Traditionalmodelofentrydeterrence1Incumbentthreatenstomakeentryunpro tableforentrant Traditionalmodelofentrydeterrence2ProblemincumbentsthreatisnotcredibleFollowingthroughisnotaNashequilibriumstra ID: 362143
Download Pdf The PPT/PDF document "EntryandIncumbentAdvantageTraditionalmo..." is the property of its rightful owner. Permission is granted to download and print the materials on this web site for personal, non-commercial use only, and to display it on your personal computer provided you do not modify the materials and that you retain all copyright notices contained in the materials. By downloading content from our website, you accept the terms of this agreement.
EntryandIncumbentAdvantageTraditionalmodelofentrydeterrence:incumbent'sthreatisnotcredible.DixitmodelofentrydeterrenceOtherentrydeterrencemethods:1.Predatorypricing2.Raisingrivals'costs Traditionalmodelofentrydeterrence1Incumbentthreatenstomakeentryunpro tableforentrant: Traditionalmodelofentrydeterrence2Problem:incumbent'sthreatisnotcredible.FollowingthroughisnotaNashequilibriumstrategyofthesubgamewhen rm2enters.Note:fFight,Don'tEntergisaNE,however!Modelswhichhavesubgameperfectequilibriainwhichentryisdeterred.Modelshavetwofeatures:1.Advantage:incumbentmoves rst2.Commitment:incumbentmustfollowthroughwithitsthreatregardlessofitsrival'saction.Focusonhowincumbentcanmakehisthreatscredible. Dixitmodel1Mainidea:Incumbentcanbuildcapacity(sunkcost)todeterentryChangethemodelslightly:1.Entrycanbedeterredifcapacitylevelk1existswherec2(k1)0.2.However,ifm(k1)is\toolow"(i.e.,deterrenceistooexpensive),incumbentprefersnottodeter. Dixitmodel2Marketdemand:p=a bQ,whereQ=q1+q2.Assumethatproductionrequirestwoinputs:labor(atwperunit)andcapacity(atrperunit).Firm1'sadvantageliesinabilitytoinstallcapacityk1beforeentrantFirm1'stotalcosts(ifitinstalledk1unitsofcapacity):C(q)=(rk1+wq+r(q k1)ifq�k1rk1+wqifqk1(1)Hence, rm1'smarginalcostsarelowerwhenproducingundercapacity(graphMC).Firm2'scostsofenteringmarketandproducing:C(q)=(w+r)q+F(2)Advantage: rm2hasnochancetobuildcapacitybeforecompeting.Credible:lookforSPEwhereentryisdeterred.Solvebackwards:1.Subgamewhere rm2enters;incumbentshasinstalledk1capacityunits2.Subgamewhere rm2decideswhethertoenter3.Subgamewhere rm1decidescapacitylevelk1 Dixitmodel3:Subgamewhere rm2entersCournotquantity-settinggameafter rm1hasinstalledk1capacityunits.Firm2'sbest-responsefunction:BR2(q1)=a (w+r) bq12bFirm1'sbest-responsefunction:maxq11=((a b(q1+q2) w r)q1ifq1k1(a b(q1+q2) w)q1 rk1ifq1k1!BR1(q2)=8a (w+r) bq22bifq1k1a w bq22bifq1k1Graph.Bybuildingcapacity, rm1commitstoproducingmoreoutput;itishisbest-responsetoproducethesehigherlevelsofoutputIfcapacitycostsnotsunk(irretrievable),this\threat"isnotcredible.Youcanbuildanddismantlecapacityastheneedarises,sothatBR1(q2)=a (w+r) bq22b. Dixitmodel4:Subgamewhere rm2entersNE:intersectionofbest-responsecurves(q1;q2)Firm1willneversetq16=k1:1.ifq1k1:wastedcapacity2.ifq1�k1:canjustbuildk1=q1tobeginwithSoinNE:q1=k1,andq2=BR2(k1)=a (w+r) bk12bQ=a (w+r)+bk12b;p=a+(w+r) bk12Pro ts:1(k1)=k1a (w+r) bk122(k1)=(a (w+r) bk1)24b F Dixitmodel5:subgamewhere rm2decideswhethertoenterWehavesolvedforNEofsubgameif rm2enters.Nowgobackwards:whenwill rm2enter?Firm2entersonlywhen2=(a (w+r) bk1)24b F�0ork1a (w+r) 2pbFbql1:ql1isthelimitquantity.Graph.Whenisentrydeterrencepossible?1.Ifql1�qa,entryalwaysoccurs.BR1andBR2cannotcrossatql1,sonoequilibriumexistsinwhich rm1canproduceql1.Firm2willalwaysenter.Entryaccommodation.2.Ifql1qc, rm2neverenters,sinceitnevermakespositivepro tseveninsymmetricCournotoutcome.Blockadedentry.3.Entrydeterrencepossibleonlywhenqcql1qa.Restrictourselvestothiscase. Dixitmodel6:Subgamewhere rm1choosesk1Nowgobackwardsagain:whatk1will rm1choosein rststage?Twopossibilities:1.Installk1=ql1,deterentry,andgetmonopolypro tsm(ql1)=1(BR1(0);0;ql1).(graph)2.Allowentryandbuildcapacityk16=ql1.Getpro tsof1(k1;BR2(k1)).BestalternativetobeingmonopolististobeStackelbergleader,sok1=qs1.Nowwill rm1prefertosetcapacityequaltoql1orks1?CompareStackelbergpro ts1(k1;BR2(k1))tomonopolypro ts1(BR1(0);0;ql1).Chooseql1anddeterif1(BR1(0);0;ql1)�1(k1;BR2(k1)).Note:Ifqs1�ql1,prefertodeterentry.Summary:Giventhatqcql1qa,SPEspeci es:1.For rm1,installql1if1(BR1(0);0;ql1)�1(k1;BR2(k1)),andqs1otherwise2.For rm2,enterifk1ql1 Dixitmodel7Dixitmodelisoneexampleoflimitpricing:committingtoapriceoroutputlevelsothatentrant ndsitunpro tabletoentermarket.Otherexamplesofbehaviorthatincumbentengagesintomaintainadvantage:1.Predatorypricing:loweringpricetodriverivalsoutofmarket2.Raisingrivals'costs.Closelyrelatedto\poisonpills". Predatorypricing1Incumbentpricesbelowcompetitor'scost,anddrivesitoutofbusiness.Single-periodcase(gametree):identicalto\traditional"limitpricingmodel,exceptthatentrantisalreadyinthemarket.1.Formally:incumbentthreatenstoproducetokeepmarketpriceat(say)P,belowrival'sAC.2.Ifentrantbelievesthis,itispricetakerandproducesonitsMCcurve,atq(e).Incumbentmustproduceq(i)todepresspricetoP.3.Graph,assumingidentical rms.Incumbentsu erslargerlossesthanrival!Withidentical rms,predationisnotlikelytobecrediblethreat Illustration:PredatoryPricing(Carlton/Perlo )q(e)q(i)q(e)p*ABCDAC(1)AC(2)DFMC Predatorypricing2Somewaysincumbentcanhaveadvantagewhichmakespredationthreatcredible:Sizedi erences:Largerincumbent rmhasaccesstofundswhichsmalledrivaldoesn't.Canmakepredationapreferredstrategyinthelong-term.Imperfectinformation:uncertaintyaboutincumbent'scosts.Graph.Ifincumbent'scostsareAC(3),thenevenatquantityq(i)itismakingpositivepro t.Butifincumbentreallyhaslowercost,entrantshouldn'tbeinthemarkettobeginwith! Raisingrivals'costs1Incumbentmaydeterentry(ordriverivalsout)byactivitieswhichraiseitsrivals'costsofproduction.Incumbentadvantagealreadyassumed:diculttodisentanglecompetitivebusinesspracticesofdominant rmmalignantbehaviortowardsrivalsExample:MicrosoftforcesPCmanufacturerswhopre-installWindowsOStobundleitwithInternetExplorer.Raisesitsrival's(Netscape)sellingcostsButisthiscompetitivebusinesspractice,ormalignantbehavior? Raisingrivals'costs2Generalschematicdiagram.Ingeneral,foraRRCstrategytobecredible,youneedm C�d. Raisingrivals'costs3Governmentregulation:quotasversestari s.Industrymaypreferquotasbecauselimitednumberofimportlicenseslockspotentialrivalsoutofmarket.\Sleepingpatents":incumbenthasmoreincentivetoinventandpatent(butneverproduce)potentialsubstitutestoitsproduct.Preemptive-innovation.Isbuyingoutsmall rmsarelatedphenomenon?Raiseconsumerswitchingcosts:frequent\ryermiles,preferredcustomercards,etc.Rivalsmustpricelowertoovercomeconsumers'brandloyalty. SomeconclusionsTwocomponentsofentrydeterrencemodel:1.Incumbentadvantage:incumbentcantakeactionbeforerivalreacts2.Credibility:incumbent'sthreatmustbeaNEofthesubgamewhichfollowsrival'sentryLimitPricingandPredatoryPricingRaisingRival'sCosts:m C�d.