Inkpen

This article was downloaded by: [157.182.150.22] On: 17 April 2015, At: 03:49 Publisher: Institute for Operations Research and the Management Sciences (INFORMS) INFORMS is located in Maryland, USA

Organization Science Publication details, including instructions for authors and subscription information: http://pubsonline.informs.org

Knowledge Management Processes and International Joint Ventures Andrew C. Inkpen, Adva Dinur,

To cite this article: Andrew C. Inkpen, Adva Dinur, (1998) Knowledge Management Processes and International Joint Ventures. Organization Science 9(4):454-468. http://dx.doi.org/10.1287/orsc.9.4.454 Full terms and conditions of use: http://pubsonline.informs.org/page/terms-and-conditions This article may be used only for the purposes of research, teaching, and/or private study. Commercial use or systematic downloading (by robots or other automatic processes) is prohibited without explicit Publisher approval, unless otherwise noted. For more information, contact [email protected]. The Publisher does not warrant or guarantee the article’s accuracy, completeness, merchantability, fitness for a particular purpose, or non-infringement. Descriptions of, or references to, products or publications, or inclusion of an advertisement in this article, neither constitutes nor implies a guarantee, endorsement, or support of claims made of that product, publication, or service. © 1998 INFORMS Please scroll down for article—it is on subsequent pages

INFORMS is the largest professional society in the world for professionals in the fields of operations research, management science, and analytics. For more information on INFORMS, its publications, membership, or meetings visit http://www.informs.org

Knowledge Management Processes and International Joint Ventures Downloaded from informs.org by [157.182.150.22] on 17 April 2015, at 03:49 . For personal use only, all rights reserved.

Andrew C. Inkpen • Adva Dinur Thunderbird, The American Graduate School of International Management, Glendale, Arizona 85306 School of Business and Management, Temple University, Philadelphia, Pennsylvania 19122

K

nowledge creation is now recognized as a key, competitive advantage. In the global economy, international joint ventures are increasingly important as the way of creating knowledge interorganizationally. Yet, research into interorganizational knowledge creation has been underdeveloped. This paper offers insights about how international joint ventures can be integrated into a firm’s dynamic system of knowledge creation. Ikujiro Nonaka

Abstract The management and processing of organizational knowledge are increasingly being viewed as critical to organizational success. By exploring how firms access and exploit alliance-based knowledge, the authors provide evidence to support the argument that the firm is a dynamic system of processes involving different types of knowledge. Using data from a longitudinal study of North American-based joint ventures (JVs) between North American and Japanese firms, they address three related research questions: (1) what processes do JV partners use to gain access to alliance knowledge; (2) what types of knowledge are associated with the different processes and how should that knowledge be classified; and (3) what is the relationship between organizational levels, knowledge types, and the transfer of knowledge? Although many generalizations have been drawn about the merits of knowledge-based resources and the creation of knowledge, few efforts have been made to establish systematically how firms acquire and manage new knowledge. Moreover, prior alliance research has not addressed in detail the nature of alliance knowledge and how knowledge is managed in the alliance context. The authors examine the processes used by alliance partners to transfer knowledge from an alliance context to a partner context. They identify four key processes—technology sharing, alliance-parent interaction, personnel transfers, and strategic integration—that share a conceptual underpinning and represent a knowledge connection between parent and alliance. Each of the four processes is shown to provide an avenue for managers to gain exposure to knowledge and ideas outside their traditional organizational boundaries and to create a connection for individual managers to communicate their alliance experiences to others.

Although all of the knowledge management processes are potentially effective, the different processes involve different types of knowledge and different organizational levels. The primary types of knowledge associated with each process are identified and then linked with the organizational level affected by the transfer process. From those linkages, several propositions about organizational knowledge transfer and management are developed. The results suggest that although a variety of knowledge management strategies can be viable, some strategies lead to more effective knowledge transfer than others.

(Organizational Knowledge, Learning; Joint Ventures and Alliances, Tacit Knowledge; Knowledge Management Processes)

Increasingly, the creation of new organizational knowledge is becoming a managerial priority. New knowledge provides the basis for organizational renewal and sustainable competitive advantage (Prahalad and Hamel 1994, Quinn 1992). We draw on recent work on knowledge management (e.g., Grant 1996, Hedlund 1994, Nonaka and Takeuchi 1995, Spender 1996b), to explore how firms access and exploit alliance-based knowledge. Using data from a longitudinal study of North American-based joint venture (JVs) between North American and Japanese firms, we address three related research questions: (1) What processes do JV partners use to gain access to alliance knowledge, (2) what types of knowledge are associated with the different processes and how should that 1047-7039/98/0904/0454/$05.00

454

ORGANIZATION SCIENCE /Vol. 9, No. 4, July–August 1998

Copyright 䉷 1998, Institute for Operations Research and the Management Sciences

Downloaded from informs.org by [157.182.150.22] on 17 April 2015, at 03:49 . For personal use only, all rights reserved.

ANDREW C. INKPEN AND ADVA DINUR Knowledge Management Processes and International Joint Ventures

knowledge be classified, and (3) what is the relationship between organizational levels, knowledge types, and the transfer of knowledge? The first question provides a basis for understanding how firms access and transfer knowledge across organizational boundaries. Given that the firm can be viewed as a system processing different types of knowledge (Spender 1996b), the second and third questions are directed at exploring the organizational and strategic implications of different knowledge types. Our focus in examining the research questions is knowledge management by the North American JV parents. Over the past two decades, the formation of international strategic alliances has increased substantially. The number of domestic and international alliances has grown by more than 25 percent annually since 1990 (Bleeke and Ernst 1995). Drucker (1995) suggested that the greatest change in the way business is being conducted is the accelerating growth of relationships based not on ownership but on partnership. Researchers seeking to explain the alliance trend have argued that alliances provide a platform for organizational learning, giving partner firms access to each other’s knowledge (Grant 1996, Hamel 1991, Kogut 1988, Westney 1988). Kogut (1988, p. 323) suggested that learning can be an alliance motive under two conditions: one or all partner firms want to acquire the other’s organizational knowledge, or one firm wishes to maintain an organizational capability while benefiting from a partner’s cost advantage or knowledge. The knowledge is often organizationally embedded and causally ambiguous. If the alliance replicates partner experiential knowledge in a jointly owned organization, one or all partners may have access to knowledge that would not have been available in the absence of collaboration. In the alliance context, knowledge useful to a parent firm can be viewed from three perspectives. First, firms may acquire knowledge useful in the design and management of other alliances (see Lyles 1988). Such knowledge may be applied to future alliances. Second, firms may seek access to other firms’ knowledge and skills, but without necessarily wishing to internalize the knowledge in their own operations. As Hamel (1991) pointed out, knowledge embodied only in the specific outputs of the alliance has no value outside the narrow terms of the collaborative agreement. Third, an alliance may generate knowledge that can be used by parent companies to enhance their own strategy and operations. Our current research pertains to that type of knowledge. Knowledge useful to a parent may be knowledge transferred by an alliance partner to the alliance. The knowledge may be created independently by the alliance through its interactions with customers, competitors, and other firms. The


alliance may also create a forum for interactions between the parents that is itself a source of new knowledge. Two examples help illustrate how firms learn through alliances. Much has been written about how General Motors struggled to learn from its New United Motor Manufacturing, Inc. (NUMMI) joint venture with Toyota (Badaracco 1991, Keller 1989, Mahoney and Deckop 1993). In recent years, knowledge has been transferred successfully from NUMMI to other General Motors divisions and plants. In particular, General Motors has used its NUMMI experience as a catalyst for several successful international greenfield plants (Miller 1993).1 The knowledge transferred has been primarily in the areas of manufacturing process and human resource management. As a second example, Sony, a firm with a culture of independence in product development, has formed various alliances with computer and telecommunications firms in an effort to forge new technology linkages for its consumer electronics products (Hamilton 1995). The alliances give Sony access to a wealth of new knowledge, such as how to manage product development cycles that are much faster in the computer industry than in consumer electronics. In forming the alliances, Sony has enabled personnel at various organizational levels to gain access to new knowledge. The challenge for Sony and other firms involved in alliances, and for all firms seeking access to knowledge beyond their boundaries, is to incorporate disparate pieces of individual knowledge into a wider organizational knowledge base. A still rather small but growing body of research (Dodgson 1993, Doz 1996, Hamel 1991, Inkpen and Beamish 1997, Inkpen and Crossan 1995, Kogut 1988, Makhija and Ganesh 1997, Mowery et al. 1996, Parkhe 1991, Pucik 1991, Simonin and Helleloid 1993, Westney 1988) is addressing the issue of alliances and learning. Researchers have begun to explore some of the important questions associated with how organizations exploit alliance learning opportunities but have not examined in detail the nature of alliance knowledge and how knowledge is managed in the alliance context. In organizational studies in general, knowledge management has received limited attention, perhaps because dominant theoretical paradigms are inappropriate for addressing the issue (Hedlund 1994). Therefore, we explore alliance knowledge with the objective of developing a framework that integrates knowledge states and knowledge management processes with the organizational levels involved in the transfer processes. We focus on alliance forms that combine resources from more than one organization to create a new organizational entity (the “child”) distinct from its parents. In

455



most of the cases, such an alliance is an equity joint venture. Examining that type of alliance allows for a clear delineation of the partner relationship and the nature of alliance knowledge. Siecor, an alliance between Siemens and Coming, is an example. The partners in that alliance brought together their complementary capabilities in telecommunications and glass technology to build an independent organization with its own headquarters, CEO, board of directors, and staff.

Conceptual Background Types of Organizational Knowledge In developing an understanding of organizational knowledge, we begin with the distinction between tacit and explicit knowledge. Tacit knowledge was defined by Polanyi (1962) as knowledge that is nonverbalizable, intuitive, and unarticulated. Spender (1996a) suggested that tacit knowledge could be understood best as knowledge that has not yet been abstracted from practice. It is knowledge that has been transformed into habit and made traditional in the sense that it becomes “the way things are done around here” (Spender 1996b) Tacit knowledge is highly context specific and has a personal quality, which makes it difficult to formalize and communicate (Nonaka 1994). Explicit knowledge is knowledge that is transmittable in formal, systematic language and may include explicit facts, axiomatic propositions, and symbols (Kogut and Zander 1992). It can be codified or articulated in manuals, computer programs, training tools, and so on. The distinction between explicit and tacit should not be viewed as a dichotomy but rather as a spectrum with the two knowledge types at either end. Winter (1987) identified other taxonomic dimensions of knowledge, including complex versus simple, not teachable versus teachable, and not observable in use versus observable in use. Similarly, we argue that although the distinction between tacit and explicit is important, it does not allow us to consider any gray areas between completely tacit and completely explicit knowledge. Knowledge types, therefore, must be classified on a continuum that ranges from explicit knowledge embodied in specific products and processes to tacit knowledge acquired through experience and use and embodied in individual cognition and organization routines. Nonaka and Takeuchi (1995) argued that a key challenge for organizations is the conversion of tacit knowledge to explicit knowledge. Knowledge that is tacit and highly personal has little value until it can be converted into explicit knowledge that other organizational members can share. However, such a conversion process exposes the knowledge to the hazard of imitation by other

456

firms. Zander and Kogut (1995) discussed the tradeoff between the need to share and transfer knowledge internally and the risk of exposing the knowledge to imitation. To address the knowledge conversion dilemma, there is a need for a better, broader taxonomy of both tacit and explicit knowledge. Recognizing that firms’ idiosyncratic knowledge consists mostly of tacit, difficult to imitate knowledge, Spender (1996b) developed a more comprehensive typology of organizational knowledge encompassing individual and social levels. Whereas individuals have knowledge that is practical, communities have knowledge that constitutes the socialization and social activities of the individuals within them (Spender 1996b). Individuals constantly acquire knowledge, share it with their organizational community, and thus increase the collective store of knowledge, while maintaining a common individual knowledge with their coworkers. In Spender’s (1996b) typology, explicit knowledge stored in databanks, standard operating procedures, manuals, and so on is referred to as objectified knowledge. Tacit knowledge is separated into three subtypes: conscious, automatic, and collective. Individual tacit knowledge can be either conscious or automatic.2 Automatic knowledge is implicit knowledge that “happens by itself” and is often taken for granted. Conscious knowledge may be codified, perhaps as a set of notes, and is potentially available to other people. Collective knowledge is tacit knowledge of a social or communal nature. Organizational Levels and Knowledge Movement Clearly, organizations are repositories of knowledge. The important question is how individual and group interactions contribute to organizational knowledge creation. Organizations cannot create knowledge without individuals, but unless individual knowledge is shared with other individuals and groups, the knowledge will have a limited impact on organizational effectiveness. Hence, organizational knowledge creation should be viewed as a process whereby the knowledge held by individuals is amplified and internalized as part of an organization’s knowledge base (Nonaka 1994). As knowledge is transformed from an individual to a collective state, organizational knowledge is created (Nonaka and Takeuchi 1995). The transformation occurs in a dynamic process involving various organizational levels and carriers of knowledge. Specific learning processes are at work at each level. At the individual level, the critical process is interpreting and sense making; at the group level it is integrating; and at the organization level it is integrating and institutionalizing (Inkpen and Crossan 1995). To capture the dynamic movement of knowledge across the levels, Nonaka (1994) developed the concept of a spiral of




knowledge creation. In the spiral, knowledge moves upward in an organization, starting at the individual level, moving to the group level, and then up to the firm level. As the knowledge spirals upward in the organization, it may be enriched and amplified as individuals interact with each other and with their organizations. Hedlund and Nonaka (1993), in their analysis of Japanese and Western knowledge management, proposed a model linking organizational levels and types of knowledge. Their objective was to develop an understanding of how different knowledge types travel and change between individuals and organizations. On the basis of the Hedlund and Nonaka (1993) model, we propose Figure 1 as a framework for an empirical examination of knowledge management processes. In the framework, the organization is seen as a repository of various knowledge types in different organizational locations. The vertical dimension refers to knowledge tacitness and the horizontal dimension distinguishes between the organizational levels where knowledge resides. In Figure 1, knowledge tacitness is a continuum in which explicit knowledge has very low tacitness. The figure implies that as knowledge becomes more tacit, it becomes less teachable, less codifiable, and hence, less transferable (Kogut and Zander 1992). The key assumption underlying this framework is that organizations have a range of types of knowledge and carriers of knowledge. Where organizations differ is in their view of the importance of different types of knowledge and their ability to transform and move knowledge across organizational levels. For example, Hedlund and Nonaka (1993) argued that Western firms lose much of their potential for knowledge creation by overemphasizing explicit knowledge and the development of complex managerial hierarchies, systems, and standardization. Using JVs as the empirical context, we examine the linkages between knowledge transfer processes, knowledge types, and organizational levels. The next section is an overview of JVs and learning. Figure 1

Knowledge Transfer Classification Framework


Knowledge Management and Learning Through JVs In arguing that strategy making is a learning process, Mintzberg (1990) suggested that strategic initiatives create experiences, actions, and strategic choices that provide the foundation for learning. The focus of our study is a particular strategic initiative—the formation of a JV. The JV experience can be the action that triggers learning because it provides new stimuli that may force changes in the mental maps of the organization (Nonaka and Johansson 1985). An underlying assumption is that managers have some understanding of the causal relationships associated with knowledge, action, and outcomes. Following Inkpen and Crossan (1995) and Nonaka (1994), we view knowledge creation through JVs as a multi-stage process, analogous to the innovation diffusion process (e.g., Tushman and Scanlon 1981). The first stage begins with the formation of the JV and interactions between individuals from the two (or more) partners. The second stage and our primary focus is the transfer of knowledge from the JV to the partners. Huber (1991) referred to that process as “grafting,” whereby organizations increase their store of knowledge by internalizing knowledge not previously available within the organization. For internalization to occur, the parents must first engage in efforts to transfer partner skill-related knowledge from the JV to themselves. Those efforts create the “connections” through which individuals can share their observations and experiences (Von Krogh et al. 1994). The intensity of a parent firm’s learning efforts reflects the degree to which the parent is actively trying to internalize the skills and capabilities of its partner. Knowledge connections are formed through both formal and informal relationships between individuals and groups (Inkpen 1996). Those internal managerial relationships facilitate the sharing and communicating of new knowledge and provide a basis for transforming individual knowledge to organizational knowledge. When one individual’s or group’s knowledge connects with other knowledge, it can be discussed, debated, and possibly discarded. The knowledge may be further developed and move upward in the organization. Individual knowledge is inherently “fragile” and therefore, without knowledge connections, new knowledge may be ignored or viewed as irrelevant (Von Krogh et al. 1994). Grant (1996) argued that organization structures can be designed to maximize the efficiency of knowledge integration. In the literature on innovation, specialized personnel such as “technological gatekeepers” (Katz and Tushman 1980) and specialized organizational structures such as transfer groups (Katz and Allen 1988) have been shown to have a significant effect on the transfer of information between organizations. When a JV partner has a strategic objective

457



of acquisition and proprietary control over alliance knowledge, knowledge connections are the mechanisms for knowledge acquisition. Although the transfer of alliance knowledge is a necessary condition for knowledge creation, the parent must ensure that the transferred knowledge is moved and shared within the parent organization. The risk, particularly with tacit knowledge, is that knowledge transferred from a JV to a parent will dissipate as it spirals up to the organization level. The rate of dissipation will be influenced by a variety of factors. For example, when confronted with learning opportunities, successful firms may see little need to change behavior and thus, may become trapped by their distinctive competence (Levinthal and March 1993). The strength of a firm’s learning intent will help determine the organizational resources committed to learning (Hamel 1991). Alliance control mechanisms may influence the transfer of knowledge (Makhija and Ganesh 1997). The type of knowledge creation mechanisms plays a key role in how new knowledge is “managed” by alliance parent firms (Hedlund and Nonaka 1993). Finally, managerial belief systems permeate all levels of knowledge creation and correspondingly, contribute to knowledge dissipation (Inkpen and Crossan 1995). In summary, our study examines the processes used by firms to gain access to and transfer different types of alliance-based knowledge. Although much of the learning literature addresses the product or content of learning, the process of learning and the types of knowledge are also very important. A focus solely on content ignores the complex cognitive and behavioral changes that must occur before a learning “outcome” can be identified. Given that the question of whether or not organizations learn is controversial, studying knowledge creation may provide a more valid foundation for understanding how knowledge travels and changes within organizations (Hedlund and Nonaka 1993).

Research Methods and Data We designed a two-stage study. The first stage established the industry context and basis for the selection of cases for longitudinal study. In the second stage we used an open-ended approach of grounded theory building (Glaser and Strauss 1967) to examine types of knowledge and processes of alliance-based knowledge management. The industry, alliance, and partner contextual data from stage I were critical in interpreting the case study data from stage 2. Stage 1: Context Definition The initial research stage was designed to provide contextual understanding of the alliance learning issues and

458

to gain a cross-sectional perspective on the basic dimensions of alliance learning. A sample of North American– Japanese JVs located in North America provided the empirical base.3 The primary data collection method for stage 1 was field interviews with 58 managers associated with 40 two-partner JVs. Most managers held positions such as JV president or JV general manager and were either employed by the American partners or appointed by the American partners to senior management positions in the JVs. Geringer and Hebert (1991) found that such managers are a valid source of JV data. Being at the boundary between the JV and parent firms, those managers were expected to have an important influence on parent access to and management of alliance knowledge. (For more detail on the stage one methodology, see Inkpen, 1995a and 1995b.) All JVs were suppliers to the automotive industry and only one had less than 50 percent of its sales to automotive customers. The primary JV motive for the majority of the American partners was access to the Japanese transplant market in the United States. Most of the JVs were direct suppliers to the automotive assemblers (i.e., tier-one suppliers). With two exceptions, all JVs were startup or greenfield organizations. In terms of ownership, 17 ventures were 50-50, in 15 ventures the Japanese partners had majority equity, and in 8 ventures the American partners had majority equity. Stage 2: Longitudinal Case Study Stage 1 of the research yielded evidence that the JVs created important learning opportunities for the American JV parents. The American firms were provided an excellent “window” into their Japanese partners’ capabilities. The window had two main sources of potential value. First, all but five JVs were transplant suppliers, and generally the products supplied to the transplants were similar to products manufactured by the Japanese partners in Japan. The JVs gave the American partners access to skills created by the Japanese partner for the Japanese market and also access to how those skills could be adapted to the North American work style and infrastructure. Second, the JVs were often the American partners’ initial experience in supplying Japanese automakers. In most cases the Japanese partners had established relationships with the Japanese automakers. Therefore, the JVs afforded the American partners an opportunity to learn how to manage a long-term Japanese customer relationship, albeit adapted to the North American context. Stage 1 also provided the foundation for an emerging understanding of alliance knowledge management. The second research stage explored the knowledge management process in detail. An emphasis on process suggested




the need for a longitudinal approach that would provide deep and extensive access to the individuals involved in collaborative exchange. Therefore, a multiple case study design was used. It was based on theoretic replication (Yin 1989) because the choice of cases was directed by the emerging theory developed from stage 1. For efficiencies in data collection and to capitalize on established industry contacts, we selected a subset of five of the cases from stage 1. We used several criteria to select the cases, with the objective of finding variance across several dimensions. Table 1 reports case characteristics. Of particular interest was the learning potential created by the JVs. It was important because it influenced the knowledge

Table 1

management processes initiated by the JV parents and the motivation of the American parents to exploit the learning potential. To evaluate learning potential, we considered such factors as the nature of partner contributions to the JV, the functional similarity of products produced by the JV and the American parent, manufacturing quality differences between the JV and the American parent, geographic proximity of the JV plant to the American parent, and the willingness of the Japanese partner to share its technology. For example, the Alpha JV was classified as creating high learning potential because (1) the Japanese partner contributed key manufacturing process technology and the customer contact, (2) the JV and American

Case Characteristics

Characteristic

Alpha

Beta

Gamma

Kappa

Sigma

American partner learning potential

High

Medium

Medium

High

Low

JV performance

Low

Medium

High

Medium

High

Partner history

None prior to technology assistance agreement signed in 1988

Limited; 10-year technology relationship

Limited; licensing agreement from 1980

Extensive; 25-year relationship between partner presidents

None

Senior JV management

President from outside; Japanese president; Japanese VP American COO and plant managers

Japanese chairman; American American president president/GM; and COO; Japanese VP Japanese sales manager

American president and plant managers; Japanese sales managers

Equity shares

50/50

50/50

50/50

Key partner contributions

Japanese: Japanese: Japanese: Japanese: Japanese: sales manufacturing manufacturing manufacturing manufacturing support. American: process and process and process and process and manufacturing engineering, engineering, product engineering, engineering process, plant customer contacts, design, customer product design, customer contacts. startup, plant product design, contacts. American: customer contacts. American: raw management, access to raw plant startup, plant American: plant materials, JV administrative materials. American: management, startup, management and support. plant startup, access administration administration plant management, to American partner support support HR, legal, startup support manufacturing finance), plant equipment, plant management management, administration support

Customers

Primarily Japanese automakers

65% Japanese, 35% American

Japanese and JV automakers


Primarily Japanese automakers

50/50

Japanese and U.S. automakers

Mix of Japanese and American Customers

459



parent produced similar products, (3) the Japanese parent was open in its willingness to share technology, (4) the JV’s quality record was superior to that of the American parent, and (5) the JV plant was located in an unused American facility a short distance from the American parent headquarters. JV performance was evaluated from the perspective of the American parent and was based on the American parent’s overall satisfaction with performance. Partner history reflects the extent of previous collaborative relationships between the partners. Building on the data collected in the first stage, we began second stage site visits and interviews in May 1993 and continued to September 1994. The interviews in stage 2 were usually 90 minutes to two hours long, although a few were a half day or more. For the cases studied in this stage of the research, a total of 20 interviews were conducted with senior American managers in the JVs and parent organizations. Including the first stage interviews, observations were collected over a period of three and one-half years.

Findings Knowledge Management Processes For each of the cases, we collected data on the knowledge management processes used by the American parents. From the data, four critical processes were identified: technology sharing, JV-parent interaction, personnel transfers, and strategic integration. The four processes share a conceptual underpinning in that each represents a knowledge connection, which creates the potential for individuals to share their observations and experiences. Each of the four processes provided an avenue for managers to gain exposure to knowledge and ideas outside their traditional organizational boundaries and created a connection for individual managers to communicate their JV experiences to others. In that sense, the four processes represent the locus of knowledge creation because it is through those processes that different types of knowledge converge and become accessible. On the basis of the interview data, the cases were evaluated individually on the intensity of the processes for transferring knowledge (Table 2). The high, medium, and low classifications are a function of comparison across the cases. “Low” means that we found no evidence of a knowledge management process. “Medium” means that the process was occurring but at a lower intensity than it was in at least one of the other “high” cases. For example, we observed that all of the knowledge management processes were present for Kappa and were occurring at a high intensity level. Alpha was classified as low for three processes and medium for one. To provide examples of

460

actual knowledge management, Table 3 summarizes some of the knowledge management activities associated with the Kappa JV. Similar data were collected for each of the cases. In the following sections, each of the processes is described in detail and additional examples are provided. We then link the processes with knowledge types and the parent’s organizational levels affected by the processes. Technology Sharing. American parent firms instituted various technology sharing processes to gain access to technology resident in the JV and in the Japanese partner. The most evident knowledge transfer approach was through structured meetings between JV and parent managers. In Gamma, monthly meetings were held, with the location alternating between the JV and one of the American parent plants. In attendance at the meetings were plant managers, heads of quality control, R&D managers, the VP manufacturing at the American parent head office, and several senior JV managers. In addition, quarterly R&D meetings were held involving the JV and American parent. The manufacturing vice president of one of the American parents said that “while [he] hated to admit it, the quality of the JV product was superior to that in the parent.” As a result, a program was initiated with plant managers to address the need to improve quality and customer service. Access to partner technology skills also was available through direct linkages between Japanese and American partners. In both Kappa and Sigma, American parent personnel regularly visited Japanese parent facilities. To capitalize on the Japanese partner’s fabrication knowledge and ability to operate with fewer equipment operators, Kappa managers invited several Japanese engineers to the United States to train parent engineers. The Japanese engineers brought very detailed equipment designs that would enable the American firm to replicate their manufacturing process. When no visible progress was made in designing new equipment, the American president decided to contract equipment design and manufacturing to the Japanese partner. An American engineer would be sent to Japan to learn about the equipment so it could be installed in the United States. In another case, the partners signed a very broad global technology agreement. Both partners agreed to be completely open in sharing both product and manufacturing technology. For product technology, explicit terms on licensing and royalties were established. For manufacturing technology there were no terms. For example, the American parent might ask to borrow a Japanese partner engineer for a few weeks. When that had happened in the past there had been no financial considerations because “it all comes out in the wash.” The American partner




Table 2

Knowledge Management in the Five Cases

Technology sharing Interorganizational interaction Personnel transfers Strategic integration

Alpha

Beta

Gamma

Kappa

Low Medium Low Low

Low Medium Medium Low

High High Medium High

High High High High

recognized the need for reciprocal commitment (cf. Gulati et al. 1994) and tried to make the technology sharing a two-way relationship. JV-parent Interactions. Individual knowledge and perspectives remain personal unless they are amplified and articulated through social interaction (Nonaka 1994). Interactions between parent and JV managers beyond specific technology initiatives can create the social context necessary to bring JV knowledge into a wider arena. In this study, the interactions were primarily social and involved a variety of groups. In effect, the JV-parent interactions provided the foundation for evolving communities of practice (Brown and Duguid 1991). Community members share knowledge and may be willing to challenge the organization’s conventional wisdom. Visits and tours of JV facilities were an effective and simple interactive means for parent managers to learn about their JVs. The JV managers were generally convinced that the differences embodied in the JV were visible and parent managers would appreciate the differences if they spent time in the JV. However, visits were not always perceived as effectively utilized, as various managers indicated. A utilization of a JV visit that produced an observable change occurred when an American parent sent several managers to visit its JV to study the JV’s human resource management systems. In contrast to most of the American parent plants, the JV was a nonunion operation with a hybrid mix of Japanese and American human resource practices. The American parent was establishing a new nonunion operation and decided to use the JV as a model. With the JV managers’ support, the visiting managers spent several days studying the JV and then incorporated much of their knowledge in the new nonunion plant. One explanation for the success of the knowledge transfer is that learning was focused on a discrete system that could easily be replicated. Customer-supplier relationships between the JV and the American parent also created a basis for extensive, although not always amicable, interaction. In the Alpha case, the JV acted as both supplier and customer for the American parent. Neither relationship was considered


Table 3

Sigma Medium High Medium Medium

Examples of Knowledge Management Processes in Kappa

1. The American parent (AP) studied various aspects of the JV’s operation, including its use of employee involvement programs, kaizen teams, and scheduling system. The AP has also studied some of the JV’s process innovations, one of which the JV considers proprietary. 2. Several JV managers were promoted to positions within the AP. One manager was promoted into an AP staff training position at AP HQ. Several engineers were also promoted. 3. AP senior managers were committed to the JV and to an Asian connection. The JV is the strongest Asian connection. The president of the AP had a very close relationship with the former Japanese parent chairman. 4. The AP set up what it called “gatekeepers,” units of the company responsible for certain aspects of manufacturing. The gatekeeper was expected to be available to all units of the company on the specific process or technology. The JV was asked to be the gatekeeper for JIT. 5. The AP had several engineers temporarily working in Japan in the Japanese parent organization. 6. The AP and the Japanese parent initiated a joint engineering project. A piece of manufacturing equipment was to be made by the Japanese partner in Japan, with an American engineer visiting Japan during the project period. 7. More than 15 employees in the JV visited Japan.

satisfactory, although the JV was a rich source of knowledge for the American parent. In another example, the American parent substantially increased its quality because of pressure from the JV customer, which in turn was under pressure from its Japanese transplant supplier. Until the JV was formed, the American parent had not had any extensive interactions with Japanese customers. In supplying the JV, and indirectly becoming a transplant supplier, the American parent was forced to evaluate some of its manufacturing operations. Meetings, such as monthly sales meetings between parent divisions and the JV, were also a means of interacting

461



and exchanging knowledge. Finally, in several cases the JV relied extensively on the American parent for various services, such as purchasing, accounting, human resource management, and in one case laboratory facilities. Consequently, the American parent managers had no choice but to be involved with the JV. Personnel Transfers. Personnel transfers can be considered a process of organizational reflection (Hedlund and Nonaka 1993) and a means of mobilizing personal knowledge. Transfers and rotation of personnel help members of an organization to understand the business from a multiplicity of perspectives, which in turn makes knowledge more fluid and easier to put into practice (Nonaka 1994, p. 29). The rotation of managers through JV positions and back to the parent may encourage the “bleedthrough” of ideas from the venture to the parent (Harrigan 1985). None of the cases studied had a structured process of rotation between the JV and the parent. However, Kappa had an extensive informal system of personnel transfers between the organizations. For example, the American parent promoted a Kappa manager to a staff training position at parent headquarters. Several engineers also were promoted. In four of the cases, senior managers in the JV had been transferred to the JV when it was formed. The careers of the managers were considered closely linked to the American parent and not just the JV. In Gamma, the chief operating officer of the JV came from the American parent to act as mentor for the younger JV managers and will eventually return to the American parent. In Beta, two plant managers spent time in the JV and then returned to plant management positions in American parent plants. Strategic Integration. Doz (1996) argued that the partner interface is critical to the parent’s appreciation of the differences between the partners. A narrow and distant interface was found to be an obstacle to learning (Doz 1996). The process through which a JV strategy is linked with a parent strategy is termed strategic integration (Harrigan and Newman 1990). A JV perceived as peripheral to the parent organization’s strategy is likely to yield few opportunities for the transfer of alliance knowledge to the parent. A JV closely related to the parent strategy may receive more attention from the parent organization, leading to substantial parent-JV interaction and a greater commitment of resources to the management of the collaboration. As Hamel (1991) argued, receptivity to learning is enhanced if the parent and its alliance are closely related. Note the assumption that the linkages are consistent with the strategic goals of the parents and JVs. Through strategic linkages between the JV and the parent, the partners can gain important insights into each

462

other’s businesses. For example, one of the American parents won a contract to supply a part but was unable to meet the target cost. The parent decided to use its JV to produce the part because of the JV’s superior process technology. Such a linkage indicates that the American parent management had internalized the knowledge associated with differences between the parent and JV. The linkage also opened the door for more knowledge sharing and cooperation in the future. To maximize exposure to strategic knowledge, alliance partners must go beyond the narrow confines of the JV agreement. In Gamma and Kappa, the JV functioned like a related division of the American parent, with the parent focused on managing the partner relationship, not just the JV itself. The JV was incorporated into the parent’s strategic planning processes and was expected to contribute new ideas and provide leadership in a particular technology area. The relationship between the partners was becoming much tighter. For Kappa, the JV was formed strictly as a transplant supplier that was relatively independent of its American parent, relying extensively on the Japanese partner for product technology and marketing support. Over the years, Kappa became less independent as ties between the two partners increased. For Gamma, the JV was initially presented to the transplant customers as a Japanese company. The JV evolved into a much less “Japanese” firm and, through its American parent’s contact, developed a substantial amount of business with domestic customers. The objective, as a Gamma manager remarked, was for both the JV and the parents to benefit. Types of Knowledge Building on the preceding description of the knowledge management processes, the primary types of knowledge associated with each process were identified. Table 4, based on Spender’s (1996b) typology, shows the classification and provides examples of each type of knowledge. Because the table is based on our observations of knowledge types, it does not include certain knowledge types that might be associated with the processes in different research settings. For example, technology sharing in our study involved knowledge with low tacitness. Clearly, technology could be viewed as having a significant tacit dimension, but in the cases studied, the knowledge transferred (and the knowledge of apparent interest to the American partners) had a minimal tacit dimension. The knowledge associated with technology sharing was classified as explicit, objectified knowledge because it was related primarily to product designs or specific manufacturing processes. For example, knowledge about quality control processes and factory scheduling systems




was of potential interest to most of the American partners. For the technology-based knowledge, a strong explicit memory component was embodied in the practices of the JV and, because much of the knowledge in the JV originated in Japan, also in the practices of the Japanese parent. Most knowledge shared through JV-parent interactions was objectified, although there was a high potential for sharing tacit, collective knowledge, such as that associated with a commitment to product quality. We view product quality knowledge as tacit because it was associated with a culture and philosophy about business and was not based on specific rules or guidelines.4 The JVparent interactions provided an excellent opportunity for the American parents to acquire such knowledge, but considerable resistance to it was evident at the American parent level. In one case, the value of quality process knowledge was discounted by the parent (even after the Japanese parent had offered to facilitate the knowledge transfer) with the argument that “what the JV does would never work here,” even though the JV and parent were producing very similar products. Personnel transfers had the potential to transfer tacit, difficult to articulate knowledge, such as beliefs and norms of behaviors. For example, an important belief in the JVs was that Japanese transplant customers had greater expectations about customer satisfaction than American customers. Because that belief was usually shared across levels of the organization, it is classified as collective knowledge within the JV. Other types of knowledge were more personal, such as the nature and importance of Japanese partner approaches to human resource practices. Because our focus was managerial movements between the JV and parent, the type of individual knowledge of greatest potential for transfer was conscious knowledge. Had we addressed the transfer of personnel below the managerial level, we undoubtedly would have identified instances of automatic knowledge transfer. Although strategic integration creates linkages that are organizational, our findings suggest that potential knowledge transfers may involve both social and individual knowledge. Through strategic cooperation, JV partners can gain access to objectified, explicit knowledge as well as to culture-related, communal knowledge about organizational behaviors and norms. The knowledge may also be of a tacit, individual nature because there is no consensus about the strategic value of the information. In this study, individual JV managers learned much about the Japanese partner’s interorganizational or keiretsu relationships. Such knowledge had potential strategic value,


Table 4

Knowledge Management Processes and Types of Knowledge

Knowledge Management Processes

Types of Knowledge

Examples of Knowledge Potentially Useful to American JV Parents

Technology sharing

Explicit/objectified

Quality control processes Product designs Scheduling systems

JV-Parent interactions


Specific human resource practices

Tacit/collective

Quality Commitment

Tacit/collective

Continuous improvement objectives Commitment to customer satisfaction

Tacit/conscious

Meaning of quality from Japanese partner perspective


Market intelligence Partner’s keiretsu relationships

Tacit/collective

JV competitive advantage

Tacit/conscious

Visions for the future Implications of the partner’s keiretsu relationships

Personnel transfers

Strategic integration

although our study suggests that American partner managers were generally unsure of how to capture the value. Organizational Levels Learning and knowledge creation occur through a process involving various organizational levels and actors. For each of the knowledge management processes, Figure 2 provides a multidimensional view linking the primary knowledge type and organizational levels in the parent firms. Like Table 4, the figure summarizes the findings for the five cases and is based on the identified (as opposed to ideal) knowledge transfer. As Table 3 shows, in

463



Figure 2

Knowledge Management Processes and Type of Knowledge Transferred

two of the cases little knowledge was transferred from the JV to American parent. The boxes in Figure 2 can be interpreted by height, width, and volume. The height of the box represents the range of knowledge types associated with the knowledge management process. The higher the box, the greater the range of knowledge types. Box width represents the range of organizational levels affected by the knowledge management process. Box volume provides an indication of the overall intensity of the process. The technology sharing process involved primarily explicit, social knowledge that could be transferred to the group and organizational levels within the American parents. The personnel transfers process involved mainly knowledge with medium to high tacitness that influenced individual and group levels. Based on our findings, we

464

conclude that only a limited amount of knowledge associated with personnel transfers “spiraled” beyond the group level to the organizational level. Strategic integration involved knowledge ranging from low to high in tacitness that penetrated mainly the group-organization levels. Strategic integration also generated some individual knowledge. The JV-parent interaction process had the potential to transfer knowledge with similar tacitness to that transferred by strategic integration, but on a more limited scale and at lower organizational levels. Resistance in the American parents to the costs of learning limited the effectiveness of the process at the organization level. One explanation for that resistance was that American parent organizations were so lean that little time was available to invest in learning. This supports the view that Western organizations try to learn in large, discrete steps (Hedlund



and Nonaka 1993) and often fail to recognize the value of incremental learning.5


Discussion This study contributes to the literature on knowledge and the firm by examining knowledge management in an alliance context. We have sought to systematically establish how firms acquire and manage new knowledge. We identified specific knowledge management processes by examining firms’ efforts to exploit JV learning opportunities and linked those processes with types of knowledge and organizational levels. Proposition Development Several propositions about organizational knowledge transfer and management can be derived from the findings. Two pertain to relationships between knowledge tacitness, organizational level, and transfer effectiveness, and three pertain to transfer effectiveness when the core of knowledge transferred is highly tacit. First, we can predict that the more tacit the knowledge, the lower the organizational level through which successful transfers will occur. Highly tacit knowledge is intuitive, nonverbalizable, and related to individual experiences. First-hand experiences with tacit knowledge are critical to its successful transfer. Knowledge that is low in tacitness is often related to product and process technology transfers that can occur on a higher, more collective level. Hence, Proposition 1. PROPOSITION 1. The tacitness of transferred knowledge will have an inverse relationship to the organizational level where initial transfer takes place; the greater the tacitness, the more likely individuals will be the primary knowledge transfer agents. Similarly, we can argue that when knowledge transfers are initiated at the group and organization levels, perhaps through team visits or group seminars, the transfers will be less effective when the knowledge has a high tacit element. PROPOSITION 2. The effectiveness of knowledge transfers initiated at the collective level will be negatively related to the tacitness of the knowledge. Figure 2 shows that the identified knowledge transfer processes are primarily in quadrant 2, with personnel transfers occupying the top half of quadrant 3. Clearly, the American firms were focused on explicit knowledge. This is consistent with the argument that in their approach to organizational learning, Western firms tend to focus on explicit knowledge that can be created through analytical skills and concrete forms of oral and visual presentation


(Nonaka and Takeuchi 1995). Although in all the cases the American firms formed JVs with an objective of learning from their Japanese partners, the learning expectations revolved around “what” the Japanese knew, rather than “how” and “why” the Japanese firms knew what they knew. In other words, initial emphasis was on explicit knowledge. The American firms expected to find visible differences in the JV that could be analyzed and incorporated in the parent. The absence of highly visible differences in systems and processes was often equated with low learning potential. Because of the focus on explicit knowledge, the American firms often began their collaboration with the view that the knowledge management processes based on technology sharing and interorganizational interaction were the most viable. However, in one of the cases, the focus on narrow, technology-related learning objectives resulted in an early abandonment of technology sharing efforts. Hence, Proposition 3. PROPOSITION 3. Firms that focus their initial learning efforts on explicit knowledge will tend to ignore tacitknowledge-based learning opportunities, thereby increasing their propensity to undervalue overall learning potential. Earlier we noted that knowledge connections create the potential for individuals to share their observations and experiences. The firms most successful in knowledge transfer recognized that important knowledge could not be internalized without substantial interaction between the people in the parent and those in the JV. This was particularly true when the knowledge had a high level of tacitness. PROPOSITION 4. The more successful the transfer of tacit knowledge, the greater the individual interactions between the “learning” organization and the “teaching” organization. Firms increasingly saw the need for strategic relationships between the two organizational units as a means of solidifying the knowledge linkages. The strategic integration process involved less visibly defined objectives than the technology sharing process and in that sense, enabled the communication of more tacit knowledge. PROPOSITION 5. The more successful the transfer of tacit strategic knowledge, the greater the strategic relationship between the “learning” organization and the “teaching” organization. Managerial Implications Spender (1996b) argued that the most strategically important feature of a firm is its body of collective knowledge. We identified several knowledge management processes that firms use to exploit alliance learning

465



opportunities. Although all of the processes are potentially effective, we found that the different processes involved different types of knowledge and different organizational levels. Organizations create, store, transfer, and discard various types of knowledge. Therefore, organizations must engage in a variety of knowledge management processes. Personnel transfers can be an effective process through which to acquire tacit knowledge that can be acquired only through time and experience. The risk with personnel transfers is that if the knowledge remains individual, the potential social impact of the learning is lost. To maximize the effectiveness of personnel transfers, systems may have to be established to ensure that knowledge goes beyond the individual level. Strategic integration can be an effective higher level knowledge sharing tool. It enables meaningful communication and collaboration between organizations at the group and organizational levels rather than at the individual level. Personnel transfer schemes and strategic integration suggest a long-term basis for knowledge sharing and potentially allow for the largest amounts of knowledge to travel interorganizationally. Additionally, such long-term processes create the potential for a continuous flow of knowledge, which in turn can lead to continuous learning and change. However, as we found, personnel transfers do not always result in significant organization level knowledge transfer. The other two processes, technology sharing and JVparent interactions, are based on shorter term knowledge relationships and as such, are less effective in transferring tacit knowledge. Nevertheless, this study shows that they can be effective as a means of acquiring explicit, objectified knowledge. Moreover, given that the decision to initiate knowledge creation efforts must be balanced with the cost of doing so, technology sharing and JV-parent interaction processes may be less costly than strategic integration and personnel transfers. Visits and tours of JV facilities were identified as a simple and effective means for parent managers to interact with JV managers. In summary, this research suggests that organizations must be aware of the different types of knowledge and design appropriate systems to process the knowledge. Clearly, firms will attach different values to JV knowledge, and therefore knowledge creation and processing strategies will differ across organizations and also evolve over time. We found that a variety of knowledge management strategies can be useful, although some strategies lead to more effective knowledge transfer than others. For example, in four of the five cases (all but Sigma), the Japanese partner was responsible for the manufacturing process and product technology, which obviously influenced the type of technology knowledge that could be

466

exchanged between the partners. Also, industry conditions can influence learning intent and managerial commitment. When our research began in 1990, the U.S. automotive industry was under serious attack by Japanese firms. Suppliers had to cope with new competitors, and their traditional customers were losing market share. By 1994, the situation had changed dramatically. The domestic auto industry was recapturing some of its market share, and suppliers were reaping the benefits. The learning imperative that prevailed in 1990 was no longer as critical. That change highlights one of the problems with cross-sectional research, namely that firms and industries are in constant evolution. Longitudinal research captures the evolutionary patterns in the underlying research context.

Conclusion There are several underlying assumptions in the paper. First, there can be a significant payoff in cooperating, namely knowledge creation. Although not all knowledge creation efforts will have immediate performance payoffs, over the long term successful knowledge creation should strengthen and reinforce a firm’s competitive strategy. Second, each alliance partner has knowledge that, at least in part, should be considered valuable by the other partner(s). Third, knowledge creation is a dynamic process involving interactions at various organizational levels and an expanding community of individuals that enlarge, amplify, and internalize the alliance knowledge. Finally, knowledge creation and the upward movement of knowledge through the different organizational levels can be at least partially responsive to managerial influence. To be successful, organizations must not only process information but also create new information and knowledge. This study explored how organizations involved in alliances can use their alliance experience as the basis for managing and creating knowledge. We provide some empirical evidence to support the conceptual arguments made by Hedlund (1994) and Spender (1996b) about the firm as a dynamic system of processes involving different types of knowledge. Further research is needed to probe deeper into the relationships between organizational levels, types of knowledge, and organizational processes. In particular, there is a need to study how organizations manage highly tacit knowledge that resides at the collective level of the organization. Such an orientation can help capture the dynamics of knowledge management and greatly enrich understanding of how firms acquire and transfer knowledge. Acknowledgments Grants to the first author from the Carnegie Bosch Institute for Applied Studies in International Management, Temple University’s Center for


ANDREW C. INKPEN AND ADVA DINUR Knowledge Management Processes and International Joint Ventures East Asian Studies, and the Thunderbird Business Research Center are gratefully acknowledged. The authors thank Rob Grant, Ikujiro Nonaka, and J. C. Spender for their helpful comments.


Endnotes 1 Confirmed through discussions between the first author and several General Motors managers. 2 Spender (1996b) emphasized that the boundaries between the types of knowledge are imprecise. 3 Although the sample included two Canadian firms, for brevity we refer to the sample as American rather than North American. 4 In contrast, specific quality control systems or processes that involve rules and guidelines would be classified as objectified knowledge. 5 For a detailed discussion of Western versus Japanese knowledge management processes, see Hedlund and Nonaka (1993). Nonaka and Takeuchi (1995, ch. 8) use several Japanese company examples to describe how organizational knowledge creation takes place on a global scale.

References Badaracco, J. L. (1991), The Knowledge Link, Boston, MA: Harvard Business School Press. Bleeke, J. and D. Ernst (1995), “Is Your Strategic Alliance Really a Sale?” Harvard Business Review, 73, 1, 97–105. Brown, J. S. and P. Duguid (1991), “Organizational Learning and Communities of Practice: Towards a Unified View of Working, Learning, and Organization,” Organization Science, 2, 40–57. Dodgson, M. (1993), “Learning, Trust, and Technological Collaboration,” Human Relations, 46, 77–95. Doz, Y. (1996), “The Evolution of Cooperation in Strategic Alliances: Initial Conditions or Learning Processes?” Strategic Management Journal, 17, 55–83. Drucker, P. F. (1995), “The Network Society,” Wall Street Journal, March 29, 12. Geringer, J. M. and L. Hebert (1991), “Measuring Performance of International Joint Ventures,” Journal of International Business Studies, 22, 253–267. Glaser, B. and A. L. Strauss (1967), The Discovery of Grounded Theory: Strategies for Qualitative Research, New York: Aldine. Grant, R. M. (1996), “Prospering in Dynamically-competitive Environments: Organizational Capability as Knowledge Integration,” Organization Science, 7, 375–388. Gulati, R., T. Khanna, and N. Nohria (1994), “Unilateral Commitments and the Importance of Process in Alliances,” Sloan Management Review, 35, 3, 61–69. Hamel, G. (1991), “Competition for Competence and Inter-Partner Learning Within International Strategic Alliances,” Strategic Management Journal, 12 (special issue), 83–104. Hamilton, D. P. (1995), “Sony Expands in Computer-Linked Gear: Effort Requires Shift in Go-It-Alone Tradition,” Wall Street Journal, April 14, A8. Harrigan, K. R. (1985), Strategies for Joint Ventures, Lexington, MA: Lexington Books. —— and W. H. Newman (1990), “Bases of Interorganizational CoOperation: Propensity, Power, Persistence,” Journal of Management Studies, 27, 417–434. Hedlund, G. (1994), “A Model of Knowledge Management and the N-


Form Corporation,” Strategic Management Journal, 15, 73–90. —— and I. Nonaka (1993), “Models of Knowledge Management in the West and Japan,” in P. Lorange, B. Chakravarthy, J. Roos, and A. Van de Ven (Eds.), Implementing Strategic Processes: Change, Learning, And Co-Operation, Oxford, UK: Basil Blackwell, 117–144. Huber, G. P. (1991), “Organizational Learning: The Contributing Processes and a Review of the Literature,” Organization Science, 2, 88–117. Inkpen, A. C. (1995a), “Organizational Learning and International Joint Ventures,” Journal of International Management, 1, 165– 198. —— (1995b), The Management of International Joint Ventures: An Organizational Learning Perspective, London, UK: Routledge Press. —— (1996), “Creating Knowledge Through Collaboration,” California Management Review, 39, 1, 123–140. —— and P. W. Beamish (1997), “Knowledge, Bargaining Power and International Joint Venture Stability,” Academy of Management Review, 22, 177–202. —— and M. M. Crossan (1995), “Believing is Seeing: Joint Ventures and Organizational Learning,” Journal of Management Studies, 32, 595–618. Katz, R. and T. J. Allen (1988), “Organizational Issues in the Introduction of New Technologies,” in R. Katz (ed.), Managing Professionals In Innovative Organizations, Cambridge, MA: Ballinger, 442–456. —— and M. L. Tushman (1980), “External Communication and Project Performance: An Investigation Into the Role of Gatekeepers,” Management Science, 26, 1071–1085. Keller, M. (1989), Rude Awakening: The Rise, Fall, and Struggle for Recovery at General Motors, New York: William Morrow. Kogut, B. (1988), “Joint Ventures: Theoretical and Empirical Perspectives,” Strategic Management Journal, 9, 319–322. —— and U. Zander (1992), “Knowledge of the Firm, Combinative Capabilities, and the Replication of Technology,” Organization Science, 3, 383–397. Levinthal, D. A. and J. G. March (1993), “The Myopia of Learning,” Strategic Management Journal, 14, 95–112. Lyles, M. A. (1988), “Learning Among JV-Sophisticated Firms,” in F. Contractor and P. Lorange, (eds.), Cooperative Strategies in International Business, Toronto, Canada: Lexington Books, 301– 316. Mahoney, T. A. and J. R. Deckop (1993), “Y’gotta Believe: Lessons from American-Vs. Japanese-Run U.S. Factories,” Organizational Dynamics, 21, 4, 27–38. Makhija, M. V. and U. Ganesh (1997) “The Relationship Between Control and Partner Learning in Learning-Related Joint Ventures,” Organization Science, 8, 6, 508–525. Miller, K. L. (1993), “GM’s German Lessons,” Business Week, December 20, 67–68. Mintzberg, H. (1990), “Strategy Formation: Schools of Thought,” in J. Frederickson (ed.), Perspectives of Strategic Management, New York: Harper Business, 105–235. Mowery, D. C., J. E. Oxley, and B. S. Silverman (1996), “Strategic

467


ANDREW C. INKPEN AND ADVA DINUR Knowledge Management Processes and International Joint Ventures Alliances and Interfirm Knowledge Transfer,” Strategic Management Journal, 17 (special issue), 77–92. Nonaka, I. (1994), “A Dynamic Theory of Organizational Knowledge,” Organization Science, 5, 14–37. —— and J. K. Johansson (1985), “Organizational Learning in Japanese Companies,” in R. Lamb and P. Shrivastava (eds.), Advances in Strategic Management, Vol. 3, Greenwich, CT: JAI Press, 277– 296. —— and H. Takeuchi (1995), The Knowledge-Creating Company: How Japanese Companies Create the Dynamics of Innovation, New York: Oxford University Press. Parkhe, A. (1991), “Interfirm Diversity, Organizational Learning, and Longevity in Global Strategic Alliances,” Journal of International Business Studies, 22, 579–602. Polanyi, M. (1962), Personal Knowledge: Towards a Post-critical Philosophy, Chicago, IL: University of Chicago Press. Prahalad, C. K. and G. Hamel (1994), “Strategy as a Field of Study: Why Search for a New Paradigm?” Strategic Management Journal, 15, 5–16. Pucik, V. (1991), “Technology Transfer in Strategic Alliances: Competitive Collaboration and organizational Learning,” in T. Agmon and M. A. Von Glinow (eds.), Technology Transfer in International Business, New York: Oxford University Press, 121–138. Quinn, J. B. (1992), The Intelligent Enterprise, New York: Free Press. Simonin, B. L. and D. Helleloid (1993), “Do Organizations Learn? An Empirical Test of Organizational Learning in International Stra-

tegic Alliances,” in D. Moore (ed.), Academy of Management Best Paper Proceedings, 222–226. Spender, J. C. (1996a), “Organizational Knowledge, Learning, and Memory: Three Concepts in Search of a Theory,” Journal of Organizational Change, 9, 63–78. —— (1996b), “Making Knowledge the Basis of a Dynamic Theory of the Firm,” Strategic Management Journal (special issue), 45–62. Tushman, M. and T. J. Scanlon (1981), “Boundary-Scanning Individuals: Their Role in Information Transfer and Their Antecedents,” Academy of Management Journal, 24, 289–305. Von Krogh, G., J. Roos, and K. Slocum (1994), “An Essay on Corporate Epistemology,” Strategic Management Journal, 15, 53– 71. Westney, D. E. (1988), “Domestic and Foreign Learning Curves in Managing International Cooperative Strategies,” in F. Contractor and P. Lorange (eds.), Cooperative Strategies in International Business, Toronto, Canada: Lexington Books, 339–346. Winter, S. G. (1987), “Knowledge and Competence as Strategic Assets,” in D. Teece (ed.), The Competitive Challenge, Cambridge, MA: Ballinger Publishing, 159–184. Yin, R. K. (1989), Case Study Research: Design and Methods, revised edition, Newbury Park, CA: Sage. Zander, U. and B. Kogut (1995), “Knowledge and the Speed of Transfer and Imitation of Organizational Capabilities: An Empirical Test,” Organization Science, 6, 76–92.

Accepted by Ikujiro Nonaka; received November 5, 1996. This paper has been with the authors for one revision.

468


Inkpen

Recommend Documents