Absorptive Capacity, Learning, And Performance in International Joint Ventures_Lane_SMJ.2001

8/10/2019 Absorptive Capacity, Learning, And Performance in International Joint Ventures_Lane_SMJ.2001

1/23

Strategic Management JournalStrat. Mgmt. J. , 22 : 11391161 (2001)

DOI: 10.1002/smj.206

ABSORPTIVE CAPACITY, LEARNING, ANDPERFORMANCE IN INTERNATIONAL JOINTVENTURESPETER J. LANE 1 *, JANE E. SALK 2 and MARJORIE A. LYLES 31 College of Business, Arizona State University, Tempe, Arizona, U.S.A.2 Department of Strategy and Management, Groupe ESSEC, Cergy-Pontoise, France 3 Kelley School of Business, Indiana University, Indianapolis, Indiana, U.S.A.

This paper proposes and tests a model of IJV learning and performance that segments absorptive

capacity into the three components originally proposed by Cohen and Levinthal (1990). First,trust between an IJVs parents and the IJVs relative absorptive capacity with its foreign parent are suggested to inuence its ability to understand new knowledge held by foreign parents.Second, an IJVs learning structures and processes are proposed to inuence its ability toassimilate new knowledge from those parents. Third, the IJVs strategy and training competenceare suggested to shape its ability to apply the assimilated knowledge. Revisiting the Hungarian IJVs studied by Lyles and Salk (1996) 3 years later, we nd support for the knowledgeunderstanding and application predictions, and partial support for the knowledge assimilation prediction. Unexpectedly, our results suggest that trust and management support from foreign parents are associated with IJV performance but not learning. Our model and results offer anew perspective on IJV learning and performance as well as initial insights into how thoserelationships change over time. Copyright 2001 John Wiley & Sons, Ltd.

Over the past two decades there has been a surgein the number of international joint ventures (IJVs)as rms have strived to become more globallyinterconnected (Bleeke and Ernst, 1993; Hennart,1991; Lorange and Roos, 1993; Mjoen and Tall-man, 1997). Although rms form IJVs for a varietyof reasons, interorganizational learning has becomean important rationale for their creation (Hamel,1991; Kogut, 1988; Lyles, 1988; Parkhe, 1991;Pucik, 1991). While an IJVs parents often seek to learn from one another, IJV learning from itsforeign parents is considered to be essential forits survival and, hence, the realization of the par-ents strategic goals (Hennart, 1991; Parkhe, 1991;Inkpen and Beamish, 1997; Porter, 1986; Pearceand Branyiczki, 1997; Yan and Gray, 1994).

The research on IJV learning to date has beenlargely theoretical or case-based, and has primarily

Key words: international joint ventures; absorptive capac-ity; learning; knowledge; performance*Correspondence to: Peter J. Lane, College of Business, ArizonaState University, P.O. Box 874006 Tempe, AZ 85287-4006,U.S.A.

focused on identifying organizational characteris-tics and processes that may inuence IJV learningfrom foreign parents. In the lone empirical studyof this type of learning, Lyles and Salk (1996)found that acquiring knowledge from foreign par-ents requires a exible IJV organization, the provi-sion of training, technology, and managerial assis-tance by foreign parents as well as written businessgoals and plans for the IJV. Inkpens (1997) casestudies of another set of IJVs provides evidenceof similar antecedents to learning. Lyles and Salk (1996) also advanced the IJV literature by testingthe common assumption that the more knowledgelearned from foreign parents, the better the IJVsperformance. They found support for this acrossseveral different measures of IJV performance.

While Lyles and Salks (1996) study providesimportant theoretical and empirical insights, it hasthree limitations. First, the average age of theirIJVs is less than 5 years. This is not surpris-ing as they examined Hungarian IJVs only a fewyears after that country began to transform itseconomy. However, the performance impact of

Copyright 2001 John Wiley & Sons, Ltd. Received 3 August 1998

Final revision received 30 March 2001


2/23

1140 P. J. Lane, J. E. Salk and M. A. Lyles

new knowledge from foreign parents may decreaseover time. This could occur because the IJVbecomes more competent, because its local man-agers become more condent in their own admin-istrative heritage, or because the factors underpin-ning learning and performance change over time asonce young IJVs overcome the liability of newness(Freeman, Carroll, and Hannan, 1983). Second,Lyles and Salks data are cross-sectional in natureand they only assessed associations among staticcharacteristics. There could be a time lag betweenacquisition of knowledge and an IJVs ability tofully comprehend and utilize it. Thus, a more com-plete understanding of foreign parent-IJV learning

requires examining IJV learning and its outcomesover time, and for older, as well as younger, IJVs.Third, they tested the relationship between knowl-edge learned from foreign parents and IJV perfor-mance without considering the context in whichthat knowledge is used. New knowledge may havelittle impact on performance if the IJVs strategyor organization do not permit its application.

This paper seeks to address these theoreti-cal and empirical limitations. Theoretically, weextend and refocus their model by adopting amore comprehensive absorptive capacity perspec-tive. Lyles and Salk equate absorptive capac-ity with organizational exibility. However, thistouches on only one component of that construct.An organizations absorptive capacity is its abil-ity to (1) understand new external knowledge,(2) assimilate it, and (3) apply it to commercialends (Cohen and Levinthal, 1990: 128). Flexibil-ity and the other learning structures and processesstudied by Lyles and Salk all inuence the middlecomponent: assimilating foreign parents knowl-edge. We begin by adopting their learning con-structs to capture the middle component (Ability toAssimilate the External Knowledge) and add new

constructs to our model to incorporate the othertwo components. We incorporate the rst compo-nent, Ability to Understand the External Knowl-edge, by adding two constructs to the absorptivecapacity model: trust and relative absorptive capac-ity. The IJVs parents need to trust each other;trust entails having condence that the other rmwill refrain from exploiting your vulnerabilities,and will also contribute their valuable knowledgeto the IJV. Relative absorptive capacity theory(Lane, 1996; Lane and Lubatkin, 1998) suggeststhat the IJV and foreign parent must have suf-ciently similar knowledge bases and norms in order

for the student (IJV) to understand the teacher(foreign parent). In their performance model, Lylesand Salk (1996) looked only at the effect of learnedexternal knowledge. We extend this by more fullyincorporating the third absorptive capacity compo-nent, Ability to Apply the Knowledge, through theaddition of two more constructs: the IJVs strat-egy and its training competence. Without a strategyappropriate for its environment and the ability todisseminate learned external knowledge throughtraining, the IJV may not be able to effectivelyapply it.

Empirically, we extend the research on thestrategic management of IJVs by resurveying the

Hungarian IJVs studied by Lyles and Salk (1996).This permits us to examine IJV learning from for-eign parents over time and to control for prior IJVperformance, two issues not addressed by previ-ous research. Note that our sample is a subsetof theirs: the rms that participated in the 1993survey and participated in the 1996 survey. Thus,where Lyles and Salk (1996) tested their predic-tions using a sample of relatively new IJVs, wetest ours using a sample of relatively successful,established IJVs.

The transitional economy IJVs studied by Lyles

and Salk are a particularly appropriate context fortesting the model we propose. Our model assumesthat absorptive capacity-related factors are the pri-mary constraints on IJVs acquiring and applyingforeign parents knowledge. Prior research sug-gests that assumption is likely to be correct fortransitional economy IJVs. Their domestic parentshave developed in economies with monopolistic,state-dominated industries and are not likely tohave the management, manufacturing, and mar-keting capabilities needed to survive in a market-based economy (Inkpen and Beamish, 1997; Lyles

and Salk, 1996). This makes acquiring knowledgerelated to those capabilities from foreign parentscritical to the transitional economy IJVs survival.As noted earlier, the survival of the IJV is a pre-requisite for the domestic and foreign parents toachieve their goals. It is also a prerequisite for thedomestic government to achieve its goals (Childand Markoczy, 1993; Hisrich and Szirmai, 1993;Lyles and Baird, 1994). The common interest of allstakeholders in the survival of a transitional econ-omy IJV creates strong incentives for the IJV andits parents to focus on its learning from the foreignparent.

Copyright 2001 John Wiley & Sons, Ltd. Strat. Mgmt. J. , 22: 11391161 (2001)


3/23

IJV Learning and Performance 1141

ABSORPTIVE CAPACITY AND IJVLEARNING

IJVs can create competitive advantage by internal-izing and adapting partner skills and capabilities(Porter, 1986). Transferring knowledge betweenorganizations is always difcult (Szulanski, 1996),but differences between rms in established andemerging economies add to the challenge. Theorysuggests that forming a separate IJV organizationwill facilitate learning by providing the expec-tation of a stable, long-term relationship whichallows trust and knowledge sharing to develop(Beamish and Banks, 1987). IJVs within transi-

tional economies are not typically parent-parentlearning races as described by Hamel (1991),but rather collaborations which foster competitiveadvantage by using the joint venture organizationto create, store, and apply knowledge (Grant andBaden-Fuller, 1995). Thus, a central concern of allparties is how to manage foreign parent-IJV learn-ing to enhance the joint ventures capabilities andperformance.

Huber (1991) refers to this type of learning asgraftingadding to an organizations knowl-edge base by internalizing knowledge not previ-ously available to it. Yet the term grafting impliesa more static learning relationship than in factexists. Tiemessen et al . (1997: 374) depict an IJVas a system of iterative knowledge ows amongthe parents and the IJV. One set of ows, knowl-edge learned from foreign parents by the IJV,is the foundation of IJV success, the pump thatdrives the system of knowledge ows. Knowledgefrom foreign parents helps the IJV to interpretand respond to its environment effectively andto identify the most important types of knowl-edge and information to be shared through the IJVsystem.

Before the IJV can use its foreign parentsknowledge as a guide, however, it must be able tounderstand it. Simply creating an IJV does not byitself lead to such mastery (Nicholls-Nixon, 1993).Understanding and assimilating complex organiza-tional knowledge requires the active engagementof both parties as well as certain structural andcognitive preconditions. These factors are summa-rized in the left half of Figure 1. Some have beenpreviously discussed by the IJV literature whileothers have been developed in the context of otherforms of interorganizational learning and are beingextended to IJVs.

Ability to understand foreign parentsknowledge: Trust between parents

Interorganizational trust is a critical part of absorp-tive capacity because it encourages the teacherrm to actively help the student rm to under-stand the knowledge it is offering. This is unlikelyto happen unless the teacher is condent that itspartners are reliable and will fulll their obliga-tions (Johnson et al ., 1996). Trust functions as anongoing social control mechanism and risk reduc-tion device (Gulati, 1995; Florin, 1997). It inu-ences both the extent of knowledge exchanged inIJVs (Buckley and Casson, 1988; Inkpen, 1997;Inkpen and Currall, 1997; Johnson et al ., 1996;Lyles and Baird, 1994; Smith, Carroll, and Ash-ford, 1995), and the efciency with which it isexchanged (Kogut, 1988; Parkhe, 1993).

Trust has multiple dimensions, which vary intheir applicability to different contexts (Rousseauet al ., 1998). Two of its dimensions are particularlyrelevant to IJV learning. The rst is a willingnessto risk vulnerability, a prerequisite for opennessand sharing of valuable secret information and tacitknowledge (Makino and Delios, 1996; Inkpen andBeamish, 1997). The second is the condence thatanother party, not under your control, will refrain

from exploiting your vulnerabilities, a dimensionsometimes called forbearance (Barney and Hansen,1994). Both dimensions suggest that, the greaterthe trust in the relationship, the more willing allparties will be to share and exchange informationthat may make them vulnerable (Mayer, Davis, andSchoorman, 1996; Chiles and McMackin, 1996;Zand, 1972). 1

Of these two dimensions, condence is likely toplay the more important role in determining IJVlearning. The act of entering into an IJV agree-ment, by itself, suggests some degree of willing-ness to risk vulnerability. Having done so, thepartners then determine their condence in eachothers actions based on the execution of the agree-ment and responses to unforeseen circumstances(Ring, 1997). Inkpen (1997), for example, foundthis dimension to be the most important determi-nant of overall trust and business performance inIJVs. Hence,

Hypothesis 1: There will be a positive asso-ciation between the level of trust between an

1 The authors would like to thank one of the reviewers forpointing out these distinctions.

Copyright 2001 John Wiley & Sons, Ltd. Strat. Mgmt. J. , 22: 11391161 (2001)


4/23


K

n o w

l e d g e

L e a r n e d

f r o m

F o r e i g n

P a r e n

t a

A b s o r p t

i v e

C a p a c

i t y

C o m p o n e n

t 2

A b i l i t y

t o A s s

i m i l a t e

E x t e r n a l

K n o w

l e d g e

I J V F l e x

i b i l i t y & A d a p t a b

i l i t y

a

M a n a g e m e n

t S u p p o r t

b y F o r e

i g n

P a r e n t s

a

T r a i n i n g

b y F o r e

i g n

P a r e n t s a

F o r m a l

G o a

l s f o r I

J V a

S p e c

i a l i z a t

i o n o f

I J V s

P a r e n t s

a

I J V

P e r

f o r m a n c e

a

A b s o r p t

i v e

C a p a c i

t y C o m p o n e n

t 3

A b i l i t y

t o A p p

l y E x t e r n a l

K n o w

l e d g e

I J V s

B u s

i n e s s

S t r a t e g y

I J V s

T r a i n i n g

C o m

p e t e n c e

A b s o r p t

i v e

C a p a c

i t y

C o m

p o n e n

t 1

A b i l i t y

t o U n

d e r s t a n

d E x t e r n a l

K n o w

l e d g e

T r u s

t B e t w e e n

I J V s

P a r e n t s

C u l

t u r a

l C o m p a

t i b i l i t y w

i t h F o r e i g n

P a r e n t s

P r i o r K n o w

l e d g e

f r o m

F o r e i g n

P a r e n t s

R e l a t e d n e s s o f

I J V & F o r e

i g n

P a r e n t s

B u s

i n e s s e s

a V a r

i a b l e s u s e d

i n L y l e s a n

d S a l k ' s ( 1 9 9 6 ) m o d e l .

F i g u r e

1 . A n a b s o r p

t i v e c a p a c i t y m o d e l o f

l e a r n i n g a n

d p e r f o r m a n c e

i n I J V s

Copyright 2001 John Wiley & Sons, Ltd. Strat. Mgmt. J. , 22: 1139 1161 (2001)


5/23


IJVs parents and its learning from the foreign parents.

Ability to understand foreign parent sknowledge: Relative absorptive capacity

Most research on learning in IJVs and other learn-ing collaborations has implicitly assumed that eachorganization has a certain ability to learn fromother organizations. For example, in their discus-sion of absorptive capacity, Cohen and Levinthal(1990) suggest that it is a by-product of prior orga-nizational problem solving and is itself dependent

on the absorptive capacities of the organization smembers. Within the IJV literature, the Lyles andSalk (1996) measures of organizational structures(their Hypothesis 1) represents an approach toassess the absolute absorptive capacity of IJV orga-nizations using measures of organizational exibil-ity and creativity.

There is, however, evidence that an organiza-tion s capacity to learn is not absolute but rathervaries with the learning context. Hamel (1991)observed that alliance partners vary in the trans-parency of their organization and their skills, andargues that this transparency in uences learningbetween partners. Similarly, Parkhe (1991, 1993)found evidence that differences in partner national-ity and culture will negatively in uence the successof an alliance, particularly the ability to bene tfrom knowledge spillovers. Though these broadsimilarities between rms are important, they areunlikely to be the only determinants of absorp-tive capacity. Additionally, research on individ-uals suggests that a student s ability to learn isgreatest when the new knowledge is related towhat the student already knows, and when the stu-dent and teacher share similar cognitive structures

(e.g., Ellis, 1965; Estes, 1970; Bower and Hilgard,1981). Given Cohen and Levinthal s (1990) sug-gested link between individual and organizationalabsorptive capacity, factors are likely to in uencea rms ability to learn from another rm.

It is therefore likely that one rm will be betterable to learn from another when: (a) it is famil-iar with the type of new knowledge offered bythe teacher rm, (b) when the organizations havecompatible norms and values, and (c) they havesimilar operational priorities or dominant logics (Prahalad and Bettis, 1986; Bettis and Prahalad,1995). Lane and Lubatkin (1998) found support

for this theory of relative absorptive capacity ina study of R&D capability transfers between U.S.pharmaceutical and biotechnology rms. Further-more, their measures of relative absorptive capac-ity were shown to have greater explanatory powerthan the established measure of absolute absorptivecapacity (R&D spending). Applying the concept of relative absorptive capacity to IJVs suggests:

Hypothesis 2: An IJVs absorptive capacity rela-tive to its foreign parents will be positively asso-ciated with learning from those parents.

What constitutes relevant prior knowledge, com-patible norms, and similar dominant logics willlikely vary with the types of organizations beingstudied. These differences must be taken intoaccount when making empirical predictions basedon relative absorptive capacity. While Lane andLubatkin (1998) examined relative absorptivecapacity in the context of R&D capability trans-fers, we are concerned with the transfer of man-agerial, marketing, and manufacturing knowledgeand skills. This is an important distinction. R&Dcapabilities are built on generally known basicand applied science. Managerial, marketing, and

manufacturing knowledge and skills are developedthrough experience, and thus tend to be highlyrm-speci c. This suggests that the more an IJVhas previously learned from its foreign parents, themore effective will be its current learning efforts.

Hypothesis 3: There will be a positive associ-ation between an IJVs prior learning from its foreign parents and its current learning fromthose parents.

Because they were examining U.S. rms in

two closely allied industries (pharmaceuticals andbiotechnology), Lane and Lubatkin assessed thesimilarity of norms and values in terms of spe-cic organizational polices and compensation prac-tices. In contrast, the norms and values of foreignparents can differ from their IJV s much morebroadly. Lorange and Roos (1993) suggest thatan IJV s success is linked to a rms ability toadapt to cultural differences. Mowery, Oxley, andSilverman (1996) found that there are higher pat-terns of knowledge transfer in alliances of cultur-ally similar parents than in alliances with cross-national parents. Conversely, cultural con icts and



6/23


misunderstandings can limit the sharing of infor-mation and learning (Fiol and Lyles, 1985; Laneand Beamish, 1990; Lyles and Salk, 1996; Parkhe,1993; Salk, 1996). Consequently, the parents cul-tures need to be compatible in order for the IJV smanagers to understand the norms and valuesimplicit in the acquired know-how. Hence,

Hypothesis 4: There will be a positive associa-tion between cultural compatibility of an IJVs parents and learning from its foreign parents.

The dominant logic of a rm is a function of the types of problems that its managers regu-larly confront (Prahalad and Bettis, 1986; Bettisand Prahalad, 1995). Lane and Lubatkin assessedthe similarity of alliance partners dominant logicsbased on the sets of scienti c research problemsthey were investigating. The relatedness of a for-eign parent and IJV centers on their similarity of business objectives and strategic resources, a factorwhich few studies of IJVs have directly addressed.Lyles (1991) tested the effect of business related-ness of the parent rm and the IJV to performanceand risk reduction of the parent rm and foundno systematic relationships. Yan and Gray (1994)

suggest that there should be an overlap and similar-ity of expertise. Mjoen and Tallman (1997) showthat the higher the relatedness of the IJV with theparent, the stronger the positive effect on controlover speci c activities and the higher the perceivedIJV performance. Kogut (1989) found a positiverelationship between relatedness and IJV survival.Furthermore, Inkpen suggests that if the IJV isinvolved in an unrelated business, it is unlikelythat the primary rationale for collaboration will beknowledge creation (Inkpen, 1997: 354). Usingbusiness relatedness as a proxy for shared domi-nant logics suggests:

Hypothesis 5: There will be a positive asso-ciation between the degree to which an IJVsbusiness is related to its foreign parents and itslearning from those parents.

Ability to assimilate foreign parent sknowledge: Learning structures and processes

A major nding of Lyles and Salk (1996) isthat learning by IJVs requires certain learning

structures and processes facilitate the assimila-tion of knowledge from the foreign parent. Theyfound a signi cant association between IJV learn-ing and its organization s exibility and creativ-ity. Inkpen (1997), in studying North American-Japanese IJVs, similarly found that receptivity tolearn was important. This is thought to be espe-cially important for creating an openness to newknowledge in the IJV, for creating emergent skillsand knowledge building on what the foreign parentcan provide an IJV, and tailoring foreign parentknowledge to t the speci c needs of the IJV(Cohen and Levinthal, 1990; Lyles and Salk, 1996;Nonaka, 1994).

Learning also is facilitated when speci c pro-cesses help IJV members to target knowledge thatcould or should be transferred, and when they havea window into the workings of the parent s orga-nization. Lyles and Salk (1996) found support forthe contention that establishing explicit goals, busi-ness plans and division of labor between parentshelp focus attention of parents and the IJV onpotentially useful knowledge to be transferred andgrafted into the IJV (Huber, 1991; Nonaka, 1994).In addition, their results and those of Inkpen (1997)support the belief that active management involve-ment and training by the foreign parent not onlycan provide a vehicle for transferring explicit orcodi ed knowledge, but also can be a vehicle forlearning and knowledge transfer of tacit knowledgevia socialization by the foreign parent management(Nonaka, 1994).

Both Inkpen (1997) and Lyles and Salk, (1996)examine relatively new IJVs. It is unclear how therelationships documented in these studies mightshift over time as an IJV evolves especially in tran-sitional economies which are themselves rapidlyevolving. Earlier we proposed that an IJV s abilityto learn from its foreign parents would increase

as it accumulates understanding of their knowl-edge. This suggests that the learning structuresand processes that support learning may becomeless important as that understanding accumulates.However, it is not clear from prior research thatsuch a trade-off in fact exists, that understandingexternal knowledge makes it easier to facilitate thatknowledge. Furthermore, positing such a trade-off assumes that there is a xed quantity of knowl-edge that an IJV needs to learn from its foreignparent over its life time. This is unlikely to betrue. For example, between 1993 and 1996, Hun-gary underwent a massive privatization of banks



7/23


and industry, antitrust and bankruptcy laws wereenacted, employee stock-option plans legalized,and freedom of the press expanded. Some of thesedevelopments might reinforce, or even acceler-ate, learning from foreign parents (e.g., privatiza-tion, antitrust). Others might create new areas of knowledge that the IJV needs to acquire (e.g., useof employee stock option plans). These changesmake it dif cult to predict a priori how differentlearning structures and processes and knowledgelearned from the foreign parents will be relatedover time. For these reasons, we retest the associ-ations between structures and processes proposedby Lyles and Salk (1996) and learning for this now

older and more experienced sample of IJVs:

Hypothesis 6: Organizational exibility and adaptability, formal goals, specialization of parents roles, managerial involvement by foreign parents, and training by foreign parentswill be positively associated with learning from foreign parents.

ABSORPTIVE CAPACITY AND IJVPERFORMANCE

Cohen and Levinthal s (1990) discussion of thethird component of absorptive capacity, the rmsability to commercially apply the learned exter-nal knowledge, is considerably briefer than that of the other two. However, in both their introductionand discussion of implications they suggest thatthe rm must be able to link the learned externalknowledge to its product markets in a meaningfulway. As noted earlier the link between IJV learningperformance and product market/ nancial perfor-mance has long been assumed to exist, and Lyles

and Salk (1996) found some evidence of this forrelatively young IJVs.

The in uence of knowledge from the foreignparent on performance may decrease over timeas the IJV meets its original needs and beginsto develop its own capabilities. However, that isless likely to be true in transitional economies.Those economies are marked by the long peri-ods of rapid and discontinuous change which cre-ate incentives for continued learning from foreignparents. D Aveni (1994) makes a similar obser-vation about hypercompetitive markets, anotherenvironment marked by rapid and discontinuous

change. He argues that this type of change requiresrms to continually learn to survive, let alonethrive. This, in turn, encourages rms to learnfrom highly skilled partners to speed their devel-opment of their own knowledge and capabilities.Organizations based in transitional economies maylack the technological and managerial sophistica-tion D Aveni ascribes to hypercompetitive rms.However, rms in both environments face similarrates and scales of environmental change whichrequire similar commitments to ongoing interorga-nizational learning.

Hypothesis 7: The knowledge an IJV acquires

from its foreign parents will be positively asso-ciated with its performance.

The acquisition of valuable new knowledge doesnot by itself in uence an organization s perfor-mance. The strategic context in which the knowl-edge is used as well as the IJV organization s abil-ity to adapt the knowledge to that strategic contextwill also in uence its ability to effectively applythe speci c knowledge it has acquired. These rela-tionships are illustrated in the right half of Figure 1and their logic is developed in detail below.

Ability to apply foreign parent s knowledge:IJV business strategy

The fundamental choices in business strategy avail-able to rms in established market economies canbe categorized as cost leadership or differentiation(Porter, 1980). These generic strategies, along withmost strategic theories, assume a market economyin which the legal environment supports propertyrights and that rms have the ability to reducetransaction costs (Peng and Heath, 1996). In for-mer communist countries, however, the institu-

tional framework created is far different (Childand Markoczy, 1993). In 1996, the governmentsstill controlled much of the access to resources andto established distribution channels (Brouthers andBamossy, 1997; Peng and Heath, 1996), and polit-ical in uence plays an important role in governingtheir use.

This environment can be especially dif cultfor small- to medium-sized rms like most IJVs.They tend to lack suf cient bargaining powerto gain enough resources and access to distribu-tion channels to compete on the basis of costwith state-owned enterprises or large enterprises.



8/23


Smaller rms also frequently face dif culty obtain-ing nancing, even those af liated with foreignrms. This is in part due to some foreign rms reluctance to put assets at risk in transitionaleconomies. However, some small- to medium-sized rms in transitional economies possess otherresources which the large enterprises do not. Whenproperly utilized, these resources can produce acompetitive advantage for a small- to medium-sized rm despite the forces arrayed against it.

IJVs can potentially access both knowledge of the domestic market and foreign expertise aboutmarketing, service, and product development. Thisknowledge can be used to avoid direct competition

with large and state-owned enterprises, developdistinct competitive advantages, and help to fur-ther expand demand in the transitional economy.IJVs that acquire knowledge from foreign parentsrelated to cost leadership will face far more obsta-cles to turning that knowledge into competitiveadvantage. Thus, the more IJVs emphasize differ-entiation, the greater their performance will tendto be.

Hypothesis 8: An emphasis on differentiationin an IJVs business strategy will be positivelyassociated with its performance.

Ability to apply foreign parent s knowledge:IJV training and development

While some of the external knowledge a rmacquires may be acted on immediately, it is morelikely that it will have to be adapted and dissem-inated internally before it can be commerciallyutilized (Cohen and Levinthal, 1990). This is espe-cially true of IJVs. Boyacigiller and Adler (1991)suggest that differences in global environments

mean that knowledge transferred by foreign rmsmay not directly t the local context. This may inpart account for the lag between the acquisition of knowledge and its in uence on decision-making(Fiol and Lyles, 1985). The processes needed toadapt and disseminate new knowledge varies withthe type of knowledge. For example, the R&Dliterature documents the success of in-house con-ferences, symposia, and knowledge brokers for dis-seminating scienti c knowledge (Jain and Triandis,1990).

In transitional economies, the knowledge learnedfrom foreign parents tends to relate to skills

in management, marketing, and decision-making.These are skills that require adaptation and integra-tion into speci c social, cultural and competitivecontexts and they are best adapted and dissem-inated through the training and development of an organization s staff. Managerial training canhelp managers in transitional economies overcomebehavioral patterns developed while operating incentrally controlled economies but inappropriatefor market economies (Child and Markoczy, 1993).Thus, an IJV s ability to train and develop its ownpersonnel can help its performance in two ways:(1) by adapting and disseminating the new knowl-edge learned from foreign parents; and (2) by help-

ing managers develop the new skills needed tocompete in emerging markets. Hence:

Hypothesis 9: An IJVs degree of competencein training and personnel development will be positively associated with its performance.

Linking the ability to understand andassimilate foreign parent s knowledge to IJVperformance

Our two-stage model suggests that the relation-ship between the antecedents to learning fromforeign parents and IJV performance is an indi-rect or mediated effect (Baron and Kenny, 1986).The IJV s ability to understand its foreign parent sknowledge (based on trust and relative absorp-tive capacity) and its ability to assimilate thatknowledge (based on learning structures and pro-cesses) lead to the IJV learning a certain amountof knowledge from the foreign parents (a b).That knowledge then in uences the IJV s perfor-mance (b c). This suggests that the IJV s abilityto understand foreign parents knowledge and its

ability to assimilate that knowledge in uences per-formance (a c), but that the in uence is trans-mitted by the knowledge actually learned (a b c). The logic is that while learning abilities areimportant, it is the outcome of the application of those skills that matters for IJV performance. Thebelief that the amount or quality of learning inter-venes in (mediates) the relationship between theantecedents of IJV learning and IJV performanceis implicit in most of the research on IJV learning.Lyles and Salk (1996) tested for this relationshipbut failed to nd support. However, the two-stagemodel developed here is more comprehensive and



9/23


accounts for several factors not included in theirtests, and we are examining it in the context of older IJVs. Thus, there is reason to revisit thisrelationship:

Hypothesis 10: The relationship between theantecedents of learning from foreign parents and IJV performance will be mediated by the amount of knowledge learned from the foreign parents.

METHODS

Data

Our model and hypotheses are tested using dataon a resurveyed sample of the Hungarian IJVsstudied by Lyles and Salk (1996). One of thechallenges when researching organizations in tran-sitional or emerging markets is the absence of reliable archival data on those organizations. Thereis no Compustat equivalent nor are reliable rm-level data available from the government. Conse-quently, researchers examining those organizationsmust collect their own data through surveys orinterviews.

We built on the survey data Lyles and Salk col-lected in 1993 using data from a second surveyin 1996. They used a strati ed sample that com-prises a representative sample of small/medium joint ventures in Hungary, in terms of industriesand the country-of-origin of the foreign partners.The initial sample and sampling technique weredeveloped with the help of a government agencythat received information about the IJVs from thegovernment. Sample strati cation was based uponstatistics provided by Hungary s Central Statisti-cal Of ce, which show the percentage of rms ineach industry and the percent from each country-of-origin of the foreign partners. The rms that

participated were identi ed through directories,contacts, and the Hungarian government database.The type and extent of the information collected inthis project is unavailable elsewhere, since smallIJVs do not have stringent reporting requirementsin Hungary. Thus, the survey created a uniquedatabase for Hungary on small to mid-sized IJVs.

The 1993 and 1996 surveys were conductedin conjunction with leading Hungarian economicinstitutes and university staff under the supervisionof one of the authors. (See Lyles and Salk, 1996:886 887, for more details.) Standardized, struc-tured interview and survey protocols were used

to avoid interviewer bias. These were translatedand back-translated to avoid any bias resultingfrom language misunderstandings. The protocolswere then administered to the general manager orpresident of each IJV by carefully selected andtrained Hungarian interviewers. While all of theinterviewers were bilingual, almost all of the inter-views were conducted in Hungarian. The inter-views resulted in the accumulation of data for eachIJV that detailed its founding, its management, itsownership structure, its relationship with its par-ents, its nancial management and its competitivestrategy. We used that same data collection pro-cess for our 1996 resurvey of these organizations.

The 1996 survey contains items from the earliersurvey as well as some new items to test our morerened model of learning. Of the IJV organiza-tions surveyed in 1993, 49.3% were still operatingin 1996. Our response rate for the surviving was44%, resulting in suf ciently complete data forboth 1993 and 1996 for 78 IJVs.

The difference in sample size between our studyand Lyles and Salk s (78 vs. 201) raises concernsabout survivor bias and the comparability of thetwo samples. Table 1 compares two sub-groupsof Lyles and Salk s (1996) sample: the 78 IJVsthat are included in our current paper and the 123that are not. In 1993 the two groups were verysimilar in terms of number of employees, num-ber of expatriates working in the IJV, percentageof equity owned by the Hungarian partner, andtenure of the IJV s General Manager. The IJVsparticipating in our study were slightly older thanthose only participating in the 1993 survey. How-ever, both groups experience similar levels of dif-culty in arranging nancing and the availabilityof inputs and suppliers. Additional comparisonsbased on the measures of learning and performancedescribed below revealed no difference in the mean

levels of learning, but the IJVs participating in oursurvey did perform slightly better in 1993 than theIJVs that did not participate. Overall, there is littleevidence of a systematic differences between thetwo groups.

The IJVs in our sample operate in a number of sectors of the Hungarian economy. The industrialand manufacturing sector accounted for 60% of oursample (47 IJVs), and over half of them were con-centrated in three industries: textiles, electronics,and chemicals and pharmaceuticals. The servicesector accounted for the second largest number of IJVs (15). The agriculture related portion of the



10/23


Table 1. Characteristics of the International Joint Ventures in 1993

IJVs participating in IJVs not participating in1996 survey 1996 survey(N = 78) (N = 123 )

Average number of total employees 70.8 73.9Average number of expatriates working in IJV 0.7 0.6Average percentage of equity held by Hungarian partner 32.4% 34.3%Average percentage of IJV sales in Hungary 63.9% 64.0%Average age of the IJV (in years) 5.1 4.7

Average tenure of the IJV s General Manager (in years) 3.6 3.1Percentage of IJVs experiencing a moderate to high degree of

dif culty obtaining nancing65.3% 67.4%

Percentage of IJVs experiencing a moderate to high degree of problems with suppliers and input availability

47.4% 50.4%

Means differ by p < 0.10

sample (9 IJVs) is dominated by food processingbusinesses. Relatively few of the IJVs participatedin the construction and transportation sectors (7IJVs combined).

Dependent variables

Learning

The dependent variable for Hypotheses 1 through

6 is Current Knowledge learned from ForeignParents. It was measured by a ve-item scalebased on Likert-type responses to the questionTo what extent have you learned from yourforeign parents (a) new technological expertise,(b) new marketing expertise, (c) product develop-ment, (d) managerial techniques, and (e) manufac-turing process (1 = little and 5 = to a greatextent), which give the scale a possible range of 5 25. The Cronbach s alpha for these ve itemswas 0.96.

Performance

The dependent variable for Hypotheses 7 through9 is IJV Performance in 1996. Two different scaleswere used to capture IJV performance in 1996. TheIJV s performance in speci c areas was assessedusing four ve-point Likert items (1 = poor to5 = excellent) on the IJV s performance in termsof increasing business volume, increasing marketshare, achieving planned goals, and making prof-its. Second, the general performance of the IJV wasassessed using a three-item scale that summarizeshow the Hungarian parent, foreign parents, and IJV

general manager evaluate the IJV s overall perfor-mance. Factor analysis revealed that the items forboth performance scales were unidimensional. Theresponses for the performance items were stan-dardized and then added to create an overall IJVPerformance Index (alpha = 0.87).

Measures of an IJV s ability to understandforeign parent s knowledge

Trust The decision to trust another organization is asocial decision (Kramer, Brewer, and Hanna,1996). This has two implications for how wemeasure trust. First, it means that the social con-text of each partner, especially their national ori-gin, strongly shapes how they learn to trust oth-ers (Doney, Cannon, and Mullen, 1998). Thesenational differences mean that the trust develop-ment process of IJVs can vary widely based onthe combination of nationalities involved (Parkhe,1998). Second, the decision to trust does not arise

from deliberate calculations based on objective cri-teria but instead is learned and reinforced throughon-going interactions (Powell, 1996). This is anevolutionary process and the basis for trust (anddistrust) in the relationship is continually in transi-tion (Rousseau et al ., 1998). Taken together, theseimplications suggest that attempting to capture thevarious facets of trust between partners in our sam-ple of IJVs is neither an easy task nor an essen-tial one. What matters most at any given pointin time is the partners overall con dence in oneanother s trustworthiness. Accordingly, the levelof Trust Between Parents was measured using a



11/23


single item (1 = low level of trust, 5 = high levelof trust).

Relative absorptive capacity

We adapted the approach developed by Lane(1996) to the context of Hungarian IJVs. Rel-evant prior knowledge was measured using thelearning scale, described above. However, themeasure Prior Knowledge from Foreign Parentsutilized data from the 1993 study. The Cron-bach s alpha for the ve learning items in thatsurvey was 0.88. Cultural differences are eas-ier to detect than cultural compatibility. Accord-

ingly, the Cultural Compatibility of the domes-tic and foreign parents was measured by reversescoring a two-item scale (1 = little to 5 = verymuch) on the extent to which cultural misun-derstandings and cultural differences have beenissues in the IJV (alpha = 0.91). The distribu-tion of the scale was positively skewed, so thelog of the Cultural Compatibility was used inour analyses to correct for this. The Related-ness of IJV s and Foreign Parent s Businesseswas measured by reverse scoring a single itemfor unrelatedness (in reverse form: 1 = related,0 = unrelated).

Measures of an IJV s ability to assimilateforeign parent s knowledge

IJVs structures and processes to facilitate learningfrom foreign parents was assessed using measuresdeveloped by Lyles and Salk (1996), but based ondata collected in the 1996 survey. The IJV s Flexi-bility and Adaptability was measured using a three-item scale of 5-point Likert-type items based onagreement with normative statements on the extentto which the IJV is exible and adapting to change,

is creative, and rewards performance (alpha =0.67; possible range 3 15). Contributions of Man-agement Support by Foreign Parents to the IJV wasmeasured using a six-item scale of 5 point Likert-type items that summarizes the degree to which theforeign parent(s) contribute sales/marketing sup-port, managerial resources, administrative support,emotional support, training, and time to the IJV(alpha = 0.82; possible range 6 30). The extent towhich managers in the IJV were given Educationand Training by Foreign Parents was measured bya single item (1 = little to 5 = great extent). TheIJV s use of Formal Goals and objectives was also

measured with a single item (1 = either or bothwritten, 0 = neither written). The Specialization byForeign and Domestic Parents was measured usinga single item on the extent to which the foreignparent(s) provides technology to the IJV while thedomestic parent provides the manufacturing capa-bility (1 = little, 5 = great extent).

Measures of an IJV s ability to apply foreignparent s knowledgeCurrent knowledge learned from the foreign parent

This is the dependent variable from the learning

models. See discussion on learning on previouspage.

IJVs strategy

The degree to which the IJV s Business Strategyentailed differentiation over the past two years wasmeasured using a scale of six Likert-type items(alpha = 0.75; possible range 6 30). The itemsassessed the degree to which the IJV had empha-sized: (1) developing new products, (2) promotionand advertising expenditures above industry aver-age, (3) a broad product line, (4) extensive cus-tomer service capabilities, (5) highly trained per-sonnel, and (6) strong in uence over the channelsof distribution.

Training and development competence

This was measured using a two-item scale of 5 point Likert-type items on how effective theIJV was in the prior year in providing adequateworker training and improving management skills(alpha = 0.72; possible range 2 10).

Controls Learning models

Four variables were used to control for other fac-tors which could in uence IJV learning from for-eign parents. First, because the size of an organi-zation may contribute to its inertia and thus inhibitlearning, we included a measure of IJV size: thelog of the IJV s capitalization in forints (the Hun-garian currency). Second, the more an IJV servesexport markets, the more directly it competes withrms from more developed economies. This mayincrease the IJV s incentives for learning from its



12/23


foreign parents. To control for this, we included thepercentage of the IJV s sales made outside Hun-gary. Third, the volatility of the IJV s home marketmay slow learning by confusing the relationshipsbetween observed means and ends. We thereforeadded a control for the volatility of domestic mar-ket demand, the coef cient of variation for thedemand growth for the years 1993 through 1996(Lev and Kunitzky, 1974). Our demand volatil-ity measure was calculated using sector level datafrom the Hungarian Economic Monitor (July 17,1997). Fourth, IJVs in service industries may facedifferent incentives for learning from foreign par-ents than manufacturing IJVs as services tend to

be more culturally speci c than products. Hence,a dummy variable was created for the 15 IJVs inservice industries (1 = service industry, 0 = otherindustry).

Performance models

Four controls were used when testing the hypothe-ses related to IJV performance. We controlled forPrior IJV Performance using the 1993 responsesto the same scales used to create the perfor-mance dependent variable from the 1996 data. The1993 responses were evaluated, standardized andsummed into an index in the same manner as the1996 responses. The Cronbach s alpha for PriorIJV Performance is 0.85. The controls for exports,demand volatility, and service industries used inthe learning models were also used in the per-formance models. The competitiveness of exportmarkets could limit IJV performance, and unsta-ble demand could make it dif cult to managecosts and meet customer needs. Service industryIJV s during this period could be more challeng-ing to manage than manufacturing IJVs in Hungarybecause of the emphasis on manufacturing during

the decades of a state dominated economy. Hun-garians entered the transitional economy with farless experience with service-oriented rms.

Analysis

There were ve missing responses in the naldata set, one each for Size, Management Con-tribution, Parent Specialization, Trust, and PriorPerformance ( ve out of over 1300 data points,less than 0.01%). Deleting the associated caseswould reduce the sample size from 78 IJVs to 74(a reduction of 5%). In order to preserve statistical

power, all regression analyses were conductedusing a mean substitution correction. Substitutingfor such a small percentage of the data points willresult in a slight reduction in the variance in eachof these ve measures. Thus, any bias introducedby the mean substitutions is conservative.

The hypotheses were tested using multiple re-gression analyses. Three models tested the learn-ing hypotheses (Hypotheses 1 6) by regressingdifferent sets of learning predictors on CurrentKnowledge Learned from Foreign Parents in 1996.One model included just the measures of Abil-ity to Understand Knowledge, one model just themeasures of Ability to Assimilate Knowledge, and

one model with both sets of measures. Four mod-els regressed the measures for current knowledgelearned from foreign parents, IJV business strat-egy, and IJV training competence on IJV Perfor-mance in 1996 to test Hypotheses 7 9. The naltwo models tested for mediation (Hypothesis 10)using the controls for the performance models,both sets of independent variables for the learn-ing models, and Current Knowledge from ForeignParents, the proposed mediator.

Common method bias can be a problem whendependent and independent variable data are col-lected from a single informant. Following Pod-sakoff and Organ (1986), we used the Harman sone-factor test to examine the extent of that biasin our data. A principal components factor analy-sis reveals that there are 6 factors with eigenvaluesgreater than 1.0 which together account for 64% of the total variance. The presence of several distinctfactors combined with the relatively low amount of variance explained by the rst factor (only 19%)indicates that our data do not suffer from commonmethod variance (Podsakoff and Organ, 1986).

Following Fox (1991), the square root of vari-ance in ation factor was calculated for the inde-

pendent variables in each regression model tocheck for multicollinearity. All the variables in ouranalyses except one fell well within Fox s accept-able range indicating no multicollinearity prob-lems. Specialization of Parents was dropped fromthe two regression models used to test for medi-ation because of multicollinearity with the twoknowledge variables.

RESULTS

Table 2 presents the means, standard deviations,and correlations among the variables used in the



13/23


T a b l e

2 . D e s c r

i p t i v e s t a t

i s t i c s a n

d c o r r e

l a t i o n s

M e a n

S . D .

1

2

3

4

5

6

7

8

9

1 0

1 1

1 2

1 3

1 4

1 5

1 6

1 . C u l

t u r a

l

c o m p a

t i b i l i t y

0 . 5 3

0 . 7 2

2 . C u r r e n t

l e a r n i n g

1 3 . 7

2

5 . 7 6

0 . 2 1

3 . C u r r e n t

p e r f o r m a n c e

0 . 0 2

1 . 8 0

0 . 0 2

0 . 2 5

4 . D e m a n

d v o

l a t i l i t y

4 7 . 8

0

2 9 . 1 0

0 . 0 7

0 . 2 0

0 . 0 6

5 . E x p o r

t s a s

%

s a l e s

3 8 . 6

1

3 6 . 2 3

0 . 1 1

0 . 2 5

0 . 0 2

0 . 2 3

6 . F l e x

i b i l i t y o f

I J V

1 2 . 6

8

2 . 0 7

0 . 0 2

0 . 2 1

0 . 4 6 0 . 0 5

0 . 0 8

7 . G o a

l s &

o b j e c t

i v e s

0 . 5 3

0 . 5 0

0 . 0 6

0 . 0 2

0 . 0 9

0 . 0 2

0 . 1 3

0 . 1 9

8 . M a n a g e m e n

t

s u p p o r

t

1 8 . 0

3

5 . 2 0

0 . 0 3

0 . 1 9

0 . 4 1

0 . 0 9

0 . 0 5

0 . 3 2

0 . 1 1

9 . P r i o r

l e a r n i n g

1 2 . 4

9

5 . 9 3

0 . 0 4

0 . 3 4

0 . 0 8

0 . 0 8

0 . 0 5

0 . 0 9

0 . 0 6

0 . 2 4

1 0 . P

r i o r p e r f o r m a n c e

0 . 0 3

1 . 7 4

0 . 1 0

0 . 0 9

0 . 4 4 0 . 0 5

0 . 0 1

0 . 3 0

0 . 0 8

0 . 2 0

0 . 0 9

1 1 . R

e l a t e d n e s s

0 . 5 3

0 . 5 0

0 . 1 2

0 . 2 1

0 . 0 2

0 . 0 9

0 . 1 4

0 . 0 9

0 . 0 7

0 . 1 9

0 . 1 3

0 . 0 4

1 2 . S

i z e

( l o g

)

1 0 . 3

0

2 . 7 4

0 . 1 4

0 . 0 4

0 . 1 3

0 . 0 5

0 . 0 8

0 . 0 5

0 . 0 4

0 . 0 7

0 . 0 4

0 . 0 3

0 . 1 5

1 3 . S

p e c i a l

i z a t

i o n

2 . 5 3

1 . 5 1

0 . 1 6

0 . 3 9

0 . 1 0

0 . 2 4

0 . 2 0

0 . 1 2

0 . 0 1

0 . 0 4

0 . 4 9

0 . 0 7

0 . 1 6

0 . 0 2

1 4 . S

t r a t e g y o f

I J V

1 9 . 8

1

4 . 9 7

0 . 1 5

0 . 2 1

0 . 4 2 0 . 0 2

0 . 0 7

0 . 3 3

0 . 0 2

0 . 2 8

0 . 1 1

0 . 3 6 0 . 1 4

0 . 1 9

0 . 0 2

1 5 . T

r a i n i n g -

f o r e

i g n

p a r e n t

2 . 1 7

1 . 2 4

0 . 2 6

0 . 4 0

0 . 0 4

0 . 1 4

0 . 1 4

0 . 0 1

0 . 0 2

0 . 2 3

0 . 2 6

0 . 0 5

0 . 1 1

0 . 0 4

0 . 4 7 0 . 1 7

1 6 . T

r a i n i n g -

I J V

6 . 5 6

1 . 6 2

0 . 1 8

0 . 1 4

0 . 5 0

0 . 1 6

0 . 0 7

0 . 2 4

0 . 1 6

0 . 4 4

0 . 1 8

0 . 3 0

0 . 1 9

0 . 0 9

0 . 0 1 0 . 1 8

0 . 1 0

1 7 . T

r u s t

B e t w e e n

P a r e n t s

4 . 0 6

1 . 1 4

0 . 2 5

0 . 0 8

0 . 4 7 0 . 0 5

0 . 1 1

0 . 3 9 0 . 0 2

0 . 1 6

0 . 0 4

0 . 2 6

0 . 1 1

0 . 0 1

0 . 0 5 0 . 0 9

0 . 1 0

0 . 1 9

p

Documents

Absorptive Capacity, Learning, And Performance in International Joint Ventures_Lane_SMJ.2001