownerstoreducetheirpremiumbyreducingtheirrisk.Forexample,homeownerscouldinstallfirealarms,raiseelectricalequipmentorheatingandcoolingsystemsabovepotentialfloodlevels,orinstallanewroof.Closingthelooponwaterusewithinahouseholdcangoalongwaytowardsolvingwaterchallengesandmakingcitiesmorewater-resilient.Oneinitiativeleadingthewayinthiseffortisthe50LiterHomeCoalitionspearheadedbyProcter&GambleandsupportedbytheWorldEconomicForum,the2030WaterResourcesGroup,andtheWorldBusinessCouncilforSustainableDevelopment.Thecoalition,whichemergedfromCapeTown’s“dayzero”eventwhenhouseholdwaterusewasrestrictedto50litersofwaterperday,aimstodemonstratethatlivingon50litersofwaterperdayisnotonlyfeasiblebutalsocanbedoneatlittlesac-rificetothehousehold.Thecoalitionisdevelopingmethodstotreatwateratitspointofusetomakeitreusable.Inthisway,somewastewatercanbetreatedtobecomedrinkableandotherwatercanbetreatedforreuse,suchasshowerwaterrepurposedfortoiletreuse(Zaverietal.2021).Seebox5.1foradiscussionofcitiesandgreeninnovation,andbox5.3laterinthechapterforexamplesofincentives,includingpricing,inwater-securecities.PoliciesforPromotingGreen,Resilient,andInclusiveUrbanDevelopment279Cancitiesbeincubatorsofgreeninnovation?NobelPrize–winningeconomistsMichaelKremerandPaulRomerhavebothsuggestedthatoverthelongtermthesizeofapopulationisrelatedtoitsspeedoftechnologicaladvance(Kremer1993;Romer1990).Small,isolatedcommunitiesseetechnologicalstagnation,whereaslarge,integratedpopulationsfosterrapidinnovation.AsKahn(2010)argues,climatechangebringswithitthepotentialforendogenoustechnologi-caladvancesthatcouldplayanimportantroleinspurringmitigationandadaptation.Ina2016study,Feietal.alsodiscusstheinterdependentrelationshipbetweencitiesandgreeninnovations.Thatinterdependencearisesbecausedenselypopulatedmetropolitanareasareoftenvulnerabletotheeffectsofclimatechangeandfacethebiggestdemandforgreenareasandecosystemservices(Oijstaeijen,Passel,andCools2020).InVietnam,thecityofHuerespondedtosuchademandbyincreasinggreenspacesandvegetationinoneofitshistoriccolonialdistrictsthatservesasaresidentialareaandtouristattraction.InMalaysia,thecityofMelakaisdevelopinginnovativepublicinfrastructuresthatusesolarpowerandotherrenewableenergyasapartoftheMelakaGreenCityActionPlan.aManygreeninnovationsinurbanareasoccurattheinstitutionallevelandareimple-mentedthroughcentralandlocalauthorities.Forexample,asamajordevelopingcountrythataimsatabsolutereductionsingreenhousegasemissions,Brazilhasadoptedaseriesofsubnational-andcity-levelgreenfiscalpolicies,suchasfiscalincentivesandsubsidiesforinnovationsandsustainablepractices.Evidencefrom24manufacturingsectorsconfirmsthatthesegreenfiscalpolicieshavehelpedpromotegreeninnovationsinBrazilandcontributedtothecountry’stransitiontoagreeneconomy(GramkowandAnger-Kraavi2018).IntheArabRepublicofEgypt,byimplementinggreenpolicies,threenewcitieshavesubstantiallyimprovedtheirenvironmentalperformanceandurbanmanagement(Hegazy,Seddik,andIbrahim2017).SuchcitiescanprovideguidelinesforotherEgyptiancitiesandhelpauthoritiesselecttheproperpoliciestoestablishgreenurbansystemsinthecountry.Sustainableandnonconventionalcommercialproductsrepresentanothercrucialaspectofgreeninnovation.Inrecentyears,thedevelopmentofsustainablegreenproductshasincreasedasmanyfirmsbegintorealizethetangibleadvantagesofandprofitsfromaddressingenvironmentalimpactsinproductdesigndecisions(Lenox,King,andEhrenfeld2000).Theseinnovationsoftentakeplaceincitiesandspreadwithurbaniza-tion.Forexample,inrecognitionofthepotentialtogreatlyreducethecarbonfootprintofthehigh-emissiontransportationsector,salesofelectricvehiclesincreasedinrecentyears.IntheUnitedStates,theadoptionofelectricvehicleshasbeencloselyassociatedwithurbanization.bAnotherexampleofaninnovativegreencommercialproductisplant-basedmeat,whoseproductionemits30–90percentlessgreenhousegasthantheproductionofconven-tionalmeat(GoodFoodInstitute2021a).Theplant-basedmeatindustryhasbeenexpandingglobally.Amongthetop10plant-basedmeatbrandsbysales,mostofthemareheadquarteredinlargecitiessuchasChicago,LosAngeles,andSeattle(GoodFoodInstitute2021b).Box5.1280THRIVINGInsurancePeople,firms,andgovernmentscaninsureagainstlossesassociatedwithclimatechangeandagainstunavoidableenvironmentalshocksandstressesHowdopeopleandfirmsbehavewhenconfrontedwithrisk?Theanswermattersgreatlyforpolicy.Aneconomicagent—here,anindividualorafirm—couldtakeseveralactions,noneofthemcost-free.Theycouldpurchaseinsurancetoguaranteereceivingfundswhenadisasteroccurs.Ifmarketinsuranceisnotavailable,suchasinmanylower-incomecountries,indi-vidualsorfirmscaneitherself-insureorself-protect.Toself-insure,theysetasidesavingstomitigatethepotentialcostsoffuturedisasters.Toself-protect,theytakeactionstoreduceriskexposureby,forexample,migratingawayfromriskyareas,takingproactivestepstoprotecttheirdwellingorbusiness,ordiversifyingtheirsourcesofrevenue.SeminalworkbyEhrlichandBecker(1972)providesaneleganttreatmentoftheoptimalinsurancedecisionwhenfacedwiththeoptionsofmarketinsurance,self-insurance,andself-protection.Urbanizationshiftsthebalanceofpreventionfromindividualmeasurestocollectiveaction.Governments,bothlocalandnational,willhavealargerroletoplayinremovingtheconstraintsonindividualsandcomplementingindividualadaptationstrategies.Nationalgovernmentscouldplayanimportantrolebyaugmentingbothmarketinsuranceandself-protection.Thegrowthofglobalinsurancegiantsoffersopportunitiesforinsurerstoexpandinthedevelopingworldbyofferingpoliciesanddiversifyingtheirriskexposureusinginstrumentssuchascatastrophebonds(box5.2).Inthissense,theriseofglobalfinancialmarketshelpsprotectmoreandmoreindividualsfromtheexpostlossesassociatedwithplace-basedclimateshocks.Meanwhile,ifglobalinsurersofferagoodpremiumoncoverage,thenpolicybuyerswillinvestlessinself-insurance.UsingtheexampleofindexeddisasterfundsinMexico,DelValle,deJanvry,andSadoulet(2020)describehowpublicandprivatecoordinationininsurancemarkets,whichallowedtheeasingofweakrulesandthesupplementationofadministrativecapacitybyeasingliquidityconstraints,ledtoconsiderableaccelerationofpostdisastereconomicrecovery.Disasterriskfinanceandinsuranceinstrumentsaimtominimizethefinancialimpactsofdisastersandtosecureaccesstopostdisasterfinancingbeforeaneventstrikes,therebyensuringtherapidavailabilityofcost-effectiveresourcestofinancerecoveryandreconstruc-tionefforts.Typically,governmentsseekfinancialprotectionforfourgroups:nationalandlocalgovernments,homeownersandsmallandmediumenterprises,farmers,andthepoorestThus,metropolisesactascradlesofbrandsthatproducesustainable,innovativeproducts.Greeninnovationsincreaseafirm’sprofitwhilereducingenergycostsforsociety(Feietal.2016).Therefore,suchinnovationsareexpectedtocontinueincreasinginmetropolitanareasandplayanessentialroleinthetransitiontowardmoresustain-able,energy-efficienturbanenvironments.a.FromAsianDevelopmentBank,“GreenCities,”https://www.adb.org/green-cities/.b.Electricvehicleregistrationper10,000peopleinUSmetrocountiesrangesfrom10tomorethan100,whereasthatinnonmetrocountiesislessthan5orevenzeroinmanycases(Tolbert2021).Box5.1continuedPoliciesforPromotingGreen,Resilient,andInclusiveUrbanDevelopment281segmentsofthepopulation.RisktransfersolutionstargetinghouseholdsweredeployedmostrecentlyinTürkiyewiththeuseofcatastropheinsurancepools.Türkiyeishighlyexposedtosignificantclimateanddisasterrisks,especiallyinurbanareas.In1999aftertheMarmaraearthquake,thegovernmentestablishedtheTurkishCatastropheInsurancePool(TCIP)tosetuplong-termreservestofinancefutureearthquakelossesandalleviatethefinancialburdenofearthquakesonthegovernmentbudget.Soldseparatelyfromfireandotherindemnityinsurance,TCIPprovideshouseholdswithcompulsoryearthquakeinsurancewithaffordableinsurancepremiums.TCIPisanonprofitpublicentitywithprivatefinancing.Itsfundinghasprimarilydependedoninsurancepremiumspaidbyhomeowners.Withitscoverageexpandingovertime,in2021TCIPreachedabout10.5millionhouseholdsnationwide(AfetandKurumu2021;Gurenkoetal.2006).Localandnationalgovernmentscanalsotakeadvantageofrisktransfersolutions.Forexample,inJuly2017thegovernmentofthePhilippinespurchasedforUS$205.9millionitsfirstparametricinsurancepolicyunderthePhilippineParametricCatastropheRiskInsuranceProgram.Indoingso,itsuccessfullytransferredsomeofitsdisasterrisktotheinternationalreinsurancemarkets.In2018,itpurchasedasecondinsurancepolicy(renewal)undertheprogram,approximatelydoublingtheamountofcoverageofferedbythefirstpolicy.Thepolicycovered25localgovernmentsagainstemergencylossesfrommajortyphoonsandcoverednationalgovernmentagenciesagainstemergencylossesfrommajortyphoonsandearthquakes(WorldBank2020b).Subnationalcatbonds—Dotheywork?Acatastrophebond(catbond)isarisktransferinstrumentthatfunctionslikeinsurance.Itisstructuredasafixedincomesecuritythatholdstheprincipleinescrow,paysperiodiccouponstotheinvestorduringthelifeofthebond,andeffectivelyinsuresthesponsorofthebondagainstapredefinedsetoflossesfromhazardssuchasearthquakesorhurricanes.Ifacoveredeventoccursduringthebond’slife,thesponsoringcountryretainsthebondprincipalwithoutanydebtobligationtofundemergencyreliefandreconstructionwork.Catbondsaremosteffectiveaspartofalargerdisasterriskman-agementstrategythatinvestsinriskreductionandadaptationefforts,inadditiontootherrisktransferandfinancinginstruments.Manycountries,includingChile,Colombia,Jamaica,Mexico,Peru,andthePhilippines,haveusedcatbondstotransferrisksrelatedtodisastersfromdevelopingcountriestothecapitalmarkets.Catbondscanalsobeissuedatthesubnationalleveltotransferriskfromlocalgovernmentsandutilitycompaniesintheeventofanaturaldisaster.Forexample,in2020and2021theLosAngelesDepartmentofWaterandPowersecuredtwocatbondsintheamountsofUS$50millionandUS$30millionforwildfireinsurancecoverage.Similarly,theNewYorkMetropolitanTransitAuthority(MTA)hasreliedoncatbondstotransferriskfromstormsurgesand,morerecently,earthquakes.TheMTAissueditsfirstcatbondin2013forUS$200milliontotransfertheriskofstormsurgesafterHurricaneSandyhitNewYorkCity,leadingtoUS$4billionindamagetoMTAassetsandinfra-structure.In2020,theMTAcollateralizedreinsuranceprotectionagainststormsurgesresultingfromstormsandearthquakeriskswithintheNewYorkmetropolitanareaonaparametrictriggerandperoccurrencebasisforathree-yearterm.Inshort,catbondshavethepotentialtohelpcitiesmovesomeoftheirriskintocapitalmarkets.Deployingsuchbondssystematicallyindevelopingcountrycontexts,however,willrequiremoreinfor-mationandsophisticateddatatohelpbringthemtomarket.Box5.2282THRIVINGSecond-bestoptionsaresometimesthebestoptionsEconomistsoftentoutmarketinsurance,whichreflectsdisasterrisks,asthefirst-bestoptiontointernalizerisksandminimizedisasterimpacts.Implementingrisk-basedinsurance,however,requiresovercomingmajortechnical,social,andpoliticalchallenges.Itis,then,notalwaysrealistic,especiallyinthedevelopingworld.Butthereareotheroptions.Usingatheoreticalurbaneconomicsmodelandnumericalsimulations,AvnerandHallegatte(2019)investigatethecostsandbenefitsoftwo“second-best”exantefloodmanagementstrategies,subsidizedinsuranceandzoning,andcomparethemwithrisk-basedinsurance.Subsidizedinsurance,whichallowshouseholdcompensationintheaftermathofafloodevent,hasthebenefitofreducinghousingscarcitybecausebuildingdecisionsarethenunaffectedbythepossibilityoffloods.Suchaninstrumentreduceshousingrents,butitalsoentailsmoralhazardbecauseexcessiveconstructiontakesplacewithnoregardforrisk.Landusezoning,ifimplementedcorrectly,ensuresthatdamagesfromfloodsarezero,butitalsoreduceshousingfloorspacebecauselandbecomesscarcer.Risk-basedinsuranceincorporatesdisasterriskintoconstructiondecisionsandreducesthehousingstockmorethansubsidizedinsurancedoes.Floodzoningisclosetooptimalwhenflood-proneareasaresmall,floodsarefrequent,andhousingqualityislow.Subsidizedinsuranceisclosetooptimalwhenalargefractionofacityisflood-prone,floodsarerare,andhousingqualityishigh.Whentheimplementationofrisk-basedinsuranceisunrealistic,acombinationofzoninginhigh-riskareasandsubsidizedinsuranceforlow-riskareasmayofferagoodalternative.Althoughthecostofimplementingasecond-bestpolicygenerallyremainslowexceptinsomeextremecases,thecostofimplementingthewrongsecond-bestpolicy(thatis,zoninginsteadofsubsidizedinsuranceorviceversa)canbeveryhigh.Equityisanimportantconsiderationwhenintroducinginsuranceprograms.BlickleandSantos(2022)demonstratehowquasi-mandatoryfloodinsuranceintheUnitedStatesreducesmortgagelendingbecausebanksgravitatetowardborrowerswithhigherincomes,whichaffectshousingaffordabilityforsomeofthepooresthouseholds.Inaddition,becausethepoorestaremorelikelytoliveinareasmoreexposedtohazards(seechapter3andRossitti2022),theywouldfacethehighestriskpremium.Asaresult,thosewhoneedinsurancethemostoftenhavetheleastaccesstoitandtheprotectionitaffords(Hallegatteetal.2017).Theanswer,however,istoavoidpricinginsurancepremiumstoolow;wheninsurancepremiumsaremispriced,that,too,canleadtooverbuildinginriskerplaces.Thisfindingsuggeststhatpoliciesthatprovidemoreaffordablehousinginlessriskylocationswouldhelpnotonlyaddresstraditionalurbanstressesbutalsoreduceriskexposure.Theunintendedconsequencesofsubsidizedinsurancepoliciescould,perversely,magnifytheharmofclimatechange.Moralhazardoccurswheninsurancespurshouseholdsandfirmstotakeriskieractionsthanifnotinsured.Ifthenationalgovernmentsubsidizesdisasterinsur-ance,individualswillmorelikelyinvestlessintheirownself-protection.Forexample,theUSnationalfloodinsuranceprogramhasoftenpaidtorebuildafloodedhomemultipletimesinthesamelocation.13FindingsfromPeraltaandScott(2018)suggestthatinsuringpeopleagainstpotentialfloodlossescontributesdirectlytopopulationgrowthinflood-proneUScounties;theavailabilityoffloodinsuranceincreasesthepopulationby5percentforevery1-standard-deviationincreaseinfloodrisk.Nevertheless,rapidlygrowingcitiesthatfacelandandhousingscarcitymayseesomebenefitsfromoverconstructionintheformoflowerpricesforfloorspace.Althoughalwaysdominatedintermsofefficiencybymarket-basedinsurance,subsidizedinsurancecanundercertainconditionsserveasausefulsecond-bestbecauseitismucheasiertoimplementandentailslimitedwelfarecosts(AvnerandHallegatte2019).Finally,socialsafetynetscanalsobeanimportantsourceofsocialinsuranceforthepooresttohelpmanagetheriskofnaturalhazards.SafetynetscanbedeployedexantetopreventPoliciesforPromotingGreen,Resilient,andInclusiveUrbanDevelopment283andmitigatetheimpactofanaturaldisaster.Forexample,Bangladesh’spreplannedfloodandcycloneresponseincludesseveralsocialsafetynetprogramstoreducevulnerability(Hassanetal.2013).Ethiopia’ssafetynetprogramsarescalableinresponsetoshocks.Forexample,in2011,inresponsetodrought,thegovernmentmanagedtoexpandtheseprogramstosupport9.6millionpeople.Safetynetscanalsobedeployedafteradisastershocktocopewithitsimpacts.Asafetynetwouldtaketheformoflabororpublicworksprogramstoaidrecoveryandtosupportshiftstomoreproductiveandalternativelivelihoods,oritcouldtaketheformofcashtransfers.Forexample,Malawiincludedashock-responsivesocialprotec-tionmechanisminits2019NationalDisasterRiskFinanceStrategy.Intheeventofadrought,thismechanismenablestheexistingsocialcashtransferprogramtoreachadditionalhouse-holdsortopuppaymentstoexistingbeneficiaries(GRiF2020).Developingcountries,however,oftenhavelowcoverageofsocialprotection,andevenmoresoinurbanregionsbecausemostprogramstendtohavearuralfocus(DevereuxandCuesta2021).IntegrationWithincities,integratedplanningpromotesmorecompactenergy-efficientdevelopmentPlansareworthless—observedUSPresidentEisenhower,drawingonhismilitaryexperience—butplanningiseverything.Whenacitygrows,thepressureonitslandandhousingmarkets,itssuppliesofbasicservicesandinfrastructure,anditsenvironmentgrowsaswell.Ifnotwellmanaged,thispressurecanunderminethegreenness,resilience,andinclusivenessofdevel-opment.Flexibleandversatileurbanplanning,coordinatedwithinvestmentsininfrastruc-ture,canensurethatcitiesarenotlockedintosuboptimalphysicalformsandinvestmentsthatexacerbatethispressure.Examplesofsuchlock-insincludeenergy-orwater-intensivebuildingtechnologies,urbansettlementsinareasexposedtonaturalhazards,andsprawlingandcar-dependentpathsofdevelopment.Thefindingsinchapters1and4suggestthatmorecompactdevelopment,whichinvolvesthegrowthofcitiesverticallyratherthanhorizontally,isassociatedwithreductionsinbothCO2andPM2.5emissions,aswellaswithlessconversionoffertileagriculturalland.Asdiscussedunderthefifth“I”(investments),althoughretrofittinginfrastructureandbuildingswillbeessentialtogreenerandmoreresilientgrowth,doingsocanbecostlyandcouldbeavoidedwithearlyeffortsatintegratedplanning.Reforminglandadministrationandurbanplanninginrapidlygrowingcitiescangoalongwaytowardensuringgreatercompactness.Landadministrationservicesinlow-andmiddle-incomecountriesarefrequentlyexcessivelycentralizedand,asaresult,unresponsivetochangingconditions.Onepossiblesolutionistodecentralizelandcertificationservicesbyallowingexistinglocalstructurestoverifyownership.Forexample,in2011Burundireformeditslandadministrationsystemtoallowits97CommunalLandServicesofficestoissuelandcertificates.Previously,landregistrationwashandledbymuchlessaccessiblelandregistra-tionservicesundertheMinistryofJustice(Mukim2021).OtherexamplessuchasEthiopiaandIndonesiaillustratethebenefitsofdecentralizedlandregistrationprocessesmanagedbylocalgovernmentsinaparticipatorymanner.InIndonesia,thecommunity-supportedLandAdministrationProjectpavedthewaytoformalizingmillionsofpreviouslyunregisteredparcels(Deininger,Selod,andBurns2012).Thesecasesalsoshow,however,thatdecentraliza-tionincreasestheresponsibilitiesoflocalgovernments,requiringcommensurateincreasesinpolicyguidanceandresources.Adecentralizedapproachtolandadministrationalsoallowsadegreeofiterativepolicyexper-imentation.Selectedprimaryandsecondarycitiescanpilotimportantinterventionssuchas284THRIVINGcreatinginventoriesofallpublicland,buildings,andinfrastructure.Thefindingsfromsuchpilotscanthenbeusedtoinformbothnationalurbanmasterplanningandlocaldevelop-mentplansthatpromoteurbandensificationandregenerationoverleapfrogdevelopment.Initiativesthatsucceedinselectedcitiescanbescaledup,whereasfailedattemptscanbescuttled.Ideally,thisdynamicwill,inthelongerterm,resultincitiesofallsizeshavingfunc-tionallandadministrationandplanningservices.SuchanapproachhasbeensuccessfulinBurundi.ItsdecentralizedsystemoflandcertificationemergedfromapilotprojectfirstimplementedinfourcommuneswiththesupportoftheSwissAgencyforDevelopmentandCooperationandtheEuropeanUnion(WorldBank2014).Integrationisgoodforthepoor—andgoodforbudgetsIntegrationmatterstothepoor.Theymaybeawareofthehazardstheypossiblyface,butthepoordependmorethanthewell-offonpublicservicesthatareofteninadequate.Policiesthataimtomoveslumdwellerstolessprecariouslocationsvianewhousingdevelopments,subsi-dies,orurbanupgradesoftenfailtoworkwellbecausepoorerhouseholdsoftenvalueaccessi-bilitytoaffordableservicesandamenitiesandproximitytojobsabovesaferhomes.Inplaceswherelanduseandurbaninfrastructureanddevelopmentdecisionsarenotcoordinated,householdsendupdisconnectedfromlabormarketsandtradingoffsafetyforaccessibility.Localgovernmentsoftenstruggletoprovideessentialurbaninfrastructure,and,untiltheysucceedindoingso,thepoorwillremainvulnerable.Moresecurelandandpropertyrightswouldencourageinvestmentinupgradingand,therefore,self-protection.14Equally,however,theprovisionoflandandaffordablehousinginsaferareaswithaccessiblejobsandessentialserviceswouldgoalongwaytowardloweringpoorpeople’sriskexposure.IntegrationneedstobecoordinatedwithriskassessmentsCityleadersmayspendlesstimeonlong-termplanningbecausetheyareoftendealingwithaseriesofemergencies.Interventionsaimedatimprovingandincreasingdensitycanbedeliveredmoreeasilyandcost-effectivelythroughinvestmentsbeforesettlement.EstimatesofthecostofaffordablehousingtoaccommodatetheburgeoningpopulationofFreetown,SierraLeone,whilereducingexposuretofloodsandlandslides,findthatproactiveplanning(forexample,viainvestmentsinsitesandservices)wouldcostapproximatelyUS$375millionoverthenextdecade.Bycontrast,theprovisionofapublichousingschemewouldcostalmostninetimesasmuch—US$3.2billion(Mukim2018b).Rapid,unplannedurbanizationinlow-andmiddle-incomecountriesisexpectedtoexacer-bateclimatechangeandothershockssuchasheatwaves,flooding,andhealthemergencies.Integratingadaptationwithgoodurbanmanagementwillbeanintegralpartofamelioratingclimatechange–relatedrisksincities.Suchintegrationinvolvesreducingnotonlythelikeli-hoodofimpactoninfrastructureandotherassetsbutalsothevulnerabilityacrossnumerousinterconnectedsystems.Multihazardriskassessments15couldbedeployedtobetterprotectcitiesbybetteridentificationandunderstandingofrisks.Forexample,morethan20,000inhabitantsbenefiteddirectlyfromsuchworkcarriedoutintheMoroccanmunicipalitiesofFezandMohammedia(WorldBank2020a).Suchriskassessmentscanalsohelpincreaselocalcapabilities,beyondthoseofpublicofficials,bydrawingincommunitystakeholders,includingthosefromvulnerablecommunities.InSierraLeone,city-levelriskassessmentsconductedinBo,Freetown,andMakeniusedaparticipatoryapproachthatincludedvulnerablecommuni-ties.Theseassessmentsidentifiedpriorityneeds,investments,andfeasibilitystudiestobolsterPoliciesforPromotingGreen,Resilient,andInclusiveUrbanDevelopment285resiliencetofloodriskandlandslides.Followingthe2017floodingandlandslidesinFreetownthatkilled493people,left600moremissing,andpushedafurther3,000peopleintohome-lessness,theseriskassessmentsweredeployedaspartofpostdisasterrecoveryprogramming(GFDRR2014).Multihazardriskassessmentsalsoinformlanduseplanningandzoning.Byunderstandingwhichareasofacitymaybeadverselyaffectedbycurrentandfuturehazardssuchasfloodingandstormsurges,policymakerscanbetterplanforfutureurbangrowthanddevelopment(Hallegatte,Rentschler,andRozenberg2020).Colombia’sframeworklawpassedin2018,forexample,requiresregional,municipal,anddistrictauthoritiestoincorporateclimatechangemanagementintotheirdevelopmentandlanduseplans.16Effectiveplanningshouldalsoguidepoliciesandchoicesforstructuralinvestmentsandcommunitymeasures,ensuringthatfutureurbandevelopmentiscompatiblewithchangingrisks.Suchinvestmentsandmeasurestoreduceriskscouldincludeerosioncontrolandreforestation;retainingwalls,drainagesystems,dikes,spillways,orreservoirs;orinvestmentsinfloodbufferingzonessuchaswetlandsandswampareas(GFDRR2012).Suchprojectsareoftenmoreeffectivewhenjointlydevelopedandimplementedalongsidethecommunitiesthatstandtobenefit,sothattheycanappropriatelymanagebehaviorlinked,forexample,tolittering,clearingofdrainagecanals,orheedingearlywarningsystems.InJamaica,coastalhazardmappingisunderwaytoupdatelanduseregula-tion;intandem,localauthoritiescontinuetosuccessfullyenforceminimallyintrusive,low-costhurricanestrapsfortheroofsofresidentialbuildings(GFDRR2019).Nevertheless,risk-basedlanduseplanningpresentschallengesbecauseofthepotentialnegativeimpactsonthepoor,whomayresideinhigh-riskareasandrequirerelocationsupport(Hallegatteetal.2017).SuccessrequirescoordinationofplanningwithinformationandinstitutionsTobesuccessful,landuseplanningrequireseffectiveimplementation.Effectiveimplementa-tion,inturn,oftenreliesoncadastres,propertyrights,andotherregulationsandthewilling-nessandcapabilitiesofthepublicauthoritieswhoadministersuchregulations.Forexample,Hallegatteetal.(2017)pointoutthat,althoughlocalauthoritiesmayhaveriskinformationreadilyavailable,incorporatingsuchinformationintourbanplanningcanposeachallengebecausethecostscouldimplytheimmediatereductionoflandvaluesinareasidentifiedasrisky,andthebenefitsofplanning(intheformofavoidedlosses)wouldaccrueonlyintheunknowablefuture.Whenlandandhousingmarketsworkandinformationasymmetriesareminimized,propertyvaluesreflectthedis-amenitiesfromhazardrisks,guidingpeople’sdecisionsonwheretoliveandwhatpreventionmeasurestotake(seetheearlierdiscussiononinformation).Conversely,lackofcoordination,oftenthenorminmanylow-andmiddle-incomecountries,leadstoperverseoutcomes—mostlyforpoorpeople.InBujumbura,Burundi,almost60percentofinformalsettlementgrowthbetween1985and2019tookplaceinareasofurbanexpansiondisconnectedfromexistingagglomerationsandatriskofflooding(Mukim2021).Thelackofplanningstandardsandinfrastructurethatreflectedthegrowingrisksfurtherexacerbatedthephysicalvulnerabilitiesintheseperi-urbanareas.Thus,integratedurbanplanning,combinedwithwell-designedandimplementedregulations,includingforfacilitationoflandandhousingmarkets,canleadtogreatercompactness.286THRIVINGIntegrationwithincitiescanhaveimportantbenefitsClimatechangeisforcingindividuals,families,andevenwholecommunitiestoseekmoreviableandlessvulnerableplacestolive(Rigaudetal.2018).Migrationbecomesanadaptionoptionforpeoplewhenareasfaceadverseshocks,includingfromclimatechange.Italsoallowsdiversificationofincomesinthefaceofshocksandincreasinguncertainty.Vulnerablepeoplehavethefewestopportunitiestoadaptlocallyortomoveawayfromriskand,whenmoving,oftendosoasalastresort.Thoseevenmorevulnerablewillbeunabletomove,trappedinincreasinglyunviableareas.AccordingtotheWorldMigrationReport2022,thevastmajorityofpeoplecontinuetoliveinthecountrieswheretheywereborn—only1in30migrates(McAuliffeandTriandafyllidou2021).17Whenfacedwithaclimatechange–relatedshock,suchasdrought,asdiscussedinchapter3,theyoftenrespondbymigratingtourbanareas.In2016,theUnitedNationsHighCommissionerforRefugeesreportedthat,between2008and2016,anaverageof21.5millionpeoplewereforciblydisplacedeachyearbysudden-onsetweather-relatedhazards,andthousandsmorebyslow-onsethazardslinkedtoclimatechangeimpacts.18Migrationinresponsetoclimateimpactsneednotalwaysbeforceddisplacementinresponsetosuddenevents.Itcouldalsobelinkedtomobilityasaproactiveadaptationstrategy.Internalmigrationpatternsareheterogeneous,withmigrationoccurringsimultaneouslyfromruraltourbanareas,betweenruralareas,andfromurbantoruralareas.Somegovernmentshaveresortedtospecificlegislation,regulations,andpoliciestodiscourageorrestrictdomesticmigration.Yetagrowingbodyofevidencesuggeststhatgreaterruraltourbanmigrationcouldincreaseaggregateproductivityandfacilitateeconomicgrowthwithincountries(Lagakos2020;SelodandShilpi2021).Migrationawayfromshockscouldalsobeassociatedwithincreasingremittancestoaffectedareas,asillustratedbyGrögerandZylerberg(2016)whostudiedtheimpactsofatyphooninVietnam.Theoveralleconomiceffectofclimatechange–inducedmigrationonthereceivingcitydependsonlocalconditionsandthecapacityofthecitytoabsorbalargerlaborforceoflower-skilledworkers.Eventhoughthepolicymixwillvaryacrosscounties,thefundamentalapproachtoeasingsuchmigrationtransitionswouldlikelyremainunchanged.Forexample,decision-makerscouldseektointegratemigrantsbothtolimitimpactsonhostcommunitiesandtoensureinclusiveopportunitiesfornewmigrants(Zaverietal.2021).Citiescouldinvestaswellinbasicservicessuchaswaterandsanitation,schools,healthcare,andsafehousingforpoormigrantsinurbanareasandbroadercitypopulations.Basicwaterandsanitationservicesareespeciallyvitalinslowingthespreadofdiseases,includingCOVID-19(Zaverietal.2021).Morebroadly,proactivepoliciesthatpromotesharedeconomicprogressandaddresssocialfrictionsbetweenmigrantsandhostpopulationswillbecomeevenmorecritical(Borgomeoetal.2021;Zaverietal.2021).Humanadaptationtothechangingclimatewillalwaysincludemigration.Climateandenvironmentalstressescanactbothasadriver(increasedmotivationtomigrate)andasaninhibitor(reducedabilitytomigrate)inindividuals’andhouseholds’decisionstomigrate(Flores,Milusheva,andReichert2021;Quiñones,Liebenehm,andSharma2021;Zaverietal.2021).Thus,publicpoliciesshouldhelprelaxsomeoftheconstraintstomigration,includingliftingregulationsthatpurposefullyrestrictmigrants’accesstosocialandeconomicservicesinurbanlocations.Attheiroriginsorincontextswherelandrightsarecontested,migrantsalsocommonlyfaceinstitutionalbarrierstoout-migration,suchaslandtenureinsecurity.Governmentscouldhelpreducethecostsofmigrationby,forexample,improvingaccesstofinancialmarkets,loweringthebarrierstoassimilationinreceivingareas,andprovidingbetterinformationonjobopportunities.PoliciesforPromotingGreen,Resilient,andInclusiveUrbanDevelopment287Powerfulsocialdynamics—suchascommunalandethnicconflicts,someofwhichdatebacktothedaysbeforeclimate-inducedmigration—canalsoobstructmigration.InSomalia,forexample,settlementpatternsamongnewlyarrivedurbanresidentsarehighlysegregatedonthebasisofclanmembership(WorldBank2021c).Thissegregationcomplicatesmigrants’optionswhenitcomestointegratingintohostcommunitiesandfindingwork;scarcejobsareoftenallocatedtofellowclanmembers.Furthermore,migrationcanshiftclandynamicsindestabilizingways,pittingneighboragainstneighbor.InSomalia,propertydisputesoccurcommonlyinneighborhoodswherelandisscarce,andtheycaneasilyescalatewhencross-pollinatedwithclanrivalries.Theimplicationsforpolicymakersnavigatingsimilarconditionsareratherclear—theymustensurebasicsecurity,putinplaceclearlandregistrationpolicies,andpromoteintergroupcontactthroughpeacefulmeans.Tradeandmigrationcanactassubstituteswhenitcomestoadaptation.Conteetal.(2021)evaluatechangesinlocalspecializationinagriculturalandnonagriculturalsectorsandfindthatfreertradewouldincreasethescopeoflocalspecialization,reducinglossesfromglobalwarmingbutalsoweakeningtheincentivesforpeopletomigrateawayfromtheworld’spoorestregions,whicharemoreaffectedbyclimatechange.Trade—bothdomesticandinternational—alsoplaysacriticalroleinrecoveringfromweather-relatedcrises(BrentonandChemutai2021)byincreasingaccesstofoodandattenuatingpricevolatility.Movementofgoods,medicines,andemergencyworkerscanbecriticaltoimmediaterecoveryfromanaturaldisasterandforrebuilding.InvestmentsInvestmentsingreen,resilient,andinclusiveinfrastructure,includinginnature-basedsolutionscanhelpcitiesaddressclimatechange–relatedrisksInfrastructureinvestments,whenwelldesigned,constructed,andmaintained,canhelppreventandrespondtodisasters,reducingthelossoflifeandproperty.AccordingtoFayetal.(2019),investmentininfrastructureacrosslow-andmiddle-incomecountriesconstitutesbetween3.4and5.0percentoftheirgrossdomesticproduct.Despitethisspending,theinfra-structureoftencannotmeettheneedsofthesecountries’growing,andoftenswiftlyurbaniz-ing,populations.Climatestressescanexacerbatethesechallengesbyleadingtodisruptionsinthesupplyofservices,includingbydamagingassets.19Forexample,urbanfloodsoftendisrupttransportationservicesandknockoutenergynetworks.Importantpreventionmeasuressuchasfloodcontrolsystems,shelters,andprotectionofenvironmentalbufferscanbeembeddedininfrastructureinvestments.Someinfrastructurecouldservemultiplepurposes,suchastheschoolsinBangladeshthatalsoserveascommunitycycloneshelters.AccordingtoHallegatte,Rentschler,andRozenberg(2019),theextracostofbuildingresilienceintoexistinginfrastructuresystems(power,waterandsanitation,transpor-tation,andtelecommunications)wouldaccountfor3percentoftheoverallinvestmentneeds,buteachdollarinvestedwouldyieldfourinreturn.Thecontinuedeffectivenessofinfrastructuredependsonitsquality.Spendingonmaintenancecanbehighlycost-effective.Rentschleretal.(2019)findthatanadditionalUS$1.00spentonroadmaintenanceinOrganisationforEconomicCo-operationandDevelopmentcountriessavesUS$1.50innewinvestments.Thus,investmentoutlaysmustnotignoremaintenancebecauseitbooststheresilienceofinfrastructurewhilereducingoverallcostsinthelongrun.288THRIVINGInvestmentsarethelast“I”becausetheyarebothcostlyanddurable.Hallegatte(2009)estimatesthatacity’sphysicalstructures,onceestablished,couldremaininplaceformorethan150years.Asaresult,infrastructureinvestmentsthataffectlanduseandacity’surbanformcanhaveimplicationsfarintothefuture.Forexample,investmentsinroadsthatpromotemotorvehicleoverpublictransportationuse,therebyencouragingsprawl,cansignificantlyandpermanentlyincreasethecostsofdeliveringbasicservicessuchaswater,sanitationandelectricity,andsocialinfrastructure(clinicsandschools,amongotherthings).20Investinginbasicservicesincitiesinlow-incomecountries—nomattertheirsize—isaleaptowardintegration.Expandinginvestmentinbasicservicessuchaswater,sanitation,elec-tricity,cleanfuelsforcooking,anddigitalconnectivity,aswellasensuringaccesstofinancial,technical,andinstitutionalresources,notonlybuildsresilienceinvulnerablecommunitiesbutalsoenhancesmobilitybyconnectingsmallercitiestomediumandlargecitiesandreducingmigrationbarriersbetweenthem.Becausecitiesarethreatenedbyclimatechange–relatedrisks,decisionsregardingurbanconstructionhaveevenmoreimportantramifications.Privateinvestmentdecisionslinkedtohousinginhazard-proneareaswillintensifyrisksandincreasethenegativeeffectsoffloods,landslides,storms,andotherclimateevents.AsDesmetandJedwab(2022)pointoutinbackgroundresearchforthisreport,themostexpensiverealestatestructures—skyscrapers—arehighlydurable.Inaworldwithswiftlyshiftingandunpredictableeffectsassociatedwithclimatechange,thisdurabilitycanbeadouble-edgedsword.Verydurableinvestmentscould,perversely,increasethelong-runcostsofclimatechange.InvestmentscanhelpcitiesanticipatethechallengesofgrowthOneofthegreatbenefitsofurbanareasisthattheycreatethedensityofdemandthatcanjustifylargesunkinvestments,suchasinpublictransportation,toguidesettlementsandmanagedisasterrisks,thecostsofwhichcanbespreadacrossnumeroustaxpayers.Publictransportationsystems,forinstance,arekeytotransformingdensityintointegratedlabormarketsthatcanfosterbettermatchesbetweenemployersandjobseekers(AvnerandLall2016;Peralta-Quiros,Kerzhner,andAvner2019).Franklin(2018)findsinAddisAbabathatmoreaffordablepublictransportationandlowerjobsearchcostsresultedinmorestable,better-paying,andmoreformaljobsforyouth.PublictransportationsystemsarealsoakeyleverinreducingurbantransportationCO2emissions.Becauseoftheirhighaverageoccupancyrates,publictransportationsystemsaretypicallylessenergy-andcarbon-intensivemodesoftransportationperpassenger-kilometerthanindividualcarsormotorcycles.Suchsystemscanalsobeelectrified,makingthemveryattractivecontributorstoCO2emissionmitigationstrategiesiftheycangetpeopleoutofautomobiles.Investmentscanalsohelpguidesettlementsspatially.Rentschleretal.(2022)documentthaturbanizationinhigh-riskflood-proneareasisoutpacingsettlementgrowthinsafeareas.Onehypothesisisthatlandscarcitypusheshouseholdstosettlewherelandis(orappearstobe)available,oftenresultingininformalsettlementswithbasiclivingconditionsandhighexposuretoseveralhazards.ExamplesofsuchriskyurbangrowthincludeurbanizationintheperipheryofDakar,Senegal(WorldBank2016);settlementsinlow-lyingareas,andevenmangrovesinConakry,Guinea(WorldBank2019a);andencroachmentintowaterbedsinCapHaitien,Haiti(Rentschleretal.2022).Relocationorretrofittingoftheseneighborhoodsisdifficult,costly,andpossiblysensitive.Instead,amuchmoreeffectiveandcost-efficientoptioninvolvesanticipatingurbangrowthandguidingitspatially.PoliciesforPromotingGreen,Resilient,andInclusiveUrbanDevelopment289Howcanthisbedone?Layingoutbasicinfrastructurecanactasapowerfulsignalforhouse-holdstosettleinareasthatauthoritieshaveidentified,awayfromhighrisks.Intheearlydaysofanarea’sdevelopment,itrequiresonlythemostbasicinfrastructure,essentiallyrightsofwayforroadsandwell-demarcatedlandplots.Upscalingoftheinfrastructurecanhappeninasecondphaseoncehouseholdshavesettledin.Angel(2017)documentshowthisapproachsucceededinasquattercommunityinLima,Peru.Andseveralrecentpapershavehighlightedthelong-lastingpositiveeffectsofsitesandserviceprojectsimplementedinthe1970sand1980s—forexampleinIndia(Owens,Gulyani,andRizvi2018)andTanzania(Michaelsetal.2021).Finally,investmentsindisasterriskmanagementnotonlyhelpreducedisasterdamagesbutalsoareanessentialcomponentofgrowthanddevelopmentstrategies.Inadditiontodamagemitigation,suchinvestmentscandeliverwidereconomicbenefitsintheformofincreasedproductivityduetoareductioninbackgroundriskandotherdevelopmentco-benefits.Avneretal.(2022)showthatfloodmitigationinthecentralpartofBuenosAires,mainlyintheformofstormwaterdrainageandretentioncapacityinvestments,canraisethevalueoflandtoadegreethatcovers,andprobablyexceeds,thefloodprotectioninvestmentcosts,whileallowingforbetteruseofthepreviouslyaffectedlocationandreducingaggregatecommutingcosts.Withincreasingdocumentationofthewidereconomicbenefitsofdisasterriskmanage-mentinvestments,theapproachshouldchangefromareactivetoaproactiveone.InvestmentscanhelpcitiessecurewaterSecuresourcesofwaterforcitiesarefundamentaltodevelopmentandwell-being(Jensenetal.,forthcoming;Saltieletal.,forthcoming).Theworld’scitiesincreasinglyfacethedifficulttaskofmeetingtherisingwaterandsanitationdemandsoftheirresidentsinasustainableway.Cityplannerswillneedtorethinkurbanplanning,andwaterplannerswillneedtofactorurbanplanningintotheirowndecision-makingprocess(OECD2015b).Throughthisprocess,itwillbecriticalfordecision-makerstolookbothoutwardandinwardtoincreaseurbanwatersecurityandresilience.Theywillneedtoexpandandincreasethemenuofwatersupplyoptionswhilealsomanagingwaterdemand(Zaverietal.2021).AnemergingideafromChinaistobuildcitieslikespongessothattheycanabsorbrainwa-ter(Chanetal.2018;Wishartetal.2021).Thesystemmimicsthenaturalhydrologicalcycleandisdesignedtopassivelyabsorb,clean,anduserainfallinanecologicallyfriendlyway.Theideaistorestorewetlandsandbuildgreeninfrastructuretoretainandstorewater.Thissystemwouldnotonlydealwithasuddenexcessofstormwaterbutalsoreuseittohelpmitigatetheimpactofdrought(Zaverietal.2021).By2030,Chinaaimstohave80percentofitsurbanareasbesponge-like(Jensenetal.forthcoming;Wishartetal.2021).Citiesmustalsolookbeyondtheirboundariesandinvestinnaturalinfrastructure.Suchinvestmentiscriticalbecauselandusechangesinupstreamwatershedsaffectmorethan90percentofurbanwatersources,withdegradedwatershedsincreasingwatertreatmentoperationcostsincities(McDonaldetal.2016).Asthechallengemountstoabsorbthegrowingdemandsofurbanpopulationsandasshockstowatersuppliesincrease,multilevelgovernancestructureswillplayacriticalroleinensuringurbanwatersecurity(OECD2015b;Saltieletal.,forthcoming).Inthatrole,theymustdothefollowing:••Recognizethaturbanwaterispartofalargerhydrologicalandeconomicsystemthatencom-passesruralandperi-urbanareas.••Recognizethatanintegrated,coordinatedapproachisneededtomanageallsourcesofwatersupplies(surfaceandgroundwater,nontraditionalsourcessuchasdesalination,wastewaterreuse,andstormwaterretention)alongwiththeusesofwaterinandoutsidethecity.290THRIVING••Recognizethatachievingmutuallybeneficialwatermanagementobjectivesrequirescollab-orationacrossadministrativeboundaries,suchaswithupstreamcatchmentmanagersanddownstreamstakeholdersaffectedbywateruseinthecityandperi-urbanareas.••Recognizetheneedforpolicy,institutional,andregulatoryreformsinservicedeliveryinordertodevelopadequateincentivesforachievinguniversalaccessandimprovedservicedelivery(WorldBank2018).TheWorldBank’snewwater-securecitiesinitiativeaimstohelpdecision-makersunderstandandaddresstheseinterlinkedproblemsandopportunitiesinurbanwatersystems,sothatcitiescanprioritizeactionstoachievewatersecuritytailoredtotheirspecificconditions(seebox5.3foradescriptionoftheseinterlinkedchallenges).Buildingwater-securecities:Usinginformation,incentives,andinvestmentsCitiescanstrengthentheirwatersuppliesandincreaseaccessbydiversifyingavailablewatersources.aForexample,citiescanembracetheideaofreusingandrecyclingwastewaterandharvestingstormwaterasalternativesourcesofwatersupply.Byreturningwaterbackintotheeconomy,theycancaptureitsfullvalue—asaservice,aninputtoprocesses,asourceofenergy,andacarrierofnutrientsandothermaterials(Delgadoetal.2021).bSuchanapproachwillhelpcitieshedgeagainstwaterriskswhilereducingresourceextractionandenvironmentaldegradation.cReusedwatercanalsobenefitstreams,rivers,lakes,wetlands,andaquifers,inpartbecausetheexcesswaterreturnedtonaturalsystemsisofbetterqualitythanstandardtreatedwastewater(TortajadaandvanRensburg2020).Althoughinsomecasesexpandingthewatersupplythroughinfrastructure(suchasdesalinationplants)willbecritical,trulysolvingthisprobleminthelongtermandinanefficientmannerwillrequiredemand-sidemanagementorincentives(Saltieletal.,forth-coming;Zaverietal.2021).Thesewillincludescarcitypricingofwaterandtechnologi-calsolutionsthatreducewaterusebyhomesandbusinesses.Fewurbanservicesareassubsidizedasmunicipalwatersupplies.ArecentWorldBankstudyfindsthatsubsidiesinthewatersupplyandsanitationsectorwereasmuchas2.4percentofgrossdomesticproduct(low-incomeeconomiesareonthehigherend),withurbanexpendituresaccountingfor76percentofthissubsidy(Andresetal.2019).Yetthepoorest20percentofthepopulationcapturedamere6percentofwatersupplyandsanitationsubsidies,whereasthewealthiest20percentcaptured56percent.Insteadofensuringaccess,thesesubsidiespushthepriceofwatersolowthatinefficiencyisincentivized,servicesustain-abilityisthreatened,andresourcesareoverexploited(Zaverietal.2021).Subsidyreforminthewatersupplyandsanitationsectoristhereforecritical.Citieswillneedtoensureaffordableaccesstowaterwhilealsodiscouragingtheexcessconsumptionthatmayresultfromlowprices.Onewaypolicymakerscanbalanceequityandconservationistousetargetedsubsidiesorbloctariffsthatcanbestrategicallyemployedtoensurethatthemostvulnerableresidentsretainaccesstoaffordablewater(Damaniaetal.2017).Dynamicallyefficientvolumetricwaterpricing,whichincorporatesthescarcitypriceofwaterintothecurrentprice,canalsominimizetheaveragewatertariffpaidbyhouseholdsovertime(Grafton,Chu,andKompas2015;Zaverietal.2021).Oneofthemajorchallengesinofferingincentivesforlowerwateruseisthat,evenwithpriceincentivesinplace,behavioralchangecanoccurslowly.HouseholdsmaylackinformationBox5.3PoliciesforPromotingGreen,Resilient,andInclusiveUrbanDevelopment291InvestmentscanhelpcitiesretrofitPastpoliciescanweighheavilyonthepresent.Forexample,defectsaredifficulttodetectandhardertoremedyinmanystructuresbuiltearlier.CostaandKahn(2011,2013)documentthathomesbuiltinthe1970sinCaliforniaconsumedmoreelectricitythanmorerecentlybuilt,observationallyidenticalhomes.Builtwhenelectricitypriceswerelow,the1970shomeswerelikelytobeenergyinefficient.Thus,retrofittingresidentialhomesandbuild-ingscouldhaveimportantimpactsongreeningthroughitseffectsonenergyconsumption.Infact,theInternationalEnergyAgencyestimatesthat75percentofthetotalvaluecreatedbybuildingretrofitscomesfrompublicbenefitsthatgowellbeyondthemaingoalofenergysavings(IEA2015).Thepositivespilloversincludeemploymentopportunities(oftenforlocalsupplychains),higherpropertyvaluesandthushighermunicipalrevenue,healthbenefits,andadditionalbenefitsaccruingtoenergyutilities.Thehousingstockcanalsoberetrofittedtoincreaseitsresiliencetotropicalcyclones,landslides,floods,andotherhazards.Housingaccountedfor93percentoftotalprivatedamagesandlossesfromthe2013floodinSt.VincentandtheGrenadinesand44percentfollowingthe2010tropicalstorminGuatemala.However,retrofittingbuildingstructuresandinfrastructureindenselypopulatedareastohelpthemadapttonaturalhazardriskscanalsoonhowmuchwatertheyuseandwhichactivitiesrequirethemostwater,makingitdifficultforthemtomakedecisionsaboutlimitingwateruse(Zaverietal.2021).New“smart”metersandothertechnologicalsolutionsofferapromisingsolutiontothisproblem.Publicawarenesscampaigns,educationalprogramsinschools,andnudgestopromoteacultureofwaterconservationcanhelpconservationefforts(WorldBank2015b).dPluggingleakypipescanalsogoalongwaytowardconservingwater.Insomecities,morethanhalfofthewaterlossesinthedistributionsystemareduetophysicalleaks(Jensenetal.forthcoming).Fixingpipescouldsoonbecomeeasierandcheaperthankstoroboticsystemsthat,bysensingpressurechangesaroundleaks,candetectandrepairthemwhilepipesarestillinuse(Balch2014).a.Singaporeisaclassicexample.Despitebeingoneoftheworld’sscarcestfreshwaterlocations,ithasensuredadequatewaterprovisionbyembracingafour-tapmodel.Thefirsttapisthesupplyofwaterfromlocalcatchments.Mostofthecityisdesignedasacatchmentwithanintegratedsystemofreservoirsandanextensivedrainagesystemthatcollectsrainwaterandchannelsitintostoragereservoirs.Fromthestoragereservoirs,watertreatmentplantstreattherainwatertobesuppliedtothecityaspotablewater.Thesecondtapisimportedwater.Thethirdtapistreatedwastewater,whichistreatedtosuchhighstandardsthatitcanbedrunk.Thefourthtapisdesalination,wherebywaterfromtheseaisdesalinatedandsuppliedtothecity(Zaverietal.2021).b.Numerouscitiesreusewastewaterbothtosupportindustrialandnearbyagriculturalactivityandtoprovideutilitieswithrevenue.Examplesincludethesaleoftreatedeffluenttoindustrialusers(LingyuanCity,China)andtoapowerplant’scoolingtowers(SanLuisPotosi,Mexico);theuseofrecoverednutrientsforfertilizer(Dakar,Senegal);andtheoutsourcingofmunicipalwastewatertreatmenttominingcompaniesinexchangefortheuseoftheresultingtreatedwastewater(Arequipa,Peru)—seeDelgadoetal.(2021).c.InWindhoek,Namibia,oneofthefirstcitiestocreateadrinkingwatersupplyfromreusedwater,theGoreangabWaterReclamationPlantcontributesuptoathirdofthecity’stotalwatersupplyintimesofdroughtwhensuppliesfromnearbyreservoirsareespeciallymeager(TortajadaandvanRensburg2020).d.InBogotá,Colombia,apublicinformationcampaignthatincludedpublicationofdailyreportsofwaterconsumptioninprominentnewspapers,advertising,andminorsanctionsonbusinesseswiththehighestconsumptionlevelsachievedlong-termbehavioralchange(WorldBank2015b).Box5.3continued292THRIVINGbeacostlysolution.21Hallegatteetal.(2017)estimatethatretrofittingofanexistingvulner-ablestructurecouldaccountfor10–50percentofthestructure’svalue.EvidencefromLatinAmericasuggeststhatretrofittinginfrastructureaftersettlementhasoccurredcancostuptothreetimesmorethaninstallationalongsidehousingconstruction(Fernandes2011).Costsarenottheonlyconstrainttoretrofittingbuildings;incentivesmaynotbealigned.Ghesquiere,Jamin,andMahul(2006)demonstratethatlandlordsoftenhavelittleincentivetoretrofitbuildingsbecausetheymaynotincorporatethepotentialcoststothetenantintermsoflostassetsandlives.Informationmaybeimperfect,sotherisksofthenaturalhazard,orthecostsandbenefitsoftheinvestment,maynotbewellunderstood.Often,theresponsestoachangingclimatecanbefoundinnature.Nature-basedinvestmentsusenaturalormodifiedecosystemstopropupresiliencetodisasters.Fromtheirsurveyofmultiplenature-basedprojectsaimedatdefendingcoastalhabitats,comparingthemtoinvest-mentsin“grayinfrastructure,”22Narayanetal.(2016)findthatnature-basedinvestmentscanbehighlycost-effectiveforprotectingcoastalsettlements.Trees,wetlands,greenspaces,andriverscanalleviatetheurbanheatislandeffect(Rajetal.2020;Tan,Lau,andNg2016).Nature-basedsolutionsreducetheimpactofnaturalhazards,suchasflooding,erosion,landslides,anddrought,incities.Theyoftendosobycomplementinggrayinfrastructuresuchasstormdrains,embankments,andretainingwalls(WorldBank2021b).AvoidingsomeinvestmentscanincreaseresilienceMostriskreductioninvestmentsareno-regretactionsbecausetheysavelivesandreducesocialandeconomiclossesfromdisasters.Sometimes,however,resiliencemaybeafunctionofnotmakinganinvestment,suchasinmitigationeffortsthathavetheunintendedconse-quenceofincreasingexposuretodisasterrisks.Forexample,buildingleveesisoftenviewedasaneffectivewaytoreducetheimpactoffloods.Yetitcanalsoleadto“theleveeeffect”—theparadoxthatbuildingleveesmayincreaselossesfromdisastersbecausetheprotectioncreatesafalsesenseofsecurityandinducesdevelopmentandattractsmorepeopletoflood-proneareas(Georgic2019;Hutton2018).23Notmakinginvestmentsalsomakeseconomicsensewhenthehighinvestmentcostsyieldlowreturnsinriskreduction.AccordingtoNichollsetal.(2019),althoughthetotalinvestmentcoststobuildandupgradedefensesagainstsea-levelriseandcoastalfloodingcouldreachUS$18.3trillioninthehighest-costscenario,thesedefensesprotectonlyone-thirdorlessoftheworld’scoasts.Anotherstudysuggeststhatsomefloodprotectioninfrastructures(suchaslevees,dams,andwalls)couldyieldnegativenetbenefits,withreturnsinfloodriskreductionlowerthantheirimplementationandmaintenancecosts(WorldBank2021b).Alternatively,resiliencecouldbeimprovedbylow-costapproachesornature-basedsolutions,especiallyforcommunitiesinlow-densityareas,orbyfacilitatingmigrationtolessriskyplaces(WorldBank2019b).SequencingofthefiveI’sTheorderingofthefiveI’sisdeliberateandbalancestheneedtomaximizethedesiredeffectsofapolicyanddoitinthemosteconomicalway.Earlyandeasyinterventionslinkedtotheprovisionofinformationcouldimprovemarketoutcomes.Whenhouseholdsandfirmshavebetteraccesstoinformation,theycanbetterunderstandthebenefitsandcostsoftheiractions,therebystemmingtheneedforexpensivegovernmentinterventions.Inthesamevein,well-implementedincentivescanscalequickly(forexample,byaffectingbehaviors)atrelativelylowcosts.Economicincentivescanhavemonetaryimplicationsthatcouldleadtosnowballingcosts(suchaswithtaxrebatesorsubsidies),buttheycouldalsobedeployedtodisincentivizePoliciesforPromotingGreen,Resilient,andInclusiveUrbanDevelopment293behaviorsthroughmonetaryfines,taxes,andthelike.Bytransferringandmitigatingrisks,well-functioningmarketsforinsurancecouldlowerriskstoapointthatwouldminimizetheneedforexpensivegovernmentinterventions.Integrationentailsplanningbycities,oftenbeforetheurbansettlementsarebuilt,andreducingbarrierstomigration.Becauseofthedura-bilityofurbaninvestments,well-thought-throughintegrationcouldhavelong-lastingbenefits,notleastforthebalancesheet.Finally,investmentsininfrastructureofteninvolvelargepublicoutlays,andinmanycasesaretheprimary(andverynecessary)responsefollowingdisasters.Durableinvestmentsinmitigationandadaptationstrategiesarecruciallyimportantbecauseoftheirimplicationsforlonger-termoutcomes.ThefiveI’srepresentarelativelysimpleapproachtoorganizingthemanypolicyinstrumentsavailableintodistinctbundles.Nevertheless,manyinterdependenciesexistbetweenthesesetsofinstruments.Insomecases,theyplayoutincomplementaryways—thatis,policiesacrossthebundlesstrengthenimpactwhenimplementedtogether.Forexample,informationhelpsfacilitatemigrationdecisionsandthusintegration.Informationalsoallowspricestobetterreflectrisks,therebybetterincentivizingbehaviors.Largepublicinvestmentsthemselvesserveasaninformationsignaltoprivatefirmsandhouseholdsregardingthedirectionoffuturedevelopment.Andincentivesviapricesorregulationcandrivetheworkingofprivateandpublicinsurancemarkets,affecting,inturn,investmentdecisionsbyprivateagents.Thus,decisionsacrossthefivesetsofI’sshouldnotbemadeinisolationofeachother.Forexample,Avner,RentschlerandHallegatte(2014)showhowmuchmoreefficient,andaccept-able,acarbontaxiswhencombinedwithadensepublictransportationnetwork—thatis,com-biningincentiveswithinvestments.Similarly,ViguiéandHallegatte(2012)demonstratehowaseriesofpolicyoptions(agreenbelt,alandusezoningpolicy,andatransportationsubsidy),representinginformationandincentives,iftakeninisolationcouldhelpachieveagivenpolicygoalyetbedetrimentaltotheachievementofothers.Combiningtheseoptions,however,resultsinapolicypackagethathelpstheachievementofallobjectives.Theresilienceandreliabilityofinfrastructureassetsandservicescanalsobeenhancedbywell-designedregulations,construc-tioncodes,andprocurementrules.Hallegatte,Rentschler,andRozenberg(2019)demonstratehowtheseinterventions—whichusuallyrequireasmallinvestmentupfront—couldsignificantlyimprovetheoverallresilienceofinfrastructuresystemsandgeneratelargebenefits.Whomakesthechoices?TurningthefiveI’sintoactionableprescriptionswillrequireeffectiveinstitutions.Thereiswidespreadagreementthatclimateactionwillrequiremultilevelinvolvementnotonlybycityandnationalgovernmentsbutalsobynonstateactorssuchasmultilateralinstitutions,largemultinationalcorporations,smallenterprises,andcivilsocietygroups.Atthegrassrootslevel,communitiesareoftentheonesintheleadonclimateaction.CitygovernmentsWiththeirknowledgeoflocalcontextandtheabilitytomobilizetheircommunities,citiesandcitygovernmentswillplayacriticalroleinclimatechangemitigationandadaptation.Infact,manyareasoflocalgovernmentpolicyrelatecloselytoclimateaction—forexample,infrastruc-tureinvestmentsandserviceprovision,landuseregulations,andemergencyplanning.Becausecitieswilllikelybearanoutsizedshareofclimateimpacts,cityleadersareprobablythemostmotivatedpoliticalactorstotakeonclimatechange.Globalcreditagenciesnowconsidertheclimatechangepreparednessofcitiesinassessingtheriskoflendingtothem.24Andmanycitieshaveadoptedmoreambitiousgreenhousegasreductiontargetsthanthoseofcountries.294THRIVINGIn2017,overallcommitmentsbycitiesamountedto27percentofgreenhousegasreductions,exceedingnationalcommitmentsbyalmost7percent(Konaetal.2018).TheCovenantofMayorsforClimateandEnergyalreadyincludesalmost11,000localandregionalgovernmentsignatoriesfrom505countrieswithatotalpopulationof340million.25Themayor’swedgeLocalgovernmentsandcityleaderscanaffectclimateactionbyinfluencingandimplementingclimatepoliciesputinplacebyhigherlevelsofgovernment;designingandimplementingcity-specificpoliciesandinitiatives;and,crucially,helpingcoordinatecollectiveclimateactionintheircities(DeConincketal.2018).Howlocalgovernmentsdeliveronthisroledependsontheirscopeandtheircapacity.Thisso-calledmayor’swedgeconsistsoftherangeofpoliciesthatcityleaderscanhopetoinfluence,includingthosepredeterminedbyhigherlevelsofgovernment.26Thefirstcomponent—scope—istheadministrativeresponsibilityofthecityauthoritytodesignandadministerpolicies.Thesecondcomponent—capacity—istheavailabilityofhuman,technical,organizational,andfinancialresourcestoplanandimplementpolicies(WorldBank2015a).Scope.Localgovernmentscopeoradministrativeresponsibilitycanbeproxiedbytheleveloffiscaldecentralization.Figure5.1showsthattheshareofspendingbylocalgovernmentsvariessubstantiallyacrosscountries.Spendingbylocalauthoritiesisoftencorrelatedwiththeircontroloverkeyaspectsofservicedelivery,whichareoftenlinkedtoclimateadaptationandmitigation.Nevertheless,spendingdoesnotequalabilitytoprioritizeandredistributefinancialresources,whichinmanycountriescomeasringfencedgrants,suchasinUzbekistan(Sivaevetal.2022).Countriesdifferinthedegreetowhichtheycentralizethedesignandimplementationofclimatepolicy.InChina,climatepolicytargetsaresetintheNationalClimateChangeProgramasapartofthe11thFive-YearPlanandarethenpassedontoprovincialleadinggroupsforimplementation.Althoughtheyhaveasayinthespecificpolicytoolsselectedtoimplementthetargets,theprovincesaresubjecttonationalmonitoringandoversight.Indiahasadoptedasomewhatlesscentralizedapproachthatallowsmoresubnationalinitiative.AlongwiththeNationalActionPlanonClimateChange,eachstatehasitsownStateActionPlanonClimateChange;theseplansfocusmostlyonimplementingnationalleveltargets,buttheyallowsubstantialflexibilityinpursuingstate-levelconcerns(Somanathanetal.2014).Theauthorityoflocalgovernmentstotakeclimateactionalsodependsonregulationsandpoliciesenactedbyhigherlevelsofgovernmentinspecificsectors.Theabilityoflocalgovernmentstodesignorimplementsuchmeasuresiscloselyrelatedtotheauthoritygrantedtothemintradi-tionalareassuchaslanduse,transportationpolicy,andenergymarketregulations.Suchauthor-itydiffersbetweencountriesandcanevendifferwidelyacrosscitieswithinthesamecountry.Forexample,intheUnitedStatesthedejureabilityoflocalgovernmentstoactonvariousaspectsofclimateadaptationandmitigationdependsonwhichstatetheyarein(Blanchard2021).Capacity.Thesecondcomponentofthemayor’swedge,capacity,iscriticalforimplementa-tion.Institutionalcapacityisbroadlyrecognizedasacriticaldeterminantofalocalgovern-ment’sabilitytoplanandimplementclimateaction(Setoetal.2014).Capacitycouldalsobeendogenous.Reckienetal.(2015)suggestthathigherexposuretoclimaterisks(suchaslocationinlower-lyingareas,proximitytothecoast,orhighersummertemperatures)maybelinkedtoalowerinstitutionalcapacityoflocalgovernments.Eventhoughnocommonmetricsexistformeasuringlocalgovernmentcapacity,itcouldbeapproximatedthroughlocalgovernmentsize(numberofemployees),localtaxextractioncapacity,andtheproductivityofpublicemployees(WorldBank2015a).SuchasimplePoliciesforPromotingGreen,Resilient,andInclusiveUrbanDevelopment295approach—focusingmainlyonstaffingandfiscalcapabilities—maybesomewhatlimitedinproxyingforthehighlytechnicalnatureoftheskillsneededforclimateaction.Countriesandcitiesareoftenconstrainedbypoorcapacity—administrative,fiscal,technical—soitiscriticalthatpolicymakingtakesuchconstraintsintoaccountandacknowledgethatsomecapacityconstraintscannotberelaxedovernight.Bakeretal.(2012)evaluatelocalclimateadap-tationplansinSoutheastQueensland,Australia,andconcludethatlocalgovernmentswerenoteffectivelyplanningfortheimpactsofclimatechange.Althoughawareofexpectedimpacts,thesegovernmentshadlimitedcapacitytousethatinformationtodevelopgeographicallyspecificactionplans.MukimandTingting(2018)studyacountrywideprocessofcounty-to-cityupgrad-inginChinainthe1990stoidentifywhetherextendingthepowersofurbanlocalgovernmentsleadstobetterprivatesectorperformance.Theyfindthatamismatchbetweentheparametersoflocalgovernmentscopeandcapacitycanleadtonegativeoutcomesforprivateenterprises.Simplyput,grantingadditionalpowerstolocalgovernmentsdoesnotnecessarilytranslateintobetteroutcomeswithoutthecommensurateincreasesincapacitytousetheadditionalauthority.Expandingscopeandcapacity.Whencitiesrunintothelimitsoftheirscopeandcapacitytoactontheirpriorities(climateorotherwise),theycantakeactiontoexpandthem.Scopeisoftentheprerogativeofhigherlevelsofgovernment,soexpandingitmayrequirelobbying.Capacity,bycontrast,canbeexpandedbyseekingtechnicalsupportorbuildingpartnershipslocally,eitherwithneighboringcitiesandmunicipalitiesorwiththeprivatesectorandcivicandacademicleaders(WorldBank2015a).Forexample,buildingapublic-privateclimatecoali-tioncouldbeawaytoexpandtheinfluenceoflocalgovernments(seebox5.4fordiscussionofthepublic-privatedialogueforclimate).020406080100JamaicaPeruLatinAmericaandtheCaribbeanMaltaDenmarkWesternEuropeAzerbaijanKazakhstanEuropeandCentralAsiaMalaysiaChinaAsiaEswatiniSouthAfricaAfricaShareofgovernmentspending(%)Figure5.1Shareofgovernmentspendingbylocalgovernments,byselectedregionsandcountriesSource:OrganisationforEconomicCo-operationandDevelopmentandUnitedCitiesandLocalGovernment(OECD-UCLG)WorldObservatoryonSubnationalGovernmentFinanceandInvestmentDatabase,2016(https://www.oecd.org/regional/observatory-on-subnational-government-finance-and-investment.htm).296THRIVINGAdoptingthepublic-privatedialogueforlocalclimateactionFromdevelopinggreentechnology(Haselipetal.2015)totransitioningtomoresustain-ablebusinessmodels(Burchetal.2016),theprivatesectorcancontributetoactiononclimatechangeinmanyways.TheOxfordStreetCorridorPartnershipinGreaterManchester,UnitedKingdom,isagoodexampleofpublic-privateengagementforclimateadaptation.TheCorridorPartnershipboard,madeupofkeylocalpublicactors,largeprivatesectorfirmsinthearea,andlocaluniversities,ledtheefforttoanalyzethepossiblelocalclimatechangeimpacts.Thepartnershipthenannouncedforecastsofthreeclimatepathwaysforthearea:statusquo,furtherintensificationofdevelopment,andgreening.Afterpublicdiscussionoftheseforecasts,locallandownerscommittedtoaprogramofgreeningtheareathatwouldhelplocalclimateadaptationandlimitthepotentialheatislandeffect(Carteretal.2015).ExperiencesliketheoneinManchestercanbereplicatedusingthepublic-privatedialogue(PPD).PPDisawell-establishedapproachforprivatesectorengagementtraditionallyusedinindustrialdevelopment.ThecorestrengthofthePPDapproachisitsversatilityanditsadaptabilitytospecificcontextsandgoals.Sivaev,Herzberg,andManchanda(2015)drawonglobalexperiencetoproduceasetofobservationsonhowtoensuretheeffectivenessofalocalPPD(figureB5.4.1).Forexample,theoptimalspatialscaleofthedialogueshouldmatchthegoalbeingpursued.Ifthegoalistoimprovewalkingandcyclinginfrastructure,justneighborhoodengagementcouldproduceresults.But,ifthegoalistolimitgreenhousegasemissionsfromcommutingbylimitingjourneysandchangingthemodemix,thenengagementofactorsfromacrosstheurbanagglomer-ation(ortravel-to-workarea)wouldbenecessary.Box5.4Source:Sivaev,Herzberg,andManchanda2015.FigureB5.4.1Dimensionsofadaptabilityofthepublic-privatedialogueAreaScopeInstitutionalizationLeadershipOwnershipFocusParticipationLocalSector-specificTemporaryinitiativePrivate-drivenLocallydriven/sustainedSpecificchanges/SpecificgoalFewactorsNationalEconomywidePermanentinstitutionPublic-drivenThird-partybrokerage/supportGeneralorientations/ManygoalsManyactorsPoliciesforPromotingGreen,Resilient,andInclusiveUrbanDevelopment297LocalgovernmentactionsforadaptationandmitigationBecauselocalgovernmentsareknowledgeableaboutlocalhazardsandvulnerabilities,theycanplayacriticalroleinadaptationbynotonlyraisingawarenessandissuingearlywarningsofhazardsbutalsoadoptinglanduseplanningandurbandesignthatcansubstan-tiallyreducethescaleoffuturedamage(Carteretal.2015;Khan2013).Infact,somelocalgovernmentswereamongthefirstpoliticalactorstoinitiateclimateaction.Rotterdam,intheNetherlands,wasoneofthepioneersinadoptingacomprehensiveadaptationstrategy(CityofRotterdam2008).Almada,Portugal,hasbeenimplementingmeasurestoprotectthefragilecoastaldunesystemforoveradecade;andBurgas,Bulgaria,acityontheBlackSeacoast,hasbeencollaboratingwithlocaluniversitiesandnongovern-mentalorganizationssince2013toraiseawarenessofrisksthatthreatenthecommunity(EuropeanUnion2013).Theabilityoflocalgovernmentstoinfluencemitigationislargelyderivedfromtheirtraditionalurbanfunctions.Localauthoritiesoverseeimprovementsinlocalpublictransportationservices,includinginvestmentsinlocalroadsandcyclinginfrastructure,alongsidezoningandlanduse.Theycantheneffectivelyinfluencetheurbanlayoutofcitiesanddeterminedensityandinflu-encecommutingpatterns—bothparametersthatdrivecompactnessofcitiesandthustheiremis-sions.Andlocalgovernmentscanmanageandregulatelocalutilities,whichoftenshapeenergyconsumptionbehaviorsinthecommunity(Gerda2021).Meanwhile,localleaderscaninfluencethechoicesandbehaviorsofresidentsthroughcommunityengagementpractices.NationalgovernmentsAllthissaid,localgovernmentshaveonlyalimitedsayinmanagingclimatechange.Becausethecausesofclimatechangeandoftenthesourcesofimpactsareexternaltocities,mayorsandvotersmaybenotinclinedtosetasidealreadyscarceresourcesformitigationmeasuresthatdonotbenefitthemdirectly.Thus,higherlevelsofgovernmentmayneedtocommittopolicyandinvestmentapproachesthatsupportlocalgovernmentsandgivethemincentivestobetterplanforandinvestinaddressingclimatechangeimpacts.Nationalgovernmentsprovidestrategicoversight,facilitateaccesstoclimatefinance,andhavethecapacityandauthoritytodriveclimateactionbycreatingasupportiveenablingenviron-ment.Drivingasectoralapproach,nationalgovernmentscancreateplansthatmainstreamclimateaction(Somanathanetal.2014).Nationalprogramsonemissionsandcleanenergystan-dards,carbonpricingmechanisms,appliancestandards,andgreenfinancingaremorelikelytoachieveeconomiesofscalebycreatinglargermarketsforhigh-tolow-techcleanertechnologies.Becauseclimatechangewillcontinuetogeneratesignificantdisruptions,threateninglive-lihoods,nationalgovernmentscanimposeclimatechangeregulatorymeasuresonthelabormarket.Throughpolicyandregulatoryinterventions,thegovernmentcouldspureconomicrestructuringbymanagingtransitionandenablinggreengrowth.Thegovernmentsofindus-trializedcountriesarepositionedtopursuegreenenergypolicies,movingjobstowardtherenewablessector.Pestel(2019)pointsouttheimportanceofconsideringbig-picturepoliciesandtheirimpacts—bothdirectpositiveimpacts(suchascreatingnewjobsintherenewablessector)andindirectnegativeimpactsonthelabormarket(suchasstiflinglabordemandandimposingadditionalproductioncosts).Environmentalregulationsthatinduceinnovationoftenspurgrowthandemployment(Horbach2020).Nationalgovernmentsplaytheleadingroleinembeddingsocialprotectionintoclimateplansandshouldfocusonclimateriskswithinsocialpolicies(Costellaetal.2021).298THRIVINGMoreover,theyholdthekeytosettingpolicyframeworksforinsuranceandcanprovidecoverageforhighlevelsofphysicalandbusinessrisks.Asidefromlegislativeandregulatoryinterventions,greatercoveragecanbeachievedthroughsubsidiesandotherfinancialincentivestopromotetheaffordabilityofdisasterinsurance(OECD2015a).Workingwithinsurancestakeholders,governmentcanmonitorlosstrends,improvehazardmodeling,addresscausesofclimaterisk,andprepareforresilience(Gupta2008).Atthenexusofmajorpolicypractices,nationalgovernmentscanunlockfinancialbarriers.Theycanreducethenegativeeffectsofclimatechangethroughdisasterriskfinancing.Bysecuringprearrangedriskfinancingmechanisms,governmentscanempowersubnationalandlocalgovernmentstoavertandminimizetheimpactsofclimatechange(OECD2015a).Theycanalsoremovepoliticalandinstitutionalbarriersbecausetheyholdthepowertofacilitateactionbylocalgovernmentsandtheprivatesectorthroughtheirlegislative,executive,andjudiciarybranches.HsuandRauber(2021)suggestthatanationalgovernmentcanthusbridgethegapbetweenthecity,subnational,andnationallevels,andispositionedtosetthestageforpolycentricclimategovernancesystemstoensurepolicycohesionandintegrationandtoavoidfragmentationthatcouldundermineprogresstowardachievingclimateactiongoals.Inapolycentricclimategovernancesystem,anationalgovernmentprovidesclearmandatestoandownershipbysubnationalgovernmentsandcities,improvesintegrationandcoordination,andfast-tracksaccesstodataandinformation.Overall,althoughtherolethatnationalandlocalgovernmentscanplayinclimateadaptationandmitigationwilldependonthenationalandlocalcontext,certainprinciplesshouldhelpdefinethatrole.Thefirstprincipleisthatclimateactionineverycountrywouldrequiremultilevelgovernance;evencentralizedstateswithlimitedlocalcapacityshouldfindavenuestoengagelocalauthoritiesandmaximizethebenefitsoftheirlocalknowledgeforclimateadaptationandmitigation.Table5.1presentsasummarylistofotherprinciplesfornationalandcitygovernmentsindefiningtheirroles.Thelistinthetableisnotexhaustive,butratherispresentedasguidance.Table5.1Principlesfordefiningtheroleofnationalandcitygovernmentsinclimatemitigation/adaptationNationalgovernmentsCitygovernmentsSettingtargets·Definehigh-levelgreenhousegasreductionandclimateadaptationtargetsbysectorandterritory.·Elaboratedetailedlocaltargetsconsis-tentwithnationalgoalsbasedonlocalconditions,economicspecialization,developmentchallenges,majorgreen-housegascontributors,andexposuretoclimateandotherdisasterrisks.Planningandimplementation·EstablishcriticalnationalpolicyframeworksacrossallfiveI’s.a·Establishregulationsandincen-tivestodriveadaptationandmitigationinsectorsofcriticalimportancefortheclimate:energy,transportation,constructionandbuildingmanagement,agriculture,disasterriskmanagement,andothers.·Maximizetheuseoftraditionalurbanfunctionsforclimateadaptationandmitigation:landuseandtransportationplanning,utilitiesmanagement,urbaninfrastructuredevelopmentandmanage-ment,urbanserviceprovision,anddisas-terresponseplanningandpreparedness.·Engagethelocalcommunityandbusi-nessestodrivebehaviorchangeandmaximizetheircontributionstolocaltargets.(Continued)PoliciesforPromotingGreen,Resilient,andInclusiveUrbanDevelopment299Table5.1continuedNationalgovernmentsCitygovernmentsFinancing·Usingnationalschemesforfi-nancingurbaninfrastructureandsubnationalgrants,incentivizeinvestmentsandprogramsthatmaximizeadaptationandmitiga-tioncontributions.·Wherepossible,createfinancialincentivesforlocalgovernmentstodriveclimateaction,suchasthroughconditionalgrantschemes.·Developanationallegalframeworktoenablelocalgovernmentstocaptureanincreaseoflandvaluearisingfrominvestmentsinclimateadaptation.·Createconditionsandincentivestoattractprivatefinancingofadap-tationandmitigationinvestmentsandprograms.·Useallavailableoptionsformaximiz-ingfinancialresourcesforlocalclimateactionwhileensuringthesustainabil-ityofmunicipaldebtobligations.Thefollowingoptionsshouldbeconsidered(ifappropriate):—Usinglandvaluecaptureschemessuchasdevelopmentfees,infrastruc-turelevies,taxincrementfinancing,andvalue-basedlandtaxation.—Attractingprivatesectorfinancingbyimplementingpublic-privatepartner-shipschemesthroughinsuranceorbyprovidingfinancialincentives.—Whenappropriate,mobilizingfinancingfromcapitalmarketssuchasthroughissuanceofgreenbonds.Capacitybuilding·Setstandardsforlocalclimatestrategiesandpoliciesandprovidesufficienttechnicalandcapacity-buildingsupporttohelplocalgovernmentsachievethem.·Considertheuseofperformance-basedgrantsandasymmetricdecentralizationtoolstoprovideadditionalincentivesforlocalclimateaction.·Buildpublic-privatepartnershipstoleveragethecapacitiesofthelocalprivatesectorandacademiaforadvanc-inglocalclimateaction.·Collaboratewithneighboringjurisdic-tionstocreateconditionsforresidentsandbusinessesacrosstheagglomerationtoadoptmoresustainablepracticesandbehaviorsandpromoteadaptationtoachangingclimate.Collaborationspecifi-callyrelatestotransportationpolicy,landuseandhousingpolicy,utilitiesregulationandmanagement,andbusinesssupportandregulation.Source:WorldBank.a.ThefiveI’sarefivebroadsetsofpolicyinstruments:information,incentives,insurance,integration,andinvestments.TheimportanceoftransboundarycoordinationCoordinatedmetropolitanareagovernancepresentsanopportunity,andachallenge,forimprovinglocalclimateaction.Coordinationacrosslocalgovernmentswithinametroareaiscriticalbecauseeffectiveadaptationandmitigationactiondemandsmoreintegratedplanning,servicedelivery,andpoliciesthanindividuallocalgovernmentscanprovide.Coordinationisalsoimportantbecausethedecisionsmadebyonemunicipalitywilldirectlyaffectitsneigh-borsinanurbanagglomeration(McCarney2010).Withbroadrecognitionoftheimportanceofcoordinatedgovernanceforclimateaction,positiveexamplesareemerging.Forexample,morethantwo-thirdsofmetroareasinmembercountriesoftheOrganisationforEconomicCo-operationandDevelopmenthavemecha-nismsforcoordinatedgovernanceacrossmunicipalboundaries.Sucharrangementsrangefromspecialstatusesgrantedbynationallegislation(forexample,Daejeon,RepublicofKorea,and30metromunicipalitiesinTürkiye)andsupra-municipalauthorities(suchasPortlandMetro,300THRIVINGOregon,UnitedStates),toinformalcoordinationmechanisms(suchasDeltaMetropool,Amsterdam,andTheHague,Rotterdam,andUtrechtintheNetherlands)—seeOECD2015a.InFinland,thecityofTamperehaslaunchedaclimatestrategywithsevensurroundingmunicipalities.Thisstrategycoverscoordinatedactioninareasofpolicysuchaslanduse,trafficmanagement,housing,andmunicipalservices(McCarneyetal.2011).In2007,themet-ropolitanregionofQuitoinEcuador,whichisgovernedbyanelectedmetropolitancouncilandametromayor,adoptedandimplementedaclimatestrategy(Andersson2015).Thisstrategycentersaroundthechallengesofmanagingwaterconsumptioninacityhighlydependentonshrinkingglaciersinthesurroundingmountains(McCarneyetal.2011).Weakcoordinationmechanisms,suchasinmanycitiesinlow-andmiddle-incomecountries,makeithardertotackleissuessuchasurbansprawl.FinancingMunicipalfinancing.Citiestendtorelyontheirownsourcesofrevenue,intergovernmen-talfiscaltransfersandgrants,dedicatedinfrastructure,climatetrustfunds,andborrowingandleveraginginstruments.Ascitycapacityimprovesalongwiththeenhancementoftheoverallenablingenvironment,citiescanadoptmoreadvancedtools.Enablingconditionsarevitalbecausecertainclimatefinanceinstrumentsmaynotbeavailabletocitiesstrugglingwithlowcapacity,lowagency,orhigherpoliticalrisk.TheWorldBank(2021d)suggeststhatperhapsthemostpromisingareaforunlockingresourcesforurbaninvestmentsliesattheintersec-tionofrevenueenhancement,landvaluecapture,andleveraginginstruments.Borrowinganddebtinstrumentscanbeoutofreachforcash-strappedcities,butmechanismsthatcombinerevenuegenerationwithaccesstocapitalprovideopportunitiesforclimatefinancing,suchasthesaleofdevelopmentrightsandtaxincrementfinancing.Leveraginginstrumentsandrevenuegeneration,includingspecialassessmentdistricts,landvaluecapturemechanisms,ortaxincrementschemes,canbeusedtomobilizeadditionalsourcesoffinancesuchasdevel-operequityorin-kindvalue.Inlow-andmiddle-incomecountries,citieswithlesseragencyorweakerenablingconditionsstruggletocollectown-sourcerevenueorcompileinvestmentplans.Thesecitiesshouldfocusonstrengtheningexpenditure-sidesystemsalongsideurbanandcapitalinvestmentplanningsystems,pavingthewayforprogressiveinnovationontherevenuesidethroughaccesstogrants,impactfees,ordevelopment.Althoughcarbonmarketswilllikelybemosteffectiveonanationalorregionalscale,enjoyinggreaterliquidityandcoverageofsectors,carbonpricinginstrumentshavebeengainingtractionatthemunicipallevel.Tocreatetheoptimalenablingconditions,mobiliz-ingurbanclimatefinanceatscalewillrequireintegrationofthelocalandnationallevelsandacrossurbanplanningandthebuiltenvironment,capitalinvestmentplanning,andmunici-palfinance.ThecityofVancouverandtheprovinceofBritishColumbia,inanexampleoftheefficacyofCanada’scarbonpricingsystem,workedverticallyandhorizontallytoshifttaxesawayfromlaborandtowardenvironmentallyharmfulactivities.Workingwitheightcities,China’semissionstradingpilotsaimtocover40percentofnationalemissionsand12percentofglobalemissions.Chinawasthelargestcarbonmarketgloballyin2021(Zhang,Buote,andAcworth2021).Privatefinancing.Privatefinancialflowscancontributeinseveralwaystotacklingclimaterisks—fromportfolioequity,todirectinvestments,tocommercialbanklending,tobondfinance.27Privatecapitalcouldbealignedwithdecarbonizationtargetsornetzerocommit-ments.28Thefinancialregulatoryenvironmentisalsomovingtowardvoluntaryormanda-torydisclosureofclimate-relatedrisks,withcentralbanksandfinancialmarketsupervisorsplayingagrowingrole.Initially,thefocuswasonthedueconsiderationofclimate-relatedrisks;PoliciesforPromotingGreen,Resilient,andInclusiveUrbanDevelopment301however,sincethe2015speechinwhichMarkCarney,governoroftheBankofEngland,issuedawarningaboutclimatechange(“breakthetragedyofthehorizon”),thisfocushasexpandedtoadjustmentofmarketportfoliosinlinewithsustainabledevelopmentgoals.29Subnationalentitiesstillstruggletoaccessfinancefromcapitalmarkets—recentestimatesbyWhiteandWahba(2019)indicatethatlessthan20percentofthelargest500citiesinlow-andmiddle-incomecountriesaredeemedcreditworthy,severelyconstrictingtheircapacitytofinanceinvestments,includingclimate-linkedinvestmentsinpublicinfrastructure.Privatesectorinvestmentinadaptationiscurrentlyalarminglylow,however.TheWorldBank(2021e)findsthat,oftheUS$30billionspentonadaptationin2017–18,onlyroughlyUS$500million—amere1.6percent—camefromtheprivatesector.30Mostofthisspendingtookplaceinhigher-incomecountriesandinsectorssuchaswaterandwastewaterman-agementandenergy.31Inaddition,issueslinkedtoinvestmentrisks—politicalorlegalrisks,currencyrisks,andcreditrisks,amongothers—couldconstrainprivatecapitalflows.Acrucialroleremains,then,forpublicfundsandsupporttohelpmobilizeprivateinvestments.Inthissense,publicandprivateinvestmentsforclimateactioncomplementeachother.Publicfundscouldhelpmobilizeprivatecapitalinseveralwaysby,forexample,co-financingindivid-ualprojects(viagrants,loans,andguarantees)orprovidingfinancialincentives(taxbreaks,subsidies)or,moreindirectly,viabuildingtechnicalcapacitywithintheprivatesectorandcreatingtherightenablingconditions.Publicinvestmentscouldalsoreinforceprivateaction.Forexample,buildingseawallsandsustainingtourismreinforceeachother.Lackofcoordina-tionbetweenthetwo,however,couldleadtoperverseoutcomes—forexample,intheabsenceofpublicsectorincentivesprivate(mal)adaptationcouldoccur.Air-conditioningisaprimeexampleofadaptivecapacitycontributingtofutureeffectsbecauseofenergyuse(DavisandGertler2015).CivilsocietyorganizationsCivilsocietyorganizations(CSOs),includingnongovernmentalorganizations,areoftenatrustedsourceofinformationandcanthusbuildawarenessamongcommunitiesofclimatechangeanditslikelyimpactontheirlives,livelihoods,andhabitats.Manyoftheseorganiza-tionsaresimplyanartifactoflocalcommunitiesorganizingthemselvestotakeonchallengesthataffectselectedgroups,neighborhoods,orsectors(seebox5.5foranexampleofwomen’svoiceininfluencingpolicyinthewatersector).Infact,projectsfinancedbytheWorldBankandotherinternationalorganizationsoftenworkalongsideCSOstoidentifyandmapdataonriskswithinurbanareasandhowtheserisksinteractwithunderlyingvulnerabilityandurbanstresses.TheWorldBankhasthussupportedseveralcash-for-digital-workprogramsthatemployvulnerablegroups,includingyouth,incitiessuchasBamako,Mali(Mukim2018a),andDar-Es-Salaam,Tanzania,tocollectdigitalinformationtomaprisksacrosstheurbanspace(seebox5.6forexamplesofotherparticipatoryresponses).CSOscouldalsoprovidescientificandtechnicalexpertisetounderpinimplementationandmonitoringofexistingclimatepolicyandactasadvocatesofnewlegislationtoprotectnaturalresources.Forexample,CSOsprovidesuchfunctionsincoastalzonesinMexico,ineffectcom-plementingtheroleoflocalgovernments(Baker,Ayala-Orozco,andGarcía-Frapolli2021).Similarly,CSOsinIslamabadpromotegreeninfrastructurethroughofferingtrainingprogramsforthelocalcommunity,holdingdriveswithinschools,andhelpingbuildcapacityinthecityadministrationtoestablishpoliciesandactionplans(Mumtaz2021).Suchorganizationscanalsogoalongwaytowardmitigatingrisksby,forexample,helpingdevelopearlywarningsystemsandcontingencyplans,conductingdrills,andrespondingtodisasters.32302THRIVINGTheimportanceofvoiceindecision-makingWomencanbechangeagentsintheircommunities.Aschapter3demonstrated,women’sneedsandopinionsarecriticalwhenplanningfortheprovisionofwaterservicestowater-scarceinformalurbanareas,inwaterresilienceplanning,andinwaterrecoverymeasures.Yetwomenarestillheavilyunderrepresentedinthewaterworkforceandindecision-makingrolesinwateruserassociationsandlocalgovernments,despitebeingdisproportionatelyaffectedbyinadequatewaterprovision(Adams,Zulu,andOuellette-Kray2020;Das2014).Numerousinternationalandnationalorganizationshaverecognizedtheargumentsforimprovingandincreasingwomen’srolesandhavebeguntomakeconcertedeffortstoincreasewomen’sinfluenceondecision-makinganddiversityinthestaffingofthewaterandsanitationsector.Togoencouragestheinvolvementofwomeninthemanagementofwaterpointsinsemiurbanareasthroughdefinedquotasthatrequirelocalcommitteestohaveatleasttwowomenforeveryfivemembers(GWPandUNEP-DHI2021).InVanuatu,aprovisionintheamendedWaterResourcesManagementActrequires40percentparticipationbywomeninalllocalwatercommittees(GWPandUNEP-DHI2021).Kenyahasemphasizedwomen’srepresentationinwatermanagementorganizations,withrecommendationstomaintaingenderbalanceinwateruserassociationsandcatchmentadvisorycommittees.Nicaraguahaspromotedtheparticipationofwomensince2012aspartofthemultilevelwatergovernancestrategy(GWPandUNEP-DHI2021).Representation,however,isonlythefirststepinmovingtowardamorediversewaterworkforceandbetterwaterresourcemanagement.Onlywhenincreaseddiversitytranslatesintoanabilitytosetagendasandinfluencedecision-makingcanthefullpotentialofrepresentationbeharnessed(Parthasarthy,Rao,andPalaniswamy2019).AnanalysisoftranscriptsfromIndianvillageassembliesrevealedthat,evenwhenrepresentedinlocalgovernance,womenmayparticipateinthedeliberationatlowerratesthanmen(Parthasarathy,Rao,andPalaniswamy2019).Wherecommunitieshaveavailabletranscriptsofdeliberationsinwatercommunitygovernance,thiskindofanalysisusingtext-to-datamethodsmayofferapromisingwaytodetectimbalancesintheparticipationofwomenandmen.Moreover,thisapproachcouldhelpidentifywhatfeaturesofdiscussionpredictwomen’sparticipationandtherebyinforminterventions(BlattnerandKeener2021).Abetterunderstandingofinterventionstoincreasewomen’sparticipationinwatergovernancewillbecomeevenmorecriticalasclimatechangeandgrowingwaterscarcityincreasetheimportanceofsustainablewatermanagementpracticesandthemaximiza-tionofupstreamwaterconservation.Box5.5PoliciesforPromotingGreen,Resilient,andInclusiveUrbanDevelopment303Effectivenessofparticipatoryresponsestoprepareforclimate-relateddisastersTheKenyaFinancingLocallyLedClimateAction(FLLoCA)programisthefirstnational-scalemodelofdevolvedclimatefinance.Theprogramisbasedonthepremisethatlocallyledadaptationcanbemoreeffectivethantop-downinterven-tionsbecauselocalcommunitiesaremoreawareofthecontextandwhatisneededtodrivechange.FLLoCAsupportspartnershipsbetweenlocalgovernmentsandtheircitizenstoassessclimaterisksandidentifysociallyinclusivesolutionstailoredtolocalneeds.WithcommunitiesacrossKenyaalsodealingwiththeimpactsoftheCOVID-19pandemic,FLLoCAtakesabroadviewofresilienceandrecognizesthatcommunitieshaveexperienceinmanagingmultipleriskssimultaneously.Forexample,investmentsmayfocusonactivitiesthatsupportlivelihooddiversificationorcommunity-levelpreparednessformultiplerisks.Dependingonwhatcommuni-tiesprioritize,investmentsmayalsopromotewaterconservationandmoreefficientuseofwater,supportnaturalresourcemanagement,rehabilitatedegradedlands,orpromoteearlywarningsystems.FLLoCAbuildsontheKenyaAccountableDevolutionProgram,apilotprogramfundedbytheWorldBank,inadditiontoCountyClimateChangeFundspilotedbytheAdaptationConsortium.TheCountyClimateChangeFundspilotprojectsfinancedsome100publicgoodinvestmentsprioritizedbythecommunitiesthroughahighlyconsultativeprocess,reachingmorethan500,000beneficiaries—mostofwhomwerewomen—acrossfivecounties(Isiolo,Garissa,Kitui,Makueni,andWajir).Investmentsincludedtherehabilitationofboreholesandinstallationofsolarequip-ment;waterharvesting,storage,anddistributionsystems;sanitationfacilities;andgovernanceactivities.Alarge-scalehouseholdsurveyconductedin2018foundthattheinvestmentsresultedin100percentgreateraccesstowaterforhouseholdsandlivestock.Afollow-upassessmentoftheprogramin2019foundthattheinvestmentshadadditionaldirectandindirectbenefits,includingimprovedlivelihoods,incomes,andfoodsecurity;neweconomicopportunities;andfewerconflictswithinhouseholdsandcommunitiesandbetweenneighboringvillages.Overall,thepilotsledtosig-nificantadaptationbenefitsforhouseholdsandcommunities,whilestrengtheningcountyinstitutionsandimprovingtheresponsivenesstolocalneeds,includingofvulnerableandmarginalizedgroups.ThesuccessofthesepilotsgenerateddemandfromothercountiesandsupportfromKenya’sNationalTreasurytoscaleuptheapproachandmakeitavailabletoall47counties,resultinginthisnewnationallyscaledprogram.Source:ArnoldandSoikan2021.Box5.6304THRIVINGHowtomakechoices?Howdopolicymakerschoosebetweenthedifferentbundlesofpoliciesinawaythatwillproducethegreatestpositiveimpactforthemostpeopleinthemostefficientmanner?TheymusttogglebetweenandsandwichtogetherthebundlesofpolicyinterventionsinthefiveI’stoarriveatgreener,moreresilient,andmoreinclusiveoutcomes.Citiescanthriveintheshadowofanunpredictablychangingclimateonlyifprogramsandpoliciesaimtocombineasmuchaspossibletheobjectivesofcombatingclimatechangeandfurtheringdevelopment,therebyaligningwiththeWorldBank’sGRID(green,resilient,andinclusivedevelopment)approach.Thatapproachaimstopromoteeconomicgrowththatgoeshandinhandwithenvironmentalgoalsandinclusion(WorldBank2021a).HowallcitiesshouldapproachtheGRIDframeworkHowtogreen?Citiesandcountries,nomattertheirlevelofdevelopment,willfacesignificantpathdependenciesintheirinvestments.Evenifreducingemissionsissecondarytofosteringresilienceandinclusion,abusiness-as-usualapproachtogrowthmightentailchoicesthathavehighercostsinthefuture(entailingretrofit-ting,forexample).Pricesplayacentralroleinallocatingresources.Substitutabilitybetweenbuildingup(compacturbandevelopment)andbuildingout(sprawl)willdependontheirrelativeprices(accountingforlanduse,accessibility,andamenities).Misallocationandoveruse(forexample,ofscarceresourcessuchasenergyorwater)canresultwhenpricesarenotattachedtoallocation.And,althoughtheyfalloutsidetheauthorityofmostcities,informationalfailureslinkedtothepriceofcarboncanlimitaccesstosubnationalborrowingforlow-carboninvestments.Howtoincreaseresilience?Somecitieswillbemuchharderhitthanothersbyclimateshocks,butnocitywillremainunaffected.Thus,buildingresiliencetothedirectandindirecteffectsofclimateeventswillbeanimportantconcernformostcities.Plugginginformationgapsisamodestinterventionthatcanhavehighreturnsoninvestmentsandshouldbeunder-takenbyalllocalandnationalgovernments.Riskinformationisapublicgoodthatisvitaltotheefficientfunctioningoflandandurbanhousingmarkets.Moretargetedinterventions,suchasearlywarningsystems,wouldhavethegreatestimpactinlocationsatthehighestriskandwheretheexacttimingofclimateshocksmaybeunpredictable.Easingmovementofpeopleacrossregions,includingacrossborders,wouldalsoallowmoredynamicreallocationofrisksacrossplaces.Howtofurtherinclusion?Astheevidenceinthisreporthasemphasized,muchoftheriskassociatedwithclimateeventswillfallonpoorerplacesandmorevulnerablepopula-tions.Makingthingsworse,theseplacesandpeopleareoftentheonesleastabletomitigateortransfertherisks,andthustheyoftenbearthefullbruntoftheimpacts.Poorplacesandpoorpeoplearealsofoundinrichcountries.Forexample,thepocketsofmarginalizedRomapopulationsindecliningcitiesinBulgariaorRomaniaareoftenthemostaffectedbyenergytransitions.Thus,theprovisionofinsurancetothemostvulnerable,suchasthroughurbansafetynets,shouldbeanalmostuniversalpractice.Betterintegrationwithincities,withbetteraccesstojobsandsocialservices,willalsoprovidelargeco-benefitstopoorerpopula-tions,nomatterwhattypeofcitiestheyresidein.PoliciesforPromotingGreen,Resilient,andInclusiveUrbanDevelopment305TakingcontextualdifferencesintoaccountwhenchoosingpoliciesIntermsofpolicychoices,itmatterswherecitiesandcountriesareintermsofboththeirGRIDchallengesandtheirclimatechange–relatedrisks(seechapter2).Theresidentsofcitiesinthepoorestcountrieswillsufferearliestandthemost.Developingregionsareatageographicdisadvantagebecausetheyarealreadywarmerandsufferfromhighrainfallvariability.Moreover,theirlowincomesandcapabilitiescanmakeadaptationmorediffi-cult.Bothlocalandnationalgovernments,however,cantakeseveralactionsthattaketheseissuesintoaccount.Ingrowth,somelow-andmiddle-incomecountriesarecatchingupbyproducingveryexplosivebutsustainedburstsofprogress.Somecountriescontinuetoexperienceslowergrowththantherichestcountries.Andothershaverecentlytakennosedives.ThelatestIntergovernmentalPanelonClimateChangereportcallsforurgentandradicalactiontoavertthepotentiallycatastrophiceffectsofclimatechange(IPCC2022).Italsoemphasizes,however,thatsuchactioncannotsidesteppovertyreduction,equity,anddevelopment.Strongeremphasisongreening.Inmanycitiesinhigher-incomecountries,firmsimplementedambitiousgreenpoliciesearlyonandtookadvantageoftheeconomicopportunitiesthathavesprungfromclimatepolicies.Theseopportunitiesinmanufacturing(pollutioncontrolequipmentandmachinery,andrenewableenergycomponents),electricitygeneration(renewableenergy),construction(greenbuildings),andservices(tourism,recycling,andpublictransportation)willonlygrow;andmanyemergingeconomiesarenowbankingonthatgrowth.Greeningcitiesbyreducingairpollutionandcreatingmoregreenspace(especiallyimportantinlargercities)willalsohavelargepositivespilloversforlaborproductivityandhealthoutcomes.Greeningcouldalsoreducerisks.Forexample,morepermeablesurfacesandgreenspacescanreduceflooding.Likewise,theanalysispresentedinchapter4suggeststhatpoliciestopromotecompactnessdonotjustmakecitiesgreenerbutalsoboostoveralleconomicactivityandhelpaddressintracityinequalities.Strongeremphasisonresilience.Mitigationinlower-incomecountries(fast-growingornot)willlikelynotbeapriority.Ashighlightedinchapter1,citiesinlow-incomecountriescontributelessthan1percenttoglobalurbanCO2emissions.Meanwhile,Chinese,Indian,andIndonesiancitiesmaynotbekeenondivertingresourcesfromtheirowndevelopmenttoreducethegreenhouseeffectunlesstheyfindthattheco-benefits,suchasproductivitygains,willbesubstantial.Forcitiesinlower-incomecountries,thebestdefenseagainstclimatechangeandvulnerabilitytoweatheringeneralistheirowndevelopmentandinvestmentsinriskreductionandemergencypreparednessmeasures.Furthermore,theirimmediateenvi-ronmentalproblems—airandwaterpollution,poorsanitation,anddisease—demandearlierattention.Thesechallengeswillalsorequirebiginvestmentsinriskreductionandemergencypreparednessandresponse.Strongeremphasisoninclusion.Countriesalreadybearingthebruntofclimatechangeandexpectingtheseimpacts33toincreaseovertimearemorelikelytofocusonadaptation.Incountriesalreadyburdenedwithenvironmentalandsocialchallenges,climateshocksandstressorsareexpectedtofurtherexacerbateexistingvulnerabilities.Brazilian,Sahelian,andSouthAfricancitiesarealreadyseeingsomeoftheeffectsofclimatechangeintheformofincreasesinconflictslinkedtoagriculturalandforestlandsandwater.Insuchplaces,closeattentionwillneedtobepaidtoredistribution,especiallyforpoliciesthatcouldhavearegres-siveeffect(suchaswithdrawalofpublicsubsidiesforfossilfuelsandadaptationinplaceforwealthybeachfrontproperties).306THRIVINGTheimplicationsofcitytypologiesmightmatterExceptforcity-statessuchasSingapore,citiesarenotsmallcountries.Therefore,theydonothaveallthetoolsavailabletocountriestotackleclimate-relatedchallengesorexploittheopportunities.Thediversityofcitiesalsofarexceedsthatofcountries,withevensmallcountriesdisplayingbigdifferencesamongtheirurbanregions.Thus,tobesensible,policyrecommendationsmustcapturemultipledimensionsofcities.Theexamplesthatfollowhighlightsomegroupingsofcities(basedontheglobaltypologyfromchapter2)andthecorrespondingbundlesofinterventionsthatmaybestaddresstheiruniquesetsofchallenges.Therecommendationsareaggregatedintable5.2.Citiesinlow-incomecountriesfacinghighlevelsofrisk.Inthesecities,urgentneedsandactionswouldhavepriority.Lower-costpolicies,includingthosethatcouldbefinancedintandemwithcommunityinvestments,couldbedeployedmoreeasily.Althoughtheexacttimingofclimatechange–relatedshocksmaynotbeeasytopredict,becauseoftheirhighrisksandscarceresources,thesecities,nomattertheirsize,shouldpursueinterventionsaimedatbetterinformation.Suchinterventionscouldincludeputtinginplaceearlywarningandinformationdistributionsystemsfornaturaldisastersandensuringthesesystemsreachpoorhouseholdsinisolatedorremoteareas.Therelationshipbetweenhighriskandlowincomealsopresentsanopportunitytoinvolvevulnerablehouseholdsandcommu-nitiesinassessingandmappingdisasterandenvironmentalrisks.Suchacommunity-level,bottom-upapproachcouldincreaseawarenessandpreparednesswhileprovidingsmallcash-for-worktransferstothepoorest,therebybuildingcommunityresilienceanddevelopment,akeyinsurancemechanismfortheurbanpoor.Inpoorcitiesdealingwithhigh-frequencyandhigh-intensityclimateevents,governments,evenifcash-strapped,canputinplaceincentivestoincreaselocalresilience.Forexample,ensuringthetenuresecurityoftheresidentsofinformalsettlementswouldhelpencouragethemtoinvestintheirproperties(suchasbuildingdwellingswithdurablematerials)ortheirneighborhoods(suchasinvestinginsmall-scaleinfrastructurelikeguttersorpaving).Theseinvestmentsshouldbecombinedwiththeprovisionofpublicassistance(suchascleaningandupgradingdrainagecanals)tohelpbolsterresilience.Community-leddevelopmentcanalsoincreaseconfidenceinpublicofficials,reduceneighborhoodtensions,andputinplacethepreconditionsforenhancedpostdisasterresponses.Formediumandlargecitiesfacinggreaterrisks,theimportanceofaddressingthetraditionalurbanstressesthatgiverisetoslums,gapsinbasicserviceprovision,andcongestiontakesonevengreaterurgency.Thesecitiesshouldfocusonsupportingmoreefficient,higher-densityurbandevelopmentinlessriskyareas.Theycouldachievesuchsupportbyrelyingonacombinationofurbanplanninginstruments,includingchangingbuildingregulationsasneeded,withinvestmentsinresilientinfrastructurethathelpdirectthelocationdecisionsoffirmsandhouseholds.Citiesinlow-incomecountriesfacinglowlevelsofrisk.Becauseoftheirrelativelylowriskprofile,thesecitiesmayneedtoaccountforfastergrowthinthefuture.Theymayhavefewerpressingconcernsandthusalargerwindowinwhichtoplan.Especiallyifmeasurestoensurebetterintegrationatthenationallevelarebeingpursued,citiesthatmightfacelowerrisksofclimateshockscouldexperiencehighermigrationfromruralorotherurbanlocations,includingthosewithahigherriskprofile.Suchcitieswoulddowelltotapopportunitieslinkedwithsuchmigration.Forexample,theycouldensurebalanceddevelopmentofthelabormarketbyfacilitatingmatchingjobswithapplicantsviainforma-tionmechanismssuchasjobsfairsorlocalforums.MoretransparentinformationlinkedtoPoliciesforPromotingGreen,Resilient,andInclusiveUrbanDevelopment307theopportunitiesinurbandestinations,includingthechangingprofileofrisksandtheavail-abilityofpotentialsupport,couldalsohelpdirectmigrationinlinewithpolicymakers’plansforurbandevelopment.Inmanysuchcitiesinlow-incomecountries,deliveryofbasicpublicservicesusuallylagsneeds,andprivateprovidersandactorsfillthegap.Inresponsetothegrowingattractivenessofthecity,thefocusoflocalgovernmentsinsuchcircumstances,lesshinderedbyfrequentclimateshocks,shouldbeonrampingupaccessibilityandthequalityofpublicgoodsandservicesprovision.Thelowerriskprofileofthesecitiesmayalsopresentthemwithalargerwindowofopportu-nitytoplanforfutureurbandevelopment.Despitefewerresources,theycoulddeploypolicyreformsthatsetthestageupfrontforgreener,moreresilient,andmoreinclusivedevelopment.Forexample,therelaxationofbuildingandplanningregulationscouldgivedevelopersincentivestoprovideaffordableunits(particularlyrentalunits)tolow-incomehouseholdsinbetter-servedandlower-riskneighborhoods.Citiescouldalsousetaxreductionsorexemptionstoincentivizesuchdevelopment,althoughthesetoolswouldhavemoreclaimsonthefiscalpurse.Thedevel-opmentofaffordableandresilienttransportation,alongsidecoordinatedinvestmentsinlanduseplanning,wouldhelpenhancetheaccessibilityoflow-incomehouseholdstoeconomicoppor-tunitiesandamenities.Thistypeofplanningwillbeparticularlyimportantforgrowingcities,whetherstartingoutassmall,medium,orlarge.Itwillalsohelpreducethelikelihoodoffutureurbanfragmentation,segregatedcommunities,andinformalsettlements.Citiesinmiddle-tohigh-incomecountriesfacinghighlevelsofrisk.Thesecitieswouldhavemorespacetoputinplacemedium-tolonger-termpolicies.Thefocuscouldbeonpoliciesthatcanhelpcoordinationamongactorsforcommongoals,includingreformsandmarketmechanisms.Citiesthatenjoyhigherlevelsofincomecandeploymoreresourcestotackletheriskstheyface.Theycanwieldadditionalpolicyinstruments,andsomemayhavemorecapablelocalgovernments.Thus,theymightbeinabetterpositiontocoordinateactors.Forexample,becauseofthevariationinriskprofiles(especiallyinlargercities),localgovernmentsinsuchcitiesshouldensurethatpredictedenvironmentalriskisincludedincadastresystemsandpartofthemandatorydisclosureofpropertycharacteristics.Householdsandfirmscouldthenmakemoreinformedlocationandinvestmentdecisions.Atthesametime,developerscouldbechargedforthenegativeexternalitiesgeneratedbybuildinginhigh-riskareas.Financialincentivesandprogramscouldalsohelpsuchcitiesmitigatetheeffectsofdisasters,planforlow-carbondevelopment,andensurethatpoorpeoplearenotleftbehind.Theprovisionofsocialsafetynetprogramsforlow-incomehouseholds,includingcashtransfersandcash-for-workprograms,wouldhelpmitigatetheimpactsofclimateshocksandstrengthenthecity’s“escalatoroutofpoverty”function.Financialassistance,includingsubsidies,couldalsoincentiv-izelow-incomehouseholdstoinvestingreenerandmoreresilientupgradesoftheirproperties.Inparallel,nationalgovernmentsshouldcreatetheconditionsfortheemergenceoffunctionalinsurancemarkets.Governmentinsuranceschemescouldprovidediscountsormakeinsuranceconditionalonactionsintendedtomitigatetheimpactsofextremeclimateevents.Finally,withgreaterfiscalspaceforoutlays,suchcitieswouldalsobewellequippedtocoordinatetheirinvestmentsinlanduseandurbaninfrastructure.Forexample,theycouldsubsidizehousingforlow-incomeresidentsinsaferneighborhoods.Atthesametime,invest-mentsintransportationnetworkstolow-riskareascouldhelpmitigatethetrade-offsthatpoorerresidentsmightmakebetweenriskandaccessibility.Citiesinmiddle-tohigh-incomecountriesfacinglowlevelsofrisk.Thesecitiescouldserveasclimatehavensandhelpstrengthenthestabilityoftheurbansystemwhileexploitingtheopportunitiesthatsuchin-migrationwouldprovide.308THRIVINGCitieswithlowclimateriskandmorefiscalandtechnicalcapabilitiescouldexploitseveralopportunitiespresentedbyachangingclimate.Theywillbemoreattractivetoin-migration,domesticallyandinternationally,andcouldusethistotheiradvantage.Suchcitiescouldsupporttheinclusionofincomingmigrantsby,forexample,rapidlyprovidingidentificationtoensureaccesstoservices(health,education),establishingjobtrainingwithjobfairs,orimprov-ingthequalityofpubliceducation(tofacilitateintergenerationalmobility).Sucheffortswouldallowsmallerordecliningcitieswiththesecharacteristicstoserveasattractivedestinationsforclimatemigrants,therebyattractingtalentandexpandingtheirtaxbases.Manysmaller,single-industrytowns,includingbutnotlimitedtomining,mightalsobeaffectedbecauseofeffortstosupportdecarbonizationandthemovetorenewablesourcesofenergy(box5.7).Countriesmightfindmigrationto“climatehavens”aneffectiveinsur-ancemechanismbyallowingthedynamicspatialreallocationofpeopleandcapitalawayfromriskierplaces,therebyensuringgreaterstabilityoftheurbansystem.Remittancesfromurbantoruralareascouldalsocontributesubstantiallytonationalpovertyreduction.Thus,policiesshouldbedeployedtoreducebarrierstomigration.Thesepoliciescouldincludethosethathelpincreasethesupplyofaffordablehousinginclimatehavens,reducemovingcostsbyimprovingtransportationnetworks,andreduceskillmismatchesbyprovidingjobtrainingforlow-skilledworkers.Forcitiesinlow-incomecountries,especiallysmallandmediumcities,addressingpovertyandbuildingresiliencearekeypolicyprioritiesandinvolveaddressingthechallengeofincreas-ingaccesstobasicservices.Inlow-incomecountries,buildinginstitutionalandindividualcapacityisafundamentalprerequisitetoacceleratingdevelopment.Meanwhile,forcitiesinmiddle-incomecountriesthechallenge,especiallyforsmallandmediumcities,becomesoneAreworkerspreparedforthegreentransition?Thetransitiontogreenereconomicactivitiesandtheuseofgreentechnologieswilldependontheskillsofthepopulation.Thetransitionawayfromcoal,combinedwiththeongoingtechnologicaldisruptionanddigitaltransformation,willhaveaprofoundimpactontheemploymentlandscapeoverthecomingyears.Althoughsomejobswillfaceredundancyandotherswillgrowrapidly,manyexistingjobswillgothroughanimportantchangeintheirrequiredskillsets.Thenewtypesofjobsthatemergewilllikelyrequiredifferenttypesorhigherlevelsofskills.Consequently,withoutadequateskillsdevelopmentmeasuresinplacetomatchlaborsupplyanddemand,unemploymentmayrise,dismissedworkersmayincurincomelosses,andmigrationtrendsmayintensifytowardthelargerandmoredynamiceconomiccenters.Toassessworkers’readinessforthegreentransition,theWorldBankconductedalaborsurveyinmultiplecitiesinBulgaria(WorldBank,forthcomingb).Thesurveyidentifiedandmeasuredemployabilityskills,movementacrossoccupations,andtypol-ogiesofgroupsofworkers,includingacrossminingandothercarbon-intensivedistrictsinthecountry.Theresultsshowthatworkerswithoutspecializededucation,agingworkers,andthoseemployedinelementaryoccupationswereathighriskoflosingtheirlivelihoodswithlittleopportunityfortransition.Asaresult,firmsintransformingsectorswouldsufferfromlonger-termskillshortagesandmaybeconstrainedintheirproductiv-ityandavenuesforgrowth.Box5.7PoliciesforPromotingGreen,Resilient,andInclusiveUrbanDevelopment309ofinequalityratherthanofpovertyperse.Thisglaringcommonalitypersistsincitiesofallsizesinhigh-incomecountries.Acrossthespectrum,citiesshouldprioritizeembeddingsocialprotectionpoliciesandprogramstodiminishrisk.Thedifferencesinchallengesacrosstypesofcitiesalsosuggestthatthefocusforlow-andmiddle-incomecountrycitiesshouldbeonadaptation,whereasforcitiesinhigh-incomecountriesitmustbeonadaptationandmitigation.Pollution,however,isakeyissueforlargecitiesinbothlow-andmiddle-incomecountries,andaddressingitcomeswithessentialclimatechangemitigationco-benefits.Addressinggreenhousegas(GHG)emissionsshouldbeapriorityforlargecitiesinmiddle-incomecountriesandforallcitiesinhigh-incomecountries.Inlow-andmiddle-incomecountries,thepolicydiscourseshouldbeonissuesofairpollution,whichthenhavetheaddedbenefitofalsoreducingGHGemissions.Accountingforco-benefitsandtrade-offsMakingchoicesacrossbundlesofinterventionscanbeevenmorecomplicatedbecausepolicymakersneedtoconsidersynergies(orco-benefits),trade-offs,andinteractionsbetweenmultipleobjectivestheymayotherwiseoverlook.Suchconsiderationsarenotnew—urbanpolicieshavealwayshadmultiplegoals,spanningsocialobjectives,economiccompetitive-ness,andenvironmentalgoals.Achangingclimatehassimplysuperchargedthechallenges,addingadaptationandmitigationtothemix—challengesthatexacerbatetheunderlyingones(asdescribedintheanalyticalframeworkforthisreport).Adoptingaco-benefitlenswouldhelpensurethatpolicychoiceswouldaimtodeliversimul-taneouslyonmultiplewell-beingobjectives,includingclimate.Doingsowouldrequireaneconomywideperspectiveratherthanafocusonasingleornarrowrangeofoutput-relatedobjectives,independentofothers.Sometimes,thebenefitsaccruetothesameorasimilarsetofstakeholders.Forexample,manyinvestmentsinclimaterisk-reductionstrategiescanalsosupporteconomicdevelopment,oftenwithinthesamecommunities.ExaminingfloodmitigationinvestmentsinBuenosAires,Avneretal.(2022)findthat,whenlandmarketsarefunctional,riskreductioncanbecapturedthroughlandvalueappreciation.Inothercases,thebenefitscouldspilloveracrossmultiplegroupsandgeographies.Forexample,investmentsinpublictransitcanreducecongestion,helptackleairpollution,andthuscombineimprove-mentsinproductivitywithwidespreadbenefitsforhealth.Likewise,investmentsinrenewableenergycanreduceemissionsandimproveenergysecurity.Thenatureofclimateriskalsoinvolvesrisksfromresponsesthemselves.Policyresponsestoclimatechangecouldentailtheirownopportunitycosts,presentingthemselvesastrade-offsbetweenGRIDoutcomes.Forexample,retrofittingbuildingstoimproveenergyefficiencyincreasescostsforhouseholdsandthreatenshousingaffordability,potentiallyreducinginclusion.AvnerandHallegatte(2019)discussthepotentialtrade-offsbetweenflooddamageandhousingscarcityandhowtheydifferonthebasisoffloodmanagementpolicies.ViguiéandHallegatte(2012)providesomeearlyquantificationofthetrade-offsandsynergiesofselectedurbanclimatepolicies.UsingspecificsetsofpolicypackagesforParis,theyquantifytheinteractionofinterventionsaimedatmakingthecitygreener,moreresilient,andmoreinclusive.Forexample,agreenbeltpolicythatlimitsurbansprawlandprotectsnaturalareascouldincreaseriskasmorepeoplemovetoflood-proneareasbecauseofthegreaterscarcityofland.Theirmodelssuggestthatacarefulmixofseveralpoliciescouldmitigatetheadverseconsequencesofeachpolicy.Infact,forurgentactions,policymakersshouldexploreoptionsthatseektomanage,minimize,orreversethetrade-offsforthoseplacesandpeoplemostaffected.310THRIVINGTable5.2Tailoredpolicyoptions,bytypeofcityandinstrumentIncomeclassTypeofcityLow-incomeMiddle-incomeHigh-incomeSmallMediumLargeSmallMediumLargeSmallMediumLargeChallenges•Resilience(S)•Poverty(S)•Basicservices(S)•Resilience(S)•Poverty(S)•Basicservices(S)•Resilience(S)•Pollution(S)•Basicservices(S)•Poverty(S)•Inequality(S)•Vegetation(S)•Inequality(S)•GHG(S)•Pollution(S)•Vegetation(S)•Inequality(S)•GHG(M)•Inequality(M)•Resilience(M)•GHG(M)•Inequality(M)•Vegetation(M)•Resilience(M)•GHG(S)•Inequality(S)InstrumentInformationPolicyoptionsEarlywarningsystems;hazardmappingandassessment•Buildinstitutionalcapacity•Decentralizedlandadministrationservices•Participatoryriskawareness•Jobfairsandlocalforums•Inclusionaryzoning•Urbanplanningdocuments•Urbandesignguidelines•GHGemissionsinventories•Participatoryriskawareness•Jobfairsandlocalforums•Pollutionmonitoring•Betterzoningofpollutingindustries•Urbanplanningdocuments•Pollutionmonitoring•Betterzoningofpollutingindustries•Urbanplanningdocuments•Urbandesignguidelines•Buildingcodes•Urbanplanningdocuments•Urbandesignguidelines•Buildingcodes•Disasterrisk–informedlandvalue•Urbanplanningdocuments•Urbandesignguidelines•Buildingcodes•Disasterrisk–informedlandvalue•Disaster-risklanddevelopmentpenaltyIncentivesPolicyoptionsPhaseoutfossilfuelsubsidies•Cashtransfers•Workfareprograms•Subsidizedhousing•Congestioncontrolschemes•Parkingcharges•Reformstolowercostsofverticalconstruction;relaxedheightrestrictions•Cashtransfers•Workfareprograms•Subsidizedhousing•Inclusionaryzoning•Subsidizedhousing•Inclusionaryzoning•Carbontaxes•Congestionpricing•Parkingreform•Lowercostsofverticalconstruction;relaxedheightrestrictions•Densitybonus•Expeditedpermitting•Buildingretrofitandcleanenergysubsidiesandtaxcredits•EVtaxcredit•Inclusionaryzoning•Densitybonus•Performancezoning•Expeditedpermitting•Buildingretrofitandcleanenergysubsidiesandtaxcredits•EVtaxcredit•Inclusionaryzoning•Congestionpricing•Parkingreform•Lowercostsofverticalconstruction;relaxedheightrestrictions•Densitybonus•Performancezoning•Expeditedpermittingandfast-trackprojectreview•Retrofitincentives•Airrightsprograms•Inclusionaryzoning•Linkagefees(Continued)PoliciesforPromotingGreen,Resilient,andInclusiveUrbanDevelopment311Table5.2(Continued)IncomeclassTypeofcityLow-incomeMiddle-incomeHigh-incomeSmallMediumLargeSmallMediumLargeSmallMediumLargeInstrumentInsurancePolicyoptions•Socialprotection•Subsidizedinsurance(low-riskareas)•Catastropheinsurance•Incorporateclimateriskconsiderationsinasset(re-)pricing,newinsuranceproductlaunches,andunderwritingprocessIntegrationPolicyoptionsIntegrateclimatechangeadaptationandurbanmanagement;urbanplanningandregulation•Basicservices;education•Layingoutstreetnetworks•Flexibleurbanplanning•Compactgrowth•Connecttomediumandlargercities•Lowermigrationbarriers•Layingoutofstreetnetworkinanticipationoffutureexpansion•Securelandandpropertyrights•Integratedlanduseandtransportationplanning•Connecttomediumandlargecities•Lowermigrationbarriers•Layingoutofstreetnetworkinanticipationoffutureexpansion•Integratedlanduseandtransportationplanning•Transit-orienteddevelopment•Connecttomediumandlargecities•Lowermigrationbarriers•Transit-orienteddevelopment•Transit-orienteddevelopmentPolicyoptionsInvestments•Localbusservices•Well-locatedaffordablehousing•Landprovision•Improvebuildingstock•Climateadaptationinfrastructure•Nature-basedsolutions•Renewableenergy•Masstransit(BRT,MRT)•Well-locatedaffordablehousing•Climateadaptationinfrastructure•Nature-basedsolutions•Renewableenergy•Energy-efficientretrofits•Localbusservices•Well-locatedaffordablehousing•Masstransit(BRT)•Well-locatedaffordablehousing•Urbangreenspace•Masstransit(BRT,LRT)•Mobility•Well-locatedaffordablehousing•Urbangreenspace•Renewableenergy•Energy-efficientretrofits•Localbusservices•Mobility•Well-locatedaffordablehousing•Climateadaptationinfrastructure•Nature-basedsolutions•Renewableenergy•Energy-efficientretrofits•Masstransit(BRT,LRT)•Mobility•Well-locatedaffordablehousing•Climateadaptationinfrastructure•Nature-basedsolutions•Urbangreenspace•Renewableenergy•Energy-efficientretrofits•Masstransit(BRT,MRT)•Mobility•Well-locatedaffordablehousing•RenewableenergyPolicyactionsappliedto:AllcitiesCitiesinmiddle-orhigh-incomecountriesCitiesinlow-incomecountriesOneortwotypesofcitiesexceptforinsuranceSource:WorldBank.Note:BRT=busrapidtransit;EV=electricvehicle;GHG=greenhousegas;LRT=lightrailtransit;M=Moderatechallenge;MRT=massrapidtransit;S=severechallenge.THRIVINGTrade-offsacrossplacesAsdiscussedinchapter3,climatechangeisexpectedtohaveheterogeneouslocaleffects.Usingadynamicspatialequilibriummodel,CruzandRossi-Hansberg(2021)findhighlyasym-metriceffectsofa1°Cincreaseinlocaltemperature,withamenitiesandproductivitydecliningintheworld’shottestareasandincreasinginthecoldest.Balboni(2021)alsodemonstrateshowthebenefitsofinvestinginselectedregionsinVietnam—regionssubjecttothegrowingriskofcoastalfloods—diminishesrapidlywithrisingsealevel.AlthoughthemodelusedbyCruzandRossi-Hansberg(2021)isonaglobalscaleandtheoneusedbyBalboni(2021)isonacountryscale,themodelshavetwocharacteristicsincommon.First,theydemonstratethattheeffectsofclimatechangewillbehighlyasymmetricacrossspace,globallyandwithincoun-tries.Second,thefuturereturnsoncurrentinvestmentsinmore(negatively)affectedregionsarelowercomparedwithreturnsinotherregions.Forgovernments,thesefindingsimplymakingseriousspatialchoicesintermsofwheretoinvest,which,inturn,impliesmakingtrade-offsacrossplaces.Placesandecologicalsystemsdonothaveaninexhaustibleabilitytoadapttoclimatichazards.Withrapidlyshiftingclimateeffects,risksandlossesthatmayhavebeenaccept-ablecouldbecomeintolerable.Atippingpointmay,then,ariseatwhichagoodinvestmentbecomesawastefulone.Withoutforward-lookingpolicyandplanning,andintheabsenceofcrediblechoicesorlong-termpathwaystoviablealternatives,thereisariskthatpeoplewillremaininplaceswithdeterioratingconditions.Forexample,about20millionpeopleincoastalBangladeshalreadysufferthehealtheffectsofsaltwaterintrusionintodrinkingwatersuppliesrelatedtotheriseinsealevel.Remittancesfromfamilymembersworkingelsewherecaninducepeopleintheseareastostay,possiblyagainsttheirbestinterests.Seawallscanreduceimpactseffectivelyintheshortterm,buttheycanalsoresultinlock-insandincreasedexposuretoclimaterisks.Withoutappropriatepolicyinterventions,perverseincentivestostayinplacecouldgreatlyunderminecommunityhealthandwell-being.Nevertheless,standalonepolicieswillhavelimitedpoliticalacceptability.Theexistenceoflocalnetworks,culturalpreferences,andothersocioeconomicfactorsmayalsoaffecttheincentivetomoveawayfromriskierlocations.34Thus,policiestolimitfuturedevelopmentincertainareasandpotentialresettlementawaymayalsoneedtobepairedwithcertainelementsofadaptation.Trade-offsacrossgroupsFinally,climatepolicieswillinvolvetrade-offsamonggroupsofpeople.EvenbenefitsassociatedwiththereductionofGHGemissionswillaccruedifferentiallyacrosscountriesandpopulations.Someclimatepolicies,however,canhaveregressive,albeitoftenunin-tended,effects.Thesepoliciescouldincludecarbontaxes,certainmandatorystandards,subsidies,andregulatorytools,andmightthenrequirefurthercorrectiveinterventions.Forexample,Känzig(2021)findsthattheEuropeanUnionemissionstradingsystemshastheintendedeffectofleadingtoapersistentfallinoverallGHGemissionsbymeansofastrongandimmediateincreaseinenergyprices,butthatfallcomesatthecostofatemporarydropineconomicactivitythatisborneunequally.Thepaperfindsthat,althoughtheexpenditureofhigher-incomehouseholdsfallsonlymarginally,low-incomehouseholdsreducetheirexpen-dituresignificantlyandpersistently.35Meanwhile,thewidespreaduseofsubsidiestoachievecleanenergy,suchasforelectricvehiclesorrooftopsolar,canbenefitricherhouseholdsdisproportionately,atleastintheshortrunbecauseofhighertake-up(BorensteinandDavis2016).Vona(2021)providesacomprehensiveoverviewofthedistributionalimpactsofvariousclimateandenvironmentalpolicies.312PoliciesforPromotingGreen,Resilient,andInclusiveUrbanDevelopment313Thepotentiallyadversedistributionalimpactsofclimatepoliciessuggestanimportantrolefortargetedfiscalpoliciestoreducetheeconomiccosts(thatis,toshifttheireconomicburdenacrosspopulationgroups).Policymakerscoulddomore.Policymeasurescouldalsobedesignedtoreducetheadverseimpactsortoexpandthefeaturesthathaveprogressiveimpacts.Forexample,Zachmann,Fredriksson,andClaeys(2018)suggestthatcertainpoliciesarelessregressivethanothers(suchasfueltaxescomparedwithfuelefficiencystandardsorsubsidiesforpublictransportationversussubsidiesfordomesticair-conditioningsystems)andthatcertaindesignelementscouldmakepolicieslessregressive(forexample,auction-ingemissionpermitsinsteadofgrandfatheringthemtopolluters).Somepolicies,suchasenergyretrofitsforsocialhousingunits,alsoactivelybenefitlow-incomehouseholds.Targetedcompensationandwell-designedclimatepolicieswithprogressiveimpactswouldalsohelpincreasepublicsupport.SummaryandconclusionsThechallengesbroughtonbyachangingclimatemayseemtoointractableafterreadingthisreport.Thischapterhaslookedatwhatcanbedone,bywhomandhow,andhasdemonstratedhowstridescanbemadebeyondtimid,temporizingpolicies.Tacklingthemyriadchallenges,andinsomecasesopportunities,associatedwithclimatechangewillrequiremakinginformedandastutechoices.Thischapterpresentsasequencedsuiteofpolicyinstruments,thefiveI’s—information,incen-tives,insurance,integration,andinvestments—atthedisposalofpolicymakers.Informationhelpspeopleandfirmsbetterunderstand,andthereforebetteradaptto,climaterisksbothacrossandwithincities.Incentivesallowpeopleandfirmstointernalizeenvironmentalexter-nalitiesandgovernmentofficialstoworkbettertoaddressGRIDchallenges.Insuranceallowspeople,firms,andgovernmentstoinsureagainstlossesassociatedwithclimatechangeandunavoidableenvironmentalshocksandstressesthatcannotbeavoided.Integrationallowsmoremigrationandtrade.And,finally,investmentsareaimedatfinancinggreen,resilient,andinclusiveinfrastructure,includingnature-basedsolutions.Underpinninganypolicyresponseistheroleofinstitutions.ThefiveI’scanbeturnedintoactionableprescriptionsonlybythosewhogetthingsdone.Localgovernmentscoulddosobyensuringbalancedgrowthoftheirscopeandcapacityforadaptationandmitigation.Nationalgovernmentscouldprovidestrategicoversight,facilitateaccesstoclimatefinance,anddriveclimateactionbycreatinganenablingenvironment.Meanwhile,privatefinancialflows—fromportfolioequitytodirectinvestments,tocommercialbanklending,tobondfinance—couldcontributetotacklingclimaterisks.Finally,civilsocietygroupsshouldnotbeoverlooked.Theyareoftenatrustedsourceofinformationandcanbuildawareness,providetechnicalexpertisetounderpinimplementationandmonitoringofexistingclimatepolicy,andactasadvocatesfornewlegislation.PolicymakerswillneedtomovebetweenandpulltogetherthebundlesofpolicyinterventionspresentedinthefiveI’stoarriveattheGRIDoutcomes.Thecombinationofinterventions,theirsequencing,andtheprioritizationofoutcomeswilldependonthecharacteristicsofcities—includingprimarilytheirlevelofrisk,levelofdevelopment,andsize.AllaspectsoftheGRIDframeworkwillapplytocities,nomattertheircharacteristics,buttheemphasiswilldependstronglyonthecontextualdifferences.Thisiswheretheworkoncitytypologies(aspresentedinchapter2)willhelpclarifythepressuresonagivencity,therangeofpolicyoptionsavailabletoit,anditsabilitytodeployitsoptionsovertime.Themixofco-benefitsandtrade-offswillalsoguidethedecision-makingprocesstoensurethatscarceresourcescanhelpdeliversimultaneouslyonmultiplewell-beingobjectives,includingclimateaction.314THRIVINGTHRIVING314Notes1.AvailableonClimateCentral’swebsite,https://www.climatecentral.org.2.TheWorldBankdefinessharedprosperityusingtheannualizedgrowthrateoftheaveragepercapitaconsumptionorincomeofthepoorest40percent(thebottom40)ofthepopulationofacountry.3.Scientificinstitutionsandinsuranceagenciesoftenhavethemostdetailedandaccurateriskinformationandsophisticatedmodeling,yetlocalgovernments,communities,andbusinesses—whoneedthatinformationthemost—lackequivalentaccessandcapabilities.4.WorldBankresearchinBulgariaprovidedaspartofReimbursableAdvisoryServices.5.SeetheoverviewoftheWorldBank’sworkonadaptivesafetynetsinAfricabyBaez,Kshirsagar,andSkoufias(2020).6.SeetheWorldBank’sdevelopmentresponsetotheDisplacementImpactProjectinUganda(Mahony,Maher,andHaile2021).7.Suchsystemsprovidehazardwarningsofdifferenttypes(rapidorslowonset)andcanvaryintermsofthespatialscale(local,regional,national,orglobal)andthestakeholders(publicauthorities,media,communities,andsoon).8.ThosestudiesincludeBin,Kruse,andLandry(2008);OrtegaandTas.pınar(2018);andZhangandLeonard(2019).Likewise,detailedempiricalworkinBogotábyWorldBank(2010)showsthatpropertyvaluescapitalizedtheexposuretoseismicandotherhazardrisks,withlowervaluesinriskierareas.9.Anoft-quotedexampleisthegreenbondissuedbythecityofJohannesburg,SouthAfrica,in2014tointroducecleanbusesintoitsfleet.10.TheSternReviewarguesthatclimateisthe“greatestmarketfailure,”andmitigationrequiresplacingapriceoncarbonsothatthemarkethasanincentivetoshifttoalow-carboneconomy(Sternetal.2006).11.Formoreinformation,seetheGovernmentofBritishColumbiawebpage,“ClimateActionRevenueIncentiveProgram,”https://www2.gov.bc.ca/gov/content/governments/local-governments/grants-transfers/climate-action-revenue-incentive-program-carip.12.FromtheEco-RoofIncentiveProgramwebpage,https://www.toronto.ca/services-payments/water-environment/environmental-grants-incentives/green-your-roof.13.However,aFederalEmergencyManagementAgencyprogram,BuildingResilientInfrastructureandCommunities,nowhelpsUScitiesbuyoutriskypropertiesafteradisaster,butitalsorequiresaplaceforthoseresidentstomovesothattheydonotendupinanotherhighlydisaster-pronehome.14.Forarecentoverviewoftheliteratureonhowtenuresecurityaffectsinvestmentsinthephysicalresilienceofhomes,seeRentschler(2013).PoliciesforPromotingGreen,Resilient,andInclusiveUrbanDevelopment315315PoliciesforPromotingGreen,Resilient,andInclusiveUrbanDevelopment15.Thesetoolsusegeospatialdatatoidentifyandassessexistingassets,concentrationofpopulation,locationofcriticalinfrastructure,andfutureareasofplannedgrowthvis-à-vishistoricalhazarddata(suchasfloodandseismicrisk)andclimateprojections.16.FromtheGranthamResearchInstituteonClimateChangeandtheEnvironment’s“ClimateChangeLawsoftheWorld”webpageonColombia,https://climate-laws.org/geographies/colombia/laws/law-no-1931-establishing-guidelines-for-the-management-of-climate-change.17.Thereportestimatesthatinternationalmigrantstotaledabout281millionin2020,or3.6percentoftheglobalpopulation.18.UnitedNationsHighCommissionerforRefugees,“FiguresataGlance”webpage,https://www.unhcr.org/figures-at-a-glance.html.19.Hallegatte,Rentschler,andRozenberg(2019)findthatnaturalshocksareamongtheleadingcausesofinfrastructuredisruptionsandcancostbetweenUS$391billiontoUS$647billioninlow-andmiddle-incomecountries.20.ForAfricancities,FosterandBriceno-Garmendia(2010)estimatethatdoublingurbandensityreducesthepercapitacostofapackageofinfrastructureimprovementsby25percentorso.21.Somesimplesolutionsexist,suchastheroofstrapsusedintheCaribbeansoroofscanwithstandhurricane-forcestorms(Gibbs2000).22.Grayinfrastructurereferstobuiltstructuresandengineeringequipment(suchasreservoirs,embankments,canals,andsoon)embeddedwithinwatershedsorcoastalecosystems.23.Thiseffecthasbeenobservedinlow-incomecountries—seeFerdousetal.(2019)whostudytheeffectofinvestmentsinfloodprotectionalongtheJamunaRiverinBangladesh.Itsuggeststhatprotectivemeasuresmayneedtobecomplementedwithstrictzoningrestrictions,withbetterunderstandingoftheexpectedriskexposure.24.SeetheCovenantofMayorsforClimateandEnergy’sUrbanAdaptationSupportTool,step0-3,“GettingStarted:AdaptationtoClimateChangeinUrbanAreas,”https://climate-adapt.eea:europa.eu/knowledge/tools/urban-ast/step-0-3.25.SeetheCovenantofMayorsforClimateandEnergy’swebpage“CovenantInitiative:CovenantinFigures,”https://www.covenantofmayors.eu/about/covenant-initiative/covenant-in-figures.html.26.Butcityleadersandgovernmentscanmanagedevelopmenteffectivelyonlyiftheyhavethefunctionalmandate,revenuebase,andcapabilitiestotargetsuchdevelopment.27.Privatefinanciersincludebanks,pensionfunds,insurancecompanies,corporations,impactinvestors,andotherprivateactors.Theprivatesectorcouldalsocontributebyprovidinggoodsandservicesthatfacilitateadaptationormitigationandbyadaptingtheirownoperationsandassetstobeclimate-resilient.28.SeetheNetZeroAssetManagersinitiative,https://www.netzeroassetmanagers.org/.29.Thespeecharguesthat,justasthesolutiontothetragedyofthecommons(aclassicprobleminenvironmentaleconomics)liesinpropertyrightsandsupplymanagement,so,too,mustcentralbankstaketheleadtocombinedata,technology,andexpertjudgmenttomeasureandmanagetherisks.316THRIVINGTHRIVING31630.Toputthisnumberinperspective,theUnitedNationsEnvironmentProgrammeestimatesthatthetotalcostofadaptationwillreachUS$140billion–$300billionayearby2030(UNEP2021).Thus,actualspendingstillfallsshortofdocumentedneeds.31.Becauseadaptationspendingisoftenpartoflargerinvestmentsandbecauseofissuesoflimitedtransparency,quantifyingthecurrentlevelsofprivateinvestmentinadaptationisnotastraightforwardexercise.32.SeeAmaoetal.(2014)forexamplesfromurbanareasinGhanaandKenya.33.However,theimpactsofdisastersareoftenmeasuredasthecostofthedamagesandlosses,which,accordingtoHallegatteandWalsh(2021),doesnotproperlyreflecttherealimpactonlow-incomecommunities.Instead,HallegatteandWalshproposeanewapproachthatmeasurestheimpactofadisasteratthehouseholdlevel,whichwouldtakeintoaccountdistributionalandpovertyimpacts.Thisapproachcouldallowpolicymakerstobetteridentifywhereandinwhichsectorstoprioritizeinvestments.34.Henriqueetal.(2022)provideanexampleofhowresidentsinSouthwestAustraliatradeoffclimate-inducedlossesagainstothervalues.35.Twomechanismsaccountforthiseffect.Poorerhouseholdsspendalargershareoftheirdisposableincomeonenergy,leavinglessforotherexpenditures.Andpoorerhouseholdsseeasteeperdropinincomesbecausetheytendtoworkinsectorsthataremoreaffectedbyaclimatepolicy.ReferencesAdams,E.A.,L.Zulu,andQ.Ouellette-Kray.2020.“CommunityWaterGovernanceforUrbanWaterSecurityintheGlobalSouth:Status,Lessons,andProspects.”WIREsWater7(5):e1466.Afet,D.,andS.Kurumu.2021.“FaaliyetRaporu.”https://www.dask.gov.tr/upload/Dask/FAALİYET%20RAPORLARI/2021_Faaliyet_Raporu.pdf.Amao,O.B.,D.Ettang,U.Okeke-Uzodike,andC.Tugizamana.2014.“RevisitingtheUtilityoftheEarlyWarningandEarlyResponseMechanismsinAfrica:AnyRoleforCivilSociety?”PeaceandConflictReview8.1:77–97.Andersson,M.2015.UnpackingMetropolitanGovernanceforSustainableDevelopment.Nairobi:UN-Habitat.Andres,L.A.,M.Thibert,C.LombanaCordoba,A.V.Danilenko,G.Joseph,andC.Borja-Vega.2019.DoingMorewithLess:SmarterSubsidiesforWaterSupplyandSanitation.Washington,DC:WorldBank.Angel,S.2017.“UrbanFormsandFutureCities:ACommentary.”UrbanPlanning2(1):1–5.Arnold,M.,andN.Soikan.2021.“KenyaMovestoLocallyLedClimateAction.”WorldBankBlogs,October27,2021.https://blogs.worldbank.org/nasikiliza/kenya-moves-locally-led-climate-action.Avner,P.,andS.Hallegatte.2019.“MoralHazardvs.LandScarcity:FloodManagementPoliciesfortheRealWorld.”PolicyResearchWorkingPaper9012,WorldBank,Washington,DC.PoliciesforPromotingGreen,Resilient,andInclusiveUrbanDevelopment317317PoliciesforPromotingGreen,Resilient,andInclusiveUrbanDevelopmentAvner,P.,andS.V.Lall.2016.“MatchmakinginNairobi:TheRoleofLandUse.”PolicyResearchWorkingPaper7904,WorldBank,Washington,DC.Avner,P.,J.E.Rentschler,andS.Hallegatte.2014.“CarbonPriceEfficiency:Lock-InandPathDependenceinUrbanFormsandTransportInfrastructure.”PolicyResearchWorkingPaper6941,WorldBank,Washington,DC.Avner,P.,V.Viguié,B.A.Jafino,andS.Hallegatte.2022.“FloodProtectionandLandValueCreation:NotAllResilienceInvestmentsAreCreatedEqual.”EconomicsofDisastersandClimateChange6:417–49.https://doi.org/10.1007/s41885-022-00117-7.Baez,J.E.,V.Kshirsagar,andE.Skoufias.2020.“AdaptiveSafetyNetsforRuralAfrica:Drought-SensitiveTargetingwithSparseData.”PovertyandEquityNotes,WorldBank,Washington,DC.Baker,I.,A.Peterson,G.Brown,andC.McAlpine.2012.“LocalGovernmentResponsetotheImpactsofClimateChange:AnEvaluationofLocalClimateAdaptationPlans.”LandscapeandUrbanPlanning107(2):127–36.Baker,S.,B.Ayala-Orozco,andE.García-Frapolli.2021.“TheRoleofCivilSocietyOrganisationsinClimateChangeGovernance:LessonsfromQuintanaRoo,Mexico.”JournaloftheBritishAcademy9(s10):99–126.Balboni,C.2021.“InHarm’sWay?InfrastructureInvestmentsandthePersistenceofCoastalCities.”DepartmentofEconomics,MassachusettsInstituteofTechnology,Cambridge,MA.Balch,O.2014.“PluggingtheLeaks:HowDigitalToolsCanPreventWaterLoss.”TheGuardian,December11,2014.https://www.theguardian.com/sustainable-business/2014/dec/11/plugging-leaks-digital-tools--water-loss-leaks.Beltrán,A.,D.Maddison,andR.Elliott.2018.“IsFloodRiskCapitalisedintoPropertyValues?”EcologicalEconomics146:668–85.Bin,O.,J.B.Kruse,andC.E.Landry.2008.“FloodHazards,InsuranceRates,andAmenities:EvidenceFromtheCoastalHousingMarket.”JournalofRiskandInsurance75(1):63–82.Bin,O.,andC.E.Landry.2013.“ChangesinImplicitFloodRiskPremiums:EmpiricalEvidencefromtheHousingMarket.”JournalofEnvironmentalEconomicsandManagement65(3):361–76.Blanchard,J.2021.“LocalGovernmentsCanUseTheirPowertoCombatClimateChange.”BloombergLaw,June3,2021.https://news.bloomberglaw.com/environment-and-energy/local-governments-can-use-their-power-to-combat-climate-change-17.Blattner,A.,andS.Keener.2021.“WhatWorks?AssessingtheEvidenceBaseonGenderandInclusion.”ResearchNotefortheWaterGlobalPractice,WorldBank,Washington,DC.Blickle,K.,andJ.A.C.Santos.2022.“UnintendedConsequencesof‘Mandatory’FloodInsurance.”FRBofNewYorkStaffReportNo.1012,FederalReserveBank,NewYork.Borenstein,S.,andL.Davis.2016.“TheDistributionalEffectsofUSCleanEnergyTaxCredits.”TaxPolicyandtheEconomy30(1)191–234.Borgomeo,E.,A.Jägerskog,E.Zaveri,J.Russ,A.Khan,andR.Damania.2021.EbbandFlow,Volume2:WaterintheShadowofConflictintheMiddleEastandNorthAfrica.Washington,DC:WorldBank.Brenton,P.,andV.Chemutai.2021.TheTradeandClimateChangeNexus:TheUrgencyandOpportunitiesforDevelopingCountries.Washington,DC:WorldBank.318THRIVINGTHRIVING318Buckley,B.M.,K.J.Anchukaitis,D.Penny,R.Fletcher,E.R.Cook,M.Sano,L.C.Nam,etal.2010.“ClimateasaContributingFactorintheDemiseofAngkor,Cambodia.”PNAS107(15):6748–52.Burch,S.,M.Andrachuk,D.Carey,N.Frantzeskaki,H.Schroeder,N.Mischkowski,andD.Loorbach.2016.“GoverningandAcceleratingTransformativeEntrepreneurship:ExploringthePotentialforSmallBusinessInnovationonUrbanSustainabilityTransitions.”CurrentOpinioninEnvironmentalSustainability22:26–32.Carter,J.,G.Cavan,A.Connelly,S.Guy,J.Handley,andA.Kazmierczak.2015.“ClimateChangeandtheCity:BuildingCapacityforUrbanAdaptation.”ProgressinPlanning95:1-66.Chan,F.K.,J.Griffiths,E.Higgitt,S.Xu,F.Zhu,Y.T.Tang,Y.Xu,andC.Thorne.2018.“‘SpongeCity’inChina:ABreakthroughofPlanningandFloodRiskManagementintheUrbanContext.”LandUsePolicy76:772–78.CityofRotterdam.2008.RotterdamClimateProof:AdaptationProgrammed.RotterdamClimateInitiative.Conte,B.,K.Desmet,D.Nagy,andE.Rossi-Hansberg.2021.“LocalSectoralSpecializationinaWarmingWorld.”JournalofEconomicGeography21(4):493–530.Costa,D.L.,andM.E.Kahn.2011.“ElectricityConsumptionandDurableHousing:UnderstandingCohortEffects.”AmericanEconomicReview101(3):88–92.Costa,D.L.,andM.E.Kahn.2013.“EnergyConservation‘Nudges’andEnvironmentalistIdeology:EvidencefromaRandomizedResidentialElectricityFieldExperiment.”JournaloftheEuropeanEconomicAssociation11(3):680–702.Costella,C.,A.McCord,M.vanAalst,R.Holmes,J.Ammoun,andV.Barca.2021.“SocialProtectionandClimateChange:ScalingUpAmbition.”SocialProtectionApproachestoCOVID-19ExpertAdviceService(SPACE),DevelopmentAlternativesIncorporated(DAI)Global,LLC,Bethesda,MD.Cruz,J-L.,andE.Rossi-Hansberg.2021.“TheEconomicGeographyofGlobalWarming.”NBERWorkingPaper28466,NationalBureauofEconomicResearch,Cambridge,MA.Damania,R.,S.Desbureaux,M.Hyland,A.Islam,A.S.Rodella,J.Russ,J.,andE.Zaveri.2017.UnchartedWaters:TheNewEconomicsofWaterScarcityandVariability.Washington,DC:WorldBank.Das,P.2014.“Women’sParticipationinCommunity-LevelWaterGovernanceinUrbanIndia:TheGapBetweenMotivationandAbility.”WorldDevelopment64:206–18.Davis,L.W.,andP.J.Gertler.2015.“ContributionofAirConditioningAdoptiontoFutureEnergyUseunderGlobalWarming.”PNAS112(19):5962–67.DeConinck,H.,A.Revi,M.Babiker,P.Bertoldi,M.Buckeridge,A.Cartwright,W.Dong,etal.2018.“StrengtheningandImplementingtheGlobalResponse.”InGlobalWarmingof1.5°C,editedbyV.Masson-Delmotte,P.Zhai,H.-O.Pörtner,D.Roberts,J.Skea,P.R.Shukla,A.Pirani,etal.Geneva:IntergovernmentalPanelonClimateChange.Deininger,K.,H.Selod,andA.Burns.2012.TheLandGovernanceAssessmentFramework:IdentifyingandMonitoringGoodPracticeintheLandSector.Washington,DC:WorldBank.Delgado,A.,D.J.Rodriguez,C.A.Amadei,andM.Makino.2021.WaterinCircularEconomyandResilience.Washington,DC:WorldBank.PoliciesforPromotingGreen,Resilient,andInclusiveUrbanDevelopment319319PoliciesforPromotingGreen,Resilient,andInclusiveUrbanDevelopmentDelValle,A.,A.deJanvry,andE.Sadoulet.2020.“RulesforRecovery:ImpactofIndexedDisasterFundsonShockCopinginMexico.”AmericanEconomicJournal:AppliedEconomics12(4):164–95.Desmet,K.,andR.Jedwab.2022.“AreWeOver-Buildingin‘Bad’LocationsGlobally?FutureClimateChangeandDurableRealEstate.”Backgroundpaperpreparedforthisreport,WorldBank,Washington,DC.Devereux,S.,andJ.Cuesta.2021.“Urban-SensitiveSocialProtection:HowUniversalizedSocialProtectionCanReduceUrbanVulnerabilitiesPostCOVID-19.”ProgressinDevelopmentStudies21(4):340–60.Domon,S.,M.Hirota,T.Kono,S.Managi,andY.Matsuki.2022.“TheLong-RunEffectsofCongestionTolls,CarbonTax,andLandUseRegulationsonUrbanCO2Emissions.”RegionalScienceandUrbanEconomics92:103750.Ehlers,T.,B.Mojon,andF.Packer.2020.“GreenBondsandCarbonEmissions:ExploringtheCaseforaRatingSystemattheFirmLevel.”BISQuarterlyReview(September):31–47.Ehrlich,I.,andG.S.Becker.1972.“MarketInsurance,Self-Insurance,andSelf-Protection.”JournalofPoliticalEconomy80(4):623–48.Erman,A.,E.Motte,R.Goyal,A.Asare,S.Takamatsu,X.Chen,S.Malgioglio,etal.2020.“TheRoadtoRecovery:TheRoleofPovertyintheExposure,VulnerabilityandResiliencetoFloodsinAccra.”EconomicsofDisastersandClimateChange4(1):171–93.EuropeanUnion.2013.ClimateChangeAdaptation:EmpowermentofLocalandRegionalAuthorities,withaFocusonTheirInvolvementinMonitoringandPolicyDesign.Brussels:EuropeanUnion.Fay,M.,H.I.Lee,M.Mastruzzi,S.Han,andM.Cho.2019.“HittingtheTrillionMark:ALookatHowMuchCountriesAreSpendingonInfrastructure.”PolicyResearchWorkingPaper8730,WorldBank,Washington,DC.Fei,J.,Y.Wang,Y.Yang,S.Chen,andQ.Zhi.2016.“TowardsEco-city:TheRoleofGreenInnovation.”EnergyProcedia104:165–70.Ferdous,M.R.,A.Wesselink,L.Brandimarte,G.D.Baldassarre,andM.M.Rahman.2019.“TheLeveeEffectalongtheJamunaRiverinBangladesh.”WaterInternational44(5):496–519.Fernandes,E.2011.RegularizationofInformalSettlementsinLatinAmerica.Cambridge,MA:LincolnInstituteofLandPolicy.Flores,F.M.,S.Milusheva,andA.R.Reichert.2021.“ClimateAnomaliesandInternationalMigration:ADisaggregatedAnalysisforWestAfrica.”PolicyResearchWorkingPaper9664,WorldBank,Washington,DC.Foster,V.,andC.Briceno-Garmendia.2010.Africa’sInfrastructure:ATimeforTransformation:ATimeforTransformation.AfricaDevelopmentForum.Washington,DC:WorldBank.Franklin,S.2018.“Location,SearchCostsandYouthUnemployment:ExperimentalEvidencefromTransportSubsidies.”EconomicJournal128(614):2353–79.Georgic,W.2019.“VulnerabilityandPolicyResponse:UnintendedConsequences.”PhDdissertation,OhioStateUniversity.Gerda,N.2021.“WhatCanLocalGovernmentsDotoFightClimateChange?AHeckofaLot,SayExperts.”VoiceofOC,April8,2021.https://voiceofoc.org/2021/04/what-can-local-governments-do-to-fight-climate-change-a-heck-of-a-lot-say-experts/.320THRIVINGTHRIVING320GFDRR(GlobalFacilityforDisasterReductionandRecovery).2012.AnalysisofDisasterRiskManagementinColombia:AContributiontotheCreationofPublicPolicies.Washington,DC:WorldBank.GFDRR(GlobalFacilityforDisasterReductionandRecovery).2014.UnderstandingRiskinanEvolvingWorld.Washington,DC:WorldBank.GFDRR(GlobalFacilityforDisasterReductionandRecovery).2019.BuildingRegulationforResilience:ProgramBrief.Washington,DC:WorldBank.Ghesquiere,F.,L.Jamin,andO.Mahul.2006.EarthquakeVulnerabilityReductionPrograminColombia:AProbabilisticCost-BenefitAnalysis.Washington,DC:WorldBank.Gibbs,T.2000.DetailingforHurricanes.Washington,DC:OrganizationofAmericanStates.https://www.oas.org/pgdm/document/mhbdc/b3_text.pdf.GlobalCenteronAdaptationandWorldResourcesInstitute.2019.AdaptNow:AGlobalCallforLeadershiponClimateResilience.GlobalCenteronAdaptation,WorldResourcesInstitute.GoodFoodInstitute.2021a.Plant-BasedMeatforaGrowingWorld.GoodFoodInstitute.GoodFoodInstitute.2021b.StateoftheIndustryReport—Plant-BasedMeat,Eggs,andDairy.GoodFoodInstitute.Grafton,R.Q.,L.Chu,andT.Kompas.2015.“OptimalWaterTariffsandSupplyAugmentationforCost-of-ServiceRegulatedWaterUtilities.”UtilitiesPolicy34:54–62.Gramkow,G.,andA.Anger-Kraavi.2018.“CouldFiscalPoliciesInduceGreenInnovationinDevelopingCountries?TheCaseofBrazilianManufacturingSectors.”ClimatePolicy18(2):246–57.GRiF(GlobalRiskFinancingFacility).2020.AnnualReportJuly1,2019–June30,2020.Washington,DC:WorldBank.Gröger,A.,andY.Zylberberg.2016.“InternalLaborMigrationasaShockCopingStrategy:EvidencefromaTyphoon.”AmericanEconomicJournal:AppliedEconomics8(2):123–53.Gupta,A.2008.“ClimateChangeandInsuranceMarkets.”DevelopmentOutreach10(1):30–34.Gurenko,E.,R.Lester,O.Mahul,andS.O.Gonulal.2006.EarthquakeInsuranceinTurkey.Washington,DC:WorldBank.GWP(GlobalWaterPartnership)andUNEP-DHI(UnitedNationsEnvironmentProgramme–DHICentreonWaterandEnvironment).2021.AdvancingtowardsGenderMainstreaminginWaterResourcesManagement.Stockholm:GWPandUNEP-DHICentreonWaterandEnvironment.Hallegatte,S.2009.“StrategiestoAdapttoanUncertainClimateChange.”GlobalEnvironmentalChange19(2):240–47.Hallegatte,S.,J.Rentschler,andJ.Rozenberg.2019.Lifelines:TheResilientInfrastructureOpportunity.Washington,DC:WorldBank.Hallegatte,S.,J.Rentschler,andJ.Rozenberg.2020.AdaptationPrinciples:AGuideforDesigningStrategiesforClimateChangeAdaptationandResilience.Washington,DC:WorldBank.Hallegatte,S.,A.Vogt-Schilb,M.Bangalore,andJ.Rozenberg.2017.Unbreakable:BuildingtheResilienceofthePoorintheFaceofNaturalDisasters,ClimateChange,andDevelopment.Washington,DC:WorldBank.PoliciesforPromotingGreen,Resilient,andInclusiveUrbanDevelopment321321PoliciesforPromotingGreen,Resilient,andInclusiveUrbanDevelopmentHallegatte,S.,andB.Walsh.2021.“NaturalDisasters,PovertyandInequality.”InTheRoutledgeHandbookofthePoliticalEconomyoftheEnvironment,editedbyÉloiLaurentandKlaraZwickl.London:Routledge.Haselip,J.,U.E.Hansen,D.Puig,S.Trærup,andS.Dhar.2015.“Governance,EnablingFrameworksandPoliciesfortheTransferandDiffusionofLowCarbonandClimateAdaptationTechnologiesinDevelopingCountries.”ClimaticChange131(3):363–70.Hassan,R.,M.S.Islam,A.S.M.Saifullah,andM.Islam.2013.“EffectivenessofSocialSafetyNetProgramsonCommunityResiliencetoHazardVulnerablePopulationinBangladesh.”JournalofEnvironmentalScienceandNaturalResources6(1):123–29.Hegazy,I.,W.Seddik,andH.Ibrahim.2017.“TowardsGreenCitiesinDevelopingCountries:EgyptianNewCitiesasaCaseStudy.”InternationalJournalofLow-CarbonTechnologies12(4):358–68.Henrique,K.P.,P.Tschakert,C.BourgaultduCoudray,P.Horwitz,K.D.C.Krueger,andA.J.Wheeler.2022.“NavigatingLossandValueTrade-OffsinaChangingClimate.”ClimateRiskManagement35:100405.Hino,M.,andM.Burke.2021.“TheEffectofInformationaboutClimateRiskonPropertyValues.”PNAS118(17):e2003374118.Horbach,J.2020.“ImpactsofRegulationonEco-innovationandJobCreation.”IZAWorldofLabor,InstituteofLaborEconomics,Bonn.Hsu,A.,andR.Rauber.2021.“DiverseClimateActorsShowLimitedCoordinationinaLarge-ScaleTextAnalysisofStrategyDocuments.”CommunicationsEarthandEnvironment2:30.Hutton,N.S.2018.“TheLeveeEffectRevisited:ProcessesandPoliciesEnablingDevelopmentinYubaCounty,California.”FloodRiskManagement12(3):e12469.IEA(InternationalEnergyAgency).2015.CapturingtheMultipleBenefitsofEnergyEfficiency.Paris:IAE.IPCC(IntergovernmentalPanelonClimateChange).2022.“SummaryforPolicymakers.”InClimateChange2022:Impacts,AdaptationandVulnerability.ContributionofWorkingGroupIItotheSixthAssessmentReportoftheIntergovernmentalPanelonClimateChange.Cambridge,UK:CambridgeUniversityPress.Jensen,O.,L.P.Weiss,S.Dahan,S.Fitzgerald,A.M.Khan,andF.Armendaris.Forthcoming.“WaterSecureCities:PositionPaper.”WorldBank,Washington,DC.Kahn,M.E.2010.Climatopolis:HowOurCitiesWillThriveinaHotterFuture.NewYork:BasicBooks.Känzig,D.R.2021.“TheUnequalEconomicConsequencesofCarbonPricing.”DepartmentofEconomics,LondonBusinessSchool.http://dx.doi.org/10.2139/ssrn.3786030.Khan,M.R.2013.TowardaBindingClimateChangeAdaptationRegime:AProposedFramework.Abingdon,UK:Routledge.Kona,A.,P.Bertoldi,F.Monforti-Ferrario,S.Rivas,andJ.F.Dallemand.2018.“CovenantofMayorsSignatoriesLeadingtheWayToward1.5DegreeGlobalWarmingPathway.”SustainableCitiesandSociety41:568–75.Kremer,M.1993.“PopulationGrowthandTechnologicalChange:OneMillionB.C.to1990.”QuarterlyJournalofEconomics108(3):681–716.322THRIVINGTHRIVING322Lagakos,D.2020.“Urban-RuralGapsintheDevelopingWorld:DoesInternalMigrationOfferOpportunities?”JournalofEconomicPerspectives34(3):174–92.Lenox,M.,A.King,andJ.Ehrenfeld.2000.“AnAssessmentofDesign-for-EnvironmentPracticesinLeadingUSElectronicsFirms.”Interfaces30(3):83–94.Mahony,C.,B.Maher,andZ.Haile.2021.“MachineLearninginUgandaBringsthePowerofRiskFinancingtoStrengthenRefugeeandHostCommunityResilience.”WorldBankBlogs,January22,2021.https://blogs.worldbank.org/nasikiliza/machine-learning-uganda-brings-power-risk-financing-strengthen-refugee-and-host.McAuliffe,M.,andA.Triandafyllidou,eds.2021.WorldMigrationReport2022.Geneva:InternationalOrganizationforMigration.McCarney,P.2010.“Conclusions:GovernanceChallengesinUrbanandPeri-urbanAreas.”InPeri-urbanWaterandSanitationServices:Policy,PlanningandMethod,277–97.Berlin/Heidelberg:SpringerScienceBusinessMedia.McCarney,P.,H.Blanco,J.Carmin,andM.Colley.2011.“CitiesandClimateChange.”InClimateChangeandCities:FirstAssessmentReportoftheUrbanClimateChangeResearchNetwork,editedbyC.Rosenzweig,W.D.Solecki,S.A.Hammer,andS.Mehrotra,249–69.Cambridge,UK:CambridgeUniversityPress.McDonald,R.I.,K.F.Weber,J.Padowski,T.Boucher,andD.Shemie.2016.“EstimatingWatershedDegradationovertheLastCenturyandItsImpactonWater-TreatmentCostsfortheWorld’sLargeCities.”PNAS113(32):9117–22.Michaels,G.,D.Nigmatulina,F.Rauch,T.Regan,N.Baruah,andA.Dahlstrand.2021.“PlanningAheadforBetterNeighborhoods:Long-RunEvidencefromTanzania.”JournalofPoliticalEconomy129(7):2112–56.Mukim,M.2018a.BamakoUrbanSectorReview:AnEngineofGrowthandServiceDelivery.Washington,DC:WorldBank.Mukim,M.2018b.FreetownUrbanSectorReview:OptionsforGrowthandResilience.Washington,DC:WorldBank.Mukim,M.2021.BurundiUrbanizationReview.Washington,DC:WorldBank.Mukim,M.,andJ.Tingting.2018.“EmpoweringCities—GoodforGrowth?”AsianDevelopmentReview35(1):175–95.Mumtaz,M.2021.“RoleofCivilSocietyOrganizationsforPromotingGreenandBlueInfrastructuretoAdaptingClimateChange:EvidencefromIslamabadcity,Pakistan.”JournalofCleanerProduction309:127296.Narayan,S.,M.W.Beck,B.G.Reguero,I.J.Losada,B.vanWesenbeeck,N.Pontee,J.N.Sanchirico,etal.2016.“TheEffectiveness,CostsandCoastalProtectionBenefitsofNaturalandNature-BasedDefences.”PLoSONE11(5):e0154735.Nicholls,R.J.,J.Hinkel,D.Lincke,andT.vanderPol.2019.“GlobalInvestmentCostsforCoastalDefensethroughthe21stCentury.”PolicyResearchWorkingPaper8745,WorldBank,Washington,DC.Nogrady,B.2021.“ChinaLaunchesWorld’sLargestCarbonMarket:ButIsItAmbitiousEnough?”ScientificAmerican,July21,2021.https://www.scientificamerican.com/article/china-launches-worlds-largest-carbon-market-but-is-it-ambitious-enough/.322PoliciesforPromotingGreen,Resilient,andInclusiveUrbanDevelopment323323PoliciesforPromotingGreen,Resilient,andInclusiveUrbanDevelopmentOECD(OrganisationforEconomicCo-operationandDevelopment).2015a.GoverningtheCity.Paris:OECD.OECD(OrganisationforEconomicCo-operationandDevelopment).2015b.WaterandCities:EnsuringSustainableFutures.OECDStudiesonWater.Paris:OECD.Oijstaeijen,W.V.,S.V.Passel,andJ.Cools.2020.“UrbanGreenInfrastructure:AReviewonValuationToolkitsfromanUrbanPlanningPerspective.”JournalofEnvironmentalManagement267:110603.Ortega,F.,andS.Tas.pınar.2018.“RisingSeaLevelsandSinkingPropertyValues:HurricaneSandyandNewYork’sHousingMarket.”JournalofUrbanEconomics106(July):81–100.Owens,K.E.,S.Gulyani,andA.Rizvi.2018.“SuccessWhenWeDeemedItFailure?RevisitingSitesandServicesProjectsinMumbaiandChennai20YearsLater.”WorldDevelopment106(June):260–72.Parthasarathy,R.,V.Rao,andN.Palaniswamy.2019.“DeliberativeDemocracyinanUnequalWorld:AText-as-DataStudyofSouthIndia’sVillageAssemblies.”AmericanPoliticalScienceReview113(3):623–40.Peralta,A.,andJ.B.Scott.2018.“MovingtoFloodPlains:TheUnintendedConsequencesoftheNationalFloodInsuranceProgramonPopulationFlows.”Workingpaper.https://www.semanticscholar.org/paper/Moving-to-Flood-Plains-%3A-The-Unintended-of-the-on-Peralta-Scott/cf9e3987df6b77448ebc5207c97fa64b8952175f.Peralta-Quiros,T.,T.Kerzhner,andP.Avner.2019.“ExploringAccessibilitytoEmploymentOpportunitiesinAfricanCities:AFirstBenchmark.”PolicyResearchWorkingPaper8971,WorldBank,Washington,DC.Pestel,N.2019.“EmploymentEffectsofGreenEnergyPolicies.”CentreforEuropeanEconomicResearch,IZAWorldofLabor,Bonn.Quiñones,E.J.,S.Liebenehm,andR.Sharma.2021.“LeftHomeHighandDry—ReducedMigrationinResponsetoRepeatedDroughtsinThailandandVietnam.”PopulationandEnvironment42(4):579–621.Raj,S.,S.K.Paul,A.Chakraborty,andJ.Kuttippurath.2020.“AnthropogenicForcingExacerbatingtheUrbanHeatIslandsinIndia.”JournalofEnvironmentalManagement257:110006.ReckienD.,J.Flacke,M.Olazabal,andO.Heidrich.2015.“TheInfluenceofDriversandBarriersonUrbanAdaptationandMitigationPlans—AnEmpiricalAnalysisofEuropeanCities.”PLoSONE10(8):e0135597.Rentschler,J.E.2013.“WhyResilienceMatters—ThePovertyImpactsofDisasters.”PolicyResearchWorkingPaper6699,WorldBank,Washington,DC.Rentschler,J.,P.Avner,M.Marconcini,R.Su,E.Strano,S.Hallegatte,L.Bernard,andC.Riom.2022.“RapidUrbanGrowthinFloodZones:GlobalEvidencesince1985.”PolicyResearchWorkingPaper10014,WorldBank,Washington,DC.Rentschler,J.,M.Kornejew,S.Hallegatte,J,Braese,andM.Obolensky.2019.“UnderutilizedPotential:TheBusinessCostsofUnreliableInfrastructureinDevelopingCountries.”PolicyResearchWorkingPaper8899,WorldBank,Washington,DC.Rigaud,K.K.,A.deSherbinin,B.Jones,J.Bergmann,V.Clement,K.Ober,J.Schewe,etal.2018.Groundswell:PreparingforInternalClimateMigration.Washington,DC:WorldBank.324THRIVINGTHRIVING324Rogers,D.,andV.Tsirkunov.2010.CostsandBenefitsofEarlyWarningSystems.GlobalAssessmentReportonDisasterRiskReduction2011.Washington,DC:WorldBank.Romer,P.1990.“EndogenousTechnologicalChange.”JournalofPoliticalEconomy98(5):71–102.Rossitti,G.2022.“WhoIsAffected?WhichSegmentsofCityPopulationsAreMostExposedtoClimate-RelatedStresses?”Backgroundpaperpreparedforthisreport,WorldBank,Washington,DC.Rubin,A.J.,andS.Sengupta.2018.“‘YellowVest’ProtestsShakeFrance.Here’stheLessonforClimateChange.”NewYorkTimes,December6,2018.https://www.nytimes.com/2018/12/06/world/europe/france-fuel-carbon-tax.html.Saltiel,G.,F.Zhang,A.Khan,D.Whittington,andR.Brears.Forthcoming.DemandManagementforUrbanWaterSecurity.Washington,DC:WorldBank.Schug,G.R.,K.E.Blevins,B.Cox,K.Gray,andV.Mushrif-Tripathy.2013.“Infection,Disease,andBiosocialProcessesattheEndoftheIndusCivilization.”PLoSONE8(12):e84814.Selod,H.,andF.Shilpi.2021.“Rural-UrbanMigrationinDevelopingCountries:LessonsfromtheLiterature.”RegionalScienceandUrbanEconomics91:103713.SetoK.C.,S.Dhakal,A.Bigio,H.Blanco,G.C.Delgado,D.Dewar,L.Huang,etal.2014.HumanSettlements,InfrastructureandSpatialPlanning.InClimateChange2014:MitigationofClimateChange.ContributionofWorkingGroupIIItotheFifthAssessmentReportoftheIntergovernmentalPanelonClimateChange.Cambridge,UK:CambridgeUniversityPress.Sivaev,D.,B.Herzberg,andS.Manchanda.2015.“Public-PrivateDialogueforCityCompetitiveness.”CompetitiveCitiesforJobsandGrowth,CompanionPaper7,WorldBank,Washington,DC.Sivaev,D.,I.Kamilov,G.Rossitti,N.Orlova,andP.Vaggione.2022.TheTimeIsNow:HowCanUzbekistanLeverageUrbanizationasaDriverofSustainableDevelopment?Washington,DC:WorldBank.SomanathanE.,T.Sterner,T.Sugiyama,D.Chimanikire,N.K.Dubash,J.Essandoh-Yeddu,S.Fifita,etal.2014.“NationalandSub-nationalPoliciesandInstitutions.”InClimateChange2014:MitigationofClimateChange.ContributionofWorkingGroupIIItotheFifthAssessmentReportoftheIntergovernmentalPanelonClimateChange.Cambridge,UK:CambridgeUniversityPress.Song,S.,C.-C.Feng,andM.Diao.2020.“VehicleQuotaControl,TransportInfrastructureInvestmentandVehicleTravel:APseudoPanelAnalysis.”UrbanStudies57(12):2527–46.Stern,N.,S.Peters,V.Bakhshi,A.Bowen,C.Cameron,S.Catovsky,D.Crane,etal.2006.SternReview:TheEconomicsofClimateChange.London:HMTreasury.Tan,Z.,K.K.-L.Lau,andE.Ng.2016.“UrbanTreeDesignApproachesforMitigatingDaytimeUrbanHeatIslandEffectsinaHigh-DensityUrbanEnvironment.”EnergyandBuildings114:265–74.Tolbert,J.2021.“BeyondCities:BreakingThroughBarrierstoRuralElectricVehicleAdoption.”EnvironmentalandEnergyStudyInstitute.https://www.eesi.org/articles/view/beyond-cities-breaking-through-barriers-to-rural-electric-vehicle-adoption.Tortajada,C.,andP.vanRensburg.2020.“DrinkMoreRecycledWastewater.”Nature577:26–28.UNEP(UnitedNationsEnvironmentProgramme).2021.AdaptationGapReport2021:TheGatheringStorm—AdaptingtoClimateChangeinaPost-pandemicWorld.Nairobi:UNEP.PoliciesforPromotingGreen,Resilient,andInclusiveUrbanDevelopment325325PoliciesforPromotingGreen,Resilient,andInclusiveUrbanDevelopmentViguié,V.,andS.Hallegatte.2012.“Trade-OffsandSynergiesinUrbanClimatePolicies:NatureClimateChange.”NatureClimateChange2(5):334–37.Vona,F.2021.“ManagingtheDistributionalEffectsofEnvironmentalandClimatePolicies:TheNarrowPathforaTripleDividend.”EnvironmentWorkingPaper188,OrganisationforEconomicCo-operationandDevelopment,Paris.Weiss,J.,E.Dunkelberg,andT.Vogelpohl.2012.“ImprovingPolicyInstrumentstoBetterTapintoHomeownerRefurbishmentPotential:LessonsLearnedfromaCaseStudyinGermany.”EnergyPolicy44:406–15.White,R.,andWahba,S.2019.“AddressingConstraintstoPrivateFinancingofUrban(Climate)InfrastructureinDevelopingCountries.”InternationalJournalofUrbanSustainableDevelopment11(3):245–56.Wilson,R.2021.“MovingtoJobs:TheRoleofInformationinMigrationDecisions.”JournalofLaborEconomics39(4):1083–128.Winkler,S.L.,J.E.Anderson,L.Garza,W.C.Ruona,R.Vogt,andT.J.Wallington.2018.“VehicleCriteriaPollutant(PM,NOx,CO,HCs)Emissions:HowLongShouldWeGo?”npjClimateandAtmosphericScience1:26.Wishart,M.,T.Wong,B.Furmage,X.Liao,D.Pannell,andJ.Wang.2021.ValuingtheBenefitsofNature-BasedSolutions:AManualforIntegratedUrbanFloodManagementinChina.Washington,DC:WorldBank.WorldBank.2010.NaturalHazards,UnNaturalDisasters:TheEconomicsofEffectivePrevention.Washington,DC:WorldBank.WorldBank.2014.RepublicofBurundi.FiscalDecentralizationandLocalGovernanceManagingTrade-OffstoPromoteSustainableReforms.BurundiPublicExpenditureReview.Washington,DC:WorldBank.WorldBank.2015a.CompetitiveCitiesforJobsandGrowth.Washington,DC:WorldBank.WorldBank.2015b.WorldDevelopmentReport2015:Mind,Society,andBehavior.Washington,DC:WorldBank.WorldBank.2016.PerspectivesUrbaines:VillesÉmergentesPourUnSénégalÉmergent.Washington,DC:WorldBank.WorldBank.2018.AligningInstitutionsandIncentivesforSustainableWaterSupplyandSanitationServices.Washington,DC:WorldBank.WorldBank.2019a.RepublicofGuinea—Planning,Connecting,FinancinginConakry.UrbanSectorReview.Washington,DC:WorldBank.WorldBank.2019b.ResilientHousingBestPracticesandLessonsLearned:UnitedStatesandJapan.Washington,DC:WorldBank.WorldBank.2020a.StrengtheningUrbanResilienceinMorocco:BuildingCapacityandIdentifyingRisktoBetterProtectCitiesandManageDisasters.ResultsinResilienceSeries.Washington,DC:WorldBank.WorldBank.2020b.ThePhilippinesParametricCatastropheRiskInsuranceProgramPilot:LessonsLearned.Washington,DC:WorldBank.WorldBank.2021a.Green,Resilient,andInclusiveDevelopment.Washington,DC:WorldBank.326THRIVINGTHRIVING326WorldBank.2021b.InvestmentinDisasterRiskManagementinEuropeMakesEconomicSense:BackgroundReport.EconomicsforDisasterPreventionandPreparedness.Washington,DC:WorldBank.WorldBank.2021c.SomaliaUrbanizationReview:FosteringCitiesasAnchorsofDevelopment.Washington,DC:WorldBank.WorldBank.2021d.TheStateofCitiesClimateFinance:Part2.TheEnablingConditionsforMobilizingUrbanClimateFinance.Washington,DC:WorldBank.WorldBank.2021e.“UnlockingPrivateInvestmentinClimateAdaptationandResilience.”WorldBankNews,March4,2021.https://www.worldbank.org/en/news/feature/2021/03/04/unlocking-private-investment-in-climate-adaptation-and-resilience.WorldBank.Forthcominga.TheFeasibilityFrontier:StrategiestoMakePolicyforaNetZeroFuturePossible.Washington,DC:WorldBank.WorldBank.Forthcomingb.TerritorialDecarbonizationinBulgaria.Washington,DC:WorldBank.Zachmann,G.,G.Fredriksson,andG.Claeys.2018.TheDistributionalImpactsofClimatePolicies.BruegelBlueprintSeries28.Brussels:Bruegel.Zaveri,E.,J.Russ,A.Khan,R.Damania,E.Borgomeo,andA.Jägerskog.2021.EbbandFlow:Volume1.Water,Migration,andDevelopment.Washington,DC:WorldBank.Zhang,H.,A.Buote,andW.Acworth.2021.“China’sPilotEmissionTradingSystemsandElectricityMarkets(HubeiandShenzhen).”CaseStudyReport,GermanEnvironmentAgency,Berlin.Zhang,L.,andT.Leonard.2019.“FloodHazardsImpactonNeighborhoodHousePrices.”JournalofRealEstateFinanceandEconomics58(4):656–74.ECO-AUDITEnvironmentalBenefitsStatementTheWorldBankGroupiscommittedtoreducingitsenvironmentalfoot-print.Insupportofthiscommitment,weleverageelectronicpublishingoptionsandprint-on-demandtechnology,whichislocatedinregionalhubsworldwide.Together,theseinitiativesenableprintrunstobeloweredandshippingdistancesdecreased,resultinginreducedpaperconsumption,chem-icaluse,greenhousegasemissions,andwaste.WefollowtherecommendedstandardsforpaperusesetbytheGreenPressInitiative.ThemajorityofourbooksareprintedonForestStewardshipCouncil(FSC)–certifiedpaper,withnearlyallcontaining50–100percentrecycledcontent.Therecycledfiberinourbookpaperiseitherunbleachedorbleachedusingtotallychlorine-free(TCF),processedchlorine–free(PCF),orenhancedelementalchlorine–free(EECF)processes.MoreinformationabouttheBank’senvironmentalphilosophycanbefoundathttp://www.worldbank.org/corporateresponsibility.ISBN978-1-4648-1935-3SKU211935
VIP
