OnlineAppendix
1
PatentLicensing,TechnologyTransfer,&Innovation1
MarkA.Lemley2&RobinFeldman3
Abstract
Traditionaljustificationsforpatentsareallbasedondirectorindirectcontributiontothe
creationofnewproducts.Patentsservethesocialinterestiftheyprovidenotjustinvention,but
innovationtheworldwouldnototherwisehave.Non-practicingentities(NPEs)aswellas
product-producingcompaniescansometimesprovidesuchinnovation,eitherdirectly,through
workingthepatentortransferoftechnologytootherswhodo,orindirectly,whenotherscopy
thepatentedinnovation.Theavailableevidencesuggests,however,thatpatentlicensing
demandsandlawsuitsfromNPEsarenormallynotcasesthatinvolveanyoftheseactivities.
Somescholarshavearguedthatpatentscanbevaluableevenwithouttechnology
transferbecausetheabilitytoexcludeothersfromthemarketmaydrivecommercializationthat
wouldnototherwiseoccur.Wedemonstratethatevenifvariouscommercializationtheories
cansometimesjustifypatentprotection,theycannotjustifymostNPElawsuitsorlicensing
demands.
1©2016RobinFeldman&MarkA.Lemley.2WilliamH.NeukomProfessor,StanfordLawSchool;partner,DurieTangriLLP.3Harry&LillianHastingsProfessor,UniversityofCaliforniaHastingsCollegeoftheLaw.ThankstoRoseHagan,JoshLerner,CarlShapiro,andparticipantsattheAmericanEconomicsAssociationannualmeetingforcommentsonapriordraft.
OnlineAppendix
2
1. Innovation-RelatedJustificationsforNPEs
Thepatentsystemexiststoserveutilitariangoals,ratherthantoreflectsomenotionof
aninventor’smoralright.(Feldman2012,Lemley1997,Landes&Posner2003).Fromthemany
commercialactivitiesthatmightotherwisebeopentoanyone,weremovesomeforalimited
periodoftimeinthehopethatdedicatingthemtotheprovinceofafew,willredoundtothe
benefitofusall.(Feldman2013).
Thetraditionalutilitarianstorysupportingthepatentsystemisthatthelureofpatent
rightsencouragesinventionthatwouldnototherwiseoccur,orattheleastwouldoccurlater
butforthepatent.(Landes&Posner,2003;Duffy2004).Theinventionthesystemisdesigned
topromoteisnotwhatisknowninscienceas“basicresearch,”suchasanunderstandingof
hownatureworksorwhatforcespropeltheuniverse.Afterall,formorethanacentury,the
Courtshaveremindedusthatthepropersubjectmatterofapatentdoesnotincludelawsof
nature,naturalphenomenon,orabstractideas—nomatterhowvaluableandessentialtothe
progressofsciencethesemaybe.(AliceCorp.Pty.Ltd.v.CLSInt’l,134S.Ct.2347,2354(2014))
Rather,thepatentsystemisaimedatprotecting“applied”inventions,orinnovations,thatare
deployedintheworld.Onlywhenbroadandbasicprinciplesarereducedtoaparticular
practiceandappliedinaspecificendeavorwilltheybeeligibleforprotection.(Bilskiv.Kappos,
561U.S.593(2010)).
Thepatentsystem’sfocusisconsistentwitheconomicliterature,whichdistinguishes
invention—anidea—frominnovation—turningtheideaintoaviableproduct.Thepatent
OnlineAppendix
3
systemencouragesnotjustinventionintheabstract,butthecreationofnewproducts.Thisis
the“progress”ofthe“usefularts”mentionedinthepatentclauseoftheConstitution.
Thefocusoninnovation,notsimplyinvention,isparticularlyimportantwiththe
emergenceofthemodernnon-practicingentity(NPE)businessmodel.Colloquiallyknownas
patenttrolls,NPEsarethosewhosecoreactivityinvolveslicensingorlitigatingpatents,as
opposedtomakingproducts.Byallaccounts,themodernNPEbusinessmodelhasexpanded
rapidlysinceitsemergenceoverthelasttwodecades,anexpansionthatisparticularlyevident
inthecontextoflitigation.(Allisonetal.,2016).Differentscholarsslicethenumbers
differently,withsomeexcludingNPEsorganizedastrustsandindividualsandothersexcluding
“failedstartups”andoriginalinventors,forexample.Whenthebroaderdefinitionisapplied,
however,thedataareremarkablyconsistentacrossstudies,withallshowingthatNPEsnow
accountforthemajorityofpatentlawsuitsfiled.4
Consistentwiththegoalsofthepatentsystem,alloftheargumentssuggestingthe
benefitsofNPEsrestontheircontributions,eitherdirectlyorindirectly,tothecreationof
productssomewhereinthesystem.NPEs,unlikepracticingentities,donotdeploythe
technologyintheworldthemselves,butthatdoesnotanswerthequestionofwhetherthey
contributetoinnovationandthecreationofnewproducts.Topromoteinnovation,however,
theymustnotonlyinvent,butthatinventionmustleadtothecreationofproductsby
someone,somewhereinthesystem,atsomepoint.NPEsmaybeactingasmiddlemen4CompareFeldmanetal.,2013;Chien,2012(usingdatafromRPXCorporationandconcludingthatthepercentageoflitigationbynon-practicingentitiesin2012hasreached62%)withCotropiaetal.2014(usinganarrowerdefinitionofNPEsandfindingnorealincreaseinNPElitigationwhencomparingtheyears2010and2012,butalsonotingthat“whenwerepackageall[NPEs]intoasinglecategory,theyareresponsibleforamajorityof[patentlawsuits]in2012”);seealsoSag,2016(findingthatpatentlitigationvolumedoubledfrom2010to2013).
OnlineAppendix
4
transferringtechnologytothosewhowouldimplementit;ortheycouldbecollectingrevenue
fromthosewhocopiedtheirinventionandimplementedit.Neitherpossibility,however,
appearsbroadlysupportedbytheavailableevidence.
Thereisasubstantialliteraturethatcallsintoquestionwhetherthepatentsystemin
generalencouragesinnovationthatwouldnototherwisehappen.Thefactsthatmost
significantinnovationsaresimultaneouslycreatedbytwoormorepeople(Lemley,2012),and
thatinmostindustriesvirtuallyallpatentenforcementisdoneagainstindependentinventors
(Cotropia&Lemley,2009)castsignificantdoubtontheclaimthattheinnovationswouldnot
havehappenedbutforthelureofapatent.Theissueis,however,complicatedbythevery
differentcharacteristicsofdifferentindustries.Theremaybeindustriesinwhichinventionisso
complexanduncertainthatitwouldnotbeundertakenwithoutpatentprotection.(Burk&
Lemley,2009).Buttherealsoseemtobeindustries–perhapsmostofthem–inwhichthe
patentsystemdoesnotseemtobedrivingnewinvention,andmayevenberetardingit.
(Bessen&Meurer,2008).Thatmightleadonetoquestionthepatentsystemasawhole
(Boldrin&Levine,2007),oratleastthetraditionalinnovation-basedjustificationforit.
Inadditiontodoubtsabouthowwellpatentsingeneralactuallydriveinnovationsthat
wouldnototherwisehaveoccurred,theevidencecastssignificantdoubtontheefficacyofthe
patentdisclosureasawayofdisseminatingideasandleadingtothecreationofproducts.
Whilewritingdownandpublishingadescriptionoftheinventionhaslongbeenaquidproquo
forapatent,inthemodernworldthereisgoodreasontothinkthatengineersinmanyfields
rarelyreadpatentsinordertolearnaboutatechnology.(Lemley,2008;Lemley,2012;Harvard
OnlineAppendix
5
Note,2005).5Therearemanyreasonsforthis.Lawyersatmanycompaniesdiscouragetheir
engineersfromreadingpatentsforfearofincreasinglegalliability.(Lemley&Tangri,2003).
Thequalityofthedisclosureinthepatentmaybepoor,particularlyintheITindustries.(Burk&
Lemley,2002;Feldman;2012,othercites).Therearesimplytoomanypatentsinmanyfieldsto
possiblykeepupwith,and600,000moreapplicationsarefiledeveryyear.Andinafast-moving
industrylikeITadelayofseveralyearsbetweeninventionanddisclosuremaymakethe
technologydescribedobsoletebythetimeanyonecouldreadit.(Feldman&Lemley,2015).
Further,economicliteraturesuggeststhatinorderforeffectivetechnologytransfer—in
otherwords,transferthatcanleadtocommercialization—readingapatentaloneisnot
enough.Suchtransfergenerallymustincludenotonlytheinformationpubliclyavailableinthe
patent,butalsothetransferofknow-how,complementaryassetsandotherperipheral
disclosures.(Bessen,2015;Teece,1986;Rantanen,2012).Thus,ifpatentsactuallydrive
innovationbythirdpartieswewouldexpecttoseenotsimplypatentingbutbusiness
transactionsthatinvolvethetransferofothertypesofinformationassets.
Alternatively,NPEscoulddriveinnovationiftheyservedasefficientmiddlemen,
connectingthosewhoinventbutwhoseinventionshavenotbeendeployedwiththosewhocan
produceaproductfromthatinvention.SeveralpeoplehavearguedthatNPEsservethisrole.
(Khan2014;Spulber,2012;Crane,2009;)Here,too,aninnovationbenefitrequirestechnology
transfer.Thesocialbenefitofthemiddlemanstorydependsonthemiddlemanproviding
somethingofvaluetotheimplementer.
5Bycontrast,insomefieldspatentsmayprovidemoreusefulguidancetoengineers.(Ouellette,2012).
OnlineAppendix
6
Inshort,then,thetraditionaljustificationsforNPEscontributingtosocialwelfareall
involvesomeformoftechnologytransferorlearningdisseminationbywhichtheNPEorthe
patentitholdsteachestheimplementeratechnologyitdidnototherwisepossess.Practicing
entitiescanbenefitsocialwelfarewithouttechnologytransferbymakingandsellingthe
inventiondirectly;NPEscannot.
EarlyevidencetestingthepositiveimpactofNPEsoncommercializationgoalsisnot
encouraging.Theevidenceislargelyobservationalinnature,flowingfromsmallsamplestudies,
withalloftheattendantlimitations.Nevertheless,thedataprovideausefulwindowonthe
NPEbusinessmodelandsuggestapproachesforgeneralizableanalyses.
FeldmanandLemleyprovidesurveyevidenceofthedirectmeasureofthecreationof
newproductsasaresultofpatentassertionsbyNPEs.Wealsotestcommercializationeffects
bymeasuringothermarkersofpotentialinnovation,suchastechnologytransferbeyondthe
patent.(Feldman&Lemley,2015).Includingsuchmarkerscreatesamoredynamicpictureof
thepotentialforfuturecommercialization,evenifthatcommercializationhasyettooccur.
Weknowthatactualtechnologytransferhappenswithinthepatentsystemintheex
antecontext.(Arora,2001).BothpracticingentitiesandsomeNPEsengageinexante
technologytransfer.Inparticular,universitiesandinventorscreateallianceswithcompanies
thatcanmoreeasilydevelopandcommercializetheirinventionsthroughjointventuresand
othertypesoftechnologyandresearchsharingagreements.(Feldman&Lemley,2015).These
agreementsfrequentlyoccurbeforeapatentissuesorevenbeforeanyofthepartiesfilefora
patent.Notably,theseagreementsinvolvetechnologytransfer.Universitiesandother
inventorsinthesedealsprovidenewtechnologytothoseinapositiontoimplementit.And
OnlineAppendix
7
thattechnologyoftenincludestradesecretsandknow-howbeyondtheto-be-patented
technologyitself.
Patentlitigationandlicensingdemandsforexistingpatents,bycontrast,tendtooccur
afterthedefendanthasalreadydevelopedandimplementedthetechnology.Thisis
particularlytrueofNPEpatentassertionsandlicensingdemands,whichsomeevidence
suggeststendtohappeninthelastfewyearsofapatent’slife,althoughthepictureis
complicated.(Love,2013;Feldman,Ewing&Jeruss,2013).NPElicensingdemandsand
litigationagainstcompaniesthatareproducingproductsdonotseemtoinvolvetechnology
transferorotherindiciaofnewinnovation.
WhilesomehavearguedthatNPEsserveasefficientmiddlementhroughthisactivity,
transferringinventionsfromcreatorstocommercializers,FeldmanandLemleyfoundnosuch
evidence.Theauthorssurveyed181in-houselicensingattorneysatcompaniesthatproduce
products,onthetheorythatthesepartieshavedirectknowledgeofwhetherthecompany
implementednewtechnologyandbecausein-housecounselstendtonegotiatelicensesbothas
patentholdersandaspotentiallicensees.Thesurveyexaminedtheeffectsoflicensesthata
companytookafterreceivingapatentdemand,whichwasdefinedtoincludecallsorletters
suggestingareasofmutualinterestorjointventures,offeringtolicensepatents,threatening
litigation,givingnoticeofintenttofileaninfringementlawsuitoractuallyfilinginfringement
lawsuit.Respondentswereaskedwhetherthoselicensesledtoanymarkersofinnovation.
Directmarkersofinnovationincludedtheadditionofnewproductsorfeatures.Indirect
markersofinnovationincludedwhetherthepatentholdertransferredknow-howorother
technicalknowledgealongwiththepatent,whetherthepatentholdertransferredpersonnel
OnlineAppendix
8
(includingthroughaconsultingagreement),andwhetheranyjointventureswerecreated.
Again,thesurveyconsideredonlylicensestakeninresponsetounsolicitedlicensingrequests.It
didnotlookatthepractice,particularlyamonguniversityinventors,ofenteringintotechnology
transferagreementsbeforeembarkingondevelopmentofanewtechnology.
TheresponsessuggestthatlicensingrequestsfromNPEsrarelyleadtodirectorindirect
markersofinnovation.Ninety-twopercentofrespondentsreportingthattheyaddednew
productsorfeaturesfromzero-10%ofthetime,withNPElicensing.Theresultswereeven
strongerwhenrespondentswereaskedaboutindirectmarkersofinnovation,withrespondents
reportingwithcompleteunanimitythattheyrarereceivedtechnicalknowledge,transferof
personnelorjointventuresalongwithapatentlicense.Thus,theresultssuggestthatNPEsdo
notappeartobeplayingtheroleofefficientmiddlemen.Whileitiscertainlypossiblethata
middlemanrolecouldbereflectedinsomeothermarkersthantheonesweexamined,wedid
notfindsuchevidenceinourpreliminarywork.
Interestingly,theevidencewasalsodismalwhenexpostlicensingrequestscamefrom
thoseotherthantraditionalNPEs.Whenproductproducingcompaniesanduniversitiesmade
unsolicitedapproachesandthoseapproachesresultedinalicensingagreement,the
agreementswereunlikelytoleadtodirectorindirectmarkersofinnovation.Three-quartersof
respondentsreportednewproductsorfeaturesfromzeroto10percentofthetime;88%of
respondentsreportedtransfersofknowledgezeroto10percentofthetime;94%reported
transfersofpersonnel(includingthroughconsultingagreements)zeroto10percentofthe
time;and91%reportedjointventuresfromzeroto10%ofthetime.(Feldman&Lemley,
2015).Theseobservationalresultssuggestthatexpostpatentlicensingdemandsdonotappear
OnlineAppendix
9
toleadtotechnologytransferorothermarkersofinnovation,nomatterwhattypeofparty
initiatestheunsolicitedapproach.
Amiddlemanwhoisnotmakingaproductandnotactuallyprovidingthelicenseewith
newtechnologyisoperatingatmostasataxcollectorforinventorswhocouldnototherwisedo
battleagainstlargecompanieswhohaveimplementedtheirideas.(Feldman&Lemley,2015).
Buttransactionsarenotdesirablefortheirownsake.(Burstein,2015)Itissociallydesirableto
imposesuchataxoninnovatorsonlyiftheworldgainssomethingfromit.Thatmightbetrue
if,forinstance,theimplementerhadactuallycopiedtheideafromthepatentee.Ifwethink
copyingcouldreduceincentivestoinvent,wemightreasonablyprefertoforcecopiersinto
licensingarrangementsinstead,compensatingtheinventorwhoseworkiscopied.
Theavailableevidencesuggestsitisunlikelythatmostpatentenforcementtargetssuch
copying.Forexample,CotropiaandLemleydemonstratethatmostpatentlawsuitsarefiled
againstthosewhohavedevelopedaproductindependently,ratherthanthosewhohavetaken
theideafromapatentholder.(Cotropia&Lemley,2009).Andwhilesomehavespeculated
thatdefendantsmaycopyindirectly,learningabouttheinventionfromthepatentee’sproduct
orfromscientificdiscussionsoftheideawithouteverreadingthepatentitself(Merges,2014),
thatisfarmorelikelywhenthepatenteeactuallymakesaproductthanwhenitproduces
nothingotherthanthepatent.Further,thereisevidencethatNPEstendtoassertpatentsat
theendoftheirlives,whilepracticingentitiesassertpatentsearly,furtheringcastingdoubton
thecopyingstory.(Love,2013).6Andthereisverylittleevidencethatpatenteeshaveuseda
6We(andLove)acknowledgethattheinterpretationofhisdataiscomplicatedbythechangeinthenumberofNPEsuitsduringthetimeofhisstudy.(Feldman,Ewing&Jeruss2013).
OnlineAppendix
10
remedycreatedin1999toprotectagainstcopyingofpublishedpatentapplications.35U.S.C.
§154(d).
Somehavesuggestedadifferenttheoryoftaxcollection–thatNPEpatentsuitsprovide
analternativewayforventurecapitaliststorecoversomeinvestmentinafailedinvention.We
areskepticalthatventurecapitalistsaremuchmotivatedbytheprospectofasortof
consolationprize.Theytellusthattheyaren’t;insurveyresponsestheyindicatethatitisthe
prospectofabigwin,notthepossibilityofrecoveringsomemoneyfromafailedinvestment,
thatmotivatesthem.(Feldman,2014).Butevenifthereweresomemarginalincentiveto
additionalinvestmentfromNPEtaxation,thathastobeweighedagainstthecostimposedon
successful,product-implementingbusinesses,aswellasonstartupsthemselves.Inparticular,
venturecapitalistsnotethatifastartupcompanyhasapatentassertionagainstit,thatwould
beasignificantdeterrentforanyfundingrequest.(Feldman,2014).Finally,itisworthnoting
thatamajorityofNPEsuitstargetsmallcompaniessuchasventure-fundedstartups,notlarge
companies(Chien2014).Thus,atbest,manyoftheseNPEsuitswouldbetaxingsomeventure-
backedstartupsforthebenefitofsubsidizingothers.Thatseemsadubiouspolicyidea.
2. “Commercialization-Plus”Justifications
Respondingtosomeoftheconcernswithtraditionaljustificationsforpatentprotection
andlicensing,anumberofscholarshavearticulatedwhatwecall“commercialization-plus”
justificationsforpatentprotection.(Kitch,1977;Abramowicz,2007;Kieff,2001;Abramowicz&
Duffy,2008;Sichelman,2010).Thesejustificationsdifferfromthetraditionalinnovation-based
justificationsbecausetheyfocusontheallegedneedforearlyoradditionalprotectionsto
OnlineAppendix
11
encouragepost-inventioninvestmentincommercialization.Themostfamousofthese,known
as“prospecttheory,”recommendsgrantingstrongpatentsearlyinthelifecycleofanideaso
thatasinglepartycancontroldevelopmentoftheideamuchthewaythatamineralprospector
managesamineralclaimsite.(Kitch,1977).Otherscholarshavesuggestedthatweshouldgrant
patentstooldtechnologiesinareaslikepharmaceuticalsinordertoencouragethepatent
ownertobringtheproducttomarket.(Roin,2009).Stillothershavesuggestedgrantingnormal
patentsattheoutset,andthen,ifnoonecommercializesit,grantingextrarightstotheperson
whodoes.(Sichelman,2010).TheBayh-DoleAct,passedin1980,waspremisedontheworry
thatuniversityinventionswouldlanguishunlessonepartywasgiventherighttoturnthose
inventionsintocommercialproducts.(CouncilonGovernmentRelations,1999).
Commercialization-plustheoriesarecontroversial.Lemleyamongothershascriticized
Kitch’sprospecttheoryas“fundamentallyanti-market”becauseitpresumesthatcentral
controlissuperiortomarketallocationofexistingresources,andFeldmanhasarguedthat
patentsareentirelyunlikethemoreclearlydefinedrightsinKitch’smineralanalogy.(Lemley,
2004;Lemley,1997;Lemley,2012;Feldman2012).Othershavesuggestedthat
commercializationtheoryispoorlyfittedtoindustriesinwhichinventionproceedsbystages
andimprovementsratherthanbydiscreteadvances.(Merges&Nelson,1990).
Kitch’stheorymayjustifypatentprotectionincertainindustries.BurkandLemley
suggestthatthetheorymapsbesttothepharmaceuticalindustry,wheregovernment
regulatorybarrierssignificantlyraisethecostofentryandmayrequireexclusivity.(Burk&
Lemley,2009).AnditmayjustifypatentprotectionforNPEsinsomecases,aswithuniversity
technologytransferagreements.(Kieff,2001)
OnlineAppendix
12
Notably,though,anyformofcommercializationtheoryisself-limitingincertain
importantrespects.First,ifexclusivityisnecessarytoinduceafirmtocommercializea
technology,weshouldrarely,ifever,seemultiplecompaniesindependentlydevelopthesame
technology.Theveryconceptofcommercializationtheoryisthatnoonewouldinvestin
developingandcommercializingthetechnologyunlesstheywerefirstconfidenttheywould
haveexclusiverightsoverthattechnology.Theremaybeexceptionsinwhichcompanies
engageinpatentracing,eachhopingtobethefirsttoreachanimportantinvention.(Barzel,
1968;Tirole,1988;Dasgupta&Stiglitz,1980;Grady&Alexander,1992;Reinganum,1989).But
thehistoricalexamplesofpatentraceshavetendedtoberacestoinvent,notracesto
commercialize.(Lemley,2012).Ifcommercializationtheoryiscorrect,evenindependent
inventorswon’tcommercializethetechnologyunlessanduntiltheyareconfidenttheywill
haveexclusiverightsoverthattechnology.Andracingtocommercialize(ratherthantoinvent)
ispresumablysomethingmostlyengagedinbypracticingentities,notNPEs.Forthesame
reason,undercommercializationtheoryweshouldn’tseeopensourceorpublicdomain
companies.Ifwedo,thatisreasonableevidencethatexclusivityisnotnecessarytoinduce
commercializationinthatindustry.(Asay,2015).
Further,ifanyformofcommercializationtheoryiscorrect,andcertainlyifenhanced
commercializationtheoryis,infringementshouldberare.Independentlaterinventors
shouldn’tcommercialize,sincetheywon’thaveexclusiverightstotheinvention.Norshould
weseemuchcopyingofthepatentee’sinvention,sinceifcommercializationrequiresmarket
exclusivitythecopierwon’tgenerallybeanybetteroffthananindependentinventorwho
doesn’thaveexclusivity.Thatdoesn’tmeanwewouldneverseepatentlitigation.Butitdoes
OnlineAppendix
13
meanthatifcommercializationtheoryiscorrect,itshouldinvolveexceptionalcircumstances
suchasalowerregulatoryburdenonsecondentrants(trueofgenericpharmaceutical
companies)orsomereasontothinkthatsimplyknowingthatamarketexistssomehow
dramaticallyreducesthecostsofcommercialization.Theremaybesuchcases.Applev.
Samsungisarguablyone.Butthosecaseswouldinvolvecopyingofexistingmarket
participants.
Athirdimplicationofcommercialization-plustheoryisthatpatentownersshould
generallynotgrantnonexclusivelicenses.Ifthistypeoftheoryiscorrect,universitiesshouldbe
grantingexclusivelicensestopracticingentitiesinanygivenmarket,sincethelicenseeneeds
thatexclusivitytocommercializetheinvention.Nonexclusivelicensesshouldberareand
tightlycontrolled,sincethenonexclusivelicenseeswouldhavetocoordinatetheirproduction
andpricingdecisionsundercommercializationtheory.Andnonexclusivelicensestomultiple
partiesshouldbenonexistent,sinceallowingopenentryintoamarketisinimicaltothetheory
ofcommercialization.Infact,however,theevidencesuggestsnotonlythatnonexclusive
licensingbyuniversitiesiscommon,butthatthelackofexclusivityisanimportantdriverof
subsequentimprovementforcoreenablingtechnologies.(Lemley,2005).
Evenifcommercializationtheoryjustifiespatentprotectioninsomeindustries,itcannot
justifymostmodernpatentlitigation.NorcanitjustifyexpostlicensingdemandsbyNPEs.
Outsidethepharmaceuticalindustries,NPElicensingdoesnotlookmuchlikecommercialization
theorywouldpredict.ThereisevidencethatNPEpatentsareassertedlateinlife,(Love,2013),
andalmostalwaysagainstindependentinventors.(Cotropia&Lemley,2009).ExpostNPE
patentlicensesdon’ttransferthetechnologytoapartythatcanlatermakeuseofit.(Feldman
OnlineAppendix
14
&Lemley,2015).AndNPEpatentlicensingdemandsessentiallyalwaysseeknonexclusive
licensesfrommultipleparties,notanexclusivelicensefromasingleparty.Indeed,NPEs
commonlysuetwentyormoredefendantsinthesameindustryandthesametime,settling
witheachoftheminexchangeforanonexclusivelicense.(Allisonetal.,2009).
TheevidencealsosuggeststhatNPEsaretargetingsuccessfulcommercializers,not
facilitatingnewcommercialization.Feldman&Frondorfstudied50productcompaniesthat
hadinitialpublicofferingsbetween2007and2012.(Feldman&Frondorf,2016;Feldman,
2014;Chien,2014).Theauthorsfoundthatthat40%ofrespondentsreceivedpatentdemands
duringtheperiodsaroundthetimeoftheIPOs,withthosedemandscominglargelyfromNPEs.
Theeffectswereevenmorepronouncedforinformationtechnologycompanies,withalmost
60%ofrespondentsreportingpatentdemandsaroundthetimeoftheirIPOs.Similarly,Cohen,
Gurun,&Kominersfoundthatacompanywas50%morelikelytobesuedbyanNPEfollowinga
large,positivecashshocksuchasafundingeventoranIPO,andthatacashshockwasa
significantpredictorofthenumberoftimesacompanywassuedbyNPEs.(Cohenetal.,2015).
Theauthorsfoundfurtherthatnootherformoflitigationhasthesametypeofcashtargeting
behavior--nottorts,contracts,securities,environmental,orlaborlaw.Nothing,butpatent
law.TheresultssuggestthatNPEdemandbehaviormaybedrivenbythelureofdeeppockets
andtheleverageopportunitiesaffordedbyanIPOperiod,ratherthanthemeritorious
representationofclaimsthatawrongedinventorcouldnotbringonitsown.Suchstudiesalso
provideareminderthatanyvirtuousbenefitsofNPEactivityshouldbeevaluatedagainstany
coststoinnovationandsociety.(Bessen&Meurer,2014;Morton&Shapiro,2014).
OnlineAppendix
15
Allformsofcommercializationandproduct-basedtheorieshaveafinal,surprising
implicationforNPEsuits.Ifthereasonweneedapatentisnottoinduceinventionbutto
inducecommercializationofthatinvention,thelawshouldpreferthosewhoactually
commercializetheinventionoverthosewhomerelyinventitbutdonothingfurther.Thus,the
ownersofthosepatentshavefailedintheirpurposeiftheyhaveneithercommercializedthe
inventionthemselvesorexclusivelylicensedthepatenttosomeonewhodoes.Suchan
approach,therefore,mayjustifyaworkingrequirement,somethingthatisgenerallyconsidered
anathematopatentadvocates.(Sichelman,2010).Further,asbetweentheNPEinventorwho
doesnotengageintechnologytransferandtheindependentinventordefendant,
commercializationandproduct-basedtheoriesshouldpreferthedefendant,sinceitisthe
defendant,notthepatentee,whohasachievedthegoalofthepatentsystem.
3. IndependentInventionandPriorUserRights
MostotherIPregimes,includingcopyrightandtradesecrets,exemptindependent
developmentfromlegalliability.Patentlaw,bycontrast,punishesanyonewhopracticesthe
claimedinvention,evenindependentinventors.Anumberofscholarshavesuggestedthat
patentlawshouldadoptsomeformofindependentinventionorprioruserrightsdefense.
(Shapiro,2006;Vermont,2006).Othershaveworriedthatanindependentinventiondefense
mightinterferewithpatentracesorincentivestocommercialize.(Lemley,2007).
Ouranalysissuggeststhatthepatentsystemmightsensiblyrequirethatapatentee
showeitherthatitpracticesinthemarketorthatithasengagedintechnologytransfertothose
OnlineAppendix
16
whothenputthetechnologyintopractice.7Apatenteewhocannotshoweitherwouldstillbe
abletoenforceitspatent,butonlyagainstthoseitcouldshowcopiedtheinventionfromit,
directlyorindirectly.Thishybridapproachtracksthelegaljustificationsthathavebeenoffered
forpatents.Aninventorwhodevelopsanideaotherscopywouldbeabletoenforcethepatent
againstthosecopiers,sincecopyingisaformoftechnologytransfer(andoneweviewas
sociallyinferiortoalicenseagreement).Aninventorwhoendsupdisseminatingtechnologyto
theworld,eitherbypracticingtheinventionorbytransferringtechnologytootherswhodo,
wouldbeabletoenforcethepatentagainstbothcopiersandindependentinfringers.Anda
practicingentitywouldsimilarlybeabletoenforcepatentsagainstbothcopiersand
independentinfringersonacommercializationtheory.
Introducingevensuchalimitedindependentinventiondefensewouldrequirecourtsto
evaluatedisputedclaimsofcopyingonsomecases.Somehaveworriedthatmuch
independentinventionisreallycopyingindisguise(Merges,2014),thoughothersareskeptical.
(Lemley,2012).AsSamVermonthasobserved,though,courtsarequitegoodatresolving
factualdisputesofthissort.Independentdevelopmentwilltendtoleaveapapertrail.Andthe
partieswillhaveanincentivetocollectandpresentevidenceonthequestion.Further,we
thinkcourtscanproperlyincludeintheconceptofcopyingcasesofindirectcopyingfroman
ideaonceithasbeenpublicizedbythepatentee.(Merges,2014).
Arequirementthatpatenteeswhodonotengageinanyformofcommercializationor
technologytransferprovethatthedefendantcopiedfromthemshouldbepairedwithstricter7Whileinanidealworldacourtmightwanttodeterminewhethertherewastransferofnon-patentknow-howalongwiththepatent,anylegalrulerequiringsuchatransferwouldleadtoshamtransactionsinwhichNPEsattacheduseless“know-how”toanonexclusivepatentlicenseinordertoobtainmorefavorabletreatment.
OnlineAppendix
17
penaltiesagainstthosedeemedtohavecopiedtheinvention.Itwouldbereasonableto
require,notmerelypermit,trebledamageawardsandattorneys’feesagainstthosefoundto
copy.Increasingthepenaltiesforthosewhoopttotaketechnologyfromapatenteewithout
paying,whileeliminatingthepenaltyimposedoninnovatorswhodonotbenefitfrompatentee
technologytransfer,properlyalignsthepatentsystem’sincentiveswiththeevidenceandthe
arrayoftheoreticaljustificationsforpatents.
OnlineAppendix
18
References
Abramowicz,Michael.2007."TheDangerofUnderdevelopedPatentProspects"92CornellLaw Review.1065.Abramowicz,Michael&Duffy,F.John.2008."IntellectualPropertyforMarket Experimentation"83NewYorkUniversityLawReview.337.Allison,JohnR.,Lemley,A.Mark,&Schwartz,David.2016(forthcoming)."HowOftenDoPAEs WinPatentSuits?"(workingpaper).Allison,JohnR.,Lemley,A.Mark,&Walker,Josh.2009."ExtremeValueorTrollsonTop? EvidencefromtheMost-LitigatedPatents"158UniversityofPennsylvaniaLawReview.Arora,Ashish.2001.MarketsforTechnology,TheMassachusettsInstituteofTechnologyPress.Asay,D.Clark.2015."EnablingParentlessInnovation"74MarylandLawReview.431.Barzel,Yoram.1968."OptimalTimingofInnovations"50ReviewYoramBarzel,"OptimalTimingofInnovations"50TheReviewofEconomicsandStatistics.348,
352(note11).Bessen,James.2015.LearningbyDoing,YaleUniversityPress.Bessen,James&Meurer,J.Michael.2008.PatentFailure:HowJudges,Lawyers,and
BureaucratsPutInnovationatRisk,PrincetonUniversityPress.Bessen,James&Meurer,J.Michael.2014."TheDirectCostsfromNPEDisputes"99CornellLaw
Review.387,411.Boldrin,Michele&Levine,David.2007.AgainstIntellectualMonopoly,CambridgeUniversity
Press.Burk,L.Dan&Lemley,A.Mark.2009.ThePatentCrisisandHowtheCourtsCanSolveIt,
UniversityofChicagoPress.Burstein,Michael.2015."PatentMarkets:AFrameworkforEvaluation"47ArizonaStateLaw
Journal.Chien,V.Colleen.2012."PatentAssertionEntities"PresentationtotheDOJ/FTCHearingon
PAES"(December10),availableathttp://ssrn.com/abstract=2187314.
Chien,V.Colleen.2014."StartupsandPatentTrolls"17StanfordTechnologyLawReview.
OnlineAppendix
19
Cohen,Burun&Kominers.2015."PatentTrolls:EvidencefromTargetedFirms"Harvard
BusinessSchoolWorkingPaperNo.15-002.Cotropia,A.Christopher.2014."UnpackingPatentAssertionEntities"99MinnesotaLaw
Review.649,655.Cotropia,A.Christopher.2009."CopyinginPatentLaw"87NorthCarolinaLawReview.1421.Crane,A.Daniel.2009."IntellectualLiability"88TexasLawReview.253,286-287.CouncilonGovernmentRelations.1999."TheBayh-DoleAct:AGuidetotheLawand
ImplementingRegulations."Dasgupta,Partha&Stiglitz,Joseph.1980."Uncertainty,IndustrialStructure,andtheSpeedof
R&D"11BellJournalofEconomics.1.
Duffy,F.John.2004."TheMarginalCostControversyinIntellectualProperty"71UniversityofChicagoLawReview.37.
Feldman,Robin.2012.RethinkingPatentLaw,HarvardUniversityPress.Feldman,Robin.2013."IntellectualPropertyWrongs"18StanfordJournalofLaw,Businessand
Finance.252.Feldman,Robin,Ewing,Tome,&Jeruss,Sara.2013."TheAIA500Expanded:TheEffectsof
PatentMonetizationEntities"17UCLAJournalofLawandTechnology.1,42.Feldman,Robin&Frondorf,Evan.2016(forthcoming)."PatentDemandsandInitialPublic
Offerings"StanfordTechnologyLawReview.Feldman,Robin.2014."PatentDemands&StartupCompanies:TheViewformtheVenture
CapitalCommunity"16YaleJournalofLawandTechnology.236.Feldman,Robin&Lemley,A.Mark.2015."DoPatentLicensingDemandsMeanInnovation"101
IowaLawReview.137,139.Grady,F.Mark&Alexander,I.Jay.1992."PatentLawandRentDissipation"78VirginiaLaw
Review.305,306.Kahn,ZorinaB.2014."TrollsandOtherPatentInventions:EconomicHistoryandthePatent
ControversyintheTwenty-FirstCentury"21GeorgeMasonLawReview.825,832.Kieff,ScottF.2001."PropertyRightsandPropertyRulesforCommercializingInventions"85
OnlineAppendix
20
MinnesotaLawReview.697.Kitch,W.Edmund.1977."TheNatureandFunctionofthePatentSystem"20JournalofLawand
Economics.265,275-280.Landes,M.William&Posner,A.Richard.2003.TheEconomicStructureofIntellectualProperty
Law,HarvardUniversityPress.Lemley,A.Mark.1997."TheEconomicsofImprovementinIntellectualProperty"75TexasLaw
Review.989.Lemley,A.Mark.2012."TheMythoftheSoleInventor"110MichiganLawReview.709.Lemley,A.Mark.2005."PatentNanotechnology"58StanfordLawReview.601.Lemley,A.Mark.2004."ExAnteVersusExPostJustificationsforIntellectualProperty"71 UniversityofChicagoLawReview.129.Lemley,A.Mark.2007."ShouldPatentInfringementRequireProofofCopying?"105Michigan LawReview.1525.Lemley,A.Mark.2008."IgnoringPatents"2008MichiganStateLawReview.19.Lemley,A.Mark.2009."ACautiousDefenseofIntellectualOligopolywithFringeCompetition"5
ReviewofLawandEconomics.3.Lemley,A.Mark&Melamed,DouglasA.2013."MissingtheForestfortheTrolls"113Columbia
LawReview.2117.Lemley,A.Mark&Tangri,K.Ragesh.2003."EndingPatentlaw'sWillfulnessGame"18Berkeley
TechnologyLawJournal.1085.Love,Brian.2013."AnEmpiricalStudyofPatentLitigationTiming:CouldaPatentTerm
ReductionDecimateTrollsWithoutHarmingInnovators?"161UniversityofPennsylvaniaLawReview.1309.
Merges,P.Robert.2014."AFewKindWordsforAbsoluteInfringementLiabilityinPatentlaw,"
availableathttp://ssrn.com/abstract=2464756.Merges,P.Robert&Nelson,R.Richard.1990."OntheComplexEconomicsofPatentScope"90
ColumbiaLawReview.839.
OnlineAppendix
21
Morton,M.ScottFiona&Shapiro,Carl.2014."StrategicPatentAcquisitions"79AntitrustLaw
Journal.463,482-483.Note.2005."TheDisclosureFunctionofthePatentSystem(orLackThereof)"118HarvardLaw
Review.2007,2025-2026.Ouellette,LarrimoreLisa.2012."DoPatentsDiscloseUsefulInformation?"25HarvardJournal
ofLaw7Technology.531.Rantanen,Jason.2012."PeripheralDisclosure"74UniversityofPittsburghLawReview.1.Reinganum,F.Jennifer.1989."TheTimingofInnovation:Research,Development,and
Diffusion"1HandbookofIndustrialOrganization,eds.RichardSchmalensee&RobertD.Willing.849.
Roin,N.Benjamin.2009."PatentableDrugsandtheStandardsofPatentability"87TexasLaw
Review.503,557-558.Sag,Matthew.2016."IPLitigationinUnitedStatesDistrictCourts:1994to2014"IowaLaw
Review(forthcoming).Manuscriptat17,availableathttp://ssrn.com/abstract=2570803.Shapiro,Carl.2006."PriorUserRights"96AmericanEconomicReview.Sichelman,Ted.2010."CommercializingPatents"62StanfordLawReview.341,345.Spulber,F.Daniel.2012."IntellectualPropertyandtheTheoryoftheFirm"Perspectiveson
CommercializingInnovation,eds.F.ScottKieff&TroyA.Peredes.9,31.Teece,J.David.1986."ProfitingfromTechnologicalInnovation:ImplicationsforIntegration,
Collaboration,LicensingandPublicPolicy"15ResearchPolicy.285,293.Tirole,Jean.1988.TheTheoryofIndustrialOrganization,TheMassachusettsInstituteof
TechnologyPress.394-399.Vermont,Samson.2006."IndependentInventionasaDefensetoPatentInfringement"105
MichiganLawReview.475.
