scieee AI-readable full text Open interactive document viewer

Endogenous timing and the choice of quality in a vertically differentiated duopoly

Lambertini, Luca

Abstract

EconStor is a publication server for scholarly economic literature, provided as a non-commercial public service by the ZBW.

Full text

Lambertini, Luca Working Paper Endogenous timing and the choice of quality in a vertically differentiated duopoly Quaderni - Working Paper DSE, No. 294 Provided in Cooperation with: University of Bologna, Department of Economics Suggested Citation: Lambertini, Luca (1997) : Endogenous timing and the choice of quality in a vertically differentiated duopoly, Quaderni - Working Paper DSE, No. 294, Alma Mater Studiorum - Università di Bologna, Dipartimento di Scienze Economiche (DSE), Bologna, https://doi.org/10.6092/unibo/amsacta/5016 This Version is available at: https://hdl.handle.net/10419/159137 Standard-Nutzungsbedingungen: Die Dokumente auf EconStor dürfen zu eigenen wissenschaftlichen Zwecken und zum Privatgebrauch gespeichert und kopiert werden. Sie dürfen die Dokumente nicht für öffentliche oder kommerzielle Zwecke vervielfältigen, öffentlich ausstellen, öffentlich zugänglich machen, vertreiben oder anderweitig nutzen. Sofern die Verfasser die Dokumente unter Open-Content-Lizenzen (insbesondere CC-Lizenzen) zur Verfügung gestellt haben sollten, gelten abweichend von diesen Nutzungsbedingungen die in der dort genannten Lizenz gewährten Nutzungsrechte. Terms of use: Documents in EconStor may be saved and copied for your personal and scholarly purposes. You are not to copy documents for public or commercial purposes, to exhibit the documents publicly, to make them publicly available on the internet, or to distribute or otherwise use the documents in public. If the documents have been made available under an Open Content Licence (especially Creative Commons Licences), you may exercise further usage rights as specified in the indicated licence. https://creativecommons.org/licenses/by-nc/3.0/ EndogenousTimingand theChoiceof QualityinaVerticallyDi®erentiated Duopoly¤ LucaLambertini DepartmentofEconomics UniversityofBologna StradaMaggiore45 I-40125 Bologna,Italy fax39-51-6402664 e-mail [email protected] September28,1998 ¤ Acknowledgements.ThispaperwaswrittenwhenIwasat theInstituteofEconomics, Abstract The endogenouschoice oftimingisdiscussedinaverticallydi®erentiated duopolywherequalityimprovementrequiresa¯xedconvexcost.The timingdecisionconcernsthequalitystage.Using anextendedgamewith observabledelay, itis shownthatonlysimultaneousequilibriacanarise. ThisputsintoquestiontheabilityofStackelberg gamestodescribethe entryprocess. J.E.L.classi¯cation:D21,D43,L13 Keywords:productquality,R&Dinvestment,timing,extendedgame 1. Introduction Anestablishedresultinthetheoryofoligopolyunderverticaldi®erentiationstates that thehigh-qualityearnshigherpro¯tsthanlow-qualityrivals(see,interalia, Gabszewicz and Thisse,1979,1980;Shakedand Sutton,1982,1983;Donnenfeld and Weber,1995).Ageneralproofofthisresultforeveryconvex¯xed-cost functionofqualityimprovementisprovided byLehmann-Grube(1997).Aoki and Prusa(1997)adoptaspeci¯c caseofthe costfunctionanalysed byLehmann- Grube,toinvestigatethe consequenceson pro¯ts,consumersurplusand social UniversityofCopenhagen.Ithankanassociate editor,ananonymousreferee and Dan Sasaki forhelpfulcomments.Theusualdisclaimerapplies. 2 welfareofthetiming ofinvestmentin productqualityinaverticallydi®erentiated duopolywherethemarketstageisplayedintheprice space.Theyshowthat (a) sequentialplayinducesboth ¯rmstoinvestless in productquality,ascompared towhat theywouldiftheyweretoplaysimultaneously;(b)undersequentialplay, theleaderpreferstosupplythehigh-qualitygood;and (c)industrypro¯tsare higherundersequentialplaythan undersimultaneousplay,whiletheopposite obviouslyholdsforsocialwelfare.Hence,theauthors suggest that¯rmsmay enjoycoordination bene¯tsundersequentialqualitychoice.Though,since the ¯rmsupplyingthehigh-qualitygood prefers sequentialplaytosimultaneousplay, whiletheoppositeappliestothe¯rmproducingthelow-qualitygood,¯rmscan coordinateoversequentialplayifeither (i)asidepaymentispossible;or (ii)¯rms discoverand introduce newvarietiesat random.Inthelatterperspective,the authorsidentifyaprobabilityinterval inwhich both ¯rmsprefer thesequential movegame. Althoughsuggestiveand indeedacceptable,neitherofthetworoutesappear fullysatisfactory,fordi®erent reasons.Thepossibilityofasidepayment raisesthe issueofopportunisticbehaviouronthepart ofboth ¯rms.Astotheprobabilistic approach,onewouldliketoexploititfullyfromtheoutset,and analysethe gameasastochasticR&Drace.Intheoriginalpaper,qualitiesareproduced 3 inadeterministic environment.Moreover,randomsuccess inR&Dactivityputs stronglyintoquestion point (b).Ifinnovationis stochastic,thereisnoreasonto believea priorithat theleaderwill supplythehigh-qualitygood.Asanexample, consider the carindustry.Aninnovationintheformofanewfamilycarinthe rangeofstationwagonsmaybe easier to accomplishthananew°agshipsportcar, sothat theleadermaywell end up withthelow-qualitygood.Thispointsto apossibletradeo®betweenearlyinnovationand theattainmentofadominant positioninthemarket (see Dutta,Lachand Rustichini, 1995).1 Mypurposeistoproceedtoaselection betweensimultaneousand sequential qualitychoice byembeddingqualitycompetitioninto anextendedgamewith observabledelay(Hamiltonand Slutsky,1990;HS henceforth),whereplayers(say, ¯rms)arerequiredtosetboththeactualmovesoractionsinthequalityspace, and thetimeatwhichsuchactionsaretobeimplemented.When ¯rmschooseto actatdi®erent times,sequentialequilibria obtain,whileiftheydecidetomove at thesametime,simultaneousNashequilibria areobserved.Forsequentialplay toemergeat thesubgameperfectequilibriumofthe extendedgame, itsoutcome mustatleastweaklyPareto-dominatetheoutcomeassociatedwithsimultaneous 1Fortheanalysisoftheincentiveforthelow-quality¯rmtowardsleapfrogging,see Rosenkranz(1997). 4 moves.The choice oftiming occursinapreplaystagewhich doesnot takeplace inrealtime,sothat thereisnodiscounting associatedwithwaiting and payo®s arethesamewhether¯rmschoosetomoveas soonaspossibleor theydelayas long astheycan.Thedecisiontoplayearlyoratalater timeisnotsu±cientper setoyieldsequentialplay,since ananalogousdecisiontaken bytherival leads tosimultaneousplay.Inthepresentsetting,anotherextensioncan be envisaged, namely,theone concerningthe choice between highand lowquality, i.e., the relativepositionsof¯rmsintheproductspace.Therefore,Ishall describetwo alternativesymmetric extendedgames,resultingfromthepermutationofthe stagesdescribingthetimingchoice and thelocationof¯rmsalongthequality spectrum,respectively.It turnsout that thesubgameperfectequilibriumofboth gamesentails simultaneousplayat thequalitystage,sothatStackelberg outcomes areruledout.Thisresult,whichappearstobeincontrastwith ¯rms'behaviourin therealworld,maycastashadowonthereliabilityofthisapproachasastylised descriptionoftheprocess ofinnovationand entry. Theremainderofthepaperisstructuredasfollows.Thebasicmodelofvertical di®erentiationis summarizedinsection2.Section3describestheextendedgames. Finally,section4providesconcludingremarks. 5 2. Theverticaldi®erentiationmodel HereIshall brie°yresumethebasicsetting.Twosingle-product¯rms, labelledas 1 and 2,produce goodsofdi®erentqualities,q1and q22[0;1),throughthesame technology,C(qi)=kq2 i;withk>0:Thiscan beinterpretedas¯xedcostdue totheR&De®ort neededtoproduce acertainquality,whilevariableproduction costsareassumedaway.Productsareo®eredonamarketwhere consumershave unitdemands,and buyifand onlyifthenetsurplusfromconsumptionvµ(qi;pi)= µqi¡pi¸0;wherepiistheunitprice ofthegoodofqualityqi,purchased bya genericconsumerwhosemarginalwillingness topayisµ2[0;µ]:2Iassumethatµis uniformlydistributedwith densityoneoversuchinterval, sothat thetotalmass of consumerisµ.Firmscompeteintwostages,the¯rstbeingplayedinthequality space,thesecond intheprice space.ImaintainAokiand Prusa'sassumption thatdownstreamBertrand competitionis simultaneous.Hence,proceedingby backwardinduction,thepro¯tfunctionof¯rm1 at the¯rststagelooksasfollows (cf.Aokiand Prusa,1997:Lemma 1,p.106): ¼1(q1;q2)=4µ2q2 1(q1¡q2) (4q1¡q2)2¡kq2 1ifq1¸q2;(2.1) 2Aokiand Prusa(1997:p.106)normalisethesupport to[0,1]. 6 ¼1(q1;q2)=µ2q1q2(q2¡q1) (4q2¡q1)2¡kq2 1ifq1<q2:(2.2) AslightgeneralizationofAokiand Prusa'sresultsintermsofequilibrium qualities,pro¯tand socialwelfare can begivenonthebasisof(1)and (2).Depending onthe entrysequence,¯rmswill alternativelytakethehigh-quality(H)or the low-quality(L)position.Table1summarisestheoutcomesassociatedwithsi- multaneousplay(n),sequentialplaywiththehigh-quality¯rmleading(Hl),and sequentialplaywiththelow-quality¯rmleading(Ll),respectively.3 INSERT TABLE1HERE Observethat ¼Hl H>¼n H>¼Ll H;¼n L>¼Ll L>¼Hl L8µ;k:(2.3) Theaboveinequalitiesimplythat¯rms'choice oftherespectiverolesisindependentoftheratiobetweenthesize ofthemarketand theamountof¯xedcosts. Moreover,onthebasisofsocialwelfarelevelsinthethree cases, itisworthem- 3Aproof forthe casewherek=1=2isgiven byMotta(1993:pp.116-117).Itsgeneralisation isamatterofstraightforwardcalculations.Moreover,theabsolutelevelofkisunimportant, in that themainresultsholdindependentlyoftherelativesize ofkand µ: 7 phasizingthatapublicagencytakingcareofsocialsurpluswithoutdistributive concerns shouldfavour theleadership bythelow-quality¯rm,whiletheleader alwaysprefer toenterwithasuperiorquality.Thisdi®erence inthestructure ofpreferencescan betraced backtothefact that,whenleading,thelow-quality ¯rmprovidesahigherqualitythaninanyothersituation. 3. The extendedgamewithobservable delay Consider thefollowing generalizationofthe extendedgameproposed byHS(1990: p.32),where¯rmscansetasinglestrategicvariableinthebasicgameand must choosebetweenmoving¯rstorsecond.4Here,givensimultaneousplayintheprice stage,thebasicstagegameisplayedinthequalityspace,and twoextensionsare accountedfor.De¯neas¡=(I;-;§;T;Q;¦) the extendedgamewithobservable delay,wherealldecisionsaretaken non-cooperatively.Thesetofplayers(or¯rms) isI=f1;2g,and q1and q22Q=[0;1)aretheintervalsofRrepresentingthe actionsavailabletoplayers1 and 2inthebasicgame.Payo®s(pro¯tlevels)¼1 and ¼22¦depend ontheactionsundertakeninthequalitystage,sothat¼1and 4HS(1990)alsoproposeanextendedgamewithactioncommitment,where each playermust commit toaspeci¯cactionirrespectivelyoftherivaltryingtoleadorfollow.The equilibrium ofsucha gameisneverunique; in particular, italwaysallowsforsequentialmoves(see their TheoremVII,p.42). 8 References [1]Aoki, R.&Prusa,T.(1997).Sequentialversus simultaneouschoice withendogenousquality.InternationalJournalof Industrial Organization,15,103- 121. [2]Boom,A.(1995).Asymmetricinternationalminimum qualitystandardsand verticaldi®erentiation.Journalof IndustrialEconomics,43,101-19. [3]Crampes,C.&Hollander,A.(1995).Duopolyand qualitystandards.EuropeanEconomicReview,39,71-82. [4]Donnenfeld,S.&Weber,S.(1995).Limitqualitiesand entrydeterrence. RANDJournalof Economics,26,113-30. [5]Dutta,P.K., Lach,S.&Rustichini, A.(1995).Betterlatethanearly:vertical di®erentiationintheadoptionofanewtechnology.Journalof Economicsand ManagementStrategy,4,563-589. [6]Ecchia,G.&Lambertini, L.(1997).Minimum qualitystandardsand collusion.Journalof IndustrialEconomics,45,101-113. 15 [7]Gabszewicz,J.J.&Thisse,J.-F.(1979).Price competition,qualityand incomedisparities.Journalof EconomicTheory,20,340-359. [8]Gabszewicz,J.J.&Thisse,J.-F.(1980).Entry(and exit) inadi®erentiated industry.Journalof EconomicTheory,22,327-38. [9]Hamilton,J.&Slutsky,S.(1990).Endogenoustimingin duopolygames: Stackelberg orCournotequilibria.GamesandEconomicBehavior,2,29-47. [10]Lambertini, L.(1996).Choosingrolesinaduopolyforendogenouslydi®erentiated products.AustralianEconomicPapers,35,205-24. [11]Lehmann-Grube,U.(1997).Strategic choice ofqualitywhenqualityiscostly: thepersistence ofthehigh-qualityadvantage.RANDJournalof Economics, 28,372-84. [12]Motta,M.(1993).Endogenousqualitychoice:price vsquantitycompetition. Journalof IndustrialEconomics,41,113-32. [13]Ronnen,U.(1991).Minimumqualitystandards,¯xedcosts,and competition. RANDJournalof Economics,22,490-504. [14]Rosenkranz,S.(1997).Qualityimprovementsand theincentivetoleapfrog. InternationalJournalof Industrial Organization,15,243-61. 16 [15]Shaked,A.&Sutton,J.(1982).Relaxingprice competitionthrough product di®erentiation.Review of EconomicStudies,49,3-13. [16]Shaked,A.&Sutton,J.(1983),Naturaloligopolies.Econometrica,51,1469- 83. 17 nHl Ll qn H=0:126655µ4=k qHl H=0:122232µ2=k qLl H=0:12667µ2=k qn L=0:024119µ2=k qHl L=0:023894µ2=k qLl L=0:024197µ2=k ¼n H=0:012219µ4=k ¼Hl H=0:012235µ4=k ¼Ll H=0:0122069µ4=k ¼n L=0:000764µ4=k ¼Hl L=0:000757µ4=k ¼Ll L=0:0007637µ4=k SWn=0:034592µ4=k SWHl=0:034008µ4=k SWLl=0:034602µ4=k Table 1 18 2 FS HLHL FH¡;¡¼n H;¼n L¡;¡¼Hl H;¼Hl L 1L¼n L;¼n H¡;¡¼Ll L;¼Ll H¡;¡ SH¡;¡¼Ll H;¼Ll L¡;¡¼n H;¼n L L¼Hl L;¼Hl H¡;¡¼n L;¼n H¡;¡ Matrix 1 19 2 FS 1F0:064915µ4=k;0:064915µ4=k 0:064993µ4=k;0:064819µ4=k S0:064819µ4=k;0:064993µ4=k 0:064915µ4=k;0:064915µ4=k Matrix 1b 20 2 HL FSFS HF¡;¡ ¡;¡¼n H;¼n L¼Hl H;¼Hl L 1S¡;¡ ¡;¡¼Ll H;¼Ll L¼n H;¼n L LF¼n L;¼n H¼Ll L;¼Ll H¡;¡ ¡;¡ S¼Hl L;¼Hl H¼n L;¼n H¡;¡ ¡;¡ Matrix 2 21 2 HL 1H¡;¡¼n H;¼n L L¼n L;¼n H¡;¡ Matrix 2b 22