inkpen beamish

7/28/2019 Inkpen Beamish

1/27

Academy o Management Review1997. Vol. 22, No. 1, 17 7-202.

K N O W L E D G E B A R G A I N I N G P O W E R . A N D T H EI N S T A B I L I T Y O F I N T E R N A T I O N A LJ O I N T V E N T U R E S

A N D R E W C . IN K P E NT h u n d e r b i r dP A U L W . B E A MIS H

U n i v e r s it y of W e s t e r n O n t a r ioA lthough the high rate of instab ility of international joint ventures(IJVs) has been well documented, the underlying reasons for the in-stability need clarification. In this article, we develop a theoreticalframework for instability of IJVs grounded in a bargaining power anddependence perspective. InstabiJity is defined as a major change inpartner relationship status that is unplanned and premature from oneor both partners' perspe ctives. T he core argument is that the instabil-ity of nV s is asso ciated with shifts in partner bargaining pow er. S hiftsin the balan ce of barg aining power occur whe n partners of an IJVacquire sufficient know ledge and skills to elimina te a partner depen-dency and make the IJV bargain obsolete. O ur primary focus is on theacquisition of local knowledge by the foreign partner and the impactthat this acquisition of knowledge has on the stability of the IJV.

As competi t ion increasingly becomes more global , many firms areusing al l iances to enter new markets , obtain new ski l l s , and share r isksand resources. Despite the surge in their populari ty, international al l i-ances are often described as inherently unstable organizational forms.Porter (1990), for ex am ple , obs erve d tha t al l ia nc es involve significantcosts in terms of coordination, reconcil ing goals with an independententi ty, and creating competi tors . Porter suggested that these costs makemany al l iances t ransi t ional ra ther than s table arrangements and, there-fore, al l iances are rarely a sustainable means for creating competi t iveadvantage . Support ing this argument , authors of several empir ical s tud-ies of all ian ce s h av e found insta bility ra te s of close to 50% (e.g., Bleeke &Ernst, 19 91 ; Kogut, 1988). Ba sed on th e finding tha t 24 of the 49 inte rna-t ional ' al l ia nc es they studied w ere conside red fai lures by one or bothpartners, Bleeke and Ernst (1991) suggested that most al l iances will ter-minate, even successful ones.Al l iances a lso have bee n descr ibed as a race to leain, and the par tnerthat le arn s the fastest d om inat es the relat ion ship (Hamel, 1991). Accord-ing to this scenario of inevitable instabil i ty, there are clear winners and

T he authors thank S teve Currall, Charles H ill, and AMR's five anonymous reviewers fortheir help in developing this article.


2/27

178 Academy of Management Review Januarylosers. However, some international alliances survive and prosper forman y years, and both sides becom e more compet it ive through a win-winrela t ionship. This example ra ises the issue of why some infernat ionalal l iances are more stable than others. We address this i ssue by examin-ing equ ity- ba se d in tern atio na l joint ve ntu res (IJVs). An eq uity -ba se d JV isan alliance that combines resources from more than one organization tocreate a new organizational entity (the "child"), which is distinct from itspare nts. Equi ty-based JVs are cons idered hierarch ical b eca use they moreclosely replicate some of fhe features associated with organizational hi-erar chie s th an do other allia nc es (Gulati, 1995). Siecor, an all ian ce be-tween S iemens an d C orning, provides an exam ple . In this JV, the par tnersbrought together thei r complementary capabi l i t ies in te lecommunica-t ions and glass technology to bui ld an independent organizat ion wi th i t sown headquarters, CEO, board of direcfors, and staff. An IJV is a JV wifhfwo or m ore par en fs of differenf na tion ality . Our focus is on IJVs b as ed infhe home country of one of the partners, which we call the Joca] partner.The par tner op eratin g ou tside i ts country of domicile is referred to as theforeign partner.

A varie ty of st ra tegic object ives h ave be en sug ges ted to explainfirms' m ofives for the form ation of IJVs (Co ntracto r & Lo ran ge , 1988;Hennart , 1991; Kogut, 1988). A firm th at e n te rs a foreign m ark et for the firstfime is likely to us e a joint ventu re; this is also true for the foreign firm tha tseeks to obtain access to resources controlled by local f irms (Hennart ,1991). In both situa tion s, IJVs ar e offen ch os en b ec au se the a lte rn ativ es ofreplicafing a complefe operation via full acquisition or greenfield invesf-ment are too cost ly and because the local partner controls resourcesdee m ed u seful to the foreign partn er. An IJV m ay be de sig ne d to market aproduct in the local market, or it may be involved in the sourcing ofmaferials, componenfs, or fechnology, possibly for use in the foreign parf-ner's home market. IJVs are widely used as a mode of direct foreign in-ves tm ent. For ex am ple , M akino (1995) found th at alm ost 70% of J ap an es edirect inve stm en ts in ma nu factu ring in As ia as of 1991 w ere via IJVs.

Regardless of fhe specific venfure objective, our primary interest, liketha t of Yan a n d G ray (1994) a n d Pa rk he (1991), is IJVs in w hic h th e foreignparfner seeks a viable long-ferm presence in a counfry via direct invest-ment. In this type of IJV, it would be unusual for the local partner not tocontribute some local knowledge. The primary knowledge contribution offore ign partners general ly involves technology, management expert ise ,an d g loba l s upport (Yan & Gray, 1994). Note that we a re not su gg est ingthat the only contribution of a local parfner is local knowledge. In anofhersection of fhis article, we suggest that if local parfners fake steps to en-sure that their roles enco m pas s mo re than a one-t ime contribution of localknowledge, insfabi l i fy may be control lable . Nevertheless, even whentechnology so urcing or risk sha ring a re im portant ob jectives of the foreign


3/27

1997 nipen and Beamish 179

Unlike al l iances such as R&D p artners hips or l i cens ing agreements ,contracts between IJV par tners are often executed under condit ions ofhigh uncer tainty, and, therefore, it is highly unlikely that all future con-t ingencies in an IJV can be ant ic ipa ted at the outset . As IJVs grow, theymay develop an identity and a culture distinct from that of the par tners ,add ing to problems of coordination. Also, the collaborative motives forIJVs are often different from other alliances; therefore, statements gener-alizing about the instabil i ty of a l l i ances should be interpreted carefully.^The factors a sso ciate d w ith instabil i ty of IJVs are not nece ssari ly the sa m eas those for all types of a l l i ances .In the l i teratu re on JV instab ility, va riou s factors hav e bee n identifiedas causes of instabi l ity , including ch an ges inpartners ' s t ra tegic missions(Harrigan & Newman, 1990), c h a n g e s in importance of the JV to the pa ren ts

(Harrigan & Ne wm an, 1990), inc rea se s in the competit ive rivalry betweenpartners (Kogut, 1989), the foreign investment cl imate of the host country(Blodgett, 1992), and the exis tence of prior relationships between the part-ne rs (Blodgett, 1992). Altho ugh th es e factors may be associated with in-stability, and in some cases provide a strong indication that instabil i ty isimminent , they provide incom plete explana t ions for the instabil i ty of IJVs.Our interest is instabil i ty that results from factors endogenous to theIJV relationship. Figure 1 i l lust rates the bounda r i e s of the article and theconceptual orientation. Factors creating instabil i ty as a result of reasonsexogenous to IJVs (e.g., changes in foreign country investment or owner-ship regulations; shifts in polit ical regimes) are for the most part notcontrollable bym a n a g e r s of IJVs and are outside the scope of this art icle.We argue tha t the primary factor contributing to instabil i ty, and a factorthat can be controlled by firms in IJVs, is a shift in par tner barga in ingpower asso ciated with the acquis i tion of know ledge and ski l ls that a l lowsa firm to e l imina te a partner dependency. With a few exceptions (Hamel,1991; Parkhe, 1991; Yan & Gray, 1994), the l i terature on IJVs has not ad-dressed this issue in detai l . Thus, the dep ar ture in this art icle is the focuson factors that enable firms tom a k e the IJV ba rg ain obso lete. We explorethe quest ion of how IJV firms, and in particular foreign partners, are ab l eto create an unstable environment .

' W hen government legislat ion prohibi ts wholly owned su bsidiaries, s i lent partn ers thatcontribute little more than a m e a n s to bypass legislat ion may be used. Also, if a foreignpartner cannot legal ly become sole owner of the JV, our stabi l i ty arguments do not hold.However, as t rade and investment opportunit ies have become more l iberal , part icularly indeveloping countries, few countries continue to m a n d a t e the use of JVs for foreign invest-ment .For example, l icensing agreements and a r r angement s for sharing technology havegained notoriety as vehicles for licensee firms to build their own ski l l base and shift theal l iance of power in their favor. An art icle that describes the competi t ive implicat ions ofsharing technology with a potential competitor was written by Reich and Mankin (1986).These authors focused on high-technology agreements in which the U.S. partn er acq uired


4/27

180 Academy of Managem ent Review JanuaryFirms contribute various types of knowledge and skills to their IJVs.For local partners, a cri t ical knowledge contribution revolves around anunderstanding of local market, cultural , and environmental condit ions.

Althoug h the foreign pa rtn er do es not ne ce ssa rily e nter th e IJV with th especific objective of knowledge acquisition, access to knowledge origi-nating in the local country is an important factor in motivating the foreignpar tner to choose a n IJV investm ent rathe r tha n full o wn ership . We pro-pose that as the foreign partner 's local knowledge and i ts commitment toacquire local knowledge increase, the probabil i ty of the instabil i ty of theIJV also inc reases bec aus e of chan ge s in par tner depen dency . We alsoargue that the acquisit ion of technology-based knowledge by the localpartner, al though possible, is less l ikely to be a factor in creating insta-bil i ty than is the foreign partner 's acquisi t ion of knowledge.The loca l knowledge argument complements the wel l -developedtheme in the internat ional management l i terature concerning the f re-quent conflict between global and local optimization of strategy (e.g.,Bartlett & Sh osh al, 1989; Fran ko, 1971). Give n t he se conflicting de m an d s,two closely related questions are addressed in this art icle. First , when am ultin atio na l en terp rise (MNE) involved in an IJV a s a foreign pa rtn erdec ides to pu rsu e a localization strateg y (i.e., a strategy of opera ting au-tonomously in the local environment), how can i t happen? We argue thatthe MNE's acquisition of local knowledge enables the MNE to make thetransit ion from an IJV to a s ubs idiary. We also sug ges t tha t an impo rtantdist inction h as to be m ad e betw een the effect that local kno wle dge h as onstabil i ty an d the decision to acq uire the kn ow ledge. For exa m ple, an MNEma y dec ide that i ts IJV should be a wholly owne d su bsid iary. Un less theMNE ha s acquired the local know ledge nec essary to m an ag e the v entureautonom ously, a wholly ow ned s ub sidiary is not a viab le option. Thus, theacquisit ion of local knowledge can facil i tate the transit ion to a subsidi-ary, and it is a necessary factor in the decision.The second and equally important question involves the factors thatinfluence the foreign pa rtner 's decision an d ab il i ty to acq uire local knowl-

edge . In some IJVs, the foreign partner may choose not to pursue a local-ization strateg y. In other IJVs, localization m ay not be po ssib le b ec au sethe local partner outlearns the foreign partner, which becomes the basisfor an obsolete bargain. An examination of these questions requires ananalysis of the dynamics of collaborative relationships, so that we candevelop a comprehensive framework of the instability of IJVs.DEFINITIONS AND THEORETICAL UNDERPINNINGS

Loca l Know ledgeTo esta blis h an o pera tiona l pre sen ce in a country, a firm m ust a cc esslocal kno wled ge as a m ea ns of overcom ing market un certa inties (Stopford


5/27

1997 np en an d Beamish 181H am el, 1990). This infrastruc ture inc lud es sa le s forces, local plan ts, m ar-ket inte l ligence, an d the m arket ing presenc e necessary to un ders ta nd an dserve local markets. Local knowledge also relates to cultural tradit ions,norms, values, and inst i tut ional differences. For example, when KentuckyFried Chicken entered China, a local par tner was considered essent ia lbe cau se of the comp lexi ties asso ciated wi th obtaining op erat ing l icensesand leases , negot ia t ing employment contracts , and interpret ing invest -ment regulations. When a foreign firm does not have local knowledge, asin the ca se of Kentucky Fried C hicke n, an IJV ca n be u se d to gain quickac ce ss to a loca l partn er's kn ow ledg e b as e. Yan an d G ray (1994: 1492)quo ted a U.S. m an ag er in a U.S.-China IJV: "We hav e the techno logy an dcer ta in know-how. The Chinese par tner knows how to make things hap-pen in China. You put the two together right, it works."Instability of IIVs

As indicated previously, the primary factor contributing to instabil i tyis a shift in partner bargaining power. Several definit ions of instabil i tyhav e been us ed in the JV literature. Franko (1971) defined a JV a s unstablewh en pa rent ho ldings chan ged to include 50% or 95% ownership, a parentsold its JV interest, or the venture was liquidated. Killing (1983) consideredboth a shift in JV control an d ve ntu re term inatio n a s ev iden ce of insta bil-i ty. Other researchers have adopted a narrower view. For example, Kogut(1989) used venture termination as the sole indicator of instability. How-ever, a s Kogut indica ted, a JV can no t be con sidered un sta ble simplybecause i ts l i fespan is short . All relat ionships between firms face chal-lenges that threaten to change or terminate the basis for cooperation.Somet imes terminat ions are planned and ant ic ipated by the involvedpa rt ies . Ven tures also m ay be termin ated a s a matter of policy when thereis a ch ang e in the own ership or m ana gem ent of the parent . In other case s ,difficult ies associated with ending a relat ionship may create a rat ionalefor m ain tain ing a n exist ing JV that would otherw ise be te rm inated .

We maintain that instabil i ty should be l inked with unplanned equitychanges or major reorganizations. Usually, instabil i ty wil l result in pre-m atur e term ination of a JV, ei ther w hen one partn er a cqu ires th e JV busi-ne ss or the venture is dissolved. A com plicating factor is that term inationof a JV will not alw ay s be a mu tual decision (Hamel, 1991; Parke, 1991). Thepre m atur e term ination of a JV m ay be precip itated b y the action s of onepartner. For example, when one firm is trying to learn from its partner inorder to reduce i ts dependency, the partner that is doing the learning mayha ve very different longevity objectives than th e partner tha t is providingthe knowledge. In this scenario, a terminated relat ionship would be clas-s i f ied as unstable , because terminat ion was premature f rom the perspec-tive of the partner that is providing the knowledge. Thus, if at least one ofthe par tners ant ic ipates a long- term rela t ionship, premature terminat ion


6/27

182 Academy of Management Review January

terminat ion at the t ime the JV is formed, instability will not be an i s sueunless terminat ion occurs prematurely.In keep ing w ith the previous a rgum ents , JV instability is defined as amajor change in re la t ionship s ta tus that was unplanned and piematuiefrom the perspect ive of either one or both par tners . In most JVs, the par t -ners do not have a specific plan for the terminat ion of their ventures.^ Thepremature terminat ion of a JV can be t r aumat ic for the venture par tners .However, it is important to emphas ize tha t we do not eq ua te JV longevitywith JV success. Many firms view JVs as intentionally temporary andrecognize that their ventures wil l not last indefinitely. If a JV term inationis an orderly and mutual ly planned event , the JV may well be eva l ua t edas extremely successful . In fact, a JV th at is prematurely terminated a lsom ay be eva lua ted as successful , dep end ing on the cri teria used to evalu-a te its performance.*If an uns table JV involves cha ng es in re la t ionship s ta tus , a sta ble JV,by definition, must be one without unplanned changes in re la t ionshipstatus. However, stable JVs may be t e rmina ted when the s t ra tegic needsof the par tners change . For exampl e , in 1995, Chry sler an no un ced that itwas winding down its 25-year alliance with Mitubishi Motors (Templin,1995). One of Chrysler 's objectives in the re la t ionship was acces s to asource of smal l cars for the U.S. market . Because Chrysler developed aninternal supply of cars , the ra t ionale for the relat ionship with Mitsubishiw as no longer present.

A further cons ideration is that stabil i ty carries a posit ive connotation.In fact, in the opening section of this article, we l inked stabil i ty with awin-win re la t ionship. In our view, a s tab le JV is one in which the par tnersbel ieve the benefi ts to the re la t ionship exceed the costs of terminat ion.Never theless , a JV may h a v e the ou t ward appea rances of stability; how-ever, one par tner may view it as unsuccessful because the b a l a n c e ofpower has shifted to a degree that is undes i rab le . In another section ofthis art icle, we discuss how the acquis i t ion of knowledge by one par tnerin what looks like a stable JV can create the potent ia l for instabil i ty.Theoretical UnderpinningsOur framework is conceptual ly grounded in a ba rga i n i ng power anddependence perspect ive . This perspect ive was deve loped by Emerson(1962), and it was general ized to the organizat ional level in Pfeffer andSalan cik's (1978) resource d ep en de nc e m odel. The es sence of themodel is

^ A seldom used option is to incorporate "fade-out" provisions in the JV agreement. Thistype of venture has been used in China, but it has been largely discontinued. If longevity isnot a goal, another option is "contractual" JVs that include a specific termination date. Also,JV agreements sometimes include a specific period of time during which the partners havethe option to either renegotiate or modify their agreement.

* This definition of instability raises some interesting methodological questions. Spe-


7/27

1997 inJcpen a nd Beamish 183that the possession or control of key resources by one enti ty may makeother organizations dependent on that enti ty. According to this notion ofresource fif, relafionships are ferminafed for the inverse of the reasons forw hich fhey w ere formed (Seabrighf, Levinfhal, & Fic hm an , 1992). Yo ungand Oik (1994) found support for fhis argument in a study of U.S.-basedR&D consort ia. In this study, al l ian ce firms that h ad ga ine d k now ledge asa result of membership in the consort ium were more l ikely to leave thea l l i ance .In a cooperative relat ionship, dependence can be a source of powerfor the firm controlling key resources, because, to some degree, each firmcan increase or wi thhold resources that are a t t ract ive to i t s par tner(Bacharach & Law ler, 1980). W hen on e firm con trols an "ir rep lac ea ble " JVreso urce or inpu t, a de pe nd en cy situa tion is cre ate d (Ham el, 1991). Al-thoug h d epe nd enc y is voluntary w hen firms choose to form JVs, once theJV is ope ration al , firms de pe nd on their partn ers for specific resourc es an dinp uts. A firm fhat h a s the option to contribute or withhold a n impo rtantresource or input can use that option as leverage in bargaining with i tspartner (Pfeffer, 1981).

The not ion of "cooperat ive" may seem at odds wi th dependence andpow er. However, the key point is tha t at th e time a n IJV is formed, ea chparfner is dependent on the other(s) for critical inputs. Thus, the firmsmust cooperate to ensure that these cri t ical inputs are transformed into aproductive enti ty ( i.e. , the IJV). Over t ime, the de pe nd en ce ma y ch an ge ,and as a resul t , the bargaining power of one par tner may be enhanced.When that hap pe ns, the JV par tne r wi th the increas ed ba rga inin g powerhas access to more par tners and opt ions , including terminat ing the ven-ture, compared to the partner with l imited bargaining power (Yan & Gray,1994).The bargaining power perspective is part icularly appropriate for fheex am ina tion of fhe sfabilify of JVs, be ca u se a ll JVs involve a ne go tiate dba rga in b etwe en the pa r tners . At a g enera l level, bar gain ing power in JVsis based on the relat ive urgency of cooperation, available resources, com-mitments , other a l ternat ives , and the s t rengths and weaknesses of eachpa rtn er (S chelling, 1956). Fo cus ing m ainly on do m estic JVs, H arrig an an dNewman (1990) adopted the view that the relat ive bargaining power ofpotent ia l JV par tn ers i s determ ined pr imari ly by w hat each par tner br ingsto the venture. Rarely will these contributions be symmetrical , becauseeach partner views the potential costs and benefi ts of cooperating differ-ent ly . As par tners ' s t ra tegic miss ions , expectat ions , loyal t ies , and re-source mixes change, the balance of bargaining power in the JV shiffs(Harrigan & N ew m an, 1990). W hen fhe ba la nc e of pow er shifts, an d inevi-tably i t wil l , the need for cooperation between the partners may diminishor disappear. At this point , the parent may decide to undertake al l activi-


8/27

184 Academy o Management Review J a n u a r yboth resource-based and context-based components of bargaining power .The context-based components , al though important , are only indi rect lyrela ted to the dyn am ics of the IJV rela t ion ship. Given ou r prima ry interestin shi f ts in par tner bargaining power associated wi th the acquis i t ion ofknow ledge, the resou rce-based com ponen ts are of par t icu lar interes t . Theresources a n d ca pab i l i t ies commit ted by the par tn ers to the JV were amajor sou rce of barg ain ing power . Exper t ise in a re a s such a s local sourc-ing, domest ic dis t r ibut ion, and personnel management was the main re-source contributed by the local partners (Yan & Gray, 1994). For the foreignpar tners , resource contr ibut ions included exper t ise and technology forproduc t ion management and g lobal suppor t .To l ink the pa r tn er resou rce contr ibut ions di rect ly to bar ga in in gpower an d to un ders ta nd the process of ba rga inin g power shif ts , conceptsof organizat ional knowledge management must be incorporated in theframework. The pa ce of kn ow ledg e acq uisi t ion by one IJV pa rtn er is a nimportant process dimension, because, as Hamel (1991) argued, this di-mension is very much within the f irm's control . Therefore, Hamel identi-f ied learning as the most important e lement in determining re lat ive bar-gaining power . Substant ial knowledge acquis i t ion by one par tner overt ime can erode the value of the knowledge contributed by the other part-ne r , b reak ing down the bargaining re la t ionsh ip be tween the par tne rs .This argument forms the fundamental conceptual premise of this ar t ic le :The acq uisi t ion of know ledge a n d sk il ls ca n shift an IJV pa rtn er 's ba r-ga inin g power and m ay en ab le the firm to e l imin ate i ts dep end en cy on i t spar tne r .

Partners in IJVs acquire knowledge and skil ls through a process ofknow ledge ma na gem en t an d creat ion. Al though s ti ll rather sm al l , there i sa growing body of theoretical (Kogut, 1988; Parkhe , 1991; Pucik, 1991; West-ney , 1988) a n d em piric al stu die s (Dodgson, 1993; Ha m el, 1991; Inkp en,1995a; Inkp en & C ros sa n, 1995; Sim onin & Helleloid, 1993) a dd re ss in g theissue of JVs and al l iances as mechanisms for gaining access to par tners 'know ledge an d ski l ls . When a JV is created, organiza t ional bou nd ar ie sbecome permeable . This permeabi l i ty provides f i rms wi th a "window ontheir pa rtn ers ' broa d ca pa bil i t ie s" (Ham el, Doz, & Pr ah ala d, 1989: 134).Consequent ly, knowledge creat ion and learning should be viewed as po-tential strategic benefi ts of joint venturing.By forming a JV, pa rtn er f irms ma y ga in a cce ss to the em be dd edknowledge of other organizat ions and, therefore , to new organizat ionalskil ls and capabil i t ies. Through their involvement in the operation of theJV, firms can learn from their partners. Huber (1991) called this grafting.the process by which firms internalize knowledge not previously avail-able wi thin the organizat ion. Knowledge perceived as potent ial ly usefulfor the organizat ion may be acquired by individual managers or organi-


9/27

1997 Inkpen and Beamish 185

^ 0115>^> l l


10/27

188 Academy o Management Review January

0)

n

of

a>>'ga

rs

Pn

cu.'9.Fo

red

2iston

2.u


11/27

1997 JnJcpen and Beamish 187

giv en sh are d org an izat ion al m ea ni n g (Daft & W eick, 1984; No naka , 1994).The t ranslat ion of new know ledge into organizat ional act ion is the b a s i sfor creating new skil ls that underpin a f i rm's compet i t ive advantage. Ofparticular interest in this art icle is the knowledge that enables f i rms in-volved in IJVs to e l iminate dependencies on thei r par tners .

A FRAMEWORK OF INSTABILITY IN IJVsThrough a focus on partner knowledge acquisit ion that shifts bar-gain ing power. Figure 2 i l lust rates the framework of the instabil i ty of theIJV. As discussed in the previous section, the ini t ia l balance of powerbetween par tners of the IJV will inevitably shift. In the next sections, weexamine the factors asso ciate d w ith thes e shifts in barga in ing power, and

from this discussion, we generate specific proposit ions.Knowledge Acquisition by the Foreign PartnerKnowledge of the local environment is usual ly a key resource of localpar tners ; it is a l so a key source of barga in ing power, beca use it m a k e s theforeign par tner dependent on the local par tner (Yan & Gray, 1994). As aforeign partner increases its knowledge of the local market, instability ofthe IJV relat ionship becom es m ore probable, be cau se the foreign partnerga ins barga in ing power . When the JV was formed, local knowledge mayhave contr ibuted to the mutual needs of the partners (Beamish, 1988).

When the need d i ss ipa tes , and the foreign partner acquires local knowl-e dge , the foreign partner may view a JV as unnecessary . In effect, overt ime, the unique domain of the local partner shifts from being comple-mentary to the foreign partner to being undis t inguished (Ring & Van deVen, 1994). This line of reasoning leads to the following proposition:Proposition 1: As the foreign paitnei's local knowledgeincieases, the ioieign paitnei's dependence on the localpaitnei decieases, leading to a shiit in baigainingpower and gieatei likelihood o the instability o the IJV.

A reduction in the foreign partner 's commitment to the IJV and needfor its partner need does not a lways mean the foreign partner will acquirethe IJV bus ines s or es tabl i sh a subsidiary. As the foreign partner in-c reases its local knowledge, var ious outcomes are possible . For example ,an increase in knowledge that erased a par tner dependency p lus a desireto mainta in a p r e s e n c e in the local market were the c i r cums tancesprompting Ralston Purina, Bayer AG, Monsanto Co., and Sandoz to con-vert a l l iances to subs id iar ies in Japan (Ono, 1991).^ In Ralston Purina's

^ A popular stereotype is that of a "naive" U.S. firm contributing technology and aJapanese f i rm contributing local knowledge to a Japan-based JV. After several years of


12/27

188 Academy of Management Review JanuaryFIGURE 3Know ledge Acquisition and Instability

Local Partner's Evaluation ofForeign Partner's K nowled ge

Access Acquisition

Foreign Partner'sEvaluation ofLocal Partner'sKnowledge

Access

Acquisition

Very stableCooperative

Very un stab leCompetitiveRace to learn

case , the firm e nd ed i ts 20-year ventu re with Taiyo Fishery C om pan y an destabl ished a whol ly owned Japanese subsidiary. After gaining exper i -ence in Japan, Rals ton Purina 's management apparent ly decided that aJapanese par tner was no longer necessary.Alternatively, as the foreign partner learns about local market reali-t ies, it may decide to withdraw from the market. In contrast, the foreignparfner may seek a more prominent role in the management of fhe JV,lea din g to conflict over the division of confrol, which in turn could le ad toan unstable relationship. Finally, as we discuss in another section, theembedded his tory of pr ior relat ionships between the par tners may be astabil izing facfor fhaf counterbalances shiffs in bargaining power.Knowledge Acquisition by the Local Partner

In contrast to the foreign partner 's building of a local knowledgeba se , the local partner may acquire the skills of ifs foreign parfner, mak-ing the IJV redu nda nt, be ca us e the foreign partn er 's skil l-bas ed resourcecontributions are no longer needed. In their study of U.S.-China JVs, Yanand Gray (1994) found that for Chinese firms in IJVs, the overwhelminggoal in cooperating with Western firms was fo learn fhe more advancedWestern technology. The Chinese partners fully expected that manage-ment would shift from the foreigners to them. However, Yan and Gray(1994) con clud ed that the Ch ine se pa rtn er s did not significantly ga in b ar-gain ing pow er through kno wledge, beca use the U.S. par tn ers protectedevidence to support this argument. Many technology-licensing arrangem ents in Japan haveended this way, but these w ere not JVs designed to enter the Japanese market.As the Purina,Monsanto, Bayer, and Sandoz cases suggest, foreign firms in Japan are capable of makingtheir local partners obsolete. According to Jones and Shill (1993), the survival record of


13/27

1997 nipen a n d Beamish 189their tech no logie s. Also, the U.S. par tners were able to mainta in the or igi -nal balance of bargaining power by making addi t ional resource commit-ments to thei r technological capabi l i t ies .Even if local pa r tne rs have un hind ere d a cce ss to the foreign par tn er ' sski l l s , the knowledge required to el iminate a foreign dependency is usu-ally more difficult to acquire for the local partner than for the foreignpa rtn er. T he New U nited Motor M an ufactu ring Inc. (NUMMI) JV in Califor-nia between General Motors and Toyota i l lus t ra tes this point . BecauseNUMMI ha d be en m an ag ed by Toyota, Toyota 's ma n ag er s we re forced tolearn how to work with American workers and labor unions . Toyota hasdeployed i t s new knowledge in a whol ly owned plant in Georgetown,Kentucky. As descr ibed by a senior manager a t Toyota , that knowledgecame directly from Toyota's NUMMI experience (Sasaki, 1993).

He dlund (1994) d i s t in gu ishe d a m on g three forms of kno w ledge :(a) cognitive knowledge in the form of mental constructs, (b) skills, and (c)knowledge embodied in products and wel l -def ined services . Hedlundsugg este d that embod ied kno wle dge w as the eas ie s t to t ransfer, fol lowedby cognit ive knowledge, and skil ls were most diff icult to transfer. Skil ls ,such as complex eng ineer ing processes , a re h igh ly embedded in organi -zatio na l rou tin es (Nelson & W inter, 1982) an d, the refore, a re difficult toextract from another f irm. Also, the foreign partner can take explici t mea-sures to protect the transparency of i ts skil ls , part icularly if the skil lscomprise explici t knowledge held by a few "experts" (e.g. , Kentucky FriedChicken's secret recipe of herbs and spices). Local knowledge, such asinvestment regulat ions , suppl ier pract ices , labor laws, and cul tural t ra-di t ions , can be c lass i f ied pr imar i ly as cogni t ive knowledge. Cogni t ivekno wled ge is usual ly eas ier to t ransfer than a re ski l ls , be cau se it i s not assens i t ive to t eam e m be dde dn ess (Zander, 1991, cite d in H ed lun d, 1994). Inm an y ca se s, it is tacit kn ow ledg e (Polanyi, 1966) that is highly contactde pe nd en t; for IJVs, the context is the local en viron m en t. Be cause thisknowledge often comes packaged in the form of individuals (Hedlund,1994), foreign p artn ers ca n g ain acc es s to the kno w ledg e through activem an age r ia l involvement in the IJV proce ss . In other words , the exp er ien ceof the managers in an IJV is the key to acquiring this type of knowledge(Nonaka & Tak eu chi, 1995).

The skil ls provided by the foreign partner, such as how to manufac-ture high-precision products, wil l consist of a combination of both tacitand explici t , or codifiable, knowledge. The explici t knowledge that can beexpressed in schemata , d iagrams, and char ts i s rela t ively easy to t rans-fer, alth ou gh a s Zan der (1991) found, codifiability d oe s not n ec es sar ilylead to faster competi tor imitat ion. Because tacit and explici t knowledgeare m utually c om plem en tary (Nonaka & Take uchi, 1995), there w il l be as t rong taci t d imension associated wi th how to use and implement the


14/27

190 Academy of Management Review JanuaryEmpirical evidence supports the previous arguments. Hamel (1991)found that market intelligence was transfered between alliance partnersmore easily than knowledge of leading-edge manufacturing skills. The

local partne rs in Yan an d G ray 's (1994) study encountered significant dif-ficulty in learning from their U.S. partners . When successfully implem ent-ing a local strategy in the United S tates, Toyota's executives ne ed ed onlyto learn how to transfer an existing management process to NorthAmerica. For General Motors to reduce its dependency on Toyota,changes in very fundamental operating philosophies were required(Badaracco, 1991; Sasaki, 1993). The local knowledge that Toyota neededwas more accessible and easier to internalize than the knowledge basethat General Motors had to build. Finally, Hennart, Roehl, and Zietlow(1995) studied Jap an ese JVs in the U nited Sta tes an d found tha t w hen JVswere terminated, the more common scenario involved the Japanese part-ner acquiring the JV bus iness, an d by implication, the local know ledgenecessary to run the business.In summary, knowledge acquisition by either partner has the poten-tial to shift the ba lance of barga ining power that, in turn, could lead to theinitiation of changes in the partner relationship. This is illustrated inFigure 3. To the extent that the b ala nc e of power shifts sufficiently to lea dto instability, it is more likely that the underlying cause is knowledgeacquisition by the foreign partner compared to knowledge acquisition bythe local partner. We also expect that the "skilling" of the local partner(leaking the foreign partner's skills to the local partner) should be lesslikely in IJVs than in other types of alliances, most notably technology-sharing relationships or R&D alliances. This is becau se a JV is a separa teentity, in which the local partner may have difficulty penetrating theventure's boundaries to gain access to the foreign partner's skills. Theforeign partner may put up barriers to keep the local partner outsidespecific skill areas. In contrast, there are fewer boundaries in a technol-ogy-sharing relationship, which increases the risk that organizationalskills may be appropriated by a partner (for a discussion of learning inR&D alliances from a dom estic perspec tive, see Young & Oik, 1994). Thereis little question that particularly in the 1960s, many Japanese firms usedlicensing arrangements as a source for learning their partner's skills(Reich & Mankin, 1986).The Value of Local Knowledge

The preceding discussion focused on the effect of IJV stab ility posedwhen partner knowledge acquisition leads to changes in partner depen-dency. An equally important issue is why, in some IJVs, foreign partnersaggressively seek to acquire local knowledge, but in other IJVs, they donot. We examine this question by starting with a key assumption: Whenan IJV is formed, the contribution of local knowledge by the local partner


15/27

1997 nipen and Beamish 191mously, the foreign partner will continue to depend on its local partner.The foreign partner's valuation of local knowledge involves twostages. The first stage occurs prior to the formation of the IJV when theforeign partner considers the value associated with gaining access to apotential partner's local knowledge. If gaining access to a partner's localknowledge satisfies the foreign partner's resource requirements, an IJVmay result. The second stage occurs after the formation of the IJV, whenthe foreign partner gains access to local knowledge. The question for theforeign partner now becomes: Is access sufficient, or should the knowl-edge be acquired?

A foreign firm involved in an IJV can choose what resources and howmuch it can devote to gaining local knowledge acquisition. Acquiringlocal knowledge enables a foreign partner to reduce its commitment tocollaboration. If a foreign p artner p laces high strategic value on access toand acquisition of local knowledge, shifts in bargaining power becomemore likely, bec ause it is likely tha t the foreign partner will not be contentwith a cce ss alone. Thus, Figure 2 shows a direct relationship between thevaluation of local knowledge and shifts in the balance of the bargainingpower.

Proposition 2: The strategic value of local knowledge tothe foreign partner w ill be positively related to shifts inIJV bargaining pow er.Proposition 2 is a refinement of Proposition 1. Proposition 1 suggeststhat the foreign partner's knowledge of local conditions will inevitablyincrease over time, simply bec ause the foreign partner is in the host coun-try. This process represents passive knowledge acquisition. Proposition 2implies active knowledge acquisition by the foreign partner that can shiftthe JV bargaining power away from the local partner.Why would a foreign partner actively seek to acquire local knowl-edge? The critical factor is the strategic intent of the foreign partner.Firms often utilize IJVs as an entry mode in an effort to balance theirdesire for international product/market diversification and growth withtheir perceived need for control. The subsequent desire to acquire localknowledge will be driven by various strategic factors. The decision toacquire local knowledge may reflect a desire to increase control over the

IJV because of a shift in the strategic importance of the venture (Harrigan& Newman, 1990) or because of the increased importance of the localmarket. Alternatively, a foreign partner's increased concerns about leak-ing proprietary technology to the local partner may result in its acquiringlocal knowledge.A partner's decision not to acquire local knowledge may reflect thecost of this acquisition. Previously, we used the example of KentuckyFried Chicken. In order for Kentucky Fried Chicken to eliminate a depen-


16/27

192 Academy of Management Review Januaryw as p ossible , it could be very expensive, and there w as no gu ara nte e ofsuccess. Finally, acquiring local knowledge may be in response to theforeign partn er 's recognit ion that the original IJV ag ree m en t w as a n ob-sole scin g ba rg ai n (Vernon, 1977), be ca us e the local pa rtn er h a s little tocontr ibute on an ongoing basis .Factors Influencing Acquisition of Local Knowledge

When a foreign partner has a strategic objective of acquisi t ion andproprietary control over local knowledge, the speed of knowledge acqui-si t ion that is necessary to shift bargaining power will be influenced bytwo key factors: (a) the foreign partner's effectiveness in acquiring localknowledge and (b) the init ial resource contributions of the partners.Effectiveness in acquiring local know ledge. Although a foreign part-ner 's acquisi t ion of local knowledge can occur passively over t ime, ex-plicit efforts to acquire this knowledge are of greater interest. This leadsto the premise that knowledge acquis i t ion is an organizat ional processthat can be m an ag ed by the foreign par tner (Hedlund & No naka, 1993;No naka, 1994). The process of acqu iring kno wled ge that o rigin ates withan IJV pa rtn er is, in effect, th e proc ess of creati ng o rga niz atio na l kn ow l-ed ge (Nonaka, 1994). Cre ating orga nizatio nal kno wle dge is not simply amatter of learning from others or acquiring knowledge from outside(Nonaka & Tak euch i, 1995). It is a comp lex orga niz atio na l pr oc ess involv-ing various organizational levels and actors. Through this process, spe-cific knowledge becomes amplified throughout the organization. To cap-ture the dynamic movement of knowledge across var ious organizat ionallevels, Non aka (1994) de ve lop ed the conc ept of a spir al of kno w ledg e cre-ation. In the spiral , knowledge moves upward in an organization; i t be-gins at the individual level , moves to the group level , and finally movesto the f irm level . As the knowledge spirals upward in the organization,individuals interact with each other and with their organizations.

In an IJV, m an ag ers a ss ign ed to the venture from th e foreign pa ren twill be exposed to various aspects of local knowledge. Clearly, thesemanagers wi l l develop new ideas about conduct ing business in the localenvironment. However, in order for that local knowledge to be internal-ized at the parent organizational level , there must be knowledge connec-t ions among the various organizational levels to create the potential forindiv idua ls to sh are their observ ations a nd exp erien ces (Von Krogh, Roos,& Slocum, 1994). The se kno wled ge conn ections pro vide a ba sis for tran s-forming individual know ledge into organizat ional know ledge. An organi-zation's set of internal managerial relat ionships facil i tates the sharingand communicat ing of new knowledge.Although the process of knowledge creation is nonlinear and inter-active, specific organizational condit ions can be identif ied that promotethis pro ces s (Nevis, DiBella, & Gould, 1995; N ona ka & Ta keu chi, 1995).


17/27

1997 Inkpen and Beamish 193

the ability of firms fo acquire knowledge associated with their IJVs (Ink-pen, In press). These factors include partner intention, top managementcommitment, and tolerance for information redundancy. The absence ofthese facfors creates obstacles to knowledge creation. Although a de-tailed examination of fhese factors is beyond the scope of this article, it isimportant to recognize that the foreign partner 's effectiveness in acquir-ing knowledge will influence ifs bargaining power. If a foreign parfner isineffective at knowledge creation (and more specifically af acquiring lo-cal knowledge), i t will not be able to reduce i ts dependency on the localpar tner .

Proposition 3: There will be a positive relationship be-tween the foreign partner's effectiveness in acquiringlocal knowledge and shifts in bargaining power.Resource contributions of the partners. W hen a JV is formed, th e p art-ne rs must al loca te respon sibil i ty for vario us decision a re as (G eringer &

Heberf, 1989; Killing, 1983). This allocafion of responsibilify is the basis fordetermining the resources and capabi l i t ies commit ted by the par tners toth e JV (Yan & Gray , 1994). Th ese reso urce co ntribu tions a re ne go tiate d bythe partners, and they are an important factor in determining the init ialbalance of power .In a n IJV, fhe loca l pa rtn er is usu ally e xp ecte d to con tribute imp ortantlocal kno wled ge. However, this do es not m ean that the foreign parfnercannof contribute resources associated with local management. For ex-am ple, in ma ny of fhe Japa nese-A me rican JVs formed to supply U.S.-basedautomotive t ransp lants , the Jap ane se par tn ers control led local m arket ingfunctions (Wom ack, 1988). The refore, fhe Am erica n parfn ers co ntrib ute dlocal knowledge confined fo areas such as legal process, workforce man-agement , and some suppl ier management . Without control over localmarketing, American partners were in a vulnerable posit ion should theirJap an es e pa rtn ers seek to acq uire furfher local kno wled ge.Init ial resource contributions of fhe partners, as determined by theallocation of sfrategic responsibil i t ies, will play an important role in sta-bility of fhe IJV, b ec a u se of fheir influe nce on fhe com plexify of the foreignpar tne r ' s know ledge acq uis i t ion task. By rel inquishing local responsibi l i-ties to the foreign parfner when fhe venture is formed, the local partnerprovides the foreign partner with an incentive to gain complete localknowledge and, hence, to increase i ts bargaining power. The acquis i t ionof local know led ge by the foreign pa rtn er a lso is simplified if fhe JV sfartsout with the responsibility for key local knowledge belonging fo the for-eign par tner .

Thus, we predict that initial resource contributions made to an IJVwill influence sub sequ enf shifts in ba rg ain ing pow er. It h as alre ad ybeen established that firms will differ in how effecfive fhey will be af


18/27

194 Academy o Management Review J a n u a r yknow ledge-acquis i t ion task nec essar y to redu ce a partn er depen den cy. Ifthe foreign partner views the task as complex, i t may be dissuaded fromacquir ing addi t ional knowledge. In contras t , i f the foreign partner viewsthe task as relat ively simple, i t is l ikely that i t will acquire more knowl-ed g e , and l ikelihood for shifts in bargaining power will be greater. Thus,the foreign partner 's responsibil i ty for local knowledge will be related tothe l ikelihood of shifts in bargaining power.

Proposition 4: The grea ter the responsibility for localknow ledge initially allocated to the foreign partner, them ore likely that the foreign partner w ill seek add itionallocal know ledge from the local partner an d the m orelikely that bargainin g power will shift.Regardless of the speed of local knowledge acquisi t ion, i t is impor-tant to emphasize that when a foreign partner has the explicit intent togain knowledge, the local partner is in a vulnerable posit ion. In IJVs indeveloping countries, the vulnerabil i ty of the local partners can be high ifthe local partner is not a competit ive threat to the foreign partner outsidethe IJVs country of domicile. In this case, the foreign partner is usuallylarger and has greater in ternat ional scope; therefore , local partners maynot be able to prevent a foreign partner from increasing i ts local knowl-ed g e . It is also unrealist ic to expect foreign partners to complacentlyignore the opportunity to increase their local knowledge. However, asDy msza (1988) no ted, the foreign firm's contribu tion m ay b eco m e les s im-portant over time if the firm from the developing country is committed toincreasing i ts skil ls .Furthermore, if the value of local knowledge declines, in terms ofeither access or acquisi t ion, the foreign partner may choose to end theventu re . For exam ple, local know ledge ma y decl ine in value if ch an ge s inlocal regulat ions open up a previously closed economy to greater compe-tition.

Attachment Between the PartnersAttachment is the binding of one party to another (Salancik, 1977).At tachment between partners develops through experience in the col-laborat ive rela t ionship and through investments the partners make in therela t ion ship over t ime (Seab right et al. , 1992). W hen the pa rtn er s ha vedeveloped a strong attachment, inert ial forces may block the pressures forch a n ge in the re lat io n shi p (Blau, 1964; Sa lan cik, 1977). If fi rms ha veworked together in the past , they wil l have a basic understanding aboutea ch other 's skil ls a n d ca pa bil i t ies (Heide & Miner, 1992). The pa rtn er s

ma y have developed com mitments to eac h other be ca us e of a re la t ion shipthat existed prior to forming the JV. According to Parkhe (1993a: 803), "The


19/27

1997 nJcpen an d Beamish 195Because of prior relationships, firms often form JVs with firms with whomthey ha v e tra ns ac ted in the pa st. For ex am ple , in a stud y of 40 IJVs, Inkpen(1995a) found that in 24 cases, the partners had previously worked to-gether. An MNE's knowledge about i ts local partner, generated throughprevious interactions, also can play an important role in determining i tsequity position in the host country (Sohn, 1994).Attachments in collaborative relat ionships may be the result of indi-vidual or structural ties that reflect the prior history of the relationship(Seab right et al., 1992). Ind ividu al atta ch m en t reflects the socializa tion byindividuals dur ing thei r involvement in exchange act ivi t ies . Seabr ightan d collea gu es (1992) sug ge sted th at indiv idua l at ta ch m en ts are impor-tant early in a relat ionship but that they diminish in significance as therelat io nsh ip persis ts . In IJVs, indiv idua l at tachm ent m ay be rep rese ntedby persona l re la t ionsh ips be tween par tners ' managers . Such managersmay ini t ia te a re la tionship b ase d on personal kn owledge a nd t rus t that inthe early y ea rs of the JV provides a buffer against the normal pressures ofcollaboration. It is comm on in IJVs for a p artn er 's sen ior m an ag er s to tak ean a ctive interest in the JV proc ess an d m ake a comm itment to workingwith top managers from the other partner(s). Ring and Van de Ven (1994)suggested that personal bonds of fr iendship can lead to norms of groupinclusion and escalate the commitment by part ies to a cooperative rela-t ionship. They proposed that the l ikelihood of termination of interorgani-zat ional re la t ionships decreased over t ime, because economic exchangesbecome t ransformed into social ly embedded re la t ionships .Individual at tachments are closely related to the tenure of individualbo un da ry-s pan nin g m an ag er s involved in the IJV (Seabright et al ., 1992).A high turnover of ma na ge rs ca n le ad to a loss of relat ionsh ip continuityand a reduction of individual at tachment. Structural at tachment reflectsthe history of organizational investments made since the formation of theIJV (Seab right et al., 1992). Struc tural a tta ch m en ts sh ou ld in cre ase withthe duration of the relat ion ship a s formalized proc edu res rep lace som e ofthe co ordin ation of IJV activ ities throu gh p ers on al con tact (Van de Ven,1976).Con t inuing busin ess re la t ionsh ips often becom e over la id wi th socialcontent that generates s t rong expectat ions of t rus t and forbearance(Grano vetter, 1985). Thu s, atta ch m en t c an lea d to the formation of JVs thathave an exist ing stock of "relat ionship assets" (Fichman & Levinthal, 1991)a n d a hig h d eg re e of inte rpa rtne r trust (Gu lati, 1995). Pa rkh e (1991) sug -ges ted that the un pla nn ed termin ation of an IJV is more l ikely w hen firmsare working together for the first time. Kogut (1989) found that structuralt ies between JV par tn ers were negat ively re la ted to JV dissolution. Kogut'svariable for structured t ies was a composite of three types of relat ion-ships: supply, other JVs, and l icensing agreements. Thus, we predict thatattachment exerts a main effect on the likelihood of instability.


20/27

198 Academy o Management Review January

Proposition 5: The greater the individual or structuralattachment between the I}Vpartners, the lower the like-lihood of IJV instability.The central theme of this article is that shifts in partner bargainingpower are the underlying ca use of instability in IJVs. As a firm acquiresknowledge that reduces its dependency on its IJV partner, the likelihood ofinstability increases. As we argued previously in the case of the foreignpartner, knowledge of local conditions will inevitably increase over time,simply beca use the foreign partner is now located in the host country. AnIJV partner also may actively e nga ge in acquiring knowledge, which re-sults in reduced partner dependency and a shift in bargaining power.However, because of attachment, the shift in bargaining power associ-ated with knowledge acquisition will notnecessarily lead to instability in

IJVs.Blau (1964) and Cook (1977) argued that attachment can lead to main-taining a relationship tha t provides fewer of the needed resources than itoriginally did. Seabright and colleagues (1992: 123) suggested that attach-ment "constitutes a counterforce to change rather than a pressure forc h a n g e . . . . In this manner, attachment may attenuate the effects ofchanges affecting resource fit on the likelihood of a relationship dissolv-ing." This discussion suggests that attachment can modera te the effects ofshifts in bargaining power. Thus, the effect of shifts in bargaining poweron the likelihood of instability decreases as attachment increases.Proposition G: The greater the individual and structuralattachment between IJV partners, the lower the effect ofshifts in bargaining pow er on the likelihood o f IJV insta-bility.The argument that relationship investments and attachments canmoderate the forces of shifts in bargaining power may appear counter tothe previous discussion on resource fit and the termination of relation-ships for the inverse of the reasons for which they were formed. IJVsinvolve many layers of interfirm complexity (Harrigan & Newman, 1990).

Our view is that in any IJV, there will be both resource fit and attachmentimplications, which may exert opposing forces on the stability of the re-lationship. The forces may lead to what we term dormant instability.^When the foreign partner acquires sufficient local knowledge to destabi-lize the IJV, a situation of dormant instability (or superficial stability fromthe opposite perspective) results. Thus, the foreign partne r may choose toreorient or restructure the relationship, or, because of attachment, it maybe willing to let the potential instability lie dormant. In any event, oncethe local knowledge is acquired by the foreign partner, the balance ofpower between the partners has shifted. This dormant instability mayultimately have a destabilizing impact on the IJV, bec aus e both partn ers


21/27

1997 JnJcpen and Beamish 197now unders tand that the foreign partner 's voluntary dependence on thelocal partner 's knowledge contributions may not las t . This s i tuat ion a lsomay lead to resentment by the local partner, which in itself is a source ofins tabil i ty .

Attachment and the valuation of know ledge. Our proposed frameworkincor pora tes th e dyn am ic na tu re of IJVs. A foreign p artn er's v alua tion oflocal knowledge will be influenced by both strategic factors and by thestrength of the partner re la t ionship . This valuatio n is repre sented in Fig-ure 2 by the feedback loop from a ttac hm ent to the valu ation of localknowledge by the foreign partner. Attachment is the result of an evolvinghistory of partne r inter action s. W hen an IJV is formed, atta ch m en ts maybe present that may have developed either through prior interfirm rela-tionships or during the negotiation stage. As the duration of an IJV in-crea ses , s tructural a t tach m ents wil l l ikely continue to develop. Individuala t tachments may increase or decrease , depending on the continuity ofm an ag er s involved in the IJV. As the level of atta ch m en t ch an ge s, wepredict that this shift will influence the foreign partner's evaluation of thestra tegic value associa ted with i ts partner 's local knowledge.Proposition 7: Changes in individual and s tructurai at-tachment between I}V paitneis will lead to updating othe ioieign paitnei's valuation o local knowledge.

Managerial ImplicationsFor a foreign firm that is inter ested in m ain tai nin g a sta ble , long-termIJV relationship, this article has several implications. First, explicit at-tem pts to build kno wle dge of the local country by capitaliz ing on the localpartner 's experience are usually transparent and may be interpreted bylocal par tne rs as comp eti tive ra ther tha n collaborative in nature . Second,IJV pa rtn ers m ay h av e to let their IJV dev elop its own c ulture a nd system s.General Electric has followed this strategy with its successful JapaneseJVs. GE has leveraged i ts Japanese partners ' names to recruit personnelfor its JVs and allowed its JVs to develop into independent companies

without interference from the parents in day-to-day management (Turpin,1993). Third, continuity in the persona l re la t ion ships b etw een the top m an-agement of both partners can be an important source of a t tachment be-tween the partners. Without continuity, there is the risk of "corporateamnesia ," whereby managers in the parent companies forget their origi-nal motivation for this alliance and the past lessons from their relation-ship (Turpin, 1993).Top pan M oore, a JV bet w een Top pan Printing of Japan a nd MooreCorporation (Moore) of Canada, provides a good example of a stable,lon g-l asti ng co llab ora tion (Beam ish & M akino, 1994). W ith 1994 rev en ue sof more than $1.5 billion and more than 3,000 employees, Toppan Moorewas Japan 's largest company that produced business forms and i ts third


22/27

198 Academy ofManagement R e v i e w January

product technologies, and the Japanese par tner assumed ini t ia l respon-sibility for sales, distr ibution, and local marketing support .^ Over t ime,the JV modified produ cts to m eet the req uirem ents of Japanese cus tomersand developed its own product ion capabi l i t ies , reducing its day-to-daydependence on Moore. Moore maintained its barga in ing power by con-tinuing to make technological and marketing contributions. Although theJapanese par tner ' s immedia te opera t iona l need for Moore lessened, ac-cess to fufure technological developments remained an important aspectof fhe relafionship. Consequenfly, the a d v a n t a g e s of collaborafion forMoore and Toppan Prinfing continued tooutweigh the need for competi-tion.Local partners and the control of instability. When a local partner isinteres ted in stability, it should consider several ideas associated wi th

con trolling IJV stab ility an d m inimizing the foreign parfn er's acq uis itionof local knowledge. First, the local partner can discourage the foreignparfner from sending large numbers of m a n a g e r s to the venture. As weindicated previously, knowledge creation in organizat ions begins wi thindividuals . The greater the number of foreign managers at the IJV, thegreater the foreign partner 's access to local knowledge. Second, the localpartn er can continue to actively invest in local know ledge . In other w ords,the local partner should not treats i ts knowledge contributions passively,but it should cont inue toupgrade the va lue of the knowledge to the for-eign partner. Third, the local partner can consider the t rack records ofpotential partners, part icularly how long other IJVs that they were a par tof have survived. This record should provide an indicafion of the foreignparfner 's expectations regarding the role of the local partner. Finally, thelocal partner can ba rga i n for greater responsibil i ty in m a n a g i n g the JVand, in effect , increase the importance of its role in the venture ' s success .

CONCLUSIONWhen JVs are formed fo exploif inferfirm differences in skills, there isa l w a y s the r isk that one par tner may acquire knowledge that it lackedwhen the alliance was formed (Parkhe, 1993b). In fact, many firms enterJVs with the objective to ga in exp licit kn ow led ge (Ham el, 1991; Inkpe n,1995a). W hen k no w led ge ac qu isitio n shiffs fhe b a l a n c e of ba r ga i n i ngpower between par tners , the cooperat ive basis for fhe JV m ay e rode, an dventure instabil i ty may result . Even though this argument applies to anyJV in which par tner knowledge contr ibut ions create a dependency , thefocus in this art icle was on IJVs and the specific nafure of local knowl-edge . When a JV is infernational and the foreign partner seeks to expand

'' Given the difficulty in penetrating Japanese markets, it is unlikely that Moore was ina position to control sales and distribution in the JV when it was formed. However, the


23/27

1997 n ipen and Beamish 199

its geographic scope of operations, the local partner's knowledge of localeconomic, political, and cultural environments will be a key contributionto the JV. Our main argument is that once the venture is formed, if theforeign partner attaches a high value to the acquisition of local knowl-edge and has the ability to acquire the knowledge, theprobability of JVinstability increases. Once the foreign partner has acqu ired local knowl-edge, unless the local partner is contributing other valuable and nonimi-table skills to the JV, the rationale for cooperation will be eliminated.Instability may be the result, although attachment between the partnersmay moderate the shifts in bargaining power. Thus, the acquisition oflocal knowledge is an enabling device for the foreign partner to operateautonomously.Although some ven tures are formed inwhich the partners ag ree on atermination plan , this article was focused on instability as an undesirableevent for one or all partners. Given its undesirability, if a local partnertakes steps to ensure that its role enco mpasses m ore than simply contrib-uting local knowledge, instability may be controllable. A foreign partneralso may choose as a viable strategy access of local knowledge ratherthan acquisition of the knowledge. Consequently, IJVs can be stable andsustainable arrangements for creating competitive adv antag e.

REFERENCESBadaracco, J. 1991. The knowledge ink. Boston: Harvard Business School Press.Bacharach, S., & Lawler, E. J. 1980. Power and politics in organizations. San Francisco: Jossey-

Bass.Bartlett, C. A., & Ghoshal, S. 1989. Managing across borders: The transnational solution.

Boston: Harvard Business School Press.Beamish, P. W. 1988. Multinational joint ventures in developing countries. London: Rout-

ledge.Beamish, P. W., & Makino, S. 1994. Toppan Moore. In P. W. Beamish, J. P. Killing, D. Lecraw,

& A. Morrison (Eds.), International management (2nd ed.): 388-402. Burr Ridge, IL: Irwin.Blau, P. M. 1964. Exchange and power in social life. New York: Wiley.Bleeke, J., & Ernst, D. 1991. The way to win in cross-border alliances. Harvard Business

Review. 69(6): 127-135.Blodgett, L. L. 1992. Factors in the instabili ty of international joint ventures: An event history

analysis. Strategic Management Journal, 13: 475-481.Contractor, F. J., & Lorange, P. 1988. Why should firms cooperate: The strategy and economics

basis for cooperative ventures. In F. Contractor & P. Lorange (Eds.), Cooperative strate-gies in international business: 3-30. Lexington, MA: Lexington Books.

Cook, K. 1977. Exchange and power in networks of interorganizational relations. SociologicalQuarterly. 18: 62-82.

Daft, R. L., & Weicke, K. E. 1984. Toward a model of organizations a s interpretation systems.Academy ot Management Review. 9: 284-295.


24/27


Doz, Y., Prahalad, C. K., & Hamel, G. 1990. Control, change, and flexibility: The dilemma oftransnational collaboration. In C. Bartlett, Y. Doz, & G. Hedlund (Eds.), Managing theglobal firm: 117-143. London: Routledge.

Dymsza, W. A. 1988. Successes and failures of joint ventures in developing countries: Lessonsfrom experience. In F. Contractor & P. Lorange (Eds.), Cooperative strategies in interna-tional business: 403-424. Lexington, MA: Lexington Books.

Emerson, R. M. 1982. Power dependence relationships. American Sociological Review,27(February): 31-41.Fichman, M., & Levinthal, D. A. 1991. Honeymoons and the liability of adolescence: A new

perspective on duration dependence in social and organizational relationships. Acad-emy of Management Review, 16: 442-468.

Franko, L. 1971. Joint venture survival in multinational companies. New York: Praeger.Geringer, J. M., & Hebert, L. 1989. Control and performance of international joint ventures.

Journal of International Business Studies, 20: 235-254.Granovetter, M. 1985. Economic action and social structure: The problem of embeddedness.

American Journalof Sociology, 78: 481-510.Gulati, R. 1995. Does familiarity breed trust? The implications of repeated ties for contractual

choice in alliances. Academy of Management Journal, 38: 85-112.Hamel, G. 1991. Competition for competence and inter-partner learning within international

strategic alliances [Special Issue]. Strategic Management Journal, 12: 83-104.Hamel, G., Doz, Y. L., & Prahalad, C. K. 1989. Collaborate with your competitorsand win.

Harvard Business Review, 67(1): 133-139.Harrigan, K. R. 1986. Managing for joint venture success. Lexington, MA: Lexington Books.Harrigan, K. R., & Newman, W. H. 1990. Bases of interorganizational cooperation: Propensity,

power, persistence. Journal of Management Studies, 27: 417-434.Hedlund, G. 1994. A model of knowledge management and the N-form corporation. Strategic

Management Journal, 15: 73-90.Hedlund, G., & Nonaka, I. 1993. Models of knowledge management in the West and Japan. In

P. Lorange, B. Chakravarthy, J. Roos, & A. Van de Ven (Eds.), Implementing strategicprocesses: Change, learning, and cooperation: 117-144. Oxford: Basil Blackwell.

Heide, J. B., & Miner, A. S. 1992. The shadow of the future: Effects of anticipated interactionand the frequency of contact on buyer-seller cooperation. Academy of ManagementJournal,35: 265-291.

Hennart, J. F. 1991. The transactions cost theory of joint ventures: An empirical study ofJapanese subsidiaries in the United States. Management Science, 37: 483-497.

Hennart, J. F., Roehl, T., & Zietlow, D. S. 1995."Trojan horse" or "workhorse"? The evolution ofU.S.-Japanese joint ventures in the United StatesRevised. CIBER Working paper 95-103,College of Commerce and Business Administration, University of Illinois, Urbana-Champaign, IL.

Huber, G. P. 1991. Organizational learning: The contributing processes and a review of theliteratures. Organization Science, 2: 88-117.

Inkpen, A. C. 1995a. The management of international joint ventures: An organizationalearning perspective. London: Routledge.

Inkpen, A. C. 1995b. Organizational learning and international joint ventures. Journal of


25/27

1997 Inkpen and Beamish 201ventures. In P. W. Beamish & J. P. Killing (Eds.), Cooperative sfraiegies; Worth Americanperspectives. San Francisco: New Lexington Press.

Inkpen, A. C, & Crossan, M. M. 1995. Believing is seeing: Joint ventures and organizationlearning. Journal o Management Studies, 32: 595-618.

Jones, K. K., & Shill, W. E. 1993. Japan: Allying for advantage. In J. Bleeke & D. Ernst (Eds.),Collaborating to compete; Using strategic alliances and acquisitions in the global mar-ketplace: 115-144. New York: Wiley.

Killing, J. P. 1983. Strategies /or joint venture success. New York: Praeger.Kogut, B. 1988. Joint ventures: Theoretical and empirical perspectives. Strategic Management

Journal 9: 319-322.Kogut, B. 1989. The stability of joint ventures: Reciprocity and competitive rivalry. Journal o

Industrial Economics. 38: 183-198.Makino, S. 1995. Joint venture ownership structure and periormance: Japanese joint ventures

in Asia. Unpublished doctoral dissertation. University of Western Ontario, London, ON.Nelson, R. R., & Winter, S. G. 1982. An evolutionary theory o economic change. Cambridge,

MA: Harvard University Press.Nevis, E. C, DiBella, A., & Gould, J. M. 1995. Understanding organizations as learning sys-

tems. Sloan Management Review. 23(2): 73-85.Nonaka, . 1994. A dynamic theory of organizational knowledge. Organization Science, 5:

14-37.Nonaka, 1., & Takeuchi, H. 1995.The inowledge-creating company: HowJapanese companiescreate the dynamics o innovation. New York: Oxford University Press.Ono, Y. 1991. Borden's breakup with Meiji Milk shows how a Japanese partnership cancurdle. Wall Street Journal February 21: Bl, B8.Parkhe, A. 1991. Interfirm diversity, organizational learning, and longevity in global strategic

alliances, /ournal o International Business Studies. 22: 579-602.Parkhe, A. 1993a. Strategic alliance structuring: A game theoretic and transaction cost ex-

amination of interfirm cooperation. Academy o Management Journal 38: 794-829.Parkhe, A. 1993b. "Messy" research, methodological predispositions, and theory develop-

ment in international joint ventures. Academy o Management Review. 18: 227-268.Pfeffer, J. 1981. Power in organizations. New York: Pitman.Pfeffer, J., & Salancik, G. R. 1978. The external control o organizations: A resource depen-

dence perspective. New York: Harper and Row.Polanyi, M. 1966. The tacit dimension. London: Routledge & Kegan Paul.Porter, M. E. 1990. The competitive advantage o nations. New York: Free Press.Pucik, V. 1991. Technology transfer in strategic alliances. Competitive collaboration and

organizational learning. In T. Agmon & M. A. Von Glinow (Eds.), Technology transier ininternational business: 121-138. New York: Oxford University Press.

Reich, R., & Mankin, E. 1986. Joint ventures with Japan give away our future. Harvard BusinessReview. 84(2): 78-86.

Ring, P. S., & Van de Ven, A. 1994. Developmental processes of cooperative interorganiza-tional relationships. Academy o Management Review. 19: 90-118.

Salancik, G. R. 1977. Commitment and the control of organizational behavior and belief. In


26/27


Sasaki, T. 1993. What the Japanese have learned from strategic alliances. Long Aange PJan-ning, 26(6): 41-53.

Schelling, T. C. 1956. An essay on bargaining. American Economic Review, 48: 281-306.Seabright, M. A., Levinthal, D. A., & Fichman, M. 1992. Role of individual attachments in the

dissolution of interorganizational relationships. Academy of Management Journal, 35:122-160.

Simonin, B. L., & Helleloid, D. 1993. Do organizations learn? An empirical test of organiza-tional learning in international strategic alliances. In D. Moore (Ed.), Academy of Man-agement Best Paper Proceedings: 222-226.

Sohn, J. H. D. 1994. Social knowledge as a control system: A proposition and evidence fromthe Japanese FDI behavior. Journal of International Business Studies, 25: 295-324.

Stopford, J. M., & Wells, L. T. 1972. Managing the multinational enterprise. New York: BasicBooks.

Templin, N. 1995. Strange bedfellows: More and more firms enter joint ventures with bigcompetitors. WaJJ Street Journal, November 1: Al, A8.

Turpin, D. 1993. Strategic alliances with Japanese firms: Myths and realities. Long RangePlanning, 26(5): 11-16.

Van de Ven, A. H. 1976. On the nature, formation, and maintenance of relations amongorganizations. Academy of Management Review, 1: 24-36.

Vernon, R. 1977. Storm over multinationals. Cambridge, MA: Harvard University Press.Von Krogh, G., Roos, J., & Slocum, K. 1994. An essay on corporate epistemology. Strategic

Management Journal, 15:53-71.Westney, D. E. 1988. Domestic and foreign learning curves in managing international coop-

erative strategies. In F. Contractor & P. Lorange (Eds.), Cooperative strategies in inter-nationaJ business: 339-346. Lexington, MA: Lexington Books.Womack, J. P. 1988. Multinational joint ventures in motor vehicles. In D. Mower (Ed.), Inter-

national collaborative ventures in U.S. manufacturing: 301-348. Cambridge, MA: Ball-inger.

Yan, A., & Gray, B. 1994. Bargaining power, management control, and performance in UnitedStates-China joint ventures: A comparative case study. Academy of Management Jour-nal, 37: 1478-1517.

Young, C, & Oik, P. 1994. Why dissatisfied members stay and satisfied members leave:Options available and embeddedness mitigating the performance commitment rela-tionship in strategic alliance. Academy of Management Best Papers Proceedings: 57-61.

Zander, U. 1991. Exploiting a technological edge7oJuntary and invoJuntary disseminationof technology. Unpublished doctoral dissertation. Institute of International Business,Stockholm.

Andrew C. Inkpen received his Ph.D. from the University of Western Ontario. He is anassistant professor of management at Thunderbird, The American Graduate Schoolof International Management. His current research interests include the manage-ment of knowledge in strategic alliances, trust in international joint ventures, andknowledge transfer in multinational corporations.Paul W. Beamish received his Ph.D. from the University of Western Ontario. He is theRoyal Bank Professor of international business at the Richard Ivey School of Busi-ness, University of Western Ontario. He currently serves as Editor-in-Chief of the


27/27

Copyright of Academy of Management Review is the property of Academy of Management and its content may

not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written

permission. However, users may print, download, or email articles for individual use.

inkpen beamish

Documents