Governing MR&D alliances: formal and social mechanisms

(1)

Governing MR&D alliances:

formal and social mechanisms

By Laurens Visser,

MSc BA: Strategic Innovation Management

University of Groningen, Faculty of Business Economics

January, 2016

Petrus Campersingel 191A 9713AL, Groningen (06) 81 72 41 76 l.visser.8@student.rug.nl Student number: 2200406

Word count: 12.824

Abstract

(2)

2

Introduction

It is widely known in literature, that strategic alliances have a high failure rate. Studies show that alliances have failure rates of 50 per cent (Porter, Goold & Luchs, 1996; Dyer, Kale & Singh, 2001). In alliance literature there is an important distinction between dyadic and multi-partner alliances. Multi-partner alliances have not received a lot of attention in the literature (Das & Teng, 2002; Thorgren, Wincent & Eriksson, 2010; Li, Eden, Hitt, Ireland & Garret, 2012). Dyadic alliances exist out of two partners, whereas multi-partner alliances exist out of more than two partners (Das & Teng, 2002; Lavie, Lechner & Singh, 2007). Studies on alliances show that about one third of the alliances have more than two partners (Gulati 1995, Gulati & Singh 1998). Furthermore, those multi-partner alliances are visible across different industries, such as the biopharmaceuticals, automotive sector and the new materials sector.

The distinction of dyadic and multi-partner alliances is important for several reasons. First, collaboration with more than two partners increases the complexity of alliance design and governance (Doz & Hamel, 1998). Second, the pattern of exchange between dyadic and multi-partner alliances is different. In dyadic alliances there is a direct link between the multi-partner and the exchange, whereas in multi-partner alliances exchange is generalized (Das & Teng, 2002). As the exchange in multi-partner alliances is generalized, there is no direct reciprocity amongst partners. Therefore, several challenges exist in multi-partner alliances. For instance, the risk of opportunism is higher (Takahashi, 2000; Das & Teng, 2002; Doz & Hamel, 1998). Multi-partner alliances are thus important to research, as they include a profound base of challenges as compared to dyadic alliances.

(5)

5 sharing is challenging to govern in R&D alliances, as too much knowledge sharing can lead to undesired knowledge leakage (Oxley & Sampson, 2004).

I use two theoretical streams to explain the governance in MR&D alliances. On the one hand, Transaction Cost Economics (TCE) addresses formal governance mechanisms in R&D alliances. In TCE, formal safeguards are known for preventing opportunistic behavior, such as knowledge leakage (Williamson, 1979; Parkhe, 1993; Das & Teng, 1998, 2001). However, research on TCE in alliances is solely done on dyadic alliances (Faems, Janssens, Madhok & Van Looy, 2008). We cannot assume that the governance of dyadic and partner alliances is the same, as in multi-partner alliances more challenges exist. Another limitation of TCE is that is mainly focused on equity alliances (Li, Eden, Hitt & Ireland, 2008; Pisano, 1989; Oxley & Sampson, 2004; Li et al., 2012; Garcia-Canal, Valdes-Llaneza & Arino, 2003), whereas I focus on non-equity alliances. The second theoretical perspective in this study is focused on the social aspect of cooperation. Social Exchange Theory (SET) defines social exchange mechanisms (generalized reciprocity, social sanctions and a cooperative macro culture) that reduce the challenges of multi-partner alliances (Das & Teng, 2002). However, previous studies focus on social exchange, without a single focus on R&D alliances (Das & Teng, 2002; Thorgren et al., 2010).

Both TCE and SET consist of certain limits, and therefore provide an incomplete picture (Poppo & Zenger, 2002; Faems et al., 2008; Das & Teng, 1998). TCE is too much focused on the economics of cooperation, and underestimates the value of human interaction (Faems et al., 2008). SET is criticized for being too time-consuming and costly to develop, and overestimates the value of human interaction (Das & Teng, 1998; Faems et al., 2008). By including both theoretical streams, the study provides a more complete view on the governance of MR&D alliances. This study is among the first to use TCE and SET to explain governance of MR&D alliances. Based on the two theoretical streams, the following research question is defined:

How are formal governance and social exchange mechanisms applied effectively in multi-partner Research & Development alliances?

(6)

6 18 partners (both public and private) and aims to develop scientific models for the biogas industry. The scientific models are developed with the aim of creating sustainability in the biogas industry. The study makes some key contributions to the governance of MR&D alliances. First, flexible contracts and steering committees are effective formal governance mechanisms. Second, exchange relationships in MR&D alliances are improved by generalized reciprocity, theme meetings, and a cooperative macro culture and expectation management.

The article is structured as follows. First, a literature review is presented, including definitions of MR&D alliances, its benefits and challenges and the formal governance and social exchange mechanisms. Second, I describe the research method, case information data collection and data-analysis. Third, I describe the findings of the study. Fourth, I discuss the findings with existing literature and build propositions. Fifth, the research concludes with the key contributions for both theory and practice, and discusses the specific limitations and directions for future research.

Literature review

In this section the theoretical background is described. First, the literature on MR&D alliances, its definitions, benefits and challenges are explored. Second, the formal governance mechanisms are described. Third, the social exchange mechanisms are discussed.

MR&D Alliances

To define a MR&D alliance properly, the following four distinctions are important. First, partners in alliances can be competitors or non-competitors, the former and latter present two separate fields of literature (Mitchell, Dussauge, Garrette, 2002).1 Second, R&D alliances can be placed in two categories, either equity or non-equity (contractual) alliances (Hagedoorn, 2002).2 Third, alliances can be between firms, or consist of public and private partners (cf. Bruneel, d’Este & Salter, 2010; Li et al., 2012). 3 Fourth, alliance scope is important to take into account. Alliances

1

In this study I focus on an alliance with non-competing firms.

2

In this study, I focus on non-equity alliances.

3

(7)

7 can be either pure R&D alliances, or combine several activities, for instance, joint production, marketing and R&D (Oxley & Sampson, 2004).4

A MR&D alliance is defined as a collective, voluntary organizational association that interactively engages its multiple members in activities, such as collaborative research and development (Lavie et al., 2007). This research focuses on the cooperation on R&D, which is an important vehicle for acquiring and leveraging technological capabilities (Oxley & Sampson, 2004). In order to have a successful R&D alliance, knowledge has to be shared. The balance between knowledge sharing and protection is rather difficult. As undesired knowledge exchange can also become part of the R&D alliance (Li et al., 2012; Oxley & Sampson, 2004). There are various terms of importance in this study: R&D consortium (Mothe & Quelin, 2001; Doz et al., 2000), multilateral alliance (Doz et al., 2000) and constellations (Das & Teng, 2002). An important similarity amongst all the described terms is that they all involve two or more partners.

Dyadic and multi-partner alliances

In the alliance literature, the distinction of dyadic and multi-partner alliances is neglected (Das & Teng, 2002; Thorgren et al., 2010). An important difference between dyadic and multi-partner alliances can be explained via SET. Social exchange can be defined as the “voluntary actions of individuals that are motivated by the returns they are expected to bring and typically in fact bring from others” (Blau, 1964). Exchanges can be divided in restricted exchanges and generalized exchanges (Ekeh, 1974). The exchange in dyadic alliances shows a direct link between the process of giving and taking. In multi-partner alliances more than two partners are involved, consequently generalized exchanges occur in multi-partner alliances (Das & Teng, 2002; Thorgren et al., 2010). There is no direct reciprocity amongst partners, in generalized exchange. Therefore, in multi-partner alliances there is no direct link between the process of giving and taking. To illustrate: What A receives from B is not contingent on what A gives to B (Das & Teng, 2002; Thorgren et al., 2010).

Because of the different patterns of social exchange, multi-partner alliances are on several aspects more challenging than dyadic alliances. Three specific challenges can be defined for multi-partner alliances. First, trust building is difficult, as multi-partners have to trust that all multi-partners contribute equally. Second, conflicts might arise, due to the risk of free-riding. For instance,

(8)

8 partners might abuse trust or ignore reciprocity, because of the absence of a direct link (Das & Teng, 2002). Also, because of the absence of direct reciprocity, there is a higher risk of opportunistic behavior in MR&D alliances (Das & Teng, 2002; Lavie et al., 2007; Thorgren et al., 2010; Li et al., 2012). Third, as the exchange process is less transparent and monitoring is difficult, coordination is difficult and costly (Das & Teng, 2002). An additional difference between dyadic and multi-partner alliances is that both the alliance design and governance is more complex in multi-partner settings (Doz & Hamel, 1998).

Benefits and challenges

Aside from the increased complexity and challenges in multi-partners alliances, this section elaborates on the benefits and challenges existing in alliances. First, numerous benefits could be mentioned on why firms form strategic alliances. To illustrate, firms form strategic alliances for numerous reasons: pursuing strategic objectives (Hamel, Doz & Prahalad, 1989), having similar characteristics or operating in a similar industry, lowering the costs of R&D (Hagedoorn, 2002), conducting several projects together and sharing the results (Doz & Hamel, 2000; Ouchi & Bolton, 1988) and creating knowledge spillovers (Teece, 1986). Second, the challenge of opportunism is visible across all alliances (Das & Teng, 1998). Furthermore, some additional challenges exist in alliances with private and public partners (Bruneel et al., 2010), as private and public partners differ in their orientations and the way transactions are done (Bruneel et al., 2010). Public partners usually have a long term focus and private partners have a short-term focus (Cyert & Goodman, 1997).

Governance of MR&D alliances

(9)

9

Formal governance mechanisms

The identified complexity of alliance design and governance for MR&D alliances can be addressed by TCE. Two key elements of TCE are incentive alignment and administrative control (Williamson, 1991). TCE is the theory that explains how governance structures can be applied in alliances. This section is divided in three parts. First, I describe the formal governance from previous work on dyadic alliances. Second, I review the work on multi-partner alliances and governance structures (Li et al., 2012; Garcia-Canal et al., 2003). Third, I recap on the existing work on governance and point to the specific limitations.

Dyadic alliances

Alliances can be divided in equity and non-equity alliances. Pisano (1989) studies R&D collaboration in the biotechnology industry and compares the likelihood of pursuing an equity alliance or a non-equity alliance. Pisano (1989) finds that R&D collaboration usually involves a more equity based structure. Also, Li et al. (2008) studies R&D alliances in the high technology industry and find more equity alliances. Oxley & Sampson (2004) study the governance of R&D alliances in the telecommunications equipment and the electronics sector. They find that the scope of activities of alliances influences the choice between equity and non-equity alliance, when the scope of activities (e.g. joint R&D and joint manufacturing) increases, an equity alliance is more likely. Equity alliances provide effective governance of R&D alliances as they mitigate the potential knowledge leakage by aligning incentives and enhancing monitoring (Pisano, 1989; Oxley & Sampson, 2004).

(10)

10 Second, Faems et al. (2008) study the nature of contracts in exploratory R&D alliances. From a TCE perspective this is interesting, as Faems et al. (2008) state that most researchers focused on the number of contractual clauses. In R&D alliances contracts are rather complex to specify, monitor, and enforce, as lot of tacit and complex knowledge is exchanged in R&D alliances (Oxley & Sampson, 2004). Faems et al. (2008) specifies broad and narrow contractual interfaces. Broad contractual interfaces consist of an overlapping task division, obligation to share information and mechanisms for performance and behavior monitoring. On the other hand, a narrow contractual interface has a mutually exclusive task division, absence of obligations to share information and only performance oriented monitoring mechanisms. According to Faems et al. (2008), a broad contractual interface is most effective in R&D alliances. This finding does not close the book on contractual interfaces as the findings of Faems et al. (2008) should be studied across different settings as their research is not conclusive.

Contracts described such as above, provide a one-sided picture, as contractsare unavoidably incomplete (Williamson, 1996). Contracts are able to foresee and provide protection against hazards, but unable to anticipate to future contingencies. A new possible solution in literature, to enhance the adaptive limits of contracts, might be a steering committee (Reuer & Devorakonda, 2015). Steering committees can fulfill an important coordinating function in alliances (Reuer & Devorakonda, 2015). To date, there is no empirical evidence that understands the structure and functioning of a steering committee as a governance mechanism for alliances (Reuer & Devorakonda, 2015; Contractor & Lorange, 2014).

Multi-partner alliances

(11)

11 governance structures (Garcia-Canal et al., 2003; Oxley, 1997; Pisano, 1989; Reuer & Devorakonda, 2015).

Second, Li et al. (2012) study the governance of dyadic and MR&D alliances and reveal the difference between those two regarding the application of governance mechanisms. In effect, their results are similar to Garcia-Canal et al. (2003). Along with other researchers (e.g. Pisano, 1989), Li et al. (2012) state that equity based governance structures are more effective in protecting against opportunism than non-equity based structures. However, their work introduces some questions. Li et al. (2012) studies trilateral alliances and they refer to the fact that the complexity of governance might increase when more than three partners are involved. Furthermore, it is not clear how the governance structure of non-equity alliances should address the existence of multiple partners.

Research limitations

(12)

12 Table 1: Overview of proposed formal governance mechanisms

Social exchange mechanisms

To recap, three important areas of managing multi-partner alliances exist: trust building, conflict resolution and coordination (Doz & Hamel, 1998). As SET describes inter-organizational relationships, relational governance mechanisms cannot be ignored. SET is defined as the voluntary actions of individuals that are motivated by the returns they are expected to bring and typically in fact bring from others (Blau, 1964). Relational governance can be defined as interfirm exchange which includes significant relationship-specific assets, combined with a high level of inter-organizational trust (Zaheer & Venkatraman, 1995). To illustrate, relational mechanisms can enhance the building of trust, which is important in SET (Dyer & Singh, 1998). Although SET might have a more specific focus on reciprocity, I do expect that relational mechanisms can help enhance the understanding of social exchange mechanisms in MR&D alliances.

(13)

13 in multi-partner alliances is described. Third, I recap on the existing work of social exchange and point to the specific limitations.

Dyadic alliances

Thorgren et al. (2010) state that most of the research exploring trust is focused on dyadic alliances (e.g. Das & Teng, 1998; Gulati, 1995). Trust plays an important role in R&D alliances (e.g. Faems et al., 2008; Li et al. 2008) and alliances with public and private partners (Bruneel et al., 2010). Trust entails the willingness of managers to accept the vulnerability in partnerships (Rousseau, Sitkin, Burt & Camerer, 1998). Das & Teng (1998) develop four different trust building mechanisms: risk taking, equity preservation, communication and interfirm adaptation. First, if firms gradually start taking risks, they will start trusting each other. Second, firms should have a sense of fairness, with regard to the exchanges in the alliances. For example, that partner firms contribute an equal amount of resources. Third, communication is important for three reasons: communication is an essential prerequisite of alliances, a constant information flow leads to a more trusting environment and communication provides a basis for continued interaction. Fourth, interfirm adaptation involves adapting to the needs of the alliance, it is believed that acting in the interests of the alliance creates trust (Das & Teng, 1998).

Mohr & Spekman (1994) study successful partnerships and specified several key attributes of successful partnerships. In line with Das & Teng (1998), trust and communication is important in alliances. Communication quality is referred to the extent to which communication is timely, accurate, adequate, complete and credible (Mohr & Spekman, 1994). Furthermore, joint problem solving grievances should be aired and underlying issues are brought to the surface (Mohr & Spekman, 1994). Joint problem solving is interesting, as it can be seen as an effective conflict resolution technique (Mohr & Spekman, 1994).

(14)

14 emphasizes the need of regularly interaction between partners, enabling familiarization of each other’s cultures (Das & Teng, 1998).

Multi-partner alliances

Below, I describe the social exchange mechanisms that are identified for multi-partner alliances. First, generalized reciprocity can help build trust, which involves that all partners of an alliance quid pro quo contribute to the alliance (Das & Teng, 2002). Thorgren et al. (2010) study the trust building in multi-partner alliances and find that the effort partners take in establishing generalized exchange efforts, leads to actively trust building. However, their research is not focused on specific types of multi-partner alliances, such as R&D consortia (Thorgren et al., 2010).

Conflicts can arise in multi-partner alliances when partners abuse trust or ignore reciprocity. Due to the existence of multiple partners, collective actions are desirable when social exchange is violated (Das & Teng, 2002). In conflict resolution, monitoring and dissemination of information on the credibility of alliance partners is important (Hagen & Choe, 1998). When the information of a potential free-riding partner is recognized in the alliance, social sanctions can be taken. In effect, social sanctions are defined as the collective punishment of partners that deviate from the norms of behavior in the alliance. Social sanctions are an effective conflict resolution technique as it may lead to the exclusion of the violating partner (Coleman, 1994).

Coordination is a key component of successful partnerships as coordination is costly and difficult. Therefore, Das & Teng (2002) refer to the need of developing a cooperative macro culture. A macro culture is defined as the existence of shared values and beliefs, which are beneficial for the stability of exchange relationships (Nord, 1969). A cooperative macro culture can improve the social exchange in two ways. First, coordination costs can be reduced, by increasing the ease of exchange among parties (Das & Teng, 2002; Jones, Hesterly & Borgatti, 1997). Second, a cooperative macro culture facilitates generalized reciprocity and social sanctions. Firms will be more likely to accept indirect reciprocity when they have shared values and beliefs (Ekeh, 1974). In addition, social sanctions are more likely to be pursued collectively, as information flow is effective in the alliance.

Research limitations

(15)

15 relational mechanisms in dyadic alliances that might enhance social exchange, consists of two limitations. First, it should be investigated whether or not they are applicable to multi-partner settings. Second, it is unclear whether or not the presented mechanism are useful for R&D alliances. Similar, the proposed mechanisms for multi-partner alliances need empirical validation in the context of R&D (Das & Teng 2002; Thorgren et al., 2010).

Table 2. Overview of proposed social exchange mechanisms

Methodology

(16)

16

Research method

To study the business phenomenon, which is the governance of MR&D alliances, a single case study is done. It is a useful method to compare existing theory with practice and thereby complement theory (Eisenhardt, 1989). A case study is a valid method to study the phenomenon of interest as there is a low amount of case studies on alliance governance in R&D alliances (Faems et al., 2008). The process of building theories from cases, includes four basic steps (Eisenhardt, 1989): 1. an existing business phenomenon which is insufficiently explained in academic literature; 2. Study of the business phenomenon via either a single case study, or multiple cases; 3. comparison of the findings with the literature and 4. building propositions, when findings contribute to existing literature.

Case description

Case selection

In order to study the phenomenon of MR&D alliances, the ‘GreenGas Alliance’(pseudonym)5 is studied. The GreenGas Alliance was an agreement between 12 firms and 6 knowledge/research institutions, with the purpose of doing collective research and development in the biogas industry. An important motive for forming the GreenGas Alliance was the existence of public funding. The alliance was active in the Northern part of The Netherlands for a period of five years. The alliance was structured around five distinctive themes. Within these themes different projects were done. This study focused on one theme of the alliance, which focused on one project, referred to as theme A. The project of theme A was to develop scientific models for the biogas industry. The scientific models were developed with the aim of creating sustainability in the biogas industry. Alliances are often formed to conduct several projects together, therefore the terms ‘project’ and ‘alliance’ are used intertwined in describing the results. In this theme six partners were active, for a detailed description of each partner, see Appendix I. The GreenGas Alliance was selected, as it included six partners and can thus be considered as a MR&D alliance.

The alliance was not only structured around themes, also the theme itself was structured in a certain way (Figure 1). The alliance was initiated by the scientific director of a foundation (Partner Z). This foundation hired an overall project manager to manage the alliance.

5

(17)

17 Consequently, the overall project manager divided the alliance in themes and assigned theme leaders (project managers of the themes). In addition, another important figure to mention is the overall coordinator of the foundation, which also fulfilled important coordinating tasks in theme A of the GreenGas Alliance (Table 3).

In theme A three different types of partners were identified: research institutions, knowledge institutions and firms (Appendix). Partner R1 was responsible for conducting the research in this alliance. Three researchers of partner R1 worked on three different (connected) studies, in order to be able to produce scientific models for the biogas industry. Partner R2, the university, was responsible for supervising the researchers. Partner K1 was responsible for developing an ICT model, which was able to make a simulation of the produced research of partner R1. The other partners (K2, F1 and F2) had the role to deliver data as input and give advice based on their specific expertise. It can thus be derived that the two most important partners in this alliance were partner R1 and partner K1. In effect, those parties created the final products in this theme of the alliance.

Figure 1. Structure theme A. GreenGas Alliance

The biogas industry

(18)

18 2014 the consumptions increased to 5,6% and the goal for 2020 is to use 14 % of renewable energy in 2020. Biogas is also increasing as sustainable energy source, as it grew in one year, from 5,8% to 6,5%.6 Furthermore, in light of the current increase of partnering for cleaner energy industries7, the biogas industry is an interesting industry to study.

Data collection

The use of multiple data sources infers a high quality case study (Yin, 1994). Therefore, I triangulated data of conducted interviews (Table 3) and documents (e.g. contract and project plan).

Interviews

In total seven face to face in-depth interviews were held. Two different types of managers were interviewed. First, the interviewees from Partner Z and the interviewees from Partner R1, fulfilled roles on the managerial and operational level. Second, the other interviewees were active in the operational level of the alliance. Another important aspect of the partners, was the time period the interviewee was involved in the alliance, as some interviewees were not involved throughout the whole five years (Table 3). Due to time and/or locational limits, three interviews were held via a telephone call and one interview was conducted via e-mail. The interviews lasted between 45 and 60 minutes and were recorded with an audiotape recorder. Transcriptions were conducted within 24 hours after the interviews, in order to create as rich information from the transcriptions as possible. In the orienting stage, two meetings with the overall coordinator were planned and one meeting with the (external) project manager was planned (Table 3). To prepare for these interviews, available secondary documents (e.g. final report, PhD thesis) were studied. The purpose of the orienting interviews was to identify the (key) partners and to establish the focus of the study. Consequently, in-depth interviews could be conducted with the key figures in the alliance (Table 3). In-depth interviews are a valuable method of data-collection, because of the possibility to address the how and why questions (Bloomberg, Cooper & Schindler, 2011).

(19)

19 Table 3. Interviews

(20)

20 interviews was visible in two ways. First, questions were adapted to the role/function one had in the project. Second, specific information that was collected, was used in order to confirm previous interviews.

Secondary data sources

As mentioned before, it is beneficial to go beyond the primary source of data collection and integrate all existing information. Some of the alliance information was presented on the website of the alliance, other information was retrieved from internet8. Table 4 presents an overview of all the secondary documents that were used for the analysis. This table also includes a description of the content of those documents.

Table 4. Secondary documents

8

(21)

21

Data analysis

In analyzing the data, the orienting interviews were helpful in establishing the most important topics of the alliance. Both the interviews and secondary documents were analyzed via a step by step approach. The interviews were analyzed using codes (Appendix III). First, important quotes from the interviews were placed in four categories. These four categories were: 1. Benefits; 2. Challenges; 3. Formal governance mechanisms; 4. Social exchange mechanisms. The secondary documents fulfilled assisted in the analysis of the formal governance mechanisms. The secondary documents were analyzed by summarizing them (Appendix IV) and the relevant quotes were included in the results. The content of the secondary documents was also analyzed using the Code Book (Appendix III). In order to establish valid results, I analyzed the amount of quotes on the benefits, challenges and the formal governance and social exchange mechanisms (Appendix V). The second step in the step by step approach, was to include the most important concepts and analyze and compared them to the existing literature. Both existing and unknown theoretical concepts were identified here. Third, an effort was done to connect the four distinct categories of codes. In order to make the connections between the categories, different categories had to be linked to each other. This step by step approach can be found in Appendix VI. In the final step, the results were crystallized and the most important results were gathered. If important patterns were found they were compared to existing literature and propositions were developed (van Aken et al., 2012).

After the data-analysis an important step was taken to validate the findings. A draft version of the study was sent to the key partners within the alliance. This was a double-sided interest, my interest was to check the findings, theirs was to check for anonymity issues. This step allowed me to incorporate valuable feedback. It allowed me to check the validity of the case and improve the study.

Results

(22)

22 exchange mechanisms. An important analysis in this section is the effectiveness of the formal governance mechanism and the social exchange mechanisms. The effectiveness of the mechanisms is defined as the ability of the mechanisms to capture and or deal with the benefits and or challenges.

Formal governance mechanisms

In this section, the formal governance mechanisms are identified, exemplified and their effectiveness is determined. First, the benefits and challenges related to the formal governance are described. Second, the contract is described, differentiating between flexibility of contracts and contractual interfaces. Third, the existence of a steering committee is presented.

An important benefit of the contract was that it was rather broad. The data illustrates that the contract was used as a guideline and was not used to a large extent (Table 5):

“A broad contract is beneficial, as the most important clauses that are present stimulate an effective collaboration.” (Project manager 1, Partner K1)

This data already provides some directions into the flexibility of the contract. A broad contract was beneficial as it led to an effective collaboration.

Next, the challenging side of the formal governance mechanisms is described. The contract of the GreenGas Alliance referred to two formal documents, which were therefore legally binding (the Application form and the Project plan). All partners had to apply for public funding in order to be able to participate in alliance and along with that, they had to constantly report their progress in the alliance.

Partners experienced problems with the reporting standards as stipulated by the Application form. First, the reporting standards were very strict and demanding. Consequently, reports were sometimes incomplete, or demanded excessively detailed information (Table 5):

“Reports were always on time, but sometimes with incomplete information.” (Overall coordinator, Partner Z)

(23)

23 frustrated some partners (Table 5). As Project manager 1 (Partner K1) states:

“The level of information the funding party demanded was too detailed and even conflicting with privacy issues.”

Second, the alliance failed to create a clear plan for the future, along with a clear division of tasks (Table 5). This is illustrated by the following quote:

“In the beginning it was not clear to us what was expected from us.” (Asset manager, Partner F2)

The data illustrates that the contract did not clearly specify the task division between partners (Table 5).

Table 5. Benefits and challenges formal governance

Contract

(24)

24 Contract flexibility

In literature it is stated that a contract is desired to be flexible in alliances, as contractual clauses are often ambiguous. From the data it can be derived that the contract was used as a guideline. (Table 6). The contract allowed managers to discuss unclear aspects of the contract (Table 6). For example, an important contractual clause was the use of property rights. The contract stipulated the following:

“It is not allowed to use the co-created parts of the alliance outside the alliance.” (Contract)

Eventually, managers agreed upon a verbal agreement that illustrated the flexibility of the contract:

“There was a verbal agreement that stated that the co-created parts of the project, were for common use.” (Overall coordinator, Partner Z)

The data illustrates that the contract was used as a guideline and the managers were able to discuss the clause on property rights contract. Therefore, I state that the contract was flexible.

Table 6. Flexibility of the contract

Contractual interface

According to the literature, the contractual interface can be determined via three aspects: division of tasks, obligations to share information and the presence of monitoring mechanisms. First, the contractual interface mainly indicated an overlapping task division (Table 7):

“Partner R1 develops a scientific model…. Partner K1 develops an energy management system…. The other parties contribute to the work of Partner R1 and K1 by delivering data on technical, financial and economic aspects.” (Project plan)

Second, the contract did include clear obligations to share information. To illustrate:

(25)

25

foster the collaboration…. Each party is obliged to report to the other partners and the management, when they experience delays in fulfilling their tasks.” (Contract)

Third, both performance and behavior monitoring mechanisms were found. First, the alliance consisted of clear performance monitoring mechanisms. The alliance had a clear end date as it applied a subsidy for five years, after that the alliance was finished. Also, the alliance had clear target dates. To illustrate:

“The scientific model in the alliance had a time span of two years, to be completed.” (Application form)

However, no performance standards were described in the Project plan. The scientific model in the alliance, was linked to a PhD thesis. It is straightforward, that the researcher was best able to determine the performance standards. Therefore, the only performance standard was to present the model in time:

“The only prerequisite was that the scientific model I made, was finished in time.” (Theme leader, Partner R1)

In addition, the broad program milestones, were described as follows :

“Research had to be done, but no clear research questions existed yet…. Optimization model; energy management system, applicable knowledge for the biogas industry.” (Project plan)

Second, behavior monitoring mechanisms were also present in the alliance. This example shows the importance of behavior monitoring mechanisms:

“As Partner K1 was specialist in ICT, they are assigned as frontrunner in the development of an energy management system. In the Project plan three activities were described: Identification of possibility to share information in an automated way; A simulation of a smart gas grid; An architecture description of automated controlling engineered system.” (Project Plan)

(26)

26 Table 7. Broad contractual interface

Effectiveness of the contract

In establishing the effectiveness of the contract, the flexibility of contracts and the broad contractual interface are discussed. First, the effectiveness of the flexible application of the contract is discussed. The data illustrated that it allowed managers to discuss clauses in the contract and that the contract was used the contract as a guideline.

At the operational level the flexible contracts enabled cooperation between partners:

“We were not tied to cooperating with a specific party via the contracts. Therefore we were able to cooperate on an intensive basis with partner R1.” (Project manager 2, Partner K1)

Furthermore, managers stated that the contract was not of huge importance and it allowed managers to discuss the ambiguous contractual clauses. Therefore, I state that the flexibility of contracts was effective.

On the other hand, the broad contractual interface led to some problems. These two problems can be defined as: reporting standards and the division of tasks. At the operational level managers experienced that the monitoring was too detailed and excessive and the division of tasks was not sufficient. Therefore, I state that the broad contractual interface was ineffective in the alliance.

Steering committee

The contract included a clause that a steering committee should be assigned in the alliance. In this steering committee a representative from each partner (all the 18 partners) was present (Contract). The overall project manager was in charge of leading the steering committee. The managerial level stated how the steering committee was supposed to function:

(27)

27

tasks and addressed the requirements we had to fulfill. You could call me the manager of a company, in this setting…. In the steering committee it was made clear what tasks were divided, and which partner(s) was/were planning to do those tasks, either alone or together.” (Overall project manager, Partner Z)

Effectiveness of the steering committee

According to the operational level, the division of tasks in the alliance was not clear in the beginning. One of the steering committee’s functions was to divide tasks. For instance, the steering committee created clarity by agreeing upon certain deliverables that had to be delivered by each partner (Table 8). Also, partners stated that the report-oriented meetings, which were in the committee, were one of the most important formal structures:

“Intensive collaboration is not effective, therefore it is better to have certain predetermined moments were partners come together, to report the progress being made.” (Program manager, Partner R1)

In the literature it was stated that a steering committee might fill in the adaptive limits of contracts. In fact, the steering committee filled in a limit of the contract, by creating a clearer task division. Therefore, I state that the steering committee was effective.

Table 8. Effectiveness of the steering committee

Social exchange mechanisms

(28)

28

It was found that a lot of partners shared knowledge on an intensive basis and gaining knowledge was important for each partner (Table 9):

“We shared our information as much as possible.” (Director, Partner F1)

From the data it can be derived, that a lot of partners achieved to exchange valuable information Contrary, the challenging side of knowledge exchange was also visible in the alliance (Table 9):

“There was a less contributing partner, who was not willing to share all the information they had.” (Project manager 1, Partner K1).

Overall, the data illustrates that a lot of knowledge was shared (Table 9).

Conflicts could originate from partners that free-ride by abusing trust or ignore reciprocity. As described above, there was a less contributing partner. One would expect that the fact of the less contributing partner, could potentially lead to a conflict. However, as described below no conflicts were found in the alliance. As the overall coordinator (Partner Z) and other managers stated (Table 9):

“There was one party that had a conflict with other parties. This conflict was however not in this theme.”

Although, conflicts were present at other themes, the data clearly illustrates that no conflicts were present in this theme.

Coordination with multiple partners was expected to be difficult and costly in the MR&D alliance. The costly aspect of coordination was not found, but the difficulty is illustrated as follows (Table 9):

“We realized that it was a huge challenge to connect the world of research with parties that solely focus on the market.” (Scientific director, Partner Z)

Furthermore, coordination was challenging as partners were very different from each other. It became apparent that the time frames and desired pace of technological or knowledge development differed between the parties (Table 9). Also, the theme leader experienced problems, due to the fact that partners were different:

“I found it difficult to achieve the balance between my own job as a researcher and to serve the interests of the other partners.” (Theme leader, Partner R1)

(29)

29 Table 9. Benefits and challenges social exchange

Trust building

According to the literature, several mechanisms exist to build trust in an alliance (risk taking, equity preservation, communication, interfirm adaptation, generalized reciprocity and effort). Two of those mechanisms were found in the GreenGas Alliance: Generalized reciprocity and communication.

In MR&D alliance generalized reciprocity is challenging to achieve, as there is no direct link in the pattern of exchange. The theme meetings reflect the existence of generalized reciprocity. For example, the theme meetings were used to give workshops (Table 10):

(30)

30

manager 2, Partner K1)

To establish whether or not direct reciprocity was present, I looked at the knowledge exchange and found that partners contributed quid pro quo. Managers at the operational level supported this (Table 10):

“To us it was important to follow the trends of the research in the biogas industry, we did not expect something in return in the short run.” (Asset manager, Partner F2)

Moreover, the data illustrates that all partners gained and contributed knowledge. Also, partners accepted the fact that there was no direct reciprocity needed in order to exchange information (Table 10).

The theme meetings are in line with the need for communication in alliances, as stipulated in literature. According to the managers of the alliance, it was important to stimulate repeated interaction between partners:

“We found that intensifying the contact was effective and waiting for six months was too long.” (Scientific director, Partner Z)

(31)

31 Table 10. Trust building

Effectiveness of the trust building

To establish the effectiveness of trust building it is important to find out how the intensive knowledge sharing was created in the alliance. Both generalized reciprocity and (intensive) communication were visible in the alliance. I look at the goals as established by the managerial level. First, the goal of the alliance was to create an interactive alliance, that fostered knowledge exchange between partners:

“I wanted to create an alliance, where all parties could innovate and interact with each other, in different parts of the supply chain of biogas industry.” (Scientific director, Partner Z)

Second, the goal of the alliance was to create a sense of harmony between the partners:

“With more than two partners, striving towards a sense of harmony is easier, since one wants to be part of the group and doing things together creates a group feeling.” (Overall project manager, Partner Z)

(32)

32 Table 11. Effectiveness of the trust building

Conflict resolution

According to the literature joint problem solving or social sanctions assist in the resolution of conflicts. Managers addressed several explanations for the fact that no conflicts were found in this case. However, neither social sanctions nor joint problem, solving were found as a conflict solving technique. A first explanation is, that the cooperation was very broad, where each partner had a lot of freedom to fulfill its tasks.

“Probably because of the amount of freedom partners experienced there were no conflicts.” (Theme leader, Partner R1)

Also, the partners were no direct competitors, who created a lower potential for conflicts:

“Partners were not mutually competitive, therefore no conflicts were present.” (Director, Partner F1).

The data illustrates that due to the fact that partners had a lot of freedom and there was low competitiveness in the alliance, it seems logical that no conflicts existed in this theme. Therefore, no specific mechanisms can be linked to the benefit of the absence of conflicts.

Coordination

(33)

33 and beliefs (cooperative macro culture). Third, I point to the importance of expectations in the alliance.

First, it was found that several efforts were made to socialize with each other. The theme meetings reflect an important effort to socialize. In the trust building it was found to be an effective method. The theme meetings appeared to be important as they led to discussions and understanding of cooperating together and were found to be effective for cooperation (Table 10). Second, establishing a cooperative macro culture, was addressed at the managerial level (Table 12):

“Talking about cooperation, managers should emphasize the feeling of doing it together and should also point to the significance of that. Most important of managing the informal side of cooperation is to create a shared vision. People should have a sense of that vision and should know how they can play a role in pursuing that vision, on their own as well as together with their partners.” (Overall project manager, Partner Z)

Third, both the managerial level and the operational level pointed to the importance of expectations (Table 12). To illustrate:

“It is important to manage that partners have the right attitude and expectations towards the alliance. The most important expectations are set at the start of an alliance.” (Overall project manager, Partner Z)

The results extend the view of a cooperative macro culture, where partners share values and beliefs, as to improve coordination, it is also important to set the right expectations.

(34)

34 Effectiveness of the coordination

To establish the effectiveness of coordination, I describe whether or not the coordination has improved and if the problem of the different cultures is tackled effectively. Also, I highlight the importance of knowledge sharing. I link the described mechanisms to the benefits and challenges and where possible to each other.

The data shows that a cooperative macro culture was important to establish in the alliance. First, an important indication of the effectiveness of the cooperative macro culture, was the referral to the willingness of partners to cooperate together. Second, an important element of the cooperative was the existence of knowledge sharing, which also indicates an effective cooperation. Third, the theme meetings showed the importance of socialization, enabling partners to build understanding of their different values and beliefs. Fourth, the data illustrates that expectation management is important, both at the managerial and operational level, those expectations were sufficiently met (Table 13).

Therefore, the coordination of the alliance was effective, as the problem of different cultures has been tackled appropriately, by the creation of knowledge sharing, the theme meetings and expectation management.

(35)

35

Visualization of results

Below an overview (Table 14) is presented of the effectiveness of mechanisms in the GreenGas Alliance. Several benefits and challenges were linked to the formal governance or social exchange mechanisms, to explain why they were effective. However, not all benefits and or challenges could be linked to a mechanism.

*= no explanatory mechanism was found

Table 14. Visualization of results

Discussion

This study explores the phenomenon of governance in MR&D alliances through a TCE and a SET lens. In the visualization of results, the effective governance mechanisms are presented. Below, I describe explanations on the results and compare them to the theory.

Formal governance mechanisms

(36)

36 1996). The results show that flexible contracts are effective as the contracts are used as a guideline. Consequently, the contract allows partners to cooperate on an intensive basis, by an effective division of property rights. Theory can explain the importance of the division of property rights as contracts should address knowledge protection in MR&D alliances (Oxley & Sampson, 2004; Li et al., 2012). However, the results show that the co-created knowledge in a MR&D alliance should not by protected, but should be for everyone’s use. This can be explained by the higher complexity in governance of MR&D alliances (Doz & Hamel, 1998; Oxley & Sampson, 2004), which leads to the need of a flexible contract. Consequently, I propose the effect of flexible contracts in MR&D alliances as follows:

Proposition 1: In MR&D alliances a flexible contract is an effective formal governance mechanism, as it addresses ambiguous contractual clauses

Faems et al. (2008) state that broad contractual interfaces are effective in R&D alliances. Contrary, I find in the results that a broad contractual interface is ineffective, for two reasons. First, it leads to an unclear division of tasks. Literature provides two explanations for the unclear division of tasks. Contracts in R&D alliances are difficult to set at the start of an alliance (Oxley & Sampson, 2004) and contracts are bound to be incomplete (Williamson, 1996; Reuer & Devorakonda, 2015). The second reason that a broad contractual interface is ineffective, is because it leads to an excessive degree of monitoring. The existence of monitoring and reporting standards are rooted in the demands of the funding parties. This is surprising as appropriation concerns and consequently monitoring are higher in alliances with competitors (Oxley & Sampson, 2004). Contrary, in this alliance the partners were non-competing. No mechanism is found to be effective in addressing the challenge of the reporting standards.

(37)

37 enhancing administrative control in non-equity alliances. However, the results do not show sufficient ground to assume that the steering committee is the alternative for a board in an equity alliance. Therefore, the following proposition is stated:

Proposition 2: In MR&D alliances, a steering committee is an effective formal governance mechanism, as it addresses the incompleteness of contracts

Social exchange mechanisms

Generalized reciprocity means that all partners quid pro quo contribute to the alliance, which leads to trust building (Das & Teng, 2002). First, the study finds that knowledge can be shared intensively when generalized reciprocity exists, because of the existence of trust. Furthermore, the results show that generalized reciprocity leads to trust building and thus validates the findings of Das & Teng (2002), in the context of MR&D alliances. Second, theme meetings lead to the desired communication and the opportunity to achieve generalized reciprocity. This finding is in line with literature on strategic alliances, as well as alliances between academia and industry (Das & Teng, 1998; Bruneel et al., 2010). Communication is essential for strategic alliances as a constant information flow leads to a more trusting environment and it provides a basis for continued interaction (Das & Teng, 1998). Moreover, literature on alliances between academia and industry emphasizes that repeated interactions are important to build trust (Bruneel et al., 2010). Therefore, I state the following propositions on trust building:

Proposition 3: In MR&D alliances, generalized reciprocity is an effective social exchange mechanism, as it leads to trust building

Proposition 4: In MR&D alliances, theme meetings are an effective social exchange mechanism, as it leads to trust building

(38)

38 Teng, 1998). Furthermore, the importance of socialization can be explained by the importance of mutual understanding between academia and industry (Bruneel et al., 2010). Due to the familiarization of cultures and increased mutual understanding, theme meetings assist in developing a cooperative macro culture. Second, expectation management is a mechanism that helps to build a cooperative macro culture, and creates stable exchange relationships. Expectation management is an unidentified mechanism in the literature. The importance of setting the right expectations can be explained by Doz (1996). Failing projects in alliances are often accommodated with frustrated expectations. When expectation management is effective, the potential of frustrated expectations is reduced. By expectation management the set of shared values and beliefs increases and the exchange relationships become more stable. Therefore, I formulate the following three propositions:

Proposition 5: In MR&D alliances, a cooperative macro culture is an effective social exchange mechanism, as it leads to improved coordination

Proposition 6: In MR&D alliances, theme meetings are an effective social exchange mechanism in developing a cooperative macro culture

Proposition 7: In MR&D alliances, expectation management is an effective social exchange mechanism in developing a cooperative macro culture

Conclusion

Governance of MR&D alliances has not received a sufficient amount of attention in the literature (Li et al., 2012). A lot of authors focus on the governance of dyadic alliances. However, MR&D alliances are far more complex (Doz & Hamel, 1998; Das & Teng, 2002). Also, the amount of alliances that fail and the rise of R&D alliances indicates the importance of this study (Porter et al., 1996; Dyer & Singh, 2001; Hagedoorn, 2002). This case study on a MR&D alliance in the biogas industry reveals an enhanced understanding of governance mechanisms.

Theoretical and managerial implications

(39)

39 and provides insights for managers on effective governance in MR&D alliances. In the literature on TCE two limitations exist. First, literature fails to make a clear distinction between dyadic and MR&D alliances (Hagedoorn & Hesen, 2007; Pisano, 1989; Li et al., 2008; Oxley & Sampson, 2004). This study finds that a flexible contract is an effective formal governance mechanism in MR&D alliances. Furthermore, the study finds that a steering committee is an effective formal governance mechanism in MR&D alliances. This makes managers aware of the importance of the contract and the steering committee. The second limitation of TCE is that authors mainly focus on the use of equity alliances when cooperating on R&D (Garcia-Canal et al., 2003; Li et al., 2012; Pisano, 1989; Li et al., 2008; Oxley & Sampson, 2004). This study extend the work on R&D alliance literature, by stating that the steering committee can be seen as a potential alternative formal governance mechanism in non-equity MR&D alliances. This finding could assist managers in choosing to cooperate in an equity or non-equity alliance.

SET explains that in multi-partner alliances exchange is generalized (Das & Teng, 2002; Thorgren et al., 2010). Therefore, three difficulties in managing multi-partner alliances exist: trust building, conflict resolution and coordination (Das & Teng, 2002). This study enhances the understanding of SET by providing evidence on the applicability of additional relational mechanisms. Two additional mechanisms can assist in reducing the challenge of generalized exchange in MR&D alliances: theme meetings and expectation management. This could be valuable mechanisms for managers to implement in a MR&D alliance. Second, there is a lack of focus on R&D alliances in SET. In addressing the lack of focus on R&D alliances in SET, evidence on the effectiveness of generalized reciprocity and a cooperative macro culture in MR&D alliances is provided. In order create effective knowledge exchange in the MR&D alliance, managers should create set of shared values and beliefs and quid pro quo contribution between alliance partners.

Limitations and future research directions

(40)

40 studies should also be studied in other industries. Moreover, future studies should focus on a larger extent on the distinction between MR&D alliances with competing and non-competing partners (Bruneel et al., 2010; Oxley & Sampson, 2004). For instance, the biopharmaceuticals, automotive and new materials sector (Gulati 1995; Gulati & Singh, 1998) as in these industries a lot of multi-partner alliances exist.

The second limitation of this study is based on the fact that formal governance mechanisms and social exchange mechanisms are researched separately. However, formal and relational governance/social exchange mechanisms are often connected in literature and described as mechanisms that complement each other (e.g. Poppo & Zenger, 2002; Faems et al., 2008). Specifically, future research should explore the complementary or substitute relationship between steering committees and trust as this is a relatively unexplored formal governance mechanism (Reuer & Devorakonda, 2015). An approach could be to study the connections between TCE and SET, by using a longitudinal design (Faems et al., 2008). This allows researchers to discover how both types of mechanisms coevolve with each other over time.

The study shows that steering committee might be an alternative form for formal mechanisms in equity alliances, such as a board (Garcia-Canal et al., 2003). However, research should explore the functioning of a steering committee more in-depth (Reuer & Devorakonda, 2015). Empirical data is needed to find out on what grounds MR&D alliances use steering committees. Previous research has shown that equity alliances are more effective for knowledge protection in MR&D alliances (Li et al., 20120). With an enhanced understanding of the steering committee, the choice for equity and non-equity alliances might be more prone towards non-equity.

Acknowledgement

(41)

41

References

Arino, A. (2003). Measures of Strategic Alliance Performance: An Analysis of Construct Validity. Journal of International Business Studies, 34(1), 66-79.

Baxter, P., & Jack, S. (2008). Qualitative Case Study Methodology: Study Design and Implementation for Novice Researchers . The Qualitative Report, 13(4), 544-559.

Blau, P.M. (1964). Exchange and power in social life. New York: Wiley.

Bloomberg, B., Cooper, D.R., & Schindler, P.S. (2011). Business Research Methods. Mcgraw- Hill Education.

Coleman, J.S. (1994). Foundations of social theory. Cambridge: Harvard University Press.

Cyert, R.M., & Goodman, P.S. (1997). Creating Effective University-Industry Alliances: An Organizational Learning Perspective. Organizational Dynamics, 25(4), 45-57.

Das, T.K., & Teng, B.S. (1998). Between trust and control: Developing Confidence in Partner Cooperation in Alliances. The Academy of Management Review, 23(3), 491-512.

Das, T.K., & Teng, B.S. (2001). Trust, control and risk in strategic alliances: An integrated framework. Organization Studies, 22(2): 251-283.

Das, T.K., & Teng, B.S. (2002). Alliance constellations: A social exchange perspective. Academy of Management Review, 27(3), 445‐456.

Doz, Y.L. (1996). The evolution of cooperation in strategic alliances: Initial conditions or learning processes? Strategic Management Journal, 17, 55-83.

Doz, Y.L., & Hamel, G. (1998). Alliance advantage: the art of creating value through

(42)

42 Doz, Y. L., Olk, P. M., & Ring, P. S. (2000). Formation processes of R&D consortia: Which path to take? Where does it lead? Strategic Management Journal, 21(3), 239‐266.

Dyer, J.H., & Singh, H. (1998). The relational view: Cooperative strategy and sources of

interorganizational competitive advantage. Academy of Management Review, 23(4), 660- 679.

Dyer, J.H., Kale, P., & Singh, H. (2001). How to make strategic alliances work. Sloan

Management Review, 42(4), 37-43.

Eisenhardt, K.M. (1989). Building Theories from Case Study Research. The Academy of

Emans, B.J.M. (2004). Interviewing techniques and training. Leiden/Groningen: H.E. Stenfert Kroese.

Faems, D., Janssens, M., Madhok, A., & Van Looy, B. (2008). Toward an integrative perspective on alliance governance: Connecting contract design, trust dynamics, and contract

application. Academy of management journal, 51(6), 1053-1078.

Garcia-Canal, E., Valdes-Llaneza, A., & Arino, A. (2003). Effectiveness of dyadic and multi- party joint ventures. Organization Studies, 24(5), 743-770.

Gulati, R. (1995). Does familiarity breed trust? The implications of repeated ties for contractual choice in alliances. Academy of Management Journal, 38(1), 85-112.

Gulati, R., & Singh, H. (1998). The Architecture of Cooperation: Managing Coordination Costs and Appropriation Concerns in Strategic Alliances. Administrative Science

Quarterly, 43(4), 781-814.

(43)

43 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(3), 342-366.

Hagen, J.M., & Choe, S. (1998). Trust in Japanese interfirm relations: Institutional sanctions matter. Academy of Management Review, 23, 589-600.

Hamel, G., Doz, Y.L., & Prahalad, C.K. (1989). Collaborate with your competitors and win. Harvard Business Review, 67(1), 133-139.

Inkpen, A.C., & Tsang, E.K. (2005). SOCIAL CAPITAL, NETWORKS, AND KNOWLEDGE TRANSFER. Academy Of Management Review, 30(1), 146-165.

Jones, C., Hesterly, W.S., & Borgatti, S.P. (1997). A GENERAL THEORY OF NETWORK

GOVERNANCE: EXCHANGE CONDITIONS AND SOCIAL

MECHANISMS. Academy Of Management Review, 22(4), 911-945.

Kumar, R., & Nti, K.O. (1998). Differential learning and interaction in alliance dynamics: a process and outcome discrepancy model. Organization science, 9(3), 356-367.

Lavie, D., Lechner, C., & Singh, H. (2007). The performance implications of timing of entry and involvement in multi-partner alliances. Academy of Management Journal, 50(3), 578‐604.

Li, D., Eden, L., Hitt, M. A., & Ireland, R. D. (2008). Friends, acquaintances, or strangers? Partner selection in R&D alliances. Academy of Management Journal, 51(2), 315-334.

Li, D., Eden, L., Hitt, M.A., Ireland, R.D., & Garret, R.P. (2012). Governance in Multilateral R&D Alliances. Organization Science, 23(4), 1191-1210.

(44)

44 Mothe, C., & Quelin, B.V. (2001). Resource creation and partnership in R&D consortia. The

Journal of High Technology Management Research, 12(1), 113‐138.

Nord, W.R. (1969). An integrative approach to social conformity. Psychological Bulletin, 71, 174-208.

Ouchi, W.G., & Bolton, M.K. (1988). The Logic of Joint Research and Development. California

Oxley, J.E. (1997). Appropriability hazards and governance in strategic alliances: A transaction cost approach. Journal of Law, Economics, and Organization, 13(2), 387-409.

Oxley, J.E., & Sampson, R.C. 2004. The scope and governance of international R&D alliances. Strategic Management Journal, 25(8-9), 723-749.

Parkhe, A. (1993). Strategic alliance structures: A game theoretic and transaction cost

examination of interfirm cooperation. Academy of Management Journal, 36(4), 794-829.

Pisano, G.P. (1989). Using equity participation to support exchange: Evidence from the biotechnology industry. Journal of Law, Economics, and Organization, 5(1), 109-126.

Poppo, L. & Zenger, T. (2002). Do formal contracts and relational governance function as substitutes or complements? Strategic Management Journal, 23, 707-725.

Porter, M.E., Goold, M., & Luchs, K. (1996). Managing the multibusiness company: Strategic

issues for diversified groups, New York: Routledge.

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(3), 393-404.

Governing MR&D alliances: formal and social mechanisms