nPropertyweaknessisperhapsthemostchallenginggrowthheadwindamidhinasongoingpostreopeningrecoveryandthusthemomentumandsentimentinthepropertysectorhavesigni?cantimplicationsforgrowthandpolicies.Webelievehomepriceswarrantmoreattentionamidthetug-of-warinthepropertysectorbetweenpersistentdownwardpressuresandincreasingeasinghopes.nInthisnotewereviewhomepricemeasuresfromvarioussourcestogaugerecenttrends,explaintheirmajordifferences,anddiscusstheimplications.Specicallywefocusonhomepricemeasuresfrombothof?cial(NBS)andsigni?cantchallengesforcompilingaportfoliothatisrelativelystableforpricetracking,implyingnoperfecthomepricemeasure.nDespitesomeimperfections,NBSnewhomepriceshavebeensetbypolicymakersasonedeterminantforcity-levelmortgagerates.Itsfuturetrendsmayalsoaffectpolicymakers’decisiononwhen,howandwheretoexpandthepilotcitiesforpropertytaxoverthelongrun,inourview.NBShomepricemetricshaveahighercorrelationwithpropertyactivityandmarketsentimentthanprivatemeasures.nBasedonourestimates,followingthedecade-longincreases,homepriceshavebeenlargelystableintheprimarymarketsincemid-2021buttrendeddowninthesecondarymarket.Homepricesinlargecitiescontinuedtooutperformthoseinsmallcities,andrecentsecondaryhomepricedeclinesweremainlyledbylower-tiercities.Homepricesineasterncoastalprovinceshaveoutperformedmostinlandregionsinrecentyears.nTocounteractpersistentweaknessinthepropertysector,weexpectmorehousingeasingmeasuresincomingmonths,butthemagnitudeshouldbesmallerthanpreviouseasingcycles.AlthoughnationwideimpositionofpropertytaxmightbepartofChina’send-gameforpropertysectorpolicy,asubstantialappearsunlikelythisyearPolicymakersmayrelatedtoprivatedeveloperslocalgovernmentimplicitdebtandsomesmallbanksinstead.+852-3966-4004|lisheng.wang@GoldmanSachs(Asia)L.L.C.+852-2978-7283|yuting.y.yang@GoldmanSachs(Asia)L.L.C.MaggieWei+852-2978-6962|maggie.wei@GoldmanSachs(Asia)L.L.C.en+852-2978-2418|xinquan.chen@GoldmanSachs(Asia)L.L.C.AndrewTilton+852-2978-1802|andrew.tilton@GoldmanSachs(Asia)L.L.C.derthisreportasonlyasinglefactorinmakingtheirinvestmentdecisionForRegACcerti?cationandotherimportantdisclosures,seetheDisclosureAppendix,orgoto/research/hedge.html.July2homestartsandsales-20-4060Percentchange,yoyPercentchange,yoyhomestartsandsales-20-4060Percentchange,yoyPercentchange,yoy604020Newprice,70-citynewhomeprice NewhomestartsFloorspacesold604020-20-40--11121314151617181920212223PropertyweaknessisperhapsthemostchallenginggrowthheadwindamidChina’songoingpost-reopeningrecovery,andthusthemomentumandsentimentinthepropertysectorhavesigni?cantimplicationsforgrowthandpolicies.Webelievehomepriceswarrantmoreattentionamidthetug-of-warinthepropertysectorbetweenpersistentdownwardpressuresandincreasingeasinghopesforthreereasons:nHomepricescoulddrivehousingactivity,andtherefore,economicgrowth(Exhibitnthelargesttransactionhouseholdsmakepricemovementsandpriceexpectationsareimportanttopurchasedecisions.nHomepricesmayalsomatterfortheriskspillovertothebankingsystem(mainlygageloansandpropertydeveloperloansandtohouseholdconsumption(mainlythroughwealtheffects,moving-inrelatedpurchases,andconsumercon?dencechannels).nCertainhousingpolicies,suchasmortgagerateadjustments,aredirectlyaffectedbysuredhomepricechangesRecentmediareportsonfallingsecondaryhomepricesinsomecities(Exhibit2)andmarketdebatesonhousehold“balancesheetrecession”alsosuggestthathomepricescouldhaveindirectbutbroaderimplicationsformarketsentimentandmacropolicies.inashomepricesseehereandhereinthisnotewereviewhomepricemeasuresfromvarioussourcestogaugerecenttrends,explaintheirmajordifferences,anddiscusstheimplicationsforgrowthandpolicies.%momsa,non-annualized%momsa,non-annualized2.52.5NBS70-cityaveragehomeprices2.0PrimarySecondary2.01.51.5 1.01.00.00.0-0.5-0.511121314151617181920212223vebeencloselytrackedbyinvestorsandecentyearsthereareanincreasingnumberofprivatesourcesExhibit3,wecomparevarioushomepricemeasuresintermsoftheirsamples,compilationmethodologies,andrecenttrends.July3ehousingmarket,asit’sestimatedasnewhomesalesvaluedividedbyvolume,butitnotanaccurateestimateforsequentialhomepricechangesasthelocationandqualityofhomessoldarenotcomparableovertime.1Bycomparison,NBS70-citypriceindexandCREIScitynewhomepriceindexhavecontrolledforpropertylocationandqualitywhenestimatingpricechanges,althoughtheiroriginaldatasourcesdifferwitheachothertosomedegree.2Wecautionnewhomepriceshereisstillalargedifferencebetweennewandsecondaryhomeprices,evenaftercontrollingforhomelocationandquality.nOnsecondary-markethomeprices,besidesNBS70-citysecondaryhomepriceindexsomeprivateentitiessuchasCentalineZhugeandBeikemostlypropertyagents,informationplatformsandconsulting?rms)alsohavetheirownmeasures.HomepricedatafromprivatesourcesusuallyhavemuchshortertimeseriesthantheNBSseries,althoughmostofthemarereleasedearlierthantheNBSone.Inepriceshavebeenaffectedbymanyventionssuchasgovernmentguidedpricesandreferenceprices(relatedtotaxpaymentsandmortgageborrowingforsecondaryabricatedsalescontractsandevenunderthetablecashpaymentstoevadetaxesanddetourmortgagerestrictions,ichdistortedsecondaryhomepricesoccasionallyHoweverduetolackofdataavailability,thereisnogoodwayforfullyadjustingforthesedistortions.1NewhomeshavebeendevelopedfurtherandfurtherawayfromcitycentersandthusASPmeasureshomepriceswithincreasinglyremotelocationsoverthelongrun.Speci?cally,itmayunderestimatepricegainsduringpropertypriceupcycles,overestimatepricedeclinesduringdowncyclesandusuallypresentmuchhighervolatilitythanothermeasures.2AsshowninExhibit3,NBS70-citynewhomepriceindexisestimatedbasedonlocalhousingauthorities,whileCREIScitynewhomepriceindexisbasedonrealestateagencyreportsand?eldinvestigationsJuly4IndicatorIndicatornameinChineseOfficial/PrivateCity-sampleCity-leveldatasourceTransactionprices?Propertylocation/qualitycontrolled?Weightedorunweightedforsampleaverage?Year-on-yearpricechange(%)202020212022Jan-May2023PrimarymarkethousingpricesNBS70-citynewhomepricedex統(tǒng)計(jì)局70個大中城市新建商品住宅價格指數(shù)Official70cities,mostlylargeandmedium-sizedcitiesNewhometransactiondatafiledwithlocalhousingauthoritiesYesYesNonationwideaveragereleasedbyNBSdirectly,simpleaverage(unweighted)bysomedatavendors(e.g.,Bloomberg,Wind),population-weightedbyGS-0.6NBSaveragesalesprice(ASP)統(tǒng)計(jì)局商品房平均銷售Official價格Nationwide,basedontop-downcalculationReportsbypropertydevelopersdirectlytoSNo;impliedASPUnweighted.EstimatedasNBSnewhomesalesvaluedividedbynewhomevolume5.63.9-3.79.3CREIS100-city中指研究院newhomeprice商品住宅銷index售價格rivate100cities,includingmoresmallcitiesthanNBS70-citysampleRawdatabasedmainlyonfieldinvestigation,plusreportsfromdevelopersandrealestateagenciesYesYesWeighted.Citywideandnationwideaveragehomepricesestimatedusinggrossfloorareaastheweighting-0.1SecondarymarkethousingpricesNBS70-citysecondaryhomepriceindex統(tǒng)計(jì)局70個大中城市二手住宅銷售價格指數(shù)Official70cities,mostlylargeandmedium-sizedcitiesRawdatabasedmainlyonreportsfromrealestateagencies,plusfieldinvestigationsandlocalhousingauthorityregistriesYesYesNonationwideaveragereleasedbyNBSdirectly,simpleaverage(unweighted)bysomedatavendors(e.g.,Bloomberg,Wind),population-weightedbyGS2.44-2.3Centaline6-citysecondaryhomepriceleadingindex中原地產(chǎn)二手住宅價格領(lǐng)先指數(shù)rivatelargecitiesTiercitiesplusTianjinandChengduProvisionalcontractsforsecondaryhometransactionfiledwithCentalineYesYesWeighted.Citywideaverageestimatedusingthemarketvalueofcomponentrealestatesastheweighting,6-citysampleaveragerecompiledbyGSusingpopulationweightings3.6-0.7-2.4Zhuge100-citysecondaryhousingmarketaverageprice諸葛找房100城二手住宅Private市場均價100cities,includingmoresmallcitiesthanNBS70-citysampleZhuge'sownsystemthatlinkstosecondaryhometransactionrecordsatrealestateagenciesYesYesWeighted.Datarequiresseparatesubscription.2-0.3Beike25-citysecondaryhomeprice貝殼二手住宅成交價格指數(shù)rivate25largeandmedium-sizedcities(availableatWind),mostlyTier-2andTier-3cities;ngTiercitiesonlyGuangzhouisavailableContractsforsecondaryhometransactionfiledwithBeikeYesYesWeighted.25-citysampleaveragerecompiledbyGSusingpopulationweightings0.84.2-5.2-5.0Besidestheabove-mentioneddistortionsfornewandsecondaryhomepricemeasures,alldatasourcesfacesigni?cantchallengeforcompilingaportfoliothatisrelativelystablestoaddresstheissue,althoughthereappearsnoperfectsolution,especiallyregardingtheadjustmentsfornewbuildingswithnosalesrecord,buildingswithintermittentsalescycles,andperiodswithverythintransactionsandextremepriceseg.,duringCovidlockdowns).TakingNBS70-citynewhomepricesasanexample:fornewbuildingswithnosalesrecord,NBSestimatestheirhomepricehistorybasedonnewhomepricesforneighboringbuildingswithsalesrecord,ornewhomepricesforbuildingswithcomparablelocationandqualityinthedistrict,orsecondaryhomepricesinneighboringliedbyof?cialandprivatesourcesappearinconsistentwithmarketperceptionsonsomesionsAssuchtobettergaugehomepricedynamicswehavetorelyonvariousnecdotalevidenceandeveneldinvestigationsJuly5Despitesomeimperfections,NBSnewhomepriceshavebeensetbypolicymakersasonedeterminantforcity-levelmortgagerateadjustments(seeBox1formoredetails).FuturetrendsintheNBShomepriceseriesmayalsoaffectpolicymakers’decisiononwhen,howandwheretoexpandthepilotcitiesforpropertytaxoverthelongrun,inourewhomeASPisrelevantforgaugingthemacroriskfromthepropertysector,asitcapturesthepricefactorofdeveloperrevenuefromnewhomesales.centShareof70citieswithrisinghomepricesforthreeconsecutivemonthsonbothyear-over-yearandcentShareof70citieswithrisinghomepricesforthreeconsecutivemonthsonbothyear-over-yearandmonth-over-monthbasiseightedpulationweighted023rcent201320190210150170114040000000ercent023Shareof70citieswithfallingnewhomepricesforthreeconsecutivemonthsonbothyear-over-yearandmonth-over-monthbasisightedlationweightedrcent0170110130190150214040000000Theongoingpropertydowncyclestartedinmid-2021,resultinginasharpcontractioninmanypropertyactivityindicatorsandsequentialhomepricedeclinesinmanycitiesthereafter.InSeptember2022,PBOCgagepricingpoliciestoallowcitiesthathaveexperiencedthreeconsecutivemonthsofnewhomepricedeclines(inbothmonth-on-monthandyear-on-yearterms)tosetmortgageratesfor?rst-timehome-buyersbelowtheofcialoorratesuntiltheendof22.InJanuary2023,PBOCandthenCBIRCof?ciallyestablishedthenewmortgagerateadjustmentsystembasedontheirpoliciesearlier,andalsointroducedapolicynormalizationmechanism:Ifcitiesexperiencethreeconsecutivemonthsofnewhomepriceincreasesinbothmonthonmonthandyear-on-yearterms),thentheof?cial?oorwouldbereinstatedonallnewmortgageslocally.AccordingtothePBOCdata,weightedaverageinterestratesonnewmortgagesfellfromtheirrecentpeakof5.6%inQ42021to4.1%inQ12023.Sincethelaunchofthenewmortgagerateadjustmentsystem,weestimateonly1cityintheNBS70-citysamplestillmettheconditionsforsettingbelow-?oormortgageratesasofMay2023vs.thepreviouspeakof38citiesinNovember2022(Exhibit4).Bycomparison,16citiesintheNBS70-citysample(22.9%oftotal)havemettheconditionsforadjustingtheiruptotheoorsasofMayExhibitThissuggestsanincreasingnumberofcitiesfacetheriskofhighermortgageratesunderthecurrentmortgagerateadjustmentmechanism.ablishedNationalFinancialRegulatoryAdministrationduringtheStateCouncilInstitutionalReformearlythisyear.July6Differenthomepricemeasurespresenteddifferentmagnitudesofvolatility,partlyduetothesizeofcitysamples:HomepricemeasureswithlargersamplesizeusuallyhavesmallervolatilityinpricechangeswithNBSnewhomeASPbeingoneexceptionasitsenotpurelycomparableovertimeExhibitMoreoverhomepricechangesmayvaryacrossmeasuresimplyingdatasourceandcompilationmethodologymaychannelstocollectdata,andtheirownmethodologytocompilecity-levelhomepricengethanNBSnewandsecondaryhomepricesinrecentyears(Exhibit7).ppRatio876543210VolatilityindifferentStandarddeviationofyoychange(LHS)High-to-lowratio(RHS)housingVolatilityindifferentStandarddeviationofyoychange(LHS)High-to-lowratio(RHS) NBSNBS70-cityCREIS100-Centaline6-Beike25-cityNBS70-cityZhuge100-averagecitycitycitysalespricePrimarymarketSecondarymarket1.351.3051.101.051.00Index(Jan2019=100)Index(Jan2019=100)50pricemeasuresShanghaipricemeasuresNBSnewhomepriceNBSsecondaryhomeprice Centalinesecondaryhomeprice50Jan-19Jan-20Jan-21Jan-22Jan-23Thecorrelationcoef?cientmatrixbetweenhomepricemeasures,propertyactivity(e.g.,?oorspacesoldandnewhomestarts,bothinvolumeterms)andpropertymarketsuggestsNBShomepricemeasures–includingNBS70-citynewandsecondaryhomeprices,andaveragenewhomesalesprices–arestillthemostrelevanttopropertyactivityandmarketindices,comparedtoothermeasures(Exhibit8).Mostprivatesourcesarerelevanttopropertyactivity,buthavemuchweakercorrelationswithpropertymarketindices.Withinhomepricemeasures,Beike25-cityhasthehighestcorrelationwithof?cialandotheralternativemeasures,whileNBSnewhomeASPhasthelowestcorrelationwithmostalternativehomepricemeasures.July7PropertypricemeasuresPropertyactivityPropertymarketindicesChinapropertyUSDHYcreditspreadCentaline6-ChinapropertyUSDHYcreditspreadCentaline6-citysecondaryhomepriceCREIS100-citynewhomepriceindexZhuge100-citysecondaryhomepriceChinapropertystockpriceindexNBS70-citynewhomepriceBeike25-citysecondaryhomepriceCorrelationcoefficientNBS70-citysecondaryhomepriceFloorspacesoldNewhomestartsnewhomesalesprice(inversed)1NBS70-citynewhome1price10.31NBSaveragenew10.31salesprice0.0010.15CREIS100-0.0010.15homepriceindexPropertypricemeasures0.280.Propertypricemeasures0.280.420.951homeprice-0.070.9410.350.07-0.070.9410.350.07secondaryhomeprice0.010.010.261secondaryhomeprice0.200.910.970.30100.200.910.970.3010.850.86homeprice0.560.350.440.470.240.2310.34Propertyac0.560.350.440.470.240.2310.34Propertyactivity0.440.420.550.550.540.570.3810.92Newhomestarts0.270.300.25-0.010.090.030.270.300.25-0.010.090.030.321Propertymarketindicespriceindex0.130.370.330.270.320.070.330.270.320.090.790.370.060.361creditspread(inversed)nourestimatesfollowingthedecadelongincreaseshomepriceshavebeenlargelystableintheprimarymarketbuttrendeddowninthesecondarymarketsincehibitAmongthesemeasuresBeikecitysecondaryhomepricesshowthelargestdeclinesincemid-2021,by~10%asofMay2023.Anecdotalevidencesuggestsevenlargerdownwardpressuretosomecities,althoughafewmediareportsmighthaveexaggeratedrecentdeclinesinhomeprices(partlyasthedatatheyquotedareeithernottransaction-based,ornotcomparableovertimeduetodifferencesinpropertylocationandquality).July820020018080Tier416060140401202010000808020020080180Tier4601604014020020018080Tier416060140401202010000808020020080180Tier4601604014020120001008080Index(Jan2019=100)30252015Index(Jan2019=100)30252015100500HousingpricemeasuresfortheprimarymarketNBS70-cityNBSaveragesalesprice CREIS100-city30252015100500Index(Jan2019=100)IndexIndex(Jan2019=100)302520151005009590HousingpriceHousingpricemeasuresforthesecondarymarketNBS70-cityCentaline6-cityZhuge100-cityBeike25-city302520151005009590Jan-19Jan-20Jan-21Jan-22Jan-23BasedonNBS70-cityhomepricesdata,weestimatesampleaveragenewhomepriceshavegainedabove50%since2015,withlargecitiessigni?cantlyoutperformingsmalldownbycitytierssuggestssequentialdeclineinNBS70-citysecondaryhomepricessinceslargelystable(seeAppendix1forthecitybreakdownbycitytier).Homepricesalsodisplayedmeaningfuldivergenceacrossregionsinrecentyears,witheasterncoastalprovincesgqingeconomiczoneoutperformingmostinlandregionsExhibitIndex(Jan2015=100)IndexIndex(Jan2015=100)NBS70-cityhomepriceindexfortheprimarymarket70cityaverage70cityaverageTier1 Tier2Tier3151617181920212223Index(Jan2015=100)IndexIndex(Jan2015=100)NBS70-cityhomepriceindexforthesecondarymarket Tier270city Tier270cityaverage1516171819202122234Differenthomepricemeasuresmayarriveatvaryingmagnitudesofpricegains.July9JilinHebeiTianjinShandongQinghaiJiangsuTibetAnhuiShanhaiSichuanZhejiangJiangxiGuizhouInnerMongolia-5%orsmallerdeclineXinjiangNingxiaShanxiiChongqingananBeijingLiaoningGansuShaanxiHenanNewhomepricechangessinceJanJilinHebeiTianjinShandongQinghaiJiangsuTibetAnhuiShanhaiSichuanZhejiangJiangxiGuizhouInnerMongolia-5%orsmallerdeclineXinjiangNingxiaShanxiiChongqingananBeijingLiaoningGansuShaanxiHenanbasedonNBS70-citysample(%;asofMay2023)HeilongjiangYunnanGuangdonGuangxiDataYunnanGuangdonGuangxiGreaterthan5%declineFlatorsmallerthan+5%increase+5%orgreaterincreaseainanTheongoingpropertycyclehasbeensigni?cantlydifferentfrompreviousrounds,mainlyaspolicymakersappearverydeterminednottousethepropertysectorasashort-termstimulustool.Tocounteractpersistentweaknessinthepropertysector,weexpectmorehousingeasingmeasuresincomingmonths,especiallyinlargecitiesandforsupportingnewhomebuyersandupgradersHoweverthistimearoundhousingeasingshouldbegcyclewhenpolicymakersenovationprogramandscrappedrestrictionsonnewhomepurchasesnationwide,amongothereasingmeasures.AlthoughnationwideimpositionofpropertytaxmightbepartofChina’send-gameforpropertysectorpolicy,weviewasubstantialexpansionofpropertytaxpilotcitiesaslesslikelythisyearandevennextyeargivenstronggrowthheadwindsandweakcon?dence.Webelievepolicymakersmayprioritizeaddressing?nancialrisksrelatedtoprivatedevelopers,localgovernmentimplicitdebtandsomesmallbanksinstead.Iftheaduallystabilizesincomingyearslikelyatalowlevelnewnormaliethe“L-shaped”recoveryweexpect)–underpinnedbyastabilizationinpropertyactivityepricespolicymakersmayhavesomeroomforlargerhousingmarketJuly10reforms,inourview.WethankSabrinaLukaninternintheAsiaEconomicsteamforhercontribution.July11CityCitytierCitylistTier-1Beijing,Shanghai,Guangzhou,ShenzhenTier-2Tianjin,Dalian,Nanjing,Hangzhou,Xiamen,Qingdao,Wuhan,Changsha,Chongqing,Chengdu,Xi'anTier-2Tier-3Shijiazhuang,Taiyuan,Hohhot,Shenyang,Changchun,Harbin,Ningbo,Hefei,Fuzhou,Nanchang,Jinan,Zhengzhou,Nanning,Haikou,Guiyang,Kunming,Lanzhou,Xining,Yinchuan,Urumqi,Tangshan,Qinhuangdao,Baotou,Dandong,Jinzhou,Jilin,Mudanjiang,Wuxi,Yangzhou,Xuzhou,Wenzhou,Jinhua,Bengbu,Anqing,Quanzhou,Jiujiang,Ganzhou,Yantai,Jining,Luoyang,Pingdingshan,Yichang,Xiangyang,Yueyang,Changde,Huizhou,Zhanjiang,Guilin,Beihai,Sanya,Zunyi,Tier-3TierTier-4Shaoguan,Luzhou,NanchongJuly12We,LishengWang,YutingYang,HuiShan,MaggieWei,XinquanChenandAndrewTilton,herebycertifythatalloftheviewsexpressedinthisreportaccuratelyre?ectourpersonalviews,whichhavenotbeenin?uencedbyconsiderationsofthe?rm’sbusinessorclientrelationships.Unlessotherwisestated,theindividualslistedonthecoverpageofthisreportareanalystsinGoldmanSachs’GlobalInvestmentResearchdivision.Seecompany-speci?cregulatorydisclosuresaboveforanyofthefollowingdisclosuresrequiredastocompaniesreferredtointhisreport:managerorco-managerinapendingtransaction;1%orotherownership;compensationforcertainservices;typesofclientrelationships;managed/co-managedpublicofferingsinpriorperiods;directorships;forequitysecurities,marketmakingand/orspecialistrole.GoldmanSachstradesormaytradeasaprincipalindebtsecurities(orinrelatedderivatives)ofissuersdiscussedinthisreport.Thefollowingareadditionalrequireddisclosures:Ownershipandmaterialcon?ictsofinterest:GoldmanSachspolicyprohibitsitsanalysts,professionalsreportingtoanalystsandmembersoftheirhouseholdsfromowningsecuritiesofanycompanyintheanalyst’sareaofcoverage.Analystcompensation:Analystsarepaidinpartbase