Inter-firm Absorptive Capacity The effects of governance mechanisms and relation characteristics on inter-firm absorptive capacity

Inter-firm Absorptive Capacity

The effects of governance mechanisms and relation characteristics on

inter-firm absorptive capacity

Author:

Martin Kleis Pit

1

Inter-firm Absorptive Capacity

The effects of governance mechanisms and relation characteristics on

inter-firm absorptive capacity

Master Thesis

MSc. Marketing Management

University of Groningen

Faculty of Economics and Business

January 9

_{, 2016}

Martin Kleis Pit

2

Preface

3

Abstract

Over the past two decades, organizational knowledge has been regarded as one of the main contributors to firm competitive advantage. With the acquisition of external knowledge, firms are able to foster innovation and create differential advantages vis a vis their competitors. This study examines the effect of contractual and relational governance mechanisms on inter-firm absorptive capacity. The interaction between governance mechanisms remains an ambiguous concept, where much research has been conducted on the firm (unit) level. This study takes an outward, inter-firm perspective, and looks at the interaction of governance mechanisms and ACAP with a relation perspective in mind. A database of 166 matched-pair buyer-supplier relationships was used to test the hypotheses. The conceptual model was assessed using SmartPLS3 PLS-SEM. Results indicate that contractual governance mechanisms have significant positive effect on a firm’s absorptive capacity from a buyer perspective, whereas relational governance mechanisms have significant positive effects for both buyer and supplier perspectives. From a buyer perspective, contractual and relational governance mechanisms function as substitutes, rather than complements as hypothesized. Relationship duration was found to mainly show significant differences for relationship governance mechanisms, and the same applied for relationship importance.

4

Index

2.2 Explorative and exploitative learning performance ... 10

2.3 Governance mechanisms ... 13

2.3.1 Contractual governance ... 14

2.3.2 Relational governance ... 17

2.4 The interaction of contractual and relational governance ... 19

2.5 Relational context and governance mechanisms ... 22

3. Conceptual model ... 24

3.1 Conceptual model ... 24

3.2 Hypotheses ... 24

4. Methodology ... 25

4.1 Data collection procedures ... 25

4.2 Measures ... 26 4.2.1 Multi-group measures... 28 4.3 Descriptives ... 29 4.4 Statistical procedures... 29 4.5 Approach ... 30 5. Results ... 31

5.1 Reliability and validity – reflective scales... 31

5.2 Reliability and validity – formative scales ... 32

5.3 Discriminant validity ... 32

5.4 Hypothesis testing - structural model ... 33

5.5 Multi-group-analysis results ... 35

5.5.1 Relationship duration, ‘less than four years’ versus ‘more than four years’ ... 35

5.5.2 Relationship importance, ‘average’ versus ‘crucial’ importance ... 36

6. Discussion ... 37

6.1 The effects of inter-firm ACAP on explorative and exploitative learning ... 37

6.2 Impact of governance mechanisms on inter-firm ACAP ... 37

6.3 Impact of relational context on governance mechanisms in inter-firm relationships ... 40

7. Managerial implications. ... 41

8. Limitations and future research ... 41

9. Conclusion ... 43

References ... 44

Appendix ... 58

Appendix A. Constructs and measurement items ... 58

Appendix B. Convergent validity of reflective scales ... 62

Appendix C. Convergent validity of formative scales ... 63

Appendix D. Deleted formative indicator scores ... 63

Appendix E. Discriminant validity first and second order constructs ... 64

Appendix F. Standardized path coefficients and Multi-Group-Analysis results (PLS-SEM) ... 65

Appendix G. Interaction between contractual and relational governance ... 66

Appendix H. Standardized path coefficients and MGA results between ‘less than four years’ and ‘more than four years’ duration. ... 66

5

1. Introduction

Over the past two decades, organizational knowledge has been regarded as one of the main contributors to firm competitive advantage (Lichtenthaler, 2009; van Wijk et al, 2008; Zahra and George, 2002). With the acquisition of external knowledge, firms are able to foster innovation and create differential advantages vis a vis their competitors (Dyer and Singh 1998; Pine, Peppers, and Rogers 1995). The type of knowledge that acts as a source of these advantages is often tacit in nature and difficult to codify or imitate (Ireland, Hitt and Vaidyanath, 2002; Zollo, Reuer and Singh, 2002). It is thus of importance to the firm that they are able to effectively acquire and transfer new knowledge in order to survive, making external sources of knowledge critical to the firm (Cohen and Levinthal, 1990; van Wijk et al, 2008).

Theories on organizational learning argue that firms are more effective than markets when it comes down to knowledge acquisition and transfer, and this phenomenon is evidenced by an increased use of strategic alliances and mergers and acquisitions to acquire knowledge (Birkinshaw et al., 1999; Lane et al., 2001; Mayer and Argyres, 2004). However, being more effective than markets at knowledge transfer does not immediately guarantee that these processes will actually take place effectively on a routine basis (Gupta, 2000; Licthenthaler, 2009). Hence, firm ‘absorptive capacity’ (ACAP) or the “ability of a firm to recognize the value of new, external information, assimilate it, and apply it to commercial ends” (recognition, assimilation and exploitation, Cohen and Levinthal, 1990) is seen as a major source of sustainable competitive advantage when it comes to knowledge transfer (Lane et al, 2006; Lichtenthaler, 2009). A lack of firm ACAP can result in difficulties when it comes to the knowledge transfer processes within and between firms (Tsai, 2000).

Zahra and George (2002) have defined ACAP further and made the distinction between potential (PACAP) and realized (RACAP) absorptive capacity. It is a dynamic capability that influences the firm’s ability to create and deploy the knowledge required for other organizational capabilities (Zahra and Georce, 2002) and consists of the sequential processes of exploratory, transformative, and exploitative learning (Lane et al, 2006).

6 the shape of unilateral, formal control (i.e. contractual governance mechanisms) and bilateral control (i.e. relational norms and trust) (Cannon et al, 2000; Wallenburg and Schaffler, 2014). These form the explicit and implicit rules of the relationship (Gosh and John, 2005). Firms are constantly pressed to make decisions regarding their optimal governance structures, decisions which can have broad impact on multiple levels throughout the firm (Gereffi et al., 2005). Literature has shown that the choice of governance mechanisms depends on several factors, among others the evolution of the relationship (i.e. relationship length and importance) as well as external market characteristics which force a firm to react in a specific way (Gereffi et al., 2005; Droge et al, 2008). Throughout the course of the relationship, certain governance mechanisms could become obsolete or can be replaced by other, more effective measures.

It is in line with this reasoning that governance mechanisms can be expected to affect the transfer of knowledge and inter-firm ACAP. Therefore, it is interesting for researchers and managers to have an extensive understanding of firm politics and governance mechanisms (Blackler, 2000; Lawrence et al, 2005).

The aim of this study is to advance the understanding of governance mechanisms and their effects on ACAP, hence furthering knowledge on ACAP’s underlying dimensions and its drivers. This study will add to the existing literature by bridging the gap between intra- and inter-organizational domains, taking an outward perspective of absorptive capacity. In addition, this study addresses how the interplay between governance mechanisms differs when relational characteristics (e.g. relationship duration and relationship importance) are taken into consideration. The main research question that will guide this study is as follows:

How do governance mechanisms affect interfirm ACAP (PACAP and RACAP)?

Certain sub questions were derived, which help guide the study and further investigate the main effect:

In what way do governance mechanisms (contractual, relational) interact with each

other?

7 The contribution to existing literature by this study is as follows:

First of all, this study responds to the request for integrative research on organizational learning and politics by assessing the effects of governance mechanisms on ACAP. In doing so, this study explicitly takes into account the multidimensionality of ACAP (Berger, 2015). This study thus takes into consideration all four dimensions of ACAP in a direct manner, instead of replacements (i.e. R&D investment). So far, only few studies have empirically captured ACAP’s multidimensionality and measured the concept directly (Berger, 2015).

In addition, the relationship between governance mechanisms and ACAP is taken on from an firm perspective. Simply transferring the concepts devised from intra-firm research to the inter-organizational domain runs the risk of failing to capture the complex nature or inter-organizational relationships. In particular research on inter-firm governance mechanisms, in contrast to intra-organizational settings, are characterized by combinations of hierarchical and market elements, adding to the overall complexity of the relationships. Taking an inter-firm perspective also allows for comparisons to be made between buyer and supplier perspectives regarding the interaction of governance mechanisms and ACAP. This can provide insights into the how both parties perceive a relationship, and whether crucial differences exist regarding the right application for governance mechanisms.

Bringing politics through governance mechanisms into research on organizational learning should provide a more effective foundation for understanding why some organizations are better able to learn and why only some of the available useful innovations are embraced by organizations (Lawrence et al, 2005). For managers to understand how to effectively deal with these complex situations, it is crucial to know the interplay between governance mechanism types. Therefore, this study not only addresses the interaction between governance mechanisms and ACAP, but also the way in which a combination of both affects each other, as well as ACAP.

8

2. Theoretical framework

The following section will discuss all theoretical concepts featured in this study. First, a discussion of ACAP, its dimensions and the antecedents to ACAP will be discussed. Afterwards, the dimensions and interactions of the different types of governance mechanisms will be discussed, followed by an application of both concepts to the interfirm relationships. In the section thereafter, a conceptual model with corresponding hypotheses will be presented.

2.1 Absorptive capacity

The roots of absorptive capacity (ACAP) can be found in the organizational learning literature from the 1980s (Volberda et al., 2010). The concept promised to explain the process through which firms learn, develop, and assimilate new knowledge necessary for competitive advantage. It presented ACAP as being critical to firm innovative capabilities. (Cohen and Levinthal, 1990). In their original work on ACAP, Cohen and Levinthal (1990) first defined the concept as “the ability of a firm to recognize the value of new, external information, assimilate it, and apply it to commercial ends” (Cohen and Levinthal 1990).

The premise of absorptive capacity is that the organization needs prior related knowledge to assimilate and use new knowledge. Without prior knowledge, the firm and its’ individual members are not able to transform the new knowledge to fit with the existing organization, and will thus fail to absorb the information. The firm’s ability to value knowledge is stated as based on past experience and investment (i.e. it develops on prior investments in building absorptive capacity in individuals), the ability to assimilate stems from knowledge characteristics and the alliance dyad characteristics and finally the ability to apply is based on technical opportunity and appropriability (Cohen and Levinthal, 1990; Zahra and George, 2002).

9 imported knowledge. Kim (1997/1998) offers a third definition of ACAP as the capacity to learn

and solve problems (Volberda et al., 2010; Zahra and George, 2002). Zahra and George (2002)

built upon these definitions by conceptualizing ACAP as a dynamic capability that relates to knowledge creation and utilization (Berger, 2015). ACAP is defined as ‘a set of organizational routines and processes by which firms acquire, assimilate, transform, and exploit knowledge to produce a dynamic organizational capability. These four dimensions were assigned to two overlapping components: potential absorptive capacity (PACAP) and realized absorptive capacity (RACAP).

The four dimensions that comprise ACAP can be seen as complementary to one another, but they play different roles in the firm. Acquisition stands for the firm’s capability to identify and acquire externally generated knowledge that is critical to its operations. Assimilation covers the firm’s routines and processes that allow it to analyze, process, interpret and understand the information obtained from external sources (Volberda et al., 2010; Zahra and George, 2002). Together, these dimensions relate to the explorative learning capabilities of the firm, and are seen as the Potential Absorptive Capacity (PACAP). They determine the firm’s receptiveness to external knowledge in the environment, however PACAP does not guarantee that the identified knowledge is subsequently exploited within the firm. This is where Realized Absorptive Capacity (RACAP) comes into play.

10 Potential and Realized ACAP coexist within a firm and are complementary to one another, however neither one nor the other concept is able to sufficiently fulfill a long-term performance for the firm on its own (Zahra and George, 2002). PACAP and RACAP build upon each other, and if a firm does not have one of both aspects sufficiently developed, it will face drawbacks in overall competence (Zahra and George, 2002). For example, a firm that has extensive PACAP, without adequate RACAP will be able to find and acquire new knowledge from its’ external environment, however it will not be able to transform this knowledge to be applicable to firm processes. On the other hand, a firm that lacks in terms of PACAP will find itself able to reap the benefits of knowledge, but it will not be able to identify external knowledge, thus leaving it stuck with its’ current internal capabilities.

2.2 Explorative and exploitative learning performance

Theory on organizational learning (OL) and absorptive capacity (ACAP) (Good, 2013; Zahra and George, 2002) has made the distinction between explorative and exploitative learning performance. Firms need to explore opportunities and exploit what is already known (March, 1991). A distinction between ACAP and OL should be made, where explorative and exploitative learning performance covers the actual outcomes of learning activities, ACAP indicates the firm’s ability to engage in such activities.

The explorative learning processes are characterized by search, discovery, experimentation, risk taking and innovation and the capability to perform explorative actions is expressed through the firm’s PACAP (Cheng and Van de Ven 1996, March 1991; Zahra and George, 2002). Exploitative learning processes are characterized by matching the retrieved knowledge to what is known in the firm, integrating and institutionalizing it (Jansen et al., 2005; Lawrence et al., 2005) and subsequently applying it to the markets (Lenox and King, 2004; Rothaermel and Deeds, 2004). The firm’s capability to do so is expressed to the firms’ RACAP.

11 implies the further practical significance of understanding these relationships, and it is also the reason why the relationship between ACAP and explorative and exploitative learning has been included in this study. Two hypotheses have been developed to support the practical significance of the study:

H1: PACAP will be positively related to explorative learning performance H2: RACAP will be positively related to exploitative learning performance

Even though both explorative and exploitative learning are positively associated with organizational performance, managing the processes of exploration and exploitation is a difficult task for firms. Implementing structures for both learning processes can result in dealing with conflicting cognitive processes, given that exploration and exploitation are fundamentally different in the architectures and structures they require (Jansen et al., 2009). And even though a balance is suggested, little evidence is available to suggest what would be able to do it well (Eisenhardt et al., 2010), a key limitation in organizational literature (Mom et al., 2009; Raisch et al., 2009). A firm that shows underdeveloped capabilities in either one of these dimensions will eventually end up in a suboptimal, stable equilibrium, in which the competency of the firm does not reach its full potential.

12 This paradox is interesting for managers to further develop, and will also form the basis on which this study is developed. Interfirm relationship governance on the one hand (from a TCE perspective) aims at maximizing efficiency and minimizing costs associated from opportunism. On the other hand (RBV perspective), governance mechanisms facilitate the exchange of knowledge and resources between partners in a relationship. Uncovering the interactions between these governance structures lies at the core of this study.

Firm ambidexterity should be distinguished as being structural or contextual (Gibson and Birkinshaw, 2004). Structural ambidexterity refers to a structural solution towards the paradox of exploration and exploitation (Berger, 2015). No trade-off between both demands are put in place, but instead a dual structure where both dimensions operate separately from one another are put in place. Research on inter-firm relationships has shown that often exploration and exploitation are addressed in separate relationships. Firms manage a portfolio of exploitative and explorative alliances to achieve ambidexterity, classifying relationships per learning objective (Lavie and Rosenkopf, 2006; Koza and Lewin, 1998).

However, Im and Rai (2008) have indicated that facilitating both explorative and exploitative processes in one single inter-firm relationship has positive effects on relational performance. This is also where the concept of contextual ambidexterity comes into play, as this is defined as “the behavioral capacity to simultaneously demonstrate alignment and adaptability across an entire business unit” (Berger, 2015). In other words, it is the capability to show explorative and exploitative capacities within a single inter-firm relationship. This study will address the paradox of ambidexterity through the governance of PACAP and RACAP, to see whether both can be effectively governed in a single inter-firm relationship, whilst still maintaining the flexibility to match the demands of the market. To counter the effect of variable diffusion, the elements of explorative and exploitative learning will be considered as separate constructs in their relationship to PACAP and RACAP.

13 2.3 Governance mechanisms

Given the imperfect nature of markets, more sophisticated mechanisms of relationship governance than competitive bidding are favored to drive learning and innovation within partnerships and networks (Ahmadjian and Lincoln, 2001). Therefore, the interplay between various governance mechanisms and the exchange between partners is particularly interesting for management and research (Ghoshal and Moran, 1996). Literature suggest that firms can use governance mechanisms in buyer-supplier relationships to improve performance, innovation and facilitate organizational learning processes (Lawrence et al., 2005). Two main streams on governance mechanisms exist, the Transaction Cost Economics view (TCE) and the Resource Based View (RBV) or relational view. From a TCE perspective, asymmetry within the exchange process between buyer and supplier can cause the better-informed party to engage in opportunistic behaviors in an upstream market (i.e., Pavlou et al., 2007; Klein et al., 1990). Hence, the use of governance mechanisms in inter-firm relationships is a means to overcome opportunism and thus to lower the associated transaction costs in inter-firm relationship (i.e. improve efficiency). The RBV takes the perspective of the inter-firm relationship, but focusses on the opportunity for relational rents to be generated, ‘a supernormal profit jointly generated in an exchange relationship that cannot be generated by either firm in isolation and can only be created through the joint idiosyncratic contributions of the specific alliance partners’ (Dyer and Singh, 1998). Inter-firm collaborations can thus be seen as new opportunities for the joint creation of competitive strategic advantages, however this is only the case if the relationship can be moved away from merely an arm’s length relationship. This happens when firm’s ‘combine, exchange, or invest in idiosyncratic assets, knowledge and resources/capabilities and/or they employ effective governance mechanisms that lower transaction costs or permit the realization of rents through the synergistic combination of assets, knowledge or capabilities’.

14 Bilateral governance is defined as the reliance on a shared set of implicit behavioral principles for the coordination of interfirm exchanges (Heide 1994). This type of governance is based on relational norms and trust, and can also be interpreted as relational governance mechanisms ( Mohr and Sengupta, 2002). In the following section, the governance mechanisms (contractual and relational) will be discussed in more detail, laying focus on their relationship with inter-firm PACAP and RACAP.

2.3.1 Contractual governance

Contractual governance mechanisms focus on formal control and specification of promises, obligations, and processes for dispute resolution (Poppo and Zenger, 2002). This type of governance mechanism refers to the degree by which rules, procedures instructions and communications are formalized or written down (Jansen et al., 2005).

From a TCE perspective, opportunism can be lowered through contractual governance mechanisms in that agreed upon terms for the relationship are established. Through monitoring and formal evaluation the compliance with these terms can be assessed, which reduces the degree of information asymmetry between parties (and hence lowering the possibility for opportunistic behavior) (e.g. Celly and Frazier 1996; Ryu, 2006; Stump and Heide, 1996). In early stages of an inter-firm relationship, contractual governance mechanisms can help build cooperation between partners, as it provides a degree of transparency and willingness to cooperate (through signing formal contracts, partners show intent to engage in a relationship) (Hamel, 1991; Noordhoff et al., 2011). The shared ‘institutional framework’ that is developed through contracts can work as an enforcement mechanisms against opportunism, which in the early stages of partnership would facilitate cooperation, providing a foundation on which future business relations can evolve (Wang et al., 2011).

15 A firm’s explorative actions are characterized by search, discovery, experimentation, risk taking and innovation and the capability to perform explorative actions is expressed through the firm’s PACAP (Cheng and Van de Ven 1996, March 1991; Zahra and George, 2002). From a RBV perspective, relations will only create substantial benefits if the relationship is moved away from an arm’s reach towards a close-collaborating relation. Hence, the importance of intuition and integration, the dialogue between organizational members and the individual mental models of opportunity finding teams is of relevance from a learning perspective (Lawrence et al., 2005; Sun and Anderson, 2010). Explorative actions (and the firm’s capability to perform these) are positively facilitated by individuals diverging from their existing behavior (Jansen et al. 2006). However, if the degree of contractualization is high, the individual member’s ability to do so will be limited. Contractual obligations and formalization reduce the reliance and trust in people, instead shifting it to the system rather than the individual (Wang et al., 2011). The reliance on rules and procedures is then expected to reduce the likelihood that individuals will deviate from established behavior, thus working as a frame of reference that constricts exploration efforts, only offering limited access to the external environment (Jansen et al., 2006; Lawrence et al., 2005; Weick, 1979). In addition, contractual governance mechanisms can decrease the sense of control over work for partners, diminishing the likelihood of a partner to seek innovative and new solutions (Atuahene-Gima 2003, Damanpour 1991) and limiting the interaction between partners in an inter-firm relationship. In a case study on ambidexterity and governance mechanisms, low levels of partner interaction were seen when high degrees of formal control were put in place.

16 Low degrees of formal control could thus function as mental constructs by which individuals can recognize and value external knowledge. Taking this from an inter-firm perspective, the low degrees of contractualization and formalization could set a common standard for exploration by both partners, allowing them both to perform explorative actions more efficiently. High degrees of contractualization and formalization would negatively impact the capability of the firm to perform explorative actions (PACAP). This leads to the first hypothesis:

H3(a): Contractual governance mechanisms have a positive effect on PACAP, H3(b): at higher degrees of contractual governance will become negative.

With regards to RACAP, this dimension covers the capacity of the firm to perform exploitative processes of transformation and exploitation. Once the external knowledge has been acquired and assimilated from the environment, the knowledge has to be linked to internal processes and converted into internally applicable knowledge. Looking at Lawrence et al. (2005) and Sun and Anderson (2010), these domains are supported by the cognitive processes of integration and institutionalization. The exploitative learning capability occurs through a process of bisociation (Zahra and George, 2002), where two apparently incongruous sets of information are recognized and then combined to arrive at a new organizational schema (Jansen et al., 2005). Having a widely recognized institutional framework can help with the retrieval and recognition of valuable information, easing the bisociation process by making it clear for members what information is already available. It creates a clear shared understanding that can be used as a common framework. In addition, formalization helps partners also codify best practices so as to make knowledge more efficient to exploit, easier to apply, and faster to implement (Lin and Germain, 2003; Zander and Kogut, 1995). In inter-firm context, formalization helps translate knowledge easily between partners and helps to overcome the resistance to change (Lawrence et al., 2005; Sun and Anderson, 2010). Research by Jansen (2005/06) supported this relationship between unilateral governance mechanisms and RACAP. The following addition to hypothesis 1 is made:

17

2.3.2 Relational governance

Relational governance mechanisms are characterized by their focus on partner interaction and focus on cooperative, non-contractual logic (e.g. the implicit rules, trust and relational norms) (Gnan et al, 2015; Ryu et al., 2009). This concept in literature has also been defined as ‘relational contracting’ (Carson et al, 2006) ‘trust’ (Eberl, 2004) or ‘self-commitment’ (Frey and Osterloh, 2002). Whereas formal governance mechanisms can help create transparency and a willingness to cooperate if partner experience is still low, over time actors in inter-firm relations will engage in social interaction and exchanges so that trust may develop (Robson et al., 2008). ‘As the social interactions are repeated through time, and meet basic norms of equity and efficiency, the parties may feel increasingly secure’ (Ring and Van de Ven, 1994). In this way, they will establish implicit social norms that help govern the relationship and lower opportunity risks. These informal, relational governance mechanisms move towards a more voluntary mode of coordination than is present in a structure characterized by contractual governance (Tsai, 2002), this due to the nature of them being established over time.

A distinction exists between instrumental and maxim-based trust (Fink and Kessler, 2009), where instrumental trust draws from sanction and control and maxim-based trust is intrinsically motivated. Instrumental trust can be generated from mutual investments and contracts/formalization (i.e. through contractual governance mechanisms). However, the concept of maxim based trust is build up over time, starting with the actors mutually conceding self-commitment to their cooperation partners (Fink and Kessler, 2009). This is a reciprocal process, where as one partner shows cooperative trust, the other partner reinforces this through additional advance performance, communicating their own self-commitment in a credible manner.

18 information and knowledge (Adler, 2001). Lawrence et al. (2005) and Sun and Anderson (2010), both showed the importance of dialogue and feedback in the processes of exploration and exploitation. Given the previous functioning of trust mechanisms, both PACAP and RACAP are expected to be positively influenced by bilateral governance mechanisms.

19 With regards to explorative and exploitative capabilities (PACAP and RACAP), excessive trust can have detrimental effects on both dimensions. The high interconnectedness and inward focus can lead to a loss of external knowledge being acquired and assimilated (PACAP) and as relationship dynamics stabilize the exploitation and transformation of new knowledge can be limited (RACAP). Concluding, the following hypotheses regarding the effects of relational governance mechanisms on PACAP and RACAP are made:

H4 (a): Relational governance mechanisms will have a positive effect on inter-firm PACAP that H4 (a1): becomes negative at high levels of relational governance.

H4 (b): Relational governance mechanisms will have a positive effect on inter-firm RACAP that H4 (b1): becomes negative at high levels of relational governance.

2.4 The interaction of contractual and relational governance

The abovementioned theoretical discussion on governance mechanisms and ACAP has shown that PACAP and RACAP are influenced by varying degrees of contractual and relational governance mechanisms. However, the question that remains is in what way both types of governance mechanisms relate to one another when employed in inter-firm relationships. For more than a decade, the interplay between contractual and relational governance in inter-firm relationships has been a considerable debate (Li et al., 2010b; Liu et al., 2009; Lumineau and Henderson, 2012; Poppo and Zenger, 2002).

Research on contractual and relational governance mechanisms has assessed two different types of relationships, the mutual relationship (i.e. between contractual and relational governance) and the shared effects (i.e. the combined effect of contractual and relational governance). This research has led to several conflicting conclusions. Due to this ambiguous nature of the relationship and interaction between contractual and relational governance, the following section will set apart the arguments for both sides, and propose a viewpoint for the remainder of this study.

20 of governance decreases the use of, or the benefits of the other one (Cao and Lumineau, 2015) whereas when functioning as complements the one increases the use of or benefits of using the other. From the ‘substitutes’ perspective, the main argument focusses on two mechanisms that cause both governance mechanisms to substitute one another, namely ‘replacing’ and ‘dampening’ (Huber et al., 2013). Contractual and relational governance are seen as ‘functional equivalents’, where the use of one can dampen the effects of the other. Contractual governance is for example seen as redundant when relational norms have been built, as these are capable to govern the inter-firm relationship on its’ own. Contractual governance could harm relational governance development by showing a lack of trust (Ghoshal and Moran, 1996; Gulati and Nickerson, 2008; Malhotra, 2009).

On the other hand, the complements perspective argues that contractual governance mechanisms can be used purposefully to support relational governance mechanisms (and the development thereof) (Cao and Lumineau, 2015; Zhang and Zhou, 2013). These contractual terms can motivate parties to cooperate, furthering the opportunity for relational governance to develop (Cao and Lumineau, 2015; Poppo and Zenger, 2002). In this case, information asymmetry is reduced, which supports a cooperative climate (Yang et al., 2012). The other way around, relational governance mechanisms can also function as enablers of contractual governance mechanisms. Once the relationship has developed a certain degree of relational norms, a new ‘mindset’ can be developed, that can be used to alter the contractual terms in such a way that they better fit the newly developed relationship (i.e. create a ‘win-win’ mindset) (Bastl et al., 2012). Additionally, contractual and relational governance can function ad compensating mechanisms, where the one type of governance mechanisms can overcome shortcomings in the other type of governance mechanisms (Li et al., 2010). Parties can learn from one another, and in time could refine their contracts to take into account situations that have occurred along the relationship (Poppo and Zenger, 2002; Ryall and Sampson, 2009). In this way, contractual and relational governance are seen as complements to one another.

22 From a TCE perspective, contractual and relational governance would function as complements in reducing opportunism and transaction costs in the relationships (as presented in the previous section). From a RBV perspective, the opportunity for relational rents to be created only becomes realized once the relationship moves away from an ‘arms-length’ character. The perspective that this study takes is that governance mechanisms do not exist in a bipolar vacuum, but rather that these are spread across a continuum, that it is a fluent construct in which ideally, both can exist at the same time (e.g. Gereffi et al., 2005; Lawrence et al. 2005). In this sense, contractual governance mechanisms can function as supportive structures to improve the trust and relational norms, bringing the two parties closer in communicating/cooperating with each other. Hence, when it comes to the shared effect of contractual and relational governance on PACAP and RACAP, this study proposes the following hypotheses:

H5a: The interaction effect of contractual and relational governance mechanisms will have a positive effect on inter-firm PACAP

H5b: The interaction effect of contractual and relational governance mechanisms will have a positive effect on inter-firm RACAP

2.5 Relational context and governance mechanisms

The previous section has developed hypotheses regarding contractual and relational governance mechanisms and their interplay with PACAP and RACAP. However, up until this point the relational context has not been assessed directly. The following section will discuss the way in which the relational context (relationship duration and importance) is included in this study, and makes hypotheses regarding the

23 2010). Due to the complex nature of the interaction between governance mechanisms and ACAP, conclusions regarding relationship duration will be made in the form of a Multi-Group-Analysis (MGA) to check whether significant differences exist between early and late stages in the relationship. This will be done to gain more understanding of the complex interactions between contractual/ relational governance and ACAP.

24

3. Conceptual model

The following section portrays the conceptual model, in addition to an overview of the hypothesized relationships between the concepts.

3.1 Conceptual model

Variables included for Multi-Group-Analysis (MGA): - Relationship duration

- Relationship importance

3.2 Hypotheses

H1: PACAP will be positively related to explorative learning performance. H2: RACAP will be positively related to exploitative learning performance.

H3(a): Contractual governance mechanisms have a positive effect on PACAP which, H3(b): at higher degrees of contractual governance will become negative.

H3(c): Contractual governance mechanisms have a positive effect on inter-firm RACAP.

H4 (a): Relational governance mechanisms will have a positive effect on inter-firm PACAP that H4 (a1): becomes negative at high levels of relational governance.

H4 (b): Relational governance mechanisms will have a positive effect on inter-firm RACAP that H4 (b1): becomes negative at high levels of relational governance.

H5a: The interaction effect of contractual and relational governance mechanisms will have a positive effect on inter-firm PACAP.

25

4. Methodology

4.1 Data collection procedures

This study works with the database as collected in the study by Berger (2015) to test its hypotheses. Given that relationships are not directly observable (and hence largely susceptible to measurement error) (Selnes and Sallis, 2003), data has been obtained by means of questionnaires sent to employees central to the customer relationship, on both sides of the buyer-supplier dyad, as suggested by John and Reve (1982), Heide and John (1992), Lane and Lubatkin (1998) and Selnes and Sallis (2003).

Key informants on both sides of the dyad were asked to supply contact data for four employees within their company, in total 166 matched-pair relationships were accounted for. Service companies were excluded from the study since these are shown to have a lower probability to systematically acquire the external knowledge necessary to conceive new or improved products and or processes (Berger, 2015). Respondents range from the automotive, machinery, chemicals, pharmaceuticals, semiconductors and electronics industries. This to support variance and to account for the observation that knowledge strategies can differ per industry (Berger, 2015).

26 4.2 Measures

The scales in this study were taken from existing scales from existing literature on the supplier relationships, firm learning, intrafirm learning or interfirm learning (not related to buyer-supplier relationships) context, as presented in the study by Berger (2015). The measures were extended to vertical interfirm relationships. First- and second-order constructs are used in the study, according to Berger (2015): “A first-order construct is a latent construct that has observed variables as indicators, whereas second-order constructs have other latent (first-order) constructs as their indicators”. In this study, PACAP is an example of a second order construct, consisting of the first order constructs ‘Acquisition’ and ‘Assimilation’.

In addition, the distinction between reflective and formative indicators is made. A construct that consists out of reflective indicators uses the indicators to ‘reflect’ the construct, hence if the value of the construct changes, so do the values of the indicators. High correlations are assumed among reflective indicators (Berger, 2015). On the other hand, a construct that consists of formative indicators has different aspects of the construct measured by each indicator. These formative indicators are thus not related and changes in the value of the indicators determine changes in the latent construct (Berger, 2015).

The first order constructs were measured with several indicators. All questions were measured by means of 7-point Likert scales, ranging from ‘strongly disagree’ (1) to ‘strongly agree’ (7). Explorative learning performance is an exception to this, as it was anchored by ‘to no extend’ and ‘to a great extend’. To retain participant attention during the questionnaire, several indicators were reverse coded. These were recoded in order to perform the analysis.

To counter nonresponse, the questionnaire was translated into Dutch (from English) as well as back-translated to English to ensure both versions measure the same.

27 obtained from Desphande and Zaltman (1982), Cannon and Perrault (1999), Buvik and Reve (2002), and Jansen et al. (2006).

Relational governance (a second-order construct) is characterized by a focus on partner interaction and cooperative, non-contractual logic (e.g. trust and relational norms) (Gnan et al, 2015). This construct was assessed using measures developed by Heide and John (1992), representing flexibility, information exchange and solidarity norm types (first order constructs).

Since appropriate scales for the dimensions of ACAP were not available, they were specifically designed by Berger (2015) for the study. Given that ACAP is conceived as a dynamic capability, the definitions and components of its dimensions were obtained from Zahra and George (2002). As suggested by Lane et al. (2006), all ACAP dimensions refer to capabilities and are linked to the individual, group and relationship level (Sun and Anderson, 2010). ACAP consists of two second order constructs, PACAP and RACAP, which in turn both consist of first-order constructs (to be discussed below).

PACAP is a second order construct consisting of ‘acquisition’ and ‘assimilation’ (first order constructs). Acquisition (first order construct) is ‘a firm’s capability to identify and acquire externally generated knowledge that is critical to its operations’ (Zahra and George, 2002). Prior investments (scope of search), prior knowledge (perceptual schema), intensity (new connections), speed (of learning) and direction (quality of learning) are seen as its relevant components. Berger (2015) has taken prior investment and prior knowledge from the construct measure, since these were measured as antecedents of ACAP in his theoretical model instead of parts of the acquisition construct. Measures were based on work by Szulanski (1996) and Camisón and Forés (2010). Assimilation (first order construct) refers to the firm’s routines and processes that allow it to analyze, process, interpret and understand the information obtained from external sources (Berger, 2015). Understanding (interpretation, comprehension, learning) is its relevant component. The generated measures were based on measures by Jansen et al. (2005) and Camisón and Forés (2010).

28 transformation. Measures by Bontis et al. (2002), and Cadiz et al. (2009) were used as a basis for the generated measures. Exploitation is based on ‘the routines that allow firms to refine, extend, and leverage existing competencies or to create new ones by incorporating acquired and transformed knowledge into its operations’ (Berger, 2015). Use and Implementation (core competencies, harvesting resources) are the core components. Measures are based on work by Jansen et al. (2005), Cadiz et al. (2009), and Camisón and Forés (2010).

Whereas ACAP expresses the capability to learn, Explorative and Exploitative learning performance relates to the direct outcomes (i.e. performance) of a firm’s learning processes. Explorative processes include search, variation, risk taking, experimentation, play, flexibility, discovery, or innovation (March, 1991). Explorative learning performance is a result of ‘search, variation, risk taking, experimentation, play, flexibility, discovery, or innovation’ (March, 1991) and conveys the measurable results of cooperation and knowledge exchange. The measures are based on work by Lane et al (2001) and covers the joint creation of expertise on five dimensions (manufacturing and production, product development, technological, marketing and managerial). Exploitative learning performance is a result of ‘refinement, implementation, efficiency, production and selection’ (Cheng and Van de Ven 1996, March 1991) and shows the firms’ success in commercializing new external knowledge. The measures are based on work by Selnes and Sallis (2003).

4.2.1 Multi-group measures

29 Relationship importance was directly measured under the respondents. Given that a minimum degree of importance is required to start a relationship in the first place (Caniels and Gelderman, 2005) the respondents were asked to choose between ‘average’ or ‘crucial’ in terms of relationship importance. Morgan and Hunt (1994) stated that a firm will “warrant maximum efforts” to maintain a relationship if the firm beliefs the ongoing relationship is important, referring to this as commitment. Hence, when relationship importance would be identified as ‘low’ no relationship between buyer and supplier would be present, as the market would take preference.

4.3 Descriptives

The database consists of 166 matched-pair relationships, which provides 332 respondents for the study. From this relationship, 40.7% of the relationships operated in the automotive/machinery industry, 28.9% in the chemicals/pharmaceuticals industry and 17.2% in the semiconductors/ electronics industry (13.3% of the respondents did not specify the industry).

Concerning the lengths of the relationships studied, 31% of the respondents (n=103) had a relationship duration of less than four years, whereas 66.6% had a relationship duration of more than 4 years (n=221), 2.4% did not specify. The mean lengths of the relationship was 11.03 years (SD = 9.79). Finally, the distribution of importance was shown as 43.4% indicating an average importance (n=144) and 55.4% indicating a crucial importance (n=188), 1.2% did not specify.

4.4 Statistical procedures

30 4.5 Approach

First of all, the database was prepared for analysis in SmartPLS. This included the construction of the DURATION variable, which split the variable RELATIONSHIP DURATION (Var 0.1) into two categories: ‘1= Less than four years’, ‘2= More than four years’.

Once the database had been prepared for analysis, the reliability and validity of the indicators, constructs (reflective and formative taken separately) as well as the measurement model was tested.

Next, the second order constructs had to be approached in SmartPLS3. To do so, the first-order construct latent variable scores were calculated and saved. These were then imported and included into a new model as indicators for the second-order constructs. Once this step had been completed, the second-order constructs (i.e. PACAP, RACAP and Relational governance) were created.

Once the constructs had been established, the first analysis using the structural model had been performed. The PLS Algorithm provided the path coefficients of the hypothesized relationships between the dependent and independent variables, whereas bootstrapping provides data on the significance of the effects discovered. A curvilinear regression analysis was performed in SPSS to further test the relationships between governance modes and PACAP/RACAP.

31

5. Results

In the following section the results of the analyses will be presented. First of all, data reliability and validity will be discussed. Since reflective and formative scales are used, which require different measures for reliability and validity, these will be discussed separately. Next, the results of the structural model will be discussed, after which the multi-group-analysis will be presented.

5.1 Reliability and validity – reflective scales

Indicator reliability was assessed using the loadings of each indicator (Factor Loadings). These scores should be above the threshold of 0.5 and significant (i.e. based on p- or t- value scores, Hair et al., 2012). All indicators on the reflective scales were significant with a loading of over 0.5. Please refer to Appendix B, convergent validity of reflective scales for an overview.

32 5.2 Reliability and validity – formative scales

As for formative constructs, the indicators measure different dimensions of the same construct and are not directly related to one another (Hair et al., 2011). Deletion of formative indicators should be assessed carefully, since this could change the meaning of the construct. Formative constructs do not work with traditional measures of convergent and discriminant validity. In order to assess the convergent validity, the weights, p-values and variance inflation factors (VIF) were measured. To ensure no multicollinearity being present among the formative indicators, VIF scores of above 3.3 will be rejected (Diamantopoulus and Siguaw, 2006). In total, three questions were removed based on inadequate assessment scores. These were ‘Acquisition 2’, ‘Transformation 2’ and ‘Transformation 3’. According to the reasoning of Berger (2015), deleting these indicators from the construct does not theoretically change it, since all indicators belong to pairs of questions. For an overview of the formative construct reliability and validity scores, refer to Appendix C. Appendix D. contains an overview of the deleted constructs.

5.3 Discriminant validity

33 5.4 Hypothesis testing - structural model

The results of the structural model are based on the SmartPLS3 algorithm and bootstrapping function. The maximum number of iterations on the PLS algorithm was set to 1000, and the number of bootstrapping subsamples to 5000. These higher number of iterations were used to improve reliability in comparison to the standard settings of SmartPLS (300, 500 respectively). For an overview of the structural model analysis results, please refer to Appendix F.

First of all, to show the practical significance of the study, the relationship between PACAP, explorative learning performance and RACAP, exploitative learning performance was tested. It was hypothesized that these relationships are positive for both PACAP (H1) and RACAP (H2) (p<0.001). The results of the structural model show that for both buyer and supplier, these relationships are positive and significant, thus H1 and H2 are supported.

34 As for relational governance, significant positive relationships are found between both relational governance and PACAP as well as RACAP. This counts for the buyer as well as the supplier database (p<0.001). The MGA results show that there are significant statistical differences between buyer and supplier, where for the buyer-only database the relationship is stronger. The ‘dark side of trust’ was hypothesized through an inverted U-shaped relationship between relational governance and PACAP/RACAP. Significant support for this relationship was found at the buyer database but not for the supplier database. However, no significant statistical differences were found with the MGA analysis (p=0.722, p=0.818) as well as for the combined database. Hence, no supporting conclusions can be made regarding this relationship.

The results of the structural model show support for H4a and H4b, however additions H4a1 and H4b1 (assuming an inverted U-shaped relationship) had to be rejected.

Regarding the interaction between contractual and relational governance, this study hypothesized a complementary relationship. This was tested by measuring the interaction effect of contractual and relational governance on PACAP and RACAP, captured by H5a and H5b. For PACAP, no support for an interaction effect was found for buyer nor supplier. For RACAP, only a significant relationship was found for the buyer database (p=0.025), with the MGA results indicating statistical differences (p=0.961, two tailed). However, this relationship is negative (β=-0.154), and thus no support for a positive complementary effect between contractual and relational governance is found. H5a and H5b are thus rejected.

35 5.5 Multi-group-analysis results

Up until this point, the interaction between governance mechanisms, and their relationship with ACAP has been assessed from a buyer and supplier perspective. However, assessing whether relational characteristics make a difference has also been included as one of the goals in this study. Therefore, the relationship between governance mechanisms and ACAP will be further assessed by checking for relational characteristics.

The following section will provide conclusions on whether there are significant statistical differences between relationships with a ‘short duration’ (i.e. less than four years) or with a ‘long duration’ (i.e. more than four years). In addition, a similar analysis is conducted based on the relationship importance, to see whether significant statistical differences exist between relationships of ‘average’ or ‘crucial’ importance.

5.5.1 Relationship duration, ‘less than four years’ versus ‘more than four years’

A Multi-Group-Analysis has been performed to analyze whether significant statistical differences exist between relationships of ‘less than four years’ and ‘more than four years’. Relationship duration has been split at the four year mark, to allow for a less skewed distribution of the samples. Based on the work of Berger (2015) and Johnson et al. (2004) a distribution was made at the two year mark. However, this would leave a distribution of n=66 versus n=258. For a better distribution, a split at four years was made, which gives a distribution of n=103 versus n=221. This was based on Amoros and Bosma (2013) who indicated the first 42 months (4 years rounded up) as crucial for entrepreneurial activity in firms.

36 The only significant statistical difference found was for the effect of relational governance on PACAP (p=0.048). The path coefficient for this relationship is lower in relationships with a duration of more than four years (β=0.535 versus β=0.339). No support among both groups was found for the interaction effect between contractual and relational governance, as well as for the inverted U-shaped relationships for higher levels of contractual and relational governance on PACAP/RACAP. See Appendix H. for an overview of the results.

5.5.2 Relationship importance, ‘average’ versus ‘crucial’ importance

The other relational aspect which was assessed in the multi-group analysis was the relationship importance. Relationship importance was measured as being either (1) ‘average’ or (2) ‘crucial’. The sample consists of 144 respondents that indicate the relationship as being average, and 188 respondents with an importance indication of crucial.

Contractual governance appears to have no effect on PACAP and RACAP for a group of average importance, as opposed to the group with crucial importance. However, no significant statistical differences can be found, so no conclusions can be made on this area.

As for relational governance, the group with crucial importance sees a stronger relationship between relational governance and both PACAP (β=0.457 vs β=0.301) and RACAP (β=0.538 vs β=0.358). These differences are significant at the 90% confidence level (PACAP, p=0.917, two tailed; RACAP, p=0.925, two tailed).

37

6. Discussion

This paper aims to research the impacts of governance mechanisms (contractual and relational) on the inter-firm absorptive capacity (ACAP). In addition, this paper set to find out how these relations can differ when accounting for relational characteristics (relationship duration and relationship importance). The following section discusses the findings and compares these to the existing literature, after which the contribution to literature is discussed. First of all, the findings of the structural model are discussed, after which the impact of relational characteristics is discussed.

6.1 The effects of inter-firm ACAP on explorative and exploitative learning

The practical significance of this study has been tested by linking the dimensions of ACAP (PACAP and RACAP) directly to organizational learning by means of the explorative and exploitative learning performance. Organizational learning performance has been closely related to organizational performance, through its positive impact on new product development and enhanced creativity of new development activities (Moorman and Miner, 1997; Saban et al. 2000). The results of the structural model have shown that for all databases, PACAP has a significant positive effect on explorative learning performance whereas RACAP has a significant positive effect on Exploitative learning performance. For firms, this means that actively managing PACAP and RACAP through governance mechanisms does result in actual performance benefits. Therefore, the interaction effects between contractual and relational governance become more interesting, as managers will directly see the effects of good or bad governance. The following section will discuss the results of the interaction between governance mechanisms, PACAP and RACAP in more detail.

6.2 Impact of governance mechanisms on inter-firm ACAP

38 would positively relate to PACAP (Anderson and Sun, 2010; Jansen et al., 2006; Lawrence et al., 2005), with an inverted U-shape relationship. A positive relationship between contractual governance and RACAP was also hypothesized, as has been proven in previous studies (e.g. Jansen, 2005/06). The structural model found support for a positive effect between contractual governance and PACAP for the buyer database (not for the supplier database). In addition, support for a positive relationship of contractual governance on RACAP was found for the buyer only database (supplier only database showed no significant positive effect. The MGA found both results to be significantly different from each other. Support for an inverted U-shaped relationship between contractual governance and PACAP was not found for any of the databases, which is most likely a result of the already weak relationship between contractual governance and PACAP/RACAP.

The relationship between relational governance and PACAP/RACAP has proven to be positive by previous studies (e.g. Berger, 2015; Jansen et al., 2005) and accordingly, this study hypothesized a positive relationship between both relational governance and PACAP as well as RACAP. However, taking into account the ‘dark side of trust’ perspective, an inverted U-shaped relationship was expected, as the relationship with PACAP and RACAP could become negative given too much trust (Gargiulo and Ertug, 2006). The structural model has shown support for a significant positive effect of relational governance on both PACAP and RACAP, for the buyer as well as the supplier perspective. The relationship between relational governance and PACAP/RACAP is found to be stronger for the buyer database in comparison to the supplier database. A Multi-Group-Analysis has confirmed that the buyer and supplier databases differ significantly when it comes to relational governance. Support for the ‘dark side of trust’ has only been found among the buyer database, not in the supplier database. However, the Multi-Group-Analysis has shown there to be no significant statistical differences between both groups in this regard.

39 another. The structural model has found no support of an interaction effect between contractual and relational governance, for the buyer nor the supplier. Instead, the buyer database showed a negative relationship between the interaction effect and PACAP/RACAP. However, the Multi-Group-Analysis showed no significant statistical results between both buyer and supplier databases. Further research into the interaction between contractual and relational governance revealed that for the buyer database, they indeed function as substitutes. Both governance mechanisms were found to have a negative relationship to one another, meaning that if one increases the other decreases. The Multi-Group-Analysis has also shown that the differences between buyer and supplier database are statistically significant.

40 not been found. This means that high levels of trust are not negatively related to PACAP and RACAP, and that the presumed negative effects are possibly outweighed by the positive effects of building trust in inter-firm relationships. However, a different type of relationship could exist, one where the maximum effectiveness of trust is reached at a certain level (i.e. a trust ceiling). This relationship has not been tested for in this study.

Contractual and relational governance function as substitutes for the buyer database. This is in line with research done by Wang et al. (2011) Huber et al. (2013), Li et al. (2010) who state that these differences might be explained by differences in perceptions of the relationships. Buyers will be more likely to manage their relationship and alleviate the risks associated with it (Ambrose et al., 2010), meaning that they will put more effort into deciding the right governance structures for the firm. This would mean one relational trust is build, it takes the place of contractual governance in relationships.

6.3 Impact of relational context on governance mechanisms in inter-firm relationships

Next to the overall structural model, the effects of relational characteristics in the form of relationship duration and relationship importance were assessed. What can be seen from the individual group results is that for the longer relationships, relational governance has a weaker relationship with PACAP and RACAP. In the short duration relationships, contractual governance shows to have little to no effect on PACAP and RACAP. However, the results do not differ significantly based on the Multi-Group-Analysis. This is due to the skewed distribution of the samples, which could negatively impact the results. Based solely on the buyer and supplier only results, it is possible to conclude that in the early stages of the relationship buyers and suppliers should seek to develop trust through frequent interactions which can result in basic norms of equity and efficiency (Ring and van de Ven, 1994). This can make both parties feel increasingly secure in later stages of the relationship, improving its performance.

41 Gereffi, 2005, and Morgan and Hunt (1994) where they state that firm’s will warrant maximum efforts when they see a relationship as being more important. The high importance of the product or the relationship will motivate them to put additional effort into developing mutual benefits and sharing knowledge, as opposed to a relationship of average importance, which could be conducted more at arms’ length.

7. Managerial implications.

This study has assessed the governance of inter-firm relationships, more specifically the processes of inter-firm knowledge sharing. In order for effective knowledge sharing to take place however, the firm’s absorptive capacity should be managed accordingly. The results have shown strong significant effects of relational governance on PACAP and RACAP. Managers in inter-firm relationships should thus focus on developing and implementing relational norms and trust as one of the key governance mechanisms in relationships. This is also supported by the notion that contractual and relational governance have been found to function as substitutes to one another, instead of complements. Managers that refuse to give up or revise certain contracts could find themselves to hinder the performance of the relationship in the long run.

8. Limitations and future research

42 With regards to the relational characteristics applied to this study, the relationship duration is an area which could receive further attention. As for the sake of detail that was possible in this study, the relationship duration was proposed as being either of ‘less than four years’ or ‘more than four years’. However, if looking at the business life cycle, several stages in a firm can be found, from the initial ‘birth stage’ up to the ‘decline stage’ (e.g. Phan et al, 2014). A more detailed assessment of the relational evolution could provide more insight for managers and researchers as to how governance mechanisms relate to one another (and perhaps even change this relationship), as well as how the Absorptive Capacity could be developed over time.

Perspectives from both buyer and supplier have been taken into consideration for this study, however for the Multi-Group-Analysis this was proven to be difficult. The sample that had been collected consisted of 332 respondents, which would leave rather small samples up for analysis, would the distinction between buyer versus supplier on short and long duration/ high and low importance have been made. Therefore, more research with a bigger database (e.g. a separate buyer and a separate supplier study) would be able to add to this research. This could allow for a better understanding of the different ways in which relationships are perceived for buyer and suppliers, and can help managers engage accordingly.

43

9. Conclusion

44

References

Ahmadjian, C. L., & Lincoln, J. R. (2001). Keiretsu, Governance, and Learning: Case Studies in Change from the Japanese Automotive Industry. Organization Science, 12(6), 683-701.

Alberto F. de Toni, Fabio Nonino, (2010),"The key roles in the informal organization: a network analysis perspective", The Learning Organization, Vol. 17 Iss 1 pp. 86-103

Antia, K. D., & Frazier, G. L. (2001). The Severity of Contract Enforcement in Interfirm Channel Relationships. Journal Of Marketing, 65(4), 67-81.

Ambrose, E., Marshall, D., & Lynch, D. (2010). Buyer supplier perspectives on supply chain relationships. International Journal Of Operations & Production Management, 30(12), 1269-1290.

Amoros, J. E. & Bosma, N. (2013) Global Entrepreneurship Monitor 1013 Global Report

Atuahene-Gima, K. (2003). The Effecs of Centrifugal and Centripetal forces on Product Development Speed and Quality: How does Problem Solving Matter? Academy Of Management Journal, 46(3), 359-373.

Bastl, M., Johnson, M., Lightfoot, H., Evans, S., (2012). Buyer–supplier relationships in a servitized environment: an examination with Cannon and Perreault’s framework. International Journal of Operations & Production Management 32, 650–675.

Bello, D. C., & Gilliland, D. I. (1997). The Effect of Output Controls, Process Controls, and Flexibility on Export Channel Performance. Journal Of Marketing, 61(1), 22-38.

Berger, J. (2015). Essays on the Governance of Buyer-Supplier Relationships. Groningen: University of Groningen.

45 van den Bosch, F. J., Volberda, H. W., & De Boer, M. (1999). Coevolution of Firm Absorptive Capacity and Knowledge Environment: Organizational Forms and Combinative Capabilities. Organization Science, 10(5), 551-568.

Birkinshaw, J., Bresman, H., & Nobel, R. (2010). Knowledge transfer in international acquisitions: A retrospective. Journal Of International Business Studies, 41(1), 21-26.

Bontis, N., Crossan, M. M., & Hulland, J. (2002). Managing an organizational learning system by aligning stocks and flows. Journal of Management Studies, 39(4), 437-469.

Buvik, A., & Reve, T. (2002). Inter-firm governance and structural power in industrial relationships: The moderating effect of bargaining power on the contractual safeguarding of specific assets. Scandinavian Journal of Management, 18(3), 261.

Cadiz, D., Sawyer, J. E., & Griffith, T. L. (2009). Developing and validating field measurement scales for absorptive capacity and experienced community of practice. Educational & Psychological Measurement, 69(6), 1035-105

Camisón, C., & Forés, B. (2010). Knowledge absorptive capacity: New insights for its conceptualization and measurement. Journal of Business Research, 63(7), 707-715

Caniëls, M. C. J., & Gelderman, C. J. (2005). Purchasing strategies in the kraljic matrix—A power and dependence perspective. Journal of Purchasing & Supply Management, 11(2), 141-155.

Cannon, J. P., Ravi S., A., & Gregory, G. (2000). Contracts, Norms, and Plural Form Governance. Journal Of The Academy Of Marketing Science, 28(Spring)