CONFLICT BETWEEN ORGANIZATIONS: THE DIFFERENT INFLUENCES OF R&D AND MARKETING ALLIANCES Master Thesis

CONFLICT BETWEEN ORGANIZATIONS: THE DIFFERENT INFLUENCES OF R&D AND MARKETING ALLIANCES

Master Thesis

University of Groningen Faculty of Economics and Business MSc BA Strategic Innovation Management

Thesis supervisor: dr. F. Noseleit Second supervisor: R.A. van der Eijk

Maarten Heeres Student number: 1685252

Stadsweg 8 9917PW Wirdum

E-mail address: m.j.heeres@student.rug.nl Word count: 13.755

2 ABSTRACT

This study examines how forming a marketing or R&D alliance influences the level of conflict between partner firms and what actions partners can undertake to minimize the level of conflict arising. First, it is hypothesized more conflict will arise between alliance partners compared to firms not forming an alliance because of intensive resource exchanges in alliances. Second, by using literature streams related to transaction cost economics and the resource based view it is expected that R&D alliances using an integrated governance structure engage in less conflict. Third, the combination of both a marketing alliance and an R&D alliance is hypothesized to lead to a lower level of conflict because competition is taken away and the partner specific learning rate increases. The hypotheses were tested by using a sample of 90 firms operating in the pharmaceutical industry. Results show that although in both marketing and R&D alliances more conflict arises compared to firms not forming an alliance, this effect is only significant for marketing alliances. Second, although a more integrated R&D alliance governance form does decrease conflict, this effect was insignificant. Third, a significant interaction effect is found for being active in an alliance comprising both marketing and R&D. A marketing alliance leads to more conflict, but this appears to be less so the case when an R&D alliance is present next to the marketing alliance. Different explanations for the found relationships are presented. Finally, this paper calls for more fine-grained measures that unravel the different mechanisms which determine the level of conflict arising in (combined) marketing and R&D alliances.

3

TABLE OF CONTENTS

1. INTRODUCTION ... 4

2. LITERATURE REVIEW ... 7

3. METHODOLOGY ... 16

3.1 Data and Sample ... 16

3.2 Measures and coding of variables ... 17

3.2.1 Dependent variable ... 17 3.2.2 Independent variables ... 17 3.2.3 Moderating variable... 17 3.3 Control variables ... 18 3.4 Estimation approach ... 19 4. RESULTS ... 21 4.1 Descriptive statistics ... 21 4.2 Multicollinearity ... 21 4.3 Hypotheses testing ... 21 5. DISCUSSION ... 25

5.1 R&D and marketing alliances influence on conflict ... 25

5.2 R&D alliance governance influence on conflict... 27

5.3 R&D and concomitant marketing alliances influence ... 27

6. LIMITATIONS & RECOMMENDATIONS FOR FUTURE RESEARCH ... 30

7. IMPLICATIONS & CONCLUSIONS ... 32

7.1 R&D and marketing alliance impact the level of conflict ... 32

7.2 R&D governance structure choice influence the level of conflict ... 33

7.3 Influence concomitant marketing alliance ... 33

4

1. INTRODUCTION

In today's fast-paced, knowledge-intensive environment, research and development (R&D) alliances have become a popular way of acquiring and leveraging technological capabilities (Oxley & Sampson, 2004). However, despite the increasing popularity of inter firm collaboration, up to this date, empirical evidence indicates that 50 percent of inter-firm collaborations fail (Ernst, Lichtenthaler & Vogt 2011). The lack of success is often caused by challenges related to the protection of technological knowledge, since successful completion of alliance objectives often requires a firm to put valuable knowledge at risk of appropriation by alliance partners (Oxley & Sampson, 2004). Firms must therefore find the right match between maintaining an open knowledge exchange to fuel the technological development goals of the alliance, and controlling knowledge outflows to avoid unintended leakage of valuable technology. A subtle balance must be struck in order to avoid conflict.

In this paper I consider how alliances can influence conflict. Prior research in transaction cost economics suggests choosing an integrated alliance governance structure is one mechanism firms can use to decrease opportunistic behaviors leading to conflict (e.g., Pisano, 1990; Oxley, 1997; Sampson, 2004). However, there are circumstances where even the most integrative alliance form (i.e. an equity joint venture) does not decrease the threat of misappropriation of knowledge sufficiently (Oxley & Sampson, 2004). Under such conditions: (1) firms can decide not to cooperate and therefore not realize the possible benefits that can accrue from the cooperation, (2) they can decide to seek an alternative way of cooperation that leads to a reduction in knowledge spillovers or (3) they can search for a way that does not reduce the spillovers, but makes them less of a threat. In this thesis I will shed light on the third alternative; how appropriation of knowledge by a partner can be made less of an issue. Specifically, I argue a concomitant marketing alliance further down the value chain of the firm (next to the R&D alliance) can help to decrease the level of conflict that can arise in an alliance. The analysis of this paper is constructed in three parts that work towards assessing the influence of a concomitant marketing alliance. Each part contributes to existing theory in a different way.

The first part of the analysis focuses on the impact alliances have on the level of conflict between corporations. It is argued a firm active in an alliance faces higher levels of uncertainty compared to a firm only engaging in marketing transactions. Because of this higher uncertainty conflict is more likely to arise. This part of the research strengthens the existing knowledge regarding R&D and marketing alliance uncertainties. It refines the understanding of the uncertainties and potential for opportunism and conflict which can arise in both alliance forms.

5

are used to explain how an R&D alliance governance structure can impact the level of conflict through opportunism. Broadly, these two literatures suggests partners in an R&D alliance are less likely to display opportunistic behavior when the governance form is more integrated. This part builds on the existing evidence and expands the literatures regarding R&D alliance governance structure. It aids the existing research because it takes a large time frame (10 years) and because it focuses on both business to business and business to consumer markets in the pharmaceutical industry.

The third part of the analysis regarding the influence of a concomitant marketing alliance on the R&D alliance is less well researched. Two arguments are theorized to explain how a concomitant marketing alliance may impact the amount of conflict arising between alliance partners; through experience and elimination of competition. First, it is argued that an additional marketing alliance results in building up partner specific experience more quickly; which in turn results in a cooperation characterized by a higher level of trust; where rules become less important, knowledge is more easily shared and opportunism less likely (Blumberg, 2001; Faems, Janssens, Madhok & van Looy, 2008). Second, when organizations are cooperating in both R&D and marketing areas spillovers in the R&D alliance become less threatening because firms do no longer compete further down the value stream. This third part of the research is the most novel. It sheds light on the moderating effect a marketing alliance can have on the level of conflict arising in an R&D alliance. It explains how an additional marketing alliance can take away threats of misappropriation of knowledge and allows alliance partners to interact more freely. In doing so it fills a gap by analyzing the effect an additional (marketing) alliance further down the value chain has on the level of opportunistic behavior present in the R&D alliance.

Summarizing, the paper addresses the following research question and sub-questions:

Research question:

How does the presence and governance of an R&D alliance influence the level of conflict between organizations and how is the relationship between R&D alliance presence and conflict affected by a concomitant marketing alliance?

Sub-questions:

- How does the presence of an R&D and marketing alliance impact the level of conflict?

- How does the R&D governance structure choice influence the level of conflict?

6

This analysis will add value to different literature fields. Most directly this thesis will add value to the TCT and RBV. Further, this thesis builds on the present evidence and expands the literatures regarding R&D alliance governance structure. It helps to explain how the nature of the collaboration influences the level of opportunism in the alliance by looking at certain alliance characteristics (e.g. type of knowledge exchange, trust, bounded rationality). Further, the agency theory can be strengthened, as managers can be seen as active decision makers, determining the selected governance form.

The analysis will take place in the pharmaceutical industry. A sample of 90 firms is used to study the hypothesized effects over a time span of 10 years; ranging from 2002 till 2012. A dyad of each combination of firms is constructed to measure the difference in conflict arising between alliance forming dyads and those dyads not forming alliances. The sample consists of firms focusing on both business and consumer markets, which increases the scope of the findings.

7

2. LITERATURE REVIEW

Conflict is defined by Rahim (2011, p. 16) as "an interactive process manifested in incompatibility, disagreement or dissonance within or between social entities". Between organizations conflict is likely to arise due to divergence of interests and opposing actions (Deutsch, 1977). In today’s increasing and globalizing competition the likelihood of ending up in conflict increases, because companies face larger risks of competitors trying to exploit and steal away their knowledge. Some companies try to minimize these opportunistic actions by organizing knowledge intense activities internally, thereby limiting the access other firms have to the focal companies knowledge (e.g. Pisano, 1990). Contrasting this advantage is the lack of progress that might result from not cooperating with outside partners; i.e. a company will be less likely to capture external knowledge if it limits its external ties.

In an attempt to overcome the risks of cooperation and benefit from the advantages firms can cooperate by forming a strategic alliance (Das & Teng, 2000). Broadly defined, strategic alliances refer to inter firm cooperative arrangements aimed at pursuing mutual strategic objectives (Das & Teng, 2000). Although an ‘alliance’ can refer to anything from an arm’s length contract to a joint venture, a strategic alliance refers to pooling of resources on a strategic level. Yoshine & Rangan (1995, p. 5) use the following three components to define strategic alliances:

- Alliance partners pursue a set of agreed upon goals and remain independent after forming the alliance.

- Alliance partners share the benefits and managerial control over the performance of the assigned tasks.

- Alliance partners make continuing contributions in one or more strategic areas, such as technology or products.

8

Although alliances are becoming more and more a necessity for ensuring a competitive edge in the marketplace (Das & Rahman, 2010), a high percentage of the alliances that are formed generates conflicts and ultimately fail (Ernst, Lichtenthaler & Vogt 2011). Research has shown the failure rate is significantly higher for alliances than for firms having a single firm strategy (Bleeke & Ernst, 1991; Das and Teng, 2000). One key difference between a single firm strategy and an alliance is uncertainty. When a firm pursues market opportunities on its own, its involvement with other companies is limited to market transactions, thereby limiting their concerns regarding opportunistic behaviors which could lead to conflicts. In an alliance the cooperation is much more intense and on a strategic level, implying higher risks of the partner acting opportunistically and ending up in a conflict.

Das & Teng (2001) use relational and performance risk to describe the differences between firms that form alliances and those who do not. They argue that although every firm faces performance risks; relational risks are only present when firms are cooperating.

Relational risk is defined as the probability of not having satisfactory cooperation (Das & Teng, 1996). This risk refers to the potential opportunistic behaviors that might be displayed by both alliance partners, e.g. shirking, cheating, distorting information or unwanted appropriation of resources. In such a situation conflict can arise because firms are more focused on their own private benefits instead of common benefits. Further, partner firms may have hidden agenda’s in the alliance, e.g. secretly learning valuable knowledge outside the agreed upon scope of the alliance. Besides this relational risk, there are several other factors that can negatively influence alliance performance. These factors are referred to as performance risk and can include intensified rivalry, new entrants, demand fluctuations, changing government policies or sheer bad luck (Das & Teng, 2001). Performance risk consists of the chances that alliance objectives are not obtained, despite satisfactory cooperation between the alliance partners. Think for example of two firms in different industries forming an R&D alliance, while having too little absorptive capacity to successfully exchange knowledge.

There are different theories and analyses that built upon the analysis of Das and Teng (2001) and explain and support the claim that alliances in general lead to more conflict. They are discussed briefly. An overview of these theories is presented in table 1.

9

Here it is argued forming an alliance is comparable with the prisoners’ dilemma, where two members of a criminal gang are unsure of the intentions of their partners. Although cooperation would reduce both their sentences, if one cooperates but the other does not the non cooperating gang member’s sentence is decreased the most. The assumption in this analogy is that behaving opportunistically should lead to a higher outcome than cooperating. Because alliances are often organized in a way that firms can make more gains by cheating or exploiting the partner it seems unavoidable that these alliances end up in conflict. Third, the resource dependence theory and the bargaining power perspective indicate that firms rely on resources of others and manage inter organizational relationships in such a way that the dependence upon the external environment is minimized (Pfeffer & Salancik, 1978; Bucklin & Sengupta, 1993). Starting an alliance decreases the dependence on the environment, but increases the dependence on a partner. The dependence of each of the partners is influenced by the resources each partner possesses. The higher the value of the resources, the less partner dependence and the more bargaining power. A shift of bargaining power can alter the relationships; i.e. a partner will try to renegotiate the agreements of the alliance, leading to conflict. Further, because firms enter alliances mostly to acquire knowledge and skills it is inevitable that the need for cooperation will be diminished when this transfer is complete. Fourth, the agency theory can be extended to alliances (Jensen & Meckling, 1976; Geringer & Woodcock, 1995). This theory implies managers might be motivated to pursue their own goals, rather than the goals of the alliance, leading to conflict.

10 Table 1: Theories explaining why alliances induce conflict

Theory Explanation of theory Authors

Relational contracting Alliances are viewed as relational contracts; conflict can arise due to difficulty of establishing trust.

Macneil, 1974, 1980

Transaction cost theory

Opportunistic behaviors are difficult and costly to control; increasing the likelihood of conflict.

Williamson, 1975, 1985

Game theory Alliances can be viewed as games in which cooperation may not be the most successful strategy; thereby inducing conflict.

Axelrod, 1984

Resource dependence theory

After one of the partner firm’s has acquired the resources the alliance becomes unstable; ultimately leading to conflict.

Pfeffer & Salancik, 1978

Bargaining power perspective

When a firm has acquired resources from its partner the bargaining power shifts; a firm tries to renegotiate the terms contract; thereby increasing the likelihood of conflict.

Bucklin & Sengupta, 1993

Agency theory Agents are prone to self-serving goals, which may not directly be in the best interest of the alliance. In such a situation managerial decisions can give rise to conflict.

Jensen and Meckling, 1976; Geringer and Woodcock, 1995

Concluding, it is expected that although the level of conflict arising in an alliance (either R&D or marketing) is significantly higher than if a firm would not form an alliance, the level of conflict occurring in an R&D alliance is higher than the conflict occurring in a marketing alliance. Therefore in the first hypothesis it is argued that (a) the presence of an R&D alliance will significantly increase the level of conflict between firms, (b) the presence of a marketing alliance will significantly increase the level of conflict between firms, and (c) an R&D alliance will give rise to more conflict compared to a marketing alliance.

11

H1b: The presence of a marketing alliance will significantly increase the level conflict between firms.

H1c: The presence of an R&D alliance will give rise to more conflict compared to a marketing alliance.

Because it is expected that the threat of conflict is the highest in an R&D alliance it is interesting to assess how this conflict can be minimized. One important mechanism to minimize the level of conflict is the alliance governance structure (Oxley, 1997). An alliance governance structure can be positioned on a continuum ranging from a hierarchical based governance structure at one end of the continuum to a contract based governance structure at the other end (Oxley, 1997). In this analysis, the focus is on two of those forms: the non equity based alliance and the equity joint venture. A non equity based alliance is a contractual agreement in which partners pool their capabilities for the purpose of collaborative R&D on a strategic level but do not form a separate legal identity for the alliance. Firms can also pool capabilities under an equity joint venture. In this more integrated form a new entity is created and jointly owned and operated by two or more collaborating firms (Oxley, 1997; Pisano, Russo & Teece, 1988). The depicted alliance governance form may affect the likelihood of opportunistic behavior within the relationship, since the alliance organization affects the firm’s ability and incentives to share information with its partners.

In the literature there are different streams that can help to assess the impact of alliance governance choice on the level of opportunistic behavior and subsequently conflict. The first one is the RBV and the second one relates to TCE (Mazzolaa & Perronea, 2013). Table 2 summarizes relevant research for these two theories; the table displays the forces that determine the governance choice and the most suitable alliance governance option.

12

focusing on the protection of a firm’s own resources) becomes less of an issue because the company faces less of a threat of their partner behaving opportunistically. The RBV suggests that, because knowledge is the transacted resource, an integrated governance structure can lead to less hesitation to share the knowledge without threats of unwanted appropriation (Chen & Chen, 2003; Vilalonga & McGahan, 2005).

Table2: The main forces influencing the governance structure choice

Theory Forces Preferred alliance governance mode Authors

RBV Resources: technology intensity

Alliances concerning technological resources should have a more integrated governance form.

Vilalonga & McGahan, 2005

Resources: exchange vs. integration

R&D alliances should possess a more integrative alliance form; marketing alliances should have a more exchange oriented relationship.

Chen & Chen, 2003

Resources typology: property/knowledge

Alliances involving marketing resources are more market oriented than agreements where R&D resources are involved

Das and Teng, 2000

Resources; Appropriation

Alliances involving R&D resources should have a more hierarchical governance form due to threats of appropriation.

Teng and Das, 2008

TCE Cost minimizing strategy

Within an R&D alliance costs are minimized when an integrative governance structure is depicted

Oxley, 1997

Opportunism risk: Appropriation concern

Alliances focusing on technological product development should have a more integrated governance structure. Oxley and Sampson, 2004 Opportunism Tacit knowledge Uncertainty

The presence of; tacit knowledge, uncertainty and threats of opportunism in R&D alliances imply these agreements should have a more hierarchical governance form.

Teece, 2010

13

activities, such as organizing information, coordinating behavior, monitoring transactions and making behavior adjustments (Hennart, 1988; Kogut, 1988; Oxley, 1997). Each form of strategic alliance is explained as a firm’s arrangement to minimize the total costs required to achieve specific business goals. The transaction costs related to the exchange of resources in an alliance can be reflected by the following factors: tacit knowledge, uncertainty, opportunism, risks and bounded rationality. In an R&D alliance the cooperation is oriented towards exchanging technological knowledge which often has a tacit nature (Teece, 1992; Teece, 2010). The complexity of knowledge that needs to be transferred is often high; the knowledge is tacit and difficult to specify. Therefore, a lot of communication (e.g. face to face meetings) and coordination is necessary to successfully transfer knowledge. Here the transaction cost theory argues that closely working together also increases the possibility of unwanted spillovers or opportunistic behavior. Companies need to safeguard themselves against this threat. An equity based joint venture offers partner this protection because a separate legal entity is created (Sampson, 2007). The capabilities of both firms can be pooled, and employee rotation can ensure that learning can take place. All in all, equity-based governance in alliances helps create a common setting in which learning can take place (Teng & Das, 2008).

Based on arguments related to the RBV and the TCE it is argued that because knowledge is the resource that is exchanged in an R&D cooperation, an integrative governance structure is required to safeguard companies from opportunistic behaviors that lead to unwanted appropriation of knowledge. Hypothesis 2 is formulated as follows:

Hypothesis 2: An equity oriented R&D alliance governance structure is expected to decrease the level of conflict within an alliance.

Besides assessing the impact alliances and the R&D alliance governance form have on the level of conflict this research addresses another variable that can influence the level of conflict; a concomitant marketing alliance. A marketing alliance can have different influences on the level of opportunistic behavior that is displayed in the R&D alliance.

14

because marketing related resources are less capital intense; the knowledge is less complex and can therefore more easily be developed or acquired (Pangarkar & Klein, 2001; Shrader, 2001; Schögel, Herhausen & Schmitz, 2011). There is however a risk of working together only in R&D. Because firms compete simultaneously in other areas they struggle with the dilemma between the need to work together in order to create value and the temptation to be opportunistic. A firm can for example appropriate a greater share of the created value than legitimate, and use it in downstream activities (Lavie, 2007; Gnyawali and Park, 2011). For this reason coopetitive relationships are unstable and dynamic in nature, which causes a high level of tension for firms.

Although one way to reduce these opportunism hazards is making use of a more integrated governance structure, the threat of one of the partner companies using knowledge unrightfully can still be present. This is especially the case when competition in both R&D and end markets is intense. In such a situation there are high levels of uncertainties and interdependencies for both partners within the R&D alliance. Firms face the dilemma of the existence of very attractive opportunities, but high risks of misappropriation of the partner as well. These coopetitive relationships can be seen as relationships involving a high degree of interdependence, with a high chance of conflict but also a high potential pay-off (Gnyawali & Park, 2011). Because the products with which companies compete further down the value chain are the same products that are invented in the R&D alliance, the competitive tension in the R&D alliance rises. This tension further intensifies because direct competitors may be both aware of the other’s opportunistic behaviors and be willing to confront each other (e.g. via legal measures). This propensity to confront each other is also high because opportunistic actions by the partner can result in serious knowledge- and market loss. Think for example of Samsung collaborating with Sony to develop LCD panels (Ramstad & Dvorak, 2006). Although both companies realize they increasingly need each other to fasten the development pace and bring innovative LCD sets to customers, their rivalry is very intense, with various proclaimed patent litigations (Ramstad & Dvorak, 2006).

These conditions lead to both instability and reinforcement of the relationship. It increases the need for a more stable partnership and a more balanced cooperation and competition. In such a situation a concomitant marketing alliance can help to minimize the level of opportunistic behavior. An additional marketing alliance does not alter the interdependence present in the relationship, but does alter the chance of conflict arising. Because when an additional alliance is formed the competition further down the value chain is deleted; the anxiety of firms stealing knowledge is decreased because misappropriated knowledge cannot be used anymore to compete further downstream in the value chain.

15

R&D alliance. If for example trust is experienced in the R&D alliance, these positive experiences may extend to the marketing alliance. Relationships with the same partner in multiple points in the value chain are expected to increase the interconnectedness among the alliance partners. Relational ties are developed more quickly, which may influence the governing of the alliances. Because alliance partners’ experience develops fast, trust is enhanced between them and this in turn mitigates opportunistic actions in the partners’ behavior.

All in all, it can be expected that an additional marketing alliance (next to the R&D alliance) will influence the level of conflict by moderating the relationship between R&D alliance presence and conflict. By eliminating the threat competition and fast growing partner experience less conflict will be displayed. Therefore the following effect is hypothesized:

Hypothesis 3: The presence of R&D alliance is expected to lead to conflict. However, this is less so the case when a concomitant marketing alliance is present.

The three deducted hypotheses are graphically represented in figure 1. Each depicted relationship represents a hypothesis. The nature of the relationship is depicted by a plus or minus sign, indicating a positive (plus) or negative (minus) influence on the degree of conflict arising in the alliance. Hypothesis 1C is displayed by indicating the difference in breadth between the arrows of Hypothesis 1A and 1B. The larger breadth for hypothesis 1A indicates a stronger impact for that concept on the level of conflict compared to the concept of hypothesis 1B.

Figure 1: Conceptual model

Conflict

-16

3. METHODOLOGY

3.1 Data and Sample

A dataset of 90 firms operating in the pharmaceutical industry between 2002 and 2012 was formed by making use of the firm-level Securities Data Corporation’s (SDC) Platinum database and the RPX database. The dataset was constructed as such that it listed each possible combination of pairs of the total of 90 firms between which conflict could arise and alliances could be formed; adding up to a total of 4005 combinations (90*89/2). In total, 134 R&D alliances and 46 marketing alliances were present in this dataset. Multipartner alliances were dropped from the sample. The SDC database contained the following information: the data of the R&D alliance start; the identity of the partners; the number of partners involved in the alliance; whether or not a joint venture was formed in the R&D alliance; whether or not a marketing alliance was present between the partners and the firm size. Because the data regarding firm size could not be retrieved for 845 dyads, these were left out of the analysis; leaving 3160 total dyad observations. The RPX database contained the total number of patent litigations between each partner dyad, either as a plaintiff or defendant.

17 3.2 Measures and coding of variables

3.2.1 Dependent variable

CONFLICT: Conflict was the dependent variable used in this study and was measured by looking at the total number of patent litigations that arose between the two partners. Different scholars have identified patent litigations as a valid way of both studying and measuring conflict (e.g. Kaufmann & Stern, 1988; Winsor, Manolis, Kaufmann, & Kashyap, 2012). The RPX database from which the patent litigations were retrieved identified both individual patent litigations and campaigns. An individual patent litigation indicated each individual legal appeal regarding a patent; a campaign on the other hand could best be viewed as a bundle of patent litigations that were related to the same or a similar subject. In this thesis the total number of campaigns was used instead of the total number of individual patent litigations because the number of distinct disputes was more relevant. It gave a more accurate image of the level of conflict between two companies because the same conflict causing behavior was not filed multiple times. The variable was ranged on a ratio scale and represented the total number of campaigns initiated by both the two companies. This dyad conflict variable was constructed both for all 3160 combinations. I.e. both for companies that had formed R&D and marketing alliances and for those who did not form alliances.

3.2.2 Independent variables

R&D ALLIANCE PRESENCE: The R&D alliance presence was operationalized by making use of a dummy variable. A 0 was used to indicate that there was no strategic R&D alliance between the two firms. A 1 was used to indicate an R&D alliance did take place, with the cooperation being either a non-equity or equity based alliance.

MARKETING ALLIANCE PRESENCE: The marketing alliance functioned as an independent and moderating variable. For the independent variable another dummy variable was created, with 0 indicating no presence of a marketing alliance and 1 indicating the presence of a marketing alliance.

EQUITY ORIENTED GOVERNANCE STRUCTURE: The equity oriented governance structure concept was again operationalized by making use of a dummy variable. The governance equaled 0 when the alliance was organized by a non-equity based governance form and a 1 when the governance form is equity based and a joint venture was formed.

3.2.3 Moderating variable

18

R&D ALLIANCE PRESENCE*MARKETING ALLIANCE PRESENCE. This multiplicative variable was constructed out of the following variables: R&D alliance presence and marketing alliance presence.

3.3 Control variables

Because factors other than those included in the hypothesis might affect the level of conflict generated in an alliance, the following control variables were included in this research. This way, these factors were held constant so that the influence of the independent variables could be measured. A description is given of the variables that were found to have a strong influence on the level of conflict in previous research. These are as follows:

AVERAGE CAMPAIGNS PER ALLIANCE: The overall propensity of a firm to file patent litigations and start a campaign might have a large influence when it comes to litigation initiation. Therefore, the average campaigns of the two firms measured on a dyad level was used to control for this affect.

FIRM SIZE: Scholars have proven large firms are often more inclined to legally appeal and file a patent litigation (e.g. Bessen & Meurer, 2005). Large firms possess more financial resources, which allows them to spend more money to initiate lawsuits and to a start a patent litigation. Therefore, differences in firm size were controlled for by taking into account the number of employees present at each firm (Cloodt, Haagedoorn & van Kraanenburg, 2006). A variable was composed where the average firm size of a duo of companies is generated. Because the distribution of the average firm size data was skewed (1.31, SE = 0.04), log transformation was used to normalize the data: ln((size firm A + size firm B)/2).

19

YEAR OF FORMATION: The time range in which patents litigations could occur (ranging from 2002-2012) was controlled for. When an R&D alliance was present, the time range was defined as both during the start and after the alliance; i.e. taking place from the year of the alliance start till 2012. Although the average lifespan of an R&D alliance in the pharmaceutical industry is 3 years (Phelps, 2003), the RPX database information showed that campaigns could still arise several years after this average lifespan as well. To control for the effect that alliances with an earlier starting year would have a higher chance to file more campaigns against one another the number of campaigns were adjusted for the number of years campaigns could arise. This adjustment was performed by dividing the number of campaigns by the number of years that the alliance was in existence. If for example an alliance started in 2004 and led to 4 conflicts by the end of 2012 the number of conflicts was calculated as: 4/(2012-2004)= 0.5 campaigns. If an R&D alliance was not present between companies the total number of campaigns was divided by 10, the total time range of the dataset. By controlling the number of campaigns for the lifespan in which they occurred a comparison between alliances starting in different years could be made.

3.4 Estimation approach

First the likelihood of conflict arising, the dependent variable, was estimated. Relative to the total number of dyads considered in this analysis, the event of conflict occurring is far less common; only in 147 dyads conflict occurred compared to 3013 partner dyads in which conflict did not occur. Because the dependent variable measures the number of conflicts arising in an alliance a regression technique needs to be used that allows to deal with the count data characteristics of this measure. Count data cannot be easily analyzed with traditional regression techniques like ordinary least squares because the distribution is discrete, non continuous and it is limited to non-negative values (Gardner, Mulvey & Shaw, 1995; Calvin, 1998). Further, it is quite likely that an ordinary linear regression model will produce negative predicted values, which are theoretically impossible as conflict cannot take-on a negative values. Different tests were conducted to assess the normality of the data. A Shapiro Wilk’s test (Shapiro & Wilk, 1965) showed that normality of data cannot be assumed (S-W = 0.13, df = 3160, p < 0.01). The Skewness (13.10, SE = 0.04) and Kurtosis (237.42, SE = 0.08) as well as graphical representations of the data (e.g. box plots, Q-Q plots and histograms) indicated the data did not follow a normal distribution. The data is skewed in that many observations in the data set have a value of 0. This high number of 0’s in the data set prevents the transformation of a skewed distribution into a normal one. Both a log and square root transformation did not significantly alter the normality of the data. Therefore, a non-parametric test had to be used.

20

21

4. RESULTS

4.1 Descriptive statistics

Table 3 reports the means and standard deviations for the variables. Of the total dyads of 3160 3.3% (134 cases) included an R&D alliance and 1.5% (46 cases) included a marketing alliance. Of the R&D alliances 9.7% were governed by a joint venture, leaving the remaining 90.3% to be governed by minority or non-equity oriented governance mechanism.

Table 3: Descriptive statistics

Variables Mean S.D. Min. Max.

Conflict 0.13 0.94 0 26

R&D alliance presence 0.03 0.17 0 1

Marketing alliance presence 0.00 0.05 0 1

R&D alliance equity focus 0.00 0.06 0 1

Average campaigns per

alliance 9.89 12.79 0 88

Firm size 8.82 1.71 4.47 11.63

Size difference 8.91 2.16 -1.70 11.66

Year of formation 9.82 1.11 0 10

4.2 Multicollinearity

Previous to testing the hypotheses a multicollinearity test was performed to assess whether the predictor variables were closely correlated to one another. The tolerance and variance inflation factor (VIF) was used to determine if the values of the independent variables were within acceptable levels. Authorities differ on how high the VIF has to be to constitute a problem. A maximum value of 10 is widely used as the highest acceptable level (Kennedy, 1992). Some scholars however recommend lower values as small as 4 (Pan & Jackson, 2008). In the current dataset, none of the VIF factors exceeds the value of 1.5, suggesting multicollinearity is not an issue.

4.3 Hypotheses testing

22 Table 4: Hypotheses testing

Variables Model 1 Model 2 Model 3

B B B Constant -10.89 -10.29 -10.14 (1.92) (1.49) (1.44) Firm size 1.11*** 1.06*** 1.05*** (0.32) (0.17) (0.17) Size difference -0.57*** -0.52*** -0.52*** (2.20) (0.99) (0.10) Year of formation 0.22** 0.17 0.16 (0.16) (0.11) (0.10)

Average campaigns per 0.06*** 0.06*** 0.06***

alliance (0.01) (0.01) (0.01)

R&D alliance presence 0.16 0.21

(0.65) (0.64)

Marketing alliance presence

2.14*** 2.63***

(0.60) (0.95)

Equity orientated R&D alliance -0.26 -0.19 (0.93) (0.97) R&D alliance presence*marketing alliance presence 2.06** (0.89) Number of observations 3160 3160 3160 Log-Pseudolikelihood -748.67 -737.11 -736.98 AIC 1505,34 1488,22 1489,96 Wald-χ2 (P > χ2 ) 311.35 286.16 270.65 (0.00) (0.00) (0.00)

Note: Robust SEs in parentheses *p<.1

**p< .05

***p<.01

-23

0.57, p < 0.01), the year of formation (B = 0.22, p < 0.05) and the average campaigns per alliance (B = -0.06, p < 0.01). In the second model the direct influences of an R&D presence (hypothesis 1A), a marketing presence (hypothesis 1B) and a contractual oriented R&D (hypothesis 2) were assessed, while controlling for the effects of the average campaigns per alliance, the firm size, the size difference and the year of formation. The overall model is significant, χ2 (7, N = 3160) = 286.16, p < 0.01. For the independent variables a significant effect was found for a marketing alliance presence (B = 2.14, p < 0.01). The presence of a marketing alliance between dyads did lead to significantly more conflict. Therefore hypothesis 1B is supported. Hypothesis 1A was not supported; an R&D alliance presence did have a positive effect on the amount of conflict arising, but this effect was not significant (B = 0.16, p = 0.80). Hypothesis 1C was rejected; contrary to the expectation that the presence of an R&D alliance would induce more conflict than a marketing alliance presence the opposite effect was found. The effect of a marketing alliance presence on the level opportunistic behavior (p < 0.01) was a more stronger (and significant) predictor of the level of conflict compared to an R&D alliance presence, which did not render significant effects (p = 0.80). Further, hypothesis 2 was not confirmed; an equity oriented R&D alliance did lead to slightly less conflict compared to a non-equity governed R&D alliance, but the effect was not significant (B = -0.26, p = 0.78).

In the third step the moderating effect of ‘R&D alliance presence*marketing alliance presence’ (hypothesis 3) was added to the model. This did not significantly alter the fit of the model, χ2 (8, N = 3160) = 270.65, p < 0.01. The influence of the moderating variable ‘R&D alliance presence*marketing alliance presence’ on the level of conflict was significant (B = 2.06, p < 0.05). The interaction is graphically depicted in Figure 2.

24

25

5. DISCUSSION

5.1 R&D and marketing alliances influence on conflict

In the theoretical analysis it was predicted that companies forming a marketing alliance or R&D alliance would engage in more conflict compared to companies not cooperating on a strategic level. It was theorized that because in a strategic alliance the cooperation is much more intense (i.e. resources are shared on a strategic level), the cooperating firms face higher risks of the partner acting opportunistically and ending in a conflict. The term relational risk: the probability of not having satisfactory cooperation (Das & Teng, 1996), was used to stress the additional risk companies face when forming an alliance over and above the risks related to the normal performance risks. The results show that although both an R&D alliance and a marketing alliance elicited more conflict than between companies that did not form a strategic alliance, the conflict only significantly increased when a marketing alliance was formed. Different theories can help to explain why these different effects were found.

26

Further, the resource dependence theory and the bargaining power perspective help to explain the difference in conflict between R&D alliances and marketing alliances. Based on these literatures it was theorized alliances end up in conflict and ultimately are terminated because power imbalances occur or the desired knowledge is acquired. If such a change occurs, the need for cooperation will be diminished. According to Beamish and Inkpen (1995) however, these power imbalances are more likely to occur with resources that are easily mobile and imitable. When a resource is not mobile and difficult to imitate (i.e. when the shared resource is knowledge) power imbalances are far less likely to occur. Because in an R&D alliance knowledge is the most valuable shared resource a shift in bargaining power is less likely to take place, leading to a lower probability of generating conflict. Companies remain dependent upon one another and the dependence remains at its status quo. Even when important knowledge is transferred and acquired in an R&D alliance the power balance may not shift. This is because the goal of an R&D alliance is more oriented to accessing and learning from partners continuously rather than acquiring knowledge that exist in a certain point in time (Beamish & Inkpen, 1995). This continuous drive to share knowledge in R&D alliances strengthens the equality in dependence and leads to less conflict in comparison to a marketing alliance where knowledge is more likely to be acquired at a certain point in time.

27 5.2 R&D alliance governance influence on conflict

Although R&D alliances compared to marketing alliances are more conflict avoiding because R&D partners have a more long term perspective and are likely to be continuously mutually dependent, conflict did arise in several R&D alliances. Based on arguments related to the RBV and the TCE it was theorized this conflict could be minimized by using an integrative governance structure. This would safeguard companies from opportunistic behaviors that lead to conflicts. The results show however an equity governed R&D alliance does not significantly decrease conflict.

An important reason for this could be that a more integrated governance form does still not sufficiently safeguard the alliance partners from opportunistic behaviors because of the intense competition. Intense competition was present in the dataset because the sample consisted of companies forming a horizontal R&D alliance in the same industry. All the companies in the sample fell within the same first three digit SIC code (283) and were divided into only two different SIC code groups in total; SIC codes 2834 (biological products) and 2836 (pharmaceutical preparations). This raised the intensity of the competition because alliance partners could be perceived to be competitors as well (Wang & Zajac, 2007; Noseleit & De Faria, 2013). This higher competitive intensity implies competitors’ knowledge bases were alike; they were very well capable of (voluntarily and involuntarily) appropriating the knowledge of their partners (Park & Russo, 1996). This is because the propensity of a company to appropriate knowledge is not only related to the difficulty or complexity of the knowledge itself, but also to its absorptive capacity, i.e. a companies’ ability to identify, appreciate and assimilate such knowledge (Cohen & Levintahl, 1990). Because the competition in the market is high the knowledge bases are alike; which in turn implies alliance partners have a high understanding of their partner’s knowledge. This higher relative absorptive capacity in combination with the high value for competitors of appropriating knowledge increases the tendency to act opportunistically and end up in conflict. This tendency may be even stronger when competitors use a joint venture instead of a non-equity governed alliance. Simply because the communication is more intense and knowledge outflows and inflows between the companies are more profound when a joint venture is formed, there is more knowledge in the alliance to involuntarily appropriate.

All in all, the theorized positive effect of a joint venture on the level of conflict may not be present in the studied dataset because of intense competition. Intense competitions serves as an extra motivation to make use of the intense cooperation as a means to involuntarily appropriate and use knowledge of a partner. An equity joint venture is still not sufficiently integrated to safeguard the company from appropriability threats.

5.3 R&D and concomitant marketing alliances influence

28

the value chain the threat of misappropriation of knowledge would decrease. Next to this it was hypothesized the partner experiences could be build up more quickly, also decreasing the overall tendency of partners to behave opportunistically. Based on this reasoning it was predicted that although the presence of an R&D alliance would lead to more conflict this would be less so the case when a concomitant marketing alliance is present. Contrary to these expectations this moderating effect was not found. Instead, it was found a marketing alliance determines the nature of the effect an R&D alliance has on the level of conflict. If there is no marketing alliance, having an R&D alliance does not lead to significantly more conflict, but if a marketing alliance is present, an R&D alliance presence leads to less conflict. Different mechanisms help to explain how both an R&D alliance presence and a marketing alliance presence influence each other differently.

First, an explanation is presented for the lower level of conflict in an alliance covering both R&D and marketing compared to an alliance only covering marketing functions. The basis of this explanation builds on the reasoning that the governance of a marketing alliance in general is not integrated sufficiently. As put forward in the previous paragraph, the governance mode of marketing alliances alone often does not offer an adequate safeguard for the appropriability threats and opportunistic behaviors. In turn, this leads to high levels of conflict. If however an R&D alliance is also present, the alliance is expected to be more integrated, or at least the R&D part. This integrated R&D part signifies trust and willingness to share information to the marketing alliance department. This trust and willingness to share information throughout both alliance functions is further enhanced by the theorized effects of taking away competition and increasing partner specific experience. This way, the positive experiences generated by the R&D department are cascaded to the marketing function. The experienced trust in the R&D alliance that is broadened to the marketing alliance can be viewed as a safeguard mechanism that protects the marketing alliance from opportunistic actions. The integrated R&D alliance functions as a substitute for an integrated marketing governance form.

Second, if the previous reasoning is followed, one would expect adding a marketing function to an R&D alliance would decrease the level of conflict as well. The effect would be less strong, because an R&D alliance is typically already governed in a more integrated way, thereby being less affected by a marketing alliance because it is often more contractually governed. But there would still be a further decrease in the level of opportunism because the competition is taken away further down the value chain and experience is built up more quickly. However, the results show an R&D alliance alone does not lead to a significant different level of conflict compared to having both an R&D alliance and marketing alliance.

29

Specifically, it could be that although an additional marketing alliance decreases the level of conflict by eliminating the competition and increasing partner experience, this effect is offset by increased difficulties related to managing the larger alliance scope. Teece (1992) suggests the protection of technological knowledge becomes more challenging with increases in alliance scope. When the alliance scope increases, the tacit knowledge embedded in operating routines (i.e. both marketing and R&D) must be exposed to the alliance partner if joint operations are to cooperate effectively and efficiently. An additional function within the alliance implies that a product or service has to be brought through many more points of contact between the partner firms, with a concomitant reduction in control over information flows across the relevant organizational boundaries, and hence, a higher opportunistic threat and chances on conflict.

30

6. LIMITATIONS & RECOMMENDATIONS FOR FUTURE RESEARCH

As with any research, this one is bounded by limitations. Of these limitations, several are related to the methodology of the research. First, previous alliance experience is not taken into account in the analysis. Previous alliance experience with a partner has a positive effect on the level of experienced trust and the overall cooperative behavior (Kogut, 1989). Consequently, it decreases the level of opportunistic behaviors and the resulting conflict. For this reason the effect of previous alliance experience should have been controlled for. Due to absence in the dataset however, this variable was left out of the analysis.

Second, overall the measures are crude for both the dependent and independent variables. Future work would benefit from more fine grained and multidimensional measures. Conflict was theorized to arise through opportunistic behaviors and was measured as the total number of patent litigation campaigns started between two companies. However, not all conflicts or opportunistic actions are directly related to and represented in patent litigations. Next to this the severity or impact of each conflict (i.e. patent litigation) in this thesis was assumed to be the same. In reality however conflicts will differ in their importance. Litigations will for example differ based on court costs but the importance is also dependent upon the knowledge or technology behind the patent. Future research should incorporate more sophisticated measures of conflict and assess their individual effects. Further, this research did not measure the specific individual effects of the independent variables that were theorized. For future research it is interesting to measure individually the effects of competition elimination, building up partner-specific experience, transcending values of trust throughout the organization and governance integration. Only by measuring these effects individually in one model scholars can make confident claims about their direct (opposing) effects and maybe even more importantly, see how they interact.

Third, this study did not take into account whether the R&D alliance partners operated in different countries. This may affect the degree of coopetition. When firms have their marketing operations in different countries the competition further down the value stream is reduced. Alliance partners may perceive each other as less of a threat, share knowledge more freely and be less worried about unwanted appropriability of knowledge.

31

32

7. IMPLICATIONS & CONCLUSIONS

The findings in this paper have several conclusions and implications for organizations. The increases in patent litigations in the last decades highlight the importance of analyzing variables that predict the tendency to end up in a patent litigation (Bessen & Meurer, 2005a; Sharipo, 2001). This research assessed several of those variables. Specifically, this thesis analyzed how R&D alliance presence, R&D alliance governance structure and a concomitant marketing alliance influenced conflict. The following research question was researched:

How does the presence and governance of an R&D alliance influence the level of conflict between organizations and how is the relationship between R&D alliance presence and conflict affected by a concomitant marketing alliance?

In order to answer this research question three sub research questions were formulated. These questions divide the research question in different sub questions that were analyzed within this thesis. These three sub research questions are answered in the following three paragraphs.

- How does the presence of an R&D and marketing alliance impact the level of conflict?

- How does the R&D governance structure choice influence the level of conflict?

- How does a concomitant marketing alliance influence the level of conflict?

7.1 R&D and marketing alliance impact the level of conflict

33

consequence, the alliance needs to be managed differently. Important managerial implications are that managers should focus more on the higher potential for opportunism and appropriation hazards when establishing a marketing alliance.

7.2 R&D governance structure choice influence the level of conflict

Next to the high level of conflict in marketing alliances this research points out that R&D alliances are also still a risky game. The R&D governance structure did not significantly influence the level of conflict in the alliance. This research suggests this decreasing effect has not been found because of the fierce market competition. Due to fierce market competition even an integrated R&D governance structure does not offer sufficient protection.

Regarding the TCE and RBV literatures this research takes a critical view on the subject of the suitability of joint ventures as conflict decreasing governance mechanisms. Although the factors the TCE and RBV use to determine the level of transaction costs are suitable (i.e. tacitness, uncertainty, opportunism, bounded rationality and knowledge intensity), their nature and strength are not correctly matched with a governance form that can properly deal with them. In the pharmaceutical industry more integrated governance forms seem to be necessary if conflict is to be minimized. Managers of firms in the pharmaceutical industry need to pay extra attention to appropriability hazards and carefully assess if they can structure their alliances in a way that sufficiently minimizes conflict.

7.3 Influence concomitant marketing alliance

Further, this thesis sheds light on how a concomitant marketing alliance influences the total level of conflict between firms. It was theorized that although an R&D alliance presence would induce more conflict, this would be less so the case when an additional marketing alliance was present. This effect was not supported by the empirical analysis. Increasing the scope of the R&D alliance by adding the marketing function did not significantly influence conflict within the alliance. The analysis showed however a different interaction effect. It was found a marketing alliance determines the nature of the effect an R&D alliance has on the level of conflict. In absence of a marketing alliance an R&D alliance does not significantly influence conflict, but if a marketing alliance is present an R&D alliance presence significantly decreases conflict.

This analysis attempts to start unraveling the opposing forces that decrease and increase conflict through opportunistic actions. Specifically, three different influences are theorized to be present in alliances covering both R&D and marketing.

34

- A larger alliance scope (i.e. covering both marketing and R&D functions) can act as a safeguard by taking away competition and increasing partner specific experiences. This lowers the total level of conflict.

- On the other hand a larger alliance scope increases the potential for conflict and opportunistic actions because knowledge has to be shared throughout more operating routines (i.e. both marketing and R&D).

These variables affect the total level of conflict in contradicting ways. Further analysis is required to assess which specific mechanisms underlie these relationships and how the different variables interact.

Overall, the results in this study stress that managers in the pharmaceutical industry should diligently assess how they construct their alliances. Marketing alliances need to be built and undertaken with care as these types of alliances can bring a high level of conflict. This research suggest managers could as a means to decrease the level of conflict broaden the scope and develop the products with its marketing partner as well. In this situation the more integrated R&D alliance structure extends its positive effects of trust to the marketing alliance. Further, managers can be seen as active decision makers who influence the shape and form of the marketing alliance, directly affecting the level of conflict arising.

35

8. REFERENCES

Axelrod, R. 1984. The Evolution of Cooperation. Basic Books, New York.

Beamish, P. W., & Inkpen, A. C. (1995). Keeping international joint ventures stable and profitable. Long Range Planning, 28(3), 2-36.

Bengtsson, M., & Kock, S. (2000). Coopetition in business networks—to cooperate and compete simultaneously. Industrial Marketing Management, 29(5), 411–426.

Bessen, J.E., & Meurer, M. (2005). The Patent Litigation Explosion. Boston Univ. School of Law Working Paper, 5.

Bleeke, J., & Ernst, D. (1991). The way to win cross-border alliances. Harvard business review, 69(6), 127-135.

Blumberg, B. F. (2001). Cooperation contracts between embedded firms. Organization studies, 22(5), 825-852.

Bucklin, L. P., & Sengupta, S. (1993). Organizing successful co-marketing alliances. The Journal of Marketing, 32-46.

Calvin, J. A. (1998). Regression Models for Categorical and Limited Dependent Variables. Technometrics, 40(1), 80-81.

Chen, H., & T. J. Chen. (2003). Governance structures in strategic alliances: Transaction cost versus resource-based perspective. Journal of World Business, 38, 1–14.

Cloodt, M., Hagedoorn, J., & Van Kranenburg, H. (2006). Mergers and acquisitions: Their effect on the innovative performance of companies in high-tech industries. Research Policy, 35(5), 642-654.

36

Das, T. K., & Teng, B. S. (1996). RISK TYPES AND INTER‐FIRM ALLIANCE STRUCTURES*. Journal of management studies, 33(6), 827-843.

Das, T.K., & Teng, B.S. (2000). Instabilities of Strategic Alliances: An Internal Tensions Perspective. Organizational science, 11(1), 77-101.

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., & Rahman, N. (2010). Determinants of Partner opportunism in Strategic Alliances: A Conceptual Framework. Journal of Business and Psychology, 25(1), 55-74.

Deutsch, M. (1977). The resolution of conflict: Constructive and destructive processes. Yale University Press.

Ernst, H., Lichtenthaler, U., & Vogt, C. (2011). The Impact of Accumulating and Reactivating Technological Experience on R&D Alliance Performance. Journal of Management Studies, 48(6), 1194-1216.

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.

Gardner, W., Mulvey, E. P., & Shaw, E. C. (1995). Regression analyses of counts and rates: Poisson, overdispersed Poisson, and negative binomial models. Psychological bulletin, 118(3), 392.

Geringer, J. M., & Woodcock, C. P. (1995). Agency costs and the structure and performance of international joint ventures. In International Joint Ventures: Economic and Organizational Perspectives. Springer Netherlands.

Gnyawali, D.R., He, J., & Madhavan, R. (2008). Co-opetition: Promises and challenges. 21st century management: A reference handbook, 386-398.

37

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.

Hagedoorn, J. (1993). Understanding the rationale of strategic technology partnering: inter-organizational modes of cooperation and sectoral differences, Strategic Management Journal, 14(5), 371-385.

Hagedoorn, J. (2002). Inter-firm R&D partnerships: an overview of major trends and patterns since 1960. Research Policy, 31(4), 477-492.

Hagedoorn, J., & Duysters, G. (2002). External sources of innovative capabilities: the preferences for strategic alliances or mergers and acquisitions. Journal of management studies, 39(2), 167-188.

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

Hennart, J. F. (1988). A transaction costs theory of equity joint ventures. Strategic management journal, 9(4), 361-374.

Hoang, H., & Rothaermel, F. T. (2005). The effect of general and partner-specific alliance experience on joint R&D project performance. Academy of Management Journal, 48(2), 332-345.

Ismail, N., & Jemain, A. A. (2007). Handling overdispersion with negative binomial and generalized Poisson regression models. In Casualty Actuarial Society Forum, 103-158.

Jensen, M, C, W, H, Meckling. 1976, Theory- of the firm: Managerial behavior, agency costs and ownership structure. J. Financial Econom, 3, 305-360.

Kaufmann, P. J., & Stern, L. W. (1988). Relational exchange norms, perceptions of unfairness, and retained hostility in commercial litigation. Journal of Conflict Resolution, 32(3), 534-552.

Kennedy, P. (1992). A Guide to Econometrics. Oxford: Blackwell.

38

Kogut, B. (1989). The stability of joint ventures: Reciprocity and competitive rivalry. The Journal of Industrial Economics, 38(2), 183-198.

Lavie, D. (2007). Alliance portfolios and firm performance: A study of value creation and appropriation in the US software industry. Strategic Management Journal, 28(12), 1187-1212.

Lanjouw, J. O., & Schankerman, M. (2004). Protecting Intellectual Property Rights: Are Small Firms Handicapped?*. Journal of Law and Economics, 47(1), 45-74.

Macneil, I. R. (1974). The many futures of contracts. Southern California, Law Rev. 47, 691-816.

Mazzola, E., & Perrone, G. (2013). How the purpose of the inter-firm relationships influences the choice of the governance form: evidence from the machine tool and the pharmaceutical industries. Technology Analysis & Strategic Management, 25(9), 1009-1025.

Noseleit, F., & de Faria, P. (2013). Complementarities of internal R&D and alliances with different partner types. Journal of Business Research, 66(10), 2000-2006.

Osgood, D. W. (2000). Poisson ‐ based Regression Analysis of Aggregate Crime Rates. Journal of Quantitative Criminology, 16, 21‐44.

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.

Pan, Y., & Jackson, R. T. (2008). Ethnic difference in the relationship between acute inflammation and serum ferritin in US adult males. Epidemiology and infection, 136(3), 421-431.

Pangarkar, N., & Klein, S. (2001). The impact on alliance purpose and partner similarity on alliance governance. British Journal of Management, 12(4), 341–353.

39

Pfeffer, J., & Salancik, G. R. (1978). The external control of organizations: A resource dependence perspective. Stanford University Press.

Phelps, C. C. (2003). Technological exploration: A longitudinal study of the role of recombinatory search and social capital in alliance networks (Doctoral dissertation, New York University, Graduate School of Business Administration).

Phelps, C. C. (2010). A longitudinal study of the influence of alliance network structure and composition on firm exploratory innovation. Academy of Management Journal, 53(4), 890-913.

Pisano, G. P. (1990). The R&D boundaries of the firm: An empirical analysis. Administrative Science Quarterly,35, 153–176.

Rahim, M. A. (2011). Managing conflict in organizations. Transaction Publishers.

Ramstad, E., & Dvorak, P. (2006). Behind Sony-Samsung rivalry, an unlikely alliance develops; electronics giants join forces on flat-panel technology. Wall street journal, January 2006, http://online.wsj.com/articles/SB113625623819236122, Last accessed at 02/11/2014.

Sampson, R. C. (2004). The cost of misaligned governance in R&D alliances. Journal of Law, Economics and Organization, 20(2), 484–526.

Sampson, R. C. (2007). R&D alliances and firm performance: The impact of technological diversity and alliance organization on innovation. Academy of Management Journal, 50(2), 364-386.

Schögel, M., Herhausen, D., & Schmitz, C. (2011). Developing the Market Together: Scale Development and Performance Implications of Co-Marketing Capability, In Delivering Value in Turbulent Times, AMA Educators Proceedings, 22. New York: Curran Associates, Inc.

Shapiro, S. S., & Wilk, M. B. (1965). An analysis of variance test for normality (complete samples). Biometrika, 52(3), 591-611.