PARADIGMOF INTERNATIONAL
THEECLECTIC
PRODUCTION:
A RESTATEMENT
AND SOMEPOSSIBLEEXTENSIONS
JohnH. Dunning*
Universityof Reading and Rutgers University
Abstract.Thisarticlereviewssome of the criticismsdirectedtowards
the eclecticparadigmof international
productionoverthe pastdecade,
and restatesits maintenets.The secondpartof the articleconsiders
a numberof possibleextensionsof the paradigmand concludesby
assertingthat it remains"a robustgeneralframeworkfor explaining
andanalysingnot onlythe economicrationaleof economicproduction
andimpactissuesin relationto MNEactivity
but manyorganisational
as well."
INTRODUCTION
The concept of the eclectic paradigmof internationalproduction'was first put
forwardby the presentauthorin 1976 at a presentationto a Nobel Symposium
in Stockholmon TheInternational
Allocationof EconomicActivity.2
The intention
was to offera holisticframeworkby which it was possibleto identifyand evaluate
the significanceof the factorsinfluencingboth the initialact of foreignproduction
by enterprisesand the growthof such production.The choice of the word eclectic
was an ambitiousyet deliberateone. It was meant to convey the idea that a full
activitiesof enterprisesneedsto drawupon several
explanationof the transnational
strandsof economic theory;and that foreigndirect investmentis just one of a
numberof possiblechannelsof international
economicinvolvement,each of which
is determinedby a numberof commonfactors.
It is acceptedthat, preciselybecause of its generality,the eclectic paradigmhas
only limitedpowerto explainor predictparticularkindsof international
production;
and even less, the behaviourof individualenterprises3.
But this deficiency,if it
is a deficiency,whichsome criticshavealleged,could no less be directedat attempts
to formulatea generalbut operationallytestableparadigmof internationaltrade.
The classicaland neoclassicaltheoriesof trade, for example, while still having
wide explanatorypowersfor most kindsof inter-industry
tradeare quiteinadequate
to explainmuch of intra-industry
trade4.Indeedit is perhapsworth emphasizing
that the point at which the Heckscher-Ohlin-Samuelson
(H-O-S) theoryof trade
fails is preciselythat at which the modernparadigmof internationalproduction
starts,namelythe pointat which thereare positivetransactioncostsin intermediate
* JohnH. Dunningis currentlySethBoydenDistinguished
Professorof International
Business
at RutgersUniversity.On returnto the Universityof Readingin 1988, he will take up
an ICI ResearchProfessorship
in International
Business.
Commentson an earlierdraft of this paper by Mark Casson, John Cantwelland three anonymousreferees
aregratefullyacknowledged.
Received:June 1985; Revised:September1986 & March1987;Accepted:June 1987.
1
Palgrave Macmillan Journals
is collaborating with JSTOR to digitize, preserve, and extend access to
Journal of International Business Studies
®
www.jstor.org
2
BUSINESSSTUDIES,SPRING1988
JOURNALOF INTERNATIONAL
goods markets.5The differencebetween the neo-technologyand other modern
theoriesof trade and those of internationalproductionis that, while the former
implicitly assume that all goods are exchangedbetween independentbuyersand
sellers across nationalfrontiers,the latterexplicitlypostulatethat the transferof
intermediateproductsis undertakenwithin the same enterprises.In other words,
production
marketfailure,the raisond'etrefor international
withoutinternational
disappears.But once it exists,explanationsof tradeand productionmaybe thought
of as a part of a generalparadigmbased upon the internationaldispositionof
intermediate
modalitiesfortransacting
factorendowments,andthecostsof alternative
productsacrossnationalboundaries.Thisis the centralthemeof thispaper.
CRITICISMSOF THE ECLECTICPARADIGM
Are Competitiveor OwnershipAdvantages Necessary to Explain
International Production?
In its originalform,the eclecticparadigmstatedthatthe extent,form,and pattern
of internationalproductionwas determinedby the configurationof three sets of
First,in orderfor firmsof one nationality
advantagesas perceivedby enterprises.6
to competewith those of anotherby producingin the latter'sown countries,they
must possesscertainadvantagesspecificto the natureand/or nationalityof their
ownership.These advantages sometimescalled competitiveor monopolistic
advantages must be sufficientto compensatefor the costs of settingup and
operatinga foreignvalue-addingoperation,in additionto thosefacedby indigenous
producersor potentialproducers.
In our 1976 paperwe identifiedthreetypes of ownership-specific
advantages:(a)
those that stem from the exclusiveprivilegedpossessionof or access to particular
income generatingassets,(b) those that are normallyenjoyedby a branchplant
comparedwith a de novo firm,and (c) thosethatarea consequenceof geographical
diversificationor multinationality
per se.7 In a later typology (Dunning 1983a,
1983b), we distinguishedbetweenthe asset (Oa) and transaction (Ot) advantages
of multinationalenterprises(MNEs).8While the formerarisefrom the proprietary
ownershipof specificassetsby MNEs vis-a-visthosepossessedby otherenterprises
(i.e., of type (a) above, which can only occur in a situationof structuralmarket
the lattermirrorthe capacityof MNE hierarchiesvis-a-visexternal
distortions)9,
benefits(or lessenthe transactional
costs)arising
marketsto capturethe transactional
from the common governanceof a networkof these assets,locatedin different
countries.
The distinctionbetween structural and transactional marketimperfectionsis an
importantone (Dunning and Rugman 1985). Clearlythe relevanceof each in
determiningthe ownershipadvantagesof MNEs will vary according to the
of firms,theproductstheyproduce,themarketsin whichtheyoperate,
characteristics
and whetherthe competitiveprocessis viewedfroma staticor dynamicperspective.
Certainlyearlieranalysesof foreigndirectinvestment- particularlythose of the
Hymer(1960, 1976)tradition tendedto emphasizetheformerkindof imperfection;
but, similarly,so do contemporaryeconomistsworkingin the area of innovation
THE ECLECTICPARADIGM
3
and technologicaldevelopment(e.g., Pavitt 1987, Cantwell 1986); and business
enterprises
analystsseekingto identifythe systemicadvantagesof globally-oriented
(e.g.,Pralahadand Doz 1987;Kogut1983, 1985a).By contrast,the moderntheory
of the MNE qua MNE (e.g., as summarizedby Teece 1986 and Casson 1987)
tends to emphasizetransactionalmarket failure as the main raison d'etre for
internationalproduction.The two kinds of imperfectionare, of course, often
particularlyin a dynamicmarketsituation;'0and there is a growing
interrelated,
consensusthat the most successfulMNEs are those that are best able to nurture
ownershipadvantages.
and exploitbothassetand transactional
The second conditionfor internationalproductionis that it must be in the best
advantagesto transferthem
interestsof enterprisesthat possessownership-specific
across nationalboundarieswithintheir own organizationsratherthan sell them,
or their right of use to foreign-basedenterprises.This immediatelysuggeststhat
MNEs perceivethat the internationalmarketplace is not the best modalityfor
of
transactingintermediategoods or services.The reasonsfor the internalization
Sufficeto reiterate
detailin the literature."I
marketshasbeenexploredin considerable
here that three main kinds of marketfailureare usually identifiedas: (i) those
that arise from risk and uncertaintyas, for example, those succinctlyanalyzed
by Vernon(1983); (ii) those that stem from the ability of firms to exploit the
economies of large-scaleproduction,but only in an imperfectmarketsituation;
and (iii) those that occur where the transactionof a particulargood or service
yields costs and benefitsexternalto that transaction,but that are not reflected
The desireby firmsto integrate
in the termsagreedto by the transactingparties.12
or
differentstagesof the value-addedchain, to engage in productdiversification,
assets(Teece1986), originate
to capturethe economiesof the use of complementary
from the presenceof one or other of these forms of transactionalmarketfailure
may be expressedratherdifferently
even thoughthe motivesfor internalization
(e.g., to safeguardsuppliesof essentialinputs,to ensurethe qualityof end products,
to
to guaranteemarkets,to protectpropertyrights,to allow price discrimination,
spreadthe costs of sharedoverheadsand so on). The greaterthe perceivedcosts
marketfailure,the moreMNEsarelikelyto exploittheircompetitive
of transactional
productionratherthanby contractualagreements
advantagesthroughinternational
costs of hierarchies
with foreignfirms.By contrast,the higherthe administrative
and/or the externaldiseconomies(or disbenefits)of operatinga foreignventure
(e.g., as shown by the Bhopal disaster),the more probablethe lattervehicle (or
at leasta jointlysharedequitystake)will be preferred.
In such cases where there is no externalmarketfor the competitiveadvantages
advantagesmay
of MNEs, the distinctionbetween ownershipand internalization
Indeedsome writers(notablyBuckleyandCasson1985;andCasson
seemirrelevant.
1987) have arguedthat the failureof internationalintermediateproductmarkets
is both a necessaryand sufficientcondition to explain the existenceof MNEs.
Yet we believe it is not only useful but logically correctto distinguishbetween
the capabilityof MNEs to internalizemarkets,and theirwillingnessto do so. For
while the lattermay explainwhy hierarchiesratherthan externalmarketsare the
vehicle by which transactionalownershipadvantages(Ot) are transferredacross
nationalboundaries,it is the formerwhich explains why these advantagesare
4
JOURNALOF INTERNATIONAL
BUSINESSSTUDIES,SPRING1988
exploitedby one groupof MNEsratherthananother,or by MNEsratherthan
firmsindigenousto the countryof production.'3
This point has in fact been
acknowledged
by Casson(1986a,p. 46).
Certainlyin the exploitation
of specificintangible
assets(Oa) (e.g.,a patentor
trademark),firmsoftenhavea choicebetweenusingthe externalmarketor not.
Herethe distinction
betweenassetgeneration,or acquisition,and assetusageis
an important
one. We wouldacceptwithRugman(1981) that,if an ownership
is eithercreatedby or becomesthe exclusivepropertyof a particular
advantage
it has in some sense"internalized"
the marketfor its use;'4but we
enterprise,
believethis to be a questionable
extensionof the interpretation
of a termthat
andquitespecifically
wasintended
to conveya response
to transactional
originally
ratherthanstructural
marketfailure.'5
Locational Advantages: Structuraland TransactionalMarket Failure
Thethirdstrand
isconcerned
oftheeclecticparadigm
withthe"where"
ofproduction.
Enterprises
will engagein foreignproduction
whenevertheyperceiveit is in their
best intereststo combinespatiallytransferable
intermediate
products'6
produced
in the home country,with at leastsomeimmobilefactorendowments
or other
intermediate
productsin anothercountry.While,in the eclecticparadigm,
the
or disadvantages
of particular
advantages
locationsaretreatedseparately
fromthe
of particular
and while the marketfor these
ownershipadvantages
enterprises,
the decisionon whereto sitea mine,factoryor office,
areinternalized;
advantages
of the ownership
of theseassetsnor of the routeby which
is not independent
thechoiceof locationmaybe prompted
theyortheirrightsaretransacted.
Similarly,
market
failure:
the
has led to
by spatial
historically impositionof tradebarriers
a lot of foreignmanufacturing
investment
by MNEs.At thesametimea reduction
in transport
costsandthe formation
of customsunionsor regionaltradingblocs
of production
greater
regional
specialization
(e.g.,EECandLAFTA)haveprompted
MNEs
by
(Dunning1987b).
Oncemorea distinction
needsto be drawnbetweenthe different
kindsof market
thatmayinfluence
thelocational
decisions
of MNEs.Structural
imperfections
market
distortions e.g., thosearisingfromsome (but not all) kindsof government
whichaffectthe costsand/or revenuesof producingin differentintervention,17
ordiscourage
inwarddirectinvestment
locations mayeitherencourage
(Guisinger
MNE activitywould
1985). On the otherhand,even withoutsuch distortions
therearetransaction
stilloccurwherever
gainslikelyto resultfromthe common
of activitiesin different
locations.Suchadvantages
includeenhanced
governance
the
reduction
of
and
arbitrage leverageopportunities,
exchangerisksand better
of financialdecisiontaking,the protectionaffordedby a hedged
coordination
of gainsthroughtransfer
andthepossibility
or multiplesourcing
marketing
strategy,
andso on (Kogut1985b,Dunning
leadsandlagsin payments,
pricemanipulation,
1987a).
The abilityto generateand sustainsuchownershipadvantages
itselfstrengthens
the competitivepositionof MNEs vis-a-visuninationalfirms. But because
it has locational
marketfailureis sometimescountry-specific,
transactional
implications
as well.To thisextent,Rugmanis on the rightlineswhenhe refers
THE ECLECTICPARADIGM
5
to the MNE as "internalising
exogenousspatialimperfections"
(Rugman1981);
but his analysisbetterexplainsthe commonownership
of MNE subsidiaries
in
different
locations,ratherthanwhyparticular
subsidiaries
arelocatedwherethey
are.
Specificor GeneralTheoriesof InternationalProduction?
Itisthenthejuxtaposition
oftheownership-specific
offirmscontemplating
advantages
in foreignproduction,
foreignproduction,
oranincrease
thepropensity
to internalize
the cross-border
marketsfor these,and the attractions
of a foreignlocationfor
production
whichis the gistof the eclecticparadigm
of international
production.
Buttheidentification
andvalueof thespecificownership,
locationandinternalisation
thatwillinfluenceindividual
(OLI)parameters
MNEsin anyparticular
production
decisionwill vary accordingto the motivesunderlyingsuch production.The
a MNE to investin a copperminein New Guineaare
parameters
influencing
unlikelytobe thesameasthoseinfluencing
investment
colortelevision
by a Japanese
companyin the UnitedStates;whilethosedetermining
the patternof rationalized
production
in theEECby a largeandgeographically
US motorvehicles
diversified
MNEwillbe different
froman investment
a
Korean
construction
by
management
companyin Kuwait.
However,the eclecticparadigmdoes allowone to go a stepfurtherby relating
theOLIconfiguration
of structural
facingMNEstoa number
orcontextual
variables.
We havepreviouslyidentifiedthe moreimportantof theseas country,industry
(oractivity)andfirm-specific
(Dunning(1981)).Forexample,theassetadvantages
MNEsmaybeexpected
(Oa)ofparticular
tovaryaccording
tothefactorendowments
and othercharacteristics
of the countriesfromwhichthey originate,and/or in
whichthey operate:and the technological
and otherfeaturesof the activitiesin
whichthey engage.The factthatsuchassetsmay be the exclusivepropertyof
particular
firms,and be mobileacrossnationalboundaries,
does not negatethe
possibilitythattheirsourcemaybe explainedby the international
of
disposition
and immobileendowments.
To thisextent,one is backto H-Ocountry-specific
S type tradetheory,but with two differences.
The firstis thatthe goods and
servicestradedareintermediate
ratherthanfinalproducts;
andthe secondis that
theeclecticparadigm
allowsfortheroleof governments,
in affecting,
by thepolitical
systemstheyoperateandthe economicpoliciestheypursuethe real(as opposed
to thepotential)
valueof theresources
containedwithintheirjurisdictional
areas.18
Severalwriters,e.g., Franko(1976), Ergas(1984), Davidson(1976) and Pavitt
(1987), havedemonstrated
thatthe kindof innovatory
advantages
generated
by
MNEsreflecttheresource
endowments,
andinstitutional
markets,
culture,attitudes
framework
of theirhomecountries.19
It requires
buta smallmodification
of the factorendowment
approach
to explain
whysometypesof economicactivityaremoreproneto internalization
thanothers.
Againthe spatialdisposition
of resourceendowments
and international
transport
costsarethekeyvariables.
If thecapacityto createa particular
assetis ubiquitous,
andtherightto itsusecanbe disseminated
atzerocost,theninternational
production
is unlikely;it is also improbable
wherethe competitive
advantages
of firmsrest
not in the exclusivepossessionof specificassets,but in the accessto immobile
6
JOURNALOF INTERNATIONAL
BUSINESSSTUDIES,SPRING1988
but nonspecificfactorendowmentson favorableterms.Thus, a combinationof
the resourcerequirementsof particulareconomic activities,their geographical
disposition,and the transfercosts of their output, helps to explain some of the
operationsof MNEs.
But only some!Forexample,it does littleto explainthe cross-hauling
of investment
in the same industriesby MNEs of differentnationalities;or the fact that some
countriesdisplaysimilarpatternsof internationalproduction.The explanationis
limitedbecauseit ignorestransactionalmarketfailure,which itselfvariesbetween
countriesand typesof economicactivities.Withoutsuchfailure,but with an uneven
distributionof resourceendowments,trade in intermediateproductswould be
conducted through external markets. With an even distributionof resource
endowments,but with marketfailure,then the only advantagewhich MNEs qua
MNEs possess is their capabilityto better overcome internationaltransactional
imperfections
thantheiruninationalrivals(Dunning1986c).
Does the EclecticParadigm InsufficientlyAllow for Firm-Specific
Behavioural Differences?
We now turnto considera structuralvariable,which some businessanalystsregard
as the most crucial of all in influencingthe level and patternof international
production.This is the strategicresponseof decision takerswithin MNEs to a
set of economic and other variables;and the way the idosyncraticbehaviourof
firmsmightinfluenceand respondto cross-border
marketfailure.
A cursoryreview of the internationalprofilesof the leadingMNEs identifiedby
Stopford(1982) revealsthat in some sectors(e.g., consumerelectronics,motor
vehicles, pharmaceuticals,etc.), there are as many differencesbetween the
characteristics
of MNEs in the same sectoras thereare betweenMNEs in different
sectors.Moreover,since these firms rarelysupply identicalproductsor the same
rangeof products,or produceon the same (or similar)pointsof the value-added
chain, or sell in the same markets;and since they have differingcapabilitiesfor,
and a need of, internationalproduction;it follows that not only are they faced
with a differentset of strategicoptions,but that theirevaluationof these options,
and the risks attachedto them, will vary. Indeed,the risk diversificationthesis
(Rugman 1979) asserts that different firms may view identical investment
opportunitiesofferedby a particularcountrydifferently,inter alia, accordingto
the distributionof their existingportfoliosand their attitudestowardsuncertainty.
For these, and other reasons identifiedin the business literature,firm-specific
characteristics
may be a crucial determinantof the responseby MNEs to any
particularOLI configuration.
While there have been some attemptsto model the strategicbehaviourof firms
towardstheir foreignoperations,20
they have not generallybeen incorporatedinto
the mainstreamof international
productiontheory.The exceptionsare the product
cycle, oligopolisticstrategyand risk minimizationmodels. The first two (Vernon
1974, Knickerbocker1973) look upon much of foreignproductionas a strategy
by firms to protector gain an ownership-specific
advantagevis-a-vistheir rivals;
the implicationbeingthat,in a more competitiveand less riskyenvironment,firms
would haveless impetusto engagein international
directinvestment(Vernon1983).
THE ECLECTICPARADIGM
7
of oligopolists
includethe bunchingof the timingof
Evidenceof suchstrategies
in somesectors(Dunning1986a;Knickerbocker
foreigninvestment
1973;Graham
1978, 1985;Lake 1976aand b). The riskminimization
hypothesisarguesthat,
otherthingsbeingequal,firmswill preferto diversifythe geographical
portfolio
nonfinancial
This conceptmay be extendedto incorporate
of theirinvestments.
portfoliobehaviour.
Clearly,whetheror not a firmadoptsa globalproductor
or choosesto engagein multiplesourcing
marketing
strategy,
(Kogut1983,1985b),
reflectsnot only on its abilityto do so (whichinteralia will be functionof its
buton its perceptions
andexistingoverseascommitments),
size,productstructure
of theresulting
costsandbenefits.
in the behaviourof firmsbe embracedby the
To what extentcan differences
The answeris theycan,insofaras it is possibleto identifyand
OLIframework?
of suchbehavior.
actions
evaluate
systematic
patterns
Purelyrandomoridiosyncratic
But no less is this true
MNEscannotbe so easilyincorporated.
by particular
a generalized
if one was attempting
firm.Suchtheories
theoryof the uninational
are reallytheoriesof the behaviourof firms in the
as aboundin the literature
of any particular
firmtheyaretryingto predict
sensethatit is not the behaviour
firmof thatgroup.
butthatof a groupof firms,or of a representative
(or average)
Theyusuallyassumetwo things.First,thatfirmshavebroadlysimilargoals;and
second,thattheyrespondto economicsignalsto advancethesegoalsin a rational
and consistent
way.Whenneitherconditionexists,it is not possibleto offerany
of behaviour;
generalized
explanations
which,indeed,is exactlywhatsomebusiness
analystswouldclaim.
Wedo notacceptthatsucha drasticcourseis eitherdesirable
orjustifiable;
indeed,
we believethat,for most firms,thatpartof businessconductwhichis purely
idiosyncratic
is probablyverysmall.However,
we arepersuaded
thatthe interface
betweentheeconomicandbehavioral
theoriesof thefirmdoesneedmoreexplicit
andsystematic
analysis.Whilethereis generalagreement
aboutthe maincountry
and industrycharacteristics
likelyto influenceeach of the maincomponents
of
the eclecticparadigm,
muchless attentionhas beengivento identifying
the key
attributes
of firms- andespeciallythosethatmightbe identifiedas operational
orstrategically
based- thatmayaffecttheirresponse
to anyparticular
configuration
of OLIparameters.
Thereare now signsof this happening.
Some of the recentliterature
on global
of business21
dimensions
is repletewithattempts
to identifythestrategically-related
characteristics
of firmsmost likelyto be associatedwith a robustinternational
posture.Theseincludetheirlong-term
goalsandperspectives,
thenatureandscope
of theircoreassets,theirattitudeto innovation
and change(aretheyleadersor
in theirindustry,
followers
aretheyinnovators
or imitators?),
therangeandsegment
of criticalmarketsserved,theirattitudeto riskand uncertainty,
theiroperational
theirorganizational
and culturalethos,the entrepreneurial
flexibility,
initiativeof
theirchiefdecisiontakersand theirwillingness
and capacityto concludecrossborderalliances.
Therehasbeenlittleempirical
ontheseandotherbehavioral-related
research
variables
in influencing
the extentand patternof international
production.
In 1972 Horst
concludedthat,apartfromsize,he couldidentifyno firm-specific
variablewhich
8
BUSINESSSTUDIES,SPRING1988
JOURNALOF INTERNATIONAL
of US-ownedfirmsacross
satisfactorily
explainedthe degreeof multinationality
on the modes of
industriesand countries.More recentlya studyundertaken
transferring
technologybetweencountriesbut within firms (Davidsonand
McFeteridge1984, 1985) revealedthat such variablesas existingoverseas
research
anddegreeof product
diversification
werepositively
commitment,
intensity,
withtheextentof a firm'sinternalization.
andsignificantly
correlated
Mostrecently
of all,Porter(1986)hasdeveloped
a modelthatrelatestheextentto whichdifferent
theircross-border
with
value-adding
investments,
typesof firmsseekto coordinate
to centralise
or decentralize
Porter
thepropensity
thelocationof theseinvestments.
a globallyorgeocentrically-oriented
MNEasonethatoperates
anextensive
describes
networkof foreignaffiliates,
the activitiesof whichare subjectto a highdegree
froma multiof centralized
coordination.
Suchan MNE is to be distinguished
overseas
domesticcompany,
subsidiaries,
pursues
that,throughitslooselyorganised
a seriesof countrycenteredstrategies;
or, indeed,fromone thatadoptsa simple
concentrated
of activities.22
witha geographically
strategy
configuration
In conclusion,
it maywell be thatthereare somebehavioral-related
variables
of
intotheeclecticparadigm.
Insofar
firmsthathavenotbeensuccessfully
incorporated
as it is possibleto identifythosethatmightinfluencethe responseof groupsof
thereis no reasonwhy thiscouldnot
to a givenOLI configuration,
enterprises
orconsistent
of firmsto changes
be done.Butwhereno generalsystematic
response
in exogenousvariablescan be discovered,
any attemptto generalizeaboutthe
is thwarted
fromthestart.
causesof international
production
The Aliber Theoryof ForeignDirect Investment
withthe eclectic
Let us now brieflyturnto RobertAliber's(1983)dissatisfaction
paradigmand, indeed,with all theoriesthattake some measureof the foreign
activities
of enterprises
as theirstarting
ThisreflectsAliber'sview
pointof interest.
of a MNEis notthefactthatit engagesin foreignproduction,
thatthekeyattribute
in its homecurrency.
but thatit financesat leastpartof thisproduction
He is,
in theexportof directinvestment
as a meansof financing
interested
then,primarily
ratherthanas a channelby whichanenterprise
transfers
foreigncapitalexpenditure
nonfinancial
resources
betweencountries,
and controlsthe use of suchresources
once transferred.
He wouldappearto believethatthe extraterritorial
expansion
of firmsperse raisesno issuesnotalreadyaddressed
by thetheoryof thedomestic
firm.Rather,
of theMNEis itsabilityto dominate
itsgeographically
theuniqueness
ofstructural
assetsindifferent
andbysodoing,totakeadvantage
dispersed
currencies,
in international
markets.
or transactional
imperfections
capitalandforeignexchange
Inasmuchas scholarsare entitledto studysubjectsof interestto them,we have
hisimplicitassumption
that
no disputewithAliber.Wewould,however,
challenge
financial
in kindexistbetweennational
andinternational
whiledifferences
markets,
this is not the case for nonfinancial
markets,such as thesefor technologyand
hisconsideration
services.
it seemsto usthatAliberrestricts
management
Moreover,
in different
in
invest
of foreigndirectinvestment
which
to situations
enterprises
currencyareas.Whilethismaybe usuallythe case,it is by no meansuniversally
so.
THE ECLECTICPARADIGM
9
Inanyevent,we do notfindAliber'sthesisincompatible
withtheeclecticparadigm.
The veryfactthatfirms,by theirpresenceoverseas,maybe ableto denominate
couldgive thema competitive
or
theirassetsand goodsin differentcurrencies,
firms.Thisadvantage
willbe themore
an ownership-specific
edgeoveruninational
ininternational
ortransactional
pronounced
thegreater
thedegreeof structural
failure
andthebetterequippedMNEsareto internalize
capitaland/orexchangemarkets,
thesemarkets.Yet,by themselves,
theseadvantages
are not sufficientto explain
of foreigndirectinvestment.
Forexample,expected
eithertheamountordistribution
of imperfectfinancial
profits(otherthanthoseresultingfromthe internalization
of the locationsin whichinvestments
are made,
markets)are not independent
norof theabilityof MNEstoappropriate
economicrentbyinternalizing
nonfinancial
markets.
It is not our purposeto offera detailedcritiqueof the Aliberhypothesis,23
but
do existin the marketsin which
ratherto suggestthat,insofaras imperfections
he is interested,
thesemayaffectboth the way in whichcapitalexpenditure
by
of
international
MNEsis financed,and the geographical
distribution
production.
in the eclecticparadigm
as influencing
the foreign
the factorsidentified
Similarly,
activitiesof firms,maydirectly,
by theirimpacton capitalandexchangemarkets,
and/or,indirectly,
by affectingthe totalcapitalexpenditure
by MNEs,haveno
less a bearingon theirfinancingof theseactivities.We wouldthen assertthat,
insupport
ofhisowntheory,
ofthenonfinancial
Professor
Alibermusttakecognizance
of firms.
aspectsof theinternational
operations
TheKojimaHypothesis24
As originally
propounded
(Kojima1978,1982),Professor
Kojima's
theoryof foreign
is an extensionof the neoclassical
directinvestment
theoryof tradeto embrace
of intermediate
transactions
cross-border
products(e.g.,technology,
management
a normative
skillsetc.).It is primarily
theory,andviewstheMNEas an instrument
of nationstatesmaybebetteradvanced.
by whichthecomparative
trading
advantage
Hencehis prescription
thata home countryshouldinvestabroadin sectorsthat
requireintermediate
(butinternationally
mobile)productsthatit is comparatively
well suitedto supply;but thatneedto be combinedwith nontransferable
inputs
in whichthe host countryis relatively
well endowed.In thiscase,foreigndirect
investment
acts both as a catalystto tradeand as an arbitrager
for improving
of economicactivity.
allocation
theinternational
Kojimacriticizesthe eclecticparadigm
forbeinigtoo micro-or business-oriented,
and claimsit is of limiteduse for policyformation
by homeor host countries.
fromtheperspective
of theUnitedKingdom
But,as we havesoughtto demonstrate
of our paradigm
(Dunning1981,chap.6), manyof the normative
implications
areentirelyconsistent
withKojima's
Thisis particularly
recommendations.
thecase
for resource-based
and importsubstitution
investment,where the export of
intermediate
products
by MNEsto countries
bestsuitedto engagein furthervalue
addedactivities,eithercircumvents
artificially
imposedimpediments
to trade,or
betterpromotes
thedynamiccomparative
of theparticipating
countries.
advantage
However,evenas a prescriptive
macroeconomic
model,the Kojimaapproachis
deficientin two majorrespects.First,sinceit is neoclassical
in its stance,it can
10
JOURNALOF INTERNATIONAL
BUSINESSSTUDIES,SPRING1988
neitherexplain,nor evaluatethe welfareimplications
of thosetypesof foreign
directinvestment
promptedby the desireto rationalize
international
production
and to benefitfromthe commongovernance
of cross-border
activities(i.e., Ot
The eclecticparadigmcan and does embracesuch international
advantages).
production.
Second,and relatedto the first,Kojimalargelyignoresthe essential
of MNEactivity- namely,theinternalization
characteristic
of intermediate
product
and wherehe does takethisinto account,he alwaysseemsto assume
markets;
thatthe resulting
of resources
allocation
is lessdesirable
to thatwhichwouldhave
beendictatedby the market(Kojima1978,chap.9). Thisis becauseKojimais
lockedinto a neoclassical
paradigm
of perfectcompetition
thatnegatesthe very
possibilityof marketfailure.In his scenario,the MNE can neverbe the most
efficientagentfortransferring
resources
acrossnationalboundaries,
simplybecause
itsveryexistence
impliesa second-best
transactional
situation.
Again,thisdoesnot seemto be of muchpracticalvalueto governments
in their
formulation
of policytowardsMNEs.Firmsdo not existin a risklessor timeless
vacuum;manyindividualtransactions
do give riseto externalcostsor benefits;
the exploitation
of economiesof scalemaynot be possiblewithoutthe presence
of somestructural
marketdistortion;
someproductdifferentiation
maybe desirable;
and somepropertyrightsmayrequire,at leasttemporary
protection
againsttheir
if
or
infringement dissipation, they are to be suppliedat all.25The questionat
issueis surelythat,giventhe viablealternatives,
and overan appropriate
timeperiod,
can the resourceallocationbetweencountriesbe improvedby foreigndirect
of MNEs?
investment
or theoperation
theallegeddichotomy
of Japanese
betweenthepatterns
Empirically,
andU.S.direct
is a falseone. As Mason(1980) has well argued,suchdifferences
investment
as
do existreflectthe differentstagesin the evolutionof Japaneseand American
MNEsas muchas anythingelse. The eclecticparadigmwouldsuggestthat,in
a worldfreeof traderestrictions,
theinitialact of foreigndirectinvestment
would
in whichtheinvesting
occurin thosesectorsthatuseintermediate
normally
products
Thisactwouldbe welfare-creating
wherever
countryhasa comparative
advantage.
its socialopportunity
the pricechargedforthe intermediate
outputfairlyreflected
releasedwithinthe homecountryaredeployedin a way
cost,and the resources
of comparative
withtheprinciple
Wewouldacceptthatmost
consistent
advantage.
of the 1960sand '70s was of this kind.However,
Japaneseforeigninvestment
and takea moreglobalperspective
of their
as firmsbecomemoremultinational
on
theirownership
becomelessbased theexclusive
foreignoperations,
advantages
in origin,and
of particular
possession
intangible
assets,whicharecountry-specific
ofgeographical
moreontheirabilitytosuccessfully
coordinate
andmanagea network
- whichtendto be firmactivities.
Thesetransaction
costs-minimizing
advantages
in origin- werelargelythe propertyof the larger
ratherthancountry-specific
US and European
MNEsin the 1960sand 1970s;only now,in the later1980s,
to be exploited
aretheybeginning
by theirJapanese
counterparts.
A RESTATEMENT
OF THEECLECTIC
PARADIGM
In consideration
So muchfor some of the criticismsof the eclecticparadigm.
of these,and on furtherreflection,
we are now fullypersuaded
thatany holistic
THE ECLECTICPARADIGM
11
theoryof international
productionmustdrawupon two inter-related
strandsof
Thefirstis theneoclassical
economicanalysis.
extended
theoryof factorendowments,
to embraceintermediate
products,and to allow for the possibilitythat some
endowments
aremobileacrossnational
boundaries.
Ceteris
themoreuneven
paribus,
thegeographical
distribution
offactorendowments,
themoreinternational
production
is likelyto takeplace.The natureof suchproduction
will resemblethatof Hin character.
O-S tradein thatit is inter-industry
The secondstrandis the theory
of marketfailure,whichis relevantto explaining
not only the locationof some
kindsof economicactivityacrossnationalboundaries,
but also the divisionof
thatactivitybetweenmultinational
and uninational
firms.26
Ceterisparibus,the
higherthe transaction
costsof usingthe marketas a transactional
model,and
the greaterthe efficiencyof MNEsas coordinators
of geographically
dispersed
themoreinternational
is likelyto takeplace.Suchproduction
activities,
production
in character;
maybe eitherinter-or intra-industry
butthatbasedon Ot advantages
aloneis morelikelyto be of thelatterkind(DunningandNorman1985).
In Figure1, we set out the relationship
betweenthesetwo intellectual
strands
and the analyticalconstructs
in
set out thispaper.We believethis figureis selfand needsno furtherelaboration.
explanatory
the relevance
Figure2 illustrates
of thesetwo basicelementsof the eclecticparadigm
in explaining
the threemain
kindsof international
We alsowouldreiterate
production.
an earlierobservation,
thatas an enterprise
whichit treatsas
developsa networkof foreignaffiliates,
partof a globalsystemof activities,
the relativeimportance
of factorendowments
in explainingchangesin international
is likelyto decrease;and that
production
of marketfailurelikelyto increase.27
SOME POSSIBLE EXTENSIONS OF THE ECLECTICPARADIGM
It is ourcontention
thattheeclecticparadigm
a richandrobustframework
provides
not onlyforanalyzing
andexplaining
thedeterminants
of international
production
and how thisvariesbetweenfirms,industries,
and countries,
and overtime;but
forourunderstanding
of a widevarietyof otherMNE-related
issues.In thispaper,
six possibledirectionsin whichworkon the paradigmmight
we will illustrate
be further
developed.
A More Formal Modeling of the Paradigm
Thereis need for a moresystematic
and rigorousmodelingof the explanation
of differenttypes of international
productionby the use of specificallyand
operationally
testableOLI parameters.28
Giventhesetypes,the variations
within
themmaybe explained
by structural
variables
as identified
by empirical
research.29
Some workby tradeeconomists,
notablyEthier(1986),Markussen
(1984) and
Helpman(1984),is currently
proceeding
in thisdirection.
Thereis also needfor
formalmodelingof the MNE as an organizational
mechanism
and/orchoiceof
modalityof resource
transfer.
Againsomeprogress
hasbeenmadeby Grosse(1985),
Casson(1985) and Horstmanand Markussen
(1986), but moreworkrequires
to be done. Finally,thereis need for a moresystemicapproachto examining
thestrategic
behavior
of MNEs- using,forexample,suchtoolsas gametheoretic
andnetworkanalysis.
12
JOURNALOF INTERNATIONAL
BUSINESSSTUDIES,SPRING1988
FIGURE 1
The Endowment/Market Failure Paradigm
of International Production
O = Ownership advantages
L = Locational advantages
I = Internalisation advantages
represents
Oa = asset advantages
Ot = transaction advantages
linking
FACTOR ENDOWMENTS
POLICY
represents
interaction
lSYSTEM
/~~~~~~~
Intermediate
T
- -
products
Immobile
-
Mobile
FAILURE
MARKET
Structural
Government
Intervention
Entry
Barriers
I
Transactional
__
Spatial
Common Governance
(Externalities)
/ s s
/
s
'
,<
4
-
I
L
Transport
Costs
-
Etc
Production
Costs
/
_
0
-
t. _ _
>Ot.
_ _ _ eO
Scale
/
Cognitive
(Risk, lack of information)
~
~/
Effective Management
Control
of
~~~~~~~~~~~~~~~~
Control
<
~~~~~~~~~~~~~~
~~~~~~~~~~~~~
I
-
Etc
Discrimination
InvestEtc
envestm6nt
Avoidance of
buyer uncertainty
International
Arbitraging
//
Tariff Barriers
Psychic
Distance
Market
Access
Economies
STRUCTURAL
VARIABLES
Developed - Developing
Large- Small
Degree of Industrialisationetc
of
Avoidance
Aodneo
Etc
of
property right
infringement
~~~~~~~~Trademarks
High- Low Technology
Innovatory- Mature
Processing- Assembly
Competitive- Monopolistic etc
Size
Age
Strategy
Leader- Follower
Innovator- Imitator etc
THE ECLECTICPARADIGM
13
FIGURE2
Illustration of Use of Factor Endowment/Market Failure Paradigm in Explaining
Three Main Forms of International Production
Main Types of
International
Production
Factor Endowments
(Affecting geographical
distributionof L)
Market Failure
Structural
Transactional
(Affecting L and Oa)
(Affecting Ot, L and 1)
1 Market Seeking
(importsubstituting)
Home country for creation of Oa (= mobile
endowments/intermediate products)
Host country advantage
in immobile endowments with which Qa
Firm specific =
proprietaryOa (eg.
Knowledge) privileged access to
Inputs
Restrictions on trade
in goods
Search and negotiating costs
Protection against
misrepresentation or
infringement of property rights
Economics of bulk
(a) natural
purchasing
(b) artificial
portfolioto spread
Oligopolistic market
structure
Protection against
actions of
competitors
have to be used eg. natural resources, some
(transportcosts)
Part of international
kindsof labourPatoineainl
Market size & character
(importcontrols)
risks
2 Resource Seeking
(supply oriented)
Home country - as
above but also market
size & character
Host country.Availability
of resources, natural,
labour (export processing) technology (eg.
investment by ldcs in
dcs)
As above, but also
privileged access to
markets
Incentives offered
by Government to
fdi (also relevant for
1 & 3)
Oligopolistic market
structure
Avoidance of risks of
breach of contract
and interruptionof
supplies
Absence of future
markets
Economics of vertical
integration
3 Efficiency Seeking
(rationalised
investment)
VERTICAL
Mainly as 1 & 2 above
HORIZONTAL
As above but as
investment influenced more by
As with 2 above
Economies of scale
and scope
of
Usuallydistribution
factorendowmentsnot
supplythan market
considerations.
Government
Riskreduction
throughproduct
very relevant, as interO
Ownership
advantagePs
L =L a-dvantages
Locational
advantages
advantages
Oa= asset
advantages
Ot = transaction
advantages
O
Ownership
advantages
L= Locational
advantages
I= Internalisation
advantages
Oa asset
advantages
Ot= transaction
advantages
nationalproductionin
countrieswith similar
resountrieswih
st ures
resource structures
induced structural
imefconlkly
likely
imnperfections
to
be of considerable importance eg.
LATERAL
tax differentials,
Of limited importance in
effect
investment incentives, performance
requirements etc.
Note that as above
regional integration
and reduction of
trade barrieraids
rationalised
investment
diversification
Aabv,utn
above, butin
|As
respect ofeg.
ancillary
activities
various
serivicies eg shripping
consultancy etc.
14
BUSINESSSTUDIES,SPRING1988
JOURNALOF INTERNATIONAL
Dynamic and Development Aspects of International Production
Some commentators(e.g., Vernon1985) have allegedthat the eclecticparadigm
is couched in statictermsand is unableto explainthe dynamicsor the process
and modeled
production.Dynamicscan be interpreted
of changeof international
in variousways; Vernon'sparticularconcern is that the eclectic paradigmfails
to allow for the behavioralinteractionbetween internationaloligopolists,which
both affect and is affectedby their foreignactivities.In other words, faced with
the same set of OLI parameters,not only would the responseof MNEs vary
accordingto their strategicpostures;but this responsemight triggeroff reactions
on the part of their competitors,that themselvesmay cause a change in one or
other of these parameters.In the real world of uncertaintyabout futuremarkets,
the actionsof government,the conductof competitors,suppliers,consumersand
or industrially
those thatare geographically
laborunions,firms- and particularly
diversified- havea varietyof strategicoptions,simplybecausethey do not know
with certaintywhat is its best option. This is a very differentscenariofrom the
one assumedby the neoclassicalmodels, where, once the value of the relevant
parametersis known,the firstbest solutionis both identifiableand assumedalways
to be adoptedby MNEs.
The literatureidentifiesvariousfactorslikely to influencethe strategyof MNEs
towardstheir foreign operations.These include the structureof their existing
investmentportfoliosand riskexposures,theircompetitivestrengthsandweaknesses,
their bargainingpower with governments,their productportfolios,their liquidity
position and so on. However,these are at best partialbehavioralexplanations.
The crucialquestionis whethera generaltheoryof businessstrategycan be devised
that can be used alongsidethe eclecticparadigmto explainthe actionsof MNEs
in a dynamicsituation.Perhapsthe best hope for progresshere lies in some of
the concepts in industrialorganizationtheory, e.g., that of dynamic market
and forthe transactional
modelitselfto embracethe type of market
contestability;30
failureinherentin interactivebehavioralsituations.
A somewhatdifferentbutnonethelessrelatedinterpretation
of dynamicsmightsuggest
and
that the eclecticparadigmshouldembracethe economicsof entrepreneurship
technologicalinnovationand change.Mark Casson (1986a) has forciblyargued
that any satisfactoryexplanationof the dynamicsof ownershipadvantagemust
to a centralpositionin
of the role of the entrepreneur
rest on the reinstatement
the theoryof the firm.Like Casson,Cantwell(1986) and Dunningand Cantwell
(1986) view the economyas an evolutionarysystem,and, haveappliedthe eclectic
paradigmto analyzingthe way in which MNEs both generateand respondto
technologicalchange.Economicand businesshistorians,too, are makinga useful
of the growthof international
contributionto our understanding
producton,using
especiallya transactioncost approach(North1985;Nicholas1986).
Viewing growth and developmentfrom the perspectiveof countriesratherthan
firms,more progresshas been made, using mainly the tools of the development
economist.Here,the conceptof an investmentdevelopmentpath or cycle, as first
extendedin Dunning(1986c),andmodified
set out in Dunning(1979), subsequently
by Tolentino(1987) is especiallyrelevant.3"
THE ECLECTICPARADIGM
15
of theinvestment
Thebasichypothesis
development
pathor cycleis thata country's
or be investedin by foreign
propensity
to engagein outwarddirectinvestment,
to (i) itsstageof economicdevelopment,
firms,willvaryaccording
(ii) thestructure
and markets,(iii) its politicaland economicsystems,
of its factorendowments
of intermediate
and (iv) the natureandextentof marketfailurein the transaction
It suggeststhat,as a country'seconomic
productsacrossnationalboundaries.
its international
directinvestment
development
proceeds,
positionwillpassthrough
a numberof stages.In the firststage,therewill be neitherinwardnor outward
MNE activity,partlybecauseits marketsand factorendowments
are insufficient
or resource-based
inwardinvestment;
andpartly,
to attracteitherimportsubstituting
infrastructure
is unabletogenerate
commercial
andtechnological
becauseitspolitical,
thekindof supportservicesrequired
by foreigndirectinvestors
(or,forthatmatter,
As the infrastructure
firmsengagedin-similaractivities).
by indigenous
improves,
on the economicstructure
of the countryandgovernment
then,depending
policy,
intermediate
productswill startto be imported;however,becauseof the high
transaction
costsof usingexternalmarkets,thesewill tendto be internalized
by
Inthisfirststageof inwardinvestment,
theownership
theforeign
suppliers.
advantages
of MNEsare morelikelyto derivefromthe possessionof individualintangible
assets(Oa) (vis-a-visthoseof indigenousfirms),ratherthan on the economies
of coordinating
multipleactivities(Ot);but this will partlydependon whether
in othercountriesand the extentof its
the MNE alreadyhas relatedinvestment
intra-firm
trade(Dunning1986c).
is markedby the abilityof a developing
Thethirdstageof development
country's
firmsto generate
theirown ownership-specific
advantages,
which,initiallyat least,
of the country's
factorendowments.
are likelyto reflectthe structure
Depending
on the natureof theseadvantages,
the relativeattractions
of a foreignlocation,
thesefirmsmaygo abroadas marketor resourceand theirstrategicpriorities,
whereasin theirearlyventuresabroad,industrialized
seekers.However,
countries
andlow costlaborin whichtheirhomecountry
normallysoughtnaturalresources
wasdisadvantaged,
arecurrently
thosefromdeveloping
countries
seekingto acquire
technology(i.e.,the resourcein whichtheyarecomparatively
poorlyendowed).32
Alternatively,
developing
countriesmayexportthe kindof intermediate
products
in whichtheyhavecomparative
thatrequireendowments
In the case
advantage.
of SouthKorea,Turkey
andthePhilippines,
forexample,
thishassometimes
involved
the export of unskilledlabor services-
notably of constructionworkers-
an
intermediate
product,
traditionally
thoughtto be immobileacrossfrontiers.
It shouldbe notedthatthe pointat whicha countryreachesthe thirdstageof
the investment
development
cycle- if indeedit is reachedat all - restslargely
on the structure
of its resourceendowments,
and the attitudesof its government
in generaland inwardand outward
towardsinternational
economicinvolvement
in particular.33
A countrysuchas India,with its sightsset on
directinvestment
industrial
self-sufficiency,
mightwellprefer
toindigenize
activities,
initially
undertaken
in an international
ratherthanto participate
divisionof labor
by foreignaffiliates,
in whichits own firmsbecomeforeigninvestors.By contrast,economiessuch
asSingapore,
wouldseemtofavorbuilding
HongKong,andTaiwan,
upa comparative
in the production
of intermediate
advantage
products,that,in partat least,may
16
JOURNALOF INTERNATIONAL
BUSINESSSTUDIES,SPRING1988
best be usedin conjunction
with immobileresourcesin othercountries;
while,
at the sametime,theyseekto fosterinwardinvestment
in activitiesthatrequire
immobile
resources
inwhichtheyareevolving(e.g.,bywayof appropriate
education,
and
training innovating
policies)a comparative
advantage.
Thefourthstageoftheinvestment
development
cycleoccurswhena country
becomes
a netoutwardinvestor.
Sinceby definition
theoutwardcapitalstockor investment
flowsof all countriesmustequalthe inwardcapitalstockor investment
flows
of all countries,
it followsthat,at a givenmomentof time,only somecountries
can be net outwardinvestors.Therefore,
any correlation
betweennet outward
investment
and economicdevelopment
can only holdgood whenmakingcrosscountrycomparisons.
Usingtime-series
data,the correlation
maybe positivefor
somecountriesbut not for others.Thisproblem,however,maybe overcomeby
normalizing
thepercapitaincomeof particular
countries
by theaverage
percapita
forallcountries.
But how farcan the eclecticparadigm
predictwhichcountrieswill becomenet
outwardinvestors,and/or the pointon the investment
development
cycle that
this will occur?And how far can it explainthe reductionin the net outward
positionof somehigh(andrising)incomecountries
(e.g.,theU.S.)in recentyears?
The answerslie in the changinginternational
distribution
of factorendowments,
thosethataretransferable
especially
acrossnationalfrontiers;
andin the changing
andmarkets
efficacyof hierarchies
as transnational
modes.As a generalhypothesis,
the less evenlyassets(whichhelp producemobileintermediate
products)are
distributed
acrossnationalboundaries,
andthe greaterthe transactional
failureof
marketsin theseproducts,the widerwill be the dispersion
of the net outward
investment
positionof countries34
(arounda zeronetoutwardinvestment
position).
The moreevenlyresources
are distributed,
and the less the transactional
market
inthenetoutward
thenarrower
thedispersion
failure,
investment
positionofcountries
is likelyto be.
Thegradualconvergence
of percapitaincomelevelsandthe economicstructure
of the advancedindustrialized
economies,togetherwith someharmonization
of
government
policies,is makingformoresymmetrical
trans-border
directinvestment
Thisphenomena
is associated
patterns.
withanother,namelythe growthof intraindustryproduction.
Like intra-industry
reflectsless the
trade,such production
and morethe advantages
of scaleeconomies
dispositionof factorendowments
in production
in consumer
andmarketing,
anddifferences
tastesbetweencountries;
theremaycontinueto
althoughwithinsomesectors(e.g.,consumerelectronics),
be someinternational
divisionof laborbasedon thedistribution
of country-specific
endowments.35
Likeinter-industry
it sometimesreplacestrade(e.g.,
production,
whereit is prompted
andsometimes
orchanges
byimportrestrictions);
complements
the patternof trade(e.g.,wherethereis specialization
of productsor production
in different
processes
locations).
Againin thislattercase,thecompetitive
advantages
of the participating
firmsarelessthoseof the rent-seeking
kind,andmorethose
thatarisefromthe commonoversight
of complementary
assets.Thefact,too,that
intra-industry
is largelyin thehandsof largeanddiversified
production
multinational
addsfurther
to thislikelihood.
oligopolists
THE ECLECTICPARADIGM
17
The case of Japanis a particularly
interesting
applicationof the investment
development
path,in spiteof (or perhapsbecauseof) the factthatgovernment
curtailed
the roleof inward
intervention
by the Japanesegovernment
deliberately
formostof thesecondstageof thecycle.In termsof theOLIparadigm,
investment
disallowed
theinternalization
of mostintermediate
Japaninitially
products
(especially
marketsby fore;ign
technology)
MNEs;instead,it acquiredtheseproductsin other
Thisprocesscontinueduntilthe
ways,or promotedtheirindigenous
production.
and
Japaneseeconomyhadevolveda strongindigenous
technological
capability,
in
markets.
At
the sametime,
its firms,distinctive
ownershipadvantages world
of thisprocess,Japan'slocationalattractions
and partlyas a consequence
began
to change.Risingrealwagesreducedthe competitive
edgeof its unitlaborcosts;
and material-saving
activitiesbecamecomparatively
more
technology-intensive
The net resultof thesechangeswas thatJapanbothneededto export
attractive.
tohelpto relocate
thekindof production
andintermediate
mobileresources
products
in whichhercomparative
immobileresources
wasdeclining,
thatrequired
advantage
while importingmobileresourcesand intermediate
productsthatcouldbe used
in whichhercomparative
was rising.Butfor
withimmobileresources
advantage
it alsohadto acceptthepresenceof foreign
its ownfirmsto becomemultinational,
itrequired
werenotoftenforthcoming
MNEs.Moreover,
thetypeof mobileresources
in
over
the
via the non-equity
route; consequence,
yearsJapanesepolicytowards
inwarddirectinvestment
hasbeenliberalized.
In the mid-1980s,Japanenteredthefifthstageof development
cycle.36
Thispoint
is reachedwheretwo thingshappen.Firstthe ownership-specific
of
advantages
a country's
MNEsbecomemorefirm-specific
(i.e.,of a transaction
cost-minimizing
andsecond,thelocational
kind)andlesscounty-specific
(i.e.,asset-based);
decisions
anddomestic
MNEsbecomelessbasedonthecomparative
bybothforeign
advantage
of factorendowments,
andmoreon thestrategies
of competitors
supplying
regional
orglobalmarkets,
thedesireto fullyexploittheeconomies
of large-scale
production,
the needto reducemarketinstabilities
anduncertainty,
andthe incentiveto reap
the gainsfromintegrating
relatedactivitiesoverspace.To thesefeaturesanother
maybe added,whichrestslesson the development
stageof a countryandmore
on its economicpositionvis-a-visthatof othercountries.Here,the proposition
is thatas countriesconvergein theirincomelevelsand economicstructures,
the
moresymmetrical
cross-investment
flowsarelikelytobe.Therelatively
fastergrowth
of European
intheU.S.thanof U.S.directinvestment
andJapanese
directinvestment
in Europeand Japanin the 1970s and the early 1980s lendssupportto this
proposition;
while the dramaticimprovement
of U.S. economicperformance
in
the mid-1980s,the accelerated
of
path technological
advance,andthe realitiesof
is causinga resurgence
globalcompetition
of foreignactivityby U.S.MNEs.
ThefactthatJapaneseparticipation
in European
andU.S.manufacturing
industry
is currently
growingveryquickly,and especiallyin thosesectors(e.g.,electronics
and motorvehicles)thattend to be dominatedby MNEs;and that European
and U.S. firmsin similarindustries
are formingallianceswith the Japanesein
theirhomemarkets(Ohmae1985)suggests
thatthestructure
of Japaneseoutward
directinvestment
is now increasingly
resembling
thatof the U.S. and European
18
BUSINESSSTUDIES,SPRING1988
JOURNALOF INTERNATIONAL
economy
thatastheJapanese
predicted
And,indeed,it maybe reasonably
countries.
economy,yet more
becomesan increasingly
high-wageand technology-intensive
of JapaneseMNEswill changein two
oriented,thatthe character
internationally
of theirinvestments
willbe directed
to thedeveloped
ways.Firsta higherproportion
in
similar
sectors
to
those
invested
by
foreign
companies
worldand be within
will cometo dependmoreon their
strengths
in Japan.Second,theircompetitive
activitiesthanon
a globalnetworkof interrelated
abilityto operatesuccessfully
assets.At the sametime,whilethereis some
of particular
the favoredpossession
MNEs are currently
suggestion(Dunning1986a)thatJapanesemanufacturing
theirhigh-value
activitiesin theirhomeplants,a needto tap and
concentrating
and
advancesin such sectorsas biotechnology
monitorthe latesttechnological
thesesameMNEsto set up (or sharewithlocalfirms)
is encouraging
telematics
in EuropeandtheU.S.(Ohmae1985).
anddesignfacilities
research,
development
ExplainingDifferentFormsof InternationalEconomicInvolvement
might
in whichtheeclecticparadigm
production
of international
Thethirddirection
business
transactions,
be extendedis forit to incorporate
otherformsof international
agreements.
contractual
trade,joint venturesand non-equity
notablyarms-length
the latter
To date,however,while some progresshas been madein embracing
two subjectareas37
-indeed, some authors(e.g.,Casson1986d)wouldgo so
be it partof an equityor non-equity
farto arguethatthecontractual
relationship,
is thekeyto ourunderstanding
of international
business
formof businesassociation,
involvement
only limitedheadwayhas been madein unifyingexplanations
We suggestthereare two mainreasonsfor this.The
of tradeand production.38
derivesits analytical
production
firstis thatthe moderntheoryof international
ratherthan
organization
fromthe theoriesof the firmand industrial
framework
atleast,alltradeable
activities
trade;and,implicitly,
fromthetheoryof international
betweenindependent
Secondly,
areassumedto be conducted
buyersandsellers.39
whilethe lattertheorytakesas its unitof interestthe nationstate,the focusof
of students
of theMNEis thefirmor groupsof firms.
interest
thosedesigned
It is truethatseveralof themoderntheoriesof trade- particularly
- explicitly
the roleof market
transactions
acknowledge
to explainintra-industry
is strongly
buttheemphasis
of interest
directed
as determining
factors;
imperfections
ratherthantransactional
to structural
1981,1983).Much
(Krugman
imperfections
trademaybe explainedby the
and monopolistic
of neo-technology
competitive
but
of resources
thatgaveriseto ownership-specific
advantages,
spatialdistribution
in
located the
with immobileresources
whichare usedby firmsin conjunction
assetsor their
homecountry.Whena firmgoesabroad,it exportstheseintangible
with foreignresources
thatit can obtainat
rights,and usesthemin conjunction
marketfailure
theimplications
of transactional
lowercostthanat home.However,
of
and
assets
across
national
make
for
the
common
boundaries,
that
ownership
of firms,havebeen
which,in turn,mayimpingeon the tradingcompetitiveness
toproduction
Itisherethatanintegrated
approach
bytradetheorists.
largelyneglected
andtradeoffersparticular
promise(DunningandNorman1985).
THE ECLECTICPARADIGM
19
thereare manyparallelsin the way in whichthe patternsof trade
Historically,
andinternational
haveevolved.Tostartwith,mosttradewasintersectoral
production
and largelyexplainable
by the international
distribution
of factorendowments.
Likewise,as we havealreadysuggested,
althoughan elementof marketfailure
is necessary
to explainthe ownership
of international
its structure
production,
and
locationinitiallyfollowthedictates
of theH-0-S paradigm
asappliedto intermediate
products.
Lateras tradebecamemoreintrasectoral,
new explanations
weresought
andfound(GrubelandLloyd1975,Tharakan
1984).Intrasectoral
production
also
of intra-industry
possessesmanyof theattributes
market
trade,butwithadditional
imperfections
(notablythosewhichareuniqueto thecommonownership
of assets
in differentcountriesand the internalization
of tradein intermediate
products).
unlikeintra-industry
Moreover,
trade,intra-industry
production
impliesintra-firm
tradeas well. Indeed,the abilityof a firmto tradeinternally
may itselfafford
thatfirmcertainadvantages
overits competitors
of gainsfrom
(e.g.,thepossibility
It is, then,not surprising
specialized
sourcingor transfer
pricemanipulation).
that
a firmbecomes,
toengageininternalized
themoremultinational
themoreitisinclined
trade(DunningandPearce,1985).
The economictheoryof intra-firm
trade,andhow it differsfrominter-firm
trade
is now beginningto receivesome attentionin the literature;
but again,mostly
in theorganization
of transactions
ratherthanthetransactions
by scholarsinterested
insofaras intra-firm
tradeis marketreplacing,
it has been
per se. Normatively,
and is commonlyperceived
viewedwith some suspicionby welfareeconomists,
as a meansthroughwhichMNEsmanipulate
transferpricesin a way inimical
to the intereststo one or other(or both)of the tradingcountries.Thoughthe
of tradeneed not affectits extent,pattern,or terms,the countryorganization
in theperceived
specificdifferences
gainswhichtransfer
pricingmanipulation
may
in countries
in whichthegainsarethought
MNEsto locateproduction
encourage
to be mostlikely.On theotherhand,MNEsmayengagein intra-firm
international
firmsmayengageinintra-firm
tradeforexactlythesamereasonasdomestic
national
the externaleconomiesof individualtransactions.
trade,namely,to internalize
Whetherthisbenefitsor adversely
affectsthedistribution
of international
economic
activityand/orthe welfareof the participating
countries,
dependson the nature
of the marketfailurebeinginternalized,
the consequences
of suchinternalization,
and how the gainsor lossesresultingfromit are distributed.
But thereis no a
priorireasonto supposethatintra-firm
tradein finalgoodsproducts
is lessbeneficial
to international
resourceallocationthan eitherinter-firmtrade conductedin
markets
or intra-firm
tradein intermediate
imperfectly
competitive
products.
The factthatan optimumsolution(in the Paretosense)is so difficultto identify
is that,becausemarketfailureexists,one is forcedto comparea numberof second
bestalternatives.
Oncematterssuchas the distribution
of benefitsovertime,risk
andgovernment
intervention
entertheequation,
one'scriteria
forjudgingoptimality
becomesmultifaceted.
Whilein principle,
inevitably
transactions
betweendifferent
thegainsor lossesresulting
partsof a domesticfirmposeidenticalproblems,
from
the transactions
areat leastcontainedin thatcountry.In the caseof tradewithin
hierarchies
acrossnationalboundaries,
inter-country
distributional
questions
cannot
20
BUSINESSSTUDIES,SPRING1988
JOURNALOF INTERNATIONAL
be ignored;
mayjudgeinternalallocativeefficiencyin terms
just as governments
theimpactof MNEs
of theirowneconomicandsocialgoals,so theywillevaluate
on theextentandpattern
of tradeflows.
maybe madeto one of the fastest
thissection,briefreference
Beforeconcluding
non-equity
namely,the cross-border
arrangements,
growingformsof institutional
ventureat a productor projectlevelbetweenMNEs,andparticularly
collaborative
in the OECDarea.Thesehavearisenfor a varietyof economic
thoseoperating
on the
to a recentsymposium
and strategicreasons,but as severalcontributors
the exploitationof
subject(Contractorand Lorange1987) demonstrated,
is not
or verticalintegration,
advantages
by horizontal
ownership
complementary
with the premisesof the eclecticparadigm,
but pointsto a need
only consistent
for broadeningits scope to embracequite specific,and perhapstemporary,
as part and parcelof their wider
international
alliancesbetweenenterprises,
international
strategies.
TheLocusof Decision-taking
outsidethe domainof the economistbut
One subjectarea,normallyconsidered
in
framework,
one whichwe believetheeclecticparadigm
offersa usefulconceptual
locusof decision-taking
withinthe MNE.Letus focus
concernsthe geographical
on the question,Why and underwhatconditionsare decisionson the way in
or influenced
by foreignaffiliates
of MNEscontrolled
whichresources
areallocated
Thisquestionmaybe brokendown
locatedin theparentcompany?
by managers40
The formeris mainlyan
and"who"aspectsof decision-taking.
intothe "where"
issueof locationaleconomics,and concernsthe priceand efficiencyof decisionin different
andthe costsof transcountries,
support)resources
taking(including
and
(e.g., especiallyadministration
borderinter-and intra-firmtransactions
areinvolved.Herea factorendowment
communication
costs),in whichmanagers
toolof analysis.
modelmaybe theappropriate
thatparalleltwo questions
The"who"aspectmaybe dividedintotwocomponents
in a foreign
in
does
a
MNE
undertake
earlier
this
production
asked
paper:Why
firmin thatcountry?,
and,Whydoesa MNE
countryratherthanan indigenous
transfer
ofintermediate
forthecross-border
themarket
products?
choosetointernalize
restsin the capacityto
of centralized
decision-taking
The competitive
advantage
take (whatare perceivedto be) the rightdecisionsfor the MNE as a whole.
to
If, for one reasonor another,such capacitycannotbe efficientlytransferred
or by exporting
foreignaffiliates-either throughthe trainingof localmanagers
willbe locatedin thehomecountry.
resources
expatriates thenthedecision-taking
natureof information
ornon-codifiable
themorethespecificity,
idiosyncrasy,
Clearly,
and relatedmanagerial
assets,the moredifficultit is to ensurean efficientuse
of scaleeconomies
of themin a foreignaffiliate.And the greaterthe advantages
thataccrueto the MNE in toto then
or benefitsof centralized
decision-taking,
is to be delegated.On the otherhand,the more
the less likelydecision-taking
to localneeds,ordependon indigenous
to be customized
support
decisionsrequired
facilitiesandexpertisefortheirefficientexecution(e.g.,withrespectto personnel
andpublicrelations),
themorelikely
distribution
industrial
relations,
management,
theyareto be decentralized.
PARADIGM
THEECLECTIC
21
the transfer
Butevenassumingthereis capacityin the hostcountryto assimilate
between
unlessthereis a mutualityof interests
advantages,
of ownership-specific
may
decision-taking
and theiragentsin the affiliates,
managersat headquarters
arise.The firstis
When mightthis occur?Two possibilities
still be centralized.
therearecostsandbenefits
as a consequence
of theactionsof an affiliate,
wherever,
is part;and the second
of
which
it
to
the
rest
of
the
organization
thataccrue
of objectivesand/or risksby local and
is wherethereis a differentperception
Take a simpleexample.Supposea MNE operatestwo
centralmanagement.
plants,one in the U.S.and the otherin France.Assumethe aim
manufacturing
decisionlocalprofits,andthateachactsas an independent
of eachis to maximize
favorthesitingof managerial
advantages
takingunit.Thenas longas thelocational
resourcesin France,and theseresourcesare efficientlyused,decisionswill be
decentralized.
Now supposethe parentplantadoptsa new strategyaimedat
is comparable
groupprofits.To do so it mayneed(andthe situation
maximizing
decisionsso as to bothrationalize
to centralize
by onefirmof another)
to a takeover
resourceallocationand captureany benefitsexternalto eitherof the production
asa whole.Inthisway,by centralizing
decisiontotheorganization
unitsbutinternal
an hierarchical
processverysimilarto thatof
taking,the MNE is undergoing
of intermediate
theinternalization
productmarkets.
then,it maybe possibleto constructan economic
Usingthe OLI framework,
(andas a variantof this,withina location,
theoryof the locusof decision-taking
is thatdecision-taking
will
Thehypothesis
of the decisiontaker).4'
the nationality
and non-codifiable
be morecentralized:
specificity
(i) the greaterthe uniqueness,
fromthe homecountry;(ii) the
emanating
advantages
natureof decision-taking
betweenthe parentcompanyand
greaterthe likelihoodof a conflictof interests
and (b) externalities;
withrespectespeciallyto (a) riskperception
the subsidiary,
favorsthehome
resources
and(iii)the morethelocationalcostsof decision-taking
in transferring
attitudes
management
country,whichmightreflectbothdifficulties
services.
andmanagement-related
andthepriceof management
andpractices,
Theseelementswill clearlyvaryaccordingto countryindustryandfirm-specific
Thetendency
tocentralize
tothedecision-taking
functions.
andalsoaccording
factors,
and accounting
and development,
decisionsrelatingto research
capitalbudgeting
those relatingto personnelmattersand sourcing
methods,but to decentralize
in theseterms.Some illustrations
of the use
be
explained
might
arrangements,
of thisapproachis containedin Dunning(1986b),whichexaminesthe locusof
in
as betweenU.S.parentcompaniesand theirU.K.subsidiaries
decision-taking
in the 1980S.42
the 1950sandtheirJapanese
counterparts
Divestment by MNEs
A fifthpossibleareafor furtherstudyrelatesto our understanding
of divestment
or a reductionof foreignproduction
by MNEs.Someprogressalongtheselines
hasbeenmadeby Boddewyn(1983)andCasson(1986d),butby andlarge,the
has so fartreateddivestment
as a discreteact of assetdisposal(i.e.,the'
literature
of
ratherthanpartand parcelof a continualreappraisal
reverseof acquisition)
of the assetsa firmwishesto hold.The processof
the amountand disposition
is differentfroman initial
of foreignproduction,
a reduction,or disintegration
22
BUSINESSSTUDIES,SPRING1988
JOURNALOF INTERNATIONAL
act of entryin two ways.First,it requiresthe absenceof only one of the three
barriers
toexitthatdonotcorrespond
andsecond,theremaybecertain
OLIvariables;
to entry.43
to barriers
set out earlierin thispaper,
Usingthe "Mark2" versionof the eclecticparadigm
we mightpredictthatMNEswouldwishto reducetheirpresencein a particular
First,wherea changein thedistribution
countryor sectorundertwocircumstances.
(or the efficiencywith whichtheseare used)(i) weakens
of factorendowments
relativeto thoseof firmsin host countries,or (ii)
advantages,
theircompetitive
fromthe hostto home(or indeed,otherhost
causesthemto switchproduction
benefits
(costs)orusingtheexternal
Second,wherethenettransactional
countries)."4
increase(fall)relative
of
advantages
for
these
competitive
exploitation
markets the
hierarchies.45
to thoseofferedadministered
of changes
approachto an understanding
As a startingpointfor an integrated
abroad,
let us assumethatonce a firmis established
in international
production,
relatedto the size and pattern
decisionsare organically
investment
its sequential
andto its viewsabout(a) its existingandlikelyfuture
of its existinginvestments,
and
abouttechnological
andweaknesses,
(b) its expectations
strengths
competitive
reactionsto (a)
of its competitors'
marketopportunities,
and (c) its perception
thisis likelyto resultin a continualreassessment,
and(b). In a dynamicsituation,
positionchanges;
As a firm'scompetitive
of activities.
andreorganization
relocation
strategies
evolve;
as new core skillsreplaceexistingones;as new management
between
as new marketsopenup andothersdie;andas thebalanceof advantage
of its
usinginternaland externalmarketsshifts,so will the level and structure
Althoughin somecases,this mayleadto a divestment
production.
international
of the entireforeignassetsof a firm,moreoften,it will resultin a restructuring
or sectorshelpingto finance
witha saleof assetsin somecountries
of its portfolio,
withinMNEsoften
and disintegration
an increaseof assetsin others.Integration
go handin handwitheachother,justas do thebirthanddeathof firms.46
investors
in the
of leadinginternational
of the OLI advantages
The realignment
of the relevanceof the factor
last two decadesprovidesampleconfirmation
of Japan
of the eclecticparadigm.
Theemergence
endowment
modelcomponent
investorhas resultedin a fall of the shareof U.S.
international
as a significant
andconsumer
MNEsinseveralindustrial
notablyautomobiles
andEuropean
sectors,
in some developing
countries
electronics.
The growthof offshoremanufacturing
in the 1970sto takeadvantage
division
changinginternational
of an apparently
domesticsectors(both
the declineof labor-intensive
of laborhelpedaccelerate
in developed
recent
countries.
anduninational
However,
of multinational
companies)
on skilledlabor,haveat thesame
whileplacinga premium
advances,
technological
This
of laborcostsin manymanufacturing
processes.
timereducedthesignificance
of
the
in
of
of
activities.47
Within
some
a
return
home
some
these
has resulted
more rapidlydevelopingidustrializing
developingcountries(e.g., SouthKorea,
in low value-added
and
one has also seen divestment
Taiwanand Singapore),
in highvalue-added
activities.In some primarysectors,mainly
new investment
a markeddeclinein inward
onehaswitnessed
of hostgovernments,
attheinsistence
investment;
in others,and particularly
in someserviceindustries,
domesticand/
THE ECLECTICPARADIGM
23
verticalintegration
has sharplyincreased.
or international
Justas the volumeand
patternof tradeof a firmor countryis affectedby changesin the distribution
so willthatof international
of factorendowments,
production.
Thequestionof the changingownership
of assetsin particular
is perhaps
countries
moreinteresting.
WhyshouldMNEssellassetsandacquireothers?Clearly,when
therearechangesin the relativetransaction
costsof individual
capitalandforeign
exchangemarkets(notleastdueto uncertainties
aboutinterest
andexchangerates),
MNEswill reappraise
theirinternational
financialportfolios.
Butwhataboutlongtermrealforces?Theanswermustbe that,whereincentives
forthedirectinvestment
are reducedor advantages
of commongovernance
so divestment
disappear,
will
occur,providingthatthe exit costs(whichthemselvesinvolvetransaction
costs)
do not outweighthe savingsof usingthe market.The transaction
costsof exit
requirefurther
study,butanythingthatreducestheriskanduncertainty
of external
markets,lessensthe importance
of scaleeconomiesin production,
or reducesthe
willmakefordivestment.48
of particular
transactions
externalities
Thisshiftingof the balanceof advantages
of hierarchies
andexternalmarketsfor
international
transactions,
togetherwith the emergenceof new contractual
whichpossesssomeof thecharacteristics
of each(Oman1984,Casson
arrangements
of thefunctions
andboundaries
of MNEs.
1986c),hasledto frequent
realignments
Whileverticalintegration
has noticeablyincreasedin somesectors(e.g.,vehicles
andelectronics)
andfallenin others(e.g.,hotelsandshipping),49
thegeneraltrend
builtarounda groupof core
has beentowardsnew collaborative
arrangements
This has elsewherebeen termedquasi-integration
and
technologies.
(Contractor
andorganizational
advanceshave
Lorange1987).At thesametime,technological
in serviceswithinvestment
in goods;theinformation
increasingly
linkedinvestment
industryis a classicexample.In secondary
new alliancesbetweenfirms
industry,
along the value-addedchain have been fostered,to exploit complementary
andto linkcomputer-aided
withthatof later
technologies
designof components
manufacturing
processes(Hayesand Wheelright1984).On the otherhand,the
sizeableamountsof capitalinvolvedand the dangerof an integrated
firmbeing
lockedintooneparticular
theriskof internalization.50
sourceof supplyincreases
the raisond'etreof horizontal
Similarly,
integration
maychangeas the importance
of synergisticownershipadvantagesshifts with advancesin technologyand
information.
The extentof cross-hauling
of hierarchical
or quasi-hierarchical
in the technologically
advancedsectorsbetweenthe U.S.andJapan
arrangements
is testimonyof this.Indeedthe growthof globalindustries,
characterized
by a
amountof intra-industry
substantial
andintra-firm
andan interlocking
production
is a feature
networkof cross-border
alliances
thatindustrial
organization
economists
areonlyjust startingto get to gripswith.Butheretoo, an organic
andstrategists
to bothdivestment
andinvestment
approach
by MNEsis required.
The Consequencesof MNE Activity
One finalareain whichthe eclecticparadigmcan be a usefulframework
for
the impactof MNEactivityon homeand hostcountry
analysisis in examining
economicgoals.Letus illustrate
fromtheviewpointof a hostcountry.
24
BUSINESSSTUDIES,SPRING1988
JOURNALOF INTERNATIONAL
The argumentruns somethinglike this. Inwarddirect investmentis welcomed
for the resourcesand/or accessto marketsit bringsto the recipientcountry,and
the way it maypromotethe upgradingandbetterdeploymentof existingindigenous
endowments.The concernover this particularvehicle of importingintermediate
productsis two-fold.First,thatbecauseof its strongbargainingpower,the investing
firm is able to capturean undesirablyhigh share of the value added or created
by its subsidiaryin the host country;and second, that decisionstaken about the
and about the use of these resources,
amount and kind of resourcestransferred,
may yield less benefitto the recipientcountrythan that which might arise from
some otherpatternof resourceallocation.
mechanismby which
In the one case, the MNE is regardedas an organizational
or costlyto producein thehostcountry,
products,whichareunavailable
intermediate
areefficientlyacquiredandused;in theother,as a vehicleof economicrationalization
and economic power,which it uses to promoteits global goals in a way which
distortsor inhibitsthe desireddispositionof resourcesby the host country.
The debateover the impactof inwarddirectinvestmenton host countriesis now
entering a new phase as markets and production become increasingly
internationalized.
Moreover,notonlyareMNEstakinga globalview of theirstrategies
and view the locationalattributesof countriesfrom this perspective- but
andcompetitive
strategies
countriesalsoarebeginningto recognizethattheirindustrial
dimension.Since the industrieswhich most
posturesmusttake on an international
desirableare largelydominatedby MNEs,it follows
countriesview as strategically
thatconflictbetweenmultinationaloligopoliespursuingglobal economicstrategies
and countriespursuingdomesticpoliticalstrategies- and both withina changing
and increasinglycompetitiveinternationalenvironment- is inevitable,and, in
In the 1950s and '60s, the interestcenteredlargelyon whether
part,irreconcilable.
the types of resourcesprovidedby MNEs were appropriateto needs of recipient
countries;and if, comparedwith alternativeroutesof acquiringthese resources,
their benefitsexceededtheir costs. The debatein the 1980s has much more to
do with the way in which MNEs use theirworldwideassetsto achievetheirlongterm economicgoals;and whetherthe resultingallocationof activityis consistent
with that which the countriesin which they operateare seekingto achieve.As
a frameworkfor analyzingthese questions,we believe that the eclecticparadigm
has a greatdeal to offer.
CONCLUSIONS
This paperhas soughtto demonstratethat,a decadeafterits inception,the eclectic
paradigmremainsa usefulandrobustgeneralframeworkforexplainingandanalyzing
productionbut manyorganizational
not only the economicrationaleof international
and impactissuesrelatingto MNE activityas well. Conceptually,thereare close
parallelsbetween the main tenets of the paradigmand that of modern theory
of businessstrategy;51
thoughneitherapproachis sufficientto explaintheinternational
profileof any particularMNE. It is likely, however,that new theorizingin the
next decadewill take a differentform to that of the last ten years,if for no other
reason than the characterand organizationof internationalproductionis itself
undergoingfundamentalchange. More especially,we foreseea more systematic
25
THE ECLECTICPARADIGM
marketfailureinto more general
effortby tradeeconomiststo model transactional
theories of internationaleconomic involvement,while industrialand business
economistsare likely to become more interestedin the dynamicsof the OLI
configuration,and its impact on the strategyof individualfirms, throughsuch
techniquesas game theoreticanalysisand networkmodels(Johansonand Mattson
1987). We expect more attentionto be paid to the determinantsand effectsof
collaborativeventuresnow being formedbetweenMNEs from advancedcountries
andLorange1987),andbetweenMNEsandtheircustomersandsuppliers.
(Contrator
We foreseea renewedinterestin identifyingand evaluatingthe ownership-specific
the ability of
advantagesof firms, with particularfocus on entrepreneurship,
managementto identifyand coordinatea rangeof core skills and assetsthrough
routes,and to promoteoperationalflexibilityin a volatile
a varietyof organizational
world environment;global marketingnetworks;the creationand use made of
informationand communicationstechnology;and a varietyof
computer-related
issues. Finally, we would perceive a gradual
cross-culturalmanagement-related
of the approachesof the economist,businessanalystand organizational
interweaving
theoristto our understandingof internationalproduction,althoughwithin these
and relateddisciplines,theorizingand empiricalwork will become both more
and morepolicyoriented.
technicallysophisticated
NOTES
1. Definedas productionfinancedby foreigndirectinvestmentand undertakenby multinationalenterprises.
2. The proceedsof which werepublishedunderthe editorshipof Ohlin et al. (1977).
3. Butsee our remarksin the secondsectionof this paper.
4. Fora recentreviewof the literatureon intra-industry
trade,see Tharakan(1984).
5. Foran elaborationof this thesissee Dunningand Norman(1985).
6. A full account of the eclecticparadigmis given in Dunning(1981). See also Dunning (1983a and 1986d),
Dunning and McQueen (1981), Dunning and Norman(1985), and Norman,Flanaganand Seymour (1985)
for some extensionsand applicationsof the concept.
7. Dunning(1981) page 27.
8. Teece(1983) usesa ratherdifferentterminology,namelyproductionand transactionadvantages.
9. For example,as identifiedby Bain (1956) as monopolypower,productdifferentiation,
absolutecost barriers
and governmentintervention.
10. See, for example,an interestingpaperby Buckley(1986) presentedto the Londonmeetingof the Academy
of International
Businessin November1986.
11. See especiallyCasson (1979), Buckley and Casson (1985), Teece (1981) and (1985), Hennart(1986),
and Rugman(1986). The word"failure"is an unfortunate
one as it impliesthatthereis an alternativetransactional
mechanismwhich is superiorto the market.This is not necessarilythe case.
12. As, forexample,occursin the case of "naturalmonopolies"and in industrieswith highsunkor developmental
costs.
13. For example, in the U.K., Japanese affiliatesdominatethe color TV sector partly because their parent
companiespreferthis route of entry ratherthan licensingtheir competitiveadvantagesto U.K. producers;but
partlytoo, becausethey are more successfulat internalizingintermediateproducttransactions,than,for example
are theirU.S. counterparts.
14. In thatwhat mightbe (or was) exogenousthe firmis (or now becomes)endogenous.
15. We would also accept with Hennart(1982, 1986) that a privilegedaccess to technologyor capital may
enable a firm to internalizecross-borderintermediateproduct markets.We would, however,suggest that the
privilegedaccessper se arisesbecause of an operationalratherthan an organizationalfailureof that market,
in the sense that it discriminatesin favorof particulargroupsof transactors,ratherthan it failsas a transactional
mode, c.f. some othermode, e.g., a firm.
26
BUSINESSSTUDIES,SPRING1988
JOURNALOF INTERNATIONAL
16. For a usefuldistinctionbetweentransferable
and nontransferable
intermediate
products,see Lall (1980).
17. It is possiblefor governmentinterventionto be directedto counteractingor alleviatingthe affectsof market
failureas well as distortingthe patternof resourceallocationto meet its social and strategicgoals. See especially
page 113, Table5.2 in Dunning(1981).
18. The concept of 'induced'or 'managed'comparativeadvantageis gaining strengthin the literature.(See
for example Scott and Lodge 1985, Teece 1987.) It suggeststhat, at a given moment of time, a country's
productiveassets consist not only of the naturalresourcesthen available(defined in the H-O-S sense) but
the accumulatedmanmadeassets of the past; the way in which these are organized;and the attitudesof its
people towardswealthcreationand economic security,and towardsthe rest of the world.The literaturefurther
suggeststhat governmentmay and does play a major role in fashioningthe strategyfor resourceallocation
and income distribution;the transactioncosts of markets;and of the ideological,culturaland work ethic of
its population.
19. Sometimesreferredto as the environment,systemsand policy (ESP) paradigm.See Koopmanand Montias
(1971) and Dunning(1981). See also a paperby Boddewyn(1986).
20. See, for example,Robockand Simmonds(1983) and Rugman,Lecrawand Booth(1985).
21. See particularlyPorter(1985) and (1986), Hamel and Prahalad(1987) Lahdepaaand Ansoff (1987).
The differencebetween the firm-specificvariablesidentifiedby these scholarsand those of their predecessors
is that the latter concentratedon the structuralcharacteristicsof firms, e.g., size, age, product composition,
such as those identifiedby
etc., while the formerfocused on operational,i.e., strategic-related
characteristics,
above.
of academicresearchers
22. It is one of the moreirritatingcharacteristics
(and the presentauthoris no exception!)
for themto inventtheirown nomenclatureforconceptsthatarefamiliarunderdifferentnamesto otherresearchers.
Thus Porter'sconcept of configurationis for all intentsand purposesthe same as our location advantages;
thatrelateto the commongovernance
while his coordinationdimensionis similarto thoseaspectsof internalization
of value-addingactivities.Finally Porter'scompetitiveadvantagesmake up a major part of our concept of
ownershipadvantages.The only difference,as far as I can see, is that our concept of ownershipadvantages
may include attributesof firms that more accuratelyreflect their monopolisticpower than their competitive
prowess.
23. Some of these were identifiedby the author when Aliber'sthesis (Aliber (1970) was first put forward
(Dunning(1971)). See also morerecentcriticismsby Gray(1982) and Teece(1986).
24. For a more extendedanalysisof the differencesand similaritiesbetweenthe Kojimaand the eclectic and
internalization
paradigm,see Buckley(1983) and (1985) and Gray(1985).
25. Unlessthe assetis sold outrightin the firstplace.
26. This suggeststhat economistsinterestedin explainingthe internationalallocationof activityboth by firms
and countriesneeded to be versed in modern trade and location theory, industrialorganizationtheory and
the theory of the firm, and on the way in which each interactswith the other. Both Casson (1985a), and
Normanand Dunning(1984) take up and illustratethis point.See also page 27-31 of thispaper.
27. See Kogut(1983). At the same time, the extentand patternof marketfailuremay itselfbe country-specific.
Compare,for example, the organizationalstructuresand sub-contractingrelationshipsof U.S. and Japanese
firms(Imai 1985;Sullivanand Nonaka1986).
28. A firstattemptto do this by use of clusteranalysiswas madein Chapter5 of Dunning(1981).
29. For example, while it may be possible to identifythe major variablesinfluencingall import substitution
FDI, theremaybe additionalfactorsspecificto (say) Swedishinvestmentin the Thai carindustry.
30. See, for example,some work on theselinesby Graham(1986) and Johansonand Mattson(1987).
31. For a full explanation,see Chapter5 of Dunning (1981 and 1986c). The term "cycle"was used in that
it was predicatedthat a countryboth startedas a zero net outwardand, at a later stage in its development,
returnedto that positionwhen its inwarddirectinvestmentstake was balancedby its (growing)outwarddirect
investmentstake. In retrospect,the term "path"might have been a better word to describethe process of
directinvestmentposition.
changein a country'sinternational
32. Hence the acquisitionor part acquisitionof Europeanand U.S. high technologyfirms (particularlythose
in difficultfinancialstraits)by, e.g., Indian,MiddleEastern,and Chinesefirms.
33. The role of governmentis one of the most idiosyncraticto evaluate.In the '70s and '80s, for example,
there have been quite dramaticswings in the policies of individualcountriesto inwardinvestment,according
to the governmentin power.Insofaras it is possible to generalize,the more right (left) wing a government
is in relationto its predecessor,the more (less) liberalits attitudeis likely to be towardsinwardand outward
foreigndirectinvestmentand indeedto privateenterpriseas a whole.
27
PARADIGM
THEECLECTIC
34. Most countriesare in factnegativenet outwardinvestors,i.e., net inwardinvestors.
35. For example,MNEs may continueto concentratetheirhigh-value(e.g., researchand development)activities
in countriesthat have a comparativeadvantagein the supply of highly trainedmanpower,while locatingthe
low-value-addedactivitiesin countriesthat have a comparativeadvantagein the supply of low- or semiskilled
labor.
36. Not consideredin the originalversionof the theory.Foran elaborationsee Dunning(1986c).
37. Fora reviewof the literaturesee Oman (1984) and Buckleyand Casson(1985).
38. With some noticeableexceptions,e.g., the work of Hirsch(1976), Gray(1982), Ethier(1986) and Markusen
(1986).
39. And no distinctionappearsto be made betweenmulti-activityand single-activitytradingfirms.
40. Using managersas a generictermfor decisiontakers.
subsidiariesin the U.K. headedby a Japaneseexpatriate
41. For example,why are most Japanesemanufacturing
whereasmost U.S.subsidiarieshavea U.K. nationalas chief executive?
42. See also some interestingwork on the structureof decision-takingin U.K. subsidiariesin the 1970s by
Young,Hood and Hamill(1985).
43. Porter(1980) identifiessix of these, namely the presenceof specializedassets, fixed costs, strategicexit
barriers.
barriers,informationconstraints,managerialemotionsand pride,and government-related
42. The questionof a reductionin foreignproductionin toto and that in a particularcountryand/or sector
needsmorecarefuldistinctionthanit has been given up to now.
45. We use the term net benefitsand costs because there are often costs and benefits of using both routes
for transactinggoods and services.
46. It is interestingthat thereis abundantliteratureon the theoryof the growthof the firm (which, in practice,
often occurs as a result of the expansionof part of a firm'sactivities),but very little on the decline of the
firm. Yet, particularlyin times of rapid technologicalchange, growth and decline are handmaidensto each
other.
47. Ohmae (1985) suggeststhat while labor costs in many developingcountriesare only one-thirdof those
companiesrepresentless
in developedcountries,the directlabor costs in the majorcompetitivemanufacturing
than 10%of total costs.The savingson costs in producingin developingcountriesis often more than outweighed
by the transportcostsbetweendevelopingand developedcountries.Moreover,as realwagecostsrisein developing
countries,thesesavingsare likelyto fall.
48. It is worth emphasizingat this point that internalizationis not without its costs and multinationalitymay
bringdiseconomies,externalto particularaffiliatesand internalto the MNE.
49. See Dunningand McQueen(1981) and Casson(1986b). In both these lattercases,contractualagreements
of one kind or anotherhave enabled the contractorto gain many of the benefitsof integrationwithout the
costs.
50. For a summaryof some interestingwork on the integrationof informationsystemsin manufacturingsee
DeMeyer and Ferdows(1984). For a generalview of changingcorporatestrategiesin a time of technological
change,see Ergas(1985) and Dunning(1986c).
51. As, for example,set out by Porter(1980, 1985 and 1986) and Kogut(1985a). The expression"competitive
advantage,while the comparativeadvantageof countries
advantage"may be interpretedas ownership-specific
appearssynonymouswith our location advantage.While the businessstrategyapproachgives more emphasis
to the positioningof firms in the sectors in which they compete, the eclectic paradigmplaces more stress
formof transactionalrelationships.
on the organizational
REFERENCES
Aliber, R. 1970. A theory of foreign direct investment.In C.P. Kindleberger,The internationalcorporation.
Cambridge:MIT Press.
.- 1983. Money,multinationalsand sovereigns.In C.P. Kindlbergerand D.B. Audresch,The multinational
corporationin the 1980s. Cambridge:MIT Press.
Bain,J.S. 1956. Barriersto new competition.Cambridge,Mass.:HarvardUniversityPress.
Archiv,
Boddewyn,J. 1983. Foreigndivestmenttheory:Is it the reverseof FDI theory? Weltwirtschaftliches
119:345-55.
28
BUSINESSSTUDIES,SPRING 1988
JOURNALOF INTERNATIONAL
. 1986. Internationalpoliticalstrategy:A fourth'generic'strategy?Paperpresentedto the Academy of
BusinessAnnualMeeting,London,November1986.
International
businessapproachto directforeigninvestment:A comment
Buckley,P.J.1983.Macroeconomicversusinternational
HitosubashiJournalof Economics,24:95-100.
on professorKojima'sinterpretation.
.- 1985. The economic analysisof the multinationalenterprise:Readingv. Japan. HitosubashiJournal
of Economics,26:117-124.
.- 1986. The limits of explanation.Testsof the theoryof the multinationalenterprise.Paper presented
BusinessAnnualMeeting,London,November1986.
to Academyof International
Cantwell,J.A. 1986. Technologicalinnovationand internationalproductionin the industrialworld.A study
networks.Ph.D. thesis,Universityof Reading,1986.
of the accumulationof capitalin international
enterprise.London:Macmillan.
Casson,M.C. 1979. Alternativesto the multinational
__
__
__
__
___
__
. 1985. The theory of foreigndirectinvestment.In P.J. Buckley& M.C. Casson, The economictheory
of the multinationalenterprise.London:Macmillan.
. 1986a. Generaltheoriesof the multinationalenterprise:A criticalexamination.In P. Hertner& G.
Theoryand history.Aldershot& Brookfield,Vt.:Gower.
Jones,eds.,Multinationals:
. 1986b. The role of verticalintegrationin the shippingindustry.Journalof TransportEconomicsand
Policy,22 (January):7-29.
for technologytransfer:New evidencefrom businesshistory.
. 1986c. Alternativecontractualarrangements
Investmentand BusinessStudiesNo. 95 (May).
Universityof Reading,DiscussionPapersin International
The sale of Chrysler(UK) to Peugot.
. 1986d. Foreign divestmentand internationalrationalization:
In J. Coyne& N. Wright,eds.,Divestmentand strategicchange.Oxford:PhilipAllan.
. 1987. Thefirm and the market.Oxford:BasilBlackwell.
Coase,R.H. 1937. The natureof the firm.Economica,4 (November):386-405.
ContractorF. & P. Lorange.1987. Cooperativestrategiesin internationalbusiness.Lexington,Mass.:Lexington
Books.
Davidson, W.D. 1976. Patternsof factor saving innovationin the industrializedworld. EuropeanEconomic
Review,8:207-17.
technologytransactionandtheoryof thefirm.Journalof Industrial
& D.G. McFeteridge.1984.International
Economics,32:253-64.
__
Business
in the choice of internationaltransfermode. Journalof International
. 1985. Key characteristics
Studies,16 (Summer).
DeMeyer,A. & K. Ferdows.1984. Integrationof informationsystemsin manufacturing.INSEAD Research
WorkingPapersNo. 13, December.
enterprise.London:Allen & Unwin.
Dunning,J.H.,ed. 1971. Themultinational
enterprise.London:Allen & Unwin.
productionand the multinational
. 1981. International
businessin a changingworld environment.Banco Nazionaledel LavoroQuarterly
. 1982. International
Review,143:351-373.
1983a. Changesin the structureof internationalproduction:The last 100 years. In M.C. Casson,
___.
business.London:Allen & Unwin.
ed., Thegrowthof international
___
___.
__
__
. 1983b. Marketpower of the firm and internationaltransferof technology.InternationalJournal of
IndustrialOrganisation,1:333-351.
. 1986a.Japaneseparticipationin Britishindustry.London:CroomHelm.
. 1986b. Decision-makingstructuresin US and Japanese manufacturingaffiliates in the UK: Some
similaritiesand contrasts.Geneva:ILO WorkingPaperNo. 41.
In K.M.Khan,ed.,Multinationals
1986c.Theinvestmentdevelopmentcycleandthirdworldmultinationals.
of the south.London:FrancisPorter.
In N. Hood & J.E. Vahlne,eds., Strategies
businessand economic restructuring.
. 1987a. International
in globalcompetition.London:CroomHelm.
corporationintegrationand regionalintegration.Universityof ReadingDiscussion
. 1987b. Cross-border
Investmentand BusinessStudiesNo. 105, July.
Papersin International
creation,transfer
& J. Cantwell.1986. The changingrole of multinationalenterprisesin the international
Investmentand Business
and diffusionof technology.Universityof ReadingDiscussionPapersin International
StudiesNo. 101, December.
THE ECLECTICPARADIGM
_____
29
& M. McQueen. 1981. The eclectictheoryof internationalproduction:A case study of the international
hotel industry.ManagerialandDecisionEconomics,2 (December):197-210.
& R.D. Pearce.1985. The world'slargestindustrialenterprises1962-83. Aldershot:Graver.
& G. Norman. 1985. Intra-industry
productionas a form of internationaleconomic involvement.In
A. Erdilek,ed., Multinationals
as mutualinvaders.London:CroomHelm.
_____
& A. Rugman. 1985. The influenceof Hymer'sdissertationon theoriesof foreigndirect investment.
AmericanEconomicReview,May:228-32.
Ergas,H. 1984. Whydo some countriesinnovatemorethanothers?Brussels:Centerfor EuropeanPolicyStudies.
_
- 1985. Corporatestrategiesin transition.In A. Jacquemin,Industrialpolicy and internationaltrade.
London:CambridgeUniversityPress.
Ethier,W.J. 1986. The multinationalfirm.QuarterlyJournalof Economics,101:805-833.
New York:Harper.
Franko,L. 1976. TheEuropeanmultinationals.
investmentby multinationalfirms:A rivalisticphenomenon.Journalof Post
Graham,E.M. 1978. Transatlantic
KeynesianEconomics,I (Fall).
__
___.
direct investment,marketstructure,firm rivalryand technologicalperformance.
. 1985. Intra-industry
In A. Erdilek,ed., Multinationals
as mutualinvaders.London,CroomHelm.
1986. Internaleconomies,oligopolyreactionand dynamiccontestabilityin global industries:A first
cut at synthesis.Mimeo.
Gray,H.P. 1982. Towardsa unifiedtheory of internationaltrade, internationalproductionand direct foreign
investment.In J. Black& J.H. Dunning,eds.,Internationalcapitalmovements.London:MacMillan.
.- 1985. Multinationalcorporationsand global welfare:An extensionof Kojimaand Ozawa.Hitosubashi
Journalof Economics,26 (December):125-132.
Grosse, R.R. 1985. An imperfectcompetitiontheory of the MNE. Journalof InternationalBusiness Studies,
16 (Spring):57-80.
Grubel,H.G. & P.J. Lloyd. 1975. Intra-industrytrade, the theoryand measurementof internationaltrade in
differentiated
products.London:MacMillan.
New York:Praeger.
Guisinger,S. 1985. Investmentincentivesandperformancerequirements.
Hamel,G. & C.K. Prahalad.1987. Creatingglobalstrategiccapability.In N. Hood and J.E. Vahne,eds.,Strategies
in globalcompetition.London:CroomHelm.
Hayes,R.H. & S.C. Wheelwright.Restoringour competitiveedge:Competingthroughmanufacturing.
Chichester
& New York:John Wiley.
Helpman,E. 1984. A simple theoryof internationaltradewith multinationalcorporations.Journalof Political
Economy,92:451-47.
Hennart,J.F. 1982. A theoryof multinational
enterprise.Ann Arbor:Universityof MichiganPress.
___.
1986. Whatis internalization?
Weltwirtschaftliches
Archiv,122: 791-804.
Hertner,P. & G. Jones. 1986. Multinationals:
Theoryand history.Aldershot& Brookfield,Vt.:Gower.
Hirsch,S. 1976. An internationaltrade and investmenttheory of the firm. OxfordEconomicPapers,28:25870.
Horst,T.O. 1972. Firm and industrydeterminantsof the decision to invest abroad.Reviewof Economicsand
Statistics,54 (August).
Horstman,I. & J.R. Markusen.1986. Licensingv directinvestment:
A modelof internalization
by the multinational
enterprise.Universityof WesternOntario,mimeo.
Hymer,S. 1960. The internafionaloperationsof nationalfirms: A study of direct investment.Ph.D. thesis,
MIT (publishedby MIT Pressunderthe same title in 1976).
Imai, K. 1985. Network organizationand incrementalinnovationin Japan. Instituteof Business Research,
HitotubashiUniversity.DiscussionPaperNo. 122, July.
JohansonJ. & L.G. Mattsson.1987. Internationalization
in industrialsystems- A networkapproach.In N.
Hood & J.E. Vahne,ed., Strategiesin global competition.London:CroomHelm.
Knickerbocker,F.T. 1973. Oligopolisticreacfionand the multinationalenterprise.Cambridge,Mass.:Harvard
UniversityPress.
Kogut,B. 1983. Foreigndirectinvestmentas a sequentialprocess.In C.P. Kindleberger& D. Audretsch,eds.,
The multinationalcorporationin the 1980s. Cambridge:MIT Press.
30
BUSINESSSTUDIES,SPRING1988
JOURNALOF INTERNATIONAL
. 1985a. Designingglobal strategies:Corporateand competitivevalue added chain. Sloan Management
Review,25 (Summer):15-28.
__
. 1985b. Designingglobal strategies:Profitingfrom operationalflexibility.Sloan ManagementReview,
26 (Fall):27-38.
Kojima,K. 1978. Directforeign investment.London:CroomHelm.
.- 1982. Macroeconomicversusinternationalbusinessapproachto foreigndirectinvestment.Hitosubashi
Journalof Economics,23:1-19.
_____
& T. Ozawa.1984. Microand macro-economicmodelsof directforeigninvestment:Towardsa synthesis.
HitosubashiJournalof Economics,25 (2):1-20.
.- 1985. Towardsa theory of industrialrestructuring
and dynamic comparativeadvantage.Hitosubashi
Journalof Economics,26 (December):135-145.
Koopman,K & J.M. Montias.1971. On the descriptionand comparisonof economicsystems.In A. Eckstein,
Comparisonof economicsystems.California:Universityof CaliforniaPress.
Krugman,P.M. 1981. Intra-industryspecializationand the gains from trade. Journal of PoliticalEconomy,
89.
___.
1983. The new theoriesof internationaltrade and the multinationalenterprise.In C.P. Kindleberger
& D. Audretsch,eds., The multinational
corporationin the 1980s. Cambridge:MIT Press.
Lahdepaa,J. & K. Ansoff. 1987. A networkanalysisof internationaltransactions.In N. Hood & J.E. Vahne,
eds., Strategiesin globalcompetition.London:CroomHelm.
Lake, A.W. 1976a. Transnationalactivityand marketentry in the semi-conductorindustry.NBER Working
Paper126, March.
___.
1976b. Foreigncompetitionand UK pharmaceutical
industry.NBER WorkingPaper127, March.
Lall,S. 1980. Monopolisticadvantagesand foreigninvolvementby US manufacturing
industry.OxfordEconomic
Papers,32:102-122.
Markusen,J.R. 1984. Multinationals,
multi-planteconomiesand the gain from trade.Journalof International
Economics,16:205-216.
Mason, R.H. 1980. A comment on professorKojima'sJapanesetype versus Americantype of technology
transfer.HitosubashiJournalof Economics,20:242-52.
Nicholas, S. 1986. The theoryof multinationalenterpriseas a transactionalmode. In P. Hertner& G. Jones,
eds.,Multinationals:
Theoryand history.Aldershot:Gower.
in theinternational
construction
Norman,G., R. Flanagan,& H. Seymour.1985.A theoryof internationalproduction
industry.Universityof ReadingDiscussion Papers in InternationalInvestmentand BusinessStudies No.
87.
_____
& J.H. Dunning. 1985. Intra-industryforeign direct investment:Its rationale and trade effects.
Weltwirtschaftliches
Archiv,120:522-540.
costs in history.Journalof EuropeanEconomicHistory,14:557-574.
North,D.C. 1985. Transaction
Ohlin, B., P.O. Hesselbom & P.M. Wiskman, eds. 1977. The internationalallocationof economic activity.
London:MacMillan.
Ohmae,K. 1985. Triadpower.New York:The FreePress.
investmentin developingcountries.Paris:OECD.
Oman,C. 1984. Newforms of international
Pavitt,K. 1987. International
patternsof technologicalaccumulation.In N. Hood & J.E. Vahne,eds., Strategies
in globalcompetition.London:CroomHelm.
Porter,M.E. 1980. Competitive
strategy.New York:The FreePress.
. 1985. Competitive
advantage.New York:The FreePress.
, ed. 1986. Competitionin globalindustries.Boston:HarvardBusinessSchool Press.
Pralahad,C.K. & Y.L.Doz. 1987. Qualityof management:An emergingsourceof globalcompetitiveadvantage.
In N. Hood & J.E. Vahne,eds., Strategiesin globalcompetition.London:CroomHelm.
Robock, S.F. & K. Simmonds. 1983. Internationalbusiness and the multinationalenterprise,third edition.
Homewood:R.D. Irwin.
and the multinational
Rugman,A.M. 1979. Internationaldiversification
enterprise.Farnborough:
Lexington.
The economicsof internalmarkets.London:CroomHelm.
. 1981. Insidethe multinationals:
. 1986. New theoriesof multinationalenterprises:An assessmentof internalisationtheory.Bulletinof
EconomicResearch,38:101-118.
THE ECLECTICPARADIGM
31
D.J. Lecraw& L.D. Booth. 1985. Internationalbusiness.Firm and environment.New York:McGraw
,___
Hill.
Scott, B.R. & G.R. Lodge, eds. US competitivenessin the worldeconomy.Boston:HarvardBusinessSchool
Press.
enterprises.Basingstoke:MacMillan.
Stopford,J. 1982. The worlddirectoryof multinational
Sullivan,J.J. and I. Nonaka. 1986. The applicationof organizationallearningtheoryto Japaneseand American
management.Journalof International
BusinessStudies,Fall:127-147.
Teece,D.J. 1981.The multinational
enterprise:
Marketfailureandmarketpowerconsiderations.
SloanManagement
Review,22:3-17.
.- 1983. Technologicaland organizationalfactorsin the theory of the multinationalenterprise.In M.C.
business.London:Allen & Unwin.
Casson,ed., Growthof international
.- 1986. Transactioncost economics and the multinationalenterprise.Journal of EconomicBehaviour
and Organization,7:21-45.
___,
ed. 1987. The competitivechallenge.Cambridge,Mass.:Ballinger.
Tharakan,P.K.M.1984. The economicsof intra-industry
trade.Amsterdam:NorthHolland.
Tolentino,P. 1987. The global shift in internationalproduction:The growth of multinationalenterprisesfrom
the developingcountries:The Philippines.Ph.D. thesis,Universityof Reading.
Vernon,R. 1974. The locationof economic activity.In J.H. Dunning,Economicanalysisand the multinational
enterprise.London:Allen and Unwin.
.- 1983. Organizationaland institutionalresponsesto internationalrisk. In R.J. Herring,ed., Managing
internationalrisk. Cambridge:CUP.
.- 1985. Commenton chapterby J.H. Dunning and G. Norman.In A. Erdilek,ed., Multinationalsas
mutualinvaders.London:CroomHelm.
Williamson,0. 1981. The moderncorporation:Origins,evolution,attributes.Journalof EconomicLiterature,
19:1537-68.
Young, S., N. Hood & J. Hamill. 1985. Decision making.In Foreignownedmultinationalsubsidiariesin the
UnitedKingdom.Geneva:ILO WorkingPaperNo. 35.