International is at Ion as a Strategy

8/2/2019 International is at Ion as a Strategy

1/21

Internationalization as a Strategy ProcessAuthor(s): Leif MelinSource: Strategic Management Journal, Vol. 13, Special Issue: Fundamental Themes in StrategyProcess Research (Winter, 1992), pp. 99-118Published by: John Wiley & SonsStable URL: http://www.jstor.org/stable/2486368 .

Accessed: 21/01/2011 10:11

Your use of the JSTOR archive indicates your acceptance of JSTOR's Terms and Conditions of Use, available at .http://www.jstor.org/page/info/about/policies/terms.jsp. JSTOR's Terms and Conditions of Use provides, in part, that unless

you have obtained prior permission, you may not download an entire issue of a journal or multiple copies of articles, and you

may use content in the JSTOR archive only for your personal, non-commercial use.

Please contact the publisher regarding any further use of this work. Publisher contact information may be obtained at .http://www.jstor.org/action/showPublisher?publisherCode=jwiley. .

Each copy of any part of a JSTOR transmission must contain the same copyright notice that appears on the screen or printed

page of such transmission.

JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range of

content in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new forms

of scholarship. For more information about JSTOR, please contact [email protected].

John Wiley & Sons is collaborating with JSTOR to digitize, preserve and extend access to Strategic

Management Journal.

http://www.jstor.org
http://www.jstor.org/action/showPublisher?publisherCode=jwileyhttp://www.jstor.org/stable/2486368?origin=JSTOR-pdfhttp://www.jstor.org/page/info/about/policies/terms.jsphttp://www.jstor.org/action/showPublisher?publisherCode=jwileyhttp://www.jstor.org/action/showPublisher?publisherCode=jwileyhttp://www.jstor.org/page/info/about/policies/terms.jsphttp://www.jstor.org/stable/2486368?origin=JSTOR-pdfhttp://www.jstor.org/action/showPublisher?publisherCode=jwiley


2/21

StrategicManagement Journal, Vol. 13, 99-118 (1992)

INTERNATIONALIZATIONS A STRATEGYPROCESSLEIFMELINDepartment of Management and Economics, Linkoping University, Linkoping,Sweden

This paper critically reviews the field of international business research. The field ischaracterized by considerable intellectual diversity, where theoreticalfocus is blurred by themultidisciplinary nature of the field. The review focuses on three themes that help shedlight on internationalization as a strategy process; stage models of internationalization,studies of the link between strategy and structure in MNCs, and studies of administrativeprocesses in MNCs and recent organizational models for MNCs. Sequential stages modelsare too deterministic and stress only early stages of internationalization. Conceptualcontributions from research on structuresfollowing strategies have a very static character.Research on management processes in MNCs have a questionable empirical base andnormative bent. Three key themes for future research on internationalization as a processare suggested. These themes, dealing with major omissions and weaknesses identified in thereview, are: the study of acquisition processes and internationalization, the study of dynamicprocesses in MNCs, and the study of internationalizationprocesses in their outer contexts.

INTRODUCTIONThe corpus of international business researchproduced over the last three decades reveals afield characterized by considerable intellectualdiversity drawing on a wide span of disciplines.The purpose of the paper is to review that corpusand explore the meaning of internationalizationas a process. Key themes for future research oninternationalization as a strategy process aresuggested. Our review is based on an extensivereading of works categorized as internationalbusiness research. The review task includesan ambition to discover and analyze certainunderlying assumptions concerning theoreticaland methodological issues.A brief characterization of the broad field ofKey words: Internationalization,strategy process,multinational orporation,stage models, global inte-gration, local responsiveness, acquisition, differen-tiation, longitudinalapproach0143-2095/92/100099-20$15.00(? 1992by John Wiley & Sons, Ltd.

international business will help to clarify thechoice of subfields and themes in this review.As Toyne (1989) stated, there seems to be alack of consensus about the conceptual domainof intemational business. Theoretical focus isblurred by the multidisciplinary nature of thefield. Compare, for instance the content ofarticles published in a single volume of theJournal of International Business Studies (JIBS,1991): analysis of information content in multi-national advertisements; theories underlyingJapanese lifetime employment; measurement ofperformance in international joint ventures;nontariff barriers for foreign investments; andfactors related to expatriate and spouse repatri-ation adjustment. This not only demonstratesdiversity but indicates a natural lack of a commontheoretical base. To more systematically illustratethe broad scope of research on internationalbusiness, we have categorized all articles pub-lished in the last three full volumes of JIBS(1989-91). Here are the seven most frequent


3/21

100 L. Melinareas of international business published in thethree volumes of JIBS (with a total of 76 articles):-Finance and banking, (9 articles)-Cross-cultural aspects, (8 articles)-International joint ventures, (7 articles)-Human resources (including expatriates), (6articles)-Foreign direct investments (5 articles)-Coordination and control in MNCs (5 articles)-Host government relationships (3 articles)

These seven areas represent alternative theor-etical frameworks of international businessresearch, each with fundamental differences. Inaddition, few empirical studies reported in thesevolumes were founded on longitudinal studiesand/or dynamic theory, i.e., the process dimen-sion of internationalization was weakly rep-resented. Most articles were based on cross-sectional methods and static models. Internationalbusiness research as a whole does not have thepreponderance of U.S. researchers found inmany other subfields of management research.This may help to explain the greater methodolog-ical diversity to be found in the field ofinternational business.The field of international business has its rootsin international economics, including a numberof theories on international trade, such as thecomparative advantage of nations. However,from the 1960s onward we find other researchthemes. In Figure 1, most major key themes in

m~~oundaries Withinoganizations Betweenor tionsActivities - across bordes - acrossbordersA B

Entrymodes * Intemationtion stage models*Internalization; C *Internationalrade DTransactions ftrasaction cost*snterational marketng*Intraorganizationalrade; *Foreigndirect nvestnentstansfer prices .interational businessnetworks* Transferof tichnology

* GlobalstrategiesCoordination eStructual forms *Alliances andmechanisms OFormal ontrolsystems joint ventures*Informal/social ntegrationand *Host govemmentcoordination relationship

Figure 1. The field of international business-activities across borders

international business research are cataloged(exceptions include the international finance andbanking theme). These key themes are structuredalong two lines: the type of activity in focus andwhether the activities are intraorganizational orinterorganizational. Figure 1 illustrates that theresearch on international business deals withorganizational activities across borders. A largeproportion of the work focuses on the intraorgan-izational activities of large firms, with the obviousintent to explain and characterize the structuralform of the typical multinational corporation(MNC). In fact, research on strategy, structureand administrative processes in MNCs formsthe basis for a named subfield, internationalmanagement (IM).

Since a brief paper cannot fairly reviewthe whole field of international business andmanagement, we will limit our discussion tothree themes (boxes A, B and E in Figure 1)that help shed light on internationalization as astrategy process:-stage models of internationalization-studies of the link between strategy andstructure in MNCs-studies of administrative processes in MNCsand recent organizational models for MNCs.

We restrict ourselves to research themes thateither address the process dimension of theinternational enterprise's formation and continu-ance (Toyne, 1989) or are kindred to this


4/21

Internationalization as a Strategy Process 101dimension. Consequently, themes rooted in eco-nomics (such as the eclectic theory with theinternalization theory, Dunning, 1981; Buckley,1988) are excluded because of their static viewof the multinational enterprise. Content-orientedresearch on international strategy is ruled outfor the same reason. Our focus on the firmlevel automatically excludes such themes as thecompetitiveness of nations (Porter, 1990).The paper has five sections. In the first sectionthe significance and certain methodological conse-quences of internationalization as a strategyprocess is discussed. The three following sectionscontain a review of major contributions and theunderlying assumptions and weaknesses of thethree selected themes of internationalization:stage models, strategy-structure relationships,and management processes in MNCs. The con-cluding section suggests three key themes andquestions for future research.

INTERNATIONALIZATION AS ASTRATEGY PROCESSStrategy making is about changing perspectivesand/or positions (Mintzberg, 1987).Internationalization-the process of increasinginvolvement in international operations acrossborders (Welch and Luostarinen, 1988)-com-prise both changed perspectives and changedpositions. Thus internationalization is a majordimension of the ongoing strategy process ofmost business firms. The strategy process deter-mines the ongoing development and change inthe international firm in terms of scope, businessidea, action orientation, organizing principles,nature of managerial work, dominating valuesand converging norms. The internationalizationdimension is related to all these aspects of thestrategy process. As is shown in this paper,structural theory on MNCs has advanced muchfurther than dynamic theory on internationali-zation as a strategy process. In a survey ofacademic researchers in strategic management,Lyles (1990) argued that the internationalizationtheme regarding global competition was viewedas the coming decade's most important area ofstrategic management research (see also Bettis,1991). This notwithstanding, the research instrategic management currently pays little atten-tion, says Lyles, to internationalization. Further-

more, the strategic management field is stilldominated by cross-sectional research that 'pro-ceeds from a distance, with a remote researchergathering data from organizations he knowsalmost nothing about' (Miller and Friesen, 1982:1014). At the same time, there seems to beincreasing consensus among researchers thatlongitudinal research would enable a betterunderstanding of organizations. According toHuff and Reger (1987: 227) ... there is dangerin believing that statistically rigorous, narrowlyfocused studies are superior to the rich, compli-cated understanding that results from carefulunderstanding of a few organizations.In sum, there remains a considerable need forresearch that is responsive to the longitudinalcharacterof internationalization as a developmentprocess through time (Welch and Luostarinen,1988). However, different meanings of develop-ment processes can be identified, depending onthe methodological approach used to revealthe process. In fact, the longitudinal approachincorporates at least four different approaches(graphically illustrated in Figure 2):-In type A, process is a time-series of detachedcritical events, or states, e.g., structural oreconomic. Most management studies that includethe time dimension in their explanatory modelsseem to use this approach, analyzing situationsdisjointed in time. Typical are the large number ofstudies of the correspondence between strategy,structure, and performance. Recent illustrationsinclude a study by Gomez-Mejia (1992) whorelated corporate performance during a 5-yearperiod to changed strategies, and a study byHabib and Victor (1991) where the fit betweenstrategy and structure in MNCs was related toperformance of the two following years. We seethat the process dimension is weakly developedin this type of longitudinal approach.-In type B, process is relatively short episodes.The approach here may be to study a singleepisode such as an acquisition from the preacqui-sition phase to the postacquisition phase(Haspeslagh and Jemison, 1991), or to studyseveral episodes in sequence or in parallel. Thetime period for an episode may vary from a fewweeks to a few years.-In type C, process is lengthy epochs. An epochmay be the strategic development of a companyunder an influentialCEO. A series of epochs maybe long periods of evolutionary change disrupted


5/21

102 L. Melin

Event A Event B Evet C

Type A: rune senes of events IIP, I ,1970 1990

Type B: Relatvely short episodes 11970 1990

Type C: Longerpochs1970 1990

TypeD: Birphic htory history1930 1990

Figure2. Four types of process capturedby different ongitudinalapproaches

by shorter episodes of revolutionary change. Anepoch may be from 2 to 20 years. Illustrativeexamples of epoch-oriented studies are found inPettigrew (1985a) and Pettigrewand Whipp (1991).-In type D, process is seen as biographic history.Here the biography of a firm captures the wholedevelopment from the time of its founding tothe present time, (see for example the classiccase studies of Chandler, 1962). The time periodmay vary considerably but for large MNCs itwill extend over several decades.We use this process typology to characterizethe process studies covered by our review andto determine the types of process studies thatmight be desirable in future research on inter-nationalization. A broad reading of empiricalstudies in the field indicates that longitudinalstudies of internationalization processes rep-resenting types C and D are very infrequent.Models and methods in the IM field are overlystatic. This is a weakness since the IM field, likethe (overlapping) field of strategic-managementfocuses 'on managerial problems, for whichtiming, sequence and change are key elements'(Hambrick, 1990: 244).Descriptions of internationalization processes,as of other strategy processes, include informationabout change. The degree of change may differfrom a state of more or less status quo to radicaltransformation. Between these extremes, we findthree kinds of change: expanded reproduction,incremental change and evolutionary transition(Wilson, 1992). Change must always be under-

stood within a process context. Change concernsthe dynamics in development over time(Pettigrew, 1985b). Change is implicitly relatedto content, in other words, some particular thingbecomes different. When we study longitudinalprocesses such as the internationalization processof a firm, we must also focus on content. Thedichotomy of process vs. content in strategyresearch has been misleading and should beavoided in international management research.

INTERNATIONALIZATION AS ASEQUENCE OF STAGESSignificant models in the field of internationalbusiness describe the internationalization processas a gradual development taking place in distinctstages and over a relatively long period of time.Two often cited models will be discussed here,the product (life) cycle model (Vernon 1966) andthe (Uppsala) internationalization process model(Johanson and Vahlne, 1977). Both models stillexert considerable influence on the field and theinternationalization process model in particularattracts many followers. Their longitudinalcharacter satisfies one criterion of internationaliz-ation as a process. Their shortcomings includecertain problems inherent to stage models labelledby Stubbart (1992) as the 'deceptive allure ofdevelopmental models'.The product cycle model developed by Vernon(1966) is an attempt to bridge a country-based


6/21

Internationalization as a Strategy Process 103perspective of international trade theory with anindividual firm's perspective of internationalinvestment theory. Vernon wished to remedya lack of realism found in the dominatingcomparative-costs-advantage theory by emphasiz-ing the role of product innovation, the effects ofscale economies and the role of uncertainty ininfluencing trade patterns across national borders(Vernon 1966: 190). He identified several stagesin the life cycle of a product, each of which hasdifferent implications for the internationalizationof the innovative company and the product itself.The introduction stage is domestic, having itsorientation in the country where the product wasdeveloped. Export to other industrial countriesmay support the emerging goal of achievingeconomies of scale in production. In the growthstage, export activities increase and foreign directinvestments in manufacturing plants are made incountries with an expanding demand for theproduct. In the maturity stage, when majormarkets are saturated and the product stand-ardized, the manufacture is relocated to countrieswith low labor costs. Finally, in the stage ofdecline, manufacture, and in some cases also thedemand, definitely leaves the industrial countrywhich was home to the original innovation(Vernon, 1966; McKiernan, 1992).Although the product cycle model takes thecompany level into account, it has its main focuson the country level. Its main contribution isthe developmental view on relocalization ofproduction activities; such change is related tovarious national characteristics such as technologi-cal know-how, demand and labor costs. A generalconclusion is that increasing product maturitymakes it less critical to have a short distancebetween the production facilities and the corpor-ate centers for decision-making and productdevelopment. However, the descriptive value ofthis stage model is weak for products with shortlife cycles, a circumstance which applies to moreand more products (McKiernan, 1992). Theapplicability of this model is also limited if newproducts are developed in companies that alreadyhave considerable operations in foreign countries(see Vernon, 1979). Furthermore, as we will seebelow, new localization patterns emerge in MNCsas new organizational forms are introduced,implying a new way of thinking about whereto establish product development centers andmanufacturing units.

The second stage model of importance is theinternationalization process model (Johanson andVahlne, 1977). This process, whereby a firmgradually increases its international involvement,is described as being sequential from the initialexport activities to the setting up of foreignproduction units (Johanson and Wiedersheim-Paul, 1975). Each firm goes through a numberof logical steps of international behavior, basedon 'its gradual acquisition, integration and use ofknowledge about foreign markets and operations,and on its successively increasing commitment toforeign markets' (Johanson and Vahlne, 1977).The focus is on market knowledge and marketcommitment (through engaged resources). Thelearning through development of experientialknowledge about foreign markets is necessary inorder to overcome the 'psychic distance' to thesemarkets, i.e., differences between any twocountries in terms of language, culture, educationlevel, business practice and legislation. Accordingto Johanson and Vahlne (1990), the firm entersnew markets with successively greater psychicdistance. This perceived distance is expected todisturb the flow of information between the firmand the foreign market. Therefore firms starttheir internationalization on markets with thelowest perceived market uncertainty, in otherwords, markets that they can rather easilyunderstand, often in neighboring countries.Similar models which build on the behavioraltheory of the firm (Cyert and March, 1963, laterapplied to international investments by Aharoni,1966) have been introduced by Bilkey (1978)and Cavusgil (1980). The internationalizationprocess model contributes an alternative andsignificant view on internationalization as com-pared to the dominating theory on foreigninvestments, the 'eclectic paradigm' (Dunning,1980, 1988). The eclectic paradigm argues thatthree distinct sets of advantages explain theinvolvement of firms in foreign production; theseare ownership-specific advantages, internalizationadvantages (further emphasized by Buckley, 1988and Rugman, 1980), and localization advantages.The eclectic paradigm is based on economictheory and has transaction costs and factor costsas its main explanatory variables together withthe assumption of rational decision-making ininternational firms which make foreign directinvestments. However it explains the existenceof the multinational enterprise rather than the


7/21

104 L. Melinprocess of internationalization even if theambition is to make the internalization theorymore dynamic and more behavioral in itsorientation (Buckley, 1990; 1991). The inter-nationalization process model on the otherhand explains internationalization as one distinctpattern of growth. It is based on behavioraltheories, with assumptions about lack of infor-mation and the importance of perceived risk anduncertainty (Cavusgil, 1980). Within this view,the internationalization process is not seen as asequence of deliberately planned steps foundedon rational analysis. Instead the incrementalnature of successive learning through stages ofincreasing commitment to foreign markets is themain characteristic of the internationalizationprocess, which according to Johanson and Vahlne(1990) will proceed along the presented stagesregardless of whether strategic decisions in thisdirection are made or not. This process modelmatches two of the four types of processes inFigure 2, the episodical and the epochal, sincethe internationalization stage can be seen as asequential chain of episodes or epochs.The internationalization process model has anumber of shortcomings which have beenaccepted by Johanson and Vahlne (1990: it istoo deterministic; its significance is limited tothe early stages of internationalization; and asthe world becomes more homogeneous, theexplanatory value of psychic distance tends todecrease. The deterministic, sequential nature ofthis process model excludes other options ofstrategic choices, e.g., to initiate local productionin a foreign country without having gone throughthe steps of exporting or having local salessubsidiaries. Leapfrogging of intermediate stagesis in fact quite common (Hedlund and Kverne-land, 1984; Bjorkman, 1989; McKiernan, 1992).The internationalization model is based onempirical research about Swedish firms, but hasgained strong support in several other studies(Bilkey, 1978, Cavusgil, 1980; Denis and Depel-teau, 1985; Johanson and Nonaka, 1983), showingthat this model's validity is not limited to smallcountries which are highly dependent on export.However, most empirical support comes fromstudies on the early stages of internationalizationrepresented by the model. The model tells us littleabout the internationalizationprocess taking placein experienced companies which have learnedthroughdecades of international activities. Further-

more the internationalization process model doesnot pay enough attention to the acquisition choiceas a route to internationalization, a shortcomingshared with the mainstream of foreign directinvestment theory (Forsgren, 1990).In more general terms, both stage modelsdiscussed in this section suffer from distinctlimitations. The focus of stage models is on thenature, sequence and order of activities (Van deVen, 1992). According to Van de Ven (p. 172)these stage models represent a unitary pro-gression, which means a sequence of the formU -* V -- W where U, V and W representqualitatively different stages that must occur inan ordered progression. Each stage of develop-ment is seen as 'a necessary precursor of

succeeding stages' (Van de Ven, 1992: 177).The implication is that stage models describedevelopmental history as the result of predeter-mined factors and preprogrammed forces. Butthe consequences of unforeseen environmentalinteractions are hardly predetermined (Stubbart,1992). Furthermore, stage models downplay thepossibility for managers to make voluntarystrategic choices. According to Stubbart (1992),another problem related to stage models is theirdisregard of individual differences. Variationsand differences are not made a part of thepattern of sequential stages.

STRATEGY, STRUCTURE ANDCONTROL IN THE MULTINATIONALCORPORATIONDuring the last two decades a large share of theresearch in the field of international managementhas been devoted to structural aspects of MNCs.During the 1970s, interest mainly focused onstructural forms and formal control mechanisms,but during the 1980s emphasis shifted towardless formal ways of coordinating the MNCs. Inan exhaustive literature survey, Martinez andJarillo (1989) identify 82 empirical studies pub-lished between 1964 and 1988 on coordinationmechanisms in MNCs. A notable share of these-about 25 percent-have their roots in twobusiness schools, the Harvard Business Schooland the Stockholm School of Economics. Earliercontributions from two distinguished Harvardscholars, Alfred Chandler and Joe Bower, havebeen especially influential in this part of the IM


8/21

Internationalization as a Strategy Process 105field. In fact, the influence of these two scholarshas generated two rather distinct schools ofthought. The text below is organized in line withthis evolution. The first part deals with studiesfollowing Chandler's (1962) work on strategyand structure, and the second part reviewsthe so-called process school of internationalmanagement, based on Bower's (1970) work onstrategy process.The structure of MNCs follows strategy(research following Chandler)The origin of this stream of research is made clearin the first paragraphof an AdministrativeScienceQuarterly article by Fouraker and Stopford:

One of the landmark studies in the field ofbusinessadministrations Strategy nd Structureby A.D. Chandler, Jr. A central propositionin Chandler's book is that the strategy ofdiversification ed to organizationalproblemsand eventually to the emergence of a newcorporate structure... International businessactivity is a form of diversification. ... In somesense, this development may be consideredareplicationagainstwhich Chandler's hesis maybe tested.... the new diversification hould leadto new problemsof organizationand, finally,todifferenttructuralccommodations1968:47-48).

The early research on internationalization strat-egy and organizational structure starts with theassumption that structure follows strategy. Notsurprisingly, this assumption is given strongsupport in the first major study, where Stopfordand Wells found that managers in multinationalenterprises 'following similar strategies in quitedifferent industries have developed similar organ-izational structures (1972: 5). Stopford and Wellsstudied epochs of internationalization (i.e., pro-cess type B in Figure 2) in order to identify thestructural state of 187 large U.S. manufacturingfirms, all internationally dispersed and eachhaving manufacturing facilities in at least sixforeign countries. An initial phase of internationalstructure with relatively autonomous subsidiarieswas followed by a second phase in whichthe MNCs organized an international businessdivision intended to increase control and coordi-nation of the expanding international activities.This division covered all foreign activities of thefirm and became its locus of internationalexpertise (Stopford and Wells, 1972: 21 and

Aman, forthcoming). The international divisiongrew and as the ratio of foreign sales tototal sales continued to increase, top managersrecognized that the enterprise needed a globalperspective. The presence of an autonomousinternational division actually served to constrainthe integration process since only the generalmanager of the international division hadresponsibility that transcended national bound-aries. The predominant international division wassuperseded by two new types of global structures:worldwide product division and geographic areadivision for all products. Worldwide productdivision was the choice of firms with high productdiversity, in which each domestic product divisionwas assigned responsibility for the worldwideactivities of its particular line of products. Areadivisions emerged in firms with greater geographicdiversity. Each such area division was maderesponsible for one geographic region of theworld market. In some cases a combination ofthese two types of global structures was found,where some product lines were managed on aworldwide basis and others through area divisions(Stopford and Wells, 1972: 26-53). However, inthe beginning of the 1980s, the internationaldivision structure still seemed to be the mostfrequent in MNCs based in the United States(Daniels, Pitts and Tretter, 1985).In a parallel study, Franko (1976) comparedthe 85 largest industrial firms of the western partof continental Europe, focusing on (in histerminology) their international networks. Theorganizational structure of 70 of these firms wasanalyzed in greater detail. A biographic approachwas used to identify epochs of dominatingstructural forms (see Figure 2). In contrast tomultinational enterprises based in the UnitedStates, 26 of the European MNCs retained theparent-subsidiary form, with direct ties betweenthe corporate head office and each autonomoussubsidiary. The route to global structure differedfrom that found in the U.S. firms studied byStopford and Wells. Most European multination-als did not choose the international division asan intermediate form, but went directly fromthe parent-subsidiary structure to a worldwideproduct structure (24 enterprises did so). As 84percent of the European enterprises in Franko'sstudy were diversified outside the home country,one would expect that an even greater numberof product division structures would have been


9/21

106 L. Melinintroduced. Franko's conclusion is that 'in conti-nental Europe, structure did not follow strategyuntil there was a (new) change in the competitiveenvironment' (Franko, 1976: 204). ScandinavianMNCs show a similar pattern to the Europeanfirms studied by Franco. The Scandinavian firmsretained the parent-subsidiary structure until the1970s at which time most big firms went directlyto the global product division structure withoutever having introduced the international division(Hedlund, 1984; Hedlund and Aman, 1984;Aman, forthcoming).Also Stopford and Wells (1972) softened theirconclusion on the strategy-structure fit despitetheir empirical evidence. They emphasized thatthe relationship between strategy and organi-zation 'is clearly not the simple one in whichstrategy is determined first and then a structureestablished to implement it' (p. 6). However, forStopford and Wells, strategy -_ structure wasstill the natural sequence, a statement laterquestioned by several authors (e.g., Hall andSaias, 1980; Peters, 1984; and Melin, 1989).The studies of Stopford and Wells (1972) andFranko(1976) representa landmark n internationalmanagement research. These studies give a valuableunderstanding of the historic development ofinternational strategies and structures for multin-ational enterprises. However, their major findingshave to do with which strategies and structuresactually emerged. Their contribution is far moremodest when they attempt to deal with whystrategies and structuresare changed. The processdimensionof how they changedis almostcompletelyabsent. Although these studies take a longitudinalapproach, they can hardly claim to capture theessence of internationalizationas a process. Ratherthey present steady states of structuralforms andfail to describe the processes of formation andimplementation related to these structures. Inaddition, they emphasize mainly formal aspects ofstructure.Notably, both studies contain early obser-vations about a new structural form for theMNC, the global matrix. Stopford and Wells(1972) called this the 'grid structure' whichapparently opened their eyes for a necessaryshift from the formal view on structure to astronger emphasis on informal and fluid aspectsof structure. They made several significantobservations about the grid structure: Boundariesbetween divisions and departments become

increasingly blurred in a global grid structure.Emphasis shifts from hierarchical to contractualmanagement. Grid structures can function assystems of creative conflicts. In a grid structure,'the creation among the managers of a sense ofcooperation and shared values is one way ofcombating the control problem' (op. cit., p. 173).Unfortunately, these significant and quite novelobservations did not stake a direction to sub-sequent research in international managementduring the 1970s. Another important discussionoften ignored by (U.S.) researchers in bothstrategic management and international manage-ment is the skepticism expressed by Stopfordand Wells about the possibility of drawinglinear conclusions on the relationship betweenstrategy-structure and performance. Theseauthors conclude that it is very difficult to makeany satisfactory analysis that relates structure toperformance. The reason given is that 'perform-ance is an amalgam of many factors..., even ifone could decide how best to measure perform-ance, one could never be sure that the resultswere influenced by the choice of structure'(Stopford and Wells, 1972: 79). From theirnoteworthy findings that some multinationalenterprises with mismatched strategies and struc-tures grew faster and showed a greater foreignprofit than others with matched strategies andstructures, Stopford and Wells allege that thesuccess of an enterprise relies more on managerialskills and abilities than on the structural formwithin which these managers work.Several IM researchers later studied the strat-egy and structure relationship in MNCs, amongothers Davidson and Haspeslagh (1982) andEgelhoff (1988). Furthermore, Bjorkman (1990)put forth an alternative explanation for changedstructures in MNCs. He regards the changedstructure as being the result of mimetic behaviorrather than of the influence exerted by acompany's individual strategy. A new structuralform becomes the fashion, or norm, and isimitated by several followers (DiMaggio andPowell, 1983).Research on informal coordination mechanismsin MNCsIn addition to the focus on structure of theMNC, there are two other coordination dimen-sions which received attention from several IM


10/21

Internationalization as a Strategy Process 107researchers (Martinez and Jarillo, 1989). Onedimension concerns the degree of centralizationof decision-making and bureaucratic controlmechanisms (e.g., Brooke and Remmers, 1970and Schollhammer, 1971). The other dimensionis the less formal and more elusive mechanismsof control and coordination. Wiechmann (1974)found, that formal and informal integrativedevices were closely interrelated in multinationalenterprises. Notably, one of the devices wascorporate acculturation. This relates to findingsregarding the Japanese type of MNC ('type Z')that maintained control through a process ofacculturation and socialization of employees(Ouchi and Johnson, 1978). In subsequentstudies, the cultural control mechanisms weregiven extended meaning by Jaeger and Baliga(1985). In the same vein, Edstrom and Galbraith(1977) presented arguments for transfer ofmanagers in multinational organizations as adeliberate means to develop control through asocializing process and the creation of a verbalinformation network. The objective seemed tobe to create a somewhat uniform corporateculture that transcended national boundaries(see also Jaeger, 1983), today a questionableproposition. That notwithstanding, Edstrom andGalbraith (1977: 248) wanted their article to beunderstood as a call for further research.However, the internationalization theme onwhich they focused (as did Ouchi and others) hasnot yet become a major stream in internationalmanagement research, even if the 'processschool' (see below) shows interest in informalmanagement systems. Martinez and Jarillo(1989) suggest that the informal and socialmode of control has emerged in MNCs as anecessary result of the increased integrationand dispersion of these firms' operations toseveral units in many countries. However, arecent study of Swedish MNCs (Rolander,Zander, and Hedlund, 1989) draws a differentpicture. This study claims that these firmsinstead introduce more formal mechanisms ofcontrol. One explanation of this divergencymay be that Swedish MNCs made early use ofsocial control mechanisms and only recentlyhave adjusted to the combined mode of formaland informal control. If so, these differentpictures instead illustrate a converging move-ment regarding control mechanisms throughoutthe entire MNC population.

THE 'PROCESS SCHOOL' OF MNCsThe emergence of a process perspective in policyresearch (Bower, 1970) has resulted in a largebody of empirical research dealing with mana-gerial issues of MNCs. The doctoral dissertationsof Prahalad, Doz and Bartlett (all at HarvardBusiness School between 1975 and 1979) seemto have laid the foundation for a visible streamof research in international management.Researchers in this stream are referred to as'belonging to the process school' of the diversifiedmultinational corporation (DMNC), (Bartlett andGhoshal, 1991; Doz and Prahalad, 1991). Todaythis school holds paradigmatic qualities, accordingto Doz and Prahalad (1991: 145): 'The develop-ment of a "process school" of research on theDMNC over the past 15 years has led to theemergence of a new paradigm.'This new so-called paradigm represents knowl-edge on DMNC management, which emphasizesthe global integration/local responsiveness frame-work. The basic unit of analysis is considered tobe the individual manager. Most scholars withinthis paradigm, it is claimed, have 'put managerialrelevance before theoretical elegance,' 'haveunderexploited the theories available to them'and 'often been engrossed in the complexity ofwhat they studied, and failed to develop, orborrow, a sufficiently powerful conceptual frame-work' (op. cit., p. 161). According to Doz andPrahalad, the result has been a missing linkbetween the DMNC phenomenon and availableorganization theory. However, this is notregarded as being a problem since the emergingparadigm constitutes a theory on its own:

Taken together the work of Prahalad, Doz,Bartlett, Ghoshal,Hedlund, Hamel, and othersfollowing the same researchapproachprovidesus witha richorganizationheoryof the DMNCand with detailed understanding f managerialtasks of DMNCs (Doz andPrahalad,1991:158).

According to Doz and Prahalad, this organi-zation theory contains mid-range constructs andprovides a set of such integrative constructs forobservation, analysis and understanding, as wellas for normative assessment of DMNC managers.The main conceptual contributions from this newparadigm, are briefly presented below, but itmust first be emphasized that it is quite unusual


11/21

108 L. Melinfor a rather small group of researchers to appointthemselves as constituting a new paradigm. Doesnot a new paradigm emerge in a rather lessdeliberate way, as a result of an institutionali-zation process drawing on ideas and debatesfrom several related fields?Doz and Prahalad (1984) want to supplementthe multinational management literature, whichin their opinion is too 'architectural' in its searchfor the right structure. Their own contributionbuilds on the identification of the constantbalancing of MNC managers between (a) theeconomic imperative, i.e., the impact of globalcompetition that pressures the MNC to transcendthe boundaries of national markets, develop aglobal strategy and rationalize global operationsthrough central control and coordination and (b)the political imperative, i.e., the adjustmentsmade necessary by host government demands,which work to give greater autonomy to thelocal subsidiary, and through diversity amongnational markets, regarding customer needs,distribution channels etc., that call for increasedlocal responsiveness (Doz, 1980; Doz, Bartlett,and Prahalad, 1981; Prahalad and Doz, 1981a;Doz and Prahalad, 1984). No specific structureprovides the solution to this dilemma. Directsubstantive control is not possible. The alternativefor top management is to create an organizationalcontext as a more subtle strategic control mechan-ism. The design of this organizational contexthas been in focus in most of the work presentedjointly by Doz and Prahalad (1981, 1984, 1987and Prahalad and Doz 1981a, 1981b and 1987).Their framework for strategic control in multi-national companies can be depicted as a matrixof (a) subprocesses of change and (b) a collectionof management tools. To be more precise, thesubprocesses of change has three dimensions:cognitive perspective, strategic priorities andpower allocation. The management tools are ofthree kinds. Data management tools providedata for critical decisions. Managers' managementtools express the rules of the game that shapeexecutive perception and expectations. Conflictresolution tools are used to create decisionstructures for careful trade-off between prioritiesof global integration and national responsiveness.Alongside this framework, Doz argues thatthe diversified multinational company shouldstrive for multifocal strategies. This means that'responsiveness and integration needs are

weighted one against the other separately foreach decision, with no a priori assumption ofdominance of one over the other' (1986: 214).The multifocal MNC will trade-off the costs andbenefits regarding national responsiveness andmultinational integration in a flexible manner.The key organizational capability of the multifocalstrategy firm is the capability to 'shift the locusand logic of decision from a national concern toa global view, and vice versa, from decision todecision' (1986: 214). Doz believes that a matrixstructure that assigns equal power to executiveswith differing responsibilities would lead tocorporate paralysis. Instead, conflicting viewsshould be confronted, which leaves the relativepower of various executives unspecified. Theideal situation for Doz is constant tensionbetween defenders of national responsivenessand managers supporting multinational inte-gration, and this is something quite apart frombalance and power symmetry.Bartlett and Ghoshal (1987, 1989) focus theirinterest on how large crossborder enterprisesorganize their activities. The studies of theseauthors emphasize the functioning and capabilitiesof the MNC, not its structural form. Importantelements are informal mechanisms of coordi-nation, including normative values, and modesand patterns of internal communication. Bartlettand Ghoshal's work represents a surprisinglystrong belief in the environment-strategy-struc-ture paradigm which they claim 'provides apowerful way to understand differences in corpor-ate performance' (1987: 53). They mean, how-ever, that as the global industrial world becomesmore complex, the unidimensional concepts ofstrategic fit should be replaced by a more dynamicview. The main driving forces for changes inMNCs, according to Bartlett and Ghoshal, areenvironmental. But Bartlett (1986) also callsattention to the administrative heritage, i.e., thehistory and background of the firm. The basicassumption in the research of Bartlett and Ghoshalconcerning large MNCs is that environmentalforces shape the strategic profile of a business,while a company's administrative heritage moldsits organizationalform and capabilities.In an empirical project, Bartlett and Ghoshalfound three clearly distinguishable organizationalmodels among nine multinational companies(three each from Japan, Europe and the U.S.).Each model is 'characterized by distinct structural


12/21

Internationalization as a Strategy Process 109configurations, administrative processes and man-agement mentalities' (1989: 49). The importanceof a company's administrative history is evidentin these organization models, as each of them isrelated to specific dependence on their respectiveEuropean, Japanese or U.S. cultural background.The multinational organization model fits themanagement norms of many European companiesthat grew internationally during the prewarperiod. Bartlett and Ghoshal label this model 'adecentralized federation of assets and responsibili-ties' (1989: 49) which allow foreign units torespond to local differences. The internationalorganization model-which suited the manage-ment norms of many U.S. firms, especially duringthe 1950s and 1960s-is a coordinated federation,where the parent company transfers knowledgeand expertise to foreign markets which layopen to exploitation. The subsidiaries are moredependent on the parent company, compared withthe multinational model, while the internationalmodel requires more formal systems for controlof subsidiaries. The global organization modelfits the managerial norms of many Japanesefirms, which reached global competitive positionsduring the 1970s and early 1980s. It is charac-terized as a centralized hub, where most assetsand decisions are centralized. Foreign operationsare used as delivery pipelines to what is regardedas a unified global market. The subsidiaries arehighly dependent on the parent company andare tightly controlled. This empirically foundedtypology gives an importantunderstandingof threedistinct forms of MNCs and supplements the old,but still valid, conceptions of the global enterprisepresented by Perlmutter (1969); the ethnocentric,the polycentric, and the geocentric types. However,the framework by Bartlett and Ghoshal lacksprocess dimensions, such as the dynamic transitionfrom one form to another, and characteristics ofthe strategy process within each model.According to Bartlett and Ghoshal all threeMNC models have problems in today's complexglobal environments, because they cannot meetthe demand to simultaneously achieve nationalresponsiveness, global efficient integration, andan ability to develop and transfer knowledgeworldwide. Bartlett and Ghoshal have constructeda solution to this demand, the transnationalorganization model, which is not derived fromempirical evidence as are the other three models.Instead the transnational solution-the configur-

ation of tomorrow's competitive MNC-is a morespeculative suggestion impelled by the studies ofthese two scholars. They found that several of thenine MNCs in their study increasingly convergedto the structuralform of a differentiated network(Ghoshal and Bartlett, 1990). In the transnationalcompany, efficiency is a means to achieve globalcompetitiveness, while local responsiveness is ameans to achieve increased flexibility, and com-pany-wide organizational learning is a means todevelop innovations vital to corporate survival.The transnational firm should function as anintegratednetwork. The subsidiaries should havedifferentiatedand specialized roles and they shouldparticipate n a worldwidesharingand developmentof knowledge. This type of organization is not easyto manage. The threatof internal fragmentationanddissipationis obvious, because of the strong degreeof dispersion. Furthermore, the interdependencemay counteract the need for flexibility, since thecomplexity can obstruct the necessary learningcapability. However, according to Bartlett andGhoshal, these problems can be resolved by topmanagement if they succeed in legitimizing diverseperspectives, developing multiple coordination andinnovation processes, and building shared visionand individual commitment.

Hedlund (1986) has developed a similar,speculative model for the modern multinationalcompany, the heterarchy. He sees the unidimen-sionality within the traditional structure-follows-strategy view as being too narrow and believesthat multinationals must transform their hier-archial form to something more flexible if theyare to survive (Hedlund and Rolander, 1990).Some of the main characteristicsof the heterarch-ial MNC are: many centers, in which traditionalheadquarter functions are dispersed; strategicroles for foreign subsidiaries; flexibility in govern-ance modes; global integration by normativemechanisms, such as shared culture and ethics;a holographic organization, where each part ofthe company shares information about the wholeand has access to detailed information; strategicaction, based on heuristic search orientation andexploitation of current potential (Hedlund andRolander, 1990: 25-27).To complete the picture of the 'process school'in international management, the new conceptsfor understanding strategic capabilities of MNCs(introduced by Hamel and Prahalad) should alsobe presented. These concepts emphasize the


13/21

110 L. Melinrole of strategic ntentin strategicdevelopmentprocesses (Hamel and Prahalad, 1989); theimportanceof developing core competencies ndiversified MNCs in order to reach globalcompetitiveness (Prahalad and Hamel, 1990);and the role of resource leverage and theimportanceof aspirationsoutstrippingresourcesfor proactive strategy making (Hamel and Pra-halad, 1992). The conceptsintroducedby Hameland Prahalad are intuitively quite reasonable,but when considered more carefully, they arevague and lack descriptiveprecision.The configurations uggested by the 'processschool' are complex and require an extremelyskillful and powerful top management eam thatcan orchestrate all internal forces in a largeworldwide corporation. The attitudes of themembers of the 'process school' range fromcritique of the possibility to find the rightstructuralorm, to the suggestingof waysto findthe right balance and compromisesbetween allinterests and units and thus simultaneouslyachieve global efficiency, local responsiveness,flexibilityand learning.The ambition to capturethe complexity in large, global business firms isdesirable(see Melin, 1987and 1989). However,the strong belief-expressed by Bartlett andGhoshal, 1989,andby Doz andPrahalad,1984-in a new type of fit, much more dynamicthan the old structure-follows-strategy-follows-environment is based on an extended use ofcultural control. A rather monolithic cultureseems to be necessary to unite the dispersedMNCs. This implies a view on culture as avariablethat can be controlledfrom the top ofthe organizationand based on an underlyingassumption hat manipulation f culturaldimen-sions will lead to intendedchangesin the mindsets of organizationmembers.These assumptionsarequestionedby manystudentsof organizationalculture(e.g., Smircich,1983;AlvessonandBerg,1992).In a critical review of 'Managingthe GlobalFirm' withBarlett,Doz andHedlund,aseditors,1990), Forsgren (1992) identifies two mainproblemsin the models of transnational,heter-archic and multifocalfirms:

First, heneedandpossibilitiesf integrationsstrongly veremphasizedomparedo differen-tiation. There is a tendencyfor integration(especiallyncombinationith hewordglobal')

to be understood as a more progressive devicethan differentiation. There is an apparent riskthat this tendency will lead to a neglect of thelimitations and costs of integration. Second,even though several authors maintain that theglobal firm is moving away from a hierarchytowards a multi-center structure, there seems tobe little doubt about the headquarter's possibilityto design the structure and systems in such away that a beneficial integration is created...The problem is primarily presented as analyticaland a question of design, rather than a politicalproblem (Forsgren, 1992: 479-480).

The power dimension is of utmost importancein theory-building about multinational enter-prises, as shown by Larsson (1985) in his politicalanalysis of foreign acquisitions made by severalSwedish multinational firms. A valid descriptionconcerning the dynamics of MNCs must containmore of the power struggle of individualsand groups on different levels that actuallycharacterizes these large organizations. We needto better understand how and why MNCs areable to perform comparatively well despitethe counteractive forces and inertia that areembedded in the culture of these mature organiza-tions.

The 'process school' has further weaknesses.The researchers found within this school havemainly used a clinical (or phenomenological)research method which, however, is seldomexplained in detail. The reader does not getmuch help when the bridge between empiricalwork and conceptual development is describedas 'intensive research suggests' or 'detailedanalysis was performed' (Doz and Prahalad,1984: 59-60). There are exceptions, of course,like the methodology appendix in Bartlett andGhoshal (1989). The conceptual language emanat-ing from this school is built up from empiricalresearch, and with the expressed ambition ofputting managerial relevance before theoreticalelegance. Organization and strategic managementtheory have been avoided (with a few exceptions)in the interpretation and concept generatingphases of research. The developed frameworkon managing MNCs has seldom been related toprevailing concepts describing similar phenomenain other contexts than that of the internationalfirm. The striving for managerial relevance hasalso led to a normative bias. The normative bentseems to be increasing with an increased degree


14/21

Internationalization as a Strategy Process 111of speculation in conceptual frameworks withoutempirical support. This is the case with themodels for the emerging new MNC: the multifocalstrategy, the heterarchy, and the transnationalsolution (Hagstrom, 1991).

The 'process school' emphasizes the dynamicdimension of strategy and structure. But, apartfrom the change process study (by Doz andPrahalad, 1987) of the transition in sixteen MNCsfrom local to cross-national orientation, theprocess orientation is in fact rather restricted.Doz and Prahalad developed in their study a four-stage process model of strategic redirection-incubation, variety generation, power shift, andrefocusing (Doz and Prahalad, 1987: 69-74). Theempirical focus in the 'process school' is onglobal MNCs, but their conceptual contributionsare expressed in rather general terms, and shouldbe seen as belonging to the fields of organizationtheory and strategic management rather thanconstituting an isolated IM theory.This notwithstanding, the research group rep-resenting the 'process school' of internationalmanagement has increased the knowledge in thisfield in several aspects. The researchers haveidentified new organizational patterns of MNCs.They have contributed a richer conceptual charac-terization of the global multinational firm withits coordination mechanisms and strategic logicin use in complex MNCs. They have articulatedthe strong need for more multidimensionalapproaches in (international) managementresearch. Furthermore, the empirical research isbased on access to several top managers in globalMNCs, a fresh contrast to much of the U.S.-based strategy research with its very weakempirical connection to strategists in action.

KEY THEMES FOR FUTURERESEARCH ONINTERNATIONALIZATION PROCESSESThree major themes in the field of internationalmanagement have been reviewed. The firsttheme, regarding stage models of internationali-zation, reflects the conventional mode of inter-nationalization that goes from export activitiesto foreign direct investment. The sequential stagemodel is too deterministic and stresses only earlystages of internationalization. This model shouldbe supplemented with research on new patterns

of internationalization of the 1980s and 1990s,covering also the process whereby internationallymature firms further increase their degree ofinternationalization. The second theme, researchon structures following strategies in MNCs,describes the historical development of structuralforms of MNCs such as the parent-subsidiarystructure, the international division and theworldwide product division. Major weaknesseswithin this theme are its attachment to thelinear thinking evident in strategy-structure-fitarguments, and the static character of theirconceptual contributions, in spite of the apparentlongitudinal nature of their research approach.The role of learning and political processes inthe internationalizationof the firmis not accountedfor at all. The third theme focuses on themanagement of DMNCs and constitutes the so-called process school in international management.This school of thought contributes a rich set ofnew and thought-provoking concepts which bothdescribe and prescribe management processes inMNCs. Major weaknesses of this school are itsquestionable empirical base, and weak relationshipwith existing management theory and its overpres-criptive nature. Furthermore, despite the label ofthe school, the process orientationin the theoreticalbody of the school is undeveloped.This concluding section suggests three keythemes for future research: the study of acqui-sition processes and internationalization, thestudy of dynamic processes in MNCs, and thestudy of internationalization processes in theirouter contexts. Each theme will be oper-ationalized through a number of questions perti-nent to research. These themes deal with majoromissions and weaknesses identified in the reviewof the field. Research on these themes shouldproduce additive knowledge on internationali-zation as a strategy process. This knowledgeshould be of a descriptive nature, based onthe belief that the purpose of (international)management research is to develop new modelsand frameworks that will help us understand howorganizations work rather than how organizationsshould work (see Daft and Buenger, 1990).The acquisition theme: Research on thedominant mode of internationalizationAcquisitions have been a predominant featureof internationalization during the last decade


15/21

112 L. Melinaccording to recent empirical evidences(McKiernan, 1992). Between 1983 and 1988,acquisitions worldwide rose by more than 20percent annually. This implies a far-reachingconcentration process with a remarkable increasein foreign ownership on many markets. In thissituation, research in international business isreally lagging behind. Conventional theorieson internationalization, including foreign directinvestment theory have paid little attention toacquisition as a major route to internationali-zation (Forsgren, 1989; 1990). The significantrole of acquisitions as a major mode ofinternationalization demands comprehensiveresearch efforts. In general, process theory onacquisitions is still weakly developed (Trautwein,1990). We need descriptions and conceptualiza-tions on crossborder acquisitions in order todevelop our understanding of the processes andmechanisms of internationalization. Some keyresearch questions on international acquisitionprocesses are suggested below:What are the typical patterns shown bycrossborder acquisition processes, in terms oftheir different strategic and organizational charac-teristics? What driving forces impel these acqui-sition processes? What strategic logic is embracedin crossborder acquisitions? The strategic logicmay express not only rational formulas, but eveninstitutionalized myths and subjective rationales.An illustration is the announcement of the lateeighties about E.C.-92 which led non-E.C.companies to escalate acquisitions in the CommonMarket. Research about these acquisition pro-cesses might reveal not only cases of purposefulstrategic intent to use acquisition as a means ofsecuring a competitive European position, buteven cases which were a sudden reaction tothe then-prevalent E.C. hysteria and whichilluminates mimetic or legitimating strategicbehavior (see Hellgren and Melin, 1991). Suchreactive ways of acting show that managersmay be more concerned with jumping on theacquisition bandwagon than with corporate profit(see Daft and Buenger, 1990).What characterizes the strategic choices ininternational acquisition processes? One suchstrategic choice concerns the selection of a foreignmarket to be entered by taking over some otherfirm which may be a rational choice of a specificcountry or a more inherent consequence of acompany being available for acquisition (Lindell

and Melin, 1991). Another strategic choice isbetween foreign acquisitions and organic growththrough foreign greenfield investments (Hennartand Park, 1992). An acquisition does not increasethe overall manufacturing capacity and is there-fore a suitable action in a mature and decliningindustry. In such industries, crossborder acqui-sitions as a means of global restructuring willincrease the degree of internationalization andat the same time reduce the population of firms.In which phases of a firm's internationalizationare acquisitions to be found? Regarding new-comers on international markets, recent studiesshow that these home-country-based firms maygo international directly through acquisition of aforeign competitor (Sharma, 1991; Luostarinen,1991), contrary to the prediction of the sequentialstage model. Forsgren claims that 'the lessinternationalized the firm, the higher its acqui-sition propensity' (1990: 262). These new obser-vations should be followed by more empiricalstudies in order to develop descriptive models ofinternationalization phases, where the acquisitionmode is included and where the order of phasesis not sequentially predetermined.In conclusion, an empirical focus on acquisitionprocesses creates good opportunities to betterunderstand acculturation processes as oneimportant dimension of internationalization.Acculturation is an ongoing process in allinternational firms with crossborder operations.But the acculturation process should be moreevident and visible when two different culturesmeet in the integration process that follows everycrossborder acquisition.The process theme: A process perspective onMNCsThe IM field has been dominated by studies onstructure, control and coordination of MNCs.The research is characterized by cross-sectionalapproaches, or at least findings expressed interms of structural constructs. The processperspective is (with few exceptions) absent inanalyses and theory generation. Knowledge aboutstructural forms of international business isunusually well-developed, while the knowledgeabout the transition of these structures is certainlyinsufficient. Five clusters of research questionscovering different aspects of a process perspectiveon MNCs are suggested:


16/21

Internationalization as a Strategy Process 113A common assumption in the IM field isthat structures change as a result of strategicadaptation to a changing environment. As dis-cussed earlier, this linear view should be ques-tioned: What internal forces impel change in

MNCs, regarding strategies, organizational formsand coordination mechanisms? How do internaland external forces interplay in the strategicprocesses of MNCs? How and why do controland coordination mechanisms change over timeand how do these changes interplay with strategicactions? These questions imply a need forresearch on strategic and structural change inMNCs based on a more reciprocal, dynamic andcomplex view than the model ofenvironment-strategy-structure fit (Pettigrew,1987; Melin, 1985; 1989; Hedlund and Rolander,1990).Several researchers in the IM field, fromStopford and Wells and onward, have identifieddistinct structural forms of MNCs. In accordancewith Child (1972), in his early critique of thecontingency school, we argue that research onMNCs should be more concerned with processesover time and should regard structures as rathertemporary manifestations of such processes. Whatcharacterizes the continuous change of thesetemporary MNC structures? If more stablestructures are to be revealed, what determinesthe steady state of each form and the transform-ation from one identified form to the following?According to Aman (forthcoming), who hasstudied the transition from a traditional parent-subsidiary structure to an across-the-border typeof MNC, this transition is a long-term evolution-ary process with several 'in between' phases.The research on coordinating mechanismsin MNCs describes the development in theseorganizations towards global integration, even iflocal responsiveness is part of the game (seeGhoshal and Nohria, 1989). This researchdescribes a rather homogeneous and well-unitedinternational organization. However, a dispersedorganization such as the MNC is a plurality ofsystems, where intended actions of powerfulmembers as well as unintended consequences ofthe social action structure play important roles(Brown, 1992). Besides coordination, anotherinherent but less controllable mechanism continu-ously affects the function of the MNC: thedifferentiation mechanism. How do differen-tiation processes influence the management of

MNCs? Which differentiation mechanisms arefound in political and cultural processes inMNCs? Research on differentiation processesshould pay attention to political and culturalforces that do not originate from the MNCheadquarters. These forces may have their originsnot only inside various units of the dispersedMNC but also outside, in the local environmentsof these units.The focus on heterogeneity and diversity raisesfurther questions on power processes and fluidboundaries of MNCs: Will MNCs remain hier-archial or will they grow into a multicenter form,in other words, have several power centers(characterized by Forsgren, 1992 as inter-nationalization of the second degree; see alsoHedlund, 1986)? This multicenter model is basedon the observation that different units of theMNC are integrated in different local businessnetworks (Forsgren, 1990; Forsgren and Johan-son, 1992). How does the power distribution inthese networks influence the strategic freedomof action for the MNC headquarters? How dounits in the internal network of the MNC becomeintegrated in these external business networks(see Ghoshal and Bartlett, 1990; Kogut, 1989)?What characterizes the interplay regardingexchange processes and political processesbetween these two types of networks?The need for more research on long-terminternationalization processes of MNCs shouldalso be noticed. Prevailing knowledge on inter-nationalization as a process focuses on firms inthe early phases of internationalization. But mostMNCs are internationally very experienced: Whatcharacterizes ongoing globalizing processes ofmature MNCs? What new patterns of phasesand modes of internationalization can be foundin these processes (see Bouchikhi and Kimberly,1992)? A research effort taking this perspectiveis a comparative project of internationalizationpatterns and organizational capabilities in Britishand Swedish firms (see Macdonald et al., 1991,and Gustavsson et al., 1992). To allow theillumination of the full internationalization pro-cess of a firm and to uncover developmentaltrajectories of this process, we need comprehen-sive studies that capture long epochs and evenbiographic histories (see Figure 2) of the wholedevelopment of the firm. Such an approach shouldincrease our understandingof internationalizationas a 'fundamental engine of organizational


17/21

114 L. Melindevelopment' (Kimberly and Bouchikhi, 1991:25).The context theme: Understanding theembeddedness of internationalization in outercontextsThis third theme stresses the importance ofrelating the internationalization process of thefirm to its surrounding context, which thereviewed research fails to do. Two suggestedsubthemes focus on internationalization embed-ded in two outer contexts; the institutional andsocial context and the sectoral context.The first subtheme concerns the embeddednessof internationalization processes in its socialand political outer context. Internationalizationinherently means that the international firm actsin a number of different social and culturalcontexts, often country-based. Hofstede (1983,1991) has shown the diversity in cross-culturalpatterns. Whitley (1992) have shown the diversityin the institutionalized structure of businesssystems in different countries. What characterizesthe link between the internationalization processof an individual firm and this institutional, socialand cultural diversity? How does the MNC thatpursues a rather uniform and integrated globalstrategy maneuver such diverse institutionalenvironments? Mutabazi (1992) argues, for exam-ple, that all attempts to reach cultural uniformityin MNCs are doomed to failure. According toJones (1991: 13), mainstream theories of theMNC are unable to access the institutional andsocial environment 'due to their undersocializedconstruction,' which underlines the need for newresearch efforts on this subtheme.When studying internationalization within astrategy process framework, it is crucial to focuson 'organizations in their sectors' (Child, 1988).The second subtheme, contrasting the first,emphasizes the embeddedness of internationali-zation processes in the context of industrialsectors crossing borders. Most industrial sectorsof today are international and characterized byglobal competition where rivals within one sectorcompete against one another on a worldwide basis(Porter, 1986). Furthermore, most internationalsectors are interlaced through widespread collab-orative arrangements between competitors indifferent overlapping constellations. The globalauto industry is a good illustration. The inter-

nationalization process of MNCs, or rather ofstrategic business units within MNCs must beunderstood in a dynamic and global sectorperspective, where collaborative forces with'win-win' relationships and competitive forces ofzero-sum character act simultaneously acrossborders (Hellgren and Melin, 1991).In addition to the three suggested researchthemes, future research on international manage-ment should also take into consideration thatall organizational processes are engrained bysubjective sensemaking (Gioia and Chittipeddi,1991). As the MNC acts in the perhaps mostcomplex and turbulent of all environments,the management process through creation andtransfer of meaning should here be especiallycritical (cp Salzer, 1992). In order to capture theessence of symbolic interaction and sensemakingprocesses we need radically different researchmethodologies that have been used in the fieldof internationalization, such as an ethnographicand interpretive field work approach (cp Botti,1992).I will end with some final concluding points.The research themes suggested above illustratethe need for a better connection betweeninternational management and theoretically moremature subdisciplines of management. This callfor a closer theoretical relationship with organi-zation and (strategic) management theory raisesa question not only about the boundaries but alsoabout the autonomy of the field of internationalmanagement. The boundaries between organi-zation theory, strategic management and inter-national management are fluid and blurred. Thisreview points toward a conclusion that theinternationalization dimension should beregarded as an empirical focus and not form thebasis for a theoretical field of management onits own. The general issues of business strategy,organizational structures, and coordination mech-anisms that have been the core of internationalmanagement should hardly be divided into onesubtheory for international firms and another fornoninternational firms. But this is what IMresearchers have tended to do, often not relatingtheir results and conceptual models to existingtheories. On the other hand, the multinationalenterprise represents one category of organiza-tions to which scholars of organization theoryshould devote more interest and research efforts.Students of organizations are served an important


18/21

Internationalization as a Strategy Process 115challenge by this type of large and dispersedorganization, with an inherent mosaic of cultures,unclear and ambiguous boundaries (within theintraorganizational network and also in relationto its outer context), a complex power structure,and multiple organizational identities.In conclusion, studies of internationalizationas a strategy process must capture the develop-ment and dynamics over time, the driving forcesof the process, and the content of the process.Internationalization processes are characterizedby a high degree of complexity, variability andheterogeneity, which taken together requireholistic research and truly longitudinal approach-es.

ACKNOWLEDGEMENTSThanks are due to Andrew Pettigrew for numer-ous suggestions and to my colleagues in theStrategic Change Research Group at LinkopingUniversity for all their support. In addition Iwould like to thank the Axel and MargaretAx:son Johnson Foundation for funding theresearch on which this article is based.

REFERENCESAharoni, Y. (1966). The Foreign Investment DecisionProcess. Division of Research, Harvard University,Boston, MA.Alvesson, M. and P.O. Berg (1992). Corporate Cultureand Organizational Symbolism. Walter de Gruyter,Berlin.Aman, P. The Emergence of a Cross-border Firm.The Transformation of Alfa Laval AB. IIB,Stockholm School of Economics, forthcoming.Bartlett, C. (1986). 'Building and managing thetransnational: The new organizational challenge'.In M. Porter (ed.), Competition in Global Industries.Harvard Business School Press, Boston, MA,pp. 367-401.Bartlett, C., Y. Doz and G. Hedlund (eds.) (1990).Managing the Global Firm. Routledge, London.Bartlett, C. and S. Ghoshal (Fall 1987). 'Managingacross borders: New organizational responses'.Sloan Management Review, pp. 43-53.Bartlett, C. and S. Ghoshal (1987). Managing AcrossBorders. The Transnational Solution. Harvard Busi-ness School Press, Boston MA.Bartlett, C. and S. Ghoshal (Summer 1991). 'Globalstrategic management: Impact on the new frontiersof strategy research', StrategicManagement Journal,12, pp. 5-16.

Bettis, R. (1991). 'Strategic management and thestraightjacket: An editorial essay', OrganizationScience, 2(3), pp. 315-319.Bilkey, W.J. (1978). 'An attempted integration of theliterature on the export behavior of firms', Journalof International Business Studies, pp. 33-46.Bjorkman, I. (1989). 'Foreign direct investments. Anempirical analysis of decision making in sevenFinnish firms'. Doctoral dissertation, No 42, TheSwedish School of Economics and Business Admin-istration, Helsinki.Bjorkman, I. (1990). 'Foreign direct investments:An organizational learning perspective'. Workingpaper, The Swedish School of Economics andBusiness Administration, Helsinki.Botti, H. (1992). 'The internationalization paradox inParodi: A research tale', Scandinavian Journal ofManagement, 8(2), pp. 85-112.Bouchikhi, H. and J. Kimberly (1992). 'Large Frenchfirms' FDI in the USA: Radical and incrementalprocesses associated with two management con-figurations'. In Proceedings from the first Inter-national Federation of Scholarly Associations ofManagement Conference, Tokyo, pp. 448-451.Bower, J. L. (1970). Managing the Resource AllocationProcess: A Study of Corporate Planning andInvestment. Harvard University Press, Boston, MA.Brooke, M.S. and H.L. Remmers (1970). The Strategyof Multinational Enterprise. American Elsevier,New York.Brown, R. (1992). Understanding Industrial Organiza-tions: Theoretical Perspectives in Industrial Soci-ology. Routledge, London.Buckley, P. J. (1988). 'The limits of explanation;Testing the internalization theory of the multi-national enterprise', Journal of International Busi-ness Studies, 19(2), pp. 181-184.Buckley, P. J. (1990). 'Problems and developments inthe core theory of international business', Journalof International Business, 21(4), pp. 657-665.Buckley, P. J. (1991). 'The analysis of internationalcorporate strategy: An integrated framework andresearch agenda'. In H. Vestergaard (ed.), AnEnlarged Europe in the Global Economy,(Proceedings of the 17th annual EIBA conference).The Copenhagen Business School, Vol. 1,

pp. 541-575.Burns, T. and G. M. Stalker (1961). The Managementof Innovation. Tavistock, London.Cavusgil, T. (November 1980). 'On the internationali-zation process of firms', European Research,pp. 273-281.Chandler, A. D. (1962). Strategy and Structure. MITPress, Cambridge MA.Child, J. (1972). 'Organization structure and strategiesof control: A replication of the Aston study',Administrative Science Quarterly, 17, pp. 163-177.Child, J. (1988). 'On organizations in their sectors',Organization Studies, 9, pp. 13-19.Cyert, R. and J. March (1963). A Behavioral Theoryof the Firm. Prentice Hall, Englewood Cliffs, NJ.Daft, R. and V. Buenger (1990). 'Hitching a ride ona fast train to nowhere: The past and future of


19/21

116 L. Melinstrategic management research'. In J. Fredrickson(ed.), Perspectives on Strategic Management, HarperBusiness, New York, pp. 81-103.Daniels, J. D., R. A. Pitts and M. J. Tretter (1985).'Organizing for dual strategies of product diversityand international expansion', Strategic ManagementJournal, 6, pp. 223-237.Davidson, W. H. and P. Haspeslagh (1982). 'Shapinga global product organization', Harvard BusinessReview, 60(4), pp. 125-132.Denis, J. E. and D. Depelteau (1985). 'Marketknowledge, diversification and export expansion',Journal of International Business Studies, 16,pp. 77-89.DiMaggio, P. and W. Powell (1983). 'The iron cagerevisited: Instituional isomorphism and collectiverationality in organizational fields', American Socio-logical Review, 48, pp. 147-160.Doz, Y. (Winter 1980). 'Strategic management inmultinational companies'. Sloan ManagementReview, 21, pp. 27-46.Doz, Y. (1986). Strategic Management in MultinationalCompanies. Pergamon Press, Oxford.Doz, Y., C. Bartlett and C. K. Prahalad (Spring1981). 'Global competitive pressures vs. hostcountry demands: Managing tensions in multi-national corporations', California ManagementReview, 23, pp. 63-74.Doz, Y. and C. K. Prahalad (1981). 'Headquarterinfluence and strategic control in MNCs', SloanManagement Review, 23(1), pp. 15-29.Doz, Y. and C. K. Prahalad (Fall 1984). 'Patterns ofstrategic control within multinational corporations',Journal of International Business Studies, 15,pp. 55-72.Doz, Y. and C. K. Prahalad (1987). 'A Process modelof strategic redirection in large complex firms: Thecase of multinational corporations'. In A. Pettigrew(ed.), The Management of Strategic Change. BasilBlackwell, Oxford, pp. 63-83.Doz, Y. and C. K. Prahalad (Summer 1991). 'ManagingDMNCs: A search for a new paradigm', StrategicManagement Journal, 12, pp. 145-164.Dunning, J. H. (Spring/Summer 1980). 'Toward aneclectic theory of international production: Someempirical tests', Journal of International BusinessStudies, 11, pp. 9-30.Dunning, J. H. (1981). The Eclectic theory of theMNC. Allen & Unwin, London.Dunning, J. H. (1988). 'The eclectic paradigm ofinternational production: A restatement and somepossible extensions', Journal of International Busi-ness Studies, 19, pp. 1-31.Edstrom, A. and J. R. Galbraith (June 1977). 'Transferof managers as a coordination and control strategy inmultinational organizations'. Administrative ScienceQuarterly, 22, pp. 248-263.Egelhoff, W. G. (1988). 'Strategy and structure inmultinational corporations: A revision of theStopford and Wells model'. Strategic ManagementJournal, 9(1), pp. 1-14.Forsgren, M. (1989). Managing the InternationalisationProcess: The Swedish Case. Routledge, London.

Forsgren, M. (1990). 'Managing the internationalmulti-centre firm', European Management Journal,8(2), pp. 261-267.Forsgren, M. (1992). 'Book review: C. A. Bartlett,Y. Doz and G. Hedlund (eds.): Managing theglobal firm', Organization Studies, pp. 477-480.Forsgren, M., U. Holm and J. Johanson (1992).'Internationalization of the second degree: Theemergence of European-based centres in Swedishfirms'. Europe and the Multinationals. EdwardElgar, Aldershof, pp. 235-253.Forsgren, M. and J. Johanson (eds.) (1992). ManagingNetworks in International Business. Gordon andBreach, Philadelphia.Fouraker, L. E. and J. M. Stopford (1968). 'Organi-zation structure and multinational strategy', Admin-istrative Science Quarterly, 13(1), pp. 47-64.Franko, L. G. (1976). The European Multinationals:A Renewed Challenge to American and British BigBusiness. Greylock, Stamford, CT.Ghoshal, S. and C. Bartlett (1990). 'The multinationalcorporation as an interorganizational network',Academy of Management Review, 15(4),pp. 603-625.Ghoshal, S. and N. Nohria (1989). 'Internal differen-tiation within multinational corporations', StrategicManagement Journal, 10(4), pp. 323-337.Gioia, D. A. and K. Chittipeddi (1991). 'Sensemakingand sensegiving in strategic change initiation',Strategic Management Journal, 12, pp. 433-448.Gomez-Mejia, L. (1992). 'Structure and process ofdiversification, compensation strategy, and firmperformance', Strategic Management Journal, 13(5),

pp. 381-397.Gustavsson, P., L. Melin, S. Macdonald and A.Pettigrew (1992). 'Global uniformity versus glocalis-ing through learning'. Paper presented at the annualStrategic Management Society Conference, London.Habib, M. and B. Victor (1991). 'Strategy, structure,and performance of U.S. manufacturing and serviceMNCs: A comparative analysis', Strategic Manage-ment Journal, 12(8), pp. 589-606.Hagstrom, P. (1991). The 'wired' MNC. The role ofinformation systems for structural change in complexorganizations. Doctoral dissertation, StockholmSchool of Economics.Hall, D. and M. Saias (1980). 'Strategy followsstructure!', Strategic Management Journal, 1,pp. 149-163.Hambrick, D. (1990). 'The adolescence of strategicmanagement, 1980-1985: Critical perceptions andreality'. In J. Frederickson (ed.), Perspectives onStrategic Management. Harper Business, New York,pp. 237-253.Hamel, G. and C. K. Prahalad (1989). 'Strategicintent', Harvard Business Review, 89(3), pp. 63-76.Hamel, G. and C. K. Prahalad (1992). Presentationsat the annual Strategic Management Society Confer-ence, London.Haspeslagh, P. and D. Jemison (1991). ManagingAcquisitions: Creating Value through CorporateRenewal. The Free Press, New York.Hedlund, G. (1984). 'Organization in-between: The


20/21

Internationalization as a Strategy Process 117evolution of the mother-daughter structure ofmanaging foreign subsidiaries in Swedish MNCs',Journal of International Business Studies, 15(2),pp. 109-123.Hedlund, G. (Spring 1986). 'The hypermodern MNC:A heterarchy?, Human Resource Management,pp. 9-35.Hedlund, G. and A. Kverneland (1984). Investingin Japan-Experience of Swedish Firms. IBB,Stockholm School of Economics, Stockholm.Hedlund, G. and D. Rolander (1990). 'Action inheterarchies: New approaches to managing theMNC'. In C. Bartlett, Y. Doz and G. Hedlund(eds.), Managing the Global Firm. Routledge,London, pp. 15-46.Hedlund, G. and P. Aman (1984). Managing Relation-ships with Foreign Subsidiaries. Sveriges Mekanfor-bund, Stockholm.Hellgren, B. and L. Melin (1991). 'Corporate strategiesin Nordic firms facing Europe-Acquisitions andother collaborative strategies'. In L. G. Mattssonand B. Stymne (eds.), Corporate and IndustryStrategies for Europe. Elsevier, Amsterdam,pp. 327-351.Hennart, J. F. and Y. R. Park (1992). 'Greenfield vsacquisition: The strategy of Japanese investors inthe United States'. Paper presented at the annualmeeting of the Academy of Management, LasVegas, NV.Hofstede, G. (1983). 'National culture in four dimen-sions', International Studies of Management andOrganizations, 13, pp. 97-118.Hofstede, G. (1991). Cultures and Organizations.Software of the Mind. McGraw-Hill, London.Huff, T. and R. Reger (1987). 'A review of strategicprocess', Journal of Management, 13(2),pp. 211-236.Jaeger, A. M. (Fall 1983). 'The transfer of organi-zational culture overseas: An approach to controlin the multinational corporation'. Journal of Inter-national Business Studies, pp. 91-114.Jaeger, A. M. and B. R. Baliga (1985). 'Controlsystems and strategic adaptation: Lessons fromthe Japanese experience', Strategic ManagementJournal, 6, pp. 115-134.Johanson, J. and J. Vahlne (Spring-Summer 1977).'The internationalisation process of the firm: amodel of knowledge development on increasingforeign commitments', Journal of InternationalBusiness Studies, pp. 23-32.Johanson, J. and J. E. Vahlne (1990). 'The mechanismof internationalisation', International MarketingReview, 7(4), pp. 11-24.Johanson, J. and F. Wiedersheim-Paul (October 1975).'The internationalisation of the firm: Four Swedishcase studies', Journal of Management Studies,pp. 305-322.Johanson, J. U. and F. Nonaka (1983). 'Japaneseexport marketing: Structures, strategies, counter-strategies', International Marketing Review, 1pp. 12-25.Jones, M. (1992). 'Mainstream and radical theories ofthe multinational enterprise: Towards a synthesis'.

Paper presented at the annual meeting of theAcademy of Management, Las Vegas, NV.Kimberly, J. and H. Bouchikhi (1991). 'The enginesof organizational development and change: Howthe past shapes the present and constrains thefuture'. Working paper, Wharton School, Philadel-phia, PA.Kogut, B. (1989). 'A note on global strategies',Strategic Management Journal, 10(4), pp. 383-389.Larsson, A. (1985). Structure and change: Power intransnational corporations'. Doctoral dissertation,Acta Universitatis Upsaliensis, Uppsala.Lindell, M. and L. Melin (1991). 'Acquisition andrealizationof vision: Diversification through strategicbridging'. Paper presented at the annual StrategicManagement Society Conference, Toronto.Luostarinen, R. (1991). 'Development of strategicthinking in international business'. In J. Nasi,Arenas of Strategic Thinking. Foundation forEconomic Education, Helsinki.Lyles, M. (1990). 'A research agenda for strategicmanagement in the 1990s', Journal of ManagementStudies, 27(4), pp. 363-375.Macdonald, S., A. Pettigrew, P. Gustavsson and L.Melin (1991). 'The learning process behind theEuropean activities'. In H. Vestergard (ed.), AnEnlarged Europe in the Global Economy,(Proceedings of the 17th annual EIBA conference).The Copenhagen Business School, 1, pp. 1-30.Martinez, J. I. and J. C. Jarillo (Fall 1989). 'Theevolution of research on coordination mechanisms inmultinational corporations', Journal of InternationalBusiness Studies, 20(3), pp. 489-514.

McKiernan, P. (1992). Strategies of Growth: Maturity,Recovery and Internationalization. Routledge, Lon-don.Melin, L. (1985). 'Strategies in managing turnaround:The Scandinavian TV industry', Long Range Plan-ning, 18(1), pp. 80-86.Melin, L. (1987). 'Commentary on chapter 4: Under-standing strategic change'. In A. Pettigrew (ed.),Management of Strategic Change. Basil Blackwell,Oxford, pp. 154-165.Melin, L. (1989). 'The field-of-force metaphor'. In T.Cavusgil (ed.), Advances in InternationalMarketing,Vol. 3, JAI Press, Greenwich, CT, pp. 161-179.Miller, D. and P. H. Friesen (1982). 'The longitudinalanalysisof organizations: A methodological perspec-tive', Management Science, 28(9), pp. 1013-1034.Mintzberg, H. (Fall 1987). 'The strategy concept I:Five PS for strategy'. California ManagementReview, 30, pp. 11-23.Mutabazi, E. (January 1992). 'What are the bestmethods for dealing with the intercultural aspectof management?' In Group ESC Newsletter, Lyon,No 8.Ouchi, W. G. and J. B. Johnson (June 1978). 'Typesof organizational control and their relationshipsto emotional well being', Administrative ScienceQuarterly, 23, pp. 293-317.Perlmutter, H. V. (1969). 'The tortuous evolution ofthe multinational corporation', Columbia Journalof World Business, pp. 9-18.


21/21

118 L. MelinPeters, T. (Spring 1984). 'Strategy follows structure:Developing distinctive skills', California Manage-ment Review, 26(3), pp. 111-125.Pettigrew, A. M. (1985a). The Awakening Giant:Continuity and Change in ICI. Basil Blackwell,Oxford.Pettigrew, A. M. (1985b). 'Examining change in thelong-term context of culture and politics'. In J.Pennings (ed.), Organizational Strategyand Change.Jossey-Bass, S

Documents

International is at Ion as a Strategy