电信服务行业每用户平均收入处于5G到来之前的低谷恢复中移动评级并评为买入中电信中联通评为中性

PMSummary15Giscomingandwillbedifferentthan4G6Chinatelcocapextoturnpositivein20197Financials:DeceleratinggrowthandsteadyEBITDAmargin11Valuation:FactoringinROEandcashdividends14M&Aframework18ChinaMobile(0941.HK):5Gandcashdividendexpectedtodriveshareprice;reinstateatBuy20ChinaMobileFinancials21ChinaTelecom(0728.HK):ARPUunderpressure;reinstateatNeutral24ChinaTelecomFinancials25ChinaUnicom(0762.HK/600050.SS):Consensusoptimisticonshareincentiveplan;reinstateatNeutral28ChinaUnicomFinancials29Appendix32信息披露附录332018年10月9日2全球投资研究中国电信服务目录38%-56%and7%-46%oftotalrevenuesasof1H18.However,ratherthanprotectreturnson4GandFTTH(fibertohome)investments,thecompanieshaveengagedinaggressivecompetitiveactivity.Specifically,CUstartedalimitedpricewaronwirelessdataservicesinearly2017,CTlaunchedanunlimiteddatapackageinlate2017,andCMsharplycutpricesinmid-2018.Inaddition,regulatorycurbsonvoiceanddataroamingwereimplementedin3Q17and3Q18,respectively.Meanwhile,CMhasaggressivelygainedmarketshareinwirelinedataservicewithlowARPUandcontinuestopressureCTandCU.Accordingly,weforecasttotalservicerevenuegrowthtodeceleratein2H18and2019.Catalyst#1:Weexpectmoremixed-ownershipreformbutapotentialmergerofCTandCUseemsunlikely.OnSeptember4,2018,Bloomberg,citingpeoplefamiliarwiththematter,reportedthatChina’stopleadersarereviewingaproposaltomergeCUandCTtospeedup5Gdevelopment.BothCUandCThavesincedeniedsuchacombinationiscurrentlyintheworksandwenotethatsuchapotentialtransactionwouldrepresentareversalofChina’sSOEreformsince1990andcouldpresentchallengesintermsofheadcountredundancyandnetworksharingin2Gand3G.Whilewetakenoviewonthepotentialoutcomeofanyproposedtransaction,wethinksuchconsiderationsmakeitappearunlikelyintheintermediateterm.Furthermore,basedonrecentmediareports,weexpectCTandCUtoobtaintheirrespective5Gspectrumallocationsin2H18,indicatingthegovernment’sintentionforCTandCUtobuildtheirownseparate5Gnetworks.Meanwhile,China’sgovernmenthasexpandeditsmixed-ownershipreformmeasuresfromsixcompanies(includingCU)inOctober2016to200companiesasofSeptember2018.WebelievebothCMandCTcouldbenefitfrompotentialmixedownershipreformonimprovedoperatingefficiency.Catalyst#2:WeexpectCMtoraiseitscashdividendpayoutratiofrom48%in2018to50%in2019.InAugust2018,ChinaTowerstatedinitsIPOprospectusthatitwouldstarttopaycashdividendsata50%payoutratiobeginningin2019.Tothebestofourknowledge,thisisthefirsttimeanSOEhaspubliclyguideditscashdividend.Inaddition,50%isthehighestpayoutratioamongalllargeChinaSOEs,indicatingSOEsmaynowhavemoreflexibilityinpayingcashdividends.CMisthelargestshareholderofChinaTower,owninga28.5%stake.CMalsohasamuchstrongerbalancesheetwith2019EnetcashofRMB441bnversusnetdebtofRMB96bnin2019EforChinaTower.Accordingly,wethinkminorityinvestorsmaypressureCMtoconsidermatchingChinaTower’spayoutratioof50%in2019E.Catalyst#3:5Giscomingandwillbeverydifferentthan4G.Unlike4G,whichhasenabledbettersmartphonedatacommunicationforconsumers,webelievetheinitialusersof5Garelikelytobecorporatesandgovernmentagencieslookingtoboostautonomousdriving,industrialIoT,videosurveillance,andedgecomputingcapabilities.Corporateusers,inparticular,tendtobelesspricesensitivebutmorequalitysensitivethanconsumersandarethereforelikelytopreferworkingexclusivelywiththetopserviceproviderintermsofcoverageandservice(i.e.,CM).Accordingly,weexpectCMtocapturethelion’sshareofChina’s5Gmarketin2020-22.2018年10月9日3全球投资研究中国电信服务Catalyst#4:China’s5Gcapexrampislikelytobeslowerthanthatof4G.China’sthreemajortelcoshaveallindicatedthattheir5Gbusinessmodelsremaininfluxandthat5Gnetworkrolloutsarelikelytotrackthedevelopmentof5Gapplications,incontrasttotherapidrolloutoffullcoveragewith4G.In2020-21,weexpecttheChinesetelcostofocusinitiallyon5Goutdoorcoverage,alongroadsandareasspecializinginindustrialIoTaswellasedgecomputingactivities,withindoorcoveragerolloutstocomeafter2021.Assuch,ourcapexestimatesare4%-13%belowconsensusfor2019E-20E.Our12mTPsarebasedonaglobalP/BandROEregression.CM,CT,andCUshareshaveunderperformedHSCEIby16%-22%overthelast12monthsmainlyduetopoorROEandinsufficientcashdividends.WeuseaglobalP/BandROEregressiontofactorROEandcashdividendsintoourvaluationframework.WefindglobaltargetAdj.P/B(definedas(12mTP-DPS)/(2019EBVPS))tobehighlycorrelatedwithROE(r2=0.83)amongthe33globaltelcostockscoveredbyGSintheUS,Europe,EMEA,andAsia.Meanwhile,overthelast12months,theaverageP/BforChina’sBigThreetelcoshastrendedata4%-39%discount(middleleveldiscountof21.5%totheregressionofthetradingAdj.P/BandROEofglobalpeers(r2=0.64)).Weexpectthismiddle21.5%discounttocontinueoverthenext12monthsduetoARPUpressureand5Guncertainty.Our12mTPsofHK$96.6,HK$4.6,andHK$8.9forCM,CT,andCUarebasedon1.6x,0.9x,and0.8x2019EP/Bandimply4.4x,3.5x,and2.4x2019EEV/EBITDA,respectively.WealsoreinstatecoverageofCU-Awitha12mTPofRMB5.4assuminganA-Hpremiumof16.5%(averagedailyA-Hpricepremiumoverthelastoneyear).Risks:Weseefurtherregulatorypressureontariffsandpotentialdelaysinthe5GrolloutscheduleasrisksforChina’sBigThreetelcos.ForCM,downsiderisksare:1)CMhasahighrevenuemixofwirelessdataandisvulnerab