[PDF] Chapter 17 Network Effects - Cornell University

Network Effects - Research Partnership

the network effect concerns all the agents in the same manner But there areother types of network effects that Katz and Shapiro (1985) have called indirect Indeed, in some situations, the number of users does not directly affect the utility derived by users but only affects them via the number of services or complementary products available

Network Effects and - Boston University

Back in the 1990s, network effects were looked to as a new source of potential market power, especially in digital mar-kets Economists use “network effects” to describe contexts in which a good or service offers increasing benefits the more users it has Network effects can be direct—for example, a fax

Chapter 17 Network Effects - Cornell University

17 1 THE ECONOMY WITHOUT NETWORK EFFECTS 511 consumers as the set of all real numbers in the interval strictly between 0 and 1 That is, each consumer is named by a different real number, and the total mass of consumers is 1

Network Effects and Social Inequality

Social network: a set of actors (nodes) connected by a set of relations (ties or edges) Network effect: effect on ego of alters’ behavior Diffusion: the process by which a practice spreads throughout (is adopted by) members of a population Homophily: the tendency of actors in a network to form ties to alters who are similar to themselves

The Power of Bitcoin’s Network Effect

Network Effect GREG CIPOLARO ROSS STEVENS, PH D Global Head of Research Founder & Executive Chairman EECUTIVE SUMMARY Q Bitcoin’ provin twor undamentals, increa invest iods, reliabl ompellin g-ter vestment Q Adopt Bitc a all-t high, wit t addresse tha bitc 25 M Q Engag apidl ising, it ail tiv c esse approachin 00,000

INDIRECT NETWORK EFFECTS AND THE PRODUCT CYCLE: VIDEO GAMES

2The direct network effect model is most appropriate for something like a telephone network As more consumers use telephones, the value of the telephone to an individual consumer increases because it is possible to call more people It is as if the quality of the telephone is increasing in the number of consumers 516 MATTHEWT CLEMENTS AND

Systems Competition and Network Effects

In a communications network, such as a network of electronic-mail users or a network of people who exchange WordPerfect files, each user desires to link directly to other users 2 Consequently, as has long been recognized, the demand for a network good is a function of both its price, and the expected size of the network

How Good is Your Network? The Effect of Network Quality on

network actors on the indicators of innovativeness, dependability, trustworthiness, commitment, reputation, and overall satisfaction Trust is particularly important to network quality because networks are built on trust (O‟Toole 1997) Individual network actors that posses a large amount of these traits may be consider of high quality

[PDF] neurone et fibre musculaire la communication nerveuse correction

[PDF] neurone et fibre musculaire la communication nerveuse corrigé

[PDF] neurone et fibre musculaire la communication nerveuse cours

[PDF] neurone et fibre musculaire la communication nerveuse qcm

[PDF] neurone et fibre musculaire la communication nerveuse terminale s

[PDF] Neurones cerveau humain avec 100 milliards de neurones

[PDF] Neutralisation et espèces chimiques

[PDF] neutralité carbone 2050

[PDF] neutralité carbone définition

[PDF] neutre ou acide

[PDF] new array php

[PDF] new bridges 1ere 2011 corrigé

[PDF] new bridges terminale nathan 2012 corrigé

[PDF] new deal

[PDF] new enjoy english 6e - workbook pdf

Chapter17

NetworkE

ff ects Fromthebook Networks,Crowds, andMarkets:Reasoningab outaHighly Connecte dWorld. ByDavid EasleyandJonKleinb erg.Cambridge University Press,2010. Completepreprint on-lineathttp://www.cs.cornell.edu/home/kle inber/net works-b ook/ Atthebeginningof Chapter16,w ediscussedtwofundamen tallydi ff erentreasonswh y individualsmight imitatethebehavior ofothers. Onereasonw asbasedoninformational e ff ects:sincethe behavior ofotherp eopleconveysinformationaboutwhatthey know, observ- ingthisb ehaviorand copyingit(evenagainsttheevidenceof one'so wnprivateinformation) cansometimesb earational decision.Thiswasourfo cusinChapter 16.The otherreason wasbasedondirect-benefiteffects,alsocalled networkeffects:forsome kindsofdecisions, youincuranexplicitb enefitwhen youalign yourb ehaviorwiththebeha viorofothers. This iswhatw ewillconsider inthischapter.

Anaturalsetting wherenet worke

ff ectsariseis intheadoption oftechnologies forwhich interactionorcompatibilitywith othersisimp ortant.Forexample,when thefax machine wasfirstintroduced asapro duct,itsvaluetoapotential consumerdep endedonho wmany otherswere alsousingthesametec hnology.The valueof asocial-net workingor media- sharingsiteexhibits thesame properties:it's valuableto theextent thatotherpeopleare usingitas well. Similarly,a computeroperatingsystemcanb emoreusefulifmany other peopleareusingit: evenif theprimarypurp oseofthe operatingsystemitselfisnot to interactwithothers,an operatingsystem withmoreusers willtendto havealargeramoun t ofsoftw arewrittenforit,andwillusefileformats (e.g.fordo cuments,images, andmo vies) thatmorep eoplecaneasily read.

NetworkE

ff ectsasExter nalities.Theeffectswe aredescribingherearecalledpositive externalities.Anexternalityisany situationinwhichthew elfareofan individualisa ffected bytheactionsofother individuals,without amutually agreed-uponcomp ensation.For example,theb enefittoy oufromasocialnet workingsite isdirectlyrelatedtothe total numberofpeoplewhousethesite.When someoneelsejoins thesite,theyhav eincreasedy our

Draftversion: June10,2010

509

510CHAPTER17.NETW ORKEFFECTS

welfareeventhoughno explicitcompensationaccountsfor this.Thisis anexternalit y,and itispositiveinthesense thatyour welfareincreases. Inthisc hapter,wewillb econsidering theconsequencesof positive externalitiesdueto networke ff ects.Inthe settingswe analyze here,pay o ff sdepend onthenumber ofothers whousea goodandnotonthedetailsofhow theyareconnected. InChapter19, we willlook atthedetails ofnetw orkconnectivityand askhow theya ff ectthep ositive externalitiesthatresult. Noticethatw ehav ealsoseenexamplesofnegativeexternalitiesearlierinthe book - thesearecases whereanexternalit ycausesa decreaseinw elfare.Tra ffi ccongestion asdis- cussedinChapter 8is anexamplein whichy ouruseof a(transportation orcommunication) networkdecreasesthepayo ff tootherusers ofthenet work,again despitethelac kof com- pensationamongthea ff ectedparties. Inthefinal sectionofthis chapter,w ewilllo okata directcomparisonof positive andnegative externalitiesinmoredetail. It'simportan t,also,tonotethatnoteverythingis anexternality - the keypart isthat thee ff ecthasto be uncompensated.For example,ifyoudrinka canofDiet Cokethen thereisone lesscanof DietCoke fortherest ofthew orldto consume,soy oudecreasethe welfareofothersb yyour action.Butin thiscase,inorderto drinkthecanofDietCok e youhaveto payforit,andifyoupa ywhatit coststomakeanother canofDiet Coke,then youhaveexactly compensatedtherestoftheworldfor youraction. Thatis,thereisno uncompensatede ff ect,andhence noexternality .We explorethein teractionofexternalities andcompensation furtherwhenwediscussprop ertyrigh tsin Chapter24.

17.1TheEconom yWithout NetworkE

ff ects Ourcanonicalsetting inthis chapterwill bethe marketfor agood:we willfirstconsider howthemarketfunctions whenthereis nonetworke ff ect - that is,whenconsumers donot carehow manyotherusersof thegoodthereare - andthen wewill seehow thingschange whenanet worke ff ectispresen t. Wewantto analyzemarketswithahugenum berof potential purchasers,each ofwhom issmallenough relative totheen tiremarketthatheor shecanmak eindividualdecisions withouta ff ectingtheaggregate behavior. For example,eachindividualconsideringthe purchaseofaloaf ofbreaddo essowithout worryingab outwhetherher individualdecision - allelse remainingthe same - will a ff ecttheprice ofbread.(Note thatthisis di ff erent fromworrying aboutwhetherdecisionsmade byalargenum berof peoplewill hav eane ff ect, whichtheycertainlycan.)Of course,in realmarkets thenum berof consumersisfinite, and eachindividualdecisiondoes hav eav ery,verysmalle ff ectonthe aggregate.Buteac h purchaser'simpactissosmall relative tothemark etthatw ecanmo delindividualsasnot takingthisin toaccoun twhentheymakea decision.

Formally,wemodelthe lackofindividuale

ff ectson theaggregateb yrepresenting the

17.1.THEECONOMY WITHOUTNETWORK EFFECTS511

consumersasthe setofall realnum bersin theinterv alstrictly betw een0and1.Thatis,each consumerisnamed bya di ff erentrealnumb er,andthe totalmassofconsumersis1.This namingofthe consumersb yrealn umbers willbenotationallyuseful - forexample,the set ofconsumerswith namesb etween 0andx<1represents anxfractionofthe population. A goodwaytothink ofthismodelofconsumersisasa continuous approximationto amarket withav erylarge,but finite,numberof consumers;thecon tinuous modelwill beusefulin variousplacestoav oidha vingtodeal withtheexplicite ff ectofan yoneindividual onthe overallpopulation. Eachconsumerwants atmostone unitofthegood;eachconsumer hasa personalin trinsic interestinobtainingthego od thatcanv aryfromone consumertoanother.Whenthereare nonetw orke ff ectsatw ork,we modelaconsumer's willingnesstopayasb eingdetermined entirelybythisin trinsicinterest.Whenthere arenetw orke ff ects,aconsumer's willingness topay isdeterminedbytw othings: •intrinsicinterest;and •thenum berofotherpeopleusingthegood - thelarger theuserp opulation,themore sheiswilling topay .

Ourstudyof network e

ff ectsherecan beview edasan analysisofhowthings changeonce thissecondfactor comesinto play. Tostartunderstandingthisissue, wefirst considerhow amarket works whenthereare nonetw orke ff ects. ReservationPrices.Withnonet work effects,eac hconsumer'sinterestinthe good is specifiedbya singlereservationprice:themaxim umamount sheiswillingtopay forone unitofthe good. We'llassume thattheindividualsarearrangedintheinterv albetween 0 and1in orderof decreasingreservation price,sothat ifconsumerxhasahigher reservation pricethanconsumer y,thenx512CHAPTER17.NETW ORKEFFECTS

01r(1)r(0)PriceConsumersprice pxy = r(x)y = p

Figure17.1:When thereareno network e

ff orts,thedemand forapro ductata fixedmarket pricepcanbe foundbylocating thepoin twherethecurvey=r(x)intersects thehorizontal liney=p. Figure17.1illustrates, sincer(·)isa continuous functionthat strictlydecreases,itmust crossthehorizon talline y=psomewhere. Thismeansthat allconsumersb etween 0andxbuythepro duct,and allconsumers abovexdon't - so anxfractionofthe population buysthepro duct.Wecandothis for everypricep:thereis anxdependingonpthatspecifies thefractionofthep opulationthat willpurchase atpricep.Thisw ay ofreadingtherelationbetweenprice andquantit y(for anypricethequantit ythat willbe demanded)isusuallycalledthe(market)demandforthe good,anditisav eryusefulw ayto thinkofthe relationbet weenthepriceandthe num ber ofunitspurc hased. 1 TheEquilibriumQuan tity oftheGood.Let'ssuppose thatthisgoodcan bepro duced ataconstan tcostof p perunit,andthat,as isthe caseforconsumers, thereareman y potentialproducersofthegoo dsothatnoneofthem islarge enoughtob eabletoinfluence themarket priceofthegoo d.Then,in aggregate,thepro ducerswillb ewillingtosupply anyamountofthe goodataprice ofp perunit,andnoneof thego odat anyprice below p Moreovertheassumptionofalarge numb erofp otentialpro ducerswhocan createnewcopies 1 Inthelanguage ofmicro economics,thefunction r(·)describes theinversedemandfunction .The inverse ofr(·),givingxintermsof p,isthe demandfunction.

17.1.THEECONOMY WITHOUTNETWORK EFFECTS513

01r(1)r(0)PriceConsumersp*y = r(x)constant cost per unit p*

equilibrium quantity x* Figure17.2:When copiesofa good canbe produced ataconstan tcostp perunit,the equilibriumquantit yconsumedwillbethenum berx forwhich r(x )=p ofthego od ataconstantcostofp impliesthatthe pricecannot remainabo vep ,sincean y profittoa producer wouldb edriventozerob ycompetitionfromotherpro ducers.Thus,we canassumea market priceofp ,regardlessof then umber ofunitsof thegoodproduced. 2

Asabo ve,casesinwhichp

isabo ver(0)orb elow r(1)arenot particularlyin teresting,since theneitherev eryoneor noonebuysthegood.Therefore,w eassume thatr(0)>p >r(1). Tocompletethepictureof how themarket operateswithout network e ff ects,we now determinethesupplyofthego od. Sincep isbet weenthehighestandlowestreservation prices,we canfindauniquex between0and1sothatr(x )=p .We callx theequilibrium quantityofthego od, giventhereservationpricesandthecost p .Figure17.2 revisits

Figure17.1,including thecost p

andtheequilibrium quantit yx

Noticethesense inwhic hx

representsanequilibriumin thepopulation's consumption ofthego od. Iflessthananx fractionofthe population purchasedthe good,therewould beconsumerswhohav enot purchasedbut whowouldhaveanincentiv etodo so,because ofreservation pricesabovep .Inother words,there would be"upwardpress ure"onthe consumptionofthe product,since thereisa portionofthep opulationthatw ouldnotha ve purchasedbutwishedthey had.Onthe otherhand,if morethanan x fractionofthe 2 Continuingwiththemicroeconomiclanguage ,thisis thelong-runcomp etitivesupplyforany good producedbyaconstan t-costindustry.

514CHAPTER17.NETW ORKEFFECTS

populationpurchasedthego od,therewouldb econsumerswho hadpurchasedthegood butwishedthey hadnot,b ecauseofreserv ationpricesb elowp .Inthis case,we'd have "downwardpressure"ontheconsumptionofthego od. Oneattractive featureofthisequilibriumisthat itisso ciallyoptimal(as definedin Chapter6).T oseewh y,let'sconsidertheso cialwelfare oftheallocation,whichwe can thinkofas thedi ff erenceb etweenthetotalreservationpricesoftheconsumerswho receive acopy ofthegoodand thetotalcost ofproducing thecorrespondingquantityofthe good. Now,ifsociety were goingtoproduceenoughofthegoo dtogiveitto anxfractionofthe population,thensocial welfarew ouldbemaximizedby givingittoallconsumersb etween

0andx,sincethey correspond tothexfractionofthe population thatvalues thegoodthe

most.Which valueofxwouldbetheb estchoice?Sincethe contributionof aconsumerx totheso cialwelfare isthedi ff erencer(x )-p ,we canthinkofthesocial welfare,when consumers0through xgetcopiesof thegoo d,asthe (signed)areab etweenthecurvey=r(x) andthehorizon talline y=p .It'ssigned inthe sensethat portionsof thecurve y=r(x) thatdropb elow y=p contributenegativelytothe area.Giventhis,we'd want toc hoosex sothatw ecollect allthepositivearea betw eeny=r(x)andy=p ,andnone ofthenegativ e area.Thisis achieved byc hoosingxtobe theequilibriumx .Hencethe equilibriumquan tity x issocially optimal.

Wenowintro ducenetworke

ff ects;we'llseethat thiscausesseveralimportan tfeatures ofthemark ettoc hangeinfundamentalw ays.

17.2TheEconom ywith NetworkE

ff ects

Inthissection, we discussamo delfornetworke

ff ectsinthe marketfor agoo d.Wewill followageneralapproach suggestedb yKatz,Shapiro, andVarian[235,368];see alsothe writingsofBrian Arthur [25,27]for influentialearlydiscussionsofthese ideas.

Withnetw orke

ffquotesdbs_dbs47.pdfusesText_47