hillbun (dixon) ho & shankar ganesan does knowledge base … · 2020-03-07 · ganesan is...

Hillbun (Dixon) Ho & Shankar Ganesan

Does Knowledge Base CompatibilityHelp or Hurt Knowledge Sharing

Between Suppliers in Coopetition?The Role of Customer Participation

Competing suppliers that collaborate to serve downstreann original equipment manufacturer customers oftenencounter partners with overlapping and compatible knowledge bases. Such knowledge base compatibilityprovides supplier partners the opportunity to exchange knowledge efficiently, leading to greater knowledge sharing.However, the ease of misappropriation of the shared knowledge can offset this beneficial effect. This researchproposes that the effect of knowledge base compatibility on supplier partners' knowledge sharing is moderated bythe customer's participation in the collaborative effort and by the customer value such effort creates. The results oftwo empirical studies show that when levels of both customer participation and customer value are high, knowledgebase compatibility between supplier partners leads to greater knowledge sharing. In contrast, when customerparticipation is high but customer value is low, knowledge base compatibility leads to lower levels of supplierknowledge sharing. This investigation validates the importance of key factors related to supplier partners'opportunity and motivation to share knowledge in coopetitive partnerships.

Keywords: buyer-seller relationships, knowledge governance, knowledge sharing, customer participation, surveymethodology

Online Supplement http://dx.doi.org/10.1509/jm.11.570

I n today's turbulent business environment, firms often enterinto collaborative relationships to access and leveragepartners' knowledge assets. Extant literature has shown

that collaborative relationships between suppliers and buyerscreate mutual value and opportunities for interorganizationallearning (Cannon and Perreault 1999; Ganesan 1994;Palmatier, Dant, and Crewal 2007; Wuyts et al. 2004).Although these studies advance the understanding of verticalcollaborations, knowledge about horizontal collaborations(i.e., partnerships between competing firms) remains scarce.This limitation is critical because horizontal collaborationshave become a popular vehicle for firms to access and lever-age partners' technological capabilities and tacit know-how(Capron and Hulland 1999; Heiman and Nickerson 2004).

Hillbun (Dixon) Ho is Assistant Professor, Division of Marketing and inter-national Business, and feilow of the Institute on Asian Consumer Insight,Nanyang Technological University (e-maii: [email protected]). Shani<arGanesan is Professor of Marketing, Mendoza Coiiege of Business, Uni-versity of Notre Dame (e-mail: sganesan§nd.edu).The authors thank theInstitute for Supply Management for generously funding part of thisresearch. The authors aiso appreciate the insightful comments from thefour anonymous JM reviewers during the review process and heipfui peerreviews from Robert Lusch, Frederick Webster, Yubo Chen, KennethKoput, Aian Malter, and Mrinai Ghosh. Inge Geyskens served as area edi-tor for this article.

© 2013, American Marketing AssociationISSN: 0022-2429 (print), 1547-7185 (electronic) 91

In business markets, horizontal collaborations' betweenupstream suppliers are often established to satisfy thesophisticated procurement needs of downstream customers,typically original equipment manufacturers (OEMs) intechnology-based industries (Liker and Choi 2004). Suppli-ers collaborate by pooling resources and technologicalexpertise. Sharing specialized knowledge enables supplierparmers to learn from each other, leverage partner expertise,and create collective knowledge, resulting in superior solu-tions for customers (Mesquita, Anand, and Brush 2008).For example, in the semiconductor industry, an increasinglypopular practice is for suppliers to develop new products orproduction processes for OEMs through the sharing of tech-nological knowledge (Clark 2010). These developmentsrequire marketing scholars to understand conditions thatfacilitate or hinder supplier knowledge sharing.

Previous research has shown that collaborating part-ners' knowledge sharing and learning depends on opportu-nity and motivation (Inkpen 2000; Nahapiet and Ghoshal

'Conceptually, "collaboration" can refer to either long-termpartnerships (often known as collaborative relationships) orepisodic collaborative efforts aimed at solving relatively short-term problems (Zacharia, Nix, and Lusch 2011). Although ourtheoretical arguments apply to both situations, our discussion andempirical studies focus on the episodic efforts.

Journal of MarketingVol. 77 (November 2013), 91-107

1998) .2 Effective knowledge sharing requires collaboratingsuppliers to understand, value, and absorb the partner'sknowledge (Cohen and Levinthal 1990). However, organi-zational knowledge is often sticky, tacit, and causallyambiguous, requiring supplier partners to possess overlap-ping knowledge bases to overcome the barriers in theknowledge exchange processes (Szulanzki 1996). We referto this as "knowledge base compatibility," which at highlevels provides supplier partners the opportunity to shareknowledge effectively in the collaborative effort. Extendingprior research that emphasizes the benefits of knowledgebase compatibility (Dyer and Singh 1998), our study pointsto the increased exchange hazards that knowledge basecompatibility can create. These hazards discourage suppli-ers from sharing valuable knowledge (Makhija and Ganesh1997), especially when the supplier partners compete inoverlapping market segments or for the same OEM cus-tomer's business. Thus, knowledge base compatibilitycould either facilitate or impede supplier partners' knowl-edge sharing; its eventual effect is likely to depend on otherimportant conditions, such as suppliers' motivation. With-out sufficient incentives, supplier partners are unlikely toidentify and make use of the knowledge exchange opportu-nities available in the collaborative partnership.

Therefore, the present study examines two factorsrelated to supplier partners' motivation to share knowledgewith each other. Because a collaborative partnershipbetween two competing suppliers serving a downstreamcustomer constitutes a triadic "compound relationship"(Ross and Robertson 2007), the customer could directly orindirectly motivate the supplier partners to share knowl-edge. The first factor we examine is the customer's partici-pation in the supplier collaboration. Customer participationcould engender a cooperative atmosphere that mitigatessupplier partners' concern regarding exchange hazards,increasing their motivation to share knowledge. The secondfactor is the strategic value of the supplier collaboration tothe customer. When their collaborative effort creates sig-nificant customer value, supplier partners can benefit fromrepeat business from the customer, thus motivating them toparticipate in knowledge exchange activities.

This research makes several key contributions to themarketing literature. First, in response to the increasingimportance of supplier partnerships to OEM customers, ourstudy helps academics and managers gain a deeper under-standing of the key conditions that promote supplier part-ners' knowledge sharing in coopetitive partnerships. Sec-ond, by integrating the literature on interfirm learning andtransaction cost economics, this study provides theoreticalexplanations for and empirical evidence on the divergent

2For example, in the case of New United Motor ManufacturingInc. (NUMMI), a joint venture between General Motors (GM) andToyota, GM initially struggled to learn the Toyota production sys-tem. The knowledge transfer took a decade to materialize becausethe operations of Toyota's system differed substantially fromGM's experience, culture, and systems, thus limiting opportunitiesfor learning. In addition, a "not-invented-here" syndrome inhibitedthe GM engineers' motivation to leam Toyota's tacit know-how(Inkpen 2005).

effects of knowledge base compatibility under differentconditions. In particular, this research shows how factorsrelated to supplier partners' motivation (including customerparticipation and customer value) moderate the effects ofknowledge base compatibility on suppliers' knowledgesharing. Thus, this research helps supplier firms understandthe positive and negative consequences of selecting partnerswith overlapping knowledge bases. Third, this study shedslight on the governance implications of the customer'sinvolvement in its suppliers' collaborative effort when thesupplier partners are also competitors.

The remainder of the article is organized as follows. Webegin with an explication of how the present research adds toextant literature. We then lay out the theoretical foundationsof our conceptual framework and develop hypotheses basedon a synthesis of the literature. Subsequently, we present themethodology, results, and analysis of two empirical studies.We conclude by discussing the theoretical and managerialimplications of our findings and the limitations of the study.

Literature ReviewThe growing importance of interfirm collaborations andlearning alliances has stimulated extensive research exam-ining the determinants, management, governance, and out-comes of the collaborating parties' knowledge-sharingefforts. The literature on learning alliances suggests thatknowledge sharing and learning between alliance partnersdepends heavily on the accessibility of the partners' knowl-edge assets (with such access depending on the partners'motivation to share knowledge), characteristics of the part-ners' relationship, and the governance mechanisms used tocontrol knowledge flow between the partners (Hamel 1991;Inkpen 2000; Li et al. 2008; Phelps 2010; Srivastava andGnyawali 2011). Among these factors, attributes of partnerrelationships are widely examined determinants of knowl-edge sharing (Van Wijk, Jansen, and Lyles 2008). Partner-ships characterized by trust, shared goals, and similar orga-nizational processes promote knowledge sharing becausethese factors can broaden communication scope, nurturecooperative norms, and develop relationship-specific routines(Frazier et al. 2009; Heiman and Nickerson 2004; Lane,Salk, and Lyles 2001; Rowley, Behrens, and Krackhardt2000). Thus, alliance partners can more readily comprehendand leam each other's specialized knowledge and tacitknow-how in the presence of relational ties (Hansen 1999).This stream of literature mainly emphasizes the advantagesof partnerships between compatible firms. However, it doesnot address the negative aspects of such partnerships.

The literature also examines knowledge spillover andmisappropriation as hazards to alliance success (Hamel1991). Scholars contend that firms should manage suchhazards by choosing specific governance mechanisms,including formal and informal safeguards, that align withthe attributes of partnerships (Dyer and Singh 1998; Lavie2006). Informal safeguards such as trust are self-enforcingand essential, because they engender relational behaviorsand lower transaction costs in crafting contracts and insti-tuting monitoring mechanisms (Das and Teng 1998). Infor-mal safeguards can complement formal safeguards in miti-

92 / Journal of Marketing, November 2013

gating opportunism (Inkpen and Tsang 2005; Rowley,Behrens, and Krackhardt 2000). In addition, previousresearch has shown that the severity of hazards in partner-ships focused on knowledge exchange determines gover-nance choices (e.g., equity-based vs. non-equity-basedalliances, repeated partners vs. new partners) (Li et al.2008; Oxley and Sampson 2004). Although this literatureacknowledges the threat of knowledge misappropriationand its implications for governance, prior empkical studieshave examined limited types of governance choices, such asforms of alliances.

Previous research on collaborative buyer-supplier rela-tionships and supplier development programs has high-lighted the increasing importance of knowledge sharing andits impact on partner performance (e.g., Cheung, Myers,and Mentzer 2011; Dyer and Hatch 2007; Kotabe, Martin,and Domoto 2003). Empirical evidence indicates that buy-ers' sharing of operational and technological knowledgeboosts supplier performance. This literature stream hasstressed how customers' actions help suppliers developcapabilities and how knowledge transfer results in superiorsupplier and customer performance. However, these studieshave not adequately addressed the hazards of suppliers'misappropriation of knowledge shared by the customer orother knowledge providers.

This review suggests that several important issuesdeserve attention from marketing scholars. First, previousresearch has inadequately addressed the negative ramifica-tions of collaborative partners possessing similar knowl-edge and skills. Investigating this aspect of partnerships canprovide insights into firms' selection of partners whenentering into collaborative relationships. Second, althoughsupplier collaboration involving a customer (i.e., a triadicpartnership) is an increasingly important supplier manage-ment practice, theoretical expositions and empirical evi-dence on the governance functions of customer involve-ment remain scarce. Examination of supplier collaborationsserving a single customer can broaden academics' under-standing of both the viability of current practices for man-aging such partnerships and the applicability of theoriesused for explaining firm behaviors in dyadic partnerships totriadic partnerships.

Conceptual Framework andHypotheses

Knowledge Sharing In Supplier Coopetition

We define "knowledge sharing" as the intensity of supplierpartners' exchange of valuable knowledge (e.g., technicalskills, product knowledge, manufacturing processes) in theircoopetitive partnership. A high level of knowledge sharingimplies that the supplier partners not only share substantialand valuable knowledge but also help each other leveragesuch knowledge into productive use for collaborative effort.Thus, this definition is consistent with extant literature thatemphasizes not only the exchange of knowledge but alsothe utilization and the quality of the shared knowledge(Haas and Hansen 2007; Hansen, Mors, and L0vâs 2005).

Knowledge that contributes to firms' competitiveadvantage is usually tacit and embedded in organizationalcapabilities (Grant 1996). Thus, suppliers can exploit theirpartners' specialized knowledge and capabilities only whena significant degree of knowledge sharing occurs (Hansen1999). Substantial knowledge sharing in collaborative inter-actions would help supplier partners better understand thecustomer's sophisticated procurement needs and enablethem to formulate solutions to meet those requirements.Moreover, because knowledge sharing reflects supplierpartners' commitment to collaborating toward specificgoals for the customer (e.g., new product development), itis likely to produce positive collaborative outcomes. Previ-ous research has provided ample evidence of the positiveoutcomes of interfirm collaborations in technological inno-vation, new product introduction, and capability develop-ment (Ganesan, Maker, and Rindfleisch 2005; Lane, Salk,and Lyles 2001; Mesquita, Anand, and Brush 2008).

We suggested previously that the occurrence of knowl-edge sharing between supplier partners depends on theirmotivation and the opportunities available in the collabora-tive effort. Our conceptual framework focUses on theopportunity offered by supplier partners' knowledge basecompatibility, because this factor may either facilitate orimpair knowledge sharing. The motivating factors are thecustomer value that the supplier collaboration could gener-ate and the customer's participation in the collaboration.These three factors individually provide a necessary butinsufficient condition for supplier knowledge sharing tooccur. We suggest that it is important to consider their jointeffects. Figure 1 depicts the conceptual framework.

Knowledge Base Compatibility Between SupplierPartners

Firms' knowledge bases are the codified and tacit knowl-edge embedded in organizational capabilities, practices, androutines (Grant 1996). Knowledge base compatibility is adyad-level factor that captures the extent to which the col-laborating parties possess overlapping knowledge and com-patible skills. We suggest that knowledge base compati-bility can facilitate or hinder knowledge sharing betweensupplier partners, and we explain these effects in the fol-lowing subsections.

Benefits of knowledge base compatibility. Although col-laborative partnerships provide firms with opportunities toaccess their partners' knowledge resources and skills,knowledge sharing and collective learning do not necessar-ily occur, because valuable knowledge such as a manufac-turing process is usually sticky, embedded in organizationalroutines, and difficult to transfer (Szulanski 1996). Whenthe shared knowledge is tacit and complex, the recipient canonly learn it from direct experience, firsthand observation,and trial and error (Hansen 1999). In addition, the knowl-edge provider must customize the shared knowledge to thereceiving firm's unique circumstances. Therefore, to over-come obstacles in knowledge exchange, the collaboratingpartners must have similar cognitive frameworks andexperience and use technical languages that both parties canunderstand (Hansen 1999; Heiman and Nickerson 2004).

Knowledge Base Compatibility / 93

FIGURE 1Conceptual Framework

Anticipated CustomerValue

Knowledge BaseCompatibility Between

Supplier Partners

H2and

Customer Participation

Supplier KnowledgeSharing

CollaborativePerformance

Control Variables•Transaction-specific investments between supplierpartners

•Relationship length between supplier partners•Relative power between supplier partners•Relationship closeness between focal supplier and thecustomer

•Focal supplier dependence on the customer•Competitive intensity between supplier partners•Duration of collaborative project•Industry

When supplier partners possess overlapping knowledgeand compatible skills, they can communicate effectively,draw on the partner's expertise and experience, and usesimilar approaches to solve problems and address the learn-ing environment. Therefore, knowledge base compatibilityis likely to result in greater comprehension of partnerknowledge and appreciation of its value. Overlappingknowledge bases also imply greater cross-understanding(i.e., understanding what the partner knows and does notknow) between the boundary-spanning personnel of the col-laborating firms. Cross-understanding can enhance theemergence and elaboration of task-relevant information,resulting in greater learning and superior collaborative out-puts (Huber and Lewis 2010). In contrast, when the knowl-edge bases of the collaborating partners are too dissimilar,high costs and the effort involved in explaining, under-standing, and assimilating the disparate knowledge prohibitthe partners from sharing knowledge effectively.

Our assertion that knowledge base compatibilityincreases learning efficiency is in line with the absorptivecapacity literature (Lane, Koka, and Pathak 2006), whichsuggests that the extent to which new external knowledgerelates to a firm's knowledge bases determines the firm'spotential to absorb new knowledge (Zahra and George2002). Alliance partners' ability to recognize, assimilate,and use partner knowledge depends on the similarity oftheir basic knowledge, organizational structures and poli-cies, and technological interests (Lane and Lubatkin 1998;Lane, Salk, and Lyles 2001 ; Mowery, Oxley, and Silverman1996). This evidence supports our contention that high lev-els of knowledge base compatibility offer opportunities forsupplier partners to achieve greater knowledge sharing.

Hazards arising from knowledge base compatibility.Paradoxically, overlapping knowledge bases may also ham-

per knowledge sharing by exposing supplier partners to thehazard of knowledge misappropriation due to spillover.Spillover is the unintended leakage of knowledge and skillsfrom the knowledge provider to the recipient during day-to-day collaborative activities (Kang, Mahoney, and Tan2009). When suppliers collaborate, their boundary-spanningemployees, such as engineers and technical specialists,interact frequently in professional and social arenas, mak-ing the organizational boundaries permeable (Kale, Singh,and Perlmutter 2000). Although firms may establish formalprocedures and policies to control the now of proprietaryinformation and knowledge, tight control systems can resultin little collective learning and may even compromise theperformance of the partnerships (Makhija and Ganesh1997). Thus, when supplier partners' knowledge basesoverlap significantly, partners could easily acquire andassimilate partner knowledge that is beyond the purview ofthe collaborative arrangement (Li et al. 2008). In addition,the knowledge recipient can use the knowledge obtainedfrom spillover for private benefits, making knowledge shar-ing a risky endeavor for the knowledge provider (Hamel1991; Khanna, Gulati, and Nohria 1998). The hazard ofknowledge misappropriation intensifies when the partnersare also competitors and the collaboration involves technol-ogy exchange (Gulati and Singh 1998). As a result, highlevels of knowledge base compatibility are likely to makesupplier partners alert to partner opportunism and cautiousabout the potential hazards and, thus, reduce the intensityand effectiveness of knowledge exchange effort.

This discussion suggests that supplier partners' knowl-edge base compatibility enhances learning efficiency butalso increases the hazards of knowledge misappropriation.Because these two opposing forces exist simultaneously,the eventual effect of knowledge base compatibility on sup-


plier knowledge sharing depends on other factors pertinentto the supplier collaboration. Previously, we commentedthat in addition to opportunity, motivation is another essen-tial condition for knowledge sharing to occur. Thus, thisinvestigation focuses on two motivating factors—customerparticipation and customer value—that provide supplierpartners with incentives to share knowledge. The study alsoexamines how these factors moderate the effects of knowl-edge base compatibility.

The Moderating Effect of Customer Participation

According to transaction cost economics, informationasymmetry and behavioral uncertainty in collaborative part-nerships lead to opportunistic behaviors such as shirking,cheating, and withholding information (Rindfleisch andHeide 1997). In addition, because collaborative activitiesinevitably involve sharing of valuable tacit know-how andskills, which are difficult to circumscribe and codify, thepotential for opportunism increases. Although transactioncost economics suggests that collaborating firms canrestrain their counterparts' opportunism through formalsafeguards, the ambiguity in assessing the value and flow ofknowledge compromises the effectiveness of formal safe-guards in partnerships focused on knowledge creation andexchange (Gulati and Singh 1998). Whereas previous stud-ies have suggested that social relationships can temper part-ners' misappropriation of knowledge assets (Kale, Singh,and Perlmutter 2000), competitive tension between the sup-plier partners makes the development of trust and commit-ment difficult; thus, supplier partners may need to seekother safeguards.

Case studies of supplier networks have proposed thatpowerful customers can take on managerial and governanceroles in their supplier networks. For example, in Toyota'ssupplier network, Toyota establishes collective goals, nur-tures trust among members, develops a shared identity, andexecutes various managerial functions (Dyer and Nobeoka2000). A powerful customer may also be involved in select-ing qualified suppliers, conducting joint improvementactivities, and implementing control systems (Liker andChoi 2004). Finally, the customer can orchestrate knowl-edge transfer within its supplier network (Dhanaraj andParkhe 2006; Dyer and Nobeoka 2000). Overall, thesequalitative studies suggest that customer participation canenhance supplier capabilities, encourage cooperative behav-ior, and motivate suppliers to work toward collective goals.These collaborative advantages provide customers withincentives to maintain their engagement.

We propose that in a triadic partnership, a customer'sparticipation in its suppliers' collaborative effort couldmotivate the suppliers to share knowledge more intensively.We define customer participation as the customer's engage-ment in its suppliers' collaborative efforts, including suchactions as coordinating collaborative activities, mediatingconflicts between supplier partners, and providing technicalassistance. Customer participation could suppress supplieropportunism through social control processes. First, thecustomer's participation puts social pressure on the supplierpartners, increasing self-deterrence (Rindfleisch and Moor-

man 2003). The suppliers' reputation and repeat businessopportunities with the customer are held hostage (Parkhe1993). Suppliers may fear that the customer's discovery oftheir opportunistic actions against the partner would lead toexclusion from future business (Lavie 2006) and negativeword of mouth among other buying firms regarding thesuppliers' trustworthiness.

Second, intensive interactions between the supplier part-ners and the customer enable collective trust and cooperativenorms to develop in the customer-supplier-supplier triad,as in clan controls (Inkpen and Tsang 2005; Ouchi 1979).Clan controls are essentially a normative process in whichthe members of a system adopt the norms, values, and goalsof the system through socialization processes. Conse-quently, opportunism is suppressed through members' self-control (Ouchi 1979). The network literature has also sug-gested that alliance network members who share commonthird-party ties embrace reciprocity norms and have gener-alized trust, both of which are transferable from one dyad toanother (Phelps 2010). For example, when the customerprovides technical assistance to a supplier, the supplier canreciprocate by offering comparable assistance to its collabo-rating partners within the customer's supplier network(Dyer and Nobeoka 2000).

These arguments suggest that customer participationcould attenuate the exchange hazards that supplier partnersencounter in their collaboration. The avoidance of knowl-edge misappropriation assures suppliers of an equitablereturn on the sharing of valuable knowledge with their part-ners. Therefore, when a customer participates intensively inits suppliers' collaborative efforts, the supplier partners aremotivated to take part in knowledge exchange that leads tothe fulfillment of the customer's procurement needs.

We stated previously that customer participation andknowledge base compatibility are factors that affect suppli-ers' motivation and opportunity to share knowledge. Wepredict that high levels of customer participation mitigatesuppliers' concerns about knowledge misappropriation. Asa result, the learning efficiency afforded by knowledge basecompatibility becomes a sufficient and facultative conditionfor supplier partners to share knowledge. Thus, compati-bility has a positive impact on supplier knowledge sharing.In contrast, when customer participation is minimal, self-deterrence and social controls are absent. As a result, sup-pliers' concern regarding knowledge misappropriation aris-ing from overlapping knowledge bases is elevated, andknowledge base compatibility alone is not a sufficient con-dition for supplier knowledge sharing to take place.

H|: Under high levels of customer participation, knowledgebase compatibility is positively related to knowledge shar-ing between supplier partners. Under low levels of cus-tomer participation, knowledge base compatibility is notrelated to supplier knowledge sharing.

The Moderating Effect of Anticipated CustomerValue

Previous research has shown that firms can capitalize ontheir partnerships to generate both collaborative and partner-specific value (Kale, Dyer, and Singh 2002). Collaborative


value refers to the value the partner firms create collec-tively, such as innovative outputs and new products (Mad-hok and Tallman 1998). In supplier collaborations, supplierpartners pursue a common goal of producing superior prod-ucts, services, or solutions for the customer. Thus, collec-tive value manifests as customer benefits created, such ascost reduction, quality improvement, or short inventorycycle. We refer to the value that supplier collaborations gen-erate for the customer as "anticipated customer value." Wefocus on anticipated customer value as a motivating factorfor supplier partners to participate in knowledge sharing,because this factor is closely related to the future businessthat the supplier partners may gain from the customer.

Knowledge sharing is a costly and risky endeavor withuncertain results, and in projecting the long-term payoffs,supplier partners must assess their collaboration's strategicvalue to the customer. When the supplier collaboration cre-ates significant customer value, the supplier partners arelikely to gain repeated and expanded business from the cus-tomer. According to expectancy theory, the amount of efforta person exerts in a given task or course of action dependson the value he or she assigns to the expected outcome (VanEerde and Thierry 1996). Thus, the anticipated rewards ofspecific actions motivate a person to participate in suchactions. Likewise, game theory posits that the behaviors oractions of organizational actors result from their evaluationand comparison of the expected payoffs of alternativestrategic options (Parkhe 1993). Previous research has alsoshown that tangible incentives strongly motivate managersto share knowledge with members in the same organiza-tional unit (Quigiey et al. 2007). In keeping with these theo-retical perspectives, we suggest that the customer value thatsupplier partners expect their collaboration to create pro-vides incentives for them to take part in knowledge sharing.When anticipated customer value is high, supplier partnersforesee greater future business opportunities. Thus, supplierpartners are more eager to put effort into knowledgeexchange activities in the current collaborative effort.

We noted previously that anticipated customer valueand knowledge base compatibility are complementary fac-tors that affect supplier partners' motivation and opportu-nity to share knowledge. Thus, we predict that these twofactors have an interaction effect. Specifically, supplierpartners that anticipate high customer value are more likelyto focus on the leaming efficiency afforded by overlappingknowledge bases and on how to exploit this advantage. As aresult, knowledge base compatibility should have a positiveimpact on supplier knowledge sharing. In contrast, supplierpartners that anticipate low customer value would have lessincentive to participate in knowledge sharing. They mightbecome more concerned with the misappropriation risksassociated with overlapping knowledge bases, renderingknowledge base compatibility insufficient to drive supplierknowledge sharing.

H2: Under high levels of anticipated customer value, knowl-edge base compatibility is positively related to supplierknowledge sharing. Under low levels of anticipated cus-tomer value, knowledge base compatibility is not relatedto supplier knowledge sharing.

We also argue that when supplier partners anticipatethat collaboration will create significant customer value,they will consider the customer's participation more accept-able and legitimate because the customer has a high stake inthe collaborative arrangement. As we discussed previously,customer participation is likely to increase social interac-tions among the employees ofthe customer and the supplierpartners, and as a result, cooperative norms develop in thecustomer-supplier-supplier triad. More importantly, cus-tomer participation in terms of technical support, sharing ofproduct information and tacit know-how, and mediation ofsupplier conflicts can improve the effectiveness of collabo-rative interactions. Therefore, high levels of customer valuecoupled with high levels of customer participation stronglymotivate supplier partners to engage in knowledgeexchange activities. In this situation, supplier partners arelikely to focus on exploiting the learning opportunities pro-vided by knowledge base compatibility. As a result, the sup-plier partners would consider the facilitative power ofknowledge base compatibility to outweigh the potentialhazards, leading to a positive effect of compatibility on sup-piier knowledge sharing.

In contrast, when customer participation is low, supplierpartners would have little incentive to take part in knowl-edge sharing because of the absence of adequate safeguards.As a result, even when suppliers anticipate that customervalue to be generated by collaborative effort is likely to behigh, they would fail to exploit the benefits of knowledgebase compatibility because of a fear of potential hazards,resulting in a nonsignificant relationship between knowl-edge base compatibility and supplier knowledge sharing.

H3a: Under high levels of anticipated customer value, knowl-edge base compatibility is positively related to supplierknowledge sharing when customer participation is highand not related to supplier knowledge sharing when cus-tomer participation is low.

We also suggest that when supplier partners anticipatethat their collaborative effort will have limited strategicvalue to the customer, they foresee the future payoffs andbusiness opportunities arising from the collaboration asuncertain and as not justifying the risks involved in workingwith a competitor. Thus, supplier partners are not willing toparticipate in costly and risky knowledge exchange activi-ties. In addition, because the supplier partners perceive thecustomer as having a low stake in the collaborative arrange-ment, they are likely to view any participatory actions of thecustomer as unnecessary intervention and may question thecustomer's intentions. Customer participation often requiressuppliers to adapt their operational procedures, such as set-ting up a special task force and information-sharing systemto support the customer's involvement, and in this case, thesuppliers may view such adaptations as unjustifiable whenthey anticipate relatively low customer value and uncertainfuture payoffs. Prior research has suggested that when buy-ers' excessive intervention lacks legitimacy, suppliers reactnegatively—for example, by engaging in noncooperativebehavior and exhibiting negative sentiments (Heide,Wathne, and Rokkan 2007).


Overall, we propose that when supplier partners antici-pate that their collaborative effort could only create limitedcustomer value, they will be reluctant to accept customerparticipation and may even react against it. Thus, lowanticipated customer value coupled with high customer par-ticipation provides minimal or even negative motivation forsupplier partners to share valuable knowledge with eachother. With negative motivation, suppliers are not onlyreluctant to participate in knowledge sharing but also unap-preciative of the value of their partners' knowledge andskills. Such negative motivation and sentiments are likely tocause suppliers to view overlapping knowledge bases withtheir partners unfavorably, escalating their concerns ofknowledge misappropriation. Suppliers would becomemore averse to the damage that results from knowledgemisappropriation, and their negative attitude toward knowl-edge sharing could induce reciprocal responses from theirpartner. In these situations, knowledge base compatibilityshould be negatively related to supplier knowledge sharing.

H3I,: Under low levels of anticipated customer value, knowl-edge base compatibility is negatively related to supplierknowledge sharing when customer participation is highand not related to supplier knowledge sharing when cus-tomer participation is low.

In the following sections, we present the research meth-ods and results of two empirical studies. In Study 1, weconducted a scenario experiment with responses from stu-dents in an executive master's program in business adminis-tration (MBA). In Study 2, we used mail surveys to collectdata from executives working in technology-based firms.

Study 1In Study 1, we used a between-subjects scenario experimentto test the proposed hypotheses. Previous studies on distri-bution channels have employed scenario-based experimentsto examine managers' attitudes and decisions under differentcircumstances and roles (e.g., Ganesan et al. 2010; Wathne,Biong, and Heide 2001). Use of an experiment can controlfor extraneous factors that may confound the results, suchas supplier and industrial characteristics.

Expérimentai Design and Procedures

We recruited 121 participants enrolled in an executiveMBA program at a large public university. All were full-time executives or managers in organizations in the south-western United States. On average, they had eight years offull-time work experience, and 68% were men. A majorityof the respondents worked in information technology, phar-maceuticals, and defense industries. We employed a 2(anticipated customer value: high vs. low) x 2 (knowledgebase compatibility: high vs. low) x 2 (customer participa-tion: high vs. low) between-subjects factorial design. Eachparticipant responded to a scenario randomly drawn fromeight different scenarios that corresponded to a combinationof the three factors manipulated at either the high or the lowlevel. To assess the realism and relevance of the scenario,we pretested it among another group of participants in thesame executive MBA program. The manipulation for each

factor was grounded in its conceptual meaning and reflectedkey characteristics of the latent constructs. Web AppendixWl presents the manipulations used in Study 1.

In all conditions, participants assumed the role of thedirector of product development in a fictitious microchipmanufacturer, MicroTech. They then read the descriptionsof the key aspects of the collaboration in which the manipu-lations for the three factors were embedded. After readingthe scenario, participants indicated how willing they wereto share knowledge with a competitor participating in thecollaboration (Innova) and then responded to the manipula-tion checks. We captured participants' intention to shareknowledge with three items rated on nine-point scales: (1)"How willing is your company to share knowledge withInnova in this collaboration?" (1 = "not willing at all," and9 = "very willing"), (2) "How motivated is your companyto share knowledge with Innova?" (1 = "not motivated atall," and 9 = "highly motivated"), and (3) "How likely is itthat your firm will take the initiative to share knowledgewith Innova?" (1 = "very unlikely," and 9 - "very likely").Reliability of the measure as assessed by Cronbach's alphawas .87. In the scenario experiments, we could measureonly participants' intention to share knowledge, not theiractual knowledge sharing.

Manipulation Checks and Assumptiorts Testing

We used a full factorial analysis of variance (ANOVA) toperform the manipulation checks. The results indicate thatthe manipulations were effective: customer value (7.91 vs.5.98; F = 42.00, p < .001), knowledge base compatibility(6.93 vs. 5.97; F = 11.39,/7 < .001), and customer participa-tion (6.28 vs. 4.92; F = 17.23,/? < .001). In each check, onlythe focal variable had a significant main effect on thedependent measure, and no other main or interaction effectswere significant. We tested the assumption that knowledgebase compatibility has a positive effect on both learningefficiency and knowledge spillover risk. We used a singleitem to assess learning efficiency ("The employees ofInnova can learn and acquire knowledge from MicroTecheffectively") and a single item to assess spillover risk ("Pro-prietary knowledge can be leaked from MicroTech toInnova unintentionally"). We used a seven-point Likertscale (1 = "strongly disagree," and 7 - "strongly agree") tomeasure both items. The results of a one-way ANOVAshow that learning efficiency (6.47 vs. 5.02; F = 12.52,/? <.01) and spillover risk (7.03 vs. 4.17; F = 21.11, /? < .01)were greater in the high knowledge base compatibility con-dition than in the low knowledge base compatibility condi-tion. A single item also tested the assumption that customerparticipation deters suppliers' opportunistic behavior ("Thecustomer's involvement in its suppliers' collaborative activ-ities may prevent the suppliers from acting opportunisti-cally, such as stealing their partner's valuable knowledge").We measured this item on a seven-point Likert scale (1 ="strongly disagree," and 7 = "strongly agree"). The results ofthe one-way ANOVA show that perceived opportunism waslower in the high customer participation condition than inthe low customer participation condition (6.04 vs. 5.01; F -7.05,/7<.01).


ResultsThe results of a 2 x 2 between-subjects ANOVA indicatethat both anticipated customer value (F = 22.35, p < .01)and customer participation (F = 19.03, p < .01) have a sig-nificant and positive effect on suppliers' intent to shareknowledge. As predicted in H2, the two-way interaction ofanticipated customer value x knowledge base compatibilityis significant (F = 6.45, p < .01). However, the interactionof customer participation x knowledge base compatibility isnot significant (F = .29, n.s.). Thus, the two-way interactionhypothesized in Hi is not supported. More important, theresults indicate a significant three-way interaction betweenanticipated customer value, knowledge base compatibility,and customer participation (F = 1.69, p < .01). As Figure 2illustrates, this three-way interaction involves two two-wayinteractions of opposing patterns.

Under the high customer value condition (right panel),the knowledge base compatibility x customer participationinteraction is significant (F = 4.01,/? < .05), consistent withH3a. To understand this interaction further, we conducted asimple effects analysis. The results indicate that when cus-tomer participation is high, knowledge-sharing intent is sig-nificantly greater in the high knowledge base compatibilitycondition than in the low knowledge base compatibilitycondition (6.67 vs. 5.75; F = 9.43, p < .01). In contrast,when customer participation is low, we find no significantdifference in knowledge-sharing intent between the highand low compatibility conditions (5.52 vs. 5.50, n.s.).

An opposing pattern of results appears in the lowcustomer value condition .3 When customer value is low (Fig-ure 2, Panel A), the knowledge base compatibility x customerparticipation interaction is significant (F = 4.46, p < .05), con-sistent with H3b. The follow-up simple-effects analysis indi-cates that when customer participation is high, knowledge-sharing intent is significantly lower in the high compatibilitycondition than in the low compatibility condition (3.94 vs.5.16; F = 11.07,/? < .01). In contrast, when both customervalue and customer participation are low, results indicate nosignificant difference in knowledge-sharing intent betweenthe high knowledge base compatibility and the low knowl-edge base compatibility conditions (5.18 vs. 5.02, n.s.).

Discussion

Study 1 shows how knowledge base compatibility, cus-tomer participation, and anticipated customer value jointlyaffect suppliers' intention to share knowledge with theirpartner in a coopetitive partnership. The results suggest thatcustomer participation can temper suppliers' concern forknowledge misappropriation. In addition, suppliers viewoverlapping knowledge bases as both facilitating knowl-edge exchange and creating the misappropriation hazard.

The significant three-way interaction shows that whenboth customer value and customer participation are high,knowledge base compatibility has a positive effect on sup-pliers' intention to share knowledge. This finding is consis-

two opposing patterns of two-way interactions betweencustomer participation and knowledge base compatibility mayexplain the insignificant result for H|.

FIGURE 2Study 1 : Anticipated Customer Value x

Knowledge Base Compatibility xCustomer Participation Interaction

A: Low Customer Value Condition

7 -,

£ 6-O)

cCO 5 .

4 -

•5.2a3(0

7-,

£ 6O)

COa>O)-a

oc

Q.

3

5.18

3.94

I High customer participation' Low customer participation

Low HighKnowledge Base Compatibility

B: High Customer Value Condition

5.75

5.50

6.67

5.52

^mm High customer participation• • • Low customer participation


tent with our conjecture that when both customer value andcustomer participation are high, suppliers focus on theleaming efficiency benefits that arise from the presence ofoverlapping knowledge bases with the partner rather thanfocusing on the misappropriation hazard. More important,when customer participation is high but customer value islow, knowledge base compatibility has a negative effect onsuppliers' intention to share knowledge. This result suggeststhat high customer participation combined with low cus-tomer value could trigger suppliers' negative motivation.Therefore, suppliers are likely to perceive the misappropria-tion hazard as detrimental, superseding the advantages thatstem from knowledge base compatibility. In addition, theresults show that when customer participation is low,regardless of the level of customer value, the effect of


knowledge base compatibility on suppliers' intention toshare knowledge is insignificant. This result suggests that inthe absence of a social control, suppliers have a low incen-tive to share knowledge.

Overall, the empirical results in Study 1 support most ofour hypotheses. The findings identify the conditions underwhich customer participation is beneficial, benign, or harm-ful. The findings also show that knowledge base compati-bility can have both positive and negative impacts on sup-plier knowledge sharing—conditional on the level ofmotivating factors such as customer participation andanticipated customer value.

Using experiments as the method in Study 1 ensures thatour empirical results have high internal validity by keepingkey extraneous factors constant across experimental condi-tions. However, the scenarios we used might be too restric-tive, diminishing the generalizability of the findings. Toaddress this limitation, in Study 2 we conducted a mail sur-vey of supplier firms. The use of primary company data totest the research hypotheses can demonstrate the robustnessof our empirical findings and increase their generalizability.

Study 2

Sample and ProceduresIn Study 2, we used mail surveys to collect data fromoptics, computing, and automotive firms to test thehypotheses. We chose these technology-based industriesbecause formal and informal collaborations among firmsare common and previous studies have examined knowl-edge transfer and collaborations in these industries (Dyerand Hatch 2007; Ganesan, Malter, and Rindfleisch 2005).Collecting data from supplier firms of various sizes acrossmultiple industries enables us to assess the generalizabilityof the findings in Study 1. In line with previous studies ofinterfirm collaborations (Rindfleisch and Moorman 2001;Rowley, Behrens, and Krackhardt 2000), we constructed adatabase of manufacturers from the membership directoriesof the largest associations for each industry (e.g., AmericanPrecision Optics Manufacturers Association, OriginalEquipment Suppliers Association). The sampled firms areall manufacturers in the private sector and thus excludegovernment organizations, trading companies, distributors,and services companies.

After compiling the sampling frame, we identified aspecific key informant in each firm by searching companydatabases, social network websites, and company websites.Informants were vice presidents, directors of research anddevelopment or manufacturing, or people in executive posi-tions who had an extensive understanding of their firms'collaborative activities with other firms. Our final samplingframe consisted of 610 manufacturers for which we couldidentify informants who were likely to be involved in theirfirm's collaborations with partners.

Before mailing the surveys, we contacted each infor-mant by telephone to solicit voluntary participation andassess whether the informant possessed the relevant knowl-edge. If the informant was not involved in the firm's collab-orative efforts, we asked him or her to provide the contact

information of an appropriate person. We then mailed eachinformant a cover letter, questionnaire, and postage-paidreply envelope as well as a $10 gift card as an incentive.Three weeks later, we sent a reminder letter, followed bytelephone calls. We mailed a second set of surveys to infor-mants who did not respond within eight weeks. After col-lecting the surveys and discarding those with substantialmissing data, we had a final sample of 110 of the 610 firmsin the sampling frame, representing a response rate of 18%.This response rate is comparable to studies of distributionchannels using mail surveys (Heide, Wathne, and Rokkan2007; Noordhoff et al. 2011). The size of the sample firmsranged from fewer than 200 employees to more than10,000, and annual sales ranged from less than $5 million tomore than $1 billion. Among the sample firms, 54% camefrom the optics industry, 13% from the computing industry,and 33% from the automobile industry.

We assessed potential nonresponse bias by comparingearly and late respondents and found no significantdifferences in means for key constructs, suggesting thatnonresponse bias was not a major problem (Armstrong andOverton 1977). In response to our validity check, respon-dents reported that they were highly involved with andknowledgeable about the collaborative project (M = 5.8 ona seven-point scale) and had worked for their firms for anaverage of 15.4 years. Of the respondents, 75% were chiefexecutive officers, presidents, or directors.

In the survey, we asked the informants to report theirexperience with a recently completed or nearly completedcollaborative project with assessable performance indica-tors between their firm (i.e., the focal supplier) and anothersupplier firm (i.e., the partner supplier) that they viewed asa direct or indirect competitor. The collaborative projectneeded to meet several criteria. First, the purpose of the col-laboration was to provide services or goods to a specificcustomer firm. Second, the partner firm was either a directcompetitor (both firms supplied the same products to thecustomer) or an indirect competitor (with overlapping targetmarkets). Third, the ownerships of the informants' firms,the partner firm, and the customer were independent.Because either the customer or the supplier could initiatethe supplier collaboration, we assessed whether the level ofsupplier knowledge sharing differed between these twogroups. The ANOVA results indicated no significant differ-ence (Mcustomer initiated = 4.81 VS. MsuppUer initiated = 4.68,n.s.) in the sample.

Measures and Validation

We developed and refined the measurement scales usingfield interviews and a pretest of the survey with purchasingpersonnel and marketing faculty at a public university. TheAppendix presents specific items of these measures and theirreliabilities. Descriptive statistics of the measures appear inWeb Appendix W2. The assessment of scale properties andcommon method variance appears in Web Appendix W3.

Dependent variables. Supplier knowledge sharing refersto the intensity of supplier partners' exchange of valuableknowledge that can be used for collaborative effort. Thus,the operationalization of this construct assesses three ele-


ments of knowledge sharing: the intensity of the exchange,the value/quality of the shared knowledge, and the use ofsuch knowledge. Because no existing scale for knowledgesharing assesses all three of the aforementioned elements,we developed a seven-item scale after reviewing variousstudies in the learning alliances literature. We assessed col-laborative project performance, which refers to the extent towhich the collaborative project achieves target goals andfulfills the customer's procurement needs from the perspec-tive of the focal supplier. Because of the diverse nature ofthe collaborative projects among the firms sampled, wecould not determine specific project goals a priori. There-fore, we developed a three-item scale to assess this con-struct without referring to specific project goals.

Independent variables. Knowledge base compatibilityrefers to the extent to which the collaborating parties possesssimilar and compatible knowledge, skills, and capabilities thatenable them to work together effectively. No existing scalesfor this construct appear in the literature. The most relevantone is the scale from Rindfleisch and Moorman (2001) thatmeasures knowledge redundancy between alliance partners.Therefore, we developed a five-item scale to assess this con-struct. Customer participation refers to the customer's activeinvolvement in its suppliers' collaboration, such as gather-ing production information, mediating supplier partners'conflicts, providing technical assistance, and coordinatingcollaborative activities. In the absence of an existing scale,we developed a five-item scale that assessed the customer'sfacilitative and supportive role in collaborative projectsbetween the suppliers. Anticipated customer value refers tothe tangible value the focal supplier expects its collaborativeeffort to create for the customer. We used a new three-itemscale to assess the focal supplier's expected impact of thecollaborative project on the customer's profitability and theextent to which the project was valuable to the customer.

Control variables related to alternative governancemechanisms. To rule out the possibility that altemative gov-ernance mechanisms might motivate supplier partners toparticipate in knowledge sharing, we controlled for two fac-tors: supplier partners' relationship length and transaction-specific investments. Prior research has shown that collabo-rating partners' relationship length and transaction-specificinvestments are safeguards that mitigate opportunism andencourage cooperative behaviors (Kale, Singh, and Perl-mutter 2000; Rindfleisch and Moorman 2001). Firms usegovernance modes such as equity-based alliances (high lev-els of transaction-specific investment) to cope with severehazards in knowledge exchange partnerships (Li et al. 2008;Sampson 2004). Relationship length fosters trust and rela-tional norms and reduces information asymmetry, whiletransaction-specific investments act as mutual hostages thatinduce self-deterrence. Thus, these two factors serve asgovernance functions similar to customer participation.Adding these control variables enables us to determinewhether customer participation has unique explanatorypower after the effects of these safeguards are partialed out.We assessed relationship length in terms of number of yearsthe supplier partners had business dealings before theexamined collaborative project. We measured transaction-

specific investments using a two-item scale that capturedthe level of supplier partners' investment of resources in thecollaborative project.

Additional control variables. We controlled for two fac-tors related to the relationship between the focal supplier andthe customer: relationship closeness and the focal supplier'sdependence on the customer The nature of the customer-focal supplier relationship might motivate the focal supplierto make a greater effort in the collaborative project toensure relationship continuance. We measured relationshipcloseness using a three-item scale adapted from Rindfleischand Moorman (2001) and focal supplier dependence using athree-item scale adapted from Ganesan (1994). We con-trolled for competitive intensity between supplier partnersby assessing the extent of competition between them acrossdifferent product markets as well as for the customer's busi-ness using a new three-item scale. Because prior researchhas indicated that competitive intensity increases exchangehazards between collaborating partners (Sampson 2004),this factor would reduce supplier partners' willingness toshare knowledge. In addition, we controlled for the relativepower between the supplier partners in the collaborativeeffort, because this factor might affect suppliers' motivationto share knowledge. We used a single item to assess theasymmetric benefits that the supplier partners obtainedfrom the collaborative project. We also controlled for theduration of collaborative projects, assessing project dura-tion in terms of number of months. Finally, we controlledfor industrial heterogeneity using a binary categoricalvariable (optics related vs. automobile). Because the samplefirms from the computing industry were related to the pro-duction of optical drives, we combined firms from thesetwo industries into one group.

Two-Stage Least Squares Regression Results

To test the hypotheses, we conducted a two-stage leastsquares (2SLS) regression analysis. This estimation proce-dure enabled us to correct for endogeneity through the useof instrumental variables. Endogeneity is a common estima-tion problem due to omitted variables and self-selection(Básele 2008). In our conceptual model, customer participa-tion may be endogenous and thus may lead to biased coeffl-cients if estimated through ordinary least squares (OLS)regression. In the first stage, we regressed customer partici-pation and all interaction terms involving customer partici-pation on the instrumental variables and the controlvariables. We chose three instrumental variables^ on the

"̂ To meet the condition that the number of instruments must begreater than the number of endogenous variables, we created addi-tional instrumental variables by multiplying the three instrumentswith customer value and knowledge base compatibility. We testedthe validity of the set of instrumental variables using the Sargantest (Bascle 2008). The result indicated that the set of instrumentalvariables were exogenous (x^ = 7.49, p > .10). We assessed thestrength of the set of instrumental variables using the Anderson-Rubin test that applies to multiple endogenous variables (Bascle2008; Wooldridge 2010). The result showed that the set of instru-mental variables was strong (F = 2.27,p < .05) (i.e., the error termis uncorrelated with all instruments).


basis of their relevance to customer participation: relation-ship length between the customer and the focal supplier, thecustomer's dependence on the focal supplier, and whetherthe customer initiated the collaborative project.

In the second stage, we replaced the original values ofcustomer participation and the interaction terms involvingcustomer participation with predicted values estimated fromthe first-stage regression. Subsequently, we regressed sup-plier knowledge sharing on the antecedents, interactionterms, and control variables. We assessed endogeneity ofcustomer participation and the related interaction termsusing the Wu-Hausman test (Wooldridge 2010). The resultsindicate that at least one of these variables was endogenous(F = 1.58, p < .10), suggesting that 2SLS regression is theappropriate estimation procedure. We report the results inTable 1.

Before the 2SLS regression analysis, we mean-centeredthe independent variables that were part of the interactionterm and then multiplied these variables to create the two-and three-way interaction terms (Aiken and West 1991). Asa result, for all the variables that form the interaction terms,the regression coefficients are interpreted as simple (condi-tional) effects. These coefficients represent the effect of anindependent variable on knowledge sharing at the mean

TABLE 12SLS Results: Study 2

Variables

Independent VariablesAnticipated customer valueCustomer participationKnowledge base compatibilityCustomer value

X customer participationCustomer value

X knowledge base compatibilityCustomer participation

X knowledge base compatibilityCustomer value

X customer participationX knowledge base compatibility

Control VariablesTransaction-specific investmentsRelationship length between

supplier partnersRelative power between

supplier partnersCompetitive intensity between

supplier partnersRelationship closeness between

focal supplier and customerFocal supplier dependence on

the customerProject durationIndustry

Adjusted R-square

SupplierKnowledge Sharing

.05 (.41)-.05 (.31)-.15 (1.17)

-.11 (.99)

.23 (2.08)**

.15 (1.76)*

.18 (2.63)**

.50 (6.26)***

.39 (1.30)

-.13 (1.65)*

.12 (1.55)

-.02 (.2)

.18 (1.86)*-.55 (1.32)

.11 (.47)

.52

*p<.10.**p < .05.***p < .01.Notes: Numbers are unstandardized coefficients with t-values in

parentheses.

value of Other independent variables in the interaction term.We first assessed the significance of the three-way interac-tions in the regression model and then decomposed the sig-nificant interactions and tested the simple effects embeddedin the significant interactions.

Supplier knowledge sharing. The 2SLS regressionresults indicate that the main effects of knowledge basecompatibility (b = - .15 , n.s.), customer participation (b =- .05 , n.s.), and anticipated customer value (b = .05, n.s.) arenot significant. As we hypothesized, the customer value xknowledge base compatibility x customer participationthree-way interaction is highly significant (b = .18; t = 2.63,p < .01), suggesting that the knowledge base compatibility xcustomer participation interaction differs between the highand low customer value conditions. To determine the natureof the interaction effect, we tested the simple effect ofknowledge base compatibility on knowledge sharing condi-tional on different levels of customer participation and cus-tomer value. We defined high and low levels of customerparticipation as one standard deviation above and below themean, respectively. Similady, high and low customer valuewas one standard deviation above and one below the mean,respectively. We then calculated the slopes (coefficients)and standard errors of knowledge base compatibility at highand low levels of customer participation along with highand low levels of customer value. We tested the signifi-cance of these coefficients using t-tests. Figure 3 presentsthese simple effects.

As we predicted in H3a, when both customer participationand customer value are high, knowledge base compatibilityincreases knowledge sharing between supplier partners (b =.63; t = 4.04,/? < .01). In contrast, as we predicted in H31,,when customer participation is high but customer value islow, knowledge base compatibility is negatively associatedwith knowledge sharing (b - - .48; t = 2.01, p < .05). Thesedivergent responses show that the customer's involvementin the suppliers' collaborative effort can be either construc-tive or destructive. The results are consistent with our con-jecture that customer participation could mitigate suppliers'concern for opportunism and motivate them to take advan-tage of having overlapping knowledge bases with the part-ner or that it could arouse suppliers' negative sentimentsand discourage them from sharing knowledge due to misap-propriation risk. Nevertheless, when customer participationis low, knowledge base compatibility has no significanteffect on knowledge sharing across different levels of cus-tomer value (b - - .32 when customer value is low, and b =-.41 when customer value is high). This result suggests thatin the absence of customer participation as a social control,supplier partners lack the motivation to exploit their over-lapping knowledge bases to facilitate knowledge transfer.

In addition to the significant three-way interactions, theinteraction between customer participation and knowledgebase compatibility is marginally significant and positive(b = .15; t = 1.76, p < .10), suggesting that the simple slopeof knowledge base compatibility is greater in the high cus-tomer participation condition than in the low customer par-ticipation condition, while customer value is at the mean.This result supports H^. Likewise, the interaction between

Knowledge Base Compatibility /101

FIGURE 3Study 2: Anticipated Customer Vaiue x

Knowledge Base Compatibiiity xCustomer Participation interaction

A: Low Customer Value Condition

o>c

0)0)•o

oc

aa

b = -.32

b = -.48(p<.05)

^ ^ High customer participation• • • « Low customer participation

Low HighKnowledge Base Compatibiiity

B: High Customer Vaiue Condition

c

0)0}

T3

oc

Q.a3

b = .61{p<-01)

(n.s.)

^ ^ High customer participation• • • • Low customer participation


customer value and knowledge base compatibility is signifi-cant (b = .23; t = 2.08, p < .05), suggesting that the simpleslope of knowledge base compatibility is greater in the highcustomer value condition than in the low customer valuecondition, while customer participation is at the mean.Thus, H2 is supported.

Finally, among the control variables, transaction-specificinvestments have a significant and positive effect on sup-plier knowledge sharing (b = .50; t = 6.26, p < .01). Thisresult suggests that transaction-specific investments act assafeguards that can promote greater knowledge sharing. Inaddition, the focal supplier's dependence on the customer(b = .18; t = 1.86,/? < .10) and the relative power betweenthe supplier partners (b = -.13; t = 1.65, p < .10) are also

marginally associated with supplier knowledge sharing.These results indicate that knowledge sharing is greaterwhen the focal supplier is substantially dependent on thecustomer for revenue and profits and when the focal sup-plier gains greater benefits from the collaborative effortthan its partner does. No other control variable has a signifi-cant effect.

Post Hoc Anaiysis

Customer participation. In a post hoc analysis, we usedOLS regression to find factors that increase the customer'sparticipation in suppliers' collaborative projects. Weregressed customer participation on all the control variablesas well as three customer-related variables: the length of thecustomer's relationship with the focal supplier, the depen-dence of the customer on the focal supplier, and whether thecustomer (vs. suppliers) initiated the collaborative project(the same as the instrumental variables used in the 2SLSregression analysis). The results indicate that the level ofcustomer participation is greater in supplier collaborationsinitiated by the customer than in those initiated by suppliers(b = 1.11; t = 3.20, p < .01). In addition, the customer'sdependence on the focal supplier (b = .32; t = 2.31,p < .05)and the length of the customer's relationship with the focalsupplier (b =-.90; t = I . 8 9 , / J < .10) are related to customerparticipation. No control variable has a significant effect.

Collaborative performance. Although we do not for-mally hypothesize the effect of supplier knowledge sharingon collaborative project performance, we tested this impor-tant relationship. Using OLS regression, we regressed col-laborative project performance on supplier knowledge shar-ing and all the control variables. The results indicate thatknowledge sharing has a significant and positive effect onperformance (b = .25; t = 2.12,/? < .05; R2 = .41; F = 6.24).Among the control variables, the focal supplier-customerrelationship closeness (b = .52; t = 4.40, p < .01) and sup-plier partners' trans action-specific investments (b = .27; t =1.78,p < .10) have significant and positive effects. No othercontrol variable has a significant effect. These results indi-cate that knowledge sharing helps supplier partners achievegreater collaborative performance and that the performanceincrement can be explained only partially by supplier part-ners' resource commitment and the focal supplier's rela-tionship with the customer.

DiscussionCollaborations among competing suppliers within a cus-tomer's supplier network have become increasingly importatitto improve supplier productivity and meet the customer'ssophisticated procurement needs. However, existing empiri-cal studies on such collaborations are scarce and confinedto reports of best practices and analysis of limited cases(e.g.. Dyer and Nobeoka 2000; Wu, Choi, and Rungtu-sanatham 2010).

To advance the understanding of this phenomenon foracademics and managers, this research investigates the col-laborative effort between two competing suppliers serving adownstream OEM customer in a business-to-business con-


text. We focus on supplier partners' knowledge sharing incollaborative efforts and examine a few major antecedents.In our conceptual framework, knowledge base compati-bility between the supplier partners represents the opportu-nity to share knowledge effectively with each other,because this factor helps them overcome communicationand leaming barriers. In addition, the customer value thecollaboration is expected to create and the customer's par-ticipation in the collaboration are two motivating factorsthat moderate the impact of knowledge base compatibilityon supplier partners' knowledge sharing. Thus, these threefactors reflect the opportunity for and motivation of sup-plier partners to share valuable knowledge in their collabo-rative effort aimed to fulfill the customer's procurementrequirements.

We examine these three factors jointly because each ofthem provides a necessary but insufficient condition forsupplier knowledge sharing to occur. We tested the researchhypotheses using data gathered from a scenario-basedexperiment and a survey study. In these two studies, wecontrolled for key extraneous factors including those relatedto the relationship between the supplier partners and therelationship between the focal supplier and the customer.Similar patterns of results appear across the two studies.Therefore, our empirical findings are robust and valid.

The results from the two studies provide strong empiri-cal evidence in support of the hypothesized three-way inter-action between knowledge base compatibility, customerparticipation, and anticipated customer value. Specifically,the results show that knowledge base compatibility has apositive impact on supplier partners' knowledge sharingwhen both anticipated customer value and customer partici-pation are high. In contrast, knowledge base compatibilityis negatively associated with knowledge sharing whenanticipated customer value is low but customer participa-tion is high. In addition, when customer participation is low,irrespective of the level of anticipated customer value,knowledge base compatibility has no significant effect onsupplier knowledge sharing. These findings demonstratethat knowledge base compatibility can help or hinder sup-plier knowledge sharing under different motivating condi-tions. Importantly, our findings reveal that in a supplier col-laboration serving a common customer, customer-relatedfactors such as customer participation and anticipated cus-tomer value can alter the impact of knowledge base com-patibility on supplier knowledge sharing from positive tonull to negative. Previous research on knowledge transfer incollaborative partnerships has mainly used secondary datasuch as patent citations as a proxy for knowledge transfer(e.g., Mowery, Oxley, and Silverman 1996). In this regard,the present study provides new and direct evidence of boththe beneficial and detrimental consequences of supplierpartners' overlapping knowledge bases for knowledge shar-ing. In addition, this study identifies the boundary condi-tions of the effects of knowledge base compatibility.

Our research sheds light on the governance role of thecustomer in coopetitive supplier partnerships. We argue thatthe hazards arising from overlapping knowledge bases dis-courage supplier partners from participating in knowledge

exchange activities. Drawing on extant literature, we con-jecture that customer participation in the suppliers' collabo-rative effort could engender collective trust and cooperativenorms within the triadic partnership (Wu, Choi, and Rung-tusanatham 2010). We also suggest that supplier partners'reputation and future business opportunities lead to self-control (Parkhe 1993). These social processes would miti-gate supplier partners' concerns for opportunism. As aresult, customer participation motivates supplier partners tofocus on exploiting the leaming efficiency advantages ofcompatible knowledge bases rather than on the potentialhazards, resulting in greater supplier knowledge sharing.Although the two empirical studies do not explicitly exam-ine the occurrence of the micro govemance processes, theresults after ruling out the effects of altemative govemancemechanisms, including transaction-specific investments andprior relationships, are consistent with our predictions.Therefore, customer participation has a unique explanatorypower for supplier knowledge sharing, and the results pro-vide tentative evidence in support of the govemance role ofcustomer participation.

This research also identifies the conditions that couldincrease a customer's participation in its suppliers' collabo-rative efforts. The post hoc analysis results suggest thatwhen the customer initiates the supplier collaboration, thecustomer depends on the suppliers for resources, and therelationship length between the customer and the suppliersis relatively short, the customer is more likely to engage insuppliers' collaborative arrangements. Overall, this researchnot only demonstrates the importance of customer partici-pation in supplier collaboration but also provides prelimi-nary evidence on factors that drive such participation.

Finally, the post hoc analysis shows that supplierknowledge sharing significantly contributes to collaborativeperformance in terms of fulfilling the customer's procure-ment needs. This performance enhancement can only bepartially explained by the amount of resources the supplierpartners invest into the collaborative effort.

Managerial Implications

This research offers several broad managerial insights.First, we discuss the dilemma that suppliers encounter whenthey collaborate with partners that have overlapping andcompatible knowledge bases. Although compatibility sim-plifies and facilitates the exchange of information, ideas,and tacit know-how, resulting in greater knowledge transferand positive collaborative outcomes, it also creates a greaterrisk of knowledge misappropriation. Therefore, suppliersmust understand the implications of choosing a partner withsimilar know-how and capabilities. Suppliers should alsonote that when the collaboration involves exchangingmostly tacit know-how such as that embedded in capabili-ties (Ganesan, Malter, and Rindfleisch 2005), high knowl-edge base compatibility is an essential condition for effec-tive knowledge transfer to occur. Paradoxically, becauseknowledge embedded in capabilities holds strategic value,any unintended leakage could cause substantial damage tothe suppliers. Thus, suppliers face a serious dilemma whensharing skills and capabilities with their partners. The


potential threats are further aggravated when the supplierpartners are also competitors.

Second, suppliers should systematically evaluate thebenefits and costs of sharing knowledge with partners.Although knowledge sharing contributes to positive collabo-rative outcomes, it also entails substantial costs in establishingroutines and governance, especially for long-term partner-ships. Suppliers must develop knowledge-sharing routinesto facilitate knowledge flow across organization boundariesand set up task forces or assign specific personnel to over-see knowledge-sharing activities. Simultaneously, suppliersmust assess the likelihood and potential damage ofunintended knowledge spillover and choose an effectivegovernance mechanism. Suppliers should understand theefficacy and implementation costs of different types of formaland informal safeguards and select the governance mecha-nism that aligns with the goals of the collaborative effortand the collaborative history of the partners.

Third, our findings suggest that customer participationlikely tempers supplier partners' concern for partner oppor-tunism, thus ensuring the learning efficiency advantages ofcompatible knowledge bases between supplier partners.Nevertheless, suppliers should understand the circum-stances under which they can rely on customer participationas a feasible safeguard. First, a customer might not havesufficient motivation or legitimate power to participate inits suppliers' collaborative arrangement. Second, suppliersmight lack the bargaining power to prescribe the activitieswith which the customer should get involved. Therefore,suppliers face challenges in evaluating the efficacy of gover-nance processes the customer can introduce in the collabora-tive arrangement before its formation. Suppliers should takethese issues into consideration in planning for partnerships.

For the customer firm, this research underscores theimportance of motivating supplier partners to share valu-able knowledge in their collaborative efforts. Because rivalscannot imitate or easily acquire effective supplier partner-ships, customers should attempt to facilitate the develop-ment of collaborative partnerships among their suppliers.To motivate supplier partners to exchange valuable know-how, the customer (if it has legitimate power and expertise)should take part in such activities as mediating conflicts,providing technical consultation, and dedicating engineersto supplier sites. Importantly, the customer's participatoryactions should aim not only to improve collaborative effec-tiveness but also to nurture collective trust and cooperativenorms among the suppliers. In addition, customers mustunderstand suppliers' opportunity and motivation concernswith respect to knowledge sharing so that they can offerincentives and formulate policies to overcome suppliers'fears and resistance. For example, customers can establish asupplier association and use it as a platform for increasingopportunities for suppliers to access one another's technicalexpertise and problem-solving experience. Customers canalso share new product, manufacturing, and technologicalknowledge with suppliers through supplier developmentprograms. As the literature has suggested, such programs

not only enhance supplier capabilities but also promotesocial interactions and reciprocity norms among the cus-tomer and suppliers (Cheung, Myers, and Mentzer 2011).Over time, a cohesive supplier community can emerge, con-ferring competitive advantages on the customer (Dyer andNobeoka 2000).

This research also suggests that a customer should recog-nize that the value the supplier partners expect their collabo-ration to create and its own participation level are motivatingfactors that increase supplier partners' willingness to takepart in knowledge-sharing efforts. However, the customermust be aware that if supplier partners are uncertain aboutthe strategic value of the collaboration for the customer,they may consider some of the customer's participatoryactivities intrusive and overbearing, resulting in resistanceand negative consequences. Therefore, to benefit from thelearning-efficiency advantages of supplier partners' over-lapping knowledge bases, a customer should explicitly dis-close to its suppliers the long-term strategic value of thecollaborative partnership and convince them that its partici-patory actions could help prevent partner opportunism andmitigate exchange hazards.

Limitations

This research has several limitations. First, we examinedknowledge sharing between supplier partners in a collabo-rative partnership aimed at serving a downstream OEMcustomer. However, we did not evaluate how collaborativegoals and other factors affect suppliers' intention to form apartnership in the first place. Second, we assessed the inde-pendent and dependent variables of the conceptual modelfrom a single supplier's perspective. Because the data col-lected are not dyadic, the focal supplier's perceptions maynot be identical to those of the partner supplier. In addition,the items used to measure collaborative project perfor-mance and supplier knowledge sharing in Study 2 wererather broad. We did not capture specific types of knowl-edge (e.g., product or process knowledge) shared betweenthe supplier partners. Because we did not restrict the scopeand purpose (e.g., new product development, processrefinement) of the supplier collaborations we examined, wecould not develop measures that would reliably capture spe-cific types of shared knowledge and specific collaborativeperformance criteria.

Third, this study did not address the implications ofvarious knowledge characteristics, such as complexity,depth, and breadth, for supplier knowledge sharing. Thesecharacteristics may facilitate or impede the process ofknowledge sharing independently or interact with knowl-edge base compatibility. Fourth, this study examined theimpact of customer participation on suppliers' collaborativeefforts without addressing the customer's motivations forparticipation. In addition, although we argue that customerparticipation can mitigate supplier opportunism, our empiri-cal studies did not directly capture the incidences of suchmicroprocesses in the supplier collaboration.


APPENDIXStudy 2 Measures

Construct and Composite Items a

AverageVariance Composite FactorExtracted Reliability Loading

Collaborative Project Performance .94 .84 .871. The collaborative project has reached our pre-specified goals. .942. The collaborative project has achieved our target performance. .973. The collaborative project has met the procurement needs of the

customer. .82Knowledge Sharing .92 .61 .87

1. My firm and the partner have shared a significant amount ofknowledge with each other. .86

2. My firm and the partner have created new skills and knowledge byworking together. .87

3. My firm has exchanged many ideas with the partner about how toimprove each other's capabilities (in manufacturing, research anddevelopment, logistics, etc.). .82

4. In general, the skills and knowledge that have been shared betweenmy firm and the partner are:a. A limited amount/substantial amount. .83b. Basic/advanced. .72c. Of little value to both parties/of significant value to both parties. .64d. Of limited use/of significant use. .69

Anticipated Customer Value .70 .50 .66At the time when your firm's collaboration with the partner was ongoing,your firm expected that this collaboration would:

1. Create superior value for the customer. .742. Have a significant impact on the profitability of the customer. .733. Become a valuable resource to the customer. .63

Customer Participation .82 .61 .831. The customer's involvement is integral to the collaborative project. .652. The customer has been regularly informed of the progress of the

collaborative project. .703. The customer has provided technical assistance to my firm and the

partner. .794. The customer has helped to mediate and resolve conflicts arising in

the collaborative project. .875. The customer has been involved in coordinating the collaborative

activities between my firm and the partner. .88Knowledge Base Compatibility .73 .51 .77

1. My firm's employees understand the partner's skills and technologies. .752. My firm's skills and technologies are compatible with those of the partner .653. My firm's approach to learning new skills and technologies is similar

to those of the partner .814. My firm's capabilities (in manufacturing, research and development,

logistics, etc.) are similar to those of the partner. .635. The partner's employees understand my firm's skills and technologies. .73

Transaction-Specific Investments .89 .61 .661. My firm has committed a significant amount of resources and efforts

to the collaboration. .762. The partner has committed a significant amount of resources and

efforts to the collaboration. .81Competitive Intensity .82 .65 .77

1. My firm competes directly with the partner for the customer firm'sbusiness. .80

2. My firm's target markets are similar to those of the partner. .513. My firm considers the partner a major competitor in various product

markets. .88Reiationship Closeness Between Focal Supplier and Customer .83 .64 .76

1. My firm's relationships with this customer can be described asmutually gratifying. .84

2. My firm expects to do business with this customer in the long run. .813. My firm's staff have close, social interactions with this customer's staff. .75


APPENDIXContinued

Construct and Composite Items

AverageVariance Composite FactorExtracted Reliability Loading

Focal Supplier Dependence on Customer1. It would be difficult for my firm to replace this customer's business.2. My firm is dependent on this customer for sales.3. This customer accounts for a high percentage of my firm's sales.

Relative Power Between Supplier Partners^Which party benefits more from the collaborative effort?

.84 .72 .80.86.89.80

aMeasured on a seven-point Likert scale (1 = "my company," and 7 = "the partner company").Notes: All measures used seven-point Likert scales (1 = "strongly disagree," and 7 = "strongly agree") except for the "Knowledge Sharing" sec-

tion, which used seven-point Likert scales for the first three items and sematic differential scales for other items.

REFERENCESAiken, Leona S. and Stephen G. West (1991), Multiple Regres-

sion: Testing and Interpreting Interactions. Thousand Oaks,CA; Sage Publications.

Armstrong, Scott and Terry S. Overton (1977), "Estimating Non-response Bias in Mail Surveys," Journal of MarketingResearch, 14 (August), 396-402.

Básele, Guilhem (2008), "Controlling for Endogeneity withInstrumental Variables in Strategic Management Research,Strategic Organization, 6 (3), 285-327.

Cannon, Joseph P. and William D. Perreault Jr. (1999),"Buyer-Seller Relationships in Business Markets," Journal ofMarketing Research, 36 (November), 439-60.

Capron, Laurence and John Hulland (1999), "Redeployment ofBrands, Sales Forces, and General Marketing ManagementExpertise Following Horizontal Acquisitions: A Resource-Based View," Journal of Marketing, 63 (April), 41-54.

Cheung, Mee-Shew, Matthew B. Myers, and John T. Mentzer(2011), "The Value of Relational Learning in GlobalBuyer-Supplier Exchanges: A Dyadic Perspective and Test ofthe Pie-Sharing Premise," Strategic Management Journal, 32(10), 1061-82

Clark, Don (2010), "Chip Firms Detail Alliance to Combat Intel,"The Wall Street Journal, (February 12), (accessed July24, 2013), [available at http://online.wsj.com/article/SB10001424052748703525704575061560284134240.html].

Cohen, Wesley M. and Daniel A. Levinthai (1990), "AbsorptiveCapacity: A New Perspective on Learning and Innovation,"Administrative Science Quarterly, 35 (1), 128-52.

Das, T.K. and Bing-Sheng Teng (1998), "Developing Trust andControl: Developing Confidence in Partner Cooperation inAlliances," Academy of Management Review, 23 (3), 491-512.

Dhanaraj, Charles and Arvind Parkhe (2006), "Orchestrating Inno-vation Networks," Academy of Management Review, 31 (3),659-69.

Dyer, Jeffrey H. and Nile W. Hatch (2007), "Relation-SpecincCapabilities and Barriers to Knowledge Transfers: CreatingAdvantage Through Network Relationships," Strategic Man-agement Journal, 27 (8),701-719.

and Kentaro Nobeoka (2000), "Creating and Managing aHigh-Performance Knowledge-Sharing Network: The ToyotaCase," Strategic Management Journal, 21 (3), 345-67.

and Harbir Singh (1998), "The Relational View: Coopera-

Ganesan, Shankar (1994), "Determinants of Long-Term Orienta-tion in Buyer-Seller Relationships," Journal of Marketing, 58(April), 1-19.

, Steven P. Brown, Babu John Mariadoss, and Hitlbun(Dixon) Ho (2010), "Buffering and Amplifying Effects ofRelationship Commitment in Business-to-Business Relation-ships," Journal of Marketing Research," 47 (April), 361-73.

-, Alan Malter, and Aric Rindfleisch (2005), "Does Distance

tive Strategy and Sources of Interorganizational CompetitiveAdvantage," Academy of Management Review, 23 (4), 660-79.

Frazier, Gary L., Elliot Maltz, Kersi D. Antia, and Aric Rindfleisch(2009), "Distributor Strategic Information Sharing with Sup-pliers," Journal of Marketing, 73 (July), 31-43.

Still Matter? Geographic Proximity and New Product Develop-ment," Journal of Marketing, 69 (October), 44-60.

Grant, Robert M. (1996), "Prospering in Dynamically-CompetitiveEnvironments: Organizational Capability as Knowledge Inte-gration," Organization Science, 7 (4), 375-87.

Gulati, Ranjay and Harbir Singh (1998), "The Architecture ofCooperation: Managing Coordination Costs and AppropriationConcerns in Strategic Alliances," Administrative Science Quar-terly, 43 (4), 18\-S34.

Haas, Martine R. and Morten T. Hansen (2007), "DifferentKnowledge, Different Benefits: Toward a Productivity Per-spective on Knowledge Sharing in Organizations," StrategicManagement Journal, 28 (11), 1133-53.

Hamel, Gary (1991), "Competition for Competence and Interpart-ner Learning Within International Strategic Alliances," Strate-gic Management Journal, 12 (SI), 83-103.

Hansen, Morten T. (1999), "The Search-Transfer Problem: TheRole of Weak Ties in Sharing Knowledge Across OrganizationSubunits," Administrative Science Quarterly, 44 (1), 82-111.

, Marie Louise Mors, and Bj0m L0vâs, "Knowledge Shar-ing in Organizations: Multiple Networks, Multiple Phases,"Academy of Management Journal, 48 (5), 776-93.

Heide, Jan B., Kenneth H. Wathne, and Asksel I. Rokkan (2007),"Interfirm Monitoring, Social Contracts, and Relationship Out-comes," Journal of Marketing Research, 44 (August), 425-33.

Heiman, Bruce A. and Jack A. Nickerson (2004), "Empirical Evi-dence Regarding the Tension Between Knowledge Sharing andKnowledge Expropriation in Collaborations," Managerial andDecision Economics, 25 (6), 401-420.

Huber, George P. and Kyle Lewis (2010), "Cross-Understanding:Implications for Group Cognition and Performance," Academyof Management Review, 35 (1), 6-26.

Inkpen, Andrew C. (2000), "Learning Through Joint Ventures: AFramework of Knowledge Acquisitions," Journal of Manage-ment Studies, 37 (7), 1019-43.

(2005), "Learning Through Alliances: General Motors andNLJMMI," California Management Review, 47 (4), 114-36.

and Eric W.K. Tsang (2005), "Social Capital, Networks,and Knowledge Transfer," Academy of Management Review,30(1), 146-65.


Kale, Prashant, Jeffrey H. Dyer, and Harbir Singh (2002),"Alliance Capability, Stock Market Response, and Long-TermAlliance Success: The Role of the Alliance Function," Strate-gic Management Journal, 23 (8), 141-61.

, Harbir Singh, and Howard Perlmutter (2000), "Learningand Protection of Proprietary Assets in Strategic Alliances:Building Relational Capital," Strategic Management Journal,21 (3), 217-37.

Kang, Min-Ping, Joseph T. Mahoney, and Danchi Tan (2009),"Why Firms Make Unilateral Investments Specific to OtherFirms: The Case of OEM Suppliers," Strategic ManagementJournal, 30 (2), Ul-35.

Khanna, Tarun, Ranjay Gulati, and Nitin Nohria (1998), "TheDynamics of Learning Alliances: Competition, Cooperation,and Relative Scope," Strategic Management Journal, 19 (3),193-210.

Kotabe, Masaaki, Xavier Martin, and Hiroshi Domoto (2003),"Gaining from Vertical Partnerships: Knowledge Transfer,Relationship Duration, and Supplier Performance Improve-ment in the U.S. and Japanese Automotive Industries," Strate-gic Management Journal, 24 (4), 293-326.

Lane, Peter J., Balaji R. Koka, and Seemantini Pathak (2006),"The Reification of Absorptive Capacity: A Critical Reviewand Rejuvenation of the Construct," Academy of ManagementReview, 31 (4), 833-63.

and Michael Lubatkin (1998), "Relative Absorptive Capac-ity and Interorganizational Learning," Strategic ManagementJournal, 19 (5), 461-77.

-, Jane E. Salk, and Marjorie A. Lyles (2001), "AbsorptiveCapacity, Learning, and Performance in International JointVentures," Strategic Management Journal, 22 (12), 1139-61.

Lavie, Dovev (2006), "The Competitive Advantage of Intercon-nected Firms: An Extension of the Resource-Based View,"Academy of Management Review, 31 (3), 638-58.

Li, Dan, Lorraine Eden, Michael A. Hitt, and R. Duane Ireland(2008), "Friends, Acquaintances, or Strangers? Partner Selec-tion in R&D Alliances," Academy of Management Journal, 51(2), 315-34.

Liker, Jeffrey K. and Thomas Y. Choi (2004), "Building DeepSupplier Relationships," Harvard Business Review, 82(December), 104-113.

Madhok, Anoop and Stephen B. Tallman (1998), "Resources,Transactions and Rents: Managing Value Through InterfirmCollaborative Relationships," Organization Science, 9 (3),326-39.

Makhija, Mona V. and Usha Ganesh (1997), "The RelationshipBetween Control and Partner Learning in Learning-RelatedJoint Ventures," Organization Science, 8 (5), 508-527.

Mesquita, Luiz F., Jaideep Anand, and Thomas H. Brush (2008),"Comparing the Resource-Based and Relational Views:Knowledge Transfer and Spillover in Vertical Alliances,"Strategic Management Journal, 29 (9), 913-41.

Mowery, David C , Joanne E. Oxley, and Brian S. Silverman(1996), "Strategic Alliances and Interfirm Knowledge Trans-fer," Strategic Management Journal, 17 (Special Issue), 77-91.

Nahapiet, Janine and Sumantra Ghoshal (1998), "Social Capital,Intellectual Capital, and the Organizational Advantage," Acad-emy of Management Review, 23 (2), 242-66.

Noordhoff, Corine, Kyriakos Kyriakopoulos, Christine Moorman,Pieter Pauwels, and Benedict G.C. Dellaert (2011), "The BrightSide and Dark Side of Embedded Ties in Business-to-BusinessInnovation," Journal of Marketing, 75 (September), 34-52.

Ouchi, William G. (1979), "A Conceptual Framework for theDesign of Organizational Control Mechanisms," ManagementScience, 25 (9), S33-4S.

Oxley, Joanne E. and Rachelle C. Sampson (2004), "The Scopeand Governance of International R&D Alliances," StrategicManagement Journal, 25 (Special Issue), 723-49.

Palmatier, Robert W., Rajiv P. Dant, and Dhruv Grewal (2007), "AComparative Longitudinal Analysis of Theoretical Perspectivesof Interorganizational Relationship Performance," Journal ofMarketing, 71 (October), 172-94.

Parkhe, Arvind (1993), "Strategic Alliance Structuring: A GameTheoretic and Transaction Cost Examination of Interfirm Coop-eration," Academy of Management Journal, 36 (4), 794-829.

Phelps, Corey C. (2010), "A Longitudinal Study of the Influenceof Alliance Network Structure and Composition on FirmExploratory Innovation," Academy of Management Journal, 53(4), 890-913.

Quigley, Narda R., Paul E. Tesluk, Edwin A. Locke, and KathrynM. Bartol (2007), "A Multilevel Investigation of the Motiva-tional Mechanisms Underlying Knowledge Sharing and Perfor-mance," Organization Science, 18 (1), 71-88.

Rindfleisch, Arie and Jan B. Heide (1997), "Transaction CostAnalysis: Past, Present, and Future Applications," Journal ofMarketing, 61 (October), 30-54.

and Christine Moorman (2001), "The Acquisition and Uti-lization of Information in New Product Alliances: A Strength-of-Ties Perspective," Journal of Marketing, 65 (April), 1-18.

and (2003), "Interfirm Cooperation and CustomerOrientation," Journal of Marketing Research, 40 (November),421-36.

Ross, William T., Jr., and Diana C. Robertson (2007), "CompoundRelationships Between Firms," Journal of Marketing, 71(July), 108-123.

Rowley, Tim, Dean Behrens, and David Krackhardt (2000),"Redundant Governance Structures: An Analysis of Structuraland Relational Embeddedness in the Steel and SemiconductorIndustries," Strategic Management Journal, 21 (3), 369-86.

Sampson, Rachelle C. (2004), "Organizational Choice in R&DAlliances: Knowledge-Based and Transaction Cost Perspec-tives," Managerial and Decision Economics, 25 (6/7), 421-36.

Srivastava, Manish K. and Devi R. Gnyawali (2011), "When DoRelational Resources Matter? Leveraging Portfolio Technologi-cal Resources for Breakthrough Innovation," Academy of Man-agement Journal, 54 (4), 797-810.

Szulanski, Gabriel (1996), "Exploring Internal Stickiness: Impedi-ments to the Transfer of Best Practice Within the Firm," Strate-gic Management Journal, 17 (1), 2 7 ^ 3 .

Van Eerde, Wendelien and Henry Thierry (1996), "Vroom'sExpectancy Models and Work-Related Criteria: A Meta-Analysis," Journal of Applied Psychology, 81 (5), 575-86.

Van Wijk, Raymond, Justin J.P. Jansen, and Marjorie A. Lyles(2008), "Inter- and Intra-Organizational Knowledge Transfer: AMeta-Analytic Review and Assessment of Its Antecedents andConsequences," Journal of Management Studies, 45 (4), 830-53.

Wathne, Kenneth H., Harald Biong, and Jan B. Heide (2001),"Choice of Supplier in Embedded Markets: Relationship andMarketing Program Effects," Journal of Marketing, 65 (April),54-66.

Wooldridge, Jeffrey M. (2010), Econometric Analysis of CrossSection and Panel Data, 2d ed. Cambridge, MA: MIT Press.

Wu, Zhaohui, Thomas Y. Choi, and M. Johnny Rungtusanatham(2010), "Supplier-Supplier Relationships in Buyer-Supplier-Supplier Triads: Implications for Supplier Performance," Jour-nal of Operations Management, 28 (2), 115-23.

Wuyts, Stefan, Stefan Stremersch, Christophe Van den Bulte, andPhilip Hans Franses (2004), "Vertical Marketing Systems forComplex Products: A Triadic Perspective," Journal of Market-ing Research, 41 (November), 479-87.

Zacharia, Zach G., Nancy W. Nix, and Robert F. Lusch (2011),"Capabilities that Enhance Outcomes of an Episodic SupplyChain Collaboration," Journal of Operations Management, 29(6), 591-603.

Zahra, Shaker A. and Gerard George (2002), "Absorptive Capac-ity: A Review, Reconceptualization, and Extension," Academyof Management Review, 21 (2), 185-203.


Copyright of Journal of Marketing is the property of American Marketing Association and itscontent may not be copied or emailed to multiple sites or posted to a listserv without thecopyright holder's express written permission. However, users may print, download, or emailarticles for individual use.

hillbun (dixon) ho & shankar ganesan does knowledge base … · 2020-03-07 · ganesan is...

Documents