0 Master Thesis
Changing partner relationships in alliances: Is the effect of steering
committees set by boundaries?
Marèl Verschoor *, supervised by Marvin Hanisch, co-assessed by Pere Arque-Castells
THESIS INFO: ABSTRACT
In the biopharmaceutical industry, the costs of drug development and their risk of failure are increasing. Therefore, firms form alliances to create successful innovations and share the risks. However, the alliance environment is globalizing and diversifying, leading to uncertainties regarding their governance structures. Previous literature has focused on formal and informal mechanisms that improve control at the time the contract is drawn up. However, research about mechanisms that simultaneously control and coordinate the alliance is limited. Therefore, this study focuses on the hybrid function of joint steering committees. I propose that in the presence of joint steering committees, firms more often achieve innovation success. Furthermore, I argue that this relationship is positively moderated by size similarity and geographical proximity between partners. To test these assumptions, I used a data sample of 252 biopharmaceutical alliance contracts from 2005 to 2008. The results show no evidence that size similarity has positive impact on achieving innovation success when the alliance established a joint steering committee. Whereas a high level of geographical proximity causes a stronger positive relationship between joint steering committees and innovation success. The findings contribute to the alliance literature concerning innovation success, joint steering committees and firm attributes.
Keywords: Alliance, Joint Steering Committee, Firm Attributes, Size Similarity, Geographical Proximity, Innovation Success
July 22, 2020
Word count: 14,495
* Student number: S2835533, E-mail: m.verschoor.2@student.rug.nl. Faculty of Economics and Business, University of Groningen, The Netherlands.
1 CONTENTS
1. INTRODUCTION ... 2
2. THEORETICAL BACKGROUND AND HYPOTHESES ... 4
2.1 Innovation Success in Alliances ... 4
2.2 Formal and Informal Governance Mechanisms ... 6
2.3 Joint Steering Committees ... 7
2.4 Hypotheses Development ... 9
2.4.1 The role of joint steering committees in alliances ... 9
2.4.1 The moderating effect of size similarity and geographical distance ... 10
3. METHODOLOGY ... 14
3.1 Sample and Data ... 14
3.2 Data Measurement ... 15
3.3 Analytical Method ... 18
4. RESULTS ... 19
4.1 Descriptive Statistics and Correlations ... 19
4.2 Regression Results and Hypotheses Testing ... 22
4.3 Additional Analyses ... 24 4.4 Robustness Check ... 26 5. DISCUSSION ... 27 5.1 Theoretical Implications ... 27 5.1.1 Additional insights ... 29 5.2 Managerial Implications ... 30
5.3 Limitations and Future Research ... 30
6. CONCLUSION ... 31
7. REFERENCES ... 32
2
1. INTRODUCTION
More than 7,000 medicines and treatments are currently being developed worldwide (Deloitte, 2015). The costs of bringing a drug to the market, including the costs of failed development, have increased with 30% since 2010. This trend seems to be continuing because of the relentless pressure on healthcare budgets and sales, combined with the growing costs of discovering, researching, developing and launching innovations (Deloitte, 2015). About 12% of the drugs that participate in clinical trials succeed to be launched to the market (Deangelis, 2016). Only 20% of them make significant profits, making less than 3% of all tested drugs sufficiently profitable.
With the cost of drug development being so high, companies engage into strategic alliances because the benefits of risk sharing outweigh the disadvantage of sharing the profit with partners (Hughes & Weiss, 2007). In addition, the circumstances of these alliances have changed to address the current problems. It has increased the diversity and objectives of partnerships to deliver value in biomedical research (Deloitte, 2015). Furthermore, the number of alliances have increased by 25% per year since 1985 (Hughes & Weiss, 2007), underlining the increasing importance of successfully managing alliances, especially in the medical industry.
According to the transaction cost economics (TCE), firms face challenges in monitoring, controlling and managing their alliance (Williamson, 1979a, 1985b). To overcome these challenges it is important to reduce the appropriation concerns and manage uncertainties by exercising control and coordination (Reuer & Devarakonda, 2016). In the past decade, researchers have begun to investigate the effect of governance mechanisms on alliances (Claro, Hagelaar, & Omta, 2003; Zajac & Olsen, 1993). On the one hand, formal governance mechanisms focus primarily on using contractual safeguards for expected contingencies at the time the contract is drawn up (Yu, Liao & Lin, 2006). Informal governance mechanisms on the other hand, focus on the relational perspective to effectively execute coordination during the alliance. Finally, a form of hybrid governance mechanism is joint steering committees, which allow to design control structures specified for each alliance, while simultaneously monitoring activities and locating decision-making during the alliance (Reuer & Devarakonda, 2016).
3
changing the structures of alliances (Anderson, 1990; Deloitte, 2015). As a result, firms’ choices regarding suitable governance structures are increasing, however not all governance mechanisms fit in every situation (Yu, Liao & Lin, 2006). Despite these findings, scholars have accumulated little insight into the effect of these trends on steering committees. A supplementary examination thereof can further prove the state of research in this area. This study advances prior literature (Anderson, 1990; Gulati, 1998; Gulati & Singh, 1998; Reuer & Devarakonda, 2016a, 2018b) by researching the relationship between joint steering committees and innovation success in alliances and its boundary conditions. Therefore, the following research question has been formulated:
“Does the presence of a joint steering committee lead to a higher chance of innovation success in alliances and is the effect bounded?”
This research addresses this theoretical gap by proposing several hypotheses. The first hypothesis suggests that joint steering committees have a positive relationship with innovation success. A joint steering committee in an alliance reduces transaction costs and creates a more efficient and effective relationship resulting in a higher chance of innovation success. Furthermore, the second hypothesis proposes that this relationship is positively moderated by size similarity. Alliances with partners similar in size experience more similar perceptions, resulting in fewer appropriation concerns and more effective coordination (Gronin & Weingart, 2007). Consequently, the effectiveness of steering committees increase the chance of innovation. Finally, the third hypothesis suggests that geographical proximity between partners positively moderates the relationship between joint steering committees and innovation success. Firms that are geographically close have more similar work visions and communicate more effectively, resulting in fewer appropriation concerns and improved coordination in steering committees.
4
findings provide a better understanding of the importance of several circumstances that affect the direct relationships between joint steering committees and innovation success.
This paper contributes to the alliance literature by examining the effect of joint steering committees. It builds on the transaction cost economics, claiming that minimizing the costs of monitoring, controlling and managing the exchange creates the optimal alliance structure. The theory is advanced by demonstrating that by considering the bounding effect of geographical distance, firms can make more accurate predictions about the usefulness of joint steering committees in their alliance.
2. THEORETICAL BACKGROUND AND HYPOTHESES
This section introduces relevant theory on innovation success in alliances, governance mechanisms and joint steering committees in particular. Building on the transaction cost economics, I will further elaborate on the effects of joint steering committees and its boundary conditions, resulting in the proposition of several hypotheses.
2.1 Innovation Success in Alliances
In recent years, the number of strategic alliances is rapidly increasing and, especially in the medical industry, companies are relying more and more on the formation of those alliances to create successful innovations (Hughes & Weiss, 2007). Several studies argue that alliances provide important resources for companies to access and absorb external knowledge or to gather the knowledge required to create and develop new products (e.g. Hamel, 1991; Kale, Singh, & Perlmutter, 2000; Reuer & Devarakonda, 2018). However, by integrating and utilizing external knowledge in alliances, companies face challenges in transferring and protecting valuable and proprietary knowledge (Giarratana & Mariani, 2014). These challenges may lead to increased transaction costs in alliances.
5
In line with the transaction cost theory, there are two important facets that support the reduction of transaction costs in alliances, namely coordination and control (Gulati, Wohlgezogen and Zhelyazkov, 2011; Schilke and Lumineau 2018). Both facets focus on different types of issues. First, contractual control defines the rights and obligations of the parties involved to overcome opportunism (Schilke & Lumineau, 2018). Williamson (1985) defines opportunism as “self-interest seeking of a strategic (i.e. secretive, deceptive, or insidious) nature undertaken in order to redirect profits from vulnerable partners”. Contractual control helps partners to comply with a minimum amount of deviant behavior by exercising authority or power mechanisms (Carson, Madhok & Wu, 2006).
In an alliance, behavioral uncertainty and contracting problems lead to these appropriation concerns (Gulati & Singh, 1998). These concerns about opportunistic behavior include shirking, appropriating partners’ resources, communicating wrong information, hidden agendas, and delivering inadequate products and services (Das & Teng, 1999). Therefore, governance mechanisms are necessary to protect the alliance (Judge & Dooley, 2006).
Next to the reduction of appropriation concerns, contractual control also manages potential moral hazards, aligns incentives of the parties and monitors problems between organizations (Reuer & Devarakonda, 2018). After all, the goal of an alliance is to expand into a new market or develop new products while minimizing the transaction costs. Therefore, establishing good control mechanisms will make achieving the desired goals more predictable. However, when the risk of opportunism is high, the governance mechanisms become more elaborate and the transaction costs increase (Judge & Dooley, 2006). Therefore, it is important to find the right balance.
6
efficient way (Gulati et al., 2012). When coordination fails, alliance partners can doubt the feasibility of the alliance, reduce their commitment and give up the effort (Mohr & Spekman, 1994).
2.2 Formal and Informal Governance Mechanisms
To address control and coordination, firms use different formal and informal governance mechanisms. These mechanisms focus on routinizing activities or improving non-routine activities, as continuous learning, risk management or innovation skills (Sitkin, Sutcliffe, & Schroeder, 1994). Irrespectively of their focus, these mechanisms are used to make the achievement of the alliance goals more predictable and to yield more certain results in the most efficient way (Gulati et al., 2012; Reuer & Devarakonda, 2018). Previous literature presents a wide variety of those mechanisms to diminish the transaction costs of alliances dealing with contracting problems and uncertain environments. A few formal and informal mechanisms that stand out above the rest are discussed in this paper.
A broadly used form of formal governance mechanism are contractual safeguards (Popper and Zenger, 2002). Contractual safeguards are defined as instruments created before the alliance is formed and are included in the contract to prevent opportunistic behavior (Judge & Dooley, 2006). However, this method has its limitations due to a reduced focus on monitoring the alliance. In addition, contract incompleteness can emerge during the implementation of the alliance because contracts can only provide safeguards for expected contingencies at the time the contract is drawn up (Williamson, 1985). Therefore, there are limits to the ability of these safeguards to meet the needs for coordination and control during the alliance (Williamson, 1991).
Another formal governance mechanism is equity control, which involves ownership of the stakes of the alliance firms, varying from no equity to 50:50 to a majority equity interests (Judge & Dooley, 2006). This form of ownership creates a shared authority by providing board representation and residual control rights (Gatignon & Anderson, 1988; Reuer & Devarakonda, 2016). Consequently, the goals and objectives of the partners are more aligned and opportunistic behavior is reduced. However, it is important that both parties have an equity investment, otherwise opportunism is more likely to occur (Das & Teng, 2000; Judge & Dooley, 2006).
7
Chan & Chan, 2007). Informal governance mechanisms are non-contractual forms of governance to provide coordination and mitigate potential opportunism. The outcomes of these mechanisms cannot be specified in advance because it depends on the individuals and their interactions (Hoetker & Mellewigt, 2009).Informal governance mechanisms can include teams or task forces (Schrader, 1991), direct contact through meetings (Martinez and Jarillo, 1989), mechanisms to enable shared decision making (Saxton, 1997) and informal dispute resolution systems (Kale et al., 2000).
An important consequence of informal governance mechanisms is partner trustworthiness. Partner trustworthiness is defined as the degree to which the trusted firm has a positive attitude towards the trusting firm’s goodwill and reliability in a risky collaboration (Ring & Van de Ven, 1992). Trust between firms in strategic alliances stimulates desirable behavior (Madhok, 1995), reduces the necessity for extensive contracts (Larson, 1992) and facilitates dispute resolution (Das & Teng, 1998; Ring & Van de Ven, 1994). Moreover, TCE claim that uncertainties and unforeseen environmental and organizational contingencies are easier to deal with in the presence of trust by increasing operational flexibility (Mjoen & Tallman, 1997; Zaheer & Venkatraman, 1995). Furthermore, trust can improve communication and increase the flow of information, which has a positive influence on innovation success in alliances (Das & Teng, 1998; Zaheer, McEvily & Perone, 1998). Therefore trust in alliances reduces the transaction costs concerning control and coordination (Gulati, 1995; Williamson, 1985).
2.3 Joint Steering Committees
A less researched, hybrid governance mechanism that expands beyond procedural control and maintains flexibility to adapt to changing circumstances are joint steering committees (Mayer & Argyres, 2004). Joint steering committees are defined as tools that formalize joint administrative control by both partners through a specific defined, board-like structure and facilitate adaptation in a coordinated manner during the alliance (Reuer & Devarakonda, 2016a, 2018b). Hence, their authority is derived from contracts, but they also have some functions from hierarchical organizational structures.
8
activities, localize decision-making, and facilitate information sharing (Liebeskind, 1997). Consequently, they reduce unstructured communication and offer guidance in the event of uncertainties (Reuer & Devarakonda, 2016). Moreover, firms can design these committees at their own discretion (e.g. design, functions and performance) and define it in the contract (Smith, 2005). These steering committees differ from ad hoc task forces or project teams because of the contractual definition and their higher degree of authority and decision-making (Hoetker, & Mellewigt, 2009).
Prior research into governance mechanisms in alliances emphasize that contractual safeguards are the primary formal governance mechanism available to alliances (Dyer & Singh, 1998; Poppo & Zenger, 2002). However, little attention has been paid to the monitoring and controlling needs arising during an alliance (e.g. Gulati & Singh, 1998; Hoetker & Mellewigt, 2009). Lumineau and Schilke (2018) researched the relationship between contractual control and coordination and innovation success, nevertheless they did not evaluate the role of joint steering committees. Furthermore, Williamson (1985) states that contracts can only provide safeguards for expected contingencies at the time the contract is drawn up, which limits the ability of these contractual safeguards regarding coordination and adaptation during the alliance.
The theory of this research looks beyond the current administrative controls and focuses on the relationship between the governance mechanism joint steering committees and innovation success that can be performed in hybrid forms. Lumineau and Schilke’s (2018) research into contractual control and coordination forms a basis for this study. Furthermore, I build on the work of Williamson (1991) and on previous research on hybrid mechanisms emphasizing that administrative controls can facilitate coordinated adaptation in alliances (Reuer & Devarakonda, 2016). In addition, I expand the research stream of Devarakonda and Reuer (2018) about the positive relation between joint steering committees and knowledge transfers, by examining alliances’ innovation success in the presence or absence of joint steering committees. Hence, this study suggests that executing control and coordination in the form of joint steering committees increases the chance of successful innovations in an alliance.
9
Consequently, this study proposes several hypotheses in which I examine the relationship between joint steering committees and innovation success and how it is bounded by size similarity and geographical proximity.
2.4 Hypotheses Development
2.4.1 The role of joint steering committees in alliances
Joint steering committees are committees that have been previously formally agreed on and possess bounded authority. They are designed to jointly review and monitor the progress and performance, and make decisions about alliance activities (Reuer & Devarakonda, 2016; Hoetker & Mellewigt, 2009). Through this collaboration, the committees have the opportunity to reduce the appropriation concerns and help coordinate the alliance efficiently, which can steer the alliance towards a greater chance of innovation success.
There are several effects arising from assigning a joint steering committee that improve the coordination and help control an alliance. The first effect is that joint steering committees create trust because the relationships of the involved firms become strongly intertwined with each other (Gil, 2009). Trust limits the discretion of individuals and aligns their activities with project results. In other words, trust between firms refers to the confidence that the vulnerabilities of the other will not be exploited by a partner (Barney and Hansen, 1994). Consequently, the risk of opportunistic behavior reduces, which decreases transaction costs and facilitates adaptive responses (Poppo & Zenger, 2002). An example of a situation where trust is needed is when incomplete contracts occur. In this case, it is essential to create trust between partners and overcome control problems. Hence, when the contracting costs reduce, joint steering committees are a suited response to the incompleteness (Wang & Chen, 2006).
The second positive effect arising from joint steering committees is that they allow partners to manage bilateral dependencies by coordinating decisions about boundary-spanning resources and adapting agreed project plans to the changing circumstances of the alliance (Reuer & Devarakonda, 2016; Williamson, 1991). Therefore, firms retain the ability to steer the interactions of partners and change the behavior of activities within the alliance. They enable close decision-making powers to approve and enforce decisions and address conflicts between the parties (Smith, 2005). Hence, firms can efficiently respond to unforeseen events and prevent small conflicts from becoming larger disputes.
10
knowledge sharing, both tacit and codified, is required to meet alliance goals (Reuer & Devarakonda, 2018). The tacit form is especially important when alliances try to achieve innovation success (Arora, 1995). Therefore, firms have to find a balance in sharing the required knowledge in agreed areas and limit the unintended knowledge spillovers in other areas. Joint steering committees can on the one hand serve as a useful interface to share experiences and share tacit knowledge (Reuer & Devarakonda, 2018). On the other hand, they can mitigate appropriation concerns due to representation of both firms and joint decision-making processes. Therefore, steering committees can serve as a screen for the knowledge processes between partners and cause a reduction of monitoring and controlling costs (Aubert & Kelsey, 2000).
Fourth, these steering committees may also create inter-organizational routinized interactions and develop a shared language (Reuer et al., 2002). It allows firms to address misunderstandings and support alliance adjustment processes. Subsequently the contract costs reduce and ex post efficiency improves, contributing to the desired outcome of the collaboration.
To sum up, joint steering committees create trust, address conflicts, help to respond to contingencies and develop routines and a shared language. These positive effects of joint steering committees reduce the appropriation concerns related to opportunism and create a more efficient and effective relationship needed to create innovation success.
Based on the arguments mentioned above, the following hypothesis is formed:
Hypothesis 1: Joint steering committees in an alliance have a positive relationship with innovation success.
2.4.1 The moderating effect of size similarity and geographical distance
11
Alliances more often choose to work with partners different in size to increase their innovation capacity (Almeida & Kogut, 1997). An explanation is that small firms are twice as effective in creating innovations than larger firms (Almeida & Kogut, 1997) due to higher flexibility and the ability to execute ideas more quickly (Acs and Audretsch, 1990). However, they are limited by financial and human resources to support their innovation activities (Schumpeter, 1934). Therefore they need larger firms for leveraging these resources to enhance their capabilities or access new markets (Chen & Chen, 2002). In turn, larger firms own the necessary resources, but must take advantage of the innovativeness of the small firms.
On the other hand, regardless of the innovation capacity and with a focus cooperation, firms prefer to collaborate with partners of similar size to avoid dependency and domination of larger firms and to create a fair balance of benefits (Gil, 2009). Furthermore, the level of size difference between partners is an important indicator for differences in internal structures (De la Sierra, 1995), the power of each partner (Albers, 2019) and cultural differences (Doz, 1988). It is therefore important to consider these factors in the presence of a joint steering committee. First, when internal structures differ among partners it is harder to coordinate and control the alliance. Larger firms have a more formalized and specialized structure, which limits the autonomy, decision making and flexibility of employees (Albers, 2019). Furthermore, formal firms have a more rigid structure of accountability, roles and responsibilities, more standardized methods and procedures and have less direct communication (Slade Shantz et al., 2020). Smaller organizations on the other hand have more informal and less specialized structures, meaning that employees experience a greater degree of autonomy and perform different tasks. In these informal firms, individuals experience a higher degree of authority and have more decision-making rights. Furthermore, they experience more lateral communication with flexible communication lines that are constantly changing (Slade Shantz et al., 2020). Subsequently, the different internal structures between partners make it more difficult to define the committee according to the wishes of both parties. Furthermore, the differences in authority level and decision making rights between the representatives of the firms may lead to miscommunication and conflicts in steering committees (Gronin & Weingart, 2007).
12
incentives within steering committees (Sivakumar, Roy, Zhu & Hanvanich, 2011). Unaligned incentives increase the chance of coordination problems since they experience different pacing of work across firms (Gronin & Weingart, 2007).
Finally, corporate cultural differences between partners arise due to differences in size. Smaller firms usually function as a close group of employees, while larger firms are more internally fragmented and bureaucratic (Doz, 1988). Employees of both firms are convinced that their approach is the best way to work and struggle to admit that their partner’s approach may also have benefits. These situations create poor group dynamics and increase conflict due to incompatibility and miscommunication in the steering committee (Song, 2014). Supported by Smith (2005), who claims that it is not about the number of representatives of each side in committees, but unanimity is the norm. Subsequently, poor group dynamics due to corporate cultural differences lead reduced effectiveness of joint steering committees (Gronin & Weingart, 2007).
Collaborations with firms similar in size on the other hand, are less exposed to different internal structures, power issues and corporate cultural differences. They have no fear that a larger firm will use its power to take advantage and they are more likely to cooperate effectively because of shared languages and operational routines. These committees contain more similar perceptions (Gronin & Weingart, 2007) and tacit understanding (Bierly & Gallagher, 2007). Resulting in less conflict, improved coordination and communication and better group dynamics. Consequently, the effectiveness of steering committees to create innovation success increases.
Therefore the next hypothesis is stated:
Hypothesis 2: Size similarity positively moderates the relationship between a joint steering committee and innovation success.
Currently, there is a trend among alliances to cooperate more beyond their national borders and thereby enter into alliances with partners who are geographically more distant (Sivakumar et al., 2011). This trend is a result of globalization and rapid technological progress of which the knowledge is increasingly spread all over the world (Duysters and Lokshin, 2011).
13
same country are more likely to trust each other due to similar cultures and better communication.
First, differences in cultures lead to a reduced amount of trust. The different cultures of countries can be divided in two different dimensions, individualism and collectivism (Hofstede, 1980). Western cultures are mostly individualistic and East Asian cultures collectivist. They vary in terms of communication and individual behavior in teams (Gudykunst et al., 1997). In individualistic cultures for example, the individual interest takes precedence over the group interest (Jarvenpaa & Leidner, 1999). In collectivism this works the other way around and the group interest is put first. Gudykunst et al. (1996) found that individuals of collectivist cultures have a harder time trusting others than members of individual cultures. These differences in values and beliefs will lead to people having different perceptions (Gronin & Weingart, 2007). Consequently, two firms with different interpretations and perceptions working together can create problems concerning interpersonal interactions (Brockman, 2003). Moreover, it can result in people making moves in steering committees that work against each other’s actions. This will increase appropriation concerns and make coordination of the alliance more difficult. The second factor is communication, cultural differences create higher complexity of tacit knowledge and a reduction in knowledge transfers, resulting in difficulties for managing the alliance and communicating effectively together (Duysters & Lokshin, 2011; Mowery, Oxley, Silverman, 2011). Joint steering committees function as a communication center for alliances whereby effectively guiding knowledge exchange is crucial for developing and coordinating responses to unforeseen events (Reuer & Devarakonda, 2016a, 2018b). Therefore, when partners become too distant from each other, managing and coordinating the alliance becomes more complex.
On the other side, firms that experience geographical proximity have more similar views on the necessary tasks of the committee, on how to work well together and what their priorities are due to similar cultures (Gerwin, 2004). In addition, they can communicate and guide knowledge more effectively (Mowery et al. 2011). Therefore, geographically nearby companies experience fewer appropriation concerns and can be coordinated more effectively, which reduces the transaction costs of the alliance and strengthens the effect of steering committees.
Following from these arguments, the next hypothesis is stated:
14
The conceptual model is depicted below in Figure 1. It presents the hypothesis that are examined in this research. These three hypothesis are tested at the alliance level.
Figure 1
3. METHODOLOGY
To examine the empirical validity of the assumptions and relationships from the identified theoretical frame (Acton, Irvin, & Hopkinds, 1991), I applied the theory testing research approach. The theoretic frame of reference, transaction cost theory, contains a theoretical gap in the literature that needs to be explained. For this empirical study I use quantitative research methods because it makes extensive use of reliability and validity (Golafshani, 2003), which are requirements of theory testing (Acton et al., 1991).
3.1 Sample and Data
15
uncertainties (Hagedoorn & Hesen, 2007; Reuer & Devarakonda, 2016). Finally, the biomedical industry is extensively monitored and provides loads of information about alliances and their governance mechanisms.
The core data for this study is drawn from 10-K, 8Q and 10Q filings, which are part of the US American Securities and Exchange Commission (SEC). These filings contain contracts of formed alliances in the biopharmaceutical industry, with a maximum of two companies involved and with the aim of producing and developing drugs. This database is a comprehensive source that, next to contract information, contains registered information about the companies involved (Susarla, 2012). To enrich the data of the SEC files, press releases and other publicly available databases were used to find firm information, deal history, compound/molecule names and other alliance information.
For this research I focused on alliances formed during the 2005 and 2008 timeframe, what reduced the overall sample of 1401 contracts to 567. The sample was screened to ensure that the compound names were available, based on this criteria, the sample was narrowed down to 329 contracts. After removing mergers, acquisitions and double contracts, the sample was finally narrowed down to 252 contracts. For every collaboration in the dataset, the firm that pays for the alliance is considered as the client, and the other firm is referred to as the research and development (RnD) firm.
3.2 Data Measurement
16
entering phase III, 3 entering phase I and 4 entering phase II. In other words, 0 means unsuccessful innovation and 4 means the highest degree of innovation success.
Independent variables. Testing the first hypotheses requires measuring the relationship between a joint steering committee and innovation success. Accordingly, the first independent variable is joint steering committee, with 1 indicating when a steering committee is noted in the contract, and 0 otherwise.
As with the dependent variable, I also made a distinction here between two measurement levels to gain a deeper understanding of joint steering committees. The second measure of joint steering committees is developed based on the level authority assigned to the joint steering committee. The degree of JSC authority equals 0 when there is no committee present, 1 when the committee only reviews the alliance, 2 when they are allowed to monitor, and 3 when the committee has the right to approve decisions. This data is used as proxy to determine the level of authority of joint steering committees.
The second hypothesis states that size similarity between partners strengthens the relationship between a joint steering committee and innovation success. The variable size similarity of the firms refers to the size of one firm relative to the size of its partner in an alliance. Verwaal and Donkers (2003) argue that in larger companies, size is related to the specialization of human resources. In addition, in the health sector it is less reliable to measure size based on total assets as they contain relatively many specialized firms compared to other sectors, making their size more dependent on the amount of labor than other assets (Yoon, Rosales & Talluri, 2018). The variable size similarity is therefore measured based on the difference between the number of employees of both parties. To ensure followability of this variable, it is measured in the opposite direction as shown in the theory, namely in difference in size rather than geographical proximity. Therefore, I measured the difference in size by calculating the difference in the number employees between the firms. This gives a 0 when the amount of employees is equal, and higher values if they differ more. In the analysis and results section, the variable is referred to as ‘difference in size’.
17
km) when the companies are geographically far apart. In the analysis and results section, the variable is referred to as 'geographical distance'.
Control variables. Several controls have been included on factors that may also affect the aforementioned dependent and independent variables in order to draw more definitive conclusions. I control for alliance characteristics, for other governance mechanisms, potential sources of unobserved effects that are common in collaborative agreements, and for fixed effects.
First, I control at the alliance level for interdependence between the partners because firms with a higher level of interdependence are more likely to assign a joint steering committee in their alliance (Reuer & Devarakonda, 2016). Interdependence means that both firms are cooperating together on research, development or commercialization activities in the collaboration (e.g. co-development) (Reuer & Devarakonda, 2016; Thompson’s, 1967). The variable interdependence equals 1 when the firms cooperate on activities, and 0 otherwise. Furthermore, I control for technology overlap, as overlapping technologies lead to appropriation concerns and therefore increase the chance that joint steering committees will be established by alliance partners (Reuer & Devarakonda, 2016). This variable is measured by examining the extent to which the partners patent in the same technology classes (Jaffe, 1986; Sampson, 2007). Their patents are categorized according to the underlying technology. The variable technological overlap is equal to 0 if the technologies are similar and is equal to a value of 1 when there is the greatest possible technological diversity between partners. Next, I account for deal size at the alliance level as larger alliance deals need more coordination and easier lose oversight (Lehn, Patro & Zhao, 2009). The variable deal size is measured by adding up the upfront payments and all the contingent payments made by the client over the life of the contract (e.g. milestone payments) (Robinson & Stuart, 2007). However, eventually this measure will be incomplete because the royalty payments have not been measured and I cannot anticipate future payments. The last control variable at the alliance level is fast track alliances as they require more coordination. The United States Food and Drug Administration (FDA) determines whether a drug treats serious or life-threatening conditions and meets unmet medical needs and therefore can perform a fast track process (FDA, 2018). This process is designed to facilitate development and accelerate drug assessment. The variable fast track is equal to 1 if the FDA determined that it is a fast track alliance and 0 otherwise.
18
contract length is measured by the number of words in the contract. Next, to control for relational support I checked whether the partners had a prior relationship. Recent empirical work links the extent to which firms have relational history with the degree of trust between partners, which reduces the chance of opportunistic behavior (Saxton, 1997). For this variable prior relationship 1 equals a prior relationship between the partners and 0 otherwise.
Third, I included the potential sources of unobserved effects in the controls. I made a distinction between alliances involving partners from the same country and alliances that include partners from across the border. Collaborating with foreign firms is more difficult due to liabilities of foreignness (Gulati, 1995; Zaheer, Hernandez, and Banerjee, 2010). The variable cross-border alliances is equal to 0 when the partners’ headquarters are located in the same country and 1 when the partners work across borders.
Lastly, to control for unmeasured and stable characteristics I used the fixed effect method (Allison & Christakis, 2006). I created dummy variables for the drug’s development phase at the moment firms enter into the alliances (Lerner and Malmendier, 2010), for the typology of the disease according to DiMasi et al. (2010) and for the major technology categories of the drugs.
3.3 Analytical Method
Since the first dependent variable is a binary variable and the second is ordinal, I use a
conditional mixed process model (CMP). Therefore, the hypotheses are tested by jointly estimating a probit and an oprobit regression model with linkages among their error processes,
creating the ability to test for cross-equation constraints (Roodman, 2011).
To check for variables with outliers, I performed a Tukey boxplot and checked for flagging data points that are beyond the quartiles with a range greater than 1.5 interquartile. According to Rousseeuw, Debruyne, Engelen and Hubert (2006) something is called an outlier if it is more than three standard deviations from the mean, which is not the case for one of the continuous variables geographical distance, deal size and contract length. Hence, there is an assumption of a non-skewed distribution.
Furthermore, I applied z-standardization (z-score) to all the independent variables except the binary variables. This function calculates the probability of a score occurring within our normal distribution, making it possible to interpret the scores of two variables that have a different normal distribution (Dawson, 2014).
19
alliances with missing values, I used the single imputation technique by replacing the missing values with the mean.
4. RESULTS
This chapter shows the results of this quantitative study. First, I present the results of the descriptive statistics and the correlations between all variables. Next, I will show the results of the conditional mixed process and discuss the results of the tested hypotheses. The third paragraph shows the results of additional analyzes I conducted to determine if there are differences in results by looking at the marginal effects of the moderators. Lastly, I performed a robustness check to check the strength of the model.
4.1 Descriptive Statistics and Correlations
Table 1 provides the descriptive statistics of the dependent, independent and control variables. Columns 1, 2, 3 and 4 include the means, standard deviations, the minimum and the maximum of the variables, respectively. Table 2 presents the correlation matrix, which indicates whether variables correlate with each other or not. To make the numbers clearer, I performed some additional analysis and compared the means by splitting the ordinary and binary variables into different categories (Appendix A). The dataset is based on a data sample of 252 alliances that together form an unbalanced dataset over a period of 2005 up until 2008.
This section presents some findings from the dataset (Table 2 and Appendix A). First, it can be noted that in total 19% of the collaborations successfully progressed to the next clinical stage. Furthermore, more than two third of the agreements (67%) included a joint steering committee in their contract. Alliances that established a JSC are on average geographically further apart (r = 0.16, p < 0.01). Alliances with a higher degree of innovation success are on average geographically closer together (r = -0.17, p < 0.01). When partners are interdependent, 11% of those alliances work without a JSC and 89% with a JSC. The average deal size of the data sample is $65 million, with $50 million on average for alliances without a JSC, and $130 million for alliances with a JSC. Additionally the contracts are on average twice as long (51%) when they work with a JSC than when they do not.
20 TABLE 1
Descriptive Statistics
Mean S.D. Min Max
Dependent variables Innovation success .92 1.38 0 4 Next phase .19 .39 0 1 Independent variables JSC .48 .50 0 1 JSC authority 1.16 1.35 0 3 Geographical distance 4654 3556 1 16221 Difference in size 33931 38482 0 339785 Control variables Interdependence .24 .43 0 1 Technology overlap .38 .35 0 1
Deal size 6.50e+07 1.66e+08 0 1.42e+09
Fast track .13 .34 0 1
Contract length 20728.71 15531.90 383 109895
Prior ties .25 .43 0 1
Cross-border alliance .49 .50 0 1
Notes: n = 252
21 TABLE 2
Correlation Matrix
1 2 3 4 5 6 7 8 9 10 11 12 13
22 4.2 Regression Results and Hypotheses Testing
Table 3 presents the main effects of the conditional mixed process model for the determinants innovation success and the degree of innovation success. The log likelihood (LL) of this model confirmed that the goodness of fit increased from Model 1 (LL = 209.5) to Model 3 (LL = -193.1).
Model 1 and 2 are the baseline models and contain only the control variables and their effect on innovation success and the degree of innovation success. The variable fast track alliances shows positive significant results about the relationship with the degree of innovation success (β = 0.72, p < 0.05). Implying that fast track alliances achieve a higher degree of innovation success.
In Model 3 and 4, I added the independent variables JSC, JSC authority, size similarity, and geographical distance. The first hypothesis posited that the presence of a joint steering committee in an alliance increases the chances of innovation success. The results show a positive coefficient of the JSC variable, which explains that the presence of a joint steering committee has a positive relationship with a successful innovation. However, the coefficients of both JSC and JSC authority proved to be non-significant for both dependent variables the degree of innovation success (β = 0.09, p > 0.1 and β = -0.01, p > 0.1) and innovation success (β = 0.15, p > 0.1 and β = -0.06, p > 0.1). Therefore, Hypothesis 1 is rejected.
23
TABLE 3 - Determinants of Innovation Success and the Degree of Innovation Success
Degree of IS IS Degree of IS IS Degree of IS IS Degree of IS IS
Model 1 Model 2 Model 3 Model 4 Model 5 Model 6 Robustness Robustness
Moderator Variables
JSC Authority x Geographical Distance -0.27 -0.00 -0.27 -0.00
(0.18) (0.23) (0.19) (0.20)
JSC x Geographical Distance 0.17 -0.37 0.17 -0.37
(0.50) (0.66) (0.52) (0.52)
JSC Authority x Difference in Size 0.15 -0.04 0.15 -0.04
(0.23) (0.27) (0.22) (0.23) JSC x Difference in Size -0.25 0.08 -0.25 0.08 (0.67) (0.77) (0.65) (0.63) Independent Variables JSC 0.09 0.15 0.09 0.06 0.09 0.06 (0.45) (0.50) (0.46) (0.53) (0.46) (0.42) JSC Authority -0.01 -0.06 -0.03 -0.04 -0.03 -0.04 (0.17) (0.19) (0.17) (0.20) (0.18) (0.17) Difference in Size 0.02 0.07 -0.13 0.09 -0.13 0.09 (0.12) (0.12) (0.21) (0.16) (0.18) (0.12) Geographical Distance -0.56*** -0.60*** -0.25 -0.42+ -0.25 -0.42* (0.13) (0.16) (0.18) (0.22) 0.09 0.06 Control Variables Interdependence 0.24 0.10 0.20 0.06 0.21 0.01 0.21 0.01 (0.24) (0.27) (0.27) (0.31) (0.28) (0.33) (0.29) (0.28) Technology Overlap 0.04 -0.05 0.03 -0.07 0.07 -0.06 (0.11) (0.12) (0.10) (0.11) (0.10) (0.12) (0.11) (0.12) 0.02 0.18+ Deal Size 0.01 0.16 0.02 0.17 0.02 0.18 (0.10) (0.10) (0.10) (0.12) (0.10) (0.13) (0.11) (0.13) 0.81** 0.38 Fast Track 0.72* 0.33 0.73* 0.39 0.81* 0.38 (0.26) (0.32) (0.30) (0.33) (0.31) (0.34) (0.32) (0.36) 0.06 0.04 Contract Length 0.06 0.02 0.05 0.02 0.06 0.04 (0.12) (0.12) (0.12) (0.13) (0.14) (0.15) (0.14) (0.15) 0.13 0.01 Prior Ties 0.26 0.13 0.21 0.04 0.13 0.01 (0.22) (0.22) (0.24) (0.27) (0.25) (0.28) (0.25) (0.29) 0.28 0.67* Cross-border Alliance -0.22 0.03 0.39 0.72* 0.28 0.67* (0.26) (0.27) 0.24 0.10 0.20 0.06 0.21 0.01 0.21 0.01
Compound Phase Dummies yes yes yes yes yes yes yes yes
Typology Dummies yes yes yes yes yes yes yes yes
Technology Dummies yes yes yes yes yes yes yes yes
AIC 513 513 505.9 505.9 512.2 512.2 512.2 512.2
BIC 678.9 678.9 700 700 734.5 734.5 734.5 734.5
Log-likelihood -209.5 -209.5 -198 -198 -193.1 -193.1 -193.1 -193.1
Observations 252 252 252 252 252 252 252 252
24
Model 5 and 6 introduce the interaction terms of the moderators and show the full conditional mixed process model. The second hypothesis stated that size similarity has a positive moderating effect on the relationship between alliances with a joint steering committee and innovation success, which means that difference in size will negatively moderate this relationship. Model 6 reveals that the moderating effect of difference in size is not significantly consistent with Hypothesis 2 (β = 0.08, p > 0.1). However, according to Huang and Shields (2000) the marginal effect can differ across observations. Namely, in non-linear regressions the marginal effect of an interaction is not simply the same as the coefficient for their interaction (Hoetker, 2007). Moreover, even the sign of the coefficient can be different per observation. Therefore I do not reject Hypothesis 2 yet.
The third and final hypothesis suggested that geographical proximity is positively moderating alliances with a joint steering committee and innovation success, implying that geographical distance will negatively moderate this relationship. The estimation of Model 6 shows indeed a negative moderating relationship, however the coefficient is non-significant (β = -0.37, p > 0.1). Again, the mixed process model is not providing support for Hypothesis 3. However, as previously established, this coefficient is not sufficient to provide inference and reject Hypothesis 3 yet.
4.3 Additional Analyses
To further investigate the moderating results of the conditional mixed process model, I performed additional analyses. The effects of both moderators difference in size and geographical distance were surprisingly non-significant. However, as Hoetker (2007) mentioned, it is possible to have a significant interaction for some observations, even when the coefficient indicates otherwise. Consequently, there was a necessity for thorough examination and therefore I plotted the interaction to interpret the magnitude of the moderating variables. The exact values of these marginal effects are presented in Appendix C.
25
changes the relationship between a joint steering committee and innovation success. Since this effect is the opposite, namely positive, Hypothesis 2 can be rejected.
Figure 1
26 Figure 2
To further investigate the effect of geographical distance, I also plotted this moderating effect with the degree of JSC authority. Figure 4 shows a similar interaction as the previous graphic (Figure 3). However, there are two new findings shown in this figure. The first finding is that the absence of a JSC is more beneficial than having a JSC at 9,000 km distance, this point is lower for the level of authority of a JSC (at 5,500 km). Secondly, a JSC with a high level of authority has a stronger negative relation with innovation success when there is a large geographical distance between the firms.
4.4 Robustness Check
27 Figure 3
5. DISCUSSION
This section discusses the theoretical and managerial implications based on the results. It also discusses some interesting additional insights. Finally, I discuss the limitations of this study and provide suggestions for future research.
5.1 Theoretical Implications
28
This research is the first to examine the relationship between steering committees and innovation success in an alliance setting. Despite the assumption of this relationship, I also examined whether some differences between partners are more beneficial than others to take advantage of the assignment of a steering committee and thereby increasing the chances of innovation success in alliances. Therefore, I examine whether the presence of a joint steering committee leads to higher innovation success in alliances, and whether it is bounded by size similarity and geographical proximity between partners.
This research investigation illuminates several interesting insights, both significant and non-significant. The first insight complements Lumineau and Schilke’s (2018) research into the contractual control and coordination effect of steering committees. It also extends the study of Reuer and Devarakonda (2018) into the positive relation between steering committees and knowledge transfers. This study found no statistically significant evidence for a positive relationship between joint steering committees and innovation success, as suggested by Hypothesis 1. However, there is a positive effect between the two variables. Hypothesis 1 can therefore not be accepted, nor can the null hypothesis because it cannot be excluded that the relationship is fully absent. Literature explained that providing coordination and control has a positive effect on innovation success (Lumineau & Schilke’s, 2018). Although joint steering committees offer these capabilities (Reuer & Devarakonda, 2018), the positive effect of these capabilities may be contingent on several factors. Working with strongly intertwined personal relationships is a highly sensitive matter (Gil, 2009), differences between firms and their employees can hinder the positive effect of the committee.
29
Finally, I find support for Hypothesis 3, suggesting that geographical proximity positively moderates the relationship between a joint steering committee and innovation success. The results show a significant difference between alliances with and without a steering committee at the higher levels of geographical proximity (0 to 4,000 km). The lower levels of geographical proximity also show a difference, but cannot be confirmed as significant. Partners who are geographically closer to each other can better manage committees due to the fact that they experience more trust in domestic alliances (Bierly & Gallagher, 2007). A higher level of trust results in higher commitment and coordinated action among individuals in a joint steering committee (Arrow, 1974; Costa, Roe & Taillieu, 2001). In addition, geographically close partners communicate more effectively (Mowery, Oxley, Silverman, 2011). Since joint steering committees function as communication centers for alliances, improved communication results in better managing and coordinating the alliance (Duysters & Lokshin, 2011). Therefore, alliances with partners geographically close to each other have a higher chance of innovation success when they assign a joint steering committee. Alliances with geographical distant partners on the other hand, experience less commitment and coordinated action, resulting in ineffective communication and more management challenges in their joint steering committee. Therefore, alliances with geographical distant partners have a higher chance of innovation success without a steering committee and alliances with partner geographically close experience a lower chance of innovation success in the presence of a joint steering committee. 5.1.1 Additional insights
I will discuss some interesting additional insights provided by this research. Firstly, the regression model (Table 3, Model 3 and 4) shows a significant positive relationship between geographical proximity and alliance success, confirming prior research. Geographical distance leads to challenges within alliances. When alliances are not properly designed, understood and managed, poor communication and development activities increase the transaction costs and eventually harm the alliance and its innovation success (Williamson, 1985). This finding supports the research of Sivakumar et al. (2011), that increased geographical distance between partners negatively influences innovation output.
30
Finally, I found a significant positive relationship between fast track alliances and the degree of innovation success in all three models (Table 3), indicating that alliances with an approved fast track process by the FDA have an increased degree of innovation success. It is important to notice that fast track alliances do not generate a higher chance of innovation success in the first place. However, when innovation success is achieved, this leads to a higher degree of innovation success.
5.2 Managerial Implications
To summarize the meaning of the results in terms of actions, I indicate what action or non-action should be taken. Firstly, this research implies that firms should not blindly assign a joint steering committee to their alliance to increase the chance of innovation success. The success of alliances is contingent to several factors and can even be harmed when these factors are not taken into account.
Size similarity between partners is one of these important factors that should be taken into account. Having a joint steering committee adds no additional value in the case of high size similarity between partners, therefore it is not recommended to appoint a steering committee because the costs may not outweigh the benefits.
The second contingency factor that should be taken into account is geographical proximity between alliance partners. This study shows that alliances with partners no further apart than 8,500 km distance are better off using a joint steering committee. Alliances with partners further than 8,500 km apart are recommended to cooperate without a steering committee, as this may even reduce the likelihood of innovation success. Additional research into the authority level of joint steering committees shows that steering committees with a high level of authority harms the alliance even more in a situation of geographical distance.
The overall conclusion of this research is that the use of joint steering committees is bounded by different conditions.
5.3 Limitations and Future Research
Despite the fact that this research found several interesting results, it also subject to certain limitations that offer possibilities for interesting future research.
31
alliances and use joint steering committees in order to create innovation success. These industries offer opportunities for future research.
Secondly, the sample of this research contains contracts from 2005 up until 2008. The alliance environment is changing due to environmental and technological developments. These factors may influence the relationship between alliances and innovation success. Therefore, future research could execute the same study with a more recent contract sample.
Furthermore, the current study was subject to other data limitations which could have impacted the results. One data limitation is missing values for the variable size similarity, as some data from the 2005 to 2008 annual reports were no longer available. Another data limitation is the relatively small amount of observations at low size similarity and low geographical distance, making it difficult to prove the significance at these levels. Future research could address these limitations by focusing on a sample with highly different firms in terms of size and geographical distance.
Fourthly, this study researched the boundary conditions size similarity and geographical proximity. However, literature offers more individual firm attributes such as age, competitive position, product diversity and financial resources as important determinants for the structures of strategic alliances (Powell and Brantley, 1992; Shan, 1990; Shan et al., 1994). Therefore they provide an interesting start for future research to examine these attributes and their boundary effect on joint steering committees.
Lastly, I am aware that this research contains endogeneity, meaning that causality cannot be claimed. According to Hamilton and Nickerson (2003), endogenous variables influence the error term and can lead to biased and inconsistent estimation. To reduce potential biases stemming from omitted variables, I included control variables (Becker, 2005). However, I could not remove all potential biases. Endogenous variables to be further tested are financial resources of both firms, the age of the firms, the average level of education of their employees, the amount of prior experience of both firms with regard to alliances and their amount of previous experience in developing innovations. Future research may further reduce the omitted variable bias by controlling for these endogenous variables.
6. CONCLUSION
32
achieving alliance success is becoming more difficult and the circumstances of alliances are changing to address this problem, it is important to research the effects of this trend on the impact of joint steering committees and innovation success. Therefore, this study has attempted to examine the following research question: Does the presence of a joint steering committee lead to a higher chance of innovation success in alliances and is the effect bounded?
Differences in partners’ firm attributes can either enhance or reduce the relationship between joint steering committees. Specifically, this study examined the moderating effect of size similarity and geographical proximity between partners. The findings reveal that geographical proximity positively moderates the relationship between joint steering committees and innovation success. In contrast, size similarity between partners improves the relationship between joint steering committees and innovation success, however this finding was not significant. These findings are relevant for firms that need to determine the governance mechanisms for their alliance, as the results imply that differences between partners’ firm attributes are an important predictor for the effect of joint steering committees on innovation success. Therefore, this study contributes to the academic research on alliances, joint steering committees, firm attributes and innovation success.
The ambition of this study is that the findings will help managers consider the allocation of joint steering committees in their alliance, taking into account the size and geographical distance of their partners. In addition, despite the fact that this research is subject to certain limitations, it offers interesting possibilities for future research.
7. REFERENCES
Acs, Z. & David, A. 1990. Innovation and Small firms. Cambridge, MA: MIT P
Acton, J. G., Irvin, B. L. & Hopkinds, B. A. 1991. Theory-testing research: Building the science. Advances in nursing science, 14(1): 52-61.
Albers, S. 2019. The Design of Alliance Governance Systems. Springer Gabler.
Allison, P. D. & Christakis, N. 2006. Fixed-effects methods for the analysis of non-repeated events. Sociological Methodology, 36(1): 155-172.
33
Argyres, N. S., Bercovitz, J., & Mayer, K. J. 2007. Complementarity and evolution of contractual provisions: An empirical study of IT services contracts. Organization Science, 18(1): 3-19.
Arora, A. 1995. Licensing tacit knowledge: Intellectual property rights and the market for know-how. Economics of Innovation and New Technology, 4(1): 41–60.
Arrow, K. 1974. The limits of organization. New York: Norton.
Aubert, B. A. & Kelsey, B. L. 2000. The illusion of trust and performance.
Barney, J. B. & Hansen, M. H. 1994. Trustworthiness as a source of competitive advantage. Strategic Management Journal, 15: 157-190.
Becker, T. E. 2005. Potential problems in the statistical control of variables in organizational research: A qualitative analysis with recommendations. Organizational Research Methods, 8: 274-289.
Belsley, D. A., Kuh, E. & Welsch, R. E. 1980. Regression diagnostics: Identifying influential data and sources of collinearity. New York: John Wiley.
Bierly, P. E. & Gallagher, S. 2007. Explaining alliance partner selection: Fit, Trust and strategic expediency. Long Range Planning, 40: 134-153.
Brockman, J. L. 2003. The problem with problem-solving training in industry. Ohio State University, 26-31.
Carson, S. J., Madhok, A., & Wu, T. 2006. Uncertainty, opportunism, and governance: The effects of volatility and ambiguity on formal and relational contracting. Academy of Management Journal, 49: 1058–1077.
Chen, O. & Chen, T. J. 2002. Asymmetric strategic alliances: A network view. Journal of Business Research, 55: 1007-1013.
34
Costa, A. C., Roe, R. A. & Taillieu, T. 2001. Trust within teams: The relation with performance effectiveness. European Journal of Work and Organizational Psychology, 10(3), 225-244.
Das, T. K. & Teng, B. S. 1998. Between trust and control: Developing confidence in partner cooperation in alliances. Academy of Management Review, 23: 491–512.
Das, T. K. & Teng, B. S. 1999. Managing risk in strategic alliances. Academy of Management Executive, 13(4).
Das, T. K. & Teng, B. S. 2000. A resource-based view of strategic alliances. Journal of Management, 26(1): 31–61.
Dawson, J. F. 2014. Moderation in management research: What, why, when, and how. Journal of Business and Psychology, 29: 1-19.
De la Sierra, M. C. 1995. Managing global alliances: Key steps for successful collaboration. Addison-Wesley.
Deangelis, C. D. 2016. Big pharma profits and the public loses. The Milbank Quarterly, 94(1): 30-33.
Deloitte. 2015. Measuring the return from pharmaceutical innovation: Transforming R&D returns in uncertain times. Deloitte Centre for Health Solutions.
DiMasi, J. A., Feldman, L., Seckler, A & Wilson A. 2010. Trends in risks associated with new drug development: Success rates for investigational drugs. Clinical Pharmacology & Therapeutics 87(3): 272-7.
Doz, Y. L. 1998. Technology partnerships between larger and smaller firms. Cooperative Strategies in International Businesses.
Duysters, G., & Lokshin, B. 2011. Determinants of alliance portfolio complexity and its effect on innovative performance of companies. Journal of Product Innovation Management, 28(4): 570-585.
35
FDA, 2018. Fast track, breakthrough therapy, accelerated approval, priority review.
https://www.fda.gov/patients/learn-about-drug-and-device-approvals/fast-track-breakthrough-therapy-accelerated-approval-priority-review. Cited 2020-6-10.
Gatignon, H. & Anderson, E. 1988. The multinational corporation’s degree of control over foreign subsidiaries: An empirical test of a transaction cost explanation. Journal of Law Economics and Organization, 4: 305–336.
Gerwin, D. 2004. Coordinating New Product Development in Strategic Alliances. The Academy of Management Review, 29(2): 241-257.
Giarratana, M. S., & Mariani, M. 2014. The relationship between knowledge sourcing and fear of imitation. Strategic Management Journal, 35(8): 1144–1163.
Gil, N. 2009. Developing cooperative project client-supplier relationships: How much to expect from relational contracts? California Management Review, 51 (2), 144-169.
Gronin, M. A. & Weingart, L. R. 2007. Representational gaps, information processing, and conflict in functionally diverse teams. Academic Management Review, 32(3): 761-773. Gudykunst, W. B., Matsumoto, Y., Ting-Toomey, S., Nishida, T., Kim, K. & Heyman, S. 1997. The influence of cultural individualism-collectivism, self construals, and individual values on communication styles across cultures. Human Communication Research, 22(4): 510-543.
Gulati, R. 1995. Does familiarity breed trust? The implications of repeated ties for contractual choice in alliances. Academy of Management Journal, 38: 85–112.
Gulati, R. 1998. Alliances and networks. Strategic Management Journal, 19: 293-317. Gulati, R. & Singh, H. 1998. The Architecture of Cooperation: Managing Coordination Costs
and Appropriation Concerns in Strategic Alliances. Administrative Science Quarterly, 43(4): 781-841.
36
Hagedoorn, J. & Hesen, G. 2007. Contract law and the governance of inter-firm technology partnerships: An analysis of different modes of partnering and their contractual implications. Journal of Management Studies, 44: 342–366.
Hamel, G. 1991. Competition for competence and inter-partner learning within international strategic alliances. Strategic Management Journal, 12: 83–103.
Hamilton B. H. & Nickerson, J. H. 2003. Correcting for endogeneity in strategic management research. Strategic Organization, 1(1): 51-78.
Hoetker, G. 2007. The use of logit and probit models in strategic management research: Critical issues. Strategic Management Journal, 28(4): 331-343.
Hoetker, G. & Mellewigt, T. 2009. Choice and performance of governance mechanisms: Matching alliance governance to asset type. Strategic Management Journal, 30: 1025-1044.
Hofstede, G. 1980. Culture and organizations. Organizations and Societies, 10(4): 15-41. Huang C, Shields TG. 2000. Interpretation of interaction effects in logit and probit analyses:
reconsidering the relationship between [U.S.] registration laws, education, and voter turnout. American Politics Quarterly, 28(1): 72–79.
Hughes, J. & Weiss, J. 2007. Simple rules for making alliances work. Business Harvard Review.
Jaffe, A. B. 1986. Technological opportunity and spillovers of R&D: Evidence from firms’ patents, profits and market value. American Economic Review, 76: 984 –1001.
Jann, B. 2014. Plotting regression coefficients and other estimates. The Stata Journal, 14(4): 708-737.
Jarvenpaa, S. L. & Leidner, D. E. 1999. Communication and trust in global virtual teams. Organization Science, 10(6): 693-815.
