Discovering the nature of coopetition in a cross-national setting
through an ambidextrous lens
Abstract
This paper builds upon the ambidexterity, alliance and coopetition literature in an endeavour to gain more insight into the nature of coopetitive agreements. By taking up an ambidextrous approach, this paper determines the nature of selected cross-national coopetitive agreements and illustrates how explorative and exploitative coopetitive agreements differentiate from each other. As this is the first time such an approach has been used, classifying the nature of coopetitive agreements as either explorative or exploitative adds a new dimension to existing literature. Furthermore, this study adds structural and motivational implications to the field and provides a solid basis for further research on this particular subject. The findings show that explorative and exploitative agreements differ on a broad variety of aspects, including their primary goals, motives, organizational routines and time orientation. This differentiation provides managers with an understanding of the vital characteristics of both natures, which relate differently to the strategic, operational and environmental factors tested in this paper. Since the nature affects the structure, strategy, scope and capabilities of coopetitive agreements, insight into this enables managers to act and react accordingly.
Student Mark A. Danhof (11398294)
Track MSc. in BA – International Management
First Supervisor Dr. C. Gelhard Second Supervisor Dr. N. Pisani
Date of Submission June 22, 2017 (final)
2
Acknowledgments
I would like to thank my lead supervisor, dr. C. Gelhard, for his feedback during the thinking and writing phase. When necessary, he steered this research in the right direction, while stimulating autonomous behaviour and leaving freedom for own interpretation. Also, I would like to mention the people close to me for supporting me throughout all those years at university. My sincere gratitude to my parents, brother and girlfriend.
Statement of originality
This document is written by Mark Alexander Danhof who declares to take full responsibility for the contents of this document.
I declare that the text and the work presented in this document is original and that no sources other than those mentioned in the text and its references have been used in creating it.
The Faculty of Economics and Business is responsible solely for the supervision of completion of the work, not for the contents.
3
Contents
Introduction ... 4 Literature review ... 8 Ambidexterity ... 8 Coopetition ... 13 Exploration ... 16 Exploitation ... 17 Framework ... 18 Working propositions ... 19 Methodology ... 28 Case study... 28 Sample ... 29 Data ... 30 Data analysis ... 31 Description of dimensions ... 32 Description of cases... 33Analysis and results ... 36
Within-case analysis ... 36 Shell ... 36 Unilever ... 39 Philips ... 42 Heineken ... 44 KLM ... 47
Cross-case and cluster analysis ... 50
Findings per case ... 57
Findings per cluster ... 57
Overview of working propositions ... 58
Discussion ... 59 Academic implications ... 63 Managerial implications ... 64 Conclusion ... 66 Bibliography ... 68 Appendix: CD-matrix ... 76
4
Introduction
“Business is cooperation when it comes to creating a pie and competition when it comes to dividing it up. (…) In other words: Business is simultaneously war and peace.” (Brandenburger & Nalebuff, 1996, p. 4).
The public announcement of the cooperation between Sony and Samsung, two competitors, surprised many people in the world of international business (IB), both in a positive and a negative sense. From 2003-2009 both companies worked together intensively in the field of LCD-panel technology to boost innovation, while competing fiercely in other segments of the market. This paradoxical form of cooperation became a major success story with both companies tripling their initial investment of 1 billion each. In the long-term, this partnership resulted in a major market share leaving their competitors behind for many years to follow (Bloomberg, 2016; Gnyawali & Park, 2011).
Globalization has led to increased integration and cooperation between companies. Whereas in the past mistrust prevailed and companies predominantly conducted business themselves, the current era reveals a growing trend in which companies continuously seek new forms of cooperation and alliances. Even though caution remains commonplace, increased competition requires companies to commit themselves to unorthodox partnerships with their rivals, so-called coopetition, which forms an important part of IB nowadays (Dorn, Schweiger, & Albers, 2016). Especially in the global context, this paradoxical relationship in which multinational enterprises (MNEs) are both partners and rivals, friend and foe, has become increasingly important. Already at the turn of the century, Harbison and Pekar (1998) claimed that half of the agreements had coopetitive elements in its nature.
Coopetition can be defined as the simultaneous cooperation and competition between rivals (Brandenburger & Nalebuff, 1996; Luo, 2007). At first sight, coopetition feels
5 contradictory since it contains diverging interests (Brandenburger & Nalebuff, 1996). Whereas cooperation departs from a positive point - we benefit from each other-, competition has a negative starting point - we compete against one another- (Deutsch, 2006). Therefore, paradoxical as it sounds, it is important to understand that coopetition can be a “win-win” scenario in which both competitors are able to generate significant benefits. Even though coopetition has become increasingly present in society, academic research lagged behind for years. Only in the past decade has research into coopetition developed, thereby being more thorough and extensive (Dorn et al., 2016; Walley, 2007).
Nevertheless, specific research into the nature of the competition-cooperation paradox and coopetitive agreements is limited (Gnyawali, Madhavan, He & Bengtsson, 2015). Increased understanding of this nature is vital to understand the motives behind those collaborations between competitors. The nature of the agreement is often determinant for the scope of and can impact all further steps and factors in the agreement (Dorn et al., 2016). This importance is acknowledged by Dahl, Kock and Lundgren-Henriksson (2016) in their recent contribution, proposing that further research needs to be conducted to explore the nature of coopetitive practices at the organizational level.
To uncover the nature of coopetition, this paper integrates coopetition with ambidexterity (AD). AD deals with the simultaneous execution of both explorative and exploitative activities by companies (March, 1991). Whereas exploration focuses on new possibilities, exploitation focuses on improving existing resources and competences. Explorative activities encompass searching, innovating and experimenting, generally implying a higher degree of both flexibility and uncertainty, focusing on the long-term. Exploitation focuses on efficiency-seeking, production and implementation, involving both a lower risk and higher profits in the short-term (March, 1991; Turner, Swart, & Maylor, 2013). This
6 differentiation enables this research to classify selected agreements, gain insight into its nature and to study differences between explorative and exploitative agreements.
This paper adopts a cross-national perspective focusing on AD across borders, or international AD, which is the process by MNEs to pursue and integrate both exploitative and explorative processes when investing abroad to further strengthen their position and to increase its competitiveness (Hsu, Lien, & Chen, 2012; Luo & Rui, 2009; Prange & Verdier, 2011). An agreement in an international context involves two or more companies, in which at least one partner is headquartered outside of the country of operations (Nielsen & Gudergan, 2012). Exploration and exploitation can be pursued separately via a differentiated approach or pursued both within the same organizational unit, which is called an integrative approach (Raisch, Birkinshaw, Probst, & Tushman, 2009). For AD in general, it should be stated that firms that are able to conduct this integrative approach, pursuing both exploration and exploitation simultaneously, achieve better results than firms choosing one over the other (Hsu et al., 2012; Luo & Rui, 2009; Prange & Verdier, 2011).
However, as March (1991) noted, both processes compete for the same set of scarce resources, implying that there is a trade-off for organizations to make and a balance to find in order to manage both effectively. It is important to keep in mind that exploration and exploitation are essentially different from each other. Since both aim for fundamentally different outcomes, each requires different types of structure, strategies, capabilities and knowledge (He & Wong, 2004; March, 1991). Since this paper reviews individual agreements in isolation rather than addressing firm internationalization or ambidextrous performance in general, it is able to label agreements as either explorative or exploitative in nature. Studying individual selected agreements rather than a firm’s AD behaviour in general is in congruence with previous research and is not uncommon in this field (Koza & Lewin, 1998; Rothaermel, Hitt, Ireland, Camp, & Sexton, 2001).
7 By integrating AD-literature with coopetitive agreements, the nature of selected agreements is determined in this paper. Few studies have integrated both streams already (Luo & Rui, 2009; Strese, Meuer, Flatten, & Brettel, 2016), but no study has utilized AD to determine the nature of coopetitive agreements. Furthermore, as research on coopetition in an international setting is still in its infancy, the contextual factors have not received much attention either (Bengtsson & Kock, 2014; Bouncken, Gast, Kraus, & Bogers, 2015). Both cluster types depart from diverging purposes, involve different sets of activities and therefore require different management (He & Wong, 2004; March, 1991). This knowledge regarding the differences between explorative and exploitative coopetition is highly valuable for managers since it allows them to create or adjust their structure, strategy and capabilities accordingly. In addition to giving insight into the nature, this research adds structural and motivational implications to the study of coopetition.
In short, this paper integrates ambidexterity with coopetition in a cross-national setting. The following research question is leading throughout this paper:
Taking on an ambidextrous perspective, how do explorative coopetitive agreements and exploitative coopetitive agreements differentiate from each other?
Through an in-depth multiple case analysis on the nature of nine coopetitive agreements, the nature of each case is determined as either predominantly explorative or exploitative. Seven working propositions are developed, categorized into strategic, operational and environmental factors, to assess their relationship with the nature of the agreement. These will be assessed through a within-case, a cross-case and a cross-cluster analysis. This article builds upon existing literature and serves as a window to further understand the nature of different coopetitive agreements and its contextual factors.
8 First, key literature will be discussed. Then, a categorization of strategic, operational and environmental factors is developed into working propositions and a framework is established. Subsequently, the methodology, analysis and the results are discussed. This paper concludes with a discussion, implications and areas for further research.
Literature review
This section begins with a further elaboration upon the concepts of ambidexterity and coopetition. Then, exploration and exploitation, two core concepts of AD, will be defined and linked to the concept of coopetition. At last, a framework is visualized with working propositions which have a potential relationship with the nature of coopetitive agreements. This framework will be divided in strategic, operational and environmental factors.
Ambidexterity
The word ‘ambidexterity’ is constructed from two Latin words, namely ambos (both) and dexter (right), which together means right on both sides (Simsek, 2009). Although the term was first coined by Duncan (1976), since March’s (1991) pioneering article on AD researchers have started picking up on this topic and literature has grown exponentially. Following Gibson and Birkinshaw (2004), the concept of AD has become nearly synonymous for simultaneously doing two different or opposing things within an organization. It refers to exploring and exploiting, new and old, short-term and long-term, stability and flexibility, radical innovation and incremental innovation, and many more.
March (1991) believed that companies should balance their attention and resources between both exploitation of existing assets as well as exploration of new activities, techniques and markets. Nevertheless, exploration and exploitation are two independent concepts with diverging priorities, requirements and outcomes (He & Wong, 2004; March, 1991), influencing among others the motives, organizational routines and time orientation. Exploration involves
9 higher uncertainty and flexibility by prioritizing activities such as searching, innovation and experimentation. Exploitation involves lower risks by focusing on the firm’s existing capabilities, in which refinement and optimization play a major role (March, 1991; Turner et al., 2013). Or, to frame it differently, AD is concerned with finding the right balance between ensuring current successes while simultaneously ensuring the company’s future opportunities (Levinthal & March, 1993; March, 1991; Raisch et al., 2009). This balance, or being ambidextrous, creates synergies for companies which stimulate firm performance (Gibson & Birkinshaw, 2004; He & Wong, 2004; Tushman & O'Reilly, 1996).
Yet, being ambidextrous is easier said than done since both types require particular structures, are firm-specific and dependent upon conditions (He & Wong, 2004; March, 1991). Naturally, managers prefer exploitation given its short-term prospects and high degree of certainty. However, reality demonstrates that overreliance on exploitation could result in a so-called competency trap in which organizations can insufficiently cope with future dynamics and developments (Raisch & Birkinshaw, 2008). Intuitively AD feels opposing in nature and seemingly conflicting, but literature in this field shows us the possibilities and the accompanied benefits when executed well.
Hand in hand with the proliferation of articles on this topic, the concept has been studied from an increasing variety of research areas and literature streams (Simsek, 2009). These include perspectives from the field of innovation and technology management (Benner & Tushman, 2003; He & Wong, 2004; Tushman & O'Reilly, 1996), dynamic capabilities (Jansen, Tempelaar, van den Bosch, & Volberda, 2009), organizational learning (Auh & Menguc, 2005; Levinthal & March, 1993), organizational adaptation (Jansen, van den Bosch, & Volberda, 2005; Probst & Raisch, 2005), strategic management (Jansen, George, van den Bosch, & Volberda, 2008), and organization design (Duncan, 1976; Gibson & Birkinshaw, 2004; Jansen et al., 2005; Tushman & O’Reilly, 1996). As the subject of AD matured over the years, different
10 authors conducted comprehensive reviews or longitudinal analyses showing a variety of perceptions and multi-level understanding on this topic (O'Reilly & Tushman, 2013; Raisch & Birkinshaw, 2008; Turner et al., 2013).
Since March’s (1991) seminal article, literature on how AD can be realized within organizations has evolved. Over the years, three main types have been established, namely sequential AD, structural AD and contextual AD. At first, following Duncan’s (1976) original idea, companies were only able to allocate their attention to one type of goal or set of activities and so achieve AD in a sequential order. Over time firms were required to sequentially shift their own structures and mechanisms from exploitative (efficiency) to explorative (innovation) and back to exploitative activities (Gupta, Smith, & Shalley, 2006; O'Reilly & Tushman, 2013).
Nevertheless, sequential AD was deemed inefficient and firms were probing to conduct both types of activities simultaneously. Structural AD was introduced by Tushman and O’Reilly (1996, p. 24) as “the ability to simultaneously pursue both incremental and discontinuous innovation and change results from hosting multiple contradictory structures, processes, and cultures within the same firm.” By setting up two autonomous subunits with their own structures, processes and personnel, firms are able to disperse exploitative and explorative activities from each other and become ambidextrous (Benner & Tushman, 2003; Tushman & O'Reilly, 1996).
Whereas sequential and structural AD focused on the organizational structure and capabilities, contextual AD focuses on the individual level. In addition, whereas the first two types support the division of exploitative and explorative units, contextual AD proposes the integration of both (Gupta et al., 2006; Turner et al., 2013). Some observed this as the differentiated approach versus the integrated approach towards AD (Raisch et al., 2009). Gibson and Birkinshaw (2004) introduced contextual AD by arguing that firms are not only able to achieve AD sequentially (sequential AD) or simultaneously (structural AD), but also
11 within an individual business unit. Firms should facilitate a supportive organizational environment, which “encourages individuals to make their own judgments as to how to best divide their time between the conflict demands for alignment and adaptability” (Gibson & Birkinshaw, 2004, p. 211).
Trust, discipline and stretch are key and individuals should be stimulated by leadership, who need to find a balance between hard (discipline and stretch) and soft management (trust and support) elements. Individuals and firms become ambidextrous by portraying that their department is aligned and adaptable, implying there doesn’t have to be a trade-off between both (Gibson & Birkinshaw, 2004; Turner et al., 2013). This reliance on individual perceptions makes it both easier and more difficult depending on the context, the level of involvement and the quality of management in providing the right organizational context. Gibson and Birkinshaw (2004) analysed 41 different business units and found a positive relationship between contextual AD and financial performance within those business units.
Nevertheless, although each type is defined separately and laid out as a transformation over the years, all three are as relevant nowadays as they were in the past. Although organizational characteristics, such as the size of the company, influence the preferred type of AD, especially larger organizations use all three types interchangeably to balance exploitation and exploration over time (Raisch, 2013; Turner et al., 2013). Another observed trend is emerging firms that often start out with structural AD, but once growth catches up implement other, more integrated, forms of AD within their company. As with exploration and exploitation itself, these different types of AD should be seen as complementary rather than competing depending on the size, scope and stage of the firm (Benner & Tushman, 2003; O'Reilly & Tushman, 2013; Vahlne & Jonsson, 2017).
A broad stream of empirical research has addressed the relationship between AD and firm performance. Whether it is the relationship between AD and sales growth (Auh & Menguc,
12 2005), innovation (He & Wong, 2004; Jansen et al., 2009) or relative performance (Floyd & Lane, 2000; Gibson & Birkinshaw, 2004), the large majority of research available shows a positive link between AD and firm performance in general. Nevertheless, few studies have found no, or even a negative, relationship between AD and performance under certain circumstances (Ebben & Johnson, 2005). Over the years, research in this field has matured, consisting of both in-depth qualitative and quantitative studies and including longitudinal analyses. Furthermore, research has been conducted on different levels, including the organizational, business-unit and individual level of analysis. Overall, different types of research on multiple levels of analyses show a solid positive relationship for AD and firm performance implying that a high degree of AD leads to higher firm performance in general (Gibson & Birkinshaw, 2004; He & Wong, 2004; O'Reilly & Tushman, 2013; Tushman & O'Reilly, 1996).
Despite the maturation of research on this topic, international AD, which studies the balance between exploration and exploitation in the internationalization processes of firms, remain to be underexplored (Prange & Verdier, 2011; Raisch & Birkinshaw, 2008). Smartly borrowed from March’s seminal article, international AD reviews AD across borders involving the balance between exploring new areas and capabilities while holding on to a firm’s initial exploitative assets and capabilities (Hsu et al., 2012; Luo & Rui, 2009; Prange & Verdier, 2011). So far, limited information is available concerning the financial performance implications for firms (Hsu et al., 2012) and further research needs to be conducted upon the impact of the international context on AD (Raisch & Birkinshaw, 2008; Zhou, Lu, & Chang, 2016).
In the field of international AD, an important contribution was made by Luo and Rui (2009), who conceptualized AD as a multi-dimensional concept encompassing four dimensions, namely co-evolution, co-competence, co-orientation and co-opetition. Although
13 they specifically applied these dimensions on the internationalization process of emerging market multinationals (EMNEs), their contribution should be seen in a broader light and deserves further investigation (Hsu et al., 2012). According to Luo and Rui (2009), EMNEs’ successful global expansion can be explained by their successful ambidextrous approach in which they carefully balance exploration and exploitation in their operations. Due to their “institutionally difficult environments both at home and abroad,” EMNEs are incentivized to adopt features of the four dimensions of AD already, including a learning mentality, flexibility, business relationships and both a short and long-term focus (Luo & Rui, 2009, p. 51).
One of their dimensions is coopetition, which refers to the simultaneous cooperation and competition between the focal firm and one or more partners. Although AD is predominantly focused on intra-organizational structures and matters, exploration and exploitation can be achieved through inter-organizational partnerships as well (Kauppila, 2010). He describes how a firm’s AD constitutes of its internal balancing processes which then should be supported, complemented and augmented by external partnerships both at home and abroad. An important upcoming stream in alliance and AD-literature is that of coopetition. This next section will cover literature on coopetitive agreements, which can have considerable impact on a firm’s ambidextrous performance (Kauppila, 2010; Luo & Rui, 2009).
Coopetition
Since the turn of the century, the unorthodox partnership of coopetition, in which a firm simultaneously cooperates and competes with a competitor, has received more attention. Whereas in AD the initial trade-off constitutes of exploration versus exploitation, in coopetition the opposing forces of cooperation and competition are at the centre. Especially in the global context, coopetition is becoming more popular and academic research is rapidly increasing on this topic (Bouncken et al., 2015; Dorn et al., 2016). Although several arguments play a role, the most important reason for companies to consider coopetition is the potential positive-sum
14 outcome of coopetitive agreements (Ritala & Hurmelinna-Laukkanen, 2009). This positive-sum results from cooperation, which aims to enlarge the size of the business pie, while simultaneously and subsequently companies compete to acquire the largest possible share of this enlarged pie in the long-term (Brandenburger & Nalebuff, 1996).
Until the 2000s, research has focused on either cooperation or competition, without considering the possibility of both within an alliance. Nevertheless, the concepts of competition and coopetition were unable to describe the full-spectrum of inter-firm relationships present in the market (Bengtsson, Eriksson, & Wincent, 2010; Padula & Dagnino, 2007). Whereas competition focused more on relative gains while refusing to consider potential interdependencies, cooperation concentrated on the common benefits while underestimating the competitive forces at work (Bouncken et al., 2015; Padula & Dagnino, 2007). With the introduction of coopetition, a third stream emerged in current literature, which combined the advantages of both competition and coopetition while averting the disadvantages (Ritala, 2012).
Although the term was used before, Brandenburger and Nalebuff (1996) utilized it for a broader public in their book Co-opetition. Over the years, coopetition has been studied from a variety of theoretical perspectives, including transaction costs economics (Ritala & Hurmelinna-Laukkanen, 2009), resource-based view (Ritala & Sainio, 2013), innovation (Huang & Yu, 2011) and internationalization (Hsu et al., 2012; Luo, 2007; Luo & Rui, 2009). Furthermore, it has been studied from different levels of analysis including the individual, intra-firm, inter-firm and network-level (Dorn et al., 2016). Although coopetition has been particularly popular among EMNEs, with research following this direction, it has become increasingly common among regular firms as well (Hubbard, 2013).
Contingent upon a firm’s strategy, coopetition is possible at all levels of the organization, along the value chain and both at home and abroad. Generally, coopetitive agreements are between two rivals, but it can consist of multiple rivals as well. Also, although
15 it is a bit harder to identify, coopetition can take place between the focal firm in question and the home or host government (Luo & Rui, 2009). To put it simply, coopetitive agreements are basically strategic alliances between competitors. As the literature on coopetition and strategic alliances has considerable overlap, it is worth touching upon this stream. Especially, since literature on strategic alliances is more established and mature.
Although varied definitions of strategic alliances exist, broadly defined it includes all sorts of cooperative relationships between two or more firms to create or improve their competitive position (Ireland, Hitt, & Vaidyanath, 2002; Vapola, Paukku, & Gabrielsson, 2010). Generally, two different motives could lead to a cooperation between two firms. First, both parties have matching resources or capabilities. Second, there is a complementary strategic fit in which companies can benefit from each other (Nielsen & Gudergan, 2012). In both situations, the common outcome is potentially greater than the individual outcome stimulating both parties to enter a strategic alliance.
These two motives derive from different starting points. Whereas first-mentioned leans on similarities, the second motive triggers companies precisely because of those differences in assets between both parties. This is no different from coopetitive agreements, which can be carried by matching resources as well as complementary resources and capabilities. Next to the initial motives, particular drivers activate companies to undertake action. Research has identified four main drivers for coopetitive agreements. First, firms aim to expand current markets. Second, firms aim to create new markets, either geographically or via differentiation of products. Third, to increase efficiency in production and resource utilization. Fourth, to strengthen its position in the market and vis-à-vis other competitors (Ritala & Hurmelinna-Laukkanen, 2009; Ritala, Golnam, & Wegmann, 2014).
The concept of coopetition advanced out of the literature on AD. Thence, the identified drivers for coopetitive agreements contain both explorative and exploitative elements. Whereas
16 increasing efficiency in production is considered exploitative, entering new geographical markets or differentiation in technology is predominantly explorative in nature. Different drivers determine the nature of the agreement, which is important to establish since exploration and exploitation aim for profoundly different outcomes. Each type requires different types of structure, strategies, capabilities and knowledge (He & Wong, 2004; March, 1991). In order to successfully conduct research towards the ambidextrous nature of coopetition, it is important to carefully lay down the differences between explorative and exploitative activities.
Exploration
As mentioned before, exploration emphasizes new possibilities and centres on the idea of (radical) innovation and experimentation. In general, exploration requires higher flexibility and its outcomes are more uncertain than with exploitation (March, 1991; Turner et al., 2013). This requires a long-term mind-set in which companies are willing to suffer short-term losses. Even more since explorative agreements generally pursue a more disruptive form of innovation by adding new activities, resources or capabilities to the company’s existing set. These type of agreements often seek for or are applied in changing environments, whether economically, technologically or geographically (He & Wong, 2004; Jansen, van den Bosch, & Volberda, 2006; Lumpkin & Dess, 1996; March, 1991). Following these dynamic environments, explorative agreements generally have a more organic structure, which is looser controlled.
In short, explorative activities can be defined along three dimensions, which are 1) a focus on new customers or markets, 2) the requirement of new knowledge, and 3) a non-routine approach to innovation or business (Benner & Tushman, 2003; Jansen et al., 2006; March, 1991). These three dimensions can be typified into characteristics of explorative agreements. First, the development of new products requires companies to experiment and to improvise, resulting in higher forms of uncertainty. Also, more radical forms of innovation are pursued. Second, transparent exchange of information is important between parties since explorative
17 agreements aim to discover new markets and target new segments of people. Third, a non-routine approach is required for non-non-routine solutions. This requires flexibility in structure and urges companies to come up with new forms of cooperation.
Exploitation
Exploitation focuses on improving existing resources and competences, or incremental innovation, in which efficiency and synergies play a centre role. Exploitative agreements are generally less risky and tend to focus on existing technologies and production in familiar geographical areas (March, 1991; Turner et al., 2013). Consequently, this type of agreements leans on stability, with arrangements therefore more set-in-stone. A clear task division between the parties involved characterizes these agreements as well as a certain form of control and bureaucracy (Dorn et al., 2016; He & Wong, 2004; Jansen et al., 2006). For both parties, creating synergies or increasing efficiency is often the major goal of entering into an exploitative coopetitive agreement. Motives for exploitative agreements can therefore be cost reduction, securing production capacity and economies of scale. Common methods to achieve this are the merging or alignment of resources, capabilities or activities.
In short, exploitative activities can be defined along three dimensions, which are 1) to target existing customers or markets, 2) to focus on improving present knowledge and skills, and 3) the pursuit of efficiency or synergies (Benner & Tushman, 2003; Jansen et al., 2006; March, 1991). It is important to typify these dimensions into characteristics of exploitative agreements. First, since exploitative agreements focus on existing customers and markets, it requires both parties to be willing to share certain sets of resources and capabilities. Second, the focus on existing capabilities implies limited adjustments to the production process and products itself. Exploitative agreements generally pursue incremental innovation focusing on efficiency in the process. Third, focus on existing capabilities in known markets limits uncertainty about the outcome of the agreements and the scope of the agreement. This allows
18 the parties involved to make longer-term contracts in which the separation of tasks is clearly set out. In addition, high forms of commitment and integration increase the need for a separate entity including a more centralized governance structure (Anslinger & Jenk, 2004).
Framework
Whereas traditional studies on AD focus on the interplay between exploration, exploitation and its potential synergies, this research analyses specific coopetitive agreements in isolation. This approach is not unique within AD-research (Koza & Lewin, 1998; Rothaermel et al., 2001). For selected agreements, their nature will be determined as either predominantly explorative or exploitative. Consequently, this labelling permits this research to cluster selected agreements as explorative or exploitative in nature.
The working propositions test how selected strategic, operational and environmental factors relate to the nature of selected agreements. In short, proposition 1 and 2 focus on the strategic aspects by investigating the relationship between the chosen entry-mode, a company’s set of resource and capabilities, and the nature of the agreement. Proposition 3, 4 and 5 focus on the operational side of agreements. Proposition 3 studies the difference between upstream and downstream coopetition, proposition 4 reviews the influence of prior experience of companies’ future behaviour and proposition 5 evaluates the relationship between the number of parties involved, the type of innovation and the nature of the agreement. At last, proposition 6 and 7 take environmental dynamics into account. Whereas proposition 6 relates the degree of competition to explorative and exploitative tendencies, proposition 7 reviews the influence of CD on the selected agreements.
To build upon the literature review, a framework has been established. This framework, presented in Figure 1, relates the different factors with the nature of coopetition. Furthermore,
19 it provides a visual overview of what is discussed in this paper. The following section will outline the working propositions.
Working propositions
The literature review is the point of departure for the development of the working propositions in this section, which assess how these factors potentially relate to the nature of the agreement. Subsequently strategic, operational and environmental factors will be discussed.
Strategic
Entry mode plays an important role in discussions regarding cooperation and internationalization. The chosen entry mode has a large impact on the degree of control, the commitment of resources, risk and the future developments of the agreement between two parties. Current literature states a clear distinction between high-equity and low-equity entry modes, or high-equity and low-equity alliances. High-equity alliances imply high-equity investment by both parties and focus on combining capabilities. This category includes joint Figure 1 Framework (author)
Environmental factors
Nature of coopetitive agreement
Explorative Strategic factors Operational factors Explorative
20 ventures, acquisitions, Greenfield and equity strategic alliances (Brouthers & Hennart, 2007; Kedia, Rhew, Gaffney, & Clampit, 2015). In the end, there is no one best way for coopetition, whether it is a handshake during dinner or a full-equity investment, it largely depends upon the conditions surrounding the agreement (Witzel, 2006).
Lower or non-equity alliances are often contractual relationships, or in some cases even without a written agreement, and focus on sharing existing capabilities. They are more loosely controlled and include strategic partnerships, licensing, R&D contracts, and distribution sharing and association agreements (Brouthers & Hennart, 2007; Luo, 2007). Exports are excluded for this category since it is rather difficult to frame this as a coopetitive agreement between two parties. This paper expects equity-alliances to have an exploitative nature and non-equity-alliances to have an explorative nature.
Exploitative agreements focus on existing capabilities and knowledge in known markets in which the risk of investment is relatively low. In this, both parties agreed upon a shared goal often related to cost-efficiency or synergies. Simultaneously, exploitative agreements tend to be formalized in longer-term contracts in which stability, trust and continuation plays an important role (Dorn et al., 2016, March, 1991). Stability, low risk and long-term commitment stimulate equity alliances precisely because those alliances enhance stability and commitment and reduce risk. Both parties are more willing to make large initial investments, which is a requirement for a high-equity alliance, such as a joint venture.
On the other side, low or non-equity alliances are looser forms of partnerships. Explorative agreements deal with a relatively high degree of uncertainty since they focus on pursuing something new in changing environments. Flexibility is something that parties then highly value. Low or non-equity alliances allow for cooperation between competitors while simultaneously ensuring a certain degree of flexibility, mobility and separateness for both parties. In addition, it eases the pressure on a so-called ‘equity hostage,’ which implies the costs
21 associated with terminating an agreement that contains high equity involvement (Luo, 2007). Building on this line of reasoning, this paper proposes the following:
Proposition 1: Exploitative coopetitive agreements are more likely to be formalized in
equity-based alliances, whereas explorative coopetitive agreements are more likely to be formalized in non-equity-based alliances
This section studies the influence of the sets of resources and capabilities (sets) both parties bring to the table. Primarily this is determined by whether parties are pursuing incremental or radical innovations in their agreement. Incremental innovations, which are characteristic for exploitative agreements, are step-by-step improvements of existing processes. Radical innovations, which characterize explorative agreements, aim for disruption and focus on acquiring new knowledge. Whereas incremental innovations function best in formal and rigid agreements, radical innovations are stimulated by more flexibility in looser agreements (Menguc & Auh, 2010; Ritala & Hurmelinna-Laukkanen, 2009).
When firms possess completely different sets than the focal firm, it often implies that they are active in a different field or stage of the production process. In general, the further firms move away from their current domain, the more eager they are to find a partner with a complementary set rather than a similar set. Partnering up with a firm that possesses different resources through a coopetitive agreement often creates uncertainty. As trust, confidence and control have been identified as the areas that receive most attention in strategic alliance literature, this could partially explain why this uncertainty of diverse sets leads to looser form, or explorative, partnerships (Gomes, Barnes, & Mahmood, 2016). This diversity forces parties to utilize new sources of knowledge, which supports more radical innovations (Ritala & Hurmelinna-Laukkanen, 2009). In addition, this new domain and new partner leads to a trial-and-error period in which uncertainty is relatively high (Padula & Dagnino, 2007). The
22 diversity in sets and fields increases the likelihood of radical innovation. This in combination with a higher form of uncertainty, as well as preference for flexibility, characterize explorative agreements.
On the other side, following resource-based view and transaction cost literature, competence and resource similarity improves productivity and efficiency (Lane & Lubatkin, 1998). Sharing similar competences generates a larger form of trust and strengthens the exploitative fit between potential partners (Nielsen & Gudergan, 2012). Both parties know the rules of the game and usually implement their sets for similar goals. Generally, both parties are also active in similar fields and stages. This shared utilization of their resources and capacities is also their competitive advantage. Similarity in resources and capabilities stimulate efficiency and merging of activities, rather than disruptive innovation (Dorn et al., 2016; Ritala & Hurmelinna-Laukkanen, 2009). This, in combination with the preference for more formal and rigid agreements, characterize exploitative agreements. Building on this line of reasoning, this paper proposes the following:
Proposition 2: Exploitative coopetitive agreements will more often involve parties contributing
similar set of resource and capabilities, whereas explorative coopetitive agreements will more often involve parties contributing a diverse set of resources and capabilities
Operations
Cooperation between competitors takes place in different fields and regions, but also along different parts of the value chain. Upstream activities refer to the beginning of the value chain including finding and extracting raw materials, R&D and innovation efforts. Downstream activities capture the large part of value creation process by turning the extracted materials from upstream activities into finished products and include among others production, distribution and marketing (Lavie & Rosenkopf, 2006). Coopetitive agreements can be characterized as cooperating in one stream of activities, while competing in the other stream of activities. For
23 example, the mentioned coopetitive agreement between Samsung and Sony comprised intensive upstream cooperation in innovating LCD-technology, while simultaneously competing in the downstream TV-market (Gnyawali & Park, 2011). In this regard, the fair distinction between knowledge generation (upstream) and knowledge application (downstream) gives an indication for the nature of the agreement (Grant & Baden-Fuller, 2004). Exploitative agreements focus on existing technologies and existing knowledge. Efficiency is key rather than innovation. In general, basic downstream activities such as production and distribution are part of the core of a company and rely on existing technologies. The production process itself relies on its current capabilities and pursues efficiency in this process. Furthermore, downstream activities include marketing and commercialization processes as well. Although creativity plays an important role in these processes, in the end, these processes mainly leverage the partners’ existing capabilities (Lavie & Rosenkopf, 2006; Leung, Lau, Zhang, & Gu, 2015; Rothaermel et al., 2001). Since downstream activities primarily rely on existing capabilities, coopetition in this area is more likely to be of an exploitative nature.
Innovation and pursuit of new forms of knowledge, capabilities or markets are at the core of explorative agreements and generally result in higher forms of uncertainty about the outcome. Upstream activities, such as R&D and product innovation, require both parties to move away from its core business to pursue new capabilities and technologies. To cope with this uncertainty, companies need to share tacit knowledge for the benefit of product innovation and be willing to sacrifice short-term losses for (uncertain) long-term benefits. Since upstream activities primarily focus on acquiring new forms of knowledge and capabilities and imply a higher degree of uncertainty, coopetition in this area is more likely to be of an explorative nature (Lavie & Rosenkopf, 2006; Leung et al., 2015; Rothaermel et al., 2001). Building on this line of reasoning, this paper proposes the following:
24
Proposition 3: Exploitative coopetitive agreements are more likely in the downstream part of
the value chain, whereas explorative coopetitive agreements are more likely in the upstream part of the value chain
Following the influential work of Nelson and Winter (1982), firms build up routines and conduct behaviour, which is based on their past experience. Or in other words, prior experience plays an important role in decisions that companies make for the future. These experiences include among others the degree of cooperation, the outcome of the agreement and the manager’s perception. Firms slightly prefer to conduct future coopetition in a similar fashion to their prior experience. Or, firms that were involved in upstream alliances in the past are more likely to stick to upstream cooperation, and vice versa. Lavie and Rosenkopf (2006) referred to this behaviour based on experience as self-reinforcing and path-dependent.
Nonetheless, larger companies that repeatedly engage in exploitative alliances in one domain try to balance this by engaging in explorative alliances in other domains. In particular the more experienced companies try to balance their exploitative and explorative activities across domains and over time (Lavie & Rosenkopf, 2006). In addition, Schiavone and Simoni (2011) argue that more experienced firms add to the quality of the agreement and should pick a coordinating or leading role. Given the importance of experience and the aim of larger companies to balance across domains and over time, this paper proposes the following:
Proposition 4: The more experienced the focal firm is in coopetition, the more open it is to
conduct both explorative and exploitative agreements
The number of parties involved in the agreement is important for cooperation. Luo (2007) argues for a differentiation between two forms of coopetition, namely a two-party form and a multiparty. Other authors differentiate between dyadic (two-party), triadic (three parties) or
25 multiple coopetition (four or more parties) (Yami & Nemeh, 2014). Since only few authors include triadic coopetition, this research uses two different forms, which are two-party and multiparty coopetition. Nevertheless, research on this topic in this field is limited so far and predominantly based upon alliance literature. In general, it is assumed that multiparty alliances are more complex to coordinate due to an increased number of voices, structures (Dorn et al., 2016; Heidl & Phelps, 2010) and diverging interests (Dagnino & Rocco, 2009). This is confirmed by Ritala and Hurmelinna-Laukkanen (2009) who re-emphasized the importance of trust in coopetitive partnerships.
However, Yami and Nemeh (2014) have conducted research into the difference in goals between those two forms of coopetition in terms of innovation. On the basis of five case studies in the telecommunications industry, they observed a relationship between the number of partners and the innovation goal. Whereas parties pursuing incremental innovation tend to prefer two-party alliances, pursuing radical innovation often occurs in multiparty alliances. Although studied from a different point of view, this new potential connection in the field of coopetition is worth reviewing. Through an ambidextrous lens, radical innovation would be linked to explorative agreements, potentially creating a relationship between multiparty alliances and explorative agreements. On the other hand, incremental innovation is an important characteristic of exploitation and therefore linked to two-party agreements. Building on this line of reasoning, this paper proposes the following:
Proposition 5: Two-party agreements are more likely to be exploitative in nature, whereas
multiparty agreements are more likely to be explorative in nature.
Environment
Firms’ behaviour is largely influenced by the natural environment in which it operates. Simultaneously, it creates tension between exploration and exploitation and choices have to be made, which the company deems more important at that time (Jansen et al., 2005).
26 Generally, rising competition puts pressure on prices and margins (Jansen et al., 2006; Luo, 2007). This is particularly applicable to listed companies, which are naturally more driven by short-termism and the “ninety-day cycle,” due to the urge for constant performance by their shareholders. A highly competitive environment could move a company to (temporarily) adapt a more exploitative direction to increase its efficiency and effectiveness (Matusik & Hill, 1998). Generally, outcomes of exploitative activities are more certain, providing those companies with steady cash flows, whereas explorative investments tend to be more insecure and focused on the long-term. In addition, the more competitive the environment, the more difficult it gets for companies to pursue longer-term goals at the expense of short-term benefits (Auh & Menguc, 2005; Jansen et al., 2006).
However, if the level of competition in the host country is lower, it allows companies to invest in new technologies and capabilities. In line with traditional business literature, lower competition reduces external pressures from shareholders to gain short-term benefits and increases opportunities to take risks. Although a combination of both exploration and exploitation is preferred in highly competitive contexts (Jansen et al., 2005) and it is dependent upon firm- and environment-specific conditions (Jansen et al., 2006), for this paper the following proposition is developed:
Proposition 6: The higher the degree of competition in the market, the more likely the
agreements is of an exploitative nature, whereas the lower the degree of competition in the market, the more likely the agreements is of an explorative nature
In their research on cultural distance (CD), Kogut and Singh (1988) stated that differences in national culture result in different organizational practices and characteristics between companies. In line with this, Zeng (2003) stated that cultural differences between both parties have a strong influence on the degree and success of cooperation. Cultural similarity between
27 both parties involved in an agreement, measured as the CD between the companies’ countries of origin, is an important factor for trust. In general, exploitative agreements require a higher form of commitment and investment, or a so-called “higher degree of cooperation.” In line with this reasoning, one would expect a lower CD in exploitative agreements than in explorative agreements.
Cultural similarity has a positive effect on the level of trust and therefore benefits the degree of commitment. It implies that both companies adhere to similar rules or at least are aware of the rules of the game. This initial trust creates certainty which is of upmost importance for exploitative agreements. Furthermore, research shows that similarity in terms of culture and practices stimulates parties to leverage existing knowledge and competences with the goal of incremental innovation (Popadić & Černe, 2016).
A high CD between both parties increases the level of uncertainty since many cultural differences are unknown to both parties. Consequently, companies initially might opt for more flexibility and lower commitment through non-equity alliances (Kogut & Singh, 1988; Majocchi, Mayrhofer, & Camps, 2012). This is characteristic for explorative agreements. Also, exploration in geographical terms often implies high CD, because firms tend to expand to culturally close countries first. In addition, Popadić and Černe (2016) argue that cooperation between companies from geographically and culturally diverse locations stimulate explorative innovation. Disruptive innovation requires creativity which can be amplified by combining different cultures and practices (Nielsen & Gudergan, 2012). Building on this line of reasoning, this paper proposes the following:
Proposition 7: The higher the cultural distance between the parties involved, the more likely
the coopetitive agreement will be of an explorative nature, whereas the lower the cultural distance, the more likely the coopetitive agreement will be of an exploitative nature
28 The potential relationship between the seven propositions, coopetition and the nature of coopetitive agreements is visualized in Figure 1 in the previous section. The validity of the propositions is reviewed in the results section.
Methodology
This chapter discusses the methodology of this research. First, it will shortly describe and recap the purpose of this paper. Then, the methodological choices will be discussed, including subsequently the case-study, sample and data-collection. Next, the data analysis method is discussed. To conclude, a descriptive overview of the selected cases will be provided.
This paper integrates the literature on AD with the concept of coopetition in a cross-national setting. By taking on an ambidextrous perspective, this paper analyses how the nature of explorative coopetitive agreements and exploitative agreements differentiate from each other. Through an in-depth multiple case analysis on the nature of nine coopetitive agreements it is determined for each case whether it is predominantly explorative or exploitative in nature. Furthermore, individual cases will be analysed on the basis of the established dimensions. These dimensions, or propositions can be categorized into strategic, operational and environmental factors and are tested upon their relationship with the nature of the agreement. Subsequently, a cross-case and cross-cluster analysis focuses on the similarities and differences between the two clusters of explorative and exploitative agreements. This article further builds upon existing theory and serves as a window to further understand the nature of different coopetitive agreements and its contextual factors.
Case study
To uncover the nature of the selected agreements, it is important to consider the contextual conditions in detail. This requires an in-depth examination of coopetitive cases for which qualitative research is deemed more appropriate than survey or experimental research (Yin,
29 2009). It enables researchers to gain better insight into the behaviour of organizations and the interactions with their environment (Velde, Jansen, & Anderson, 2004). Therefore, in combination with an exploratory focus, it allows researchers to uncover the real motives and nature behind the selected coopetitive agreements (Saunders & Lewis, 2009; Yin, 2009). Since this paper focuses on a relatively new phenomenon (coopetition) from a different perspective (AD) with an exploratory focus, an inductive approach was used.
Literature on the nature of coopetitive agreements is relatively sparse, which makes a case-study method most suitable for this research. A multiple case analysis is deemed the most appropriate method given the multi-faceted core coopetition and its evolvement over time (Bengtsson et al., 2010; Gnyawali et al., 2015; Yin, 2009). For this research, nine cases have been selected ensuring to cover sufficient cases without becoming superficial. Eisenhardt (1989) argues that four to ten cases is most effective depending on the topic of the research. Furthermore, this number allows researchers to go into depth and permits cross-case comparison across settings as well (Baxter & Jack, 2008).
Sample
Four criteria were defined to select suitable companies for the cases. First, only companies with an annual revenue of over one billion euros and a minimum of 10.000 employees were qualified for this research. This allows companies to conduct both explorative and exploitative coopetitive activities on a larger basis and guarantees sufficient media attention and information available. Second, each company needed to have significant experience with coopetition. In this paper, each company concluded more than eight coopetitive agreements and was considerably transparent in its reporting via press statements, annual reports and their website. Third, only focal companies with their headquarters (partly) in the Netherlands were qualified to ensure broad access to information both in the mother language (Dutch) and in English. Four, all firms needed to cover different industries to avoid specific industry particularities.
30 The following five companies were selected: Heineken, KLM, Philips, Shell and Unilever. Each company is represented by two cases except for KLM since it predominantly focuses upon exploitative agreements. Each selected case is a coopetitive agreement between a Dutch MNE and a global rival. A cross-national agreement involves two or more companies, in which at least one partner is headquartered outside of the country of operations (Nielsen & Gudergan, 2012). To increase potential generalization of the outcome, a broad sectoral focus over a broad time span of twenty years (1996-2016) is ensured to control for variations. By selecting large MNEs with considerable international experience over a broad time span, this research aimed to pick cases carrying potential to be replicated or extended across this emerging theory (Eisenhardt, 1989).
Data
Through a qualitative and exploratory approach, this paper relies primarily on secondary data. This collection method refers to data that is publicly available and which can be used for analysis (Velde et al., 2004). Advantageous is that these published sources are most probably more objective than interviews. Published sources permit cross-checking of information across different sources and largely avoid subjectivism and one-sided stories which are inherent to interviews. By investigating cases from independent media and both the perspective of the focal firm as well as that of partners to the agreement, this research intents to increase the validity of the data analysis.
Secondary data is gathered through documentary research of corporate websites, annual reports, project reports, NGO reports, journals and newspaper articles. As additional information, relevant published interviews with relevant employees of the companies involved will be used as primary data. Furthermore, the Thomson ONE database (formerly known as SDC Platinum) has been used to gather information about deals, company and equity data.
31
Data analysis
Following the how-question posed in this research, both a within-case and a cross-case analysis were conducted. The within-case analysis was a logical consequence of the case-study research, given the large amount of data generated when reviewing data about individual cases. It facilitates researchers to cope with the avalanche of information. The within-case analysis consists primarily of descriptive write-ups of the individual cases focusing on the most relevant information available (Eisenhardt, 1989). Following this, two clusters were created, an explorative cluster and an exploitative cluster.
Eisenhardt (1989) described potential dangers of this type of analysis by jumping to conclusions too fast. She denoted the tendency of people to rely on their own intuition, to allow bias slipping in or to ignore statistical properties, which are important issues to be aware of. To counter these potential dangers, it is important to establish themes or dimensions and cluster the available information. In addition, thematic analysis through these dimensions helps to understand potential patterns in the large amounts of information available. These dimensions are, on its turn, based on the working propositions that were developed in the theoretical background (Eisenhardt, 1989; Ryan & Bernard, 2003). These propositions defined the cadres within which this research searched for patterns in the data available.
After theoretical review and thematic analysis, the following seven dimensions were established: “entry mode,” “set of resource and capabilities,” “part of value chain,” “prior experience,” “number of parties,” “degree of competition,” and “cultural distance.” These dimensions cluster the data and give a clear direction and structure. In order to review these dimensions, different measurements were established to enable objective analysis.
32
Description of dimensions
This section gives an overview of the different dimensions, their main sources and the different characteristics for both exploration and exploitation. These are described in table 1. Most dimensions and characteristics are explained in the proposition section already and do not require additional explanation. Two dimensions require further justification.
Prior experience
Unless the other six dimensions, prior experience is applicable to both explorative and exploitative agreements. To classify prior experience, three levels were established from 0 to 2. A score of 0 implies limited experience with coopetition in general. A score of 1 implies reasonable experience with coopetitive partnerships in general or extensive experience limited to one cluster (explorative or exploitative). A score of 2 implies high experience with coopetition, both in the explorative and the exploitative field.
Cultural distance
To study cultural differences among companies, this research reviews the cultural context of the company’s country of origin. Geert Hofstede (1980) developed a set of variables classifying the cultural settings in a particular country. With a focus on work-related norms and values, he differentiated between “power distance,” “uncertainty avoidance,” “individualism vs collectivism,” and “masculinity vs femininity.” Hofstede’s seminal work formed the basis for research in this field and different models, such as the GLOBE-index and the Schwartz’ survey of values. Although criticism on Hofstede’s dimensions is widely prevalent, it is still the most applied method for measuring CD. In addition, it allows researchers to easily classify, quantify and compare countries with each other (Drogendijk & Slangen, 2006; Kirkman, Lowe, & Gibson, 2006). Here for, this research uses Hofstede’s dimensions and relies on the Kogut and Singh-index (K&S-index) to calculate CD across countries (Kogut & Singh, 1988).
33 CDi stands for the CD between the home country (the Netherlands) and the selected
country. Iij is the index of the selected country’s four types of cultural dimensions, whereas Iih
is the index of the home country. Vi is the variance in the index of the selected country’s
dimension. The higher the K&G-index score, the larger the cultural difference between the home country (the Netherlands) and the selected country (Kogut & Singh, 1988). In this case, the higher the K&G-index score, the larger the CD between the Dutch MNE and its partner(s) in the agreement.
Dimensions Exploitation Exploration Main sources
P1. Entry mode High-equity Low-equity (Brouthers & Hennart, 2007; Dorn et al., 2016; Luo, 2007; March, 1991)
P2. Sets of resources and capabilities
Similar sets Diverse sets (Menguc & Auh, 2010; Nielsen & Gudergan, 2012; Padula & Dagnino, 2007; Ritala &
Hurmelinna-Laukkanen, 2009)
P3. Part of value chain
Downstream Upstream (Grant & Baden-Fuller, 2004; Lavie & Rosenkopf, 2006; Leung et al., 2015)
P4. Prior experience
Applicable to both Applicable to both (Lavie & Rosenkopf, 2006; Nelson & Winter, 1982; Schiavone & Simoni, 2011)
P5. Number of parties
Two-party Multiparty (Dagnino & Rocco, 2009 ; Dorn et al., 2016; Luo, 2007; Yami & Nemeh, 2014)
P6. Degree of competition
High Low (Aug & Menguc, 2005; Jansen et al., 2005, 2006; Luo, 2007;
P7. Cultural distance
Lower Higher Hofstede, 1980; Kogut & singh, 1988; Kirkman et al., 2006; Majocchi et al., 2012; Nielsen & Gudergan, 2012; Popadić & Černe, 2016; Zeng, 2003) Table 1: Description of dimensions
Description of cases
Nine selected coopetitive agreements from five Dutch MNEs were analysed. Below follows a short description of each agreement.
