© Academy of Management. All rights reserved. Content may NOT be copied, e-mailed, shared or otherwise transmitted without written permission. This non-copyedited article version was obtained from the Academy of Management InPress website and is intended for personal or individual use.
Towards an Integrative Perspective on Alliance Governance: Connecting Contract Design, Trust Dynamics, and Contract Application
Dries Faems University of Twente, Netherlands
[email protected] Maddy Janssens Katholieke Universiteit Leuven, Belgium
[email protected] Anoop Madhok Schulich School of Business York University, Canada
[email protected] Bart Van Looy Katholieke Universiteit Leuven, Belgium
[email protected]
1
2
Towards an Integrative Perspective on Alliance Governance: Connecting Contract Design, Trust Dynamics, and Contract Application
ABSTRACT Based on a case-study of two sequential alliances between the same firms, we develop a more integrative perspective on alliance governance, providing insights into the interactions between structural and relational aspects, both within and between transactions. In particular, we disentangle a) how contracts with a similar degree but different nature of formalization (narrow versus broad) trigger different kinds of trust dynamics (negative versus positive) at both operational and managerial levels, b) how trust dynamics and contract application (rigid versus flexible) co-evolve over time, and c) how relational dynamics in previous transactions influence the design of contracts in subsequent transactions.
3
Towards an Integrative Perspective on Alliance Governance: Connecting Contract Design, Trust Dynamics and Contract Application
INTRODUCTION During the past two decades, alliances have become an increasingly popular strategy for organizations to complement and supplement their internal activities (Doz & Hamel, 1998; Faems, Van Looy & Debackere, 2005; Hagedoorn, 2002). Although popular, their failure rates are high (Bleeke & Ernst, 1991), instigating numerous scholars to study the governance of alliances or the process of safeguarding and coordinating the alliance in order to influence its evolution and performance over time (Doz & Hamel, 1998). In this research stream, two different theoretical perspectives aiming to develop insights into effective and efficient governance have emerged (Madhok, 1995a; Powell, 1998). The first perspective focuses on the structural design of single transactions and emphasizes the importance of contracts, or ‘agreements in writing between two or more parties, which are perceived as legally binding’ (Lyons & Martha, 1997: 241), as effective and efficient governance mechanisms. The second perspective focuses on relational processes within ongoing interfirm relationships and emphasizes the importance of trust for safeguarding and coordinating the alliance, where trust is defined as ‘a psychological state comprising the intention to accept vulnerability based on positive expectations of the intentions or behavior of another’ (Rousseau, Sitkin, Burt & Camerer, 1998: 395). While alliance governance scholars have mainly relied on either the structural or relational perspective, we seek to provide a more integrative understanding of alliance governance through exploring how the design and application of structural elements (i.e., contracts) is related to relational processes such as trust dynamics (i.e., how trust evolves over time) at both operational and managerial levels. To accomplish this research objective,
4
we conduct a case study where we investigate two sequential explorative R&D alliances between the same pair of firms. Our methodological approach differs from previous research on alliance governance in four fundamental ways. First, we conduct an in-depth content analysis of the observed contracts, examining the nature of contractual formalization (i.e., the content of contractual clauses). This focus on content is different from previous alliance governance research (e.g., Mellewigt, Madhok and Weibel, 2007; Parkhe, 1993a; Poppo & Zenger, 2002; Reuer & Ariño, 2007) that only measures the degree of contractual formalization (i.e., the number of contractual clauses). Second, while previous alliance governance research has mainly taken a static view on trust, measuring it at a single point in time (Lewicki, Tomlinson & Gillespie, 2006), we apply a more dynamic view, focusing on trust dynamics or how expectations about the partners’ competencies and intentions evolve over time. Third, we deviate from previous research by interviewing both managers and engineers in both companies involved. This approach allows for a multi-level process analysis of the unfolding trust dynamics at both the operational and managerial level and their connections with contract design and contract application at the governance level. Fourth, while previous alliance governance research focuses on a single alliance transaction or the interfirm relationship (Powell, 1998), we apply a more combinative approach, observing two alliance transactions that are embedded within one particular relationship. These four points of difference from extant alliance research allow us to extend and deepen theoretical insights into the phenomenon in a novel and distinctive manner that we demonstrate in the ensuing pages. The paper is organized in six sections. First, we position our study in the existing alliance governance literature, discussing how it addresses the gap between structural and relational perspectives. Next, we discuss our methodology. We then provide a detailed description of the observed alliances. For each observed alliance transaction, we subsequently
5
develop a multi-level process model identifying particular connections between the design and application of contracts at the governance level and relational processes at both the operational and managerial level. We then present the main theoretical implications of our findings. In particular, we provide a more integrative perspective on alliance governance, connecting structural and relational aspects, both within and between transactions. Finally, we conclude by suggesting avenues for future research and pointing to the main limitations of our study. THEORETICAL BACKGROUND Structural and Relational Perspectives on Alliance Governance In the alliance literature, two different perspectives - structural and relational - have contributed to theory building on alliance governance (Madhok, 1995a, Powell, 1998). These two perspectives have their specific assumptions, theoretical bases and foci of analysis. Accordingly, they propose different mechanisms to govern alliances, and are consequently criticized for different reasons. Table 1 presents the main differences between the structural and relational perspectives. ----- Insert Table 1 about here ----Focusing on individual alliance transactions and relying on transaction cost theory (e.g., Williamson, 1985), the structural perspective assumes that alliance partners have a tendency to act opportunistically. The higher the level of transactional attributes such as asset-specificity and uncertainty1, the higher the risk that one or both partners engage in opportunistic actions (Oxley, 1997; Williamson, 1991). This perspective also assumes that the initial structural design of the alliance transaction is the most crucial factor in explaining alliance performance (Hennart, 2006).
1
The complete transaction cost framework also includes transaction frequency. However, this construct has received limited attention in empirical governance research (Geyskens, Steenkamp & Kumar, 2006).
6
In contrast, the relational perspective focuses on interfirm relationships as they evolve over time and over multiple transactions. Relying on social exchange theory (e.g., Blau, 1964), this perspective assumes that alliance partners tend to behave in a trustworthy manner, especially when a history of successful collaboration is present. In addition, they consider ongoing relational management of the interfirm relationship (i.e., fostering communication and trust) to be more important in explaining alliance performance than crafting the initial structural design (Salk, 2005). Grounded in these assumptions, each perspective points to different mechanisms to govern alliances. The structural perspective identifies complex contracts, or ‘contracts with a large number of clauses that are specified in detail’ (Ariño & Reuer, 2005), as safeguarding devices that mitigate the perceived risk of opportunistic behaviour (Deeds & Hill, 1998). In particular, through specifying what is and what is not allowed and through inflicting penalties for the commission of violating behaviors, complex contracts reduce both the ability and willingness of partners to act opportunistically (Parkhe, 1993a). It is also argued that contracts are coordination mechanisms which, through specifying a precise division of labor across partners and providing procedures for the integration of dispersed activities, simplify decision making and prevent disputes on how to achieve tasks (e.g., Mellewigt et al., 2007; Reuer & Ariño, 2007). In sum, it is argued from the structural perspective that complex contracts create a predictable collaborative environment that mitigates exchange hazards and facilitates coordinated action. In contrast, the relational perspective promotes a more relational governance strategy where partners rely on trust to address issues of safeguarding and coordination. While a broad variety of trust definitions exist, there is a converging understanding that trust refers to positive expectations regarding the other in a risky situation (Lewicki et al., 2006; Rousseau et al., 1998). In addition, scholars (e.g., Das & Teng, 2001; Nooteboom, 1996) increasingly
7
argue that trust is a multidimensional concept, encompassing positive expectations about a partner’s
ability
to
perform
according
to
the
agreement
(competence
trust) as well as her intentions to do so (goodwill trust). According to the relational perspective, trust can be used as an alternative governance mechanism for two reasons. First, trust provides alliance partners with the assurance that knowledge and information will be used for the greater good (Jones & George, 1998). In this way, the perception of opportunistic hazards is likely to be limited, reducing the need for costly and inflexible formal safeguarding mechanisms such as complex contracts (Dyer & Singh, 1998; Larson, 1992). Second, under conditions of trust, members of different partners are likely to engage in extensive communication and information sharing on an informal basis (Ring & Van de Ven, 1994; Uzzi, 1997). Coordination between partners can be consequently achieved not by referring to contractual systems and procedures, but by processes of mutual adjustment (Mintzberg, 1979). Both structural and relational perspectives on alliance governance are criticized for different reasons. The structural perspective is faulted for being acontextual and ahistorical (Granovetter, 1985; Zajac & Olsen, 1993), neglecting the social context within which the alliance transaction is embedded2. At the same time, those who promote a more relational perspective are often criticized for having too ‘rosy’ a view of human nature. In particular, they tend to overemphasize issues like relational embeddedness, ignoring transaction-situated issues like opportunism and contractual hazards (Jeffries & Reed, 2000). Connecting Structural and Relational Perspectives Although structural and relational alliance governance perspectives tend to exclusively focus on the structure of the transaction or the relational processes within the interfirm relationship,
2
Some scholars (e.g., Reuer & Ariño, 2007; Reuer, Ariño & Mellewigt, 2006) have conducted first attempts to include the social context into their analyses of the design of structural elements such as contracts. However, these studies only measure the social context by counting the number of prior alliance transactions, without providing insights into the nature and dynamics of the underlying relational processes (Contractor, 2005).
8
the notable study of Doz (1996) already indicates that these two aspects are related to each other in governing alliances. In his theoretical framework of alliance evolution, Doz (1996) suggests that relational processes in alliances mediate between initial structural conditions and alliance outcomes. Based on this framework, numerous scholars (e.g., Osborn & Hagedoorn, 1997; Madhok & Tallman, 1998; Contractor, 2005; Bell, den Ouden & Ziggers, 2006; Vlaar, Van den Bosch & Volberda, 2007) have called for empirical research that explores the connections between structural and relational governance perspectives. Some scholars have responded to this call, examining the relationship between the two governance mechanisms that have been stressed by these two perspectives: complex contracts and trust. However, the results of these studies are rather ambiguous. Some studies (Luo, 2002; Poppo & Zenger, 2002) provide evidence that complex contracts facilitate trust building, whereas other studies (Lyons & Mehta, 1997; Malhorta & Murnighan, 2002) conclude that complex contracts negatively influence the level of trust between partners. We argue that these inconsistent findings might be due to three important issues. First, existing research (e.g., Parkhe, 1993a; Deeds & Hill, 1998; Mellewigt et al., 2007; Reuer & Ariño, 2007) exclusively focuses on the degree of contractual formalization or the number of clauses that are defined in the contract. These studies assess for each observed alliance contract the absence/presence of a predefined set of contractual clauses, resulting in a score that reflects the degree of formalization of the contract. However, such an approach ignores the nature of formalization or the actual content of the contractual clauses (Klein-Woolthuis, Hillebrand & Nooteboom, 2005). This aspect is important, as shown by a recent study (Hagedoorn & Hesen, 2007). Based on in-depth content analysis of six interfirm contracts, Hagedoorn and Hesen (2007) conclude that similar types of juridical clauses, such as dispute resolution clauses, intellectual property clauses and warranties, are given a different content depending on the partnership form. For instance, in equity joint ventures and non-equity
9
partnership contracts, dispute settlement clauses favor internal and private dispute resolution whereas, in licensing contracts, dispute settlement clauses emphasize the use of court enforcement. Whereas this study clearly shows that the content of alliance contracts can significantly vary, it does not examine how such variation in the nature of contractual formalization influences relational processes such as trust dynamics. Second, these previous studies tend to focus attention on the initial design of contracts, while ignoring how such structural elements are applied during the alliance (Bell et al., 2006). However, previous case-study research on the evolution of alliances (e.g., Ariño & de la Torre, 1998; Doz, 1996; Madhok, 1995a) provides anecdotal evidence indicating that alliance contracts can be applied in different ways, ranging from strict enforcement of contracts to much more flexible approaches. In addition, some studies on collaborative dynamics within firms (Ghoshal & Moran, 1996; Sitkin & Stickel, 1996) suggest that contract application and trust are likely to co-evolve over time. In particular, they identified the occurrence of negative reinforcing cycles in which managers’ increased reliance on formal mechanisms triggers negative trust dynamics at the employee level, which in turn motivates managers to further increase the use of such formal mechanisms. In-depth process research into how the application of structural elements such as contracts and relational processes such as trust dynamics co-create rich collaborative dynamics in alliances has however remained absent. A third issue is the studies’ exclusive focus on relational processes at the managerial level. As alliance governance scholars mainly rely on managers as core informants (Currall & Inkpen, 2000), they tend to treat the operational level as a black box. However, as the roles of actors significantly vary across operational and managerial levels, the nature of relational processes such as trust dynamics might also vary (Zaheer, Lofstrom & Varghese, 2002). Moreover, in a recent conceptual study, Janowicz, Krishnan and Noorderhaven (2005)
10
suggest that relational processes at the operational level might substantially influence relational processes at the managerial level. We therefore argue that, in order to get a finegrained understanding of the connections between structural and relational aspects, it is necessary to explore the evolution of relational processes such as trust dynamics at both the managerial and operational level. In sum, previous attempts to link structural elements (i.e., contracts) to relational processes (i.e., trust dynamics) have provided inconsistent results. In this paper, we aim to create a richer understanding of the connections between these structural and relational aspects by exploring the following research questions: Research question 1: How does the content of the contract influence trust dynamics at both operational and managerial levels in alliances? Research question 2: How does the application of contracts co-evolve with trust dynamics at both operational and managerial levels in alliances? METHODOLOGY Research Design and Setting The purpose of this paper is to inductively build a more integrative perspective on alliance governance that explains how the design and application of structural elements is connected to relational processes at different levels (i.e., operational and managerial level). Although case studies have remained rather rare within the alliance governance literature3, this design is appropriate for such theory building because it allows us to 1) answer ‘how’ questions about a contemporary set of events over which the investigator has little or no control (Yin, 1984), 2) mobilize multiple observations on complex relational processes (Eisenhardt, 1989; Eisenhardt & Graebner, 2007; Parkhe, 1993b), and 3) draw in the significance of various interconnected levels of analysis (Hall, 2006; Pettigrew, 1990).
3
The studies of Ariño and de la Torre (1998), Larson (1992), Doz (1996), Yan and Gray (1994), and de Rond & Bouchikhi (2004) are notable exceptions in this respect.
11
We conducted an in-depth case study of two sequential explorative R&D alliances (i.e., Side Shooter Head (SSH) alliance and End Shooter Head (ESH) alliance) between the same two firms: Graph, one of the world’s leading imaging companies, employing over 20.000 people in 40 different countries, and Jet, a stock-quoted inkjet technology company employing 185 people. Both alliances were of an explorative nature, meaning that the purpose of each alliance was to evaluate new technological opportunities by conducting upstream activities such as fundamental research, experimenting and testing4 (Faems, Janssens, Bouwen & Van Looy, 2006; Koza & Lewin, 1998). Table 2 presents the main characteristics of both alliances. For each alliance, both Graph and Jet created an engineering team that was responsible for the execution of the project. In each company, these teams consisted of approximately five engineers5 who internally reported to two senior managers. ----- Insert Table 2 about here ----Data Collection and Analysis Data on the two explorative R&D alliances were collected in a retrospective way. Retrospective data collection allows for a more focused data-gathering process because it reduces the danger of data overload and collecting much unusable data (Leonard-Baton, 1990; Poole, Van de Ven, Dooley & Holmes, 2000). However, documenting cases in a retrospective way also has its disadvantages. For instance, respondents have the tendency to filter out events that do not fit or that render the story less coherent (Poole et al., 2000). To improve the validity of these retrospective reports and prevent accepting respondent bias, we
4
In contrast, the purpose of exploitative R&D alliances is to leverage partners’ existing technological capabilities through engaging in upstream activities such as standardization, refinement and upscaling. 5 In each team, the exact number of members fluctuated over time. For instance, additional engineers were sometimes brought into the teams on a temporary basis to help addressing particular technological problems. The engineers of the different teams also got support from technicians who carried out basic tasks such as handling the machines and constructing the prototype printheads. In this study, we did not interview these technicians as they were not involved in collaborative interactions between Graph and Jet.
12
applied a number of strategies. First, we triangulated our data, applying two data sources: interviews and documents (Eisenhardt, 1989; Yin, 1984). Second, we asked informants to reflect on concrete events rather than abstract concepts to reduce the risk of cognitive biases and impression management (Miller, Cardinal & Glick, 1997). Finally, we attempted to verify individual reports by asking similar questions to multiple informants (Cardinal, Sitkin, & Long, 2004). Following Pettigrew’s (1990) and Pentland’s (1999) suggestions, we distinguished three different stages in our theory-building efforts, going from surface level to deeper levels of data collection and analysis. In the first stage, we conducted unstructured interviews with two managers of Graph to obtain some initial information about the history and characteristics of the two explorative R&D alliances. For each alliance, we also studied relevant documents. In addition to publicly available information (i.e., annual reports, press releases), we obtained access to 126 private documents (i.e., contracts, reports of technical meetings, reports of steering meetings), representing a total amount of 1747 pages (see Table 2). Based on this information, we constructed a graphical representation of the chronology of the major events within each alliance. In the second stage, we conducted semi-structured interviews (Kvale, 1996) with informants of both organizations involved. As we wanted to analyze relational processes at both managerial and operational levels, we made an effort to interview not only managers but also engineers (see Table 2)6. Interviews were conducted individually, face-to-face and in the native language of the interviewee to maximize the informant’s ability to express his/her thoughts, feelings, and opinions. The interviews were structured along the chronology of the major events. As interviewees described these events, we asked additional ‘why’ and ‘how’ questions to get a better view on the role of structural and relational aspects in explaining the
6
1 Jet and 4 Graph informants were involved in both alliances, while others only participated in one alliance.
13
described collaborative dynamics. The average length of the interviews was between one and two hours. All the interviews were taped, transcribed and sent back to the interviewees for feedback. At this stage, we also reexamined available documents (i.e., reports of steering and technological meetings) to verify whether the content of the interviews was consistent with the content of the documents. When discrepancies occurred among data sources, we contacted respondents for additional questions. Subsequently, a case study report was written, reconstructing the history of the two explorative R&D alliances. In this report, we made extensive use of citations from both the interviews and documents in order to stay very close to the original data, achieving a high level of accuracy (Langley, 1999). These reports were sent to two managers in each company and their comments were collected via electronic mail or face-to-face conversations. In the third and last stage, we analyzed the data through an inductive approach, moving from first-order to second-order analysis. The first-order analysis was framed around the information as expressed by the informants (Gioia & Chittipeddi, 1991; Gioia, Thomas, Clark & Chittipeddi, 1994; Van Maanen, 1979). Each author individually conducted a content analysis of the SSH and ESH contract to identify differences in the nature of contractual formalization. Using statements from documents and interviews, each author also developed for each alliance a visual map (Plowman et al., 2007; Langley, 1999) of the interactions among contractual elements (i.e., contract design and application) and relational processes such as trust dynamics7 at both operational and managerial levels. Comparing our interpretations, we had at times identified the same differences in the nature of contractual
7
Previous alliance governance research tends to measure trust in a static way (Lewicki et al., 2006). However, scholars (e.g., Carson, Madhok, Varman & John, 2003; Rousseau et al., 1998) increasingly argue that, as partners receive first-hand information on the competencies and/or intentions of the other partner, the level of trust is likely to change during the alliance. We therefore applied a more dynamic approach. Statements that referred to a positive evolution of a partner’s expectations about the intentions/competencies of the other partner were categorized as positive goodwill/competence trust dynamics. Statements that referred to a negative evolution of a partner’s expectations about the intentions/ competencies of the other partner were categorized as negative goodwill/competence trust dynamics.
14
formalization but had used different labels to express them. Negotiation led to agreement on representative labels. Regarding the visual maps, some discrepancies between the authors were initially observed8. These inconsistencies were addressed by going back to the data, contacting informants, and intensive discussions among the authors9. The final result of this first-order analysis is described in the next section. During the second-order analysis, we moved to a more theoretical level, wherein the data and first-order findings were examined to arrive at an explanatory framework that allowed addressing the original research questions (Gioia & Chittipeddi, 1991; Gioia et al., 1994; Van Maanen, 1979). Such a second-order framework was first individually developed by each author. While we focused on identifying connections between our initial core concepts (i.e., contract design, trust dynamics, contract application), we also allowed for additional concepts to enter the analysis. For instance, as the second-order analysis proceeded, it became clear that we needed to add the concept of operational joint sensemaking (i.e., joint problem definition and joint problem solving; Allen, 1977) as an additional relational process to our explanatory framework to explain the connection between contract design and managerial trust dynamics. Next, we came together to discuss, compare, and integrate our individual efforts. This multi-stage process of independent, comparative, and collaborative analyses resulted in a multi-level process model, separating out governance, operational, and managerial levels and identifying connections among them. During the latter meetings, we also started confronting and comparing our empirical findings
8
For instance, while some authors had concluded that, because of intensive interaction, delays in the ESH project had been minimal, another author came to the conclusion that, despite intensive interaction, substantial delays had emerged in the ESH project. 9 By reanalyzing some technological documents and contacting one Graph engineer, we found that interaction on mechanical aspects had been intensive from the start of the ESH alliance, resulting in limited delays on these technological aspects. At the same time, it became clear that interaction on problems that were directly related to the core of Jet’s technology was lacking, resulting in substantial delays regarding electronic aspects.
15
with the two competing theoretical perspectives on alliance governance (i.e., structural and relational perspective). This ‘three-cornered fight’ (Hall, 2006: 27) finally resulted in new theoretical insights regarding the connections between structural elements and relational processes, both within as well as between alliance transactions. When a first draft of this paper was completed, we conducted feedback interviews with the two Graph managers who were interviewed at the first stage to discuss our interpretations and explanatory framework. These feedback interviews proved to be very helpful in fine-tuning our insights but also in testing the internal validity of our findings. SIDE SHOOTER HEAD AND END SHOOTER HEAD ALLIANCE In this section, we provide a description of the two explorative R&D alliances. For each alliance, we begin with a brief introduction of its initiation, followed by the negotiation of the alliance contract and the dynamics of collaboration in the alliance. Figure 1 and Figure 2 provide a summary of the dynamics of collaboration in the respective SSH and ESH alliance. The Side Shooter Head Alliance At the end of the nineties, Jet planned to initiate a new technological endeavor regarding inkjet technologies. Based on its recently invented SSH technology, the company wanted to develop a generic inkjet printhead that could be integrated into printing systems of different original equipment manufacturing (OEM) partners. Within Jet, this project was seen as a ‘revolutionary innovation project’ (Jet manager). Given Jet’s limited financial resources, the company aimed for partners to financially and technologically support the exploration of the SSH printhead. Jet recognized Graph as a promising partner because Graph possessed substantial financial resources as well as technological expertise in digital printing systems. Graph was interested in Jet’s proposal because it had recently recognized inkjet technology as an important emerging technology for printing applications.
16
Negotiation of SSH contract. At the end of 1998, negotiations began between Graph and Jet. In the interviews, both Graph and Jet managers stressed that, during these negotiations, Jet acted as the dominant partner, presenting the collaborative project as a takeit-or-leave-it proposition to Graph: Jet had orchestrated the whole scenario. This became apparent during the negotiations. This came across as quite arrogant… Their attitude communicated that, if it would not be possible to come to an agreement with Graph, they would do it with somebody else. (Graph manager) Our CEO applied a fairly bullish negotiating style. He found that projecting an air of high confidence in the technology was a winning strategy. (Jet manager) As Graph was eager to start working with this new technology, they accepted most conditions dictated by Jet. On March 25th 1999, the SSH Agreement was signed. In this contract, Graph agreed to fund Jet’s technological efforts to design and develop SSH prototype printheads. The contract stipulated that Graph would pay a ‘non-refundable contribution towards Jet’s development costs’ (SSH Agreement: 3) and that, in exchange for its financial contribution to the exploration of the SSH technology, ‘Graph would become one of at most three preferential partners to which Jet would sell the generic SSH printhead’ (SSH Agreement: 4). It was also stated in the contract that Graph would design and develop a prototype printing system which would allow testing the prototype printheads. The SSH Agreement contained a number of juridical clauses that dealt with concerns about the partners’ behavior outside of the alliance itself. First, it was stipulated that, in case of conflicts that could not be solved amicably, legal action by the partners was permitted. Second, the confidential nature of exchanged technological and commercial information was stressed, forbidding partners to pass on exchanged information to outside parties. Third, the contract stressed that the initiation of the alliance did not imply an agency relationship between the partners. Fourth, the contract contained a rule for allocating intellectual property between the partners if new inventions should emerge during their collaboration.
17
In the SSH Agreement, contractual clauses were also present that referred more directly to the actual execution of the collaborative agreement. We found contractual statements linking Graph’s payments to contractual milestones that specified deliveries of prototype printheads from Jet to Graph. Target dates and performance standards for these deliveries were also codified in the contract. Interviewees referred to these milestones, target dates and performance standards as ‘mechanisms that allowed monitoring the performance of the other partner’ (Graph manager). However, contractual clauses, specifying how partners had to achieve these milestones, remained absent in this contract. In the SSH contract, it was also stipulated that Jet’s engineering team was solely responsible for designing and developing prototype printheads, while Graph was solely responsible for designing and developing one prototype printing system into which the prototype printheads could be integrated. No contractual statements were present that obliged partners to exchange technological information during the project. Graph interviewees stressed that it was Jet’s management who had proposed to contractually define a strict task division that did not anticipate information exchange and only expected monitoring of the partners’ outputs: Jet’s CEO came to us and said ‘We propose this kind of collaboration and this is the contractual framework in which this collaboration should fit. You only need to sign it. (Graph manager) To explain Jet’s preference for this particular mode of collaboration, Jet interviewees referred to their fear of unintended knowledge spillovers. Although an explicit rule was present for the allocation of intellectual property rights, they still feared that Graph would be able to access sensitive knowledge and abuse this knowledge for their own benefits. By avoiding profound technological involvement of Graph in designing and developing the prototype printheads, Jet wanted to minimize the risk of unintended knowledge spillovers:
18
Jet had only one body of intellectual property. So the fear was that if we engaged Graph at a detailed technical level, our intellectual property would rapidly flow out of this building towards Graph. (Jet engineer) In addition, Jet interviewees mentioned that, if Graph were to become actively involved in the design and development of printheads, they would be able to influence the original design specifications in order to improve the applicability of the printhead for the particular printing systems that Graph envisioned. Such changes, however, would hamper the generic applicability of the printhead, making it much more difficult for Jet to sell the printhead to other OEM partners. As a consequence, Jet wanted to keep Graph at a distance: The attitude at Jet then was: ‘Graph is not going to be an exclusive printhead licensee. We therefore will keep them away from the core technology; we are going to keep them at a distance.’ (Jet manager) In sum, Jet managers feared that intensive technological involvement of Graph in designing and developing printheads could trigger self-interested actions by Graph. In order to mitigate these risks, Jet’s management preferred to contractually define a strict task division with limited information exchange and limited monitoring opportunities. Emergence of UTPs at Jet. After the SSH Agreement was signed, operational activities within the SSH alliance took off. As Jet started to experiment with the SSH technology, a number of unanticipated technological problems (UTPs) quickly emerged. Jet engineers faced huge operating problems regarding printing nozzles10 and experienced enormous difficulties in coating11 the prototype printheads. Graph engineers were not involved in addressing these technological issues. Rather, they were ‘watching from the sidelines’ (Graph project manager). Both Graph and Jet engineers explicitly referred to the strict task division to explain this limited involvement of Graph engineers in solving the unexpected problems:
10 11
A printing nozzle is a mechanical device designed to control the flow of ink within a printhead Coating involves the application of a thin film of functional material to a substrate, such as a printhead
19
[It] was concurrent engineering, each partner on its own domain of expertise… Jet did not ask for support. (Graph engineer) I would guess that the structure of the first agreement between Jet and Graph was limiting the involvement that Graph [engineers] could have. (Jet engineer) Graph interviewees also stressed that, because Jet perceived Graph as an investor instead of a technology development partner, Jet’s willingness to provide additional information on the technological problems was low: They [= Jet] were very reserved in telling details to us... They saw us as an investor who should be informed about the product that they were developing, but who should not know the details of the technology. (Graph engineer) At the same time, one Graph interviewee explicitly mentioned that ‘we could not really force them to provide more information’, indicating that the absence of contractual obligations to exchange information limited the ability of Graph engineers to become actively involved in addressing the emerging technological problems. Because of the strict task division between the partners and the limited interaction about the emerging technological problems, Graph engineers only had limited understanding of Jet’s technological problems. In addition, Graph engineers started to become frustrated over their inability to influence the progress of the project: That was an annoying situation. We saw that things went wrong and, at the same time, we were forced to stay on the sidelines. (Graph engineer) The frustrations of Graph engineers also had their impact on how Graph’s management assessed the intentions of Jet’s management. As Graph’s engineering team was not fully informed about the content of the emerging technological problems, Graph’s management started questioning the good intentions of Jet’s management. In particular, they got the impression that ‘Jet’s management was much more preoccupied with defending its own financial interests than the technological project itself.’ (Graph manager). At the same time, Jet’s concerns about potential opportunistic action by Graph further increased. One Jet interviewee, for instance, mentioned that ‘Graph’s demands for more information on our
20
technology were perceived as a confirmation of our theory [that established firms are after stealing knowledge].’ Emergence of serious delays. At the beginning of 2000, the UTPs at Jet resulted in considerable delays. Graph’s management, having serious doubts about the intentions of Jet’s management, opted for emphasizing the contractual milestones, perceiving this as the only available option for them to address the difficulties: We started to feel as if we were playing a game of poker with them [= Jet’s management]…The only way that we could react was by emphasizing the contractual issues. We told them that if they could not perform according to the milestones, we would not pay them. (Graph manager) The exertion of contractual pressure by Graph’s management made Jet’s management very nervous. Jet’s management feared the financial consequences of not delivering prototype printheads in a timely manner. Consequently, Jet engineers were instructed to develop and deliver prototype printheads to Graph as quickly as possible. Instead of focusing on sustainable solutions for the pending technological problems, Jet engineers started to look for technological short-cuts, allowing for the delivery of prototype heads to Graph. At the same time, Jet’s management was unwilling to deviate from the contractually agreed mode of collaboration in the project. Although Graph engineers wanted to know more about these UTPs and wanted to help Jet engineers in addressing some particular issues, Jet engineers were instructed by their own management to continue limiting the exchange of information to Graph: A lot of electronic communication was initiated. A lot of that could be characterized as the Graph guys trying to get more information… From my personal point of view, I would have liked to involve them more. But my instructions were not to… I was under instructions not to be entirely truthful. I always had to conceal part of the story. (Jet engineer) Jet interviewees again referred to the fear of Jet’s management for unintended knowledge spillovers and losing command over the project to explain this explicit order not to share information on technological problems with Graph:
21
Our management was both commercially and technically defensive. They were saying: ‘We do not want to lose technical and commercial control.’ (Jet engineer) As information exchange on the technological problems remained absent, Graph engineers started getting the feeling that ‘Jet engineers were further isolating themselves’ (Graph engineer). A few months after Graph’s management had started to increase contractual pressure, the first prototype printheads were delivered to Graph. However, Graph engineers quickly noticed that the quality of these printheads was not in accordance with the performance standards that were agreed in advance. Graph engineers consequently started to question the technological capabilities of Jet’s engineering team: In these circumstances, you get the feeling that they did not master their technology, that they do not understand the subtlety of their own inventions. (Graph engineer) As a result, Graph’s management started to have serious doubts about the technological feasibility of the SSH project. At Graph, the feeling also emerged that Jet was trying to ‘keep Graph happy’ (Graph manager) by complying with contractual commitments (i.e., milestones), even if this meant delivering printheads of poor quality. Consequently, Graph’s suspicions about the intentions of Jet’s management further increased. Because of these lowered expectations and heightened suspicions, Graph’s management started to push even harder for the milestones that were negotiated in the contract. As one Jet interviewee mentioned, this additional pressure resulted in a negative spiral: When you get put under financial pressure on such a project, you start looking for short cuts… The main questions that are raised at that moment are ‘what is the quickest and what is the lowest cost option’, and not ‘what is the option most likely to succeed’… I do not disagree with financial and time scale pressures. But when you run up against difficulties, they actually create a feedback that makes it worse and worse. This very much happened in this project. (Jet manager) As Graph’s management continued to emphasize the contractual milestones, Jet engineers continued to focus on finding short term solutions to deliver prototype printheads as quickly as possible instead of solving the fundamental technological issues at hand.
22
Termination of the SSH alliance. In December 2000, a new CEO was appointed at Jet to turnaround the decreasing stock quotation of the company. This new CEO started an indepth investigation of the UTPs in the SSH project. At Graph, these changes were seen as a positive development. However, Graph’s confidence in the technological feasibility of the SSH technology remained low. At the beginning of 2001, after benchmarking the potential business scenarios for the SSH technology, Graph officially decided to terminate the SSH alliance. ----- Insert Figure 1 about here ----The End Shooter Head Alliance In May 2000, Graph proposed Jet to jointly initiate a second project in which they would work together to explore the feasibility of Jet’s proprietary ESH technology for a number of printing systems that Graph had in mind. As mentioned before, the SSH alliance between Graph and Jet was facing serious difficulties at that moment. In addition, Graph’s management had become very suspicious of the intentions of Jet’s management. It therefore might seem odd that Graph wanted to initiate another alliance with Jet at that point in time. Graph interviewees provided two explanations in this respect. First, it was stressed that Graph was highly dependent on Jet in order to get access to inkjet technology, which was becoming very important for Graph from a strategic point of view: Researchers at Graph were convinced that inkjet technology would be the future for our Graphics Division. In addition, it was very difficult to get access to other head developers. Jet turned out to be our only option. (Graph manager) Second, it was mentioned that ‘the engineering team that was responsible for the ESH technology was completely different from the engineering team that was responsible for the SSH-related activities’ (Graph engineer). Moreover, Graph interviewees stressed that ‘this ESH engineering team already had successfully commercialized inkjet printheads in the past for other applications and customers.’ Graph consequently was quite confident in the
23
capabilities of this ESH engineering team to turn the ESH alliance into a success. As Jet was in desperate need of additional sources of revenue, Jet also had its interests in initiating this ESH alliance. Negotiation of the ESH contract. The Jet and Graph managers who had been involved in negotiating and monitoring the SSH alliance were also responsible for negotiating the ESH alliance. However, contrary to the negotiations on the SSH alliance, these negotiations were dominated by Graph’s management: At that moment, Jet faced substantial financial problems… Psychologically, this was the optimal moment for us to negotiate these conditions. (Graph manager) Jet was finding it very difficult to sell licenses… We were struggling to get income at all. Our stock was declining rapidly… These facts changed the attitude of our managing director [during the negotiations] enormously. (Jet manager) Because of Jet’s precarious financial situation at that time, Jet accepted the conditions advanced by Graph. On June 30th 2000, the ESH Agreement was signed, signaling the start of the second explorative R&D alliance between Graph and Jet. In the contract, it was stipulated that Graph would fund Jet’s explorative efforts to design and develop ESH prototype printheads. In exchange for its financial contribution, Jet granted Graph a ‘royalty-free, world-wide, nontransferable license to use the ESH printhead manufactured by Jet in any Graph application’ (ESH Agreement: 4). This license would make it possible for GRAPH to use the ESH printheads in their printing systems. The degree of formalization of the ESH contract was similar to the SSH contract. In particular, both contracts arrived at a score of 0.66 on Parkhe’s (1993a) measure of degree of formalization12. However, we also compared the content of both contracts. In this way, we
12
Numerous empirical studies (e.g., Deeds & Hill, 1998; Mellewight et al., 2007; Reuer & Ariño, 2007) have applied Parkhe’s (1993a) measure of degree of formalization. Parkhe (1993a) selected eight provisions through a computer-assisted search of the legal literature. In addition, he arranged these different types of contractual safeguards in increasing order of strength of severity, so a weighting scheme for the stringency of contractual provisions can be adopted to arrive at the global measure of degree of formalization that ranges between 0 and 1.
24
observed that, in the ESH contract, the content of the contractual clauses that referred to partners’ behavior outside the alliance (i.e., right to legal action, proprietary nature of exchanged information, no agency relationship, allocation of intellectual property rights) was identical to the SSH contract. At the same time though, we noticed that the content of the clauses that addressed how the alliance activities should be executed substantially differed from the SSH alliance. Apart from program milestones, target dates, and performance standards formalizing Jet’s expected outputs (i.e., the delivery of printheads), the ESH contract contained contractual statements regulating the activities that Jet was supposed to conduct in order to meet these milestones. For instance, it was stipulated that, during the first six months of the ESH alliance, Jet had to execute the following activities: Allocate resources to the program and establish working practices…Define the detailed printhead specification required by Graph … At the end of this stage the [Jet] development team will produce a recommendation for the printhead design, and plans and detailed financial estimates for the remainder of the development program through to full production. (ESH Agreement Annex 1: Project Description ESH) In this way, Graph could not only monitor the outputs of Jet’s activities but also follow up the activities that Jet had to conduct in order to arrive at these outputs. In addition, both task division and information exchange were contractually defined in different ways. In comparison with the SSH contract, the division of tasks was much more overlapping in the ESH contract. For instance, the contract stipulated that the engineering teams of both partners would work together in order to establish the initial design specifications for the ESH prototype printhead. In addition, it was agreed that, during the collaborative project, Jet and Graph would conduct similar technological tests to evaluate the design of the prototype printheads. In the ESH contract, clauses were also present that obliged partners to exchange information on specific technological activities including joint reviews regarding results of specific technological experiments:
25
Jet will consider possibilities for […] printhead designs and construction methods to meet the Graph requirements. Jet will review this with Graph. (ESH Agreement, Annex 1: Project Description ESH) Interviewees indicated that these contractual changes were proposed by Graph, based on the previous collaborative experiences in the SSH alliance: From the SSH project we had learned that the ESH project should be defined in an alternative way. Therefore, the ESH project was initiated from another perspective. Now, Graph wanted to know more details and wanted to understand why some processes had a low yield. (Graph manager) At Jet, enthusiasm to shift towards this alternative mode of collaboration was quite limited as they were still concerned about potential opportunistic actions by Graph: It was a completely different structure from that of the SSH alliance and it was a structure that was largely dictated by Graph... People here were still concerned about the potential of Graph abusing Jet as a result of the ESH alliance. (Jet manager) However, as one Graph manager expressed, ‘they [=Jet] had no other option than to accept the conditions put forward by Graph as they were standing with their back against the wall.’ Emergence of UTPs at Jet. In July 2000, the engineering teams of Graph and Jet started their ESH-related activities. After a few months of experimenting, Jet engineers came to the conclusion that developing a prototype head that met the predefined target specifications would be much more difficult than expected in advance. For instance, Jet’s engineers ‘faced serious problems in creating ESH prototype printheads that were able to shoot very small drop sizes’ (Jet engineer). However, in comparison with the SSH alliance, Graph engineers were now much better informed about these UTPs. Interviewees referred to the contractual structure of the ESH alliance to explain this intensive information sharing. First, it was stressed that, in the ESH alliance, Jet engineers were contractually obliged to provide information about ongoing technological activities. One Jet interviewee, for instance, mentioned that ‘now we had to tell them what was going on.’ Second, Graph engineers emphasized that Jet’s ongoing efforts
26
were now closely monitored by checking the status of Jet’s contractually defined activities. As UTPs triggered delays in the execution of these activities, additional technical meetings were organized during which Graph engineers asked for explanations: We kept them on a tight leash… When we thought that a specific problem was present, we organized a technical meeting to get additional information. (Graph engineer) Graph engineers not only received information about Jet’s UTPs but also were actively involved in defining the specific nature of the technological problems at hand. The presence of an overlapping task division, allowing Graph engineers to conduct similar tests as the Jet engineers, was identified as a condition that facilitated the emergence of such joint problem-definition processes: There definitely has been overlap in this project… They were testing the heads and we were trying to carry out similar tests to compare the results… The results of these tests were put on the table. Based on these results, decisions were made about how to move on. (Graph engineer) As technological problems were identified, ‘joint brainstorming sessions were initiated to jointly search for solutions regarding issues within the scope of the contract’ (Graph engineer). These joint efforts quickly started to pay off. Adopting suggestions that had emerged during joint brainstorm sessions, Jet engineers were able to successfully address a number of technological issues. Because of the intensive information exchange, joint problem definition and joint problem solving, Graph’s engineering team acquired a fine-grained understanding of the UTPs at Jet and felt actively involved in addressing these issues. One exception was observed however. In conducting tests, Graph had come to the conclusion that the electronic design applied by Jet was unsatisfactory. Discussing this specific technological issue was less straightforward, though. After all, this problem turned out to be directly related to the core of Jet’s ESH technology. In the ESH contract, it was not mentioned that Jet needed to exchange
27
information on the core aspects of its technology. Jet engineers were less willing to involve Graph in solving these issues: We had a number of questions regarding the waveform development of the printhead. Jet however said: ‘This is our know-how and we do not want to give much insight into it.’ (Graph engineer) Both Jet and Graph interviewees also indicated that, as the ESH alliance became operational, the quality of the relationship between the managers of both partners started to change. While Jet’s management had previously suspected Graph of having opportunistic intentions, they now started assessing the intentions of Graph’s management in a more positive way. One Jet manager, for instance, mentioned that ‘we started seeing Graph as an important partner that was committed to turning the relationship into a success.’ In a similar vein, Graph’s management became more convinced in the good intentions of Jet’s management. One Graph manager, for instance, stressed that ‘there was a different approach by the Jet management. … Now, the focus was on the technology [instead of financial interests].’ In the interviews, two reasons were provided to explain this positive relational evolution at the managerial level in the ESH alliance. First, it was stressed that the intensive and fruitful collaboration at the operational level triggered a relational turnaround at the managerial level: Clearly Graph had a significant amount of involvement in this project [=ESH project]. This drastically changed the relationship and the way that it was perceived here [at Jet]. (Jet manager) [At the operational level], meetings were regularly organized in which information was exchanged in an honest, open, and verifiable manner. In this way, you start building trust. (Graph manager) Second, the appointment of a new CEO at Jet in December 2000 was identified as a major event that further accelerated these positive relational dynamics at the managerial level. During 2001, this new CEO hired some new managers and engineers from other companies that were experienced in developing and industrializing products. He also consistently stressed the strategic importance of the collaboration with Graph in his internal
28
and external correspondence. As a result of these efforts, Graph’s expectations about the positive intentions of Jet’s management further increased: The arrival of this new CEO caused an additional jump in the relationship…..He was a huge supporter of the collaboration with Graph. (Graph manager) Emergence of delays. In October 2001, the first ESH prototype heads were built by Jet. However, Graph engineers quickly found out that fundamental problems with the electronics of these prototype printheads had remained unsolved: An initially fully functional chevron-based printhead was delivered… During the days after the meeting, tests revealed severe problems with the printhead electronics which were not experienced during the first hours we [=Graph] used these printheads. Further delivery of ESH was blocked. (Internal Graph Document: Work meeting UIJ Graph/Jet of 9 November 2001) Jet had not been able to internally address the printhead electronics problems. Because of the persistence of this technological issue, the ESH alliance began to face serious delays. However, unlike the SSH alliance, Graph’s management refrained from increasing contractual pressure on Jet: We no longer threatened that, if they would not meet the deadlines, we would end the relationship or no longer reimburse their costs. Instead, we sat together and said to them: ‘We need to find a solution; tell us about your problems.’ (Graph manager) In the ESH alliance, the original time schedule for completing the prototype printheads was adapted. As a result, Jet engineers could continue the search for qualitative solutions to the pending technological issues. At the same time, information exchange from Jet to Graph further increased. Jet became much more open in providing details. I remember meetings where Jet’s project leader projected a table that provided detailed data about the yield of their core processes, what the main problems were, and how they were trying to address them. (Graph manager) Although they were not contractually obliged, Jet’s management instructed its own engineers to start providing detailed information on the core of Jet’s ESH technology.
29
In sum, when delays emerged in the ESH alliance, the contract was applied in a different way. In contrast to the SSH alliance, contractual milestones were adjusted and additional information exchange was initiated. Interviewees referred to the positive relational dynamics at the managerial level as an important explanation in this respect: From a relational point of view, a different kind of connection had grown between us and Jet’s new management. This allowed for a different management approach. (Graph manager) The relationship between our new management and Graph’s management was much more open and honest… Now, we were told [by Jet’s management] to deliver all the information they could use. (Jet engineer) The additional information flows allowed Graph engineers to achieve a better understanding of the remaining technological problems and to make valuable contributions to the solution of the printhead electronics problems: After we had got access to the designs of the electronics, we started digging deeper. In this way, we noticed that some fundamental errors were present. However, these errors were easy to solve when you knew how to solve them. (Graph engineer) The successful solution of additional problems further improved the quality of the relationship at both operational and managerial levels. At the operational level, Jet engineers admitted that ‘Graph delivered substantial added value from a technological point of view’, while Graph engineers expressed that they ‘were impressed by the creativity and perseverance of the [Jet] engineering team.’ At the managerial level, partners’ confidence in the intentions of the other partner further increased. One Graph manager, for instance, referred to the ‘linear progression of trust between Graph and Jet’s management’ in the ESH project. At the same time, the technological successes at the operational level increased the confidence of both Graph and Jet managers in the technological feasibility of the project. According to one Graph interviewee, these relational dynamics made it possible to continue adapting milestones as well as focusing on the technological problems at hand: A feeling of mutual respect emerged. We started seeing them as competent suppliers of technology with whom we could work in a structural manner… The milestones have been adapted a lot of times. However, it was always clear why they were
30
adapted… We tried to settle this matter with them. We did not want to say to them: ‘this is what has been agreed in the contract; find a solution for it.’ (Graph manager) Delivery of functional prototypes. In March 2002, fully functional ESH prototype printheads were delivered by Jet, allowing partners to jointly move to the downstream stages of this technological innovation trajectory. ----- Insert Figure 2 about here ----MULTI-LEVEL PROCESS MODELS Based on the findings above, we developed for each observed alliance transaction a multilevel process model, separating out governance, operational, and managerial levels. These two models are presented in Figure 3. In this section, we systematically discuss the models and link them to our research questions. ----- Insert Figure 3 about here ----Content of Contract and its Impact on Trust Dynamics To answer our first research question on the impact of the content of the contract on trust dynamics, we first present two different kinds of ‘contractual interface structures’, defined as collections of contractual statements that formalize monitoring, task division, and information flows within the alliance transaction. We then discuss how these differences in the nature of contractual formalization have an impact on the quality of joint sensemaking at the operational level, which in turn influences goodwill trust dynamics at the managerial level. Narrow versus broad contractual interface structures. Although the degree of formalization was similar in both the SSH and ESH contract, their nature of contractual formalization showed some remarkable differences. In particular, we observed substantial differences in how issues of monitoring, task division and information exchange were contractually defined. Relying on concepts of organization theorists (e.g., Landau, 1969; Barney & Hesterly, 1996; Van de Ven & Fry, 1980), we summarize the observed differences in Table 3.
31
----- Insert Table 3 about here ----Relying on Barney and Hesterly’s (1996) distinction between performance and behavior monitoring mechanisms, we noticed a first difference regarding the content of both contracts. In the SSH contract, formalization of monitoring was limited to the definition of milestones, target dates and performance standards. In contrast, in the ESH contract, not only performance monitoring mechanisms but also behavior monitoring mechanisms were contractually defined. Contractual statements that specified the technological activities that JET was supposed to execute were illustrations of this type of monitoring. Besides monitoring, the SSH and ESH contracts differed regarding the formalization of task division and information flows. In the SSH contract, collaborating partners were solely responsible for different technological activities. Landau (1969) labeled such task division as ‘mutually exclusive.’ Contractual obligations for information flows or ‘work-related messages sent among members of collaborating partners through different modes of communication’ (Van de Ven & Fry, 1980) also remained absent in this contract. In contrast, in the ESH contract, task division between partners was formalized in a much more ‘overlapping’ (Landau, 1969) manner: collaborating partners shared responsibilities regarding specific technological activities and conducted similar kinds of technological activities in parallel. In addition, the ESH contract contained statements that explicitly obliged partners to exchange information on the results of technological experiments. Based on these observations, we suggest a conceptual distinction between two kinds of contractual interface structures: narrow and broad contractual interface structures. A narrow contractual interface structure is characterized by a mutually exclusive task division, absence of obligations to exchange information, and monitoring mechanisms that are mainly performance-oriented. In contrast, a broad contractual interface structure is characterized by an overlapping task division, the presence of obligations to exchange information, and
32
mechanisms that provide opportunities for not only performance but also behavior monitoring. This conceptual distinction between narrow and broad contractual interface structures provides a much more comprehensive view of the options that managers have in designing the actual content of contracts in interfirm settings. It clearly shows that they can not only vary the contractual definition of juridical clauses that focus on partners’ behavior outside the alliance (Hagedoorn & Hesen, 2007), but also make different choices in the content of contractual clauses that refer to the actual execution of the alliance agreement. Moreover, while Hagedoorn and Hesen (2007) stress that the content of contracts is likely to vary across different partnership forms, we provide evidence that, within one particular partnership form (i.e. contractual alliance), the content of particular contractual clauses can also substantially vary. Impact of contractual interface structures on operational joint sensemaking. As predicted by new product development scholars (e.g., Burgelman & Sayles, 1986; McGrath, 2001), the exploratory nature of technological activities quickly led to unanticipated UTPs in both alliances under study. Our data suggest that the nature of contractual formalization substantially influences the quality of joint sensemaking (i.e., joint problem definition and joint problem solving) on these UTPs at the operational level. In the SSH alliance, the presence of a narrow contractual interface structure severely limited information exchange on unanticipated technological problems between engineers of different partners, which resulted in what Allen (1977) calls an impoverished problem definition and solution space. In contrast, in the ESH alliance, the contractual obligation to exchange information and the presence of behavior monitoring systems stimulated discussions about emerging technological problems. In addition, the presence of an overlapping task division enhanced opportunities for joint problem definition and joint problem solving. In other words, the
33
presence of a broad contractual interface structure in the ESH alliance positively influenced the quality and amount of joint sensemaking on unanticipated technological problems between engineers. This latter finding is in line with Landau’s (1969) argument that, in organizational settings, overlap contributes to both problem definition and problem solving activities. Whereas our study indicates that the outcome of the contractual negotiation process (i.e., the negotiated contractual interface structure) substantially influences the intensity and quality of joint sensemaking on the operational level, Vlaar, Van den Bosch and Volberda (2006) recently argued that the contracting process itself functions as a sensemaking device on the managerial level. In both the SSH and ESH alliance, however, we did not find instances of intensive joint sensemaking between managers during the contractual negotiation process. Instead, we observed that, based on their dominant bargaining position, managers of one partner imposed a contract on the other partner with limited room for focusing partners’ attention, provoking articulation, and instigating interactions. Impact of operational joint sensemaking on managerial trust dynamics. Our data also indicate that the quality of joint sensemaking at the operational level influences goodwill trust dynamics at the managerial level. As joint sensemaking on unanticipated technological problems remained absent in the SSH alliance, Graph’s management started questioning Jet’s good intentions. In contrast, in the ESH alliance, the presence of joint problem definition and joint problem solving at the operational level contributed to a relational turnaround where managers of both partners started assessing the intentions of the other partner in much more positive terms. In other words, extensive joint sensemaking at the operational level contributed to positive goodwill trust dynamics at the managerial level. These findings are in line with the conceptual argument of Janowicz, Krishnan and Noorderhaven (2005) that
34
relational processes (i.e., joint sensemaking) at the operational level substantially influence relational processes (i.e., goodwill trust dynamics) at the managerial level. To sum up, we identified particular connections between the nature of contractual formalization, the quality of joint sensemaking at the operational level and goodwill trust dynamics at the managerial level. In particular, we propose that: Proposition 1: In explorative R&D alliances, a broad (narrow) contractual interface structure facilitates (hampers) joint sensemaking on unanticipated technological problems at the operational level, which in turn positively (negatively) influences goodwill trust dynamics at the managerial level. We need to remark that the extensive joint sensemaking at the operational level was not the only factor that influenced goodwill trust dynamics in the ESH alliance. We observed that the appointment of a new CEO at Jet also contributed to the emergence of positive goodwill trust dynamics. This latter finding sheds a new light on the impact of mobility of key persons in interfirm relationships. While previous research (de Rond & Bouchikhi, 2004; Doz, 1988; Ring & Van de Ven, 1994) has mainly focused on the negative consequences (i.e., destruction of informal commitments; disappearance of important tacit knowledge) of replacing key individuals, we observed that the introduction of new people at the managerial level can also have a positive impact, contributing to a positive turnaround in trust dynamics. Co-evolution of Contract Application and Trust Dynamics To address our second research question, we present how trust dynamics and contract application co-evolve over time. We first discuss how managerial goodwill trust dynamics influence the mode of contract application and then argue that different modes of contract application trigger different trust dynamics at both operational and managerial levels, resulting in negative or positive reinforcing cycles. Impact of trust dynamics on mode of contract application. In line with previous research (e.g., Ariño & de la Torre, 1998; Doz, 1996; Madhok, 1995a), our findings showed that contracts can be applied in different ways during alliances. In the SSH alliance, a rigid
35
mode of contract application was observed. When unanticipated technological problems emerged in the SSH alliance, Graph’s management was not willing to adjust the milestones and target dates. Rather, contractual pressure was increased by emphasizing the existing contractual milestones. At the same time, Jet’s management turned out to be unwilling to adjust the contractually agreed task division and to allow for information flows. In contrast, in the ESH alliance, a much more flexible contract application approach was applied when unanticipated technological problems emerged. In this case, Graph’s management was willing to adjust the existing milestones and target dates, while Jet’s management was willing to further increase information flows and overlap in task activities. While previous research remains silent on why contracts are applied in different ways, we identified one particular condition that seems to influence the mode of contract application: the nature of goodwill trust dynamics at the managerial level. In the SSH alliance, Graph’s management started questioning the good intentions of Jet’s management. At the same time, Jet’s management concerns about potential opportunistic behavior by Graph’s management further increased. Interviewees explicitly referred to these negative goodwill trust dynamics at the managerial level to explain the rigid mode of contract application. In the ESH alliance, much more positive goodwill trust dynamics were observed at the managerial level. According to the interviewees, these positive goodwill trust dynamics triggered a much more flexible mode of contract application when unanticipated technological problems emerged. We therefore propose that: Proposition 2: In explorative R&D alliances, positive (negative) goodwill trust dynamics at the managerial level increase the probability of flexible (rigid) contract application. At the same time, however, we did not find that the emergence of positive goodwill trust dynamics reduced the importance of contracts as governance mechanisms. While previous studies (e.g., Das & Teng, 1998; Dyer & Singh, 1998; Larson, 1992) suggest that, as
36
positive trust dynamics emerge in alliances, the formal contract is pushed to the background and norms of fairness, honesty, and reciprocity take on a powerful role in governing the alliance, our data showed that the contract remained an important safeguarding and coordination device, even after positive trust dynamics emerged in the ESH alliance. For instance, one Graph manager explicitly stressed that ‘even when it starts becoming very pleasant, you need to oblige yourself to continue cross-checking whether the partner’s activities are in line with the contractual arrangements.’ At the same time, this manager admitted that ‘we were now [= in the ESH alliance] much more flexible in how these contractual targets were met.’ In other words, our data suggest that positive goodwill trust dynamics do not reduce the importance of contracts in governing alliances, but rather allow for a shift from a rigid to a more flexible mode of contract application. Negative or positive reinforcing cycles. Previous studies (Ghoshal & Moran, 1996; Sitkin & Stickel, 1996) point to the risk of negative reinforcing cycles between contract application and trust dynamics. Our data on the SSH alliance seem to confirm this reasoning. In this case, the rigid application of the contract triggered ‘contract compliance behavior’ (Macaulay, 1963) at Jet (i.e., searching for technological short-cuts) and further reduced the opportunities for joint sensemaking at the operational level, which in turn increased the suspicions of Graph’s management about the intentions of the other partner. In addition, as Jet engineers started to deliver printheads of inferior quality, Graph engineers started to question the technological competencies of the Jet engineers. These negative competence trust dynamics at the operational level lowered the expectations of Graph’s management on the feasibility of the project. The further decrease of goodwill trust and the lowered expectations of the feasibility of the project motivated Graph’s management to further increase contractual pressure.
37
However, our data on the ESH alliance suggest that interactions between contract application and relational dynamics can also result in positive reinforcing cycles. In this case, the flexible application of the contractual interface structure allowed Jet engineers to focus on the search for qualitative solutions to pending technological problems and resulted in an increased sharing of sensitive technological information with Graph engineers. As a consequence, goodwill trust dynamics at the managerial level continued to be positive. In addition, the positive collaborative experiences at the operational level triggered positive competence trust dynamics at this particular level, leading to heightened expectations of the project's feasibility at the managerial level. The further increase in goodwill trust and the heightened expectations stimulated managers to continue applying the contractual interface structure in a flexible way. In sum, while previous studies emphasize the likely occurrence of negative reinforcing cycles between contract application and trust dynamics, our study suggests that application of contracts can trigger both negative and positive reinforcing cycles, depending on how contracts are applied. In particular, we argue that: Proposition 3: In explorative R&D alliances, a rigid (flexible) application of the contract is likely to trigger negative (positive) trust dynamics at both operational and managerial levels, which in turn leads to increasing rigidity (flexibility) regarding contract application. TOWARDS AN INTEGRATIVE PERSPECTIVE ON ALLIANCE GOVERNANCE In this section, we discuss the main theoretical implications of our findings. Whereas previous alliance governance research tends to focus on the structural design of single transactions or relational dynamics within interfirm relationships, we present a more integrative perspective on alliance governance. In particular, we provide in-depth theoretical insights into how the design and application of structural elements is connected to relational processes, both within and between transactions. Connections between Structural Design and Relational Dynamics within Transactions
38
The multi-level process models that we developed above suggest that, within alliance transactions, the initial contract design substantially influences relational dynamics at operational and managerial levels, which in turn determine how contracts are applied. The theoretical implications of these findings are twofold. First, we provide a process-oriented view on the contract-trust relationship. Second, we propose an alternative perspective on the role of goodwill trust in governing alliance transactions. Our study adds an additional and richer voice to the current debate on the relationship between contracts and trust and the mixed evidence to date (e.g., Luo, 2002; Lyons & Mehta, 1997; Mulharta & Murnighan, 2002; Poppo & Zenger, 2002) by providing a more processoriented view. We observed that contracts with a similar degree of formalization but a different nature of formalization triggered different kinds of joint sensemaking dynamics at the operational level, which in turn generated different kinds of goodwill trust dynamics at the managerial level. In the SSH alliance, the narrow contractual governance structure hampered joint sensemaking on unanticipated technological problems which are an inherent aspect of explorative activities. This misalignment between the nature of contractual formalization and the specific nature of explorative activities triggered negative trust dynamics at the managerial level. In contrast, the implementation of a broad contractual governance structure in the ESH alliance seemed to fit much better with the need for intensive joint sensemaking on unexpected technological problems in explorative innovation settings. This positive alignment between the nature of contractual formalization and the nature of alliance activities contributed to the emergence of much more positive goodwill trust dynamics at the managerial level in this case. We therefore argue that contracts with a similar degree of contractual formalization can trigger both positive and negative trust dynamics, depending on the extent to which the nature of contractual formalization reflects the nature of the alliance activities. In addition, while previous alliance governance studies
39
tend to treat the operational level as a black box (Currall & Inkpen, 2000), we point to relational processes at the operational level (i.e., joint sensemaking) as important intermediary processes between contract design at the governance level and goodwill trust dynamics at the managerial level. Adherents to the relational perspective (e.g., Dyer & Singh, 1998; Gulati, 1995; Larson, 1992; Parkhe, 1993a; Uzzi, 1997) typically argue that, as high levels of goodwill trust are present, the contract is de-emphasized and partners switch from contractual governance to relational governance. This line of reasoning envisions goodwill trust as an alternative governance mechanism for complex contracts. Our data provide a different perspective on the role of goodwill trust in governing alliance transactions. First, we did not find evidence that positive goodwill trust dynamics substantially reduced the importance of contracts. Further, we found strong indications that positive goodwill trust dynamics allowed for applying a flexible approach towards contract application. And finally, our data suggest that, in case of negative goodwill trust dynamics, a rigid mode of contract application is likely to emerge. Based on these findings, we do not refer to goodwill trust as an alternative governance mechanism for contracts, but rather conceptualize goodwill trust as a condition that determines how contracts are applied as governance mechanisms. Such conceptualization of trust is in accordance with the argument of Rousseau and colleagues (1998: 395) that ‘trust is not a behavior or a choice, but an underlying psychological condition that can cause or result from such [governance] actions.’ Connections between Relational Dynamics and Structural Design between Transactions Our findings also generate additional theoretical insights into how the relational history in previous alliance transactions influences the structural design in subsequent transactions between the same pair of firms. Examining two separate alliance transactions that were sequentially embedded within the same interfirm relationship, we suggest two relational
40
processes explaining partners’ need and ability to change the contractual interface structures for the governance of subsequent alliances as well as question the role of goodwill trust as a necessary condition for subsequent transactions. In line with Mayer and Argyres (2004), we found evidence that learning experiences in previous transactions trigger a need to change the content of the contract in subsequent transactions. The negative experiences with the narrow contractual interface structure in the SSH alliance was for Graph a motivation to negotiate a broader contractual interface structure in the second explorative R&D alliance. However, while Mayer and Argyres (2004) provide an example where both alliance partners experienced the need to change the content of the contract, our data indicated that learning experiences may differ between partners. While Graph managers perceived a high need to shift to another contractual interface structure, Jet managers were not eager to do so. Further, we observed that Graph was able to effectuate the shift in contractual interface structure because, during the previous transaction, it gradually had gained the necessary bargaining power to do so. This latter observation is consistent with the argument of bargaining power theorists (e.g., Makhija & Ganesh, 1997; Yan & Gray, 1994) that the partner with a greater distribution of power is able to affect the structural design more than the partner with less power. In sum, we agree with Mayer and Argyres (2004) that, because of learning experiences in previous transactions, a collaborating partner may feel the need to change the nature of the contractual interface structures in subsequent transactions. However, we argue that, in case of asymmetric learning experiences (i.e., only one partner perceives a need to change the content of the contract), a partner will primarily be able to effectuate such a shift if it has the necessary relative bargaining power to do so. Focusing on transactional attributes to explain structural design changes, adherents to the structural alliance governance perspective might suggest an alternative theoretical
41
explanation for the observed shift from narrow to broad contractual interface structures. As interviewees referred to the SSH technology as revolutionary and to the ESH technology as an extension of Jet’s existing commercial printhead technologies, the level of technological uncertainty in the ESH alliance can considered to be lower than in the SSH alliance. Such lower levels of technological uncertainty would, according to the structural perspective, decrease the risk and/or consequences of opportunistic action, which in turn reduces the need for contractual protection in interfirm relationships (e.g., Blumberg, 2001; Poppo & Zenger, 2002). Following this reasoning, one could argue that it was the presence of a lower degree of technological uncertainty in the ESH alliance that reduced the perceived risk of opportunism at the start of the ESH alliance, which in turn increased the willingness to negotiate a broader contractual interface structure. However, our interview data do not confirm this alternative explanation. As already mentioned, Jet interviewees stressed that, at the start of the ESH alliance, concerns about the potential of Graph to abuse Jet as a result of their collaboration were still substantial. In other words, we did not find indications of reduced levels of opportunism concerns at the start of the ESH alliance despite the decreased level of technological uncertainty. Finally, our findings question the role of goodwill trust as a necessary condition to continue interfirm relationships over different alliance transactions. Some conceptual studies (e.g., Madhok, 1995b; Ring, 1997) proposed that goodwill trust functions as the necessary social glue that keeps partners together. They argued that the emergence of positive goodwill trust dynamics in previous transactions induces partners to continue and even expand their relationship through negotiating new transactions. Implicitly, these scholars seem to assume that negative goodwill trust dynamics limit the incidence of future ties. Previous case study research that provides instances of unsuccessful alliances (e.g., Ariño & de la Torre, 1998; Doz, 1996) seems to affirm this argument. These studies describe how negative goodwill
42
trust dynamics in the alliance contributed to the dissolution of the relationship. However, our case study provides an example where, despite the negative goodwill trust dynamics in a previous transaction, managers of both partners negotiated a new contractual transaction. Both Jet and Graph interviewees explained this decision by referring to their respective financial and technological dependence on each other. Whereas Jet desperately needed Graph’s financial support, Graph needed Jet to gain access to inkjet technologies. A second condition to explain their willingness to negotiate a new alliance was the presence of positive expectations on the competencies of the new Jet engineering team. Based on the prior track record of Jet’s ESH team, Graph was convinced that, in contrast to Jet’s SSH team, this team was competent to turn this project into a success. Based on these findings, we suggest that mutual interdependence – the core concept of resource dependence theory (e.g., Pfeffer & Salancik, 1978) – and competence trust might be more important conditions for the continuation of interfirm relationships over multiple transactions than goodwill trust. CONCLUSION Our study contributes to an integrative perspective on alliance governance by examining the complex and dynamic connections between structural and relational aspects in governing alliances. We clearly show that structural and relational aspects are inherently linked and mutually influence each other, both within and between transactions. Based on our findings, we (a) provide a process-oriented view on the relationship between contracts and trust (b) conceptualize goodwill trust as a condition that determines how contracts are applied (c) define the contracting process as an incremental learning process that is sensitive to changes in relative bargaining power, and (d) point to mutual interdependence and competence trust as crucial conditions for subsequent transactions. Several avenues for future research can be identified. First, we stress the need for additional research on how alliance contracts are negotiated. We found strong indications
43
that, when managers are able to achieve a good fit between the content of the contract and the nature of the alliance activities, the contract is likely to function as a catalyst of positive relational processes on operational and managerial levels. These findings point to the negotiation of the contract as a vital process in alliances. However, while researchers have paid a lot of attention to the antecedents and outcomes of contracts, the actual process of negotiating alliance contracts has received limited attention (Argyres & Mayer, 2007; Ring, 1997). Second, the relationship between variations in contract design and relational processes may be extended to variations in juridical clauses. In our study, contract design differed in the way monitoring, task division, and information flows were contractually defined but was similar regarding the juridical clauses regulating the behavior of partners outside the alliance. As these clauses can also differ (Hagedoorn & Hesen, 2007), future research examining how changes in the content of juridical clauses influence relational processes might be very fruitful. Third, while previous work on trust tends to focus on the individual or the organizational level, our distinction between operational and managerial levels allowed analyzing the evolution of trust on the meso-level (i.e., group level). In this way, we found first indications that competence and goodwill trust dynamics may operate at different levels, but, at the same time, mutually influence each other. These observations point to the need for additional research on the complex linkages between competence and goodwill trust across different levels. Fourth, we point to the need for further research on the revitalization of interfirm relationships. Previous alliance research (e.g., Ariño & de la Torre, 1998; Das & Teng, 2000) tends to conceptualize interfirm relationships as rather unstable phenomena, especially when negative trust dynamics are present. Our findings, however, indicate that, after negative trust dynamics have emerged, partners may still decide to continue their relationship and may even manage to achieve a significant turnaround in trust dynamics. Moreover, while previous research (de Rond & Bouchikhi, 2004; Doz, 1988; Ring & Van de
44
Ven, 1994) emphasizes the negative consequences of mobility of key persons during collaborative projects, we found first indications that replacing key individuals and/or teams can stimulate such revitalization processes. For instance, we observed how the creation of discontinuity at the operational level (i.e., switching between engineering teams at Jet) allowed for continuity at the level of the interfirm relationship. In addition, we found strong indications that the management changes at Jet hugely contributed to the positive trust dynamics in the ESH alliance. At the same time, we acknowledge that additional research is needed in this respect. For instance, while we collected data in a retrospective way, it might be interesting to conduct real-time case study research to attain more fine-grained insights into how the mobility of key individuals at operational and managerial levels influences the collaborative dynamics in alliances. As a final reflection, we point to the main limitation of this study. Our findings are based on an in-depth examination of two sequential explorative R&D alliances between the same pair of partners. Although this research design allowed us to compare these two alliances with a minimum influence of extraneous variation, the findings are contextualized. For instance, our data suggest that a broad contractual interface structure is more effective than a narrow contractual interface structure in explorative R&D alliances. However, this does not imply that narrow interface structures cannot be useful in other collaborative settings. For instance, Gerwin and Ferris (2004) already provided indications that, in exploitative settings, where partners focus on standardization and fine-tuning and face higher time-to-market pressures, a narrow interface structure might be more cost and time efficient than a broad interface structure. In sum, this study provided a more integrative perspective on alliance governance. We hope that our insights may help managers in improving the success rate of their alliances. At the same time, we encourage scholars to further explore the interactions between structural
45
and relational aspects within and between alliance transactions in a variety of organizational settings. ACKNOWLEDGMENTS Useful comments and suggestions on this paper have been provided by associate editor Chet Miller; three anonymous reviewers; and Africa Ariño, Arvind Parkhe, Henri Dekker, Paul Vlaar and Brian den Ouden. The authors gratefully acknowledge the financial support of the Fund for Scientific Research – Flanders (FWO G.0286.06) and the Flemish Government (Policy Research Center ‘Entrepreneurship and International Entrepreneurship’) for conducting this research. Earlier versions of this paper have been presented at the Academy of Management 2006 Meeting, Atlanta, August 11-16 and the 2006 CSA Workshop, Nijmegen, October 26-27.
46
REFERENCES Allen T.J. 1977. Managing the flow of technology. MIT Press. Anand, B. & Khanna, T. 2000. Do firms learn to create value? The case of alliances. Strategic Management Journal, 21: 295-315. Argyres, N. & Mayer, K.J. 2007. Contract design as a firm capability: An integration of learning and transaction cost perspectives. Academy of Management Review, 32: 10601077. Ariño, A. & de la Torre, J. 1998. Learning from failure: towards an evolutionary model of collaborative ventures. Organization Science, 9: 306-325. Ariño, A. & Reuer, J.J. 2005. Alliance contractual design. In O. Shenkar, & J.J. Reuer (Eds.) Handbook of Strategic Alliances, 149-167. Sage. Barney, J.B. & Hesterly, W. 1996. Organizational economics: Understanding the relationship between organizations and economic analysis. In S. Clegg, C. Hardy, & W.R. Nord (Eds.) Handbook of Organization Studies, 115-147. London and Thousand Oaks: Sage Publications. Bell, J., den Ouden, B. & Ziggers, G. W. 2006. Dynamics of cooperation: At the brink of irrelevance. Journal of Management Studies, 43: 1607-1619. Blau, P. 1964. Exchange and power in social life. John Wiley & Sons, New York Bleeke, J. & Ernst, D. 1991. The way to win in cross-border alliances. Harvard Business Review, November-December: 127-135. Blumberg, B. 2001. Cooperation contracts between embedded firms. Organization Studies, 22: 825-852.
47
Burgelman, R.A. & Sayles, L.R. 1986. Inside corporate innovation. New York: The Free Press. Cardinal, L. B., Sitkin, S. B. & Long, C. P. 2004. Balancing and rebalancing in the creation and evolution of organizational control. Organization Science, 15: 411-431. Carson, S.J., Madhok, A., Varman, R. & John, G. 2003. Information processing moderators of the effectiveness of trust-based governance in interfirm R&D collaboraitons. Organizations Sccience, 14: 45-56. Contractor, F. J. 2005. Alliance structure and process: Will the two research steams ever meet in alliance research? European Management Reveiw, 2: 123-129. Currall, S. C. & Inkpen, A.C. 2000. Joint venture trust: interpersonal, inter-group, and interfirm levels. In D. O. Faulkner, & M. De Rond (Eds.) Cooperative strategy: economic, business, and organizational issues. Oxford: University Press. Das, T. K. & Teng, B-S. 1998. Between trust and control: Developing confidence in partner cooperation in alliances. Academy of Management Review, 23: 491-512. Das, T.K. & Teng, B-S. 2000. Instabilities of strategic alliances: An internal tensions perspective. Organization Science, 11: 77-101. Das, T. K. & Teng, B-S. 2001. Trust, control, and risk in strategic alliances: an integrated framework. Organization Studies, 22: 251-283. de Rond, M. & Bouchikhi, H. 2004. On the dialectics of strategic alliances. Organization Science, 15: 56-69. Deeds, D. L. & Hill, C.W. 1998. An examination of opportunistic action within research alliances: evidence from the biotechnology industry. Journal of Business Venturing, 14: 141-163.
48
Doz, Y. L. 1988. Technology partnerships between larger and smaller firms: some critical issues. In F. J. Contractor, & P. Lorange (Eds.) Cooperative strategies in international business. New York: Lexington Books. Doz, Y. L. 1996. The evolution of cooperation in strategic alliances: initial conditions or learning processes? Strategic Management Journal, 17: 55-83. Doz, Y. & Hamel, G. 1998. Alliance advantage: The art of creating value through partnering. Boston: Harvard Business School Press. Dyer, J. H. & Singh, H. 1998. The relational view: cooperative strategy and sources of interorganizational competitive advantage. Academy of Management Review, 23: 660679. Eisenhardt, K. M. 1989. Building theories from case study research. Academy of Management Review, 14: 532-550. Eisenhardt, K. M. & Graebner, M. E. 2007. Theory building from cases: Opportunities and challenges. Academy of Management Journal, 50: 25-32. Faems, D., Janssens, M., Bouwen, R. & Van Looy, B. 2006. Governing explorative R&D alliances: Searching for effective strategies. Management Revue, 17: 9-29. Faems, D., Van Looy, B. & Debackere, K. 2005. Interorganizational collaboration and innovation: toward a portfolio approach. Journal of Product Innovation Management, 22: 238-250. Gerwin, D. & Ferris, J.S. 2004. Organizing new product development projects in strategic alliances. Organization Science, 15: 22-37. Geyskens, I., Steenkamp, J-B. E. M. & Kumar, N. 2006. Make, buy, or ally: A transaction cost theory meta-analysis. Academy of Management Journal, 49: 519-543.
49
Ghoshal, S. & Moran, P. 1996. Bad for practice: A critique of the transaction cost theory. Academy of Management Review, 21: 13-47. Gioia, D. A. & Chittipeddi, K. 1991. Sensemaking and sensegiving in strategic change initiation. Strategic Management Journal, 12: 433-448. Gioia, D. A., Thomas, J. B., Clark, S. M. & Chittipeddi, K. 1994. Symblism and strategic change in academia: The dynamics of sensemaking and influence. Organization Science, 5: 363-383. Gulati, R. 1995. Does familiarity breed trust? The implications of repeated ties for contractual choice in alliances. Academy of Management Review, 38: 85-112. Granovetter, M. 1985. Economic action and social structure: the problem of embeddedness. American Journal of Sociology, 91: 481-510. Hagedoorn, J. 2002. Inter-firm R&D partnerships: an overview of major trends and patterns since 1960. Research Policy, 31: 477-492. Hagedoorn, J. & Hesen, G. 2007. Contract law and the governance of inter-firm technology partnerships: An analysis of different modes of partnering and their contractual implications. Journal of Management Studies, 44: 342-366. Hall, P. A. 2006. Systematic process analysis: When and how to use it. European Management Reveiw, 3: 24-31. Hennart, J-F. 1991. The transaction costs theory of joint ventures. Management Science, 37: 483-497. Hennart, J-F. 2006. Alliance research: Less is more. Journal of Management Studies, 43: 1621-1628.
50
Janowicz, M., Krishnan, R. & Noorderhaven, N.G. 2005. Trust repair: Taking it
to
the
next
level.
Paper
presented
at
the
21st
European
Group
for
Organization Studies (EGOS) colloquium 2005, Berlin, Germany. Jeffries, F.L. & Reed, R, 2000. Trust and adaptation in relational contracting. Academy of Management Review, 25: 873-882. Jones, G. R. & George, J.M. 1998. The experience and evolution of trust: implications for cooperation and teamwork. Academy of Management Review, 23: 531-546. Klein Woolthuis, R., Hillebrand, B. & Nooteboom, B. 2005. Trust, contract and relationship development. Organization Studies, 26: 813-840. Koza, M. P. & Lewin, A.Y. 1998. The co-evolution of strategic alliances. Organization Science, 9: 255-264. Kvale, S. 1996. InterViews: An introduction to qualitative research interviewing. London: Sage Publications. Landau, M. 1969. Redundancy, rationality, and the problem of duplication and overlap. Public Administration Review, 29: 346-358. Langley, A. 1999. Strategies for theorizing from process data. Academy of Management Review, 24: 691-710. Larson, A. 1992. Network dyads in entrepreneurial settings: a study of the governance of exchange relationships. Administrative Science Quarterly, 37: 76-104. Leonard-Barton, D. 1990. A dual methodology for case studies: synergistic use of a longitudinal single site with replicated multiple studies. Organization Science, 1: 248266.
51
Lewicki, R.J.; Tomlinson, E.C. & Gillespie, N. 2006. Models of interpersonal trust development: Theoretical approaches, empirical evidence, and future directions. Journal of Management, 32: 991-1022. Luo, Y. 2002. Contract, cooperation, and performance in international joint ventures, Strategic Management Journal, 23: 903-919. Lyons, B. & Mehta, J. 1997. Contracts, opportunism and trust: Self-interest and social orientation. Cambridge Journal of Economics, 21: 239-257. Macaulay, S. 1963. Non-contractual relations in business: a preliminary study. American Sociological Review, 28, 55-67. Madhok, A. 1995a. Opportunism and trust in joint venture relationships: an exploratory study and a model. Scandinavian Journal of Management, 11: 55-74. Madhok, A. 1995b. Revisiting multinational firms’ tolerance for joint ventures: A trust-based approach. Journal of International Business Studies, 26: 117-137. Madhok, A. & Tallman, S.B. 1998. Resources, transactions and rents: managing value through interfirm collaborative relationships. Organization Science, 9: 326-339. Makhija, M. V. & Ganesh, U. 1997. The relationship between control and partner learning in learning-related joint ventures. Organization Science, 8: 508-527. Malhorta, D. & Murnighan, J.K. 2002. The effects of contracts on interpersonal trust. Administrative Science Quarterly, 47: 534-559. Mayer, K. J. & Argyres, N.S. 2004. Learning to contract: evidence from the personal computer industry. Organization Science, 15: 394-410. McGrath, R. G. 2001. Exploratory learning, innovative capacity, and managerial oversight, Academy of Management Journal, 44: 18-131.
52
Mellewigt, T., Madhok, A. & Weibel, A. 2007. Trust and formal contracts in interorganizational relationships – substitutes and complements. Managerial and Decision Economics. Miller, C. C.; Cardinal, L. B., and Glick, W. H. 1997. Retrospective reports in organizational research: A reexamination of recent evidence. Academy of Management Journal, 40: 189-204. Mintzberg, H. 1979. The structuring of organizations: a synthesis of the research. N.J.: Prentice-Hall Englewood Cliffs. Nooteboom, B. 1996. Trust, opportunism and governance: A process and control model. Organization Studies, 17: 985-1010. Osborn, R.N. & Hagedoorn, J. 1997. The institutionalization and evolutionary dynamics of interorganizational alliances and networks. Academy of Management Journal, 40: 261278. Oxley, J. E. 1997. Appropriability hazards and governance in strategic alliances: a transaction cost approach. Journal of Law, Economics, and Organization, 13: 387-409. Parkhe, A. 1993a. Strategic alliances structuring: a game theoretic and transaction cost examination of interfirm cooperation. Academy of Management Journal, 36: 794-829. Parkhe, A. 1993b. "Messy" research, methodological predispositions, and theory development in international joint ventures. Academy of Management Review, 18: 227268. Pentland, B. T. 1999. Building process theory with narrative: From description to explanation, Academy of Management Review, 24: 711-724. Pettigrew, A. M. 1990. Longitudinal field research on change: theory and practice, Organization Science, 1: 267-292. 53
Pfeffer, J. & Salancik, G.R. 1978. The external control of organizations. New York: Harper Row. Pisano, G. P. 1990. The R&D boundaries of the firm: an empirical analysis. Administrative Science Quarterly, 35: 153-176. Pisano, G.P., Russo, M.V. & Teece, D. 1988. Joint ventures and collaborative agreements in the telecommunications equipment industry. In D. Mowery (Ed.) International collaborative ventures in U.S. manufacturing: 23-70. Cambridge: Ballinger. Plowman, D. A., Baker, L. T., Beck, T. E., Kulkarni, M., Solansky, S. T. & Travis, D. V. 2007. Radical change accidentally: The emergence and amplification of small change. Academy of Management Journal, 50: 515-543. Poole, M.S., Van De Ven, A.H., Dooley, K. & Holmes, M.E. 2000. Organizational change and innovation processes: theory and methods for research. Oxford: Oxford University Press. Poppo, L. & Zenger, T. 2002. Do formal contracts and relational governance function as substitutes or complements? Strategic Management Journal, 23, 707-725. Powell, W. W. 1998. Learning from collaboration: Knowledge and networks in the biotechnology and pharmaceutical industries. California Management Review, 40: 228240. Reuer, J.J. & Ariño, A. 2007. Strategic alliance contracts: Dimensions and determinants of contractual complexity. Strategic Management Journal, 28: 313-330. Reuer, J.J.; Ariño, A ; Mellewigt, T. 2006. Entrepreneurial alliances as contractual forms. Journal of Business Venturing, 21: 306-325.
54
Ring, P.S. 1997. Processes facilitating reliance on trust in inter-organizational networks. In M. Ebers (Ed.) The formation of inter-organizational networks. Oxford: Oxford University Press. Ring, P. S. & Van de Ven, A.H. 1992. Structuring cooperative relationships between organizations. Strategic Management Journal, 13, 483-498. Ring, P. S. & Van De Ven, A.H. 1994. Developmental processes of cooperative interorganizational relationships. Academy of Management Review, 19: 90-118. Rousseau, D. M., Sitkin, S.B., Burt, R.S. & Camerer, C. 1998. Not so different after all: a cross-discipline view of trust. Academy of Management Review, 23: 393-404. Salk, J. E. 2005. Often called for but rarely chosen: Aliance research that directly studies process. European Management Reveiw, 2: 117-122. Sampson, R. C. 2004. Organizational choice in R&D alliances: knowledge-based and transaction cost perspectives. Managerial and Decision Economics, 25, 421-436. Sitkin, S.B. & Stickel, D. 1996. The road to hell: The dynamics of distrust in an era of quliaty. In R.M. Kramer & T.R. Tyler (Eds.) Trust in organizations: Frontiers on theory and research: 196 – 215. Thousand Oaks: Sage. Uzzi, B. 1997. Social structure and competition in interfirm networks: the paradox of embeddedness. Administrative Science Quarterly, 42: 35-67. Van de Ven, A.H. & Fry, D.L. 1980. Measuring and assessing organizations. New York: Whiley. Van Maanen, J. 1979. The fact of fiction in organizational ethnoGraphy. Administrative Science Quarterly, 24: 539-550.
55
Vlaar, P. W. L., Van den Bosch, F. A. J. & Volberda, H. W. 2006. Coping with problems of understanding in interorganizational relationships: Using formalization as a means to make sense. Organization Studies, 27: 1617-1638. Vlaar, P.W.L., Van den Bosch, F.A.J. & Volberda, H.W. 2007. On the evolution of trust, distrust, and formal coordination and control in interorganizational relationships: Toward an integrative framework. Group & Organization Managemnet, 32: 407-428. Williamson, O.E. 1985. The economic institutions of capitalism. New York: The Free Press. Williamson, O. E. 1991. Comparative economic organization: the analysis of discrete structural alternatives. Administrative Science Quarterly, 36: 269-296. Yan, A. & Gray, B. 1994. Bargaining power, management control, and performance in United States-China Joint Ventures: a comparative study. Academy of Management Journal, 37: 478-1517. Yin, R. K. 1984. Case study research: designs and methods. London: Sage. Zaheer, A., Lofstrom, S. & George, V.P. 2002. Interpersonal and interorganizational trust in alliances. In F. J. Contractor, & P. Lorange (Eds.) Cooperative strategies and alliances: 347-380. Amsterdam: Pergamon. Zaheer, A. & Venkatraman, N. 1995. Relational governance as an interorganizational strategy: An empirical test of the role of trust in economic exchange. Strategic Management Journal, 16: 373-392. Zajac, E. J. & Olsen, C.P. 1993. From transaction cost to transactional value analysis: implications for the study of interorganizational strategies. Journal of Management Studies, 30: 131-145.
56
Table 1: Structural and Relational Perspectives on Alliance Governance
Structural Perspective
Relational Perspective
Focus of analysis
Single transaction
Interfirm relationship
Theoretical basis
Transaction cost theory
Social exchange theory
Main assumptions
Partners have a tendency to act opportunistically
Partners have a tendency to act in a trustworthy fashion
Alliance performance is driven by the quality of the initial structural design
Alliance performance is driven by the quality of the ongoing relational processes
Proposed governance mechanism
Complex contracts
Trust
Criticism
Undersocialized view on human action
Oversocialized view on human action
Reference publications
Pisano, Russo & Teece (1988); Pisano (1990); Williamson (1991); Hennart (1991, 2006); Parkhe (1993a); Oxley (1997); Sampson (2004);
Larson (1992); Ring & Van de Ven (1992); Zaheer & Venkatraman (1995), Gulati (1995); Uzzi (1997); Dyer & Singh (1998); Salk (2005)
57
Table 2: Main characteristics of SSH and ESH alliance
SSH Alliance (March ’99 – April ’01)
ESH Alliance (July ’00 – March ’02)
Ownership structure
Non-equity
Non-equity
Purpose of alliance
Exploration of the feasibility of Side Shooter Head (SSH) technology for printing applications
Exploration of the feasibility of End Shooter Head (ESH) technology for printing applications
Partners’ main responsibilities
Jet: Development of prototype SSH printheads Graph: Development of prototype SSH printing system
Jet: Development of prototype ESH printheads Graph: Development of prototype ESH printing system
Degree of success
Objectives not achieved Premature termination of the alliance
Successful achievement of objectives Partners negotiate new alliance transaction to jointly exploit the ESH printhead
Number of interviews
Graph managers : 2 Graph engineers: 4 Jet managers: 1 Jet engineers: 2
Graph managers: 2 Graph engineers: 5 Jet managers: 2 Jet engineers: 2
Number of documents
Contract: 1 Reports of technological meetings between Jet and Graph: 14 Reports of internal technological meetings: 33 Reports of steering meetings between Jet and Graph: 6 Other documents (Graph evaluation document, memo’s): 3
Contract: 1 Reports of technological meetings between Jet and Graph: 29 Reports of internal technological meetings: 29 Reports of steering meetings between Jet and Graph: 5 Other documents (memo’s and mail correspondence): 5
58
Table 3: Content of alliance contracts: differences between SSH and ESH contracts
Formalization of monitoring (Barney & Hesterly, 1996)
Narrow contractual interface structure SSH Alliance
Broad contractual interface structure ESH Alliance
Presence of performance monitoring mechanisms “Presence of program milestones, target dates, and performance standards for delivery of SSH printheads”
Presence of performance monitoring mechanisms “Presence of program milestones, target dates, and performance standards for delivery of ESH printheads”
Absence of behavior monitoring mechanisms
Presence of behavior monitoring mechanisms “Specification of technological activities that are supposed to be conducted by Jet”
Formalization of task division (Landau, 1969)
Overlapping task division Mutually exclusive task “Jet and Graph jointly division establish the initial design “Jet solely designs and specifications of the ESH develops the SSH prototype prototype printhead. Jet printheads, while Graph and Graph are supposed solely designs and develops to conduct similar the prototype printing system” technological tests”
Formalization of information flows (Van de Ven & Fry, 1980)
Absence of contractual obligation for information flows
59
Presence of contractual obligation for information flows “Planning of joint review meetings in which results of technological experiments need to be exchanged”
Figure 1: Overview dynamics of collaboration in the SSH alliance Negotiation of SSH contract (Dec ’98 – March ’99)
Managerial Level at Graph
* Is confronted with take-it-orleave it negotiation position * Sees alliance as a huge opportunity to get acquainted with new technology
Operational Level at Jet
Managerial Level at Jet
Emergence of delays in delivery of prototype printheads (Dec ’99 – Dec ’00)
Questions the technological competencies of Jet engineers (i.e., JET engineers do not seem to master their own technology)
Contractual definition of: * performance monitoring mechanisms * strict task division * absence of obligation to exchange information
* Contractual milestones are emphasized by Graph management * Jet management instructs Jet engineers to maintain strict task division and limits information exchanges * Independently attempts to address UTPs * Does not exchange information about UTPs toward Graph engineers
* Frames Graph as a financial investor * Questions the good intentions of Graph’s management (i.e., Graph is likely to steal knowledge and to impose changes in design specifications that reduce generic applicability of printheads)
Termination of alliance (Dec ’00 – April ’01)
* Increasingly looses its confidence in the good intentions of Jet’s management (i.e., Jet is trying to keep Graph happy) * Questions about feasibility SSH technology
Starts questioning the good intentions of Jet’s management (i.e., Jet’s management is more concerned about financial interests than about technological success)
* Has no fine-grained understanding of nature of UTPs at Jet * Is forced to stay on the sidelines
Operational Level at Graph
Governance Level
Emergence of unanticipated technological problems (UTPs) at Jet (March ’99 – Dec ’99)
Termination of SSH alliance by Graph
* Further isolates itself from Graph * Focuses on finding short-cuts to produce prototype printheads as quickly as possible
Further loses confidence in the good intentions of Graph when Graph’s starts asking for more information
60
Starts panicking because of contractual pressure
Is not able to find long-term solutions for UTPs
Starts in-depth investigation of UTPs
Figure 2: Overview dynamics of collaboration in the ESH alliance Event
Managerial Level at Graph
Negotiation of ESH contract (May ’00 – July ’00) * Possesses a dominant negotiation position * Feels the need to change contractual structure because of negative experiences in SSH alliance
Is substantially involved in solving remaining UTPs
* Faces precarious financial situation * Continues questioning good intentions of Graph management (i.e., fear of opportunistic actions by Graph is still present)
Positively assesses competencies of Jet engineers (i.e., Impressed by creativity and perseverance of Jet engineers)
* Graph’s management is wiling to adapt contractual milestones * Jet management instructs Jet engineers to provide additional information about core of ESH technology Engages in extensive information exchange on UTPs to Graph
Functional prototype ESH heads delivered (Feb ’02 – March ’02)
* Increasingly gains confidence in good intentions of Jet management (i.e., linear expansion of trust to a higher level) * Is confident in feasibility of ESH technology
Starts assessing the intentions of Jet’s management in a more positive way (i.e., Jet’s management is committed to solve UTPs)
Contractual definition of: * performance and behavior monitoring mechanisms * overlapping task division * presence of obligation to exchange information
Operational Level at Jet
Managerial Level at Jet
Emergence of serious delays in delivery of prototype printheads because of remaining UTPs (October ’01 – Feb ’02)
Is substantially involved in problem definition and problem solving processes (Exception: issues related to the core of the ESH technology)
Operational Level at Graph
Governance Level
Emergence of unanticipated technological problems (UTPs) at Jet (July ’00 – October ’01)
Successfully addresses some UTPs
* Provides additional information to Graph * Focuses on finding long-term solutions to remaining UTPs
Joint decision to continue collaboration in downstream stages of project
Positively assesses competencies of Graph engineers (i.e., technological added value of Graph engineers is substantial)
Starts assessing the intentions of Graph in a more positive way (i.e., Graph is committed partner)
Experiences significant changes (i.e., new
61
* Does not engage in panicking reactions * Increasingly gains confidence in good intentions of Graph management (i.e., Graph is strategic long-term partner)
Figure 3: Multi-level process models of dynamics of collaboration in SSH and ESH alliance Governance Level
SSH Alliance
ESH Alliance
Operational Level
Narrow contractual interface structure
Limited joint sensemaking on unexpected technological problems
Rigid application of contract
Contract compliance behavior Joint sensemaking on unexpected technological problems further reduces
Managerial Level Negative goodwill trust dynamics
Negative goodwill trust dynamics
Negative competence trust dynamics
Lowered expectations of feasibility of project
Broad contractual interface structure
Extensive joint sensemaking on unexpected technological problems
Positive goodwill trust dynamics
Flexible application of contract
Search for qualitative solutions Joint sensemaking on unexpected technological problems further increases Positive competence trust dynamics
62
Positive goodwill trust dynamics
Heightened expectations of feasibility of project
BIO STATEMENTS Dries Faems is an Assistant Professor at the Department of Operations, Organisations, and Human Resources in the Faculty of Management and Governance, University of Twente and Affiliated Researcher at the Research Centre for Organisation Studies in the Faculty of Business and Economics, Katholieke Universiteit Leuven. He obtained his Ph.D. in applied economics from Katholieke Universiteit Leuven. His current research focuses on governance processes in R&D alliances and performance implications of alliance portfolios. Maddy Janssens is a Full Professor at the Research Centre for Organisation Studies in the Faculty of Business and Economics, Katholieke Universiteit Leuven. Her main research interest focuses on ways in which organizations deal with differences. She currently studies diversity and inclusion in organizations, interorganizational collaboration, and global teams and managers. She received a PhD in organizational psychology from Katholieke Universiteit Leuven. Anoop Madhok is a Professor of Strategy at the Schulich School of Business, York University, Toronto and a Visiting Professor of Organization Theory and Fellow of the Center for Comparative Social Studies, Vrije Universiteit, Amsterdam. He obtained his Ph.D. from McGill University, Montreal. His research interests span strategy and international management and include topics such as multinational firm strategy, strategic alliances, trust and governance of interfirm collaborations, and the theory and boundaries of the firm. Bart Van Looy is Professor at the Department of Managerial Economics, Strategy and Innovation in the Faculty of Business and Economics, Katholieke Universiteit Leuven. His current research focuses on organizing innovation (company level) and (regional) innovation systems. Bart Van Looy has published on these topics in
63
journals like Research Policy, Journal of Product and Innovation Management, Organization Studies, R&D Management, Scientometrics and Journal of Technology Transfer. He received a PhD in organizational psychology from Katholieke Universiteit Leuven.
64
Organization Studies http://oss.sagepub.com
The Initiation and Evolution of Interfirm Knowledge Transfer in R&D Relationships Dries Faems, Maddy Janssens and Bart van Looy Organization Studies 2007; 28; 1699 DOI: 10.1177/0170840606082222 The online version of this article can be found at: http://oss.sagepub.com/cgi/content/abstract/28/11/1699
Published by: http://www.sagepublications.com
On behalf of:
European Group for Organizational Studies
Additional services and information for Organization Studies can be found at: Email Alerts: http://oss.sagepub.com/cgi/alerts Subscriptions: http://oss.sagepub.com/subscriptions Reprints: http://www.sagepub.com/journalsReprints.nav Permissions: http://www.sagepub.com/journalsPermissions.nav Citations (this article cites 46 articles hosted on the SAGE Journals Online and HighWire Press platforms): http://oss.sagepub.com/cgi/content/refs/28/11/1699
Downloaded from http://oss.sagepub.com at Universiteit Twente on December 18, 2007 © 2007 SAGE Publications. All rights reserved. Not for commercial use or unauthorized distribution.
article title
The Initiation and Evolution of Interfirm Knowledge Transfer in R&D Relationships Dries Faems, Maddy Janssens and Bart van Looy
Abstract Dries Faems Katholieke Universiteit Leuven, Belgium Maddy Janssens Katholieke Universiteit Leuven, Belgium Bart van Looy Katholieke Universiteit Leuven, Belgium
Organization Studies 28(11): 1699–1728 ISSN 0170–8406 Copyright © 2007 SAGE Publications (Los Angeles, London, New Delhi and Singapore) www.egosnet.org/os
In this study, we examine the process of interfirm knowledge transfer in R&D relationships. Based upon an embedded case study, we develop a model for understanding the initiation and evolution of interfirm knowledge transfer. Complementing previous cross-sectional research, this model points to the importance of legal clauses as a formal design alternative to equity governance structures, expectations of a long-term relationship as a specific indicator of trust, and similarity of technological equipment as an important facilitator for acquisition and assimilation of knowledge. This model also indicates that interfirm knowledge transfer is likely to continue in R&D relationships as long as perceived market threats remain limited and perceived technological complementarities remain extensive. In addition, we point to two strategies to organize an R&D relationship such that continuation of interfirm knowledge transfer is more likely to occur. Keywords: Interfirm knowledge transfer, R&D relationships
In R&D settings, firms increasingly rely on formal links with other organizations to increase their knowledge of innovative technologies (e.g. Faems et al. 2005; Hagedoorn 2002). Although such R&D relationships can lead to successful transfer of codified as well as tacit knowledge (Doz and Hamel 1997), failure rates of such collaborative efforts are said to be high (De Laat 1997). Previous research (e.g. Chen 2004; Lane and Lubatkin 1998; Mowery et al. 1996) has therefore examined the conditions that influence interfirm knowledge transfer in the specific setting of R&D. While this research offers valuable insights in the effectiveness of interfirm knowledge transfer in R&D relationships, several issues remain. First, these previous studies focus on the relationship between particular conditions and knowledge transfer outcomes such as the increase of patent cross-citations between collaborating partners (Mowery et al. 1996) or the perceived increase of a firm’s stock of new skills or capabilities (Chen 2004; Lane and Lubatkin 1998). As a result, the actual process of initiating interfirm knowledge transfer remains a black box (Inkpen and Pien 2006). Second, because of their mainly cross-sectional nature, these previous studies provide limited insights into how the process of interfirm knowledge transfer evolves over time. The purpose of this study is therefore twofold: (1) to explore the process of initiating knowledge transfer in R&D relationships; and (2) to investigate how and under what conditions interfirm knowledge transfer evolves once the desired knowledge is transferred. To accomplish these research objectives, we rely on an embedded case study of how one established company over the course of a DOI: 10.1177/0170840606082222 Downloaded from http://oss.sagepub.com at Universiteit Twente on December 18, 2007 © 2007 SAGE Publications. All rights reserved. Not for commercial use or unauthorized distribution.
1700
Organization Studies 28(11)
technological trajectory collaborated with three different entrepreneurial partners. Based on these data, we develop a model for understanding the process through which interfirm knowledge transfer in R&D relationships occurs. Further developing the facilitating conditions of knowledge transfer found in previous research, this model points to the importance of legal knowledge-sharing clauses as an alternative to equity governance structure, expectations of a long-term relationship as a specific indicator of trust, and similarity of technological equipment as an important additional facilitator of knowledge acquisition and assimilation. In addition, we newly identify the conditions that influence the evolution of knowledge transfer once the desired knowledge is transferred, referring to perceived technological complementarities and perceived market threats. Finally, we identify and assess two potential strategies to come to sustainable interfirm knowledge transfer in R&D relationships. This study contributes to the alliance literature by taking a process perspective on interfirm relationships, providing new insights into the complexity of knowledge transfer.
Theoretical Background The Initiation of Knowledge Transfer: Barriers and Facilitating Conditions
Following Hamel (1991), we define interfirm knowledge transfer as a process that consists of two critical steps. First, knowledge needs to be disclosed by the ‘expert partner’ or the firm that possesses the knowledge. As a second step, the disclosed knowledge needs to be acquired and assimilated by the ‘novice partner’. Knowledge acquisition is the process of identifying and evaluating the opportunities and liabilities of disclosed knowledge, while assimilation of knowledge refers to the process of embodying and internalizing the disclosed knowledge (Lane and Lubatkin 1998; Zahra and George 2002). Once knowledge is successfully transferred, the novice partner can start internally exploiting the disclosed knowledge for new markets, products and businesses (Hamel 1991; Zahra and George 2002). While R&D relationships are increasingly recognized as viable organizational structures for transferring technological knowledge between firms (Doz and Hamel 1997; Mowery et al. 1996), the initiation of interfirm knowledge transfer in R&D relationships may be difficult because of (1) the limited willingness of the expert partner to disclose knowledge, (2) the limited ability of the novice partner to acquire and assimilate knowledge. Willingness to Disclose Knowledge
When knowledge is disclosed, the other partner may be induced to abuse it for opportunistic or competitive motives (Hamel 1991; Khanna et al. 1998). For this reason, the expert partner may be hesitant to fully disclose the know-why (i.e. principles underlying the technology), know-how (i.e. procedures required to apply the technology) and know-what (i.e. specific technology configurations that different customers groups may want) (Garud 1997) of the technological knowledge to the novice partner (Larsson et al. 1998). Previous research
Downloaded from http://oss.sagepub.com at Universiteit Twente on December 18, 2007 © 2007 SAGE Publications. All rights reserved. Not for commercial use or unauthorized distribution.
Faems et al.: Interfirm Knowledge Transfer in R&D Relationships
1701
(Mowery et al. 1996; Cheng 2004) points to equity governance structure as an important formal condition that mitigates opportunistic abuse of disclosed knowledge. Equity governance structures include joint ventures, minority and majority participations; non-equity governance structures refer to all other cooperative arrangements not involving equity exchange (e.g. co-development agreement, technology sharing agreement) (Tsang 2000). Informed largely by transaction cost economics (e.g. Williamson 1985), the theoretical reasoning for this relationship is that equity governance structures create a mutual hostage situation through ex ante commitments to an alliance, reducing the incentive of the novice partner to opportunistically abuse disclosed knowledge (Dyer and Singh 1998; Williamson 1991). As a result, the expert partner is likely to be more motivated to disclose knowledge. Several researchers (e.g. Kale et al. 2000; Muthusamy and White 2005), however, criticize the transaction cost economics perspective for being acontextual and ahistorical, ignoring the impact of relational capital on interfirm knowledge transfer. They therefore highlight the importance of trust as a relational condition that motivates disclosure of knowledge. Trust refers to ‘a psychological state comprising the intention to accept vulnerability based upon positive expectations of the intentions or behavior of another’ (Rousseau et al. 1998: 395). Here, the theoretical argument is that trust tends to decrease the perceived risk of opportunism, encouraging members of the expert partner to engage in wide-ranging, continuous and intense contact with members of the novice partner (Kale et al. 2000; Muthusamy and White 2005). The study of Chen (2004), reporting a positive relationship between trust and the effectiveness of interfirm knowledge transfer, supports this argument in the specific context of R&D relationships. Ability to Acquire and Assimilate Knowledge
The second difficulty of knowledge transfer may be situated in the limited ability of the novice partner to acquire and assimilate knowledge (Hamel 1991; Lam 1997; Larsson et al. 1998). Technological knowledge is to some extent tacit and/or embedded within a specific context (Doz and Hamel 1997). However, the more tacit the knowledge, the lower the ability to structure that knowledge into a set of easily communicated rules and relationships (Kogut and Zander 1992; Nonaka and Takeuchi 1996). Valuation and internalization of disclosed knowledge may consequently be hampered. Existing research points to two organizational conditions that influence a firm’s ability to acquire and assimilate knowledge from another firm. A first condition is the similarity of the partners’ knowledge bases. Several scholars (e.g. Cohen and Levinthal 1990; Lane and Lubatkin 1998) have stressed that what can be learned from the expert partner is largely determined by what is already known by the novice partner. If the novice partner does not possess some amount of prior knowledge basic to the disclosed knowledge, it will be very difficult to value and internalize it for its own operations (Inkpen and Pien 2006). Effective acquisition and assimilation of disclosed knowledge therefore ask for some overlap in knowledge bases between the involved firms (Cohen and Levinthal 1990). A second condition is the similarity in organizational
Downloaded from http://oss.sagepub.com at Universiteit Twente on December 18, 2007 © 2007 SAGE Publications. All rights reserved. Not for commercial use or unauthorized distribution.
1702
Organization Studies 28(11)
structures and business practices. The more similar the organizational structures and business practices of expert and novice partner, the more likely the novice partner will possess the necessary knowledge-processing systems to value and internalize not only the codified but also the tacit components of the disclosed knowledge (Inkpen and Pien 2006; Lane and Lubatkin 1998; Simonin 1999). Lane and Lubatkin (1998) provide empirical evidence for the positive impact of these two conditions on the ability to acquire and assimilate knowledge in the particular setting of R&D. Using a sample of R&D alliances between pharmaceutical and biotechnology firms, they provide evidence that the effectiveness of such alliances in terms of increasing the pharmaceutical firm’s stock of knowledge is positively influenced by (1) the overlap between the novice firm’s basic knowledge and that of the expert firm; and (2) the similarity of the firm’s management formalization, management centralization and compensation practices. In sum, the above studies refer to equity governance structure, trust, similar basic knowledge and similar organizational structures and business practices as facilitating conditions of interfirm knowledge transfer in R&D relationships. However, as these findings are derived from studies focusing on the outcomes of knowledge transfer, the question remains to what extent they also facilitate the process of knowledge transfer. As mentioned previously, the first objective of this research is to examine how and under what conditions knowledge transfer in interfirm R&D relationships is initiated. The Dynamics of Knowledge Transfer: Dissolution or Continuation?
Once the desired technological knowledge is transferred, it remains unclear how the process of interfirm knowledge transfer evolves. Because previous research is mainly cross-sectional in nature, studies provide limited insights into the dynamics of interfirm knowledge transfer. Examining the alliance literature, we further notice that different studies seem to provide contradictory logics, pointing to continuation versus dissolution of knowledge transfer. Some alliance scholars (e.g. Hamel 1991; Khanna et al. 1998) seem to suggest that, after the novice partner has acquired and assimilated the desired knowledge, knowledge transfer is likely to dissolve. Applying insights from game theory (e.g. Oye 1986) and social exchange theory (e.g. Blau 1964), they argue that, as one partner succeeds in acquiring the desired tangible or intangible assets of the other partner, the expected future pay-offs of the relationship for the former partner are likely to decrease, triggering a shift from cooperative to non-cooperative behaviour. In contrast, other alliance scholars (e.g. Doz 1996; Larson 1992; Ring and Van de Ven 1994; Uzzi 1997) seem to argue that, when knowledge is successfully transferred from one partner to the other, the incentives to sustain and even expand the collaborative relationship are strong, providing new opportunities for interfirm knowledge transfer. These scholars refer to the dynamics of trust as an explanation in this respect. In line with the trust literature (e.g. Lewicki and Bunker 1996; Jones and George 1998; Ring 1997), they argue that, as partners successfully engage in knowledge transfer, the associated cycles of exchange, risk taking and successful fulfilment of expectations are likely to
Downloaded from http://oss.sagepub.com at Universiteit Twente on December 18, 2007 © 2007 SAGE Publications. All rights reserved. Not for commercial use or unauthorized distribution.
Faems et al.: Interfirm Knowledge Transfer in R&D Relationships
1703
trigger an increase from fragile to resilient trust levels. In situations of resilient trust, each party’s trustworthiness is based on confidence in the other’s values, backed up by empirical evidence derived from repeated behavioural interactions (Jones and George 1998). Such resilient trust provides individuals with the assurance that knowledge and information will be used for the greater good and that one need not exercise power or enforce contractual arrangements to protect one’s own interests (Ring 1997). As a consequence of such trust dynamics, partners become motivated to expand a single economic transaction to a diverse set of ongoing joint projects and interrelated exchange processes, leading to new opportunities for knowledge transfer (Larson 1992; Ring and Van de Ven 1994). In sum, while some scholars suggest that knowledge transfer is likely to dissolve because of changes in pay-off structures, other scholars argue that knowledge transfer is likely to continue because of positive trust dynamics. Given these contradictory logics, the second objective of this research is to examine how and under what conditions knowledge transfer evolves over time in interfirm R&D relationships. Methodology Research Design
In this study, we adopt an embedded case study approach (Yin 2003), examining three interfirm R&D relationships that were part of the same technological trajectory. This methodology suits our goal of developing a model in an area where little data or theory exist (Yin 2003). In addition, it allows us to perform a comparative analysis of three interfirm relationships, an analysis that facilitates ‘analytic generalization’ (Parkhe 1993a; Yin 2003). Through selecting interfirm R&D relationships that were part of the same technological trajectory, we minimize the influence of extraneous variation on our research findings (Eisenhardt 1989). The main limitation of our research design is that its generalizability to a larger and more diverse population might be restricted. The value of this research lies instead in its capacity to provide in-depth insights in complex and dynamic processes, which are difficult to reveal by means of more cross-sectional research designs (Larson 1992). In this study, all names of firms, products and individuals are disguised to ensure confidentiality. Research Setting
In the early 1990s, MAT, a Belgian company working on a global scale with products and systems based on metal transformation and advanced coatings, identified diamond-like coatings as a new promising technology to expand its coating activities. At the end of 2003, one of its divisions, MAT Diamond, succeeded in becoming the leading supplier of diamond-like coatings for a wide array of applications. During the development of this technological trajectory, MAT initiated interfirm relationships with three different partners: USCOAT, RES and FRCOAT (see Table 1). In all three interfirm R&D relationships, the initial objective was the unilateral (i.e. MAT–USCOAT relationship and
Downloaded from http://oss.sagepub.com at Universiteit Twente on December 18, 2007 © 2007 SAGE Publications. All rights reserved. Not for commercial use or unauthorized distribution.
1704
Organization Studies 28(11)
Table 1. Overview Interfirm R&D Relationships MAT-RES Relationship
MAT-FRCOAT Relationship
Characteristics
MAT-USCOAT Relationship
Description of partners
MAT: International group specialized in metal transformation and advanced coatings, located in Belgium USCOAT:
RES:
FRCOAT:
High-tech SME specialized in advanced coatings and advanced ceramic materials, located in the US
Research institute specialized in advanced materials, located in Belgium
University spin-off specialized in advanced coatings, located in France
Initiation of interfirm relationship
Feb 1995
Dec 1997
July 2001
Initial technological objective of interfirm relationship
Transfer of USCOAT’s DLX technology to MAT to explore the feasibility of this technology at MAT
Transfer of RES’s DLC technology to MAT to explore the feasibility of this technology at MAT
Transfer of FRCOAT’s Advanced DLC technology to MAT to further exploit MAT’s DLX/DLC technology Transfer of MAT’s DLX/DLC technology to FRCOAT to further exploit FRCOAT’s Advanced DLC technology
History of governance structure of interfirm relationship
Feb 1995 – Dec 1997: Contractual agreements Dec 1997 – Aug 2000: Joint Venture (MAT 60%; USCOAT 20%; RES 20%) Aug 2000: MAT buys out USCOAT from the Joint Venture MAT Diamond
Dec 1997 – Aug 2000: Joint Venture (MAT 60%, USCOAT 20%, RES 20%) Aug 2000 – May 2002: Joint Venture (MAT 80%, RES 20%) May 2002: MAT buys out RES from the Joint Venture MAT Diamond
July 2001 – July 2003: Minority participation (MAT holds 48% of FRCOAT’s shares) July 2003: Majority participation (MAT holds 90% of FRCOAT’s shares)
MAT–RES relationship) or bilateral (i.e. MAT–FRCOAT relationship) knowledge transfer of an existing coating technology from one partner to the other. We should point out that, between 1998 and 2000, USCOAT and RES were minority partners in the same joint venture with MAT. However, we do not consider the interfirm relationship between USCOAT and RES in this study because no knowledge transfer was intended to take place between these two firms. Data Collection and Analysis
Data on the three interfirm relationships were collected in a retrospective way which allowed for a much more focused data-gathering process (Poole, Van de Ven, Dooley and Holmes 2000). At the same time, unconsciously accepting respondent bias might occur in retrospective studies, leading to confusion about cause-and-effect relationships (Leonard-Barton 1990). To improve the validity of the retrospective reports, we applied a number of strategies. First, we triangulated our data, applying two data sources: interviews and documents (Eisenhardt 1989, Yin 1984). Second, we asked informants to reflect on concrete events rather than past opinions or beliefs to reduce the risk of cognitive biases and impression management (Miller et al. 1997). Finally, we attempted to verify individual reports by asking similar questions to multiple informants.
Downloaded from http://oss.sagepub.com at Universiteit Twente on December 18, 2007 © 2007 SAGE Publications. All rights reserved. Not for commercial use or unauthorized distribution.
1705
Faems et al.: Interfirm Knowledge Transfer in R&D Relationships Figure 1. Chronological Overview of Main Events (Simplified Depiction)
Feb 1995 MAT and USCOAT sign Technology Evaluation Agreement
Dec 1997 Initiation of Joint Venture MAT Diamond
Dec 1996 MAT starts experimenting with combinations of DLX and DLC coating technologies
August 2000 USCOAT sells its JV shares to MAT
Dec 1998 MAT decides to fully focus on commercializing its DLX/DLC technology
July 2001 MAT takes minority participation in FRCOAT
January 2001 JV activities globally expand
May 2002 RES sells its JV shares to MAT
July 2003 MAT takes majority participation in FRCOAT
January 2002 July 2002 MAT and FRCOAT MAT tries to decide to enter commercial exchange coating automotive systems market
MAT-USCOAT Relationship MAT-RES Relationship MAT-FRCOAT Relationship
Applying the suggestions of Pettigrew (1990) and Pentland (1999), we made an explicit distinction between three different stages in our theory-building process, representing an evolution of surface levels to deeper levels of data collection and analysis. In the first stage, we conducted unstructured interviews with two MAT managers who were closely involved in all three interfirm R&D relationships. The purpose of these interviews was to obtain initial information about the history and characteristics of the different interfirm R&D relationships (e.g. how successful was the relationship; what major events shaped the relationship; how would the interviewee describe the other organizations). For each interfirm R&D relationship, we also studied relevant documents. In addition to publicly available information (e.g. annual reports, press releases), we gained access to 81 private documents (e.g. contracts, slideshows of technological meetings, minutes of steering meetings) totalling 1731 pages. Based on both interviews and documents, we constructed a graphical representation of the chronology of the major events that took place within each R&D relationship. Figure 1 presents a simplified depiction of this chronology. In the second stage, we conducted semi-structured interviews (Kvale 1996) with informants from the different organizations involved. Between October 2003 and November 2004, we individually interviewed 19 persons (see Table 2). The interviews were structured along the chronology of the major events, asking respondents to describe these events and the kind of interactions they triggered between the partners. The average length of interview was between one and two hours. The transcribed interviews were sent back to the interviewees for feedback. At this stage, we also re-examined the available documents to verify whether the content of the interviews was consistent with the content of the documents. When discrepancies between these two data sources were observed, we again contacted respondents to ask for additional comments. For each interfirm relationship, a case study report was then written with extensive use of citations from both the interviews and documents to achieve a high level of accuracy (Langley 1999). These reports were sent to one manager of each company; their comments were collected via electronic mail or face-toface conversations, leading to the final case study reports.
Downloaded from http://oss.sagepub.com at Universiteit Twente on December 18, 2007 © 2007 SAGE Publications. All rights reserved. Not for commercial use or unauthorized distribution.
1706
Organization Studies 28(11)
Table 2. Overview of Interviews
Company
Function of Interviewee
MAT
Corporate Manager Project Manager Engineer/Technician Sales Manager Lawyer Corporate Manager Project Manager Engineer/Technician Project Manager Engineer/Technician Corporate/Project Manager Engineer/Technician
RES USCOAT FRCOAT
Number of Interviewees 2 2 2 1 1 2 2 1 1 3 1 1 19
In the third and last stage, we analysed the case study reports through an inductive approach, relying on an iterative process that coupled within-unit analysis with between-unit analysis (Eisenhardt 1989; Yin 2003). We started by conducting a within-unit analysis for each observed interfirm R&D relationship, focusing on the initiation and evolution of knowledge transfer. After completing the within-case analyses, we compared the findings across the three cases. Based on the identification of similarities and differences across cases, new iterations of within-case and across-case analysis were subsequently initiated. This procedure was repeated until dominant findings emerged. As we analysed our data, we sometimes felt the need to collect additional data. For instance, conducting a first analysis of the data on the MAT–FRCOAT relationship, it was unclear to us why MAT was immediately willing to disclose its coating technology to FRCOAT. We therefore contacted one of the MAT managers involved, asking him the reason for MAT’s openness in disclosing knowledge to FRCOAT. When a first draft of this paper was completed, we again conducted feedback interviews with the two MAT managers who were interviewed at the first stage to discuss the content of the study. These feedback interviews proved to be very helpful in finetuning our insights as well as testing the internal validity of our findings.
Results In this section, we turn to our two research questions, examining the initiation of interfirm knowledge transfer and its evolution after the desired transfer is completed. As it is the purpose of this study to inductively develop a model, we provide for each R&D relationship a discussion of the initiation and evolution of interfirm knowledge transfer and its conditions. Table 3 presents the main findings of all three cases. MAT–USCOAT Relationship Initiation of Knowledge Transfer
In February 1995, MAT and USCOAT, a US high-tech SME that had developed a pioneering diamond-like coating called DLX, started an R&D relationship by Downloaded from http://oss.sagepub.com at Universiteit Twente on December 18, 2007 © 2007 SAGE Publications. All rights reserved. Not for commercial use or unauthorized distribution.
Downloaded from http://oss.sagepub.com at Universiteit Twente on December 18, 2007 © 2007 SAGE Publications. All rights reserved. Not for commercial use or unauthorized distribution.
Technological equipment
Conditions of knowledge acquisition and assimilation Knowledge bases Organizational structures and business practices
Nature of knowledge acquisition and assimilation
Overlapping knowledge base Hierarchical structure and business practices versus entrepreneurial structure and business practices Similar coating systems
Similar coating systems
Dec ’97 – Dec ’98: High ability to acquire and assimilate DLC technology by MAT
Feb ’95 – Feb ’96: High ability to acquire and assimilate DLX technology by MAT
No overlapping knowledge base Hierarchical structure and business practices versus entrepreneurial structure and business practices
JV structure Specified in detail Long-term horizon
Non-equity structure Specified in detail Long-term horizon
Dec ’97 – Dec ’98: Open disclosure of know-what, know-how, and know-why of DLC technology by RES
Feb ’95 – Feb ’96: Open disclosure of know-what, know-how, and know-why of DLX technology by USCOAT
Nature of knowledge disclosure
Conditions of knowledge disclosure Governance structure Legal knowledge transfer clauses Expected time horizon
MAT-RES Relationship
MAT-USCOAT Relationship
Initiation of knowledge transfer during first year of relationship
Overlapping knowledge base Hierarchical structure and business practices versus entrepreneurial structure and business practices Different coating systems
July ’01- Jan ’02: Limited ability to acquire and assimilate DLX/DLC technology by FRCOAT Limited ability to acquire and assimilate Advanced DLC technology by MAT
(continued)
Overlapping knowledge base Hierarchical structure and business practices versus entrepreneurial structure and business practices Similar coating systems
Jan ’02 – July ’02: High ability to acquire and assimilate DLX/DLC technology by FRCOAT High ability to acquire and assimilate Advanced DLC technology by MAT
Minority equity structure Specified in detail Long-term horizon
July ’01- July ’02: Open disclosure of know-what, know-how, and know-why of DLX/DLC technology by MAT Open disclosure of know-what, know-how, and know-why of Advanced DLC technology by FRCOAT
MAT-FRCOAT Relationship
Table 3. Initiation and Evolution of Knowledge Transfer in Observed Interfirm R&D Relationships: An Overview
Faems et al.: Interfirm Knowledge Transfer in R&D Relationships
1707
Limited Limited
Limited
Extensive
Limited
Dec ’98 – May ’02: No interfirm knowledge transfer
Feb ’96 – Dec ’98: Dec ’98 – Aug ’00: Bilateral disclosure No interfirm and acquisition/ knowledge transfer assimilation of know-what, know-how, and know-why of internal experiments
Nature of knowledge evolution
Extensive
MAT-RES Relationship
MAT-USCOAT Relationship
Evolution of knowledge transfer after first year of relationship
Conditions of knowledge evolution Perceived technological complementarities Perceived market threats
MAT-RES Relationship
Initiation of knowledge transfer during MAT-USCOAT Relationship first year of relationship
Table 3 (continued)
Limited
Extensive
July ’02 – July ’03: Bilateral disclosure and acquisition/assimilation of know-what, know-how, and know-why of internal experiments
MAT-FRCOAT Relationship
MAT-FRCOAT Relationship
1708 Organization Studies 28(11)
Downloaded from http://oss.sagepub.com at Universiteit Twente on December 18, 2007 © 2007 SAGE Publications. All rights reserved. Not for commercial use or unauthorized distribution.
Faems et al.: Interfirm Knowledge Transfer in R&D Relationships
1709
signing a technology evaluation agreement. In this agreement, it was stated that the objective of the collaboration was ‘to allow MAT an initial opportunity to assess the commercial potential of the Technology [DLX coating] and to determine its interest in pursuing future joint commercial activities with USCOAT’ (Technology Evaluation Agreement: 1). After signature of the technology evaluation agreement, USCOAT began to disclose knowledge about its DLX technology. Specifically, a number of meetings with MAT engineers were organized, during which ‘there was a lot of information exchange [toward MAT engineers] including papers, journal articles and unpublished communications’ (USCOAT project manager). These documents contained detailed information about the know-why (i.e. fundamental properties such as ‘stability, adhesion, and conductivity of DLX coating’; USCOAT White Paper 95–1023: 1–2) and the know-what (i.e. ‘coating evaluations of USCOAT’s priority accounts’; USCOAT Market Development Plan for DLX: 9) of the DLX technology. In addition, MAT’s project manager visited USCOAT for one month, allowing him ‘to see with [his] own eyes how the coatings were manufactured’ (MAT project manager). In other words, USCOAT also disclosed the ‘know-how’ of its technology. The open disclosure of knowledge was also emphasized in the interviews: ‘Despite the risk of withholding information during this initial phase, they [USCOAT] played it openly.’ (MAT manager) ‘I think we were fairly open.’ (USCOAT project manager)
USCOAT interviewees provided two grounds for their openness in disclosing knowledge. First, they referred to the presence of specific contractual clauses in the technological evaluation agreement, regulating how MAT could apply the disclosed knowledge. For instance, the contract stipulated that ‘MAT shall not, without the prior written consent of USCOAT, directly or indirectly use or incorporate the technology in connection with any work or project’ (Technology Evaluation Agreement: 2). According to one USCOAT engineer, such contractual clauses provided ‘sufficient legal protection for providing information [to MAT]’. Second, USCOAT interviewees pointed to the expected time horizon of the relationship with MAT. Negotiating with MAT managers, USCOAT’s CEO was confident in MAT’s willingness to establish a long-term relationship with USCOAT to launch DLX coatings onto the market. This long-term horizon seemed to have a positive impact on their openness toward MAT. A USCOAT engineer, for instance, mentioned that ‘we were willing to provide all information because we saw it as a beneficial long-term relationship’. Next to the high motivation of USCOAT to disclose knowledge, the ability of MAT to acquire and assimilate this knowledge seemed to be extensive. Despite the fact that (1) MAT engineers did not have any experience with diamond-like coatings and (2) partners’ organizational structures and business practices were clearly different (i.e. hierarchical structure with highly formalized business practices versus entrepreneurial structure with low degree of business practice formalization), MAT interviewees stressed that that they quickly acquired a feeling for the advantages and disadvantages of the technology: ‘We swiftly started understanding what was good about it [USCOAT’s DLX technology], what was not good about it, what were the potential opportunities of it.’ (MAT manager)
Downloaded from http://oss.sagepub.com at Universiteit Twente on December 18, 2007 © 2007 SAGE Publications. All rights reserved. Not for commercial use or unauthorized distribution.
1710
Organization Studies 28(11)
MAT’s project manager also mentioned that ‘after a few months, we succeeded in internally manufacturing our first DLX samples’. In other words, MAT quickly managed to replicate USCOAT’s DLX technology, pointing to successful assimilation of the disclosed knowledge. To explain this high ability to acquire and assimilate the disclosed knowledge, MAT interviewees referred to the presence of similar coating systems at USCOAT and MAT. At the start of their interfirm relationship, MAT and USCOAT had agreed that MAT, experienced in developing vacuum systems, would develop two identical coating systems. One of these coating systems would be installed at USCOAT, while the other coating system would remain at MAT. According to the MAT interviewees, the presence of these similar coating systems was a huge advantage in acquiring and assimilating the disclosed knowledge as it allowed them to ‘speak the same language’ and ‘use similar process parameters’. Evolution of Knowledge Transfer
After MAT had acquired and assimilated USCOAT’s DLX technology, it started to internally experiment with the technology to optimize its feasibility for industrial-wear applications in the European market. At the same time, USCOAT refined the DLX technology for high-end applications in the American micro-electronics market. While disclosure of knowledge had been rather unilateral during the first year (from USCOAT to MAT), we found indications that it started to become more bilateral later on. Meetings were organized between the two partners on a regular basis. In one of the documents, the objective of these meetings was worded as follows: ‘The purpose of this meeting is to discuss the technical and commercial issues with respect to the DLX and related technologies. The format is an open dialogue of two partners with similar goals: broad commercialization of the DLX technology.’ (Minutes of MAT/ USCOAT Technology/Business Meeting, 4–5 September 1997: 1)
During these meetings, partners openly exchanged the results of their internal DLX experiments. USCOAT, for instance, provided ‘results of doping DLX coatings with various kinds of atoms’ (USCOAT slideshow Technology/Business Meeting, 20–26 March 1997:33). In a similar vein, MAT presented detailed information about its attempts to ‘combine DLX and DLC technologies to improve the hardness of coatings’ (MAT slideshow Technology/Business Meeting, 12–13 December 1996: 12). Not only the results, but also the ‘process system upgrades’ (USCOAT slideshow Technology/Business Meeting, 20–26 March 1997: 26) to execute these experiments were exchanged. Finally, partners also exchanged ‘questionnaires for DLX coatings’, which both MAT and USCOAT had sent to their potential customers to ‘better understand the [customer’s] needs and expectations for a DLX coating’ (MAT slideshow Technology/Business Meeting, 12–13 December 1996: 6). In the interviews, two grounds were provided to explain the emergence of such bilateral knowledge disclosure of know-why, know-how and know-what. First, it was stressed that, from a technological point of view, it was interesting to bilaterally exchange information. To be specific, as both partners conducted experiments with the same technology but in different contexts, ‘the exchange of information [about the experiments] provided opportunities to know more
Downloaded from http://oss.sagepub.com at Universiteit Twente on December 18, 2007 © 2007 SAGE Publications. All rights reserved. Not for commercial use or unauthorized distribution.
Faems et al.: Interfirm Knowledge Transfer in R&D Relationships
1711
about how the technology behaved in different contexts’ (MAT engineer). Second, it was emphasized that, as partners focused on clearly different applications for different markets, the risk that bilateral transfer of knowledge would bring along increased market competition was limited. A MAT manager, for instance, mentioned that: ‘From an American perspective, they looked at high-tech applications, while we focused on wear applications for the European market. In my opinion, this was a good division … If the teams of USCOAT and MAT would have targeted the same markets, they would have contacted the same customers and tried to find solutions for the same applications.’ (MAT manager)
We also found indications that the partners’ ability to acquire and assimilate the disclosed knowledge remained high. USCOAT engineers, for instance, stressed that ‘we internally managed to replicate some of MAT’s experiments regarding combining DLX and DLC coatings’. At the same time, MAT turned out to be successful in imitating some of USCOAT’s experiments regarding doping DLX coatings. After experimenting with the DLX technology for two years, MAT became convinced that this technology could offer new industrial applications. At the end of 1997, MAT bought a licence from USCOAT to commercially exploit the DLX technology in Europe and made a proposal to USCOAT to jointly start up a business activity for diamond-like coatings in Europe. In December 1997, the joint venture ‘MAT Diamond’ was officially launched. At the beginning of the joint venture, USCOAT and MAT continued their bilateral exchange of testing results during regular technical meetings. However, after the first year of the joint venture, the frequency of technical meetings dropped drastically. According to both the MAT and USCOAT interviews, it was MAT that had lost its motivation to engage in bilateral exchange of information: ‘We became more reserved concerning [exchanging information about] our [technological] developments.’ (MAT manager) ‘I think the information was initially very intense and slowed down later on … We no longer knew what was actually done at MAT.’ (USCOAT project manager)
MAT interviewees referred to the evolution of the technological activities at both partners to explain their decreased motivation to engage in bilateral exchange of testing results. After the joint venture was initiated, MAT became committed to apply a combination of DLX and more traditional DLC technologies for commercial exploitation of industrial wear applications in Europe. Although USCOAT engineers had been informed about this combined DLX/ DLC coating, they did not apply this coating for their commercial development efforts in the American market. An USCOAT engineer stressed that ‘our management was not willing to invest in the development of DLC-oriented coatings as USCOAT had recently launched an intensive marketing campaign in the US to promote its DLX technology as an alternative to DLC coatings’. USCOAT consequently remained focused on further developing pure DLX coatings for their blue-sky applications. However, USCOAT’s continued experiments with its pure DLX coating were no longer interesting from a technological point of view as they had no relevance for MAT’s DLX/DLC development efforts:
Downloaded from http://oss.sagepub.com at Universiteit Twente on December 18, 2007 © 2007 SAGE Publications. All rights reserved. Not for commercial use or unauthorized distribution.
1712
Organization Studies 28(11)
‘They worked on DLX applications for the electronics industry where you had to put Fluor in the coatings. That was none of our business. There almost was no common ground left.’ (MAT engineer)
As a result, MAT engineers became less interested in exchanging technological knowledge with the USCOAT engineers. In 2000, due to internal financial problems, USCOAT sold not only its MAT Diamond shares but also its entire coating division to MAT, representing the end of the interfirm relationship. MAT–RES Relationship Initiation of Knowledge Transfer
When MAT decided to start a diamond-like coating business activity in 1997, they asked not only USCOAT but also RES, a Flemish research institute that had developed its own DLC coating and had conducted first production of diamond-like coatings for a limited number of local customers, to become a joint venture partner. In December 1997, a joint venture agreement was signed, stipulating that RES brought its DLC technology and its existing DLC customers into the joint venture in exchange for 20% of the joint venture shares. Interviewees from both companies stressed that, from the beginning, the disclosure of knowledge from RES to MAT was very transparent: ‘Communication from RES to MAT was very open.’ (MAT engineer) ‘We passed on the complete set-up of the RES process as well as all the process parameters.’ (RES engineer)
As in the MAT–USCOAT relationship, technological meetings and visits were organized during which RES engineers provided detailed information about the know-why (i.e. fundamental coating properties and characteristics), know-how (i.e. demonstrations of how coatings are produced) and know-what (i.e. coating evaluations of RES’s existing customer base) of the DLC technology to MAT engineers. Again, RES interviewees pointed to specific contractual clauses to explain their openness in disclosing knowledge to the other partner: ‘In the JV Agreement, a number of annexes were present that described in detail which technology would be transferred to the joint venture. At that moment, we had just started working on another technology that had some linkages with the DLC technology. Through these annexes we could prevent the disappearance of this technology.’ (RES project manager)
Second, RES interviewees referred to the positive expectations that they, as the expert partner, had about the long-term orientation of the relationship with the novice partner (i.e. MAT). The RES project manager, for instance, stated: ‘RES believed that this JV could be the start of a close relationship with MAT which, in the long term, would go beyond the scope of the JV.’ In addition to the open disclosure by RES, MAT engineers seemed to have limited difficulties in acquiring and assimilating the disclosed knowledge. For instance, MAT engineers stated that they ‘quickly managed to start comparing RES’s DLC technology with [their] own DLX/DLC technology’ and ‘were able to get familiar with the technology.’ As in the MAT–USCOAT relationship, MAT interviewees referred to the presence of similar coating Downloaded from http://oss.sagepub.com at Universiteit Twente on December 18, 2007 © 2007 SAGE Publications. All rights reserved. Not for commercial use or unauthorized distribution.
Faems et al.: Interfirm Knowledge Transfer in R&D Relationships
1713
systems as a condition that facilitated the acquisition and assimilation of the disclosed knowledge. At the beginning of this relationship, MAT had bought a second-hand coating system from RES and it was ‘this coating system [that] allowed imitation of the characteristics of RES’s DLC technology’ (MAT engineer). Evolution of Knowledge Transfer
In contrast to the MAT–USCOAT relationship, no technological meetings emerged in which partners bilaterally exchanged information about their internal experiments. Instead, the intensity of communication between engineers of both partners drastically dropped after the first year. Two explanations were provided to explain this evolution. First, it was stressed that MAT engineers were no longer interested in RES’s DLC technology. After conducting experiments with RES’s DLC technology, MAT engineers quickly concluded that ‘this technology turned out to have limited opportunities for industrial commercialization because of scale-up problems’ (MAT engineer). As a consequence, ‘MAT’s technological interest in RES’s DLC technology quickly faded away’ (MAT manager). Second, it was emphasized that ‘MAT was not open about its own internal DLX/DLC experiments’ (RES engineer). Asked for their reasons, MAT interviewees expressed their fear that, if RES engineers, being research institute scientists, were to be informed about MAT’s technological progress in exploiting the DLX/DLC technology, they would be tempted to talk about it at scientific conferences and workshops, triggering the risk that potential competitors would gain access to sensitive knowledge and could become active in the same markets: ‘At RES, the risk of leakage of valuable information was large. After all, such scientists are used to giving speeches at conferences and telling everything they know … We did not want other companies to get access to our coatings.’ (MAT manager)
It needs to be stressed that, at the start of the joint venture, MAT and RES had signed a separate R&D agreement, which stipulated that MAT had to fund a certain amount of R&D projects at RES during the next five years. The purpose of these projects was that RES would apply its expertise in diamond-like coatings to generate new coating knowledge, which could then be transferred to the MAT team. RES therefore started exploring a number of new technological coating opportunities. However, it quickly became apparent that MAT engineers were not really interested in discussing the progress of these explorative R&D projects. To explain this disinterest, interviewees stressed that, as MAT was fully committed to commercializing its own DLX/DLC technology, the motivation to spend time on discussing the potential of alternative technological opportunities was limited: ‘The MAT people were fully occupied with commercializing their [DLX/DLC] technology. They had to make sure that they had sufficient revenues. Discussing new opportunities was not a priority for them … I remember that we really had to insist on having a meeting during which we could discuss our research.’ (RES project manager) ‘When we started commercializing our technology, our interest in more basic R&D faded away … We [therefore] did not supervise these projects … Regarding the research projects [of RES], there was limited communication.’ (MAT engineer)
Downloaded from http://oss.sagepub.com at Universiteit Twente on December 18, 2007 © 2007 SAGE Publications. All rights reserved. Not for commercial use or unauthorized distribution.
1714
Organization Studies 28(11)
In 2001, when USCOAT had already left the joint venture, the JV activities were expanding globally. As this expansion required additional financial investments, MAT asked RES to jointly increase MAT Diamond’s working capital. However, RES’s board of directors was not willing to contribute financial resources to the joint venture. In 2002, MAT chose to buy out RES, making themselves the sole arbiter of MAT Diamond’s future. In this way, MAT Diamond, originally a joint venture between three partners, became a fully owned subsidiary of MAT. MAT–FRCOAT Relationship Initiation of Knowledge Transfer
From 2000 on, MAT realized that, to stay competitive in this technological domain, it needed collaboration with other partners who possessed complementary coating technologies. In this respect, FRCOAT, a spin-off from a French university that had successfully commercialized a high-quality DLC coating for a particular niche market (i.e. the racing industry), was recognized as an interesting partner. In July 2001, MAT and FRCOAT signed an agreement, stipulating that MAT took a minority interest (47.8%) in FRCOAT. At the same time, the partners signed a cross-licensing agreement, stipulating that each partner disclosed its own coating technology and expertise to the other. Shortly after the agreement was signed, technical meetings and visits were organized in which, according to the interviewees, the project managers and engineers of both MAT and FRCOAT openly disclosed their coating technology: ‘I did not hide anything about FRCOAT’s process. Also the MAT people did not hide anything.’(FRCOAT engineer) ‘From the start, everything was put on the table.’ (MAT engineer)
To explain their openness in disclosing knowledge, MAT interviewees stressed that there was no risk in disclosing knowledge as ‘FRCOAT did not have the financial means to abuse such knowledge for competitive reasons’ (MAT manager). When we asked the CEO of FRCOAT why he was willing to openly disclose its technology to MAT, he emphasized his confidence in MAT’s long-term commitment to the relationship: ‘What convinced me was the straightforward discourse I had with the people I met [during the negotiation stage] … They clearly communicated that their objective was not to just acquire a technology, but to really commit to a joint project within this technological domain.’ (FRCOAT engineer)
Finally, we noticed that, as in the MAT–USCOAT and MAT–RES relationships, contractual clauses were also present that explicitly stipulated how and for which purpose partners could use the disclosed knowledge. While the disclosure of knowledge seemed to be very open, we found indications that, during the first months of the MAT–FRCOAT relationship, both partners had difficulties in acquiring and assimilating the disclosed knowledge. Although both partners had significant expertise in the field of DLC technology, interviewees from both firms indicated that they initially experienced problems in evaluating and replicating the partner’s coating technology. MAT’s project manager, for instance, stated that ‘initially we were not able to become
Downloaded from http://oss.sagepub.com at Universiteit Twente on December 18, 2007 © 2007 SAGE Publications. All rights reserved. Not for commercial use or unauthorized distribution.
Faems et al.: Interfirm Knowledge Transfer in R&D Relationships
1715
a natural at FRCOAT’s technology’. After six months, the firms therefore decided to install each other’s coating systems: a FRCOAT coating system was installed at MAT, while a MAT coating system was installed at FRCOAT. As one FRCOAT engineer expressed, the installation of each other’s technological equipment was experienced as a fundamental step in becoming able to evaluate the partner’s technology: ‘For me this [exchange of coating systems] was the fundamental step in the collaboration which meant that both parties really started working with each other’s technology … It was a very interesting step because earlier on we thought that it [partner’s technology] was very good, but by using the machines we started experiencing problems … In this way it became possible to list the strong and weak characteristics of the partner’s technology.’ (FRCOAT engineer)
At the same time, a MAT engineer explicitly stressed the importance of this equipment for their ability to imitate the disclosed knowledge: ‘During the first months, we just visited them [FRCOAT]. We got a lot of information about their machines and technology. However, we gradually realized that getting a large amount of information on a blackboard and looking at how they turn the buttons was not sufficient … After coating systems were exchanged, both partners started experimenting with it. From then on, we became familiar with the technology [of the other partner].’ (MAT engineer)
Through experimenting with the coating system of the other partner, engineers were able to develop the necessary know-how in order to understand and use the other partner’s technology. In sum, as in the two previous cases, the presence of similar coating systems seemed to be a crucial factor in generating the ability to acquire and assimilate the disclosed knowledge. Evolution of Knowledge Transfer
After MAT and FRCOAT succeeded in acquiring and assimilating each other’s coating technology, they both started to conduct experiments internally with the other partner’s technology. MAT engineers, on the one hand, started examining how FRCOAT’s advanced DLC technology could help to optimize their own DLX/DLC technology for applications in the commercial automotive market. FRCOAT engineers, on the other hand, experimented with MAT’s DLX/DLC technology to improve their own coating technology for its applications in the racing industry. As in the MAT–USCOAT relationship, engineers from both partners had regular meetings in which the set-up, results and customer evaluations of their experiments were openly exchanged and discussed: ‘Meetings between technical people [of both partners] were very frequent … We had very open discussions with them. There was no confrontation. The experience of everybody was put on the table.’ (MAT engineer)
Interviewees stressed that, as both partners were experimenting with the same technologies but for different applications in different markets, it was interesting to exchange this information without being exposed to the risk that it would trigger competitive threats: ‘They conducted experiments with our technology which we would never think of … This was very interesting.’ (FRCOAT engineer)
Downloaded from http://oss.sagepub.com at Universiteit Twente on December 18, 2007 © 2007 SAGE Publications. All rights reserved. Not for commercial use or unauthorized distribution.
1716
Organization Studies 28(11)
‘The applications of both technologies were quite different. The applications did not focus on the same markets. At that level there were few synergies. But this was also an advantage because it reduced the likelihood of conflict.’ (MAT engineer)
In July 2003, the partners agreed to increase MAT’s participation in FRCOAT to 90%, representing a quasi integration of FRCOAT into MAT’s diamond-like coating business activity, which was still expanding on a global scale.
Discussion This study aims at contributing to the alliance literature by taking a process perspective on interfirm knowledge transfer in R&D relationships. Relying on the findings of an embedded case study in which one established company collaborated with three different entrepreneurial partners, we present a theoretical model that indicates the conditions facilitating the initiation of interfirm knowledge transfer as well as conditions and strategies impacting on its evolution. This model (see Figure 2) aims to be comprehensive, presenting conditions and strategies discussed in previous research as well as those newly discovered in this study. Initiating Interfirm Knowledge Transfer in R&D Relationships
The model first identifies the conditions which facilitate the expert partner’s willingness to disclose knowledge, the first step of knowledge transfer, followed by those which facilitate the novice partner’s ability to acquire and assimilate knowledge, the second step. Motivation to Disclose Knowledge
Our data indicate that, in interfirm R&D relationships, disclosure of knowledge can be initiated via technological meetings and visits during which engineers of the novice partner receive detailed information about the technology of the expert partner. Two important conditions which facilitate this process emerge from this study: legal knowledge-transfer clauses and expectations of a long-term relationship. While previous research (Chen 2004; Mowery et al. 1996) has pointed to equity governance structure as a facilitating condition, this study suggests that this formal condition may not be a sufficient nor necessary condition to motivate the expert partner to disclose knowledge. In the two interfirm relationships that possessed an equity structure (i.e. MAT–RES and MAT–FRCOAT relationships), partners had also negotiated detailed contractual clauses, providing a legal framework for the transfer of knowledge. Asked about their willingness to disclose information, interviewees of the expert partner referred to these ‘legal knowledge-transfer clauses’ rather than to the equity structure. They stressed that the presence of specific contractual clauses protected them against possible opportunistic abuse by the expert partner. In addition, even in the case of the interfirm R&D relationship (MAT–USCOAT relationship), which was governed by a non-equity structure, the expert partner showed a high willingness to openly disclose their technological knowledge from the beginning. Again, interviewees
Downloaded from http://oss.sagepub.com at Universiteit Twente on December 18, 2007 © 2007 SAGE Publications. All rights reserved. Not for commercial use or unauthorized distribution.
Downloaded from http://oss.sagepub.com at Universiteit Twente on December 18, 2007 © 2007 SAGE Publications. All rights reserved. Not for commercial use or unauthorized distribution.
Conditions
Strategies
Organizational conditions Similarity in knowledge base Similarity in organizational structures and business practices Similarity of technological equipment
Relational conditions Expectations of long-term relationship
Formal conditions Equity structure Legal clauses regarding knowledge transfer
Outcomes
Novice partner’s ability to acquire and assimilate knowledge
Expert partner’s willingness to disclose knowledge
Initiation of Interfirm Knowledge Transfer
Figure 2. Initiation and Evolution of Interfirm Knowledge Transfer
Both partners exploit the originally transferred technologies for different product/market combinations
One partner introduces string of new technologies in R&D relationship, which subsequently can be exploited by other partner
Perceived technological complementarities decrease
OR
Perceived market threats are extensive
Perceived technological complementarities remain extensive
AND
Perceived market threats remain limited
Dissolution of Interfirm Knowledge Transfer
Continuation of Interfirm Knowledge Transfer
Faems et al.: Interfirm Knowledge Transfer in R&D Relationships
1717
1718
Organization Studies 28(11)
emphasized the presence of legal knowledge-transfer clauses to explain their transparent sharing of knowledge. These findings are in line with the argument of recent alliance contract researchers (e.g. Reuer et al. 2006) that micro-level governance mechanisms (i.e. specific contractual clauses) can be used as a formal design alternative for macro-level governance structures (i.e. equity ownership structures). Through defining expected behaviours as well as penalties in case of deviation from these behaviours, specific contractual clauses are able to mitigate the risk of opportunism in interfirm relationships. In sum, this study suggests that legal knowledge transfer clauses are an important condition for initiating knowledge transfer in interfirm R&D relationships, even when an equity governance structure is present. A second important facilitating condition refers to the relational aspect of an interfirm R&D relationship. Previous research (Chen 2004) has identified trust as a relational condition that influences partners’ willingness to disclose knowledge in R&D relationships. In accordance with the trust literature (e.g. Cummings and Bromiley 1996; Rousseau et al. 1998), Chen’s study (2004) referred to trust in quite general terms as positive expectations of the intentions and behaviours of the other partner. In our cases, however, the managers of the entrepreneurial partner pointed to one specific kind of positive expectations to explain their motivation in openly disclosing knowledge: positive expectations of a long-term duration of the relationship. These positive expectations about future cooperation seemed to increase the willingness to accept vulnerability, leading to behavioural transparency (Jones and George 1998; Lewicki and Bunker 1996). We therefore interpret the presence of positive expectations of a long-term relationship as a specific indicator of trust that motivates disclosure of knowledge in R&D relationships. Game theorists (e.g. Axelrod 1984; Oye 1986) refer to this bond between the future benefits a firm anticipates and its present actions as the ‘shadow of the future’. The longer this shadow of the future, the more likely partners prefer cooperative above non-cooperative behaviour (Parkhe 1993b). It may be that this condition is of particular importance to entrepreneurial companies. In contrast to established companies, entrepreneurial companies are more vulnerable to losing competitive learning races or to being acquired (Alvarez and Barney 2001). We therefore suggest that seeing the benefits of a long-term partnership is crucial for entrepreneurial companies before engaging in risk-taking behaviour such as the disclosure of sensitive knowledge. Ability to Acquire and Assimilate Knowledge
The second step of knowledge transfer, the acquisition and assimilation of knowledge, manifests itself in our study as an experiential process where engineers of the novice partner evaluate and internalize the technology of the expert partner through conducting experiments with the disclosed technology themselves. We also observed that the presence or absence of similar technical equipment substantially influenced the ability of the novice partner to acquire and assimilate the knowledge that was disclosed by the expert partner. For instance, in the MAT–USCOAT relationship, overlapping knowledge regarding diamond-like coating technology was initially lacking, and the partners had clearly different organizational structures and business practices. While, based on previous
Downloaded from http://oss.sagepub.com at Universiteit Twente on December 18, 2007 © 2007 SAGE Publications. All rights reserved. Not for commercial use or unauthorized distribution.
Faems et al.: Interfirm Knowledge Transfer in R&D Relationships
1719
research (e.g. Cohen and Levinthal 1990; Lane and Lubatkin 1998; Simonin 1999), one could expect that, in this case, the novice partner’s ability to acquire and assimilate knowledge would be limited, we observed a smooth acquisition and assimilation of the technology by the novice partner. Explaining this high ability to evaluate and internalize the disclosed knowledge, interviewees referred to the presence of similar technological equipment as a major facilitator. This same condition was also found in the two other R&D relationships. Based on these findings, we identify similarity of partners’ technological equipment as an additional condition that influences acquisition and assimilation of external knowledge. Nonaka and Takeuchi (1996) as well as Tyre and Von Hippel (1997) already argued that, where users’ skills are largely tacit, problem solvers need to observe actual patterns in the operating setting before they can understand and diagnose problems. Likewise, the role of participation — in order for learning to effectively occur — has been stressed by several scholars adopting a situated perspective on learning processes (e.g. Lave 1988; Lave and Wenger 1991). In line with this previous work, the physical environment and more specifically the tools and equipment one is using, can be seen as an inherent part of the setting that allows experiential learning processes to unfold. In other words, the presence of similar technological equipment constitutes an important complement of social interaction, facilitating the acquisition and assimilation of tacit knowledge. Evolution of Knowledge Transfer in Interfirm R&D Relationships
The model further specifies the evolution of interfirm knowledge transfer in terms of continuation and dissolution as well as the conditions that impact these two possible outcomes. In addition, we identify and assess two strategies that may increase the likelihood of continued interfirm knowledge transfer. Conditions that Influence Evolution of Interfirm Knowledge Transfer
In two of the three interfirm R&D relationships, partners continued knowledge transfer after the novice partner had acquired and assimilated the technology of the expert partner. In the MAT–USCOAT and MAT–FRCOAT relationships, technological meetings were regularly organized, resulting in bilateral information exchange of internal experiments; however, interfirm knowledge transfer quickly dissolved in the MAT–RES relationship. Our data point to the evolution of perceived technological complementarities between partners as one of the main drivers behind these different dynamics. In the case of dissolution, MAT quickly realized that RES’s DLC technology had limited potential value for its DLX/DLC development activities. As a result, perceived technological complementarities between partners decreased, which in turn reduced the motivation of MAT engineers to further engage in technological meetings with RES engineers. In contrast, in the two other interfirm R&D relationships, interviewees stressed that, as each partner experimented with the same technology in different settings, they both continued to generate knowledge that was potentially valuable for the other partner. This sustained presence of technological complementarities motivated continuation of interfirm knowledge transfer during regular technological meetings. The further evolvement of the MAT–USCOAT relationship provides
Downloaded from http://oss.sagepub.com at Universiteit Twente on December 18, 2007 © 2007 SAGE Publications. All rights reserved. Not for commercial use or unauthorized distribution.
1720
Organization Studies 28(11)
additional support for the importance of technological complementarities. After approximately two years of continued interfirm knowledge transfer in the MAT– USCOAT relationship, the transfer of knowledge gradually decreased. Clarifying this evolution, interviewees of both firms referred to the disappearance of perceived technological complementarities between them. In this case, the technological focus of the partners started to diverge, making it more difficult to find common ground and consequently reducing their motivation to engage in bilateral exchange of information about individual experiments. While the presence of technological complementarities seems to be a necessary condition to continue interfirm knowledge transfer, our data suggest that it is not a sufficient one. In both the MAT–USCOAT and the MAT–FRCOAT relationships, interviewees pointed to the absence of perceived market threats as an additional condition that seems to be necessary to continue knowledge transfer. It was stressed that, as both partners were focusing on clearly different applications for different markets, the risk of market competition between them was limited, allowing for the continuation of interfirm knowledge transfer. In contrast, market threats were perceived to be present in the MAT–RES relationship. In particular, MAT feared that interaction with the ‘scientists’ from RES would result in unintended knowledge spillovers to potential customers. These considerations added to the dissolution of the knowledge transfer between MAT and RES. From these findings, we argue that, in R&D relationships, knowledge transfer is likely to continue between partners if (1) perceived technological complementarities remain extensive and (2) perceived market threats remain limited. Finally, this study suggests that, in contrast to several alliance scholars (e.g. Larson 1992; Ring and Van de Ven 1994; Uzzi 1997) who argue that resilient trust stimulates partners to continue knowledge transfer, resilient trust is not a sufficient condition for the continuation of interfirm knowledge transfer in R&D relationships. For instance, in the MAT–USCOAT relationship, resilient levels of trust emerged both at the managerial level (one MAT manager referred to the emergence of a ‘friendship relationship’ with the CEO of USCOAT) and the operational level (one USCOAT engineer stressed a ‘collegial bond with engineers from the other partner’). However, despite this trusting relationship, MAT’s motivation to continue disclosing knowledge drastically decreased once the technological complementarities between the partners diminished. This latter finding corresponds with the argument of Madhok (1995a) that, when a long-term structural disequilibrium in contributions between collaborating partners emerges, the bonding properties of the social dimension (i.e. resilient trust) become inadequate, leading to a reduced willingness to invest time and effort in the relationship. We therefore argue that the presence of resilient trust may not compensate for the disappearance of technological complementarities to continue interfirm knowledge transfer in R&D relationships. Strategies that Facilitate Continuation of Interfirm Knowledge Transfer
Our model not only identifies two conditions that seem to be important to continue interfirm knowledge transfer, but it also suggests two strategies to organize
Downloaded from http://oss.sagepub.com at Universiteit Twente on December 18, 2007 © 2007 SAGE Publications. All rights reserved. Not for commercial use or unauthorized distribution.
Faems et al.: Interfirm Knowledge Transfer in R&D Relationships
1721
an R&D relationship such that market threats between partners remain limited and technological complementarities between partners remain present. A first strategy can be found in the work of Alvarez and Barney (2001). Examining multiple alliances between entrepreneurial and established firms, they observed some successful R&D relationships in which a clear division of responsibilities had emerged between the partners. In these cases, the entrepreneurial partner was responsible for continuously developing new technologies that were potentially valuable for the established partner; and the established partner, after acquiring and assimilating these new technologies, was responsible for commercially exploiting them. In other words, the entrepreneurial partner introduced a ‘string of new technologies’ (Alvarez and Barney, 2001: 145) in the relationship, which subsequently could be commercially exploited by the established partner. We acknowledge that this strategy has huge potential for continuation of unilateral interfirm knowledge transfer. As one partner continuously introduces new technologies in the relationship, technological complementarities between partners are constantly renewed. In addition, as only one partner is responsible for commercializing the technologies, there is no issue of market competition in such relationships. Alvarez and Barney (2001: 143) recognize that this strategy has one particular limitation: ‘the need for large investments in basic R&D at the entrepreneurial partner’. Introducing a continuous stream of new technologies requires broad and somewhat diversified investments in basic R&D. However, given the limited financial resources of entrepreneurial companies, such firms may be hesitant to conduct such costly and risky R&D activities. Our data point to a second limitation of this strategy. The MAT–RES relationship indicates that the motivation of the established partner to acquire a string of new technologies may be limited. In this case, partners had signed an R&D agreement, stipulating that MAT would fund basic R&D projects at RES. Theoretically, this R&D agreement gave RES the opportunity to come up with a string of new technologies that subsequently could be transferred to MAT. However, because of its focus on obtaining commercial revenues from its existing DLC/DLX technology, MAT was not eager to spend time and effort on discussing RES’s initiatives to come up with new technological opportunities. In other words, MAT’s intent to acquire and assimilate knowledge about RES’s explorative technological efforts became limited due to other priorities (Hamel 1991). As a result, these new R&D projects failed in renewing technological complementarities between RES and MAT. Next to this strategy of introducing a string of new technologies in the R&D relationship, our own data suggest another strategy to continue interfirm knowledge transfer. We observed that, in the two cases of continuation, both partners internally exploited the initially transferred knowledge for clearly different product/market combinations. As partners focused on different market applications, they perceived limited competitive threats between them. At the same time, the partners’ use of the same technology in different settings ensured that technological complementarities remained present. As a result, partners were motivated and able to continue knowledge transfer in terms of bilateral information exchange of internal experiments. However, we acknowledge that this strategy also faces some limitations. First, partners can only apply this strategy when the technology has the potential
Downloaded from http://oss.sagepub.com at Universiteit Twente on December 18, 2007 © 2007 SAGE Publications. All rights reserved. Not for commercial use or unauthorized distribution.
1722
Organization Studies 28(11)
to be exploited in different application domains and markets. Second, we observed that, when partners individually exploit the originally transferred technology for different product/market combinations, they are likely to start adjusting the technology through combining it with other kinds of knowledge. As the MAT–USCOAT relationship illustrates, such internal ‘knowledge transformation’ (Garud and Nayyar 1994) might cause a divergence of technological focus, which subsequently reduces technological complementarities between the partners. Third, we expect that this strategy is only feasible when sufficient entry barriers, restricting the ability of one partner to enter the product/market combinations of the other partner in the future (Robinson and McDougall 2001), are present. If entry barriers are not present, it would be fairly easy for one partner to switch to the other partner’s product/market combinations in the future. As a result, partners might remain concerned about competitive risks, discouraging knowledge transfer. In both the MAT–USCOAT and the MAT–FRCOAT relationship such entry barriers were perceived to be present. For instance, in the MAT–FRCOAT relationship, MAT interviewees stressed that FRCOAT was not likely to become active in industrial-wear applications because this entrepreneurial partner could not generate the necessary financial and operational resources to do so. At the same time, FRCOAT interviewees stated that, because of the particular nature of the racing industry, MAT faced huge entry obstacles.
Conclusion It is widely recognized that the formal design of relationships influences the process of interfirm knowledge transfer. Previous empirical research on interfirm knowledge transfer (e.g. Chen 2004; Mowery et al. 1996; Muthusamy and White 2005) has mainly focused on the distinction between equity and non-equity ownership structures as a formal condition that influences the effectiveness of interfirm knowledge transfer. This study, however, indicates that this well-researched dichotomy of equity/non-equity relationships does not cover the complex and numerous formal design alternatives to motivate disclosure of knowledge in R&D relationships. In particular, our data suggest that R&D relationships which do not possess an equity ownership structure can be as effective in motivating knowledge disclosure as long as detailed legal knowledge-transfer clauses are implemented in the contract. Based on these findings, we encourage scholars to examine the impact of both macrolevel (i.e. equity versus non-equity structure) and micro-level (i.e. specific contractual clauses) governance mechanisms on the potential for interfirm knowledge transfer. In addition, this study suggests that managers tend to combine formal and relational governance approaches to mitigate the risk of opportunism. While previous research has mainly conceptualized formal safeguards (e.g. legal clauses and/or financial hostages) and relational governance (e.g. building a trustful relationship) as alternative and even substitutive options (Dyer
Downloaded from http://oss.sagepub.com at Universiteit Twente on December 18, 2007 © 2007 SAGE Publications. All rights reserved. Not for commercial use or unauthorized distribution.
Faems et al.: Interfirm Knowledge Transfer in R&D Relationships
1723
and Singh 1998; Larson 1992; Madhok 1995b), we observed that partners negotiated detailed legal knowledge transfer clauses and, at the same time, spent time and effort in creating positive expectations about long-term cooperation. In our cases, this combinative approach turned out to be very successful in motivating expert partners to fully disclose their technological knowledge. This observation also seems to support recent research (e.g. Klein Woolthuis et al. 2005; Poppo and Zenger 2002), suggesting a complementary relationship between contracts and trust. This study also provides a richer perspective on the role of relation-specific investments or investments that have limited value outside the interfirm relationship (Dyer, 1997). It contributes to the transactional value perspective (Dyer 1997; Zajac and Olsen 1993) which, in contrast to transaction cost scholars (e.g. Klein et al. 1978; Williamson 1985), emphasizes the value creation properties instead of cost implications of relation-specific investments. In particular, our data indicate that the relation-specific investment of similar technological equipment for transferring knowledge between partners has value-generating properties. This particular type of investment allows creation of a physical environment in which acquisition and assimilation of tacit technological knowledge is facilitated. We therefore argue that, when collaborating partners aim to transfer technological knowledge, the advantages of investing in similar technological equipment (i.e. increased ability to acquire and assimilate knowledge) may outweigh its disadvantages (i.e. additional costs). At the same time, we acknowledge that the presence of similar technological equipment might increase the risk of unintended knowledge spillovers as the partner’s ability to evaluate and imitate knowledge is likely to be extensive. Further research examining the advantages and disadvantages of such relation-specific investments in different kinds of collaborative settings therefore seems to be necessary. A final contribution of this study is related to the stability of interfirm knowledge transfer in R&D relationships. Some alliance scholars (Khanna et al. 1998; Hamel 1991) indicate that, because of changing pay-off structures, interfirm knowledge transfer is an inherently unstable phenomenon. In contrast, other scholars (e.g. Larson 1992; Ring and Van de Ven 1994; Uzzi 1997) suggest that positive trust dynamics motivate partners to expand their relationship, triggering sustained continuation of knowledge transfer. Our study provides a more balanced view on the evolution of interfirm knowledge transfer. On the one hand, our data suggest that, after transfer of the initially desired knowledge, expected future pay-offs of the relationship can remain quite stable. In particular, we found that, as long as perceived technological complementarities remain present and perceived market threats remain limited, the perceived future gains of the relationship are likely to remain positive, allowing for the continuation of interfirm knowledge transfer. These two conditions can be realized through either the strategy of one partner introducing a string of new technologies or both partners exploiting the transferred knowledge for different product/market combinations. On the other hand, our data question the notion of resilient trust as a sufficient condition for continuation of inter-
Downloaded from http://oss.sagepub.com at Universiteit Twente on December 18, 2007 © 2007 SAGE Publications. All rights reserved. Not for commercial use or unauthorized distribution.
1724
Organization Studies 28(11)
firm knowledge transfer. We argue that the long-term social bonding properties of resilient trust may not compensate for short-term disturbances in expected pay-offs, which are caused by decreased technological complementarities and/or increased market threats. As a final reflection, we point to the need for additional research on the initiation and evolution of interfirm knowledge transfer in other organizational settings. In this study, the observed technological trajectory was one in which interdependencies between the different interfirm relationships remained limited as the core company (i.e. MAT) preferred to maintain bilateral links with the different partners. However, Dyer and Nobeoka (2000) have already pointed to the possibility of applying a more multilateral approach, where interactions among all participating partners are stimulated by the core company. Such alternative approach might require other conditions and processes to facilitate and continue knowledge transfer. We therefore stress the need for additional case studies in other contexts to develop a more general theory on the initiation and evolution of interfirm knowledge transfer. In sum, this study provided an enriched view on the process of interfirm knowledge transfer in R&D relationships. We hope that our suggestions for future research may encourage scholars to continue exploring the process of knowledge transfer in a variety of organizational settings. Note
References
Useful comments and suggestions on this paper have been provided by co-editor Raghu Garud; three anonymous reviewers; and René Bouwen, Koenraad Debackere, Anoop Madhok, Arie Rip and Luc Sels. The authors gratefully acknowledge the financial support of the Fund for Scientific Research, Flanders (FWO G.0286.06) and the Flemish government (Policy Research Centre ‘Entrepreneurship, Enterprises and Innovation’) for conducting this research. Earlier versions of this paper have been presented at the INIR meeting, Enschede, 16 June 2005 and at the Academy of Management meeting, Honolulu, 5–10 August 2005.
Alvarez, Sharon A., and Jay B. Barney 2001 ‘How entrepreneurial firms can benefit from alliances with large partners’. Academy of Management Executive 15: 139–148. Axelrod, Robert 1984 The evolution of cooperation. New York: Basic Books. Blau, Peter 1964 Exchange and power in social life. New York: Wiley. Chen, Chung-Jen 2004 ‘The effects of knowledge attribute, alliance characteristics, and absorptive capacity on knowledge transfer performance’. R&D Management 34: 311–321. Cohen, Wesley M., and Daniel A. Levinthal 1990 ‘Absorptive capacity: A new perspective on learning and
innovation’. Administrative Science Quarterly, 35: 128–152. Cummings, Larry L., and Philip Bromiley 1996 ‘The organizational trust inventory (OTI): Development and validation’ in Trust in organizations: Frontiers of theory and research. Roderick M. Kramer and Tom R. Tyler (eds), 302–320. Thousand Oaks, CA: Sage. De Laat, Paul 1997 ‘Research and development alliances: Ensuring trust by mutual commitments’ in The formation of inter-organizational networks. Mark Ebers (ed.), 146–173. Oxford: Oxford University Press. Doz, Yves 1996 ‘The evolution of cooperation in strategic alliances: Initial conditions or learning processes?’ Strategic Management Journal 17: 55–83.
Downloaded from http://oss.sagepub.com at Universiteit Twente on December 18, 2007 © 2007 SAGE Publications. All rights reserved. Not for commercial use or unauthorized distribution.
Faems et al.: Interfirm Knowledge Transfer in R&D Relationships Doz, Yves, and Gary Hamel 1997 ‘The use of alliances in implementing technology strategies’ in Managing strategic innovation and change: A collection of readings. Michael L. Tushman and Philip Anderson (eds), 556–580. New York: Oxford University Press. Dyer, Jeffrey H. 1997 ‘Effective interfirm collaboration: How firms minimize transaction costs and maximize transaction value’. Strategic Management Journal 18: 535–556. Dyer, Jeffrey H., and Harbir Singh 1998 ‘The relational view: Cooperative strategy and sources of interorganizational competitive advantage’. Academy of Management Review 23: 660–679. Dyer, Jeffrey H., and Kentaro Nobeoka 2000 ‘Creating and managing a highperformance knowledge-sharing network: The Toyota case’. Strategic Management Journal, 21: 345–367. Eisenhardt, Kathleen M. 1989 ‘Building theories from case study research’. Academy of Management Review 14: 532–550. Faems, Dries, Bart van Looy, and Koenraad Debackere 2005 ‘Interorganizational collaboration and innovation: Toward a portfolio approach’. Journal of Product Innovation Management 22: 238–250. Garud, Raghu 1997 ‘On the distinction between knowhow, know-what and know-why’ in Advances in strategic management. Anne Huff and Jim Walsh (eds), 14: 81–101. Greenwich, CT: JAI Press. Garud, Raghu, and Praveen R. Nayyar 1994 ‘Transformative capacity: Continual structuring by intertemporal technology transfer’. Strategic Management Journal 15: 365–385. Hagedoorn, John 2002 ‘Inter-firm R&D partnerships: An overview of major trends and patterns since 1960’. Research Policy 31: 477–492. Hamel, Gary 1991 ‘Competition for competence and interpartner learning within
1725
international strategic alliances’. Strategic Management Journal 12 (Summer Special Issue): 83–103. Inkpen, Andrew C., and Wang Pien 2006 ‘An examination of collaboration and knowledge transfer: China-Singapore Suzhou Industrial Park’. Journal of Management Studies 43: 779–811. Jones, Gareth R., and Jennifer M. George 1998 ‘The experience and evolution of trust: Implications for cooperation and teamwork’. Academy of Management Review 23: 531–546. Kale, Prashant, Harbir Singh, and Howard Perlmutter 2000 ‘Learning and protection of proprietary assets in strategic alliances: Building relational capital’. Strategic Management Journal 21: 217–237. Khanna, Tarun, Ranjay Gulati, and Nitin Nohria 1998 ‘The dynamics of learning alliances: competition, cooperation, and relative scope’. Strategic Management Journal 19: 193–210. Klein, Benjamin, Robert G. Crawford, and Armen A. Alchian 1978 ‘Vertical integration, appropriable rents, and the competitive contracting process’. Journal of Law and Economics 21: 297–326. Klein Woolthuis, Rosalinde, Bas Hillebrand, and Bart Nooteboom 2005 ‘Trust, contract and relationship development’. Organization Studies 26: 813–840. Kogut, Bruce, and Udo Zander 1992 ‘Knowledge of the firm, combinative capabilities and the replication of technology’. Organization Science 3: 383–397. Kvale, Steinar 1996 InterViews: An introduction to qualitative research interviewing. London: Sage Publications. Lam, Alice 1997 ‘Embedded firms, embedded knowledge: Problems of collaboration and knowledge transfer in global cooperative ventures’. Organization Studies 18: 973–997.
Downloaded from http://oss.sagepub.com at Universiteit Twente on December 18, 2007 © 2007 SAGE Publications. All rights reserved. Not for commercial use or unauthorized distribution.
1726
Organization Studies 28(11) Lane, Peter J., and Michael Lubatkin 1998 ‘Relative absorptive capacity and interorganizational learning’. Strategic Management Journal 19: 461–477.
reexamination of recent evidence’. Academy of Management Journal 40: 189–204.
Langley, Ann 1999 ‘Strategies for theorizing from process data’. Academy of Management Review 24: 691–710.
Mowery, David C., Joanne E. Oxley, and Brian S. Silverman 1996 ‘Strategic alliances and interfirm knowledge transfer’. Strategic Management Journal 17 (Winter Special Issue): 77–91.
Larson, Andrea 1992 ‘Network dyads in entrepreneurial settings: A study of the governance of exchange relationships’. Administrative Science Quarterly 37: 76–104.
Muthusamy, Senthil K., and Margaret A. White 2005 ‘Learning and knowledge transfer in strategic alliances: A social exchange view’. Organization Studies 26: 415–441.
Larsson, Rikard, Lars Bengtsson, Kristina Henriksson, and Judith Sparks 1998 ‘The interorganizational learning dilemma: Collective knowledge development in strategic alliances’. Organization Science 9: 285–305.
Nonaka, Ikujiro, and Hirotaka Takeuchi 1996 The knowledge creating company: How Japanese companies create the dynamics of innovation. Oxford: Oxford University Press.
Lave, Jean 1988 Cognition in practice. New York: Cambridge University Press. Lave, Jean, and Etienne Wenger 1991 Situated learning. New York: Cambridge University Press. Leonard-Barton, Dorothy 1990 ‘A dual methodology for case studies: Synergistic use of a longitudinal single site with replicated multiple studies’. Organization Science 1: 248–266. Lewicki, Roy J., and Barbara B. Bunker 1996 ‘Developing and maintaining trust in work relationships’ in Trust in organizations: Frontiers of theory and research. Roderick M. Kramer and Tom R. Tyler (eds), 114–139. Thousand Oa, CA: Sage. Madhok, Anoop 1995a ‘Revisiting multinational firms’ tolerance for joint ventures: A trustbased approach’. Journal of International Business Studies 26: 117–137. Madhok, Anoop 1995b ‘Opportunism and trust in joint venture relationships: An exploratory study and a model’. Scandinavian Journal of Management 11: 55–74. Miller, C. Chet, Laura B. Cardinal, and William H. Glick 1997 ‘Retrospective reports in organizational research: A
Oye, Kenneth 1986 Cooperation under anarchy. Princeton, NJ: Princeton University Press. Parkhe, Arvind 1993a ‘“Messy” research, methodological predispositions, and theory development in international joint ventures’. Academy of Management Review 18: 227–268. Parkhe, Arvind 1993b ‘Strategic alliances structuring: A game theoretic and transaction cost examination of interfirm cooperation’. Academy of Management Journal 36: 794–829. Pentland, Brian T. 1999 ‘Building process theory with narrative: From description to explanation’. Academy of Management Review 24: 711–724. Pettigrew, Andrew M. 1990 ‘Longitudinal field research on change: Theory and practice’. Organization Science 1: 267–292. Poole, Marshall S., Andrew H. van de Ven, Kevin Dooley, and Michael E. Holmes 2000 Organizational change and innovation processes: Theory and methods for research. Oxford: Oxford University Press. Poppo, Laura, and Todd Zenger 2002 ‘Do formal contracts and relational governance function as substitutes or complements?’ Strategic Management Journal 23: 707–725.
Downloaded from http://oss.sagepub.com at Universiteit Twente on December 18, 2007 © 2007 SAGE Publications. All rights reserved. Not for commercial use or unauthorized distribution.
Faems et al.: Interfirm Knowledge Transfer in R&D Relationships Reuer, Jeffrey J., Africa Arin–o, and Thomas Mellewigt 2006 ‘Entrepreneurial alliances as contractual forms’. Journal of Business Venturing 21: 306–325. Ring, Peter S. 1997 ‘Processes facilitating reliance on trust in inter-organizational networks’ in The formation of interorganizational networks. Marc Ebers (ed.), 113–145. Oxford: Oxford University Press. Ring, Peter S., and Andrew H. van de Ven 1994 ‘Developmental processes of cooperative interorganizational relationships’. Academy of Management Review 19: 90–118. Robinson, Kenneth C., and Patricia P. McDougall 2001 ‘Entry barriers and new venture performance: A comparison of universal and contingency approaches. Strategic Management Journal 22: 659–685. Rousseau, Denise M., Sim B. Sitkin, Ronald S. Burt, and Colin Camerer 1998 ‘Not so different after all: A crossdiscipline view of trust’. Academy of Management Review 23: 393–404. Simonin, Bernard L. 1999 ‘Ambiguity and the process of knowledge transfer in strategic alliances’. Strategic Management Journal 20: 595–623. Tsang, Eric W. K. 2000 ‘Transaction cost and resource-based explanations of joint ventures:
1727
A comparison and synthesis’. Organization Studies 21: 215–242. Tyre, Marcie J., and Eric von Hippel 1997 ‘The situated nature of adaptive learning in organizations’. Organization Science 8: 71–83. Uzzi, Brian 1997 ‘Social structure and competition in interfirm networks: The paradox of embeddedness’. Administrative Science Quarterly 42: 35–67. Williamson, Oliver E. 1985 The economic institutions of capitalism. New York: Free Press. Williamson, Oliver E. 1991 ‘Comparative economic organization: The analysis of discrete structural alternatives’. Administrative Science Quarterly 36: 269–296. Yin, Robert K. 2003 Case study research, design and methods, 3rd edn. Newbury Park, CA: Sage. Zahra, Shaker A., and Gerard George 2002 ‘Absorptive capacity: A review, reconceptualization, and extension’. Academy of Management Review 27: 185–203. Zajac, Edward J., and Cyrus P. Olsen 1993 ‘From transaction cost to transactional value analysis: Implications for the study of interorganizational strategies’. Journal of Management Studies 30: 131–145.
Dries Faems
Dries Faems is a Postdoctoral Researcher at the Research Centre of Organisation Studies at the Katholieke Universiteit Leuven. He became Doctor in Applied Economics (Katholieke Universiteit Leuven) in 2006 with a thesis entitled ‘Collaboration for innovation: Processes of governance and learning in R&D alliances’. He has published articles in the areas of alliance management and HR management. His current research focuses on governance processes in R&D alliances and performance implications of alliance portfolios. Address: Katholieke Universiteit Leuven, Faculty of Economics and Applied Economics, Naamsestraat 69, 3000 Leuven, Belgium. Email:
[email protected]
Maddy Janssens
Maddy Janssens is a Full Professor at the Research Centre of Organisation Studies at the Katholieke Universiteit Leuven, having received her PhD in Psychology in 1992. She studied at Northwestern University, and was a visiting faculty at INSEAD during 1996 and at the Stern School of Business, New York University, during 1999. She has
Downloaded from http://oss.sagepub.com at Universiteit Twente on December 18, 2007 © 2007 SAGE Publications. All rights reserved. Not for commercial use or unauthorized distribution.
1728
Organization Studies 28(11)
published articles in the areas of intercultural management, diversity in organizations, and critical perspectives on HRM. Her research interest focuses on ways in which organizations deal with difference. Address: Katholieke Universiteit Leuven, Faculty of Economics and Applied Economics, Naamsestraat 69, 3000 Leuven, Belgium. Email:
[email protected] Bart van Looy
Bart van Looy is a Professor in Innovation Management at the Faculty of Economics and Applied Economics, Katholieke Universiteit Leuven, Belgium. He is responsible for the research activities of INCENTIM, a research division at the university and is involved in Steunpunt O&O Indicators (Innovation Policy Research Unit funded by the Flemish government) where he is responsible for technology studies. His current research focuses on organizing innovation at the firm and regional level and the role of entrepreneurial universities within innovation systems. Address: Katholieke Universiteit Leuven, Faculty of Economics and Applied Economics, Naamsestraat 69, 3000 Leuven, Belgium. Email:
[email protected]
Downloaded from http://oss.sagepub.com at Universiteit Twente on December 18, 2007 © 2007 SAGE Publications. All rights reserved. Not for commercial use or unauthorized distribution.
