微觀經(jīng)濟(jì)學(xué)（第八版-英文）課件：Pricing with Market Power

11.1

CapturingConsumerSurplus11.2

PriceDiscrimination11.3

IntertemporalPriceDiscriminationandPeak-LoadPricing11.4

TheTwo-PartTariff11.5

Bundling11.6

Advertising

Appendix:TheVerticallyIntegratedFirm

PricingwithMarketPowerCHAPTEROUTLINECapturingConsumerSurplus11.1CAPTURINGCONSUMERSURPLUSFIGURE11.1●

pricediscrimination Practiceofchargingdifferentpricestodifferentconsumersforsimilargoods.Ifafirmcanchargeonlyonepriceforallitscustomers,thatpricewillbeP*andthequantityproducedwillbeQ*.Ideally,thefirmwouldliketochargeahigherpricetoconsumerswillingtopaymorethanP*,therebycapturingsomeoftheconsumersurplusunderregionAofthedemandcurve.ThefirmwouldalsoliketoselltoconsumerswillingtopaypriceslowerthanP*,butonlyifdoingsodoesnotentailloweringthepricetootherconsumers.Inthatway,thefirmcouldalsocapturesomeofthesurplusunderregionBofthedemandcurve.●

firstdegreepricediscrimination Practiceofchargingeachcustomerherreservationprice.First-DegreePriceDiscrimination●

variableprofit Sumofprofitsoneachincrementalunitproducedbyafirm;i.e.,profitignoringfixedcosts.●

reservationprice Maximumpricethatacustomeriswillingtopay

foragood.PriceDiscrimination11.2ADDITIONALPROFITFROMPERFECTFIRST-DEGREEPRICEDISCRIMINATIONFIGURE11.2Becausethefirmchargeseachconsumerherreservationprice,itisprofitabletoexpandoutputtoQ**.Whenonlyasingleprice,P*,ischarged,thefirm’svariableprofitistheareabetweenthemarginalrevenueandmarginalcostcurves.Withperfectpricediscrimination,thisprofitexpandstotheareabetweenthedemandcurveandthemarginalcostcurve.FIRST-DEGREEPRICEDISCRIMINATIONINPRACTICEFIGURE11.3PERFECTPRICEDISCRIMINATIONTheadditionalprofitfromproducingandsellinganincrementalunitisthedifferencebetweendemandandmarginalcost.IMPERFECTPRICEDISCRIMINATIONFirmsusuallydon’tknowthereservationpriceofeveryconsumer,butsometimesreservationpricescanberoughlyidentified.Here,sixdifferentpricesarecharged.Thefirmearnshigherprofits,butsomeconsumersmayalsobenefit.WithasinglepriceP4,therearefewerconsumers.TheconsumerswhonowpayP5orP6enjoyasurplus.SECOND-DEGREEPRICEDISCRIMINATIONFIGURE11.4Second-DegreePriceDiscrimination●

second-degreepricediscrimination Practiceofchargingdifferentpricesperunitfordifferentquantitiesofthesamegoodorservice.●

blockpricing Practiceofchargingdifferentpricesfordifferentquantitiesor“blocks”ofagood.Differentpricesarechargedfordifferentquantities,or“blocks,”ofthesamegood.Here,therearethreeblocks,withcorrespondingpricesP1,P2,andP3.Therearealsoeconomiesofscale,andaverageandmarginalcostsaredeclining.Second-degreepricediscriminationcanthenmakeconsumersbetteroffbyexpandingoutputandloweringcost.Third-DegreePriceDiscrimination●

third-degreepricediscrimination Practiceofdividingconsumersintotwoormoregroupswithseparatedemandcurvesandchargingdifferentpricestoeachgroup.CREATINGCONSUMERGROUPSIfthird-degreepricediscriminationisfeasible,howshouldthefirmdecidewhatpricetochargeeachgroupofconsumers?

Weknowthathowevermuchisproduced,totaloutputshouldbedivided

betweenthegroupsofcustomerssothatmarginalrevenuesforeach

groupareequal.

Weknowthattotaloutputmustbesuchthatthemarginalrevenueforeach

groupofconsumersisequaltothemarginalcostofproduction.DETERMININGRELATIVEPRICES

(11.1)

(11.2)

LetP1bethepricechargedtothefirstgroupofconsumers,P2the

pricechargedtothesecondgroup,andC(QT)thetotalcostof

producingoutputQT=Q1+Q2.TotalprofitisthenTHIRD-DEGREEPRICEDISCRIMINATIONFIGURE11.5Consumersaredividedintotwogroups,withseparatedemandcurvesforeachgroup.Theoptimalpricesandquantitiesaresuchthatthemarginalrevenuefromeachgroupisthesameandequaltomarginalcost.Heregroup1,withdemandcurveD1,ischargedP1,andgroup2,withthemoreelasticdemandcurveD2,ischargedthelowerpriceP2.MarginalcostdependsonthetotalquantityproducedQT.NotethatQ1andQ2arechosensothatMR1=MR2=MC.NOSALESTOSMALLERMARKETSFIGURE11.6Evenifthird-degreepricediscriminationisfeasible,itmaynotpaytoselltobothgroupsofconsumersifmarginalcostisrising.Herethefirstgroupofconsumers,withdemandD1,arenotwillingtopaymuchfortheproduct.Itisunprofitabletoselltothembecausethepricewouldhavetobetoolowtocompensatefortheresultingincreaseinmarginalcost.EXAMPLE11.1THEECONOMICSOFCOUPONSANDREBATESCouponsprovideameansofpricediscrimination.TABLE11.1PRICEELASTICITIESOFDEMANDFORUSERSVERSUSNONUSERSOFCOUPONSPRICEELASTICITYPRODUCTNONUSERSUSERSToilettissue–0.60–0.66Stuffing/dressing–0.71–0.96Shampoo–0.84–1.04Cooking/saladoil–1.22–1.32Drymixdinners–0.88–1.09Cakemix–0.21–0.43Catfood–0.49–1.13Frozenentrees–0.60–0.95Gelatin–0.97–1.25Spaghettisauce–1.65–1.81Crèmerinse/conditioner–0.82–1.12Soups–1.05–1.22Hotdogs–0.59–0.77EXAMPLE11.2AIRLINEFARESTABLE11.2

ELASTICITIESOFDEMANDFORAIRTRAVELFARECATEGORYELASTICITYFIRSTCLASSUNRESTRICTEDCOACHDISCOUNTEDPrice–0.3–0.4–0.9Income1.21.21.8Travelersareoftenamazedatthevarietyoffaresavailableforround-tripflightsfromNewYorktoLosAngeles.Recently,forexample,thefirst-classfarewasabove$2000;theregular(unrestricted)economyfarewasabout$1000,andspecialdiscountfares(oftenrequiringthepurchaseofatickettwoweeksinadvanceand/oraSaturdaynightstayover)couldbeboughtforaslittleas$200.Thesefaresprovideaprofitableformofpricediscrimination.Thegainsfromdiscriminatingarelargebecausedifferenttypesofcustomers,withverydifferentelasticitiesofdemand,purchasethesedifferenttypesoftickets.Airlinepricediscriminationhasbecomeincreasinglysophisticated.Awidevarietyoffaresisavailable.IntertemporalPriceDiscrimination

andPeak-LoadPricing11.3INTERTEMPORALPRICEDISCRIMINATIONFIGURE11.7IntertemporalPriceDiscrimination●

intertemporalpricediscrimination Spendingmoneyinsociallyunproductiveeffortstoacquire,maintain,orexercisemonopoly.●

peak-loadpricing Spendingmoneyinsociallyunproductiveeffortstoacquire,maintain,orexercisemonopoly.Consumersaredividedintogroupsbychangingthepriceovertime.Initially,thepriceishigh.Thefirmcapturessurplusfromconsumerswhohaveahighdemandforthegoodandwhoareunwillingtowaittobuyit.Laterthepriceisreducedtoappealtothemassmarket.PEAK-LOADPRICINGFIGURE11.8Peak-LoadPricingDemandsforsomegoodsandservicesincreasesharplyduringparticulartimesofthedayoryear.ChargingahigherpriceP1duringthepeakperiodsismoreprofitableforthefirmthanchargingasinglepriceatalltimes.Itisalsomoreefficientbecausemarginalcostishigherduringpeakperiods.EXAMPLE11.3HOWTOPRICEABEST-SELLINGNOVELPublishingbothhardboundandpaperback

editionsofabookallowspublisherstoprice

discriminate.Someconsumerswanttobuyanewbestseller

assoonasitisreleased,evenifthepriceis$25.

Otherconsumers,however,willwaitayearuntil

thebookisavailableinpaperbackfor$10.Thekeyistodivideconsumersintotwogroups,sothatthosewhoarewillingtopayahighpricedosoandonlythoseunwillingtopayahighpricewaitandbuythepaperback.Itisclear,however,thatthoseconsumerswillingtowaitforthepaperbackeditionhavedemandsthatarefarmoreelasticthanthoseofbibliophiles.Itisnotsurprising,then,thatpaperbackeditionssellforsomuchlessthanhardbacks.TheTwo-PartTariff11.4TWO-PARTTARIFFWITHASINGLECONSUMERFIGURE11.9●

two-parttariff Formofpricinginwhichconsumersarechargedbothanentryandausagefee.SINGLECONSUMERTheconsumerhasdemandcurveD.ThefirmmaximizesprofitbysettingusagefeePequaltomarginalcostandentryfeeT*equaltotheentiresurplusoftheconsumer.TWO-PARTTARIFFWITHTWOCONSUMERSFIGURE11.10TWOCONSUMERSTheprofit-maximizingusagefeeP*willexceedmarginalcost.TheentryfeeT*isequaltothesurplusoftheconsumerwiththesmallerdemand.Theresultingprofitis2T*+(P*?MC)(Q1+Q2).NotethatthisprofitislargerthantwicetheareaoftriangleABC.TWO-PARTTARIFFWITHMANYDIFFERENTCONSUMERSFIGURE11.11MANYCONSUMERSTotalprofitπisthesumoftheprofitfromtheentryfeeπaandtheprofitfromsalesπs.BothπaandπsdependonT,theentryfee.Therefore

π=πa+πs=n(T)T+(P?MC)Q(n)wherenisthenumberofentrants,whichdependsontheentryfeeT,andQistherateofsales,whichisgreaterthelargerisn.HereT*istheprofit-maximizingentryfee,givenP.TocalculateoptimumvaluesforPandT,wecanstartwithanumberforP,findtheoptimumT,andthenestimatetheresultingprofit.PisthenchangedandthecorrespondingTrecalculated,alongwiththenewprofitlevel.EXAMPLE11.4PRICINGCELLULARPHONESERVICEMosttelephoneserviceispricedusingatwo-parttariff:

amonthlyaccessfee,whichmayincludesomefree

minutes,plusaper-minutechargeforadditional

minutes.Thisisalsotrueforcellularphoneservice,

whichhasgrownexplosively,bothintheUnitedStates

andaroundtheworld.Inthecaseofcellularservice,providershavetakenthe

two-parttariffandturneditintoanartform.InmostpartsoftheUnitedStates,consumerscanchooseamongfournationalnetworkproviders—Verizon,T-Mobile,AT&T,andSprint.Theseproviderscompeteamongthemselvesforcustomers,buteachhassomemarketpower.Marketpowerarisesinpartfromoligopolisticpricingandoutputdecisions,butalsobecauseconsumersfaceswitchingcosts:Mostserviceprovidersimposeapenaltyupwardsof$200forearlytermination.Becauseprovidershavemarketpower,theymustthinkcarefullyaboutprofit-maximizingpricingstrategies.Thetwo-parttariffprovidesanidealmeansbywhichcellularproviderscancaptureconsumersurplusandturnitintoprofit.Thetwo-parttariffworksbestwhenconsumershaveidenticalorverysimilardemands.EXAMPLE11.4PRICINGCELLULARPHONESERVICETABLE11.3CELLULARRATEPLANS(2011)ANYTIMEMINUNTESMONTHLYACCESSCHARGESNIGHT&WEEKENDMINUTESPER-MINUTERATEAFTERALLOWANCEA.VERIZON:AMERICA’SCHOICEBASIC450$39.99Unlimited$0.45900$59.99Unlimited$0.40Unlimited$69.99UnlimitedIncludedB.SPRINT:BASICTALKPLANS200$29.99Unlimited$0.45450$39.99Unlimited$0.45900$59.99Unlimited$0.40C.AT&TINDIVIDUALPLANS450$39.995000$0.45900$59.99Unlimited$0.40Unlimited$69.99UnlimitedIncludedEXAMPLE11.4PRICINGCELLULARPHONESERVICETABLE11.3CELLULARRATEPLANS(2011)(continued)ANYTIMEMINUNTESMONTHLYACCESSCHARGESNIGHT&WEEKENDMINUTESPER-MINUTERATEAFTERALLOWANCED.ORANGE(UK)100￡10.00None2.5pence200￡15.00None2.5pence300￡20.00None2.5penceE.ORANGE(ISRAEL)None28.00NISNone0.59NIS10038.00NISNone0.59NIS40061.90NISNone0.59NISF.CHINAMOBILE15058RMBNone0.40RMB450158RMBNone0.35RMB800258RMBNone0.32RMB1200358RMBNone0.30RMB1800458RMBNone0.25RMBToconverttheinternationalpricestoU.S.dollars(asofAugust2011),usethefollowingconversionfactors:1￡=$1.60,1NIS=$0.30,and1RMB=$0.13.TABLE11.3CELLULARRATEPLANS(2011)(continued)ANYTIMEMINUNTESMONTHLYACCESSCHARGESNIGHT&WEEKENDMINUTESPER-MINUTERATEAFTERALLOWANCEBundling11.5●

bundling Practiceofsellingtwoormoreproductsasapackage.Toseehowafilmcompanycanusecustomerheterogeneitytoitsadvantage,supposethattherearetwomovietheatersandthattheirreservationpricesforourtwofilmsareasfollows:Ifthefilmsarerentedseparately,themaximumpricethatcouldbechargedforWindis$10,000becausechargingmorewouldexcludeTheaterB.Similarly,themaximumpricethatcouldbechargedforGertieis$3000.Butsupposethefilmsarebundled.TheaterAvaluesthepairoffilmsat$15,000($12,000+$3000),andTheaterBvaluesthepairat$14,000

($10,000+$4000).Therefore,wecanchargeeachtheater$14,000forthepairoffilmsandearnatotalrevenueof$28,000.GONEWITHTHEWINDGETTINGGERTIE’SGARTERTheaterA$12,000$3000TheaterB$10,000$4000RelativeValuationsWhyisbundlingmoreprofitablethansellingthefilmsseparately?Becausetherelativevaluationsofthetwofilmsarereversed.Thedemandsarenegativelycorrelated—thecustomerwillingtopaythemostforWindiswillingtopaytheleastforGertie.Supposedemandswerepositivelycorrelated—thatis,TheaterAwouldpaymoreforbothfilms:Ifwebundledthefilms,themaximumpricethatcouldbechargedforthepackageis$13,000,yieldingatotalrevenueof$26,000,thesameasbyrentingthefilmsseparately.GONEWITHTHEWINDGETTINGGERTIE’SGARTERTheaterA$12,000$4000TheaterB$10,000$3000RESERVATIONPRICESFIGURE11.12Reservationpricesr1andr2fortwogoodsareshownforthreeconsumers,labeledA,B,andC.ConsumerAiswillingtopayupto$3.25forgood1andupto$6forgood2.CONSUMPTIONDECISIONSWHENPRODUCTSARESOLDSEPARATELYFIGURE11.13ThereservationpricesofconsumersinregionIexceedthepricesP1andP2forthetwogoods,sotheseconsumersbuybothgoods.ConsumersinregionsIIandIVbuyonlyoneofthegoods,andconsumersinregionIIIbuyneithergood.CONSUMPTIONDECISIONSWHENPRODUCTSAREBUNDLEDFIGURE11.14Consumerscomparethesumoftheirreservationpricesr1

+r2,withthepriceofthebundlePB.Theybuythebundleonlyifr1+r2isatleastaslargeasPB.RESERVATIONPRICESFIGURE11.15In(a),becausedemandsareperfectlypositivelycorrelated,thefirmdoesnotgainbybundling:Itwouldearnthesameprofitbysellingthegoodsseparately.In(b),demandsareperfectlynegativelycorrelated.Bundlingistheidealstrategy—alltheconsumersurpluscanbeextracted.MOVIEEXAMPLEFIGURE11.16ConsumersAandBaretwomovietheaters.ThediagramshowstheirreservationpricesforthefilmsGonewiththeWindandGettingGertie’sGarter.Becausethedemandsarenegativelycorrelated,bundlingpays.MIXEDVERSUSPUREBUNDLINGFIGURE11.17MixedBundling●

mixedbundling Sellingtwoormoregoodsbothasapackageandindividually.●

purebundling Sellingproductsonlyasapackage.Withpositivemarginalcosts,mixedbundlingmaybemoreprofitablethanpurebundling.ConsumerAhasareservationpriceforgood1thatisbelowmarginalcostc1,andconsumerDhasareservationpriceforgood2thatisbelowmarginalcostc2.Withmixedbundling,consumerAisinducedtobuyonlygood2,andconsumerDisinducedtobuyonlygood1,thusreducingthefirm’scost.Let’scomparethreestrategies:1.SellingthegoodsseparatelyatpricesP1=$50andP2=$90.

2.Sellingthegoodsonlyasabundleatapriceof$100.3.Mixedbundling,wherebythegoodsareofferedseparatelyatpricesP1=P2=$89.95,orasabundleatapriceof$100.TABLE11.4BUNDLINGEXAMPLEP1P2P3PROFITSoldseparately$50$90—$150Purebundling——$100$200Mixedbundling$89.95$89.95$100$229.90Asweshouldexpect,purebundlingisbetterthansellingthegoodsseparatelybecauseconsumers’demandsarenegativelycorrelated.Butwhataboutmixedbundling?MIXEDBUNDLINGWITHZEROMARGINALCOSTSFIGURE11.18Ifmarginalcostsarezero,andifconsumers’demandsarenotperfectlynegativelycorrelated,mixedbundlingisstillmoreprofitablethanpurebundling.

Inthisexample,consumersBandCarewillingtopay$20moreforthebundlethanareconsumersAandD.Withpurebundling,thepriceofthebundleis$100.Withmixedbundling,thepriceofthebundlecanbeincreasedto$120andconsumersAandDcanstillbecharged$90forasinglegood.TABLE11.5MIXEDBUNDLINGWITHZEROMARGINALCOSTSP1P2P3PROFITSoldseparately$80$80—$320Purebundling——$100$400Mixedbundling$90$90$120$420MIXEDBUNDLINGINPRACTICEFIGURE11.19Thedotsinthisfigureareestimatesofreservationpricesforarepresentativesampleofconsumers.Acompanycouldfirstchooseapriceforthebundle,PB,suchthatadiagonallineconnectingthesepricespassesroughlymidwaythroughthedots.ThecompanycouldthentryindividualpricesP1andP2.GivenP1,P2,andPB,profitscanbecalculatedforthissampleofconsumers.ManagerscanthenraiseorlowerP1,P2,andPBandseewhetherthenewpricingleadstohigherprofits.Thisprocedureisrepeateduntiltotalprofitisroughlymaximized.BundlinginPracticeEXAMPLE11.5THECOMPLETEDINNERVERSUSàLACARTE:ARESTAURANTPRICINGPROBLEMForarestaurant,mixedbundlingmeansoffering

bothcompletedinners(theappetizer,main

course,anddessertcomeasapackage)and

anàlacartemenu(thecustomerbuysthe

appetizer,maincourse,anddessertseparately).Thisstrategyallowstheàlacartemenutobe

pricedtocaptureconsumersurplusfromcustomerswhovaluesomedishesmuchmorehighlythanothers.Atthesametime,thecompletedinnerretainsthosecustomerswhohavelowervariationsintheirreservationpricesfordifferentdishes(e.g.,customerswhoattachmoderatevaluestobothappetizersanddesserts).EXAMPLE11.5THECOMPLETEDINNERVERSUSàLACARTE:ARESTAURANTPRICINGPROBLEMForarestaurant,mixedbundlingmeansofferingcompletedinnersandanàlacartemenu.Thisstrategyallowstheàlacartemenutobepricedtocaptureconsumersurplusfromcustomerswhovaluesomedishesmuchmorehighlythanothers.Successfulrestaurateursknowtheircustomers’demandcharacteristicsandusethatknowledgetodesignapricingstrategythatextractsasmuchconsumersurplusaspossible.TABLE11.6

MIXEDBUNDLINGATMCDONALD’S(2011)INDIVIDUALITEMPRICEMEAL(INCLUDESSODAANDFRIES)UNBUNDLEDPRICEPRICEOFBUNDLESAVINGSChickenSandwich$5.49ChickenSandwich$10.07$7.89$2.18Filet-O-Fish$4.39Filet-O-Fish$8.97$6.79$2.18BigMac$4.69BigMac$9.27$6.99$2.28QuarterPounder$4.69QuarterPounder$9.27$7.19$2.08DoubleQuarterPounder$6.09DoubleQuarterPounder$10.67$8.39$2.2810-pieceChickenMcNuggets$5.1910-pieceChickenMcNuggets$9.77$7.59$2.18LargeFrenchFries$2.59LargeSoda$1.99Tying●

tying Practiceofrequiringacustomertopurchaseonegoodinordertopurchaseanother.Whymightfirmsusethiskindofpricingpractice?Oneofthemainbenefitsoftyingisthatitoftenallowsafirmtometerdemandandtherebypracticepricediscriminationmoreeffectively.Tyingcanalsobeusedtoextendafirm’smarketpower.Tyingcanhaveotheruses.Animportantoneistoprotectcustomergoodwillconnectedwithabrandname.Thisiswhyfranchisesareoftenrequiredtopurchaseinputsfromthefranchiser.EFFECTSOFADVERTISINGFIGURE11.20Advertising11.6ARandMRareaverageandmarginalrevenuewhenthefirmdoesn’tadvertise,andACandMCareaverageandmarginalcost.ThefirmproducesQ0andreceivesapriceP0.Itstotalprofitπ0isgivenbythegray-shadedrectangle.Ifthefirmadvertises,itsaverageandmarginalrevenuecurvesshifttotheright.Averagecostrises(toAC′)butmarginalcostremainsthesame.ThefirmnowproducesQ1(whereMR′=MC),andreceivesapriceP1.Itstotalprofit,π1,isnowlarger.ThepricePandadvertisingexpenditureAtomaximizeprofit,isgivenby:Thefirmshouldadvertiseuptothepointthat=

fullmarginalcostofadvertising(11.3)Advertisingleadstoincreasedoutput.

Butincreasedoutputinturnmeansincreasedproductioncosts,andthismustbetakenintoaccountwhencomparingthecostsandbenefitsofanextradollarofadvertising.

First,rewriteequation(11.3)asfollows:NowmultiplybothsidesofthisequationbyA/PQ,theadvertising-to-salesratio.●advertising-to-salesratioRatioofafirm’sadvertisingexpenditurestoitssales.●advertisingelasticityofdemandPercentagechangeinquantitydemandedresultingfroma1-percentincreaseinadvertisingexpenditures.ARuleofThumbforAdvertising

(11.4)

EXAMPLE11.6ADVERTISINGINPRACTICEConveniencestoreshavelowerpriceelasticitiesof

demand(around?5),buttheiradvertising-to-sales

ratiosareusuallylessthanthoseforsupermarkets

(andareoftenzero).Why?

Becauseconveniencestoresmostlyservecustomerswholivenearby;theymayneedafewitemslateatnightormaysimplynotwanttodrivetothesupermarket.Advertisingisquiteimportantformakersofdesignerjeans,whowillhaveadvertising-to-salesratiosashighas10or20percent.Laundrydetergentshaveamongthehighestadvertising-to-salesratiosofallproducts,sometimesexceeding30percent,eventhoughdemandforanyonebrandisatleastaspriceelasticasitisfordesignerjeans.Whatjustifiesalltheadvertising?Averylargeadvertisingelasticity.EXAMPLE11.6ADVERTISINGINPRACTICETABLE11.7SALESANDADVERTISINGEXPENDITURESFORLEADINGBRANDSOFOVER-THE-COUNTERDRUGS(INMILLIONSOFDOLLARS)SALESADVERTISINGRATIO(%)PainMedicationsTylenol855143.817Advil36091.726Bayer17043.826Excedrin13026.721AntacidsAlka-Seltzer16052.233Mylanta13532.824Tums13527.620EXAMPLE11.6ADVERTISINGINPRACTICETABLE11.4SALESANDADVERTISINGEXPENDITURESFORLEADINGBRANDSOFOVER-THE-COUNTERDRUGS(INMILLIONSOFDOLLARS)(continued)SALESADVERTISINGRATIO(%)ColdRemedies(decongestants)Benadryl13030.924Sudafed11528.625CoughMedicineVicks35026.68Robitussin20537.719Halls13017.413TABLE11.7SALESANDADVERTISINGEXPENDITURESFORLEADINGBRANDSOFOVER-THE-COUNTERDRUGS(INMILLIONSOFDOLLARS)(continued)AppendixtoChapter11TheVerticallyIntegratedFirm●horizontalintegration Organizationalforminwhichseveralplantsproducethesameorrelatedproductsforafirm.●verticalintegration Organizationalforminwhichafirmcontainsseveraldivisions,withsomeproducingpartsandcomponentsthatothersusetoproducefinishedproducts.●transferprices Internalpricesatwhichpartsandcomponentsfromupstreamdivisionsare“sold”todownstreamdivisionswithinafirm..WhyVerticallyIntegrate?MarketPowerandDoubleMarginalizationHowdofirmsalongaverticalchainexercisesuchmonopolypower,andhowarepricesandoutputaffected?Wouldthefirmsbenefitfromaverticalmergerthatintegratesanupstreamandarelateddownstreambusiness?Wouldconsumers?Supposeanenginemanufacturerhasmonopolypowerinthemarketforengines,andanautomobilemanufacturerthatbuystheseengineshasmonopolypowerinthemarketforitscars.Wouldthismarketpowercausethesetwofirmstobenefitinanywayiftheyweretomerge?Wouldconsumersofthefinalproduct—automobiles—bebetterorworseoffifthetwocompaniesmerged?Whenthereismarketpowerofthissort,averticalmergercanbebeneficialtothetwofirms,andalsobeneficialtoconsumers.SEPARATEFIRMSSupposeamonopolistproducerofspecialtyenginesproducesthoseenginesataconstantmarginalcostcE,andsellstheenginesatapricePE.Theenginesareboughtbyamonopolistproducerofsportscars,whichsellsthecarsatthepriceP.Demandforthecarsisgivenby

(A11.1)

withtheconstantA>cE.Ifthetwocompaniesareindependentofeachother,theautomobilemanufacturerwilltakethepriceofenginesasgiven,andchooseapriceforitscarstomaximizeitsprofits:

(A11.2)

YoucancheckthatgivenPE,theprofitmaximizingpriceofcarsis:

(A11.3)

SEPARATEFIRMSThenthenumberofcarssoldandtheautomobilecompany’sprofitare:

(A11.4)

and

(A11.5)

Whatabouttheenginemanufacturer?Itchoosesthepriceofengines,PE,tomaximizeitsprofit:

(A11.6)

Youcanconfirmthattheprofit-maximizingpriceofenginesis:

Theprofittotheenginemanufactureristhenequalto:InEquation(A11.5),substituteforthepriceofenginesfromequation(A11.7).Youwillseethattheautomobilecompany’sprofitisthen:

(A11.9)

(A11.7)

(A11.8)

Also,thepricepaidbyconsumersis:

(A11.11)

Hencethetotalprofitforthetwocompaniesis:

(A11.10)

VERTICALINTEGRATIONNowsupposethattheenginecompanyandtheautomobilecompanymergetoformaverticallyintegratedfirm.Themanagementofthisfirmwouldchooseapriceofautomobilestomaximizethefirm’sprofit:(A11.12)

whichyieldsaprofitof:

(A11.13)

Theprofit-maximizingpriceofcarsisnow:(A11.14)

Observethattheprofitfortheintegratedfirmisgreaterthanthetotalprofitforthetwoindividualfirmsthatoperateindependently.Furthermore,thepricetoconsumersforautomobilesislower.DOUBLEMARGINALIZATION●doublemarginalization Wheneachfirminaverticalchainmarksupitspriceaboveitsmarginalcost,therebyincreasingthepriceofthefinalproduct.EXAMPLEOFDOUBLEMARGINALIZATIONFIGUREA11.1(1of2)Fortheautomobilecompany,themarginalrevenuecurveforcarsisthedemandcurveforengines(thenetmarginalrevenueforengines).Correspondingtothatdemandcurveistheenginecompany’smarginalrevenuecurve,MRE.Iftheenginecompanyandautomobilecompanyareseparateentities,theenginecompanywillproduceaquantityofenginesQ′E

atthepointwhereitsmarginalrevenuecurveintersectsitsmarginalcostcurve.Theautomobilemakerwillbuythoseenginesandproduceanequalnumberofcars.Hence,thepriceof