EXPERIENTIAL KNOWLEDGE AND COST IN THE ... - CiteSeerX

EXPERIENTIALKNOWLEDGEAND COST IN THE INTERNATIONALIZATION PROCESS KentEriksson,*Jan Johanson**andAndersMajkgaird*** Uppsala University

D. Deo Sharma**** Umea Business School

Abstract. Using a behavioral approach, this study identifies and delineatescomponentsof experientialknowledgein the internationalization process.Three hypothesesare developedand tested. They center aroundthe lack of knowledgein the areas of foreignbusiness,foreign institutionsand firminternationalization, as well as the effect that this lack of knowledge has on managers' perceivedcost in the internationalizationprocess.With the help of a LISREL-basedstructural model,the threehypothesesare testedon a sampleof 362 servicefirms. The analysis shows that lack of internationalization knowledgehas a strongimpacton the lack of both businessand institutionalknowledge which,in turn, influencethe perceivedcost of internationalization. But there is no direct effect of lack of internationalizationknowledgeon perceivedcost of internationalization. INTRODUCTION In furtheringour understandingof the dynamicsof the internationalizationof firms,process models have played a significantrole [Bilkeyand Tesar 1977; *Kent Eriksson is Associate Professor at the Department of Business Studies, Uppsala University.His researchinterestsinclude internationalbusiness,marketingand strategy.His researchhas been publishedin the Journalof InternationalBusinessStudies,and the Service IndustriesJournal. **JanJohansonis Professorof InternationalBusinessat the Departmentof BusinessStudies, Uppsala University.His researchinterestscenter on internationalizationprocesses,business relationshipsand networksin multinationalenterprises.ProfessorJohansonhas publishedin severalacademicjournals includingthe Journalof InternationalBusinessStudies,Journalof Marketing,Journalof ManagementStudies, InternationalMarketingReview,International BusinessReview,and InternationalStudiesof Managementand Organization. ***AndersMajkgArdis a doctoralcandidateat the Departmentof BusinessStudies,Uppsala University.His researchinterestsinclude foreign market entry and internationalizationof serviceindustries. ****D. Deo Sharma is Professor of Marketing at the Ume'a Business School, and CopenhagenBusinessSchool. His researchinterestsare managementcontracts,governmentTNC relationships, and internationalization.His research has been published in the InternationalMarketingReview,InternationalBusinessReview,Journalof GlobalMarketing, ScandinavianJournalof Management,and Advancesin InternationalMarketing The authors are grateful to Professor Dag Sorbom, Department of Statistics, Uppsala University, for his valuable help concerning methodology, and three anonymous reviewers for their constructive suggestions. Received: February 1996; Revised: November 1996 & February 1997; Accepted: February 1997. 337

Palgrave Macmillan Journals is collaborating with JSTOR to digitize, preserve, and extend access to Journal of International Business Studies ® www.jstor.org

338

JOURNAL OF INTERNATIONALBUSINESS STUDIES, SECOND QUARTER 1997

Cavusgil 1980, 1984; Czinkota 1982; Johanson and Vahlne 1977, 1990; Luostarinen1980; Reid 1983]. In these models, market-specificexperiential knowledgeis central in explainingthe firm'sinternationalizationprocess.A vast amount of research,using the concept of experientialknowledge,on the internationalizationprocess and the choice of mode for entering foreign marketshas been accumulated[Barkema,Bell and Pennings 1996; Beamish 1990;Calof and Beamish1995;Erramilli1990, 1991;Erramilliand Rao 1990, 1993; Hirsch 1993; Kogut and Singh 1988; O'Grady and Lane 1996; Reid Olson and 1984;Root 1987;Sharmaand Johanson1987;Wiedersheim-Paul, Welch 1978].Surprisingly,none of the above-mentionedwork has explicitly process.This is surprisingsince dealt with the cost of the internationalization the managementof internationalization unavoidablygivesrise to the question of cost [Carlson 1974]. An internationalizationprocess entails risk and the investmentof resources.Herethe issue of the effectsof the criticalexperiential processbecomesimportant. knowledgeon the cost of the internationalization Cost aspectshavea bearingon the profitgeneratedby firms[Bilkey1982],on a firm'sinclinationto enter foreignmarkets[Dichtl, Koeglmayrand Mueller 1990],and on the selectionor changingof foreignmarketentrymode [Calof and Beamish1995]. In internationalbusinessresearchof other than process-basedtraditions,the cost of internationaloperationshas been discussed.In the eclecticapproach, for example, internationalizationis seen as resulting from three factors: namely,firm-specificadvantage,country-specificadvantage,and internalization advantage[Dunning1977, 1980,1988].Accordingto this paradigm,in the absenceof marketimperfections,firmswouldrelyon exportas the only foreign marketentry mode [Calvet 1981;Hymer 1976].In the real world of market restrictions[Kindleberger1969] imperfectionscausedby government-imposed and knowledge[Rugman1981],however,firmscan choose betweena variety of foreignmarketentrymodes,e.g., whollyownedsubsidiaries,joint ventures, licensing,and other contractualagreements.The joint influencesof the three factorson the selectionof foreignmarketentryis investigatedby Agarwaland Ramaswami[1992]. In the models based on the transactioncost approach [Williamson1975],cost factorsare criticalin explaininga firm'sinternational operations [Anderson and Gatignon 1986; Beamish and Banks 1987; Gatignonand Anderson1988].The sameholds true for internalizationtheory [Buckleyand Casson 1976].Neither eclectic theory nor the transactioncost approachelucidatethe process of internationalizationor, consequently,the cost in this process.Common to these approachesis their assumptionthat each foreignmarketentry is made in isolation [Hill, Hwang and Kim 1990; Kim and Hwang 1992]. Whatarethe cost consequencesof lack of experientialknowledgein the course of the internationalizationprocess? Since the behavioral models of the internationalizationprocessstressthe role of perceivedproblemsand oppor-

EXPERIENTIALKNOWLEDGE AND COST

339

tunitiesin the internationalization of a firm, this articlepays specialattention to the perceivedcost associatedwith internationalization. The models assume that managers act on the basis of their cost perceptions and that those perceptionsarebasedon past experience[Johansonand Vahlne1977].Accordingly,the purposehereis to identifyand delineatethe principalcomponentsof experientialknowledgethat influencemanagement'sperceptionof the cost in The mainconstituentsof experientialknowledge,and the internationalization. links between them and the cost of internationalizationin service industry firms,are investigated.The perceivedcost mayconcernnot only the directcost of starting up business abroad, in the form of traveling costs, salaries, collecting foreignmarketinformation,purchaseor rent of office space, etc., but also the costs due to changesin the organizationalset-upof the firm,such as trainingand retrainingof staff, and changesin routines,reportingsystems and procedures.In addition,opportunitycosts for seekingand evaluatinglocal partnersmay also be includedin the case of joint venturesor other alliances abroad. This article is, however, not concerned with operationalizingor measuringthe individualcost components. Since the internationalizationprocessmodels are based on the experienceof manufacturingfirms, most empirical studies of internationalizationalso concern such firms.Most of the studies on serviceindustriessupply limited insight into the internationalizationprocess of service firms [Lo and Yung 1988;Weinstein1974, 1977].A secondarypurpose of this article is thus to applythe experience-basedinternationalization processmodels to firmsin the serviceindustryand suggestpossible extensions.At the same time, the study tests the validityof the role of experientialknowledgein the model. The studydiffersfrompreviousresearchin the methodologyused. In previous research,experience,as indicatedby age or numberof countriesenteredby firms,has been used to explain the pattern of internationalizationobserved [Erramilli1991;Johansonand Wiedersheim-Paul1975;Yu 1990].This study goes beyond that explanationand introducesperceivedcost as a mediating variableand formulatesa structuralmodel of the influenceof a numberof aspectsof experientialknowledgeon perceivedcost of an incrementalstep in the internationalizationof the firm. Moreover,while most other studiesfocus on entrymode as the incrementalstep, this article,in the spiritof the process model, attemptsto capturea more generalincrementalstep by considering additionalcommitmentsto foreigncustomers.In this way,our approachfalls more in line with the basic assertions of the internationalizationprocess model. We begin with a discussionof the internationalizationprocess.This leads to threehypotheseson the perceivedcost in the internationalization process.The hypotheses focus on the role of experiential knowledge in this process. Following a presentationof the method and data employed, the empirical

340

JOURNAL OF INTERNATIONALBUSINESS STUDIES, SECOND QUARTER 1997

analysis is conducted in two steps: construct analysis and structuralmodel analysis.The resultsare then discussedand some implicationsexamined. THE INTERNATIONALIZATION PROCESS OF FIRMS

In a review of internationalizationprocess models, Andersen [1993]distinguishesbetweena U-model (Uppsala)by Johansonand Vahlne[1977]and the I-models (Innovation) by Bilkey and Tesar [1977], Cavusgil [1980, 1984], Czinkota [1982] and Reid [1983]. This paper is based on the U-model developedby Johanson and Vahlne [1977].The main reason is that the Umodel is assumedto be valid for firmsof any size while the I-modelsmay be applicableto small firms only [Andersen1993].The models are similar,however,in that they are behavioralin natureand in that experientialknowledgeis a prominentfactorin the internationalization process. The behavioralmodel of internationalizationis based on the theory of the growth of the firm [Penrose 1959] and the behavioraltheory of the firm [Aharoni1966;Cyert and March 1963],and assertsthat the internationalization of a firm is an incrementalprocess [Johansonand Vahlne 1977].The model restson the assumptionthat firmshaveimperfectaccessto information and explains internationalizationas a process of increasing experiential knowledge. It postulates an unpredictableincrementalinterplay between marketcommitmentand marketknowledgedevelopment.It also posits that experientialknowledge of the market, the clients, the problems, and the opportunitiesabroadare acquiredby operatingin the internationalmarket. Experientialknowledgenot only yields a reductionof the risks involvedin going abroad,but also providesa vehiclefor acquiringknowledgeof internal and externalresourcesand of opportunitiesfor combiningthem. The model identifiesstate and change aspects. The state aspects consist of marketknowledgeand marketcommitment.The marketcommitmentconcept is composed of two factors:the amount of resourcescommittedto a market and the degreeof that commitment.The degreeof commitmentrefersto the ease with which resources can be moved from one market to another. Knowledgeis dividedinto objectiveknowledgeand experientialknowledge(cf. Penrose [1959]). Objective knowledge is acquired through standardized methodsof collectingand transmittinginformation,i.e., marketresearch,and can easily be transferredto other countriesand replicatedby other firms.A criticalassumptionof the modelby Johansonand Vahlne[1977],supportedby Ayal and Zif [1979],Denis and Depelteau [1985],Reid [1984],Simpsonand Kujawa[1974],and Sunzook [1978],is that objectiveknowledgeis of minor importancein a firm's internationalizationprocess. The remainderof this articlefocuseson experientialknowledge,whichis viewedas a drivingforcein the internationalizationprocess [Johansonand Vahlne 1977]. Experiential knowledge is country-specificand cannot be transferredbetween firms or business units. Accumulatingexperientialknowledge is costly. These costs

EXPERIENTIAL KNOWLEDGEAND COST

341

arise because collection, transmissionand interpretationare all based on specificsituations[Carlson1974].Currentactivities,part of the changeaspect, are the primarysourceof experientialknowledge.The other change aspect is the decision to commit resources.Decisions are made when problems and opportunities arise. Firms apply the solutions that have been successfully appliedin the past [Cyertand March 1963]. In the internationalizationmodel, two dimensionsof internationalexpansion are identified:psychicdistanceand establishmentchain. The model postulates that as the psychic distance between marketsincreases,the more difficultit becomesfor firmsto collect and interpretincominginformationproperly.The firm'sexperientialknowledgederivedfrom the domesticmarketis of limited value in markets located at a great psychic distance. The psychic distance between the home and foreign markets affects market selection as well as choice of entry mode. Firms with little experienceof foreign marketsprefer those that are similarto their own domesticmarketand that are located at a short psychic distance. As firms accumulate experiential knowledge, the influenceof this kind of distance on the choice of entry mode decreases.In their study of Swedish manufacturing firms, Hornell, Vahlne and Wiedersheim-Paul[1972]found that the firmsfirst enteredmarketsat a short psychic distance and later went into more distant markets.This study was replicatedby Nordstrom[1990]who reportedalmost the same establishment pattern.Among U.S.-basedfirms,Vernon[1966],and Kogut and Singh [1988] reported a steady shift from culturally familiar to culturally less familiar markets.Firmswith vast stocks of experienceshow less preferencefor similar markets.Davidson [1983]reportedthat U.S.-basedfirms first preferto enter English-speakingcountries.Studies of service firms in this context are few. Two studies from the U.S., Weinstein[1977],and Erramilliand Rao [1993], found that such firmsstarttheirinternationalizationwith countriesat a short psychicdistance. The model postulatesthat in each country,the firm increasesits commitment incrementallyas it learnsaboutthe market.This is manifestedin a sequenceof operationalformslabeledthe establishmentchain. First the firmexports,then a marketing subsidiary is established, followed by foreign production. Johansonand Wiedersheim-Paul [1975]noted such a patternamong Swedish manufacturingfirms. Davidson [1980], Franko [1989] and Gatignon and Anderson [1988] have observed an increasing propensity to select wholly owned subsidiariesas experiential knowledge increases. Davidson [1980], Goodnow and Hansz [1972],and Kogutand Singh[1988]showedthat, among Americancompanies,the usage of joint venturesincreaseswith sociocultural distancefromthe domesticmarket.Similarresultswerereportedby Gatignon and Anderson[1988],Greenand Cunningham[1975],and Stopfordand Wells [1972].Calof and Beamish[1995]reportedmixedfindings,but on the whole in line with the above.

342

JOURNAL OF INTERNATIONALBUSINESS STUDIES, SECOND QUARTER 1997

Someconfusiondoes, however,exist. Maclayton,Smithand Hair [1980]found no significantcorrelationbetweena firm'sexperiencein foreignmarketsand its evaluationof individualmarkets.Daniels, Ogramand Radebaugh[1976] and Shetty [1979]reporteda shift toward licensing and joint venturesas a firm's experientialknowledge grew. Davidson and McFetridge [1985] and Hedlundand Kverneland[1985]showeda decreasein the relianceon wholly owned subsidiaries as foreign experience increases. Bureau of Industry Economics[1984],Millingtonand Bayliss[1990],and Newbould,Buckleyand Thurwell[1978]publishedsimilarfindings.Stopfordand Wells[1972],studied U.S. investmentsin Europe,concludingthat almost three-quartersof initial venturesused wholly owned subsidiaries.In general, these researchershave studiedonly the earlystages of foreignmarketentry [Erramilli1991].In light of these findings,Erramilli[1991]postulateda U-shapedrelationshipbetween experientialknowledgeand a firm'spropensityto rely on high-controlmodes of entry into foreign markets.O'Gradyand Lane [1996]demonstratedthat entering a foreign market near the domestic market may result in poor performance.On the whole, currentresearchsupports the view that, in the internationalizationprocess,experienceis importantin the selectionof entry mode. THE COST OF LACK OF EXPERIENTIAL KNOWLEDGE IN THE INTERNATIONALIZATION PROCESS: THREE HYPOTHESES

On the basis of the literatureon the internationalizationprocess of firms reviewedabove,threehypothesescan now be developedand specified. Firms start their operations in domestic markets. Initially, firms develop routinesand administrativestructuresto managedomesticmarketoperations. The routinesand administrativestructuresthus developedare not sensitiveto the signalsflowingfrom the overseasmarkets.As firms operateabroad,they accumulateexperientialknowledgeand change their routines and administrative structures.This process is gradual. The internationalizationprocess model explicitlyemphasizesthat the internationalizationof a firm entailsthe commitment of exchange-specificand market-specificresources. These exchange-specificcommitmentsare triggeredwhen firms acquireexperiential knowledgeof a particularmarket[Bilkeyand Tesar1977;Bureauof Industry Economics1984;Millingtonand Bayliss1990;Newbouldet al. 1978].Indeed, in their original model, Johanson and Vahlne [1977] explicitlypostulate a positiverelationbetweenexperientialknowledgeof a marketand the level and speed of resourcecommitmentin that market.This relationis due to a greater ability to detect the opportunities and reduce the uncertaintiesof going abroad. These market-specificresource investmentscan take the form of human resources, technology, or know-how. Resources are by definition valuableand are costly to obtain. Increasingexperientialknowledgetriggers greaterresourcecommitmentto a particularmarket.The experience-seeking

EXPERIENTIALKNOWLEDGEAND COST

343

firm must engage in foreign operations. This implies a presence abroad, exposure to the situation abroad, and interaction with specific customers, intermediariesand other firms in the internationalmarket. Acquiringlocal experientialknowledgeis time-consumingand importantfor performance,as demonstratedby a study of Japaneseventuresin Asia [Makino and Delios 1996]. A criticalconsiderationin internationalizationis the compatibilitybetweena firm's existing resourcesand those needed in a particularforeign market. Knowledge is requiredboth about the market and the firm [Johansonand Vahlne 1977; Madhok 1996, 1997]. Thus, there is a need for experiential knowledgeof the firm'scapabilityand resourcesto engage in international operations,which we label "internationalization knowledge,"and knowledge of the foreign markets in which the firm is going to operate [Yu 1990]. Experientialmarket knowledge pertains to two different aspects: business knowledgeand institutionalknowledge.By "foreignbusinessknowledge"we mean experientialknowledgeof clients,the market,and competitors."Foreign institutional knowledge"refers to experientialknowledge of government, institutionalframework,rules,norms,and values. A lack of experientialknowledgeof a particularclient'sway of working,its organizationand decisionmaking,and its particularneeds regardinggoods and services,is problematicfor an internationalizingfirm.All this calls for the cultivationof businesscontactsto acquirea first-handfeel for the preferences, practicesand customsin the market[Denisand Depelteau1985;Reid 1984].A local presence,e.g., a subsidiary,allows the internationalizingfirm to gain moredifferentiatedknowledgeof the clientsand the local business.Likewise,it enablesthe internationalizingfirm to accumulatethe kind of informationit really needs, and to interpret the information in a firm-specificcontext [Carlson 1974]. With increasingexperientialknowledge of the clients, the market,and the demandsituation,firmsbecomemore able to perceiveopportunities in foreign markets,therebyreducinguncertainty[Kogut and Singh 1988]. Firms with experientialknowledgedevelop a position in a particular market, and their engagementin that specific market graduallyincreases. Thus, Davidson [1980, 1983] and Franko [1989] noted that having a production base in a foreign market has a positive impact on the subsequent penetrationof that market. As discussed above, there are some empirical results that contradict this, but generally speaking, it can be said that experientialknowledgedoes influencea firm'sinternationalization process. The extent to which a firm'scurrentexperientialbusinessknowledgemay be used in the internationalizationprocess may vary. Firms that elicit the assistanceof their currentexchangerelationshipsin internationalizationhave access to the experientialknowledge requiredfor this internationalization. Thesefirmshavealreadycommittedclient-specificresourcesthat may be attni-

344

JOURNAL OF INTERNATIONALBUSINESS STUDIES, SECOND QUARTER 1997

buted to the internationalizationprocess to facilitatingexchange [Dunning and McQueen 1983;Jones 1981;Sagari 1992;Seymour1986;UNCTC 1981; Walter1985].Few extra resourceshave to be investedin seekingexperiential knowledge[Erramilliand Rao 1990;Sharmaand Johanson1987].Otherfirms must internationalizeon their own and spend resourceson detecting and exploitingthe businessopportunitiesin the internationalmarket.They lack experienceof new marketsabroad and of specificforeign clients, and must acquireexperientialknowledgeof the internationalmarket.This is costly.And dependingupon the level of the firm'sexperientialbusiness knowledge,the cost of the internationalization processwill vary. Hi: The greatera firm'slack of foreignbusinessknowledge,the higher

the perceivedcost of the internationalization process. As we have noted, the internationalizationmodel rests on the assumptionof imperfectknowledgeon the part of the decisionmakersin the internationalization process.This lack of knowledgeconcernsthe institutionsto be found in foreign markets, foreign governmentsand bureaucracies,and the ways in which these work. A lack of experientialinstitutional knowledge is problematic,as it is difficultfor the companyto acquirean adequateunderstanding of the technical and commerciallaws and norms that apply in a foreign market.It is not only importantto know what the statute-bookssay,but also how the law is appliedin practiceat a particulartime by a particulargovernment agency.It may be a question of the import and export of goods and services,tariffs,local taxes,generalconditionsin the market,as well as related problemsand prospects.Knowledgeof these institutionalmattersis a source of advantage[Jansson,Saqib and Sharma 1995; Lenwayand Murtha 1994; Murthaand Lenway1994;Stopfordand Strange1991;Yoffie1988].The same is trueof knowledgeof the language[Dichtlet al. 1990]and of the local culture [Hofstede 1984a,b],both of which facilitatebecomingacquaintedwith local needs and requirements.A knowledgeof local institutionalnorms and laws reducesthe cost in internationalization. H2: The greatera firm'slack of foreigninstitutionalknowledge,the process. higherthe perceivedcost of the internationalization However,the experientialcost of the internationalizationprocessis not only relatedto knowledgeof foreignbusinessand institutions,it is also a matterof learninghow to organize and manage internationalizationefforts [Ball and Tschoegl 1982;Terpstraand Yu 1988]. Organizationalroutines,procedures, and structuresmatter,and aredecisivein controllingbehaviorin organizations [Cyertand March 1963;Marchand Simon 1958].They makeup an organization'sknowledgeassets,are accumulatedover time, and constituteconditions for subsequentbehavior.Internationalizationis no different.Organizational routinesare, however,opaque [Kilduff1992, 1993;Nelson and Winter 1982; Polanyi 1962]and decisionmakers'knowledgeon these routinesis imperfect.

EXPERIENTIAL KNOWLEDGEAND COST

345

This, as pointed out by Winter[1987],is becausedecisionmakersin a firm are the "symbolprocessingbrains"of the organizations.A greatpart of organizational routinesoperateoutside the formal hierarchyand structureof a firm [Kilduff1993].Moreover,as routinesoutlast individuals,decisionmakerslack a complete knowledgeof organizationalroutines.A firm must considernew situationsand problemsthat arise in connection with internationalbusiness initiativesand how they affect the existingresourcesand routinesof the firm [Madhok 1997]. When entering a foreign market, experientialknowledge about internationalbusinessis gained and stored in the firm'sroutines and programs[Nelsonand Winter1982].This accumulatedexperientialknowledge then exertsan influenceon the futureinternationalizationof the firm through its influenceon informationsearchprocesses,e.g., what type of informationis sought, and where.It also affects the firms' ability to evaluateinternational initiatives.It can be expected that the less experiencea firm has of international business, the less knowledge it will have on how to organize internationaloperations[Madhok 1996]The perceivedcost of furtherinternationalizationwill thereforbe greater[Yu 1990].Accumulatedexperiencein is neitherspecificto a countrynor to a mode of entry.It is internationalization firm-specificand constitutesa particularfirm's"wayof going international." A firm's experience of organizing internationalization,experiential internationalizationknowledge,means knowing what knowledge is requiredin differentsituationsand differentsettingsconnectedwith internationalization, and whereto seek this knowledge. H3: The greatera firm'slack of internationalizationknowledge,the higherthe perceivedcost of the internationalization process. The threehypothesesare combinedin a structuralmodel, with perceivedcost in the internationalization processas a dependentvariable.The model is tested empiricallyon a sampleof servicefirms. As mentionedat the outset of the article,thereis reasonto discusswhetherthe argumentspresentedin the previouspages apply to firms in both the manufacturingand the servicesectors.The questionis somewhatcontroversial.On the one hand, thereare researchers[Gronroos1990;Lovelock 1988;Normann 1984] who argue that service firms differ in nature from manufacturing companies.Others [Buckley,Pass and Prescott 1992; Erramilli1991; Levitt 1972;Quinn and Gatignon 1986;Sauvantand Mallampally1993]claim that these differencesare differencesof degree,not of nature.Although we cannot hope to resolvethis controversyin a studybasedon servicefirmsalone,we can take a step towardsresolvingit by comparingfirms in which the servicesare product-relatedand firmsofferingpureservices.Thereis reasonto expectthat the need for experientialknowledgein the internationalizationof a firm may vary dependingupon the type of industryin which the firm is active.In more product-relatedindustries, a firm may benefit more from experiential

346

JOURNAL OF INTERNATIONALBUSINESS STUDIES, SECOND QUARTER 1997

knowledgeof the productand of its functioning.As the servicecontent of a experientialknowledgebecomesmore firm'sofferingincreases,market-related important. Consequently, we have reason to investigate the differences between these two types of industryas regardsthe experientialcost in the internationalization process. EMPIRICAL METHOD

The general incremental character of the internationalizationprocess, whereveror wheneverit occurs,is a fundamentalassumptionof the process model. Obviously,it is difficultto operationalizethis processin a standardized researchdesign.In orderto captureit, however,the studyis designedto focus on an incrementalinternationalizationcommitment.Respondentsin eight differentservice industrieswere asked questions relatedto their lack of the three kinds of knowledgeand about the perceivedcost associatedwith an incrementalmarketcommitment.In orderto get a measurethat did not relate decision,incrementalmarketcommitment to any specificinternationalization has been measuredas the executionof an additionalclient orderabroad.The term "additional"implies receivinga new assignmentfrom customers.This includessituationswherea firm is going abroadfor the first time or wherea firm that is already operating abroad executes an additional assignment abroad.The firmmay providethe serviceabroadwithoutmovingabroad.The respondentswere not asked to consider any specific market. This made it possibleto analyzethe generaleffectof the knowledgefactorson the perceived cost in the internationalization process. On the basis of informationfrom personalinterviews,a questionnaire-based statistical survey was conducted. We systematicallysearched for Swedish service firms engaged in internationaloperations. The Central Statistical Bureau in Sweden did not have data on the internationaloperations of Swedish service firms. Therefore,we searched for information from three 1 secondarysources;traderegisters,branchregistersand businesspublications. Wealso conductedsome seventyface-to-faceinterviewswith CEOsof Swedish servicefirms.Duringthese interviews,we requestedthe respondentsto supply the names of other firms in their line of business that were doing business abroad. As Table 1 shows, there is a higher representationof firms in engineering,architectureand managementconsulting. Altogether 774 companies were included in the mail survey.The questionnaireswereaddressedto the presidentsof these companies,who weredeemed decisionprocessof their most likelyto be involvedin the internationalization firms.The informationwas suppliedby executivesactuallyengagedin foreign operationsof the firm. They are presidents(329 cases), vice-presidentforeign operations,vice-presidentfinance and others (33 cases). A five-pointLikert scale (rangingfrom "not at all important"to "veryimportant")was used.

EXPERIENTIALKNOWLEDGEAND COST

347

TABLEI Sample Size by Industry Service Industry

n

%

Legal Engineering and architecture Computer software and data processing Advertising Accounting Education Management consulting Miscellaneous services (maintenance, leasing, etc.)

16 119 36 54 17 19 78 23

4.5 32.9 9.9 14.9 4.7 5.2 21.5 6.4

Total

362

100

Seventy-threequestionnaireswere returnedundeliveredand forty-ninecompaniesexpressedregretat theirinabilityto participatefor variousreasons,the most common of which was their no longer being engaged in international business.Of the remaining652 potential respondents,usable answerswere submittedby 409. The responserate of 62.7%comparesfavorablywith rates reported in other surveys involving service firms (e.g., Erramilli [1991]; Zeithaml,Parasuramanand Berry[1985]).An additional47 of the 409 were droppedfrom the analysisfor havinginsufficientinformationon a numberof variables.The remaining362 firmsprovideddata on all the key variables. Among the respondentfirms,34 went abroadduringthe 1960sor earlier.171 firms carriedout their first assignmentabroadin 1980 and later. One firm, VBB, carriedout their first foreignoperationas early as 1902. 152 firms had 1-10 employees,and 68 firmshad 100 or more employees. A standardtest of nonresponsebias was conducted.Earlyrespondentswere compared to late respondents,with late respondentsbeing assumed to be similarto nonrespondents[Armstrongand Overton 1977].Accordingly,the samplewas split into two categorieson the basis of surveyreturndates,with the first65%classifiedas earlyand the last 35%as late respondents.They were viewed as representativeof actual nonrespondents.We found no significant differences between these two groups on variables such as number of employees,total turnoverand industrydistribution.Nonresponsebias is thus not a problem. LISREL

The hypothesizedcausal relationswere investigatedby LISREL, which is a structuralequation modeling method. On the recommendationof J6reskog and S6rbom [1993], we used LISREL for explorativepurposes as well as hypothesis-testing.A structuralmodel with LISREL has two components. The first is a set of indicators related to a higher order variable, which representslatent, common propertiesof the indicators.The second is the

348

JOURNAL OF INTERNATIONALBUSINESS STUDIES, SECOND QUARTER 1997

definitionof causalrelationsbetweenthe latentvariablesin a structuralmodel. In technicalterms,LISRELderivescausal structuresby analyzingboth error covariancesand regular correlations[Bollen 1989; Bollen and Long 1993; Hayduk1987;Joreskogand S6rbom 1993]. The confirmatoryanalysisis performedin two steps, the first step of which investigateswhether the indicators are valid measures of the theoretically deducedconstructs.For this purpose,the indicatorsare groupedaccordingto theoreticalassumptions.In the second step, the causal relationsbetweenthe constructsare analyzedaccordingto the hypothesizedstructuralmodel. The validityof LISRELmodelsis estimatedby the validityof the entiremodel (nomologicalvalidity),and also by the extentof separationbetweenconstructs (discriminantvalidity) and the homogeneityof constructs(convergentvalidity). The overallfit of the LISREL models is assessedby x2 and degreeof freedommeasures,and a probabilityestimate(p-value)[Joreskogand Sorbom 1993:121].The x2 and degree of freedom, together, measure the distance betweendata and model, and the p-value is a significanceestimate.Together these constitute our measure of nomological validity. Discriminant and convergentvalidityarejudged by studyingthe t-valuesand R2-valuesof each relationshipin the model. The R2-valueis a measureof the strengthof a linear relationship estimate [Joreskogand Sorbom 1993:121],and t-values test significance[Joreskogand Sorbom1993:108].The resultsof the validityof our constructsare shownin Table2. Pairwisedeletion is used to account for missingvalues.The total numberof missingvalues rangedfrom 35 to 106, dependingon the item (see Appendix). We also tested the resultsby listwisedeletionwith similarresults. CONSTRUCT ANALYSIS

The construct"lack of businessknowledge"is meant to capturethe lack of business knowledge about competitors, clients and markets abroad. The constructconsists of two indicators(Table2). The two indicatorsreflecttwo important ways of gaining foreign business knowledge. They concern the respondents'evaluationof the lack of foreign subsidiariesor representative companiesabroad,or the lack of cooperativeagreements.These may include agreementswith agents and alliance partners.The t-values are 16.32 and higher,and the R2-valuesare above0.68, suggestinggood convergentvalidity for the constructs. The construct"lackof institutionalknowledge"reflectsknowledgeabout the institutional conditions of foreign markets.The construct consists of two indicators.They concerna lack of knowledgeaboutthe language,laws,norms and standardsin foreign markets.That both indicatorsconstitute a latent variableis validatedby t-valuesabove 14.42,and R2-valuesabove0.56.

EXPERIENTIALKNOWLEDGE AND COST

349

TABLE2 The Constructs and Their Indicators Constructs Lack of business knowledge Lack of institutional knowledge

Lack of internationalization knowledge Perceived cost

Indicators Lack of cooperative agreements (COOP) Lack of subsidiary (SUB) Lack of knowledge about foreign laws/norms/ standards (INSTITUTE) Lack of foreign language (LANGUAGE) Lack of foreign experience (FOREXP) Lack of unique knowledge/ competence (UNIQCOM) Perceived cost of an additional assignment abroad (COST)

A

R2

t

0.86

0.75

17.22

0.82

0.68

16.32

0.79

0.63

15.21

0.75

0.56

14.42

0.93

0.87

19.12

0.70

0.49

13.87

1.00

1.00

The construct "lack of internationalizationknowledge" represents the accumulatedinternationalizationexperiencegained by a firm in its international operations.The constructconsists of two indicators.The first is the respondent'sevaluationof the importanceof a lack of experiencein doing business abroad. The second is a lack of unique knowledge and/or competence.The constructis valid, t-valuesare above 13.87 and the R2-valuesare above0.49. The construct "perceivedcost" consists of one indicator based on the perception of the overallcost of executingan additionalclient orderabroad.This single indicator is assumed to capture the overall cost judgments made by managers. To assess discriminantvalidity, a model with no causal relations between constructs (measurementmodel) is created. Our set of constructs are discriminantlyvalid. Key statisticalestimatesshow that no pair of constructsis unidimensional. STRUCTURAL MODEL ANALYSIS

The second step of the empiricalinvestigationis to test the model of causal relationshypothesized.Figure 1 depicts the model of relationsbetweeninstitutional knowledge,businessknowledge,internationalizationknowledge,and cost. The model'skey statisticalmeasuresare acceptable,since X2=16.76 (9) andp=O0.053.T-valuesare 1.96 and higher,and R2-valuesare 0.21 and higher, except for internationalizationknowledgewhose relationto perceivedcost is insignificant.Lack of institutional(0.19) and business(0.26) knowledgeboth lead to higherperceivedcosts.

350

JOURNAL OF INTERNATIONALBUSINESS STUDIES, SECOND QUARTER 1997

FIGURE1 Model of Hypothesized Causal Relations

| usiness \ knowledge \26

0.19 (19)

acX

International-, ization

(3.72)

00

.9(.1

09

I

knowledge X2=1 6.76(9)

p=0.053

Note: The figures given are factor loadings of causal relations with t-values in parentheses.

The insignificantrelation between internationalizationknowledgeand perceived cost providessome clues to a furtherelaborationof the model. In the discussionof the hypotheses,the need for compatibilitybetweenknowledge about the firm's resources and knowledge about the market is stressed. According to this view, it is not possible to articulatethe need for market knowledgewithout knowingthe specificresourcesof the firm. This suggests that thereis reasonto expectcausalrelationsfromlack of internationalization knowledgeto lack of both businessknowledgeand institutionalknowledge. Such a structuralmodel is depicted in Figure 2. The model's key statistical estimatesare good since x2= 19.37(11)and p=0.055. All t-valuesare 3.78 or higher and the lowest R2 is 0.22. The statisticsof this model (Figure 2) are better than those of the previousmodel (Figure 1).The analysis shows that causalrelationsbetweeninternationalization knowledgeon the one hand, and businessand institutionalknowledgeon the other,are strong.Both business knowledgeand institutionalknowledge,in turn,influencethe perceivedcost of internationalization. There is no direct relation, however, between internationalization knowledge and perceived cost. Internationalization knowledge operates only via the more specific constructs of business and institutional knowledge. The LISREL estimates show, however, that the knowledgeon perceivedcost are indirecteffectsof lack of internationalization 0.36, t=7.87. The magnitudeof the costs associatedwith collectingbusiness processare knowledgeand institutionalknowledgein the internationalization similar, i.e., managers perceive that it costs as much to gather business knowledgeas it does to gatherinstitutionalknowledge.

351

EXPERIENTIALKNOWLEDGEAND COST

FIGURE2 Final Structural Model

0.83 (18.8)

kwd

0.9

(14.90)

business

Lack ofac

\

>,.92 (19.52)

Perceived 1.00 cs cs

International( ization knowledge UNIQCOM

0.71 (14.18) 0.74 (8.23)

29(.7

COST

0.25 (3.78) Lackof InstitutioIa ,knowledg

0.75 (11.03)

LANGUAGE

X2=19.37(11)

\

0.79 (10.99)

IINSTITUT

p=0.055

Note: The figuresgiven are factor loadingsof causal relationswith t-values in parentheses.

In orderto test the validity of the analysis,the samplewas dividedinto two groups,one in whichthe tangibleelementsin the servicessold aremorereadily identified, such as product-relatedbusinesses, such as computer software, engineering and architecturefirms (n= 100, 33%), and one in which the tangible elementsin the service providedare less readily identified,such as non-product-relatedfirms dealing in management consulting, education, accounting,advertising,and legal services(n=207, 67%).The analysisshows the groupsto be identical(, = 50.08(39), p=0.1). As a furthertest of the model, the samplewas dividedin largeand smallfirms. It is sometimesassumedthat largesize is associatedwith moreexperiencethan small size.With the limit set at twentyemployees,the numberof smallfirmsis 158 (62%),and of large firms 95 (38%).The group analysis shows that the causal structure is valid in both groups after three modifications (x2= 50.05(36),p=0.06). The modificationsshow that the magnitudeof indicator loadings varies betweengroups.The structuralmodel is valid for both large and smallfirms. Since severalof the empiricalstudies of the internationalizationprocess use durationof internationaloperationsas a proxyfor experience(Erramilli1991)

352

JOURNAL OF INTERNATIONALBUSINESS STUDIES, SECOND QUARTER 1997

we also dividedthe sampleinto firmswith short and long priorinternational operations.Ninety-one firms (36%)had three or less years of international experience,and 162(64%)had morethan threeyearsexperience.Resultsshow that the causal structureis valid in both groupsafter one modification(X2= 53.17(38), p=0.05). This third test of the robustness of the cost model demonstratesthat the model is valid irrespectiveof experienceof international operations. Altogether, the three group analyses provide a strong test of the general validityof the causalstructurelinkingthe experientialknowledgecomponents process. to each other and to perceivedcost in the internationalization DISCUSSION AND CONCLUDINGREMARKS In contrastto earlierempiricalresearchon internationalizationof firms,this study addresses the effect of experiential knowledge on the cost of the process.Ourresultsconfirmthat thereis a cost associated internationalization of a firm.The analysis with experientialknowledgein the internationalization shows that in the processof internationalizing,firmshaveto seek experiential knowledgeon individualclientsand markets,as well as on institutionalfactors such as local laws, local governmentsand local cultures.This informationis collectedthroughactivitiesabroadand througha presencein foreignmarkets. Activitiesand presenceabroadentail costs. These costs are relatedto collecting, encoding,transferring,and decodingknowledge,as well as changingthe resourcestructures,processesand routinesin the organization. The analysis also shows that a firm'sexperienceof the internationalization process influencesthe perceivedcost in this process.This implies that some experientialknowledgeis located in the firm, in its decisionmakingroutines and structures. Johanson and Vahlne [1977] suggest that the relevant experienceof a firmconcernsspecificmarkets.Ourfindingsindicate,however, that accumulatedinternationalizationexperiencethat affects both business knowledge and institutional knowledge, is not related to specific country markets.It is a firm-specificexperiencerelevant to all markets.But firmspecific internationalizationknowledge affects the perceivedcost in internationalization only indirectly,through experientialmarket knowledge. It seems reasonableto regardit as a kind of proceduralknowledgeconcerning, for instance,what kind of knowledgea firmneeds in differentsituations.This finding may explain the results obtained by Erramilli[1991], that scope, measuredin termsof the numberof countriesin whichthe firmhas enteredin the past, explainsthe choice of entry mode into a new foreignmarket.More differentiatedresearch on the nature of this knowledge and how it is accumulatedand transferredfrom one countryto anotheris needed. Ratherthan being conclusive,the findingsof this study open new avenuesfor furtherresearch.Both modeling and measurementcan be developed.There

EXPERIENTIALKNOWLEDGEAND COST

353

are other samplesto be investigatedand other measuresto be used. In future research, constructs such as managers' "perceptionof costs" should, for example,be measuredin a moresophisticatedmanner.The studyhas, however, indicateda possiblefruitfulapproachto furtherstudy of internationalization processes.Such an approachimpliesa focus on the perceptionsand cognitive structures of managers who enact and have personal experience of the internationalizationof firms. As shown by this study, LISREL analysis of structuralmodels can be a powerfultool in enhancingour understandingof internationalization processes. The findingsof the study also have managerialimplications.First, the view that internationalization is a matterof learningis supported.But it is not only a matterof learningabout foreignmarketsand institutions.Knowledgemust also be gainedon the internalresourcesof a firm,and whatthe firmis capable of whenexposedto new and unfamiliarconditions.The resultssuggestthat, in internationalizing,a firm must develop structures and routines that are compatiblewith its internalresourcesand competence,and that can guide the searchfor experientialknowledgeaboutforeignmarketsand institutions.On a differentlevel,this can be viewedas the need to developa cognitiveframework showingwhat furtherknowledgeabout foreignmarketsis relevant. Our resultsshow that Swedishmanagersfeel that a deficiencyin knowledgeof language is a problem in the internationalizationprocess of firms. In this respect,our findingsdifferfrom those of Beamishand Calof [1989]. One critical issue of this study is the perceptionsof the firms involved in internationalization.How do these perceptionsrelate to the real situation? This is importantfor the internationalizing firm.Managerialperceptionsguide decisionmaking,the outcome of which, in time, will reveal the success or failureof these decisions and the accuracyof the cost perceptions[O'Grady and Lane 1996].The outcomesare real and form the platformfor experiential knowledge.Successand failureteachesa firmwhatcourseof action is suitable in a specific internationalsetting. In fact, there is reason to expect that internationalization decisionsbased on inaccuratecost perceptionswill havea stronger impact on experiential learning than those based on accurate perceptions. Outcome-based knowledge is invested in the organizational routinesand processes,and thereis reasonto assumethatmanysmallmistakes in gradual internationalizationallow management to form more realistic perceptionsthan do a single great mistake in a leap-froggingapproachto internationalization. This should, however,be investigatedempirically Moreover,this demonstratesthat internationalizationis a process that is difficult to plan in advance. The structuresand routines mentioned above cannot be establishedbeforehand, but must be built gradually as a consequenceof learningboth a firm'scapabilitiesand foreignmarketneeds.In this process, understandingthe history of the firm is crucial. The planning of

JOURNAL OF INTERNATIONALBUSINESS STUDIES, SECOND QUARTER 1997

354

internationalizationefforts must allow for considerableadjustment,use of slack resources,and restructuring. The articleindicatesthe importanceof threeroadsto experientialknowledge: local presence,repetitionand variation.We arguethat experientialknowledge of foreign markets requires durable and repetitive interactions abroad. Sporadic interactionwith market actors abroad produces little experience. Variation,i.e., presenceof an internationalizingfirm in a variety of foreign markets,will enrich the organizationalroutinesand proceduresin the firm. and interpretingcapabilitiesof This will also improveinformation-searching the firm. Attention to these aspects is critical to a firm's efforts to internationalizeoperations.

APPENDIX Correlation Matrix Sample size=362 1

2

3

4

5

6

7

COST LANGUAGE INSTITUTE SUB COOP FOREXP UNIQCOM

1.00 0.29 0.30 0.36 0.34 0.35 0.27

1.00 0.60 0.25 0.35 0.51 0.40

1.00 0.34 0.35 0.54 0.37

1.00 0.71 0.46 0.32

1.00 0.48 0.40

1.00 0.65

1.00

Mean: Standard deviation: Missing values:

2.70 1.15 36

2.78 1.26 37

2.84 1.12 38

2.35 1.29 103

2.40 1.24 106

2.90 1.25 39

2.64 1.32 35

1. 2. 3. 4. 5. 6. 7.

The managerswereaskedthe following: How importantare the followingfactors as obstaclesfor the possibilities of your firmto acquireassignmentsfrom abroad? 1. High costs 2. Lack of languageknowledge 3. Lack of knowledgeof foreignlaws/norms/standards outside Sweden 4. Lack of subsidiaries/branches 5. Lack of cooperativeagreementswith foreignfirms 6. Lack of foreignexperience 7. Lack of uniqueknowledge/competence NOTE 1. The following sources were used: Advokatsamfundets matrikel, Affdrsvdrlden, Dagens Industri, Foretagskatalogen, Konsultguiden Tekniska Konsultguiden, The Association of Public Relations Consultancies in Sweden and VeckansAffairer.

EXPERIENTIALKNOWLEDGEAND COST

355

REFERENCES Agarwal,Sanjev& SridharN. Ramaswami.1992. Choice of foreignmarketentry:Impactof ownership, location and internationalizationfactors. Journal of International Business Studies, 23(1):1-27. Aharoni, Yair. 1966. The foreign investment decision process. Boston, Mass.: Division of Research,GraduateSchool of BusinessAdministration,HarvardUniversity. processof firms:A criticalanalysis.Journal Andersen,Otto. 1993.On the internationalization of International Business Studies, 24(2): 209-32. Anderson,Erin & HubertGatignon. 1986. Modes of entry:A transactionscost analysisand propositions. Journal of International Business Studies, 17(3): 1-26.

Armstrong,J. Scott & TerryS. Overton. 1977. Estimatingnon-responsebias in mail surveys. Journal of Marketing Research, 14(3): 396-402. Ayal,Igal & JehielZif. 1979.Marketexpansionstrategiesin multinationalmarketing.Journal of Marketing, 43(2): 84-94.

Ball,CliffordA. & AdrianE. Tschoegl.1982.The decisionto establisha foreignbankbranchor subsidiary:An applicationof binary classificationprocedures.Journal of Financial and 411-24. QuantitativeAnalysis. 17(September): Barkema,HarryG., John H. J. Bell & JohannesM. Pennings.1996. Foreignentry,cultural barriers,and learning.Strategic Management Journal, 17(2):151-66. processfor smallerOntariofirms:a research Beamish,Paul W 1990.The internationalization agenda.In A. Rugman,editor,Research in global business management, 1: 77-92. Greenwich, Conn.:JAI Press. & John C. Banks. 1987. Equity joint ventures and the theory of the multinational enterprise.Journal of International Business Studies, 19(2):1-16. & JonathanL. Calof. 1989.Internationalbusinesseducation:A corporationview.Journal of International Business Studies, 20(3): 553-64.

Bilkey,WarrenJ. 1982.Variablesassociatedwith export profitability.Journal of International Business Studies, 12(2): 39-55.

& George Tesar. 1977. The export behaviorof smallersized Wisconsinmanufacturing firms.Journal of InternationalBusiness Studies, 8(1):93-98. Bollen,KennethA. 1989.Structural equations with latent variables.New York:JohnWiley. & Scott J. Long, editors.1993. Testing structural equation models. NewburyPark,Calif.: Sage. Buckley,Peter J. & Mark Casson. 1976. The future of the multinational enterprise. London: Macmillan. Buckley,PeterJ., C. L. Pass & KatePrescott.1992.The internationalization of servicefirms:A comparisonwith the manufacturing sector.ScandinavianInternational Business Review, 1(1): 39-56.

Bureauof IndustryEconomics.1984.Australian direct investmentabroad. Canberra:Australian GovernmentPublishingService. Calof, Jonathan L. & Paul W Beamish. 1995. Adapting to foreign markets:Explaining internationalization. International Business Review, 4(2): 115-31. Calvet, A. L. 1981. A synthesis of foreign direct investmenttheories and theories of the multinationalfirm.Journal of International Business Studies, 12(1): 43-59.

356

JOURNAL OF INTERNATIONALBUSINESS STUDIES, SECOND QUARTER 1997

Carlson,Sune.1974.Internationaltransmissionof informationand the businessfirm.Annals of the American Academy of Political and Social Science, 412: 55-63.

Cavusgil,S. Tamer.1980. On the internationalizationprocess of firms.European Research, 8(November):273-81. . 1984. Organizationalcharacteristicsassociated with export activity. Journal of Management Studies, 21(1):3-22.

Cyert, Richard M. & JamesG. March. 1963. A behavioural theory of the firm. New York: Prentice-Hall. Czinkota,Michael. 1982. Export development strategies: US promotion policies. New York: Praeger. DanielsJ.D., E. W Ogram& L. M. Radebaugh.1976.Internationalbusiness: Environmentsand operations. Reading,Mass.:Addison-Wesley. Davidson, William H. 1980. The location of foreign direct investmentactivity: Country characteristics and experienceeffects.Journal of InternationalBusiness Studies, 11(2): 9-22. . 1983. Marketsimilarityand marketselection:Implicationsof internationalmarketing strategy.Journal of Business Research, 11: 439-56. & Donald G. McFetridge.1985. Key characteristicsin the choice of international technologytransfermode.Journal of InternationalBusiness Studies, 16(2): 5-21. Denis, Jean-Emile& Daniel Depelteau. 1985. Marketknowledge,diversificationand export expansion.Journal of InternationalBusiness Studies, 16(3):77-89. Dichtl, Erwin,Hans-GeorgKoeglmayr& StefanMueller.1990.Internationalorientationas a preconditionfor exportsuccess.Journal of International Business Studies, 21(1): 23-41. Dunning,John H. 1977.Trade,location of economicactivityand the MNE: A searchfor an eclecticapproach.In B. Ohlin, P. Hesselborn& M. Wijkman,editors, The international allocation of economic activity. New York:Holmes& Meier. . 1980. Towardsan eclectic theory of internationalproduction:Some empiricaltests. Journal of International Business Studies, 11(1):9-31.

. 1988.The eclecticparadigmof internationalproduction:A restatementand somepossible extensions.Journal of InternationalBusiness Studies, 19(1):1-31. & MatthewMcQueen.1983.The eclectictheory of the multinationalenterpriseand the internationalhotel industry.In A. M. Rugman, editor, New theories of multinational enterprise.New York:St. Martin's. Erramilli,M. Krishna.1990.Entrymode choice in serviceindustries.International Marketing Review, 7(5): 50-62. . 1991.The experiencefactorin foreignmarketentrybehaviorof servicefirms.Journal of International Business Studies, 22(3):479-501.

& C. P. Rao. 1990.Choiceof foreignmarketentrymode by servicefirms:Role of market knowledge.Management InternationalReview, 30(2): 135-50. . 1993.Servicefirmsinternationalentrymode choice:A modifiedtransaction-costanalysis approach.Journal of Marketing, 57: 19-38. Franko,LawrenceG. 1989.Use of minority50-50 joint venturesby U.S. multinationalsduring the 1970s:The interactionof host country policies and corporatestrategies.Journal of International Business Studies, 20(1): 19-40.

EXPERIENTIALKNOWLEDGE AND COST

357

Gatignon,Hubert& Erin Anderson. 1988.The multinationalcorporation'sdegreeof control overforeignsubsidiaries:An empiricaltest of a transactioncost explanation.Journal of Law, Economics, and Organization,4(2): 305-35. Goodnow, James D. & JamesE. H. Hansz. 1972. Environmentaldeterminantsof overseas marketentrystrategies.Journal of International Business Studies, 3(1):33-50. Green, R. T. & W M. Cunningham.1975. Determinantsof U.S. foreign investment:An empiricalexamination.Management International Review, 15: 113-20. Gronroos,Christian.1990.Service managementand marketing. Managing the moment of truth in service competition. Lexington,Mass.:LexingtonBooks. Hayduk,LeslieA. 1987. Structural equation modeling with LISREL: Essentials and advances. Baltimore,Md.:JohnsHopkinsUniversityPress. Hedlund,Gunnar& Arne Kverneland.1985.Are entrystrategiesfor foreignmarketschanging? The case of Swedish investmentsin Japan. International Studies of Management and Organization, 15(2): 41-59.

Hill, CharlesW L., PeterHwang & W Chan Kim. 1990. An eclectictheory of the choice of internationalentrymode.Strategic Management Journal, 11(2):117-28. Hirsch,Seev. 1993.The globalizationof servicesand service-intensivegoods industries.In Y Aharoni, editor, Coalitions and competition: The globalization of professional business services, 66-78. London:Routledge. Hofstede,Geert. 1984a.Culture'sconsequences: International differences in work related values. BeverlyHills,Calif.:Sage. . 1984b. Cultural dimensions in managementand planning. Asia Pacific Journal of Management, January:81-111.

1972.Export och utlandsetableringar H6rnell,Erik,Jan-ErikVahlne& Finn Wiedersheim-Paul. (Exportand foreignestablishments).Uppsala,Sweden:Almqvist& Wicksell. Hymer,StephenH. 1976.The internationaloperations of nationalfirms: A study of directforeign investment.Cambridge,Mass.:MIT Press. Jansson, Hans, Mohamed Saqib & D. Deo Sharma. 1995. The state and transnational corporations:A networkapproach to industrialpolicy in India. London:EdwardElgar. Johanson,Jan & Finn Wiedersheim-Paul.1975. The internationalizationof the firm- Four Swedishcases.Journal of Management Studies, 12(3):305-22. & Jan-ErikVahlne. 1977. The internationalizationprocess of the firm- A model of knowledge development and increasing foreign market commitments. Journal of International Business Studies, 8(1):23-32. . 1990. The mechanismof internationalisation.International Marketing Review, 7(4): 11-24. Jones,Edgar.1981.Accountancy and the British economy 1840-1980: The evolution of Ernst & Whinney.London:BatsfordBooks. J6reskog,Karl-Gustav& Dag Sorbom.1993.LISREL 8. Structural equation modelling with the SIMPLIS commandlanguage. Chicago:ScientificSoftwareInternational. Kilduff, Martin. 1992. Performanceand interactionroutinesin multinationalcorporations. Journal of International Business Studies, 23(1): 133-45. . 1993. The reproductionof inertiain multinationalcorporations.In S. Ghoshal & D.E Westney,editors,Organizationtheory and the multinational corporation,259-74. London:St. Martin's Press.

JOURNAL OF INTERNATIONALBUSINESS STUDIES, SECOND QUARTER 1997

358

Kim, W Chan & PeterHwang. 1992.Global strategyand multinationals'entrymode choice. Journal of International Business Studies, 23(1): 29-53. Kindleberger, Charles P. 1969. American business abroad: Six lectures on direct investment.New

Haven,Conn.:YaleUniversityPress. Kogut,Bruce& HarbirSingh. 1988.The effectof nationalcultureon the choiceof entrymode. Journal of International Business Studies, 19(3): 411-32.

Lenway,StefanieAnn & Thomas P. Murtha. 1994. The state as a strategistin international business studies. Journal of International Business Research, (25)3: 513-35.

Levitt, Theodore.1972. Production-lineapproachto service.HarvardBusinessRewiev,50(5): 41-52.

Lo, ThamisW & Amy Yung. 1988.Multinationalservicefirmsin centrally-planned economies: Foreign advertising in the PRC. Management InternationalReview, 28(1): 26-32.

Lovelock,Christopher.1988.Managingservices.EnglewoodCliffs,N.J.:Prentice-Hall. Luostarinen,Reijo.1980.Internationalization of thefirm.Helsinki,Finland:HelsinkiSchoolof Economics.

Maclayton,Darego,MickeySmith& JosephHair.1980.Determinantsof foreignmarketentry: A multivariate analysis of corporate behaviour. Management International Review, 20(3): 40-52.

Madhok, Anoop. 1996. Know-how-,experience-and competition-relatedconsiderationsin foreign market entry: An exploratory investigation. International Business Review, 5(4):

339-66. . 1997.Cost, valueand foreignmarketentrymode:The transactionand the firm.Strategic Management Journal, 18(1): 39-61.

Makino, Shige & Andrew Delios. 1996. Local knowledge transfer and performance: Implications for alliance formation in Asia. Journal of International Business Studies, 27(5), 905-27.

New York:Wiley. March,JamesG. & HerbertA. Simon. 1958.Organizations. Millington, Andrew I. & Brian T. Bayliss. 1990. The process of internationalization:UK companies in the E.C. Management InternationalReview, 30(2): 151-61.

Murtha,ThomasP. & StefanieAnn Lenway.1994.Countrycapabilitiesand the strategicstate: How national political institutionsaffect multinationalcorporations'strategies.Strategic Management Journal, 15(Summer): 113-29. Nelson, Richard R. & Sidney G. Winter. 1982. An evolutionary theory of economic change.

Cambridge,Mass.:BelknapPress. Newbould, Gerald D., Peter J. Buckley& Jane C. Thurwell.1978. Goinginternational:The experience of smaller companies overseas. New York: John Wiley.

Nordstrom,Kjell. 1990.The internationalization processof the firm:Searchingfornewpatterns and explanations. Ph.D. dissertation. Stockholm: Institute of InternationalBusiness StockholmSchool of Economics. New York:JohnWiley. Normann,Richard.1984.Servicemanagement. O'Grady, Shawna & Henry W Lane. 1996. The psychic distance paradox. Journal of International Business Studies, 27(2): 309-33. Penrose, Edith T. 1959. The theory of the growth of thefirm. Oxford, U.K.: Basil Blackwell.

Polanyi.M. 1962.Personalknowledge.London:Routledge& KeganPaul.

EXPERIENTIALKNOWLEDGE AND COST

359

Quinn,James-Brian& ChristopherE. Gatignon.1986.Will servicefollow manufacturinginto 95-103. decline?,HarvardBusiness Review, November-December: transactioncosts and strategicchoice.International Reid,Stan. 1983.Firminternationalization, Marketing Review, 1(2):44-56. . 1984. Informationacquisition and export entry decisions in small firms. Journal of Business Research, 12: 141-57.

Root, FranklinJ. 1987.Foreign market entry strategies. New York:AMACOM. Rugman,Alan M. 1981.Inside the multinationals: The economics of internalmarkets. New York: ColumbiaUniversityPress. Sagari,S. B. 1992.UnitedStatesforeigndirectinvestmentin the bankingindustry.Transnational Corporations, 1(3):93-123. Sauvant,Karl P. & PadmaMallampally.1993. Transnationalcorporations in services. London: Routledge. Seymour,H. 1986. Internationalinvestmentin the constructionindustry.Ph.D. dissertation. Reading,U.K.: Universityof Reading. Sharma, Deo D. & Jan Johanson. 1987. Technical consultancy in internationalisation. International Marketing Review, 4(Winter):20-29. Shetty,Y Krishna. 1979. Managingthe MNC: Europeanand Americanstyles. Management International Review, 19(3):39-48. Simpson,Claude L., Jr. & Duane Kujawa.1974. The export decision process:An empirical inquiry.Journal of International Business Studies, 5(1): 107-17. Stopford,JohnM. & LouisT. Wells,Jr. 1972.Managing the multinational enterprise.New York: BasicBooks. & Susan Strange. 1991. Rival states, rival firm: Competiton for world market shares. Cambridge,U.K.: CambridgeUniversityPress. Sunzook,F 1978.How Britishindustryexports.Marketing (UK), June:29-34. Terpstra,Vern& Chwo-MingYu. 1988.Determinantsof foreigninvestmentby U.S. advertising agencies.Journal of International Business Studies, 19(1):33-46. United Nations Centreon TransnationalCorporations(UNCTC). 1981. Transnationalbanks: Operations,strategies and their effects on developingcountries. New York:United Nations. Vernon,Raymond.1966.Internationalinvestmentand internationaltradein the productcycle. QuarterlyJournal of Economics, 80(2):190-207. Walter,Ingo. 1985. Barriers to trade in banking and financial services. London:TradePolicy ResearchCentre. Weinstein,Arnold K. 1974. The internationalexpansion of U.S. multinationalsadvertising agencies.MSU Business Topics, 22(Summer):29-35. . 1977.Foreigninvestmentby servicefirms:The case of multinationaladvertisingagencies. Journal of InternationalBusiness Studies, 8(1):83-91.

Wiedersheim-Paul, Finn, Hans C. Olson & LawrenceS. Welch.1978.Pre-exportactivity:The firststep in internationalization. Journal of International Business Studies, 8(1):47-58. Williamson,OliverE. 1975. Markets and hierarchies:Analysis and antitrust implications. New York:The FreePress.

JOURNAL OF INTERNATIONALBUSINESS STUDIES, SECOND QUARTER 1997

360

Winter,SidneyG. 1987.Knowledgeand competenceas strategicassets.In D. Teece,editor,The compettitive challenge. Strategies for industrial innovationand renewal. New York: Harper &

Row. Yoffie,David B. 1988. How an industrybuildspolitical advantage.HarvardBusinessReview, May-June:82-89. Yu, Chwo-Ming J. 1990. The experience effect and foreign direct investment. WeltwirtschaftlichesArchiv, 126: 561-79.

Zeithaml,Valerie,A. Parasuraman & LeonardL. Berry.1985.Problemsand strategiesin service marketing. Journal of Marketing, 49(2): 33-46.