Managing for ambidexterity
The effects of balancing high levels of transformational and transactional
leadership, the organizational context and environmental dynamism
Nadine Vissers 10070494 June 24, 2016 Final draft Msc. in Business Administration – Strategy Track ABS, University of Amsterdam Supervisor: Michiel Tempelaar
Table of contents
Abstract ... 3
Introduction ... 4
Literature review ... 11
Contextual ambidexterity and leadership ... 11
Transformational leadership ... 14
Transactional leadership ... 15
Organizational context ... 16
Environmental dynamism ... 20
Leadership, organizational context and environmental dynamism ... 21
Organizational context, ambidexterity and environmental dynamism... 22
Data and method ... 24
Research setting and data collection ... 24
Measurement and validation of constructs ... 25
Analyses and results ... 29
Robustness ... 33 Discussion ... 38 Discussion of results... 38 Implications ... 43 Further research ... 44 Limitations ... 45 Conclusion ... 49 Reference list ... 51 Appendix ... 59
Abstract
In order to meet the challenges of the increasingly complex competitive environment, organizations these days need to be ambidextrous. To achieve contextual ambidexterity, organizations should be able to manage the contradictory demands of both exploratory and exploitative activities simultaneously at a business unit level. The capacity for contextual ambidexterity emerges from the features of the firm’s organizational context, in which a supportive organizational context for ambidexterity allows for both alignment and adaptability. However, there is a lack of research that examines how contextual ambidexterity is managed exactly and how a supportive organizational context for ambidexterity is created. Since the contradictory demands in shaping a supportive organizational context and creating ambidexterity need to be joined simultaneously, it is essential to approach managing for ambidexterity as a paradox. This enables the combination and integration of the contradictory demands between exploration and exploitation and alignment and adaptability. Therefore, this research examines how balancing high levels of transformational and transactional leadership as paradoxical leadership styles affects the achievement of ambidexterity and the creation of a supportive organizational context. Thereby, the effect of the external environment in the form of environmental dynamism is tested on these relations, because in order to achieve ambidexterity different environments require different demands to be fulfilled. To test the relations between the concepts an online questionnaire was sent out to senior managers of various companies in the Netherlands. The most important findings of this research are that balancing high levels of transformational leadership relates positively to the establishment of ambidexterity and that environmental dynamism moderates the relationship between a supportive organizational context and ambidexterity, resulting in a stronger relationship for higher levels and a weaker relationship for lower levels of environmental dynamism.
Introduction
Organizations these days need to both exploit their existing business and explore new opportunities in order to meet the challenges of the increasingly complex competitive environment (Smith, Binns & Tushman, 2010). Organizational ambidexterity, through which organizations exercise both exploratory and exploitative activities simultaneously, is seen as an effective way to deal with these challenges and will therefore lead to better firm performance (Tushman & O’Reilly, 1996; Gibson & Birkinshaw, 2004; He & Wong, 2004; Lubatkin, Simsek, Ling & Veiga, 2006; O’Reilly & Tushman, 2013). In creating ambidexterity exploration refers to the search for new competences and opportunities and exploitation to the use and further development of current capabilities (March, 1991; Levinthal & March, 1993). The demands for both concepts differ, exploration for instance asks for search, variation and experimentation to create new and divergent recombinations of knowledge, and exploitation demands for efficiency and convergent thinking to use and improve existing capabilities (Andiopoulos & Lewis, 2009). However, exploration and exploitation should be practiced in balance, otherwise organizations end up in the failure trap which eventually does not lead to ambidexterity (Levinthal & March, 1993). When organizations are able to manage the contradictory demands of exploratory and exploitative activities simultaneously at a business unit level, these organizations are contextually ambidextrous (Gibson & Birkinshaw, 2004).
The capacity for contextual ambidexterity emerges from the features of the firms’ organizational context (Gibson & Birkinshaw, 2004). It is already known that a supportive organizational context has a positive and direct effect on the creation of contextual ambidexterity (Gibson & Birkinshaw, 2004). A supportive organizational context consists of sets of systems and processes that allow for both alignment and adaptability (Gibson &
Birkinshaw, 2004). In the creation of an organizational context for alignment and adaptability both must flourish simultaneously in order to establish a supportive organizational context for ambidexterity (Gibson & Birkinshaw, 2004). This means that the organizational context should have the capacity to simultaneously align all the patterns of activities in the business unit and to adapt these activities in the business unit to the demands of the environment (Gibson & Birkinshaw, 2004).
It is the role of managers to create and shape the supportive organizational context for ambidexterity (Ghoshal & Bartlett, 1994; Gibson & Birkinshaw, 2004; Lin & McDonough, 2011). Thereby, in a contextually ambidextrous organization, successful leaders should be able to manage the contradictory demands of both exploratory and exploitative activities simultaneously (Andriopoulos & Lewis, 2009; Raisch, Birkinshaw, Probst & Tushman, 2009). This makes leadership an antecedent of ambidexterity, which in turn is influenced by different behavoirs of the management team (Raisch & Birkinshaw, 2008). Leadership thus affects success in achieving ambidexterity, however how ambidexterity and the organizational context are managed exactly is not clear. Therefore it is important to know how different leadership behaviors relate to managing contradictory demands and eventually to the establishment of a supportive organizational context and the achievement of organizational ambidexterity.
In order to manage the contradictory demands in shaping the organizational context and creating ambidexterity simultaneously, these demands need to be joined. A paradoxical approach acknowledges contradictions and creates an overarching vision to bring competing demands together in order to engage these demands simultaneously (Lewis, Andriopoulos & Smith, 2014). Therefore a paradoxical approach enables the combination and integration of contradictory demands (Poole & van de Ven, 1989; Lewis, 2000). It is thus essential to approach managing for ambidexterity as a paradox, because this enables the combination and
integration of the contradictory demands between exploration and exploitation and alignment and adaptability.
Paradoxical leadership allows managers to address competing, but yet incompatible objectives simultaneously (Lavine, 2014). Managing the contradictory demands of the organizational context and ambidexterity therefore asks for a paradoxical leadership approach (Lavine, 2014; Lewis et al., 2014). Paradoxical leadership should be executed according to a both/and vision rather than an either/or vision (Lavine, 2014; Lewis et al., 2014), which encourages the practice of multiple leadership styles. Transformational and transactional leadership are paradoxical leadership styles (Burns, 1978; Vera & Crossan, 2004). Transformational leadership for instance is people-oriented and participative and transactional leadership is task-oriented and regulative (Vera & Crossan, 2004).
There is however a lack of research that examines how contextual ambidexterity and the organizational context are managed exactly. Additionally, leadership and a supportive organizational context both relate positively to ambidexterity (Gibson & Birkinshaw, 2004; Andriopoulus & Lewis, 2009; Raisch et al., 2009; Smith et al., 2010; Rosing, Frese & Bausch, 2011; Turner et al., 2013). However, the link between specific leadership behaviors and a supportive organizational context has not been studied yet. Prior research on leadership, organizational context and ambidexterity confirms the need for research that examines the effects of and links between both transformational and transactional leadership behaviors on organizational context and ambidexterity (Jansen et al., 2009b). Since the effects of and links between multiple leadership behaviors, such as transformational and transactional leadership, have not been examined yet (Jansen et al., 2008; Jansen et al., 2009b; Nemanich & Vera, 2009). Prior research argues that transformational leadership relates positively to exploration and transactional leadership to exploitation (Jansen, Vera & Crossan, 2009b). Transformational and transactional leadership are thus expected to be complementary in their
relation with managing exploratory and exploitative activities simultaneously, especially when practised at high levels and in balance. Therefore a balance between high levels of transformational and transactional leadership behaviors is likely to create a contextually ambidextrous organization (Tushman & O’Reilly, 1996; Rosing et al., 2011). Balancing high levels of both transformational and transactional leadership is likely to affect the organizational context, because these paradoxical leadership behaviors enable leaders to manage contradictory demands (Lavine, 2014; Lewis et al., 2014), but this has not been researched yet. By establishing and balancing high levels of both transformational and transactional leadership behaviors the organizational context is expected to allow for alignment as well as adaptability and thus enables the achievement of ambidexterity (Jansen et al., 2009b; Rosing et al., 2011).
In addition, the external environment was not taken into account in prior research. In practice however, the environment in which a relation takes place affects the strength of this relationship (Smith et al., 2010). The effectiveness of organizational ambidexterity is affected by contextual conditions (Raisch & Birkinshaw, 2008), thus, in order to achieve ambidexterity different environments require different demands to be fulfilled (Dutta, 2013). Furthermore, managers should create an organizational context that meets the requirements of the external environment (Gibson & Birkinshaw, 2004; Dutta, 2013). So, the external environment determines which leadership behaviors should be practiced in order to create a supportive organizational context in a certain situation. Currently organizations face changes, variations and fluctuations in their external environment. These environments are characterized as dynamic and are continuously changing (Jansen et al., 2006). The level of environmental dynamism a firm faces varies per organization. According to Jansen et al. (2009b) different levels of environmental dynamism ask for different leadership approaches. In highly dynamic environments organizational survival for instance depends on paradoxical
leadership, because by doing so managers are able to meet rising and contradictory demands (Lavine, 2014; Lewis et al., 2014). In addition, a supportive organizational context is more likely to adapt to and align with the changing demands of the environment (Eisenhardt & Martin, 2000; Birkinshaw & Gibson, 2004; Wei & Lau, 2010). Therefore it is likely that the level of environmental dynamism will affect the relation of the establishment of specific leadership behaviors on organizational context as well as the relation of the supportiveness of the organizational context on ambidexterity. Consequently, balancing high levels of transformational and transactional leadership and a supportive organizational context are more effective in highly dynamic environments.
In this research, therefore, the influence of balancing high levels of both transformational and transactional leadership on organizational context and ambidexterity is examined. By doing so, organizational context is tested as a mediator between balancing high levels of transformational and transactional leadership styles and ambidexterity. To take the external context into account, the level of environmental dynamism is tested as a moderator on both relations. The research model can be found in figure 1 and leads to the following research question:
“How does balancing high levels of transformational and transactional leadership influence the organizational context and ambidexterity and how are these relations influenced by environmental dynamism?”
Figure 1: Research model
By extending the framework of organizational context of Gibson and Birkinshaw (2004) through adding the balance between high levels of transformational and transactional leadership as predictor, this research contributes to a richer understanding of the role of leadership in shaping an organizational context that enables firms to balance the conflicting demands of ambidexterity. Besides that, this extension of the framework of organizational context allows to test the ability of combining transformational and transactional leadership to create a supportive organizational context for ambidexterity proposed by Jansen et al. (2009b). In addition, this research provides more insight into how the paradox of ambidexterity is actually managed and how combinations of paradoxical leadership behaviors affect the organizational context and eventually the balance between exploration and exploitation and thus ambidexterity. Despite the fact that positive relations were found between leadership and establishing ambidexterity (Raisch & Birkinshaw, 2008; Andriopoulus & Lewis, 2009; Raisch et al., 2009; Smith et al., 2010; Rosing et al., 2011; Turner, Swart & Maylor, 2013) and a supportive organizational context and ambidexterity
(Gibson & Birkinshaw, 2004), the external environment that is likely to influence these relations was not taken into account. It is for this reason that this research contributes to the contingency perspective as well by taking the effect of environmental context into account through examining the level of environmental dynamism as moderator on the balance of high levels of transformational and transactional leadership, organizational context and ambidexterity. Eventually, this research will bring up suggestions that need to be addressed in future research when it comes to the combination of paradoxical leadership styles, establishing a supportive organizational context and the creation of organizational ambidexterity. Thus, this research contributes to this field by providing a base for further research.
Firstly, in this thesis the existing literature regarding ambidexterity, organizational context, leadership and environmental dynamism is reviewed. Secondly, the hypotheses used to test the research question are explained within the literature review. Thirdly, the data and method of the research are discussed through the research setting and data collection and the validation of the constructs. Fourthly, the results of the questionnaire and the analyses to test the research question are explained. Fifthly, the hypothesis and research question are answered in the discussion section. In the discussion explanations regarding these results are provided as well. Followed by the managerial implications of the research, the directions for future research and the limitations of the research. Lastly, the conclusion including a short summary of the findings and the contributions of the research will complete this thesis.
Literature review
Contextual ambidexterity and leadership
Due to the changing environment organizations must be ambidextrous to remain successful in the long-term, which means organizations must practice both exploratory and exploitative activities simultaneously (Tushman & O’Reilly, 1996). Where exploration refers to exploring new competences and opportunities, exploitation is the use and further development of existing capabilities (March, 1991; Levinthal & March, 1993). Exploration and exploitation should not drive out one or another, otherwise organizations end up in the failure trap which in the end does not lead to ambidexterity (Levinthal & March, 1993). When organizations are able to align current practices and also adapt to the changing demands of the organization’s environment simultaneously at a business unit level these organizations are contextually ambidextrous (Gibson & Birkinshaw, 2004). Contextual ambidexterity arises from the features of its organizational context (Gibson & Birkinshaw, 2004). This means that individuals in contextually ambidextrous organizations have the behavioral capacity to simultaneously align all the patterns of activities in the business unit and to reconfigure activities in the business unit to meet the changing demands in the environment (Gibson & Birkinshaw, 2004). Being ambidextrous as a firm relates positively to performance (Gibson & Birkinshaw, 2004; Lubatkin et al., 2006), especially when exploration and exploitation are in balance and both high (He & Wong, 2004).
One of the main challenges in creating and sustaining an ambidextrous organization is managing the contradictory demands between exploratory and exploitative activities (Andriopoulos & Lewis, 2009; Raisch et al., 2009). Whereas exploration demands for search, variation and experimentation efforts in order to create new, divergent recombinations of knowledge exploitation asks for efficiency and convergent thinking to use current capabilities
and continuously improve product offerings (Andiopoulos & Lewis, 2009). Therefore these two concepts may require different organizational approaches (Raisch & Birkinshaw, 2008). However, in establishing ambidexterity, exploration and exploitation are only effective when their management is combined and integrated (Andiopoulos & Lewis, 2009). Leadership plays a crucial role in managing these contradictions (Smith et al., 2010; Turner et al., 2013). According to Rosing et al. (2011) managers should therefore establish leadership behaviors that are complementary to manage both exploratory and exploitative activities simultaneously. There is however a lack of research that examines how this paradox of ambidexterity is managed, therefore different leadership behaviors of senior executives as antecedents and their effects on ambidexterity should be examined (Gibson & Birkinshaw, 2004; Jansen, van den Bosch & Volberda, 2006; Raisch & Birkinshaw, 2008; Jansen, Tempelaar, van den Bosch & Volberda, 2009a; Lin & McDonough, 2011; O’Reilly & Tushman, 2013; Turner et al., 2013).
The contradictory demands between exploratory and exploitative activities, as explained above, in creating ambidexterity can be described as paradoxical. A paradox consists of contradictory, yet interrelated, concepts that seem logical when isolated but irrational when they appear simultaneously (Lewis, 2000). Thereby, an overemphasis on either one or the other concept of the paradox is dysfunctional (Cameron, 2008). By understanding tensions or oppositions as a paradox, this stimulates the development of more comprehensive theories (Poole & van de Ven, 1989). Additionally, a paradoxical approach acknowledges contradictions and therefore enables to combine and integrate these contradictions (Poole & van de Ven, 1989; Lewis, 2000). It is essential to approach managing for ambidexterity as a paradox, because this enables the combination and integration of the contradictory demands between exploration and exploitation. By combining and integrating the contradictory management demands for ambidexterity organizations develop dynamic
capabilities in order to adapt to and align with the environment (Eisenhardt & Martin, 2000). Thus, the paradoxical nature of the demands of ambidexterity provides the organization with a broad repertoire of solutions for various situations.
The competing demands for ambidexterity pose challenges that require paradoxical leadership practices engaging in these tensions (Lewis et al., 2014). A paradoxical leadership approach creates an overarching vision that brings together competing demands and seeks to engage these demands simultaneously (Lewis et al., 2014). Applied to ambidexterity, a paradoxical approach emphasizes the interdependent nature of exploration and exploitation (Lewis et al., 2014). Through paradoxical leadership managers are able to address competing, yet incompatible objectives (Lavine, 2014). Therefore, when it comes to ambidexterity, a paradoxical view of leadership is needed (Lavine, 2014; Lewis et al., 2014). According to Lavine (2014) and Lewis et al. (2014) paradoxical leadership enables leaders to manage contradictory demands simultaneously when executing leadership according to a both/and vision rather than an either/or vision. Which means managers can, in order to establish ambidexterity, best practice multiple leadership styles. Leadership styles that are typically paradoxical are transformational and transactional leadership (Burns, 1978; Vera & Crossan, 2004). Where transformational leadership for instance is relation-oriented and participative, transactional leadership is task-oriented and directive (Vera & Crossan, 2004).
There are several studies that examine the role of transformational and transactional leadership in relation to ambidexterity. Jansen et al. (2009b) argue that transformational leadership positively relates to exploratory innovation and transactional leadership to exploitative innovation. Jansen, George, van den Bosch and Volberda (2008) found that transformational leadership positively moderates the effect of senior team attributes on organizational ambidexterity. Nemanich and Vera (2009) conclude that transformational leadership directly influences the achievement of ambidexterity too. In addition, leaders do
not necessarily execute one specific leadership style, according to Avolio, Bass and Jung (1999) management is often a combination of transformational and transactional leadership. Prior research revealed that managers can indeed practise both leadership styles in different amounts and intensities (Bass, 1998; Jansen et al., 2009b). However, in what way the combination between transformational and transactional leadership relates to the creation of organizational context and ambidexterity seems unclear at present (Jansen et al., 2009b). Therefore, testing the link between transformational and transactional behaviors and ambidexterity is needed (Jansen et al., 2009b). The properties of both transformational and transactional leadership are described below.
Transformational leadership
Leaders that execute transformational leadership broaden the interests of their employees, generate awareness and acceptance and stir individuals to act for the good of the group instead of their own self-interest (Bass, 1991). Thus, transformational leaders inspire, energize and intellectually stimulate the members of their team (Bass, 1991). This is achieved through four dimensions: idealized influence, inspirational motivation, intellectual stimulation and individualized consideration (Avolio et al., 1999; Bass, 1991). First, through idealized influence leaders provide a vision and sense of mission, instill pride and gain respect and trust of their followers (Bass, 1991). Secondly through inspirational motivation leaders communicate their expectations, focus efforts and express important purposes (Bass, 1991). Through the third dimension, intellectual stimulation leaders promote intelligence, rationality and problem solving (Bass, 1991). Finally, through individualized consideration transformational leaders give personal attention to their group members and provide individual coaching and advice (Bass, 1991).
Transactional leadership
Transactional leadership is based on the transactions between a manager and his/her employees (Bass, 1991). Leaders that execute transactional leadership explain the members of their team what is required and what compensation they will receive when fulfilling these requirements (Bass, 1991). There are three characteristics that identify transactional leadership: contingent reward, active management by exception and passive avoidance leadership (Avolio et al., 1999). Through contingent reward transactional leaders recognize accomplishments, they clarify to their followers what they need to do and what effort they have to make to be rewarded (Bass, 1991; Avolio et al., 1999). Secondly, through active management by exception, managers monitor the performance of their followers for any possible problems and take action to correct these problems to maintain the current performance level (Bass, 1991; Avolio et al., 1999). Thirdly, through passive avoidant leadership transactional leaders only intervene when standards are not met and when problems have become serious to take corrective action, or avoid responsibility and decision-making (Bass, 1991; Avolio et al., 1999).
As mentioned before, in contextually ambidextrous organizations leaders manage exploratory and exploitative activities simultaneously and should therefore establish leadership behaviors that encourage both activities (Tushman & O’Reilly, 1996; Rosing et al., 2011). By executing both transformational and transactional leadership, managers practice complementary and paradoxical behaviors (Burns, 1978; Vera & Crossan, 2004; Rosing et al., 2011). This relates to the establishment of both exploratory and exploitative activities (Jansen et al., 2009b; Rosing et al., 2011). By doing so, managers are able to manage these contradictory demands simultaneously which eventually leads to ambidexterity (Andriopoulus & Lewis, 2009;
Raisch et al., 2009; Lavine, 2014; Lewis et al., 2014). Because of the positive relations between transformational leadership and exploration, and transactional leadership and exploitation (Jansen et al., 2009b), it is expected that the higher the level of the leadership style, the more positive the relation to exploration and exploitation. Additionally, paradoxical leadership styles are needed that combine and integrate the contradictory demands of exploration and exploitation and thus manage the paradoxical demands of ambidexterity (Poole & van de Ven, 1989; Lewis, 2000; Lavine, 2014; Lewis et al., 2014). Balancing high levels of transformational and transactional leadership behaviors is therefore expected to manage exploratory and exploitative activities simultaneously and is thus likely to relate positively to ambidexterity.
Hypothesis 1: Balancing high levels of transformational and transactional leadership relates positively to ambidexterity.
Organizational context
By practicing leadership, managers shape, inter alia, the organization’s context (Ghoshal & Bartlett, 1994; Gibson & Birkinshaw, 2004; Lin & McDonough, 2011). The organizational context is a combination of structural context, culture and climate and consists of systems, processes and beliefs that together shape individual behaviors in the organization (Gibson & Birkinshaw, 2004). Gibson and Birkinshaw (2004) found ambidexterity has a mediating effect between organizational context and firm performance. To measure organizational context, the concept is grouped into four behavior-framing attributes, namely discipline, stretch, trust and support (Ghoshal & Bartlett, 1994; Pastoriza, Arino & Ricart, 2009). According to Ghoshal and Bartlett (1994) the following management actions are needed to
develop these dimensions. First of all, discipline induces members to strive to meet all expectations generated by their commitments. Managers should therefore establish clear standards, feedback and consistent sanctions. Secondly, stretch induces members to strive for more ambitious objectives. To obtain stretch managers should establish a shared ambition, develop a collective identity and give personal meaning to individual contributions. Thirdly, trust induces members to rely on commitments. Fairness, equity and involvement in the decision processes and the right staffing of positions contribute to the establishment of trust. Fourthly, support induces members to assist and count on each other. Managers should allow access to resources, freedom of initiative and personal orientation to contribute to the establishment of support. Organizational context is measured by combining the four dimensions of Ghoshal and Bartlett (1994) into two concepts: ‘performance management context’, which represents discipline and stretch, and ‘social context’, representing support and trust (Gibson & Birkinshaw, 2004). Performance management context, thus, is concerned with the stimulation of people to deliver high-quality results and holding these people accountable for their actions (Birkinshaw & Gibson, 2004). And the social context provides people with the security and latitude they need to perform (Birkinshaw & Gibson, 2004).
Bass and Avolio (1993) argue that transformational leadership influences the organization’s culture, which is part of the organizational context. But the influence of other or a combination of multiple leadership behaviors on organizational context as a whole is underexposed. Given the influence of leadership on organizational context (Ghoshal & Bartlett, 1994; Gibson & Birkinshaw, 2004; Lin & McDonough, 2011) itis necessary to test the balance between high levels of transformational and transactional leadership behaviors in relation to organizational context and ambidexterity, since these behaviors, just like alignment and adaptability, relate to exploration and exploitation as well (Jansen et al., 2009b; Rosing et al., 2011). Thereby, Gibson and Birkinshaw (2004) extend the framework
of organizational context by arguing when managers in organizations create a supportive context, through the performance management context and the social context, individuals in the organization engage in both explorative and exploitative activities and this leads to ambidexterity.
The role of managers, thus, is shaping the organizational context. But it is not enough to just create a supportive organizational context, this supportive context must create the capacity for ambidexterity in order to gain performance (Ghoshal & Bartlett, 1994; Gibson & Birkinshaw, 2004; Lin & McDonough, 2011). When it comes to the development of a supportive organizational context, this is achieved by building a set of systems and processes that allow for both alignment and adaptability (Gibson & Birkinshaw, 2004). Such a context is called a high-performance context and is a balance of a high performance management context and a high social context (Birkinshaw & Gibson, 2004). How managers shape a supportive organizational context for ambidexterity depends on the specific leadership behaviors they establish. To balance high levels of the performance management context and the social context, both must flourish simultaneously (Gibson & Birkinshaw, 2004). It is expected that a combination of transformational and transactional leadership as paradoxical leadership styles relates to both the establishment of the performance management context and the social context. Since a combination of transformational and transactional leadership behaviors describes what performance is required and what can be expected when fulfilling these requirements or not, but is also person oriented and focused on the creation of awareness and acceptance within the group (Bass, 1991; Avolio et al., 1999). Thus, the either and vision of paradoxical leadership allows managers to combine and integrate the dimensions of the organizational context simultaneously (Lavine, 2014; Lewis et al., 2014). Balancing high levels of transformational and transactional leadership is therefore expected
to relate to the establishment of high levels of the performance management context and the social context and thus a high-performance context.
Hypothesis 2: Balancing high levels of transformational and transactional leadership relates positively to a supportive organizational context for ambidexterity.
However, as mentioned before, a supportive organizational context must create the capacity for ambidexterity in order to gain performance (Ghoshal & Bartlett, 1994; Gibson & Birkinshaw, 2004; Lin & McDonough, 2011). Prior research found a direct, positive link between the organizational context and ambidexterity (Gibson & Birkinshaw, 2004). Thereby positive relations were found between leadership and the establishment of ambidexterity (Andriopoulus & Lewis, 2009; Raisch et al., 2009; Smith et al., 2010; Rosing et al., 2011; Turner et al., 2013). Since both leadership and a supportive organizational context relate to ambidexterity, there may be a link between these concepts through which they indirectly affect ambidexterity. However the link between leadership, and in particular transformational and transactional leadership as paradoxical leadership styles, a supportive organizational context and ambidexterity has not been studied yet.
By balancing high levels of transformational and transactional leadership and, as a result, establishing the performance management context and the social context simultaneously, leaders create a supportive organizational context. The creation of this high-performance organizational context allows for alignment and adaptability, or in other words exploration and exploitation, and eventually develops the capacity for organizational ambidexterity (Gibson & Birkinshaw, 2004). Thus, it is expected that it is also possible to obtain ambidexterity indirectly through managing the organizational context. For this effect it is necessary to create a supportive organizational context, which can be established by
balancing high levels of transformational and transactional leadership as paradoxical leadership styles. Therefore it is likely that the effect of the balance between high levels of transformational and transactional leadership on ambidexterity is also indirectly explained through the organizational context.
Hypothesis 3: Organizational context mediates the relationship between balancing high levels of transformational and transactional leadership and ambidexterity.
Environmental dynamism
The hypotheses mentioned above, however, do not take the external environment into account, while the environmental condition of an organization is likely to influence the balance between transformational and transactional leadership as well as the management of the contradictions between exploratory and exploitative activities. Prior studies found that the effectiveness of organizational exploitation and exploration depends on, and is influenced by, different contextual conditions, therefore environmental dynamism as contingency factor should be taken into account when studying ambidexterity (Raisch & Birkinshaw, 2008; Rosing et al., 2011). In addition, it should be examined whether environmental dynamism is an important boundary condition to ambidexterity (Levinthal & March, 1993; Gibson & Birkinshaw, 2004).
Environmental dynamism is defined as the rate and unpredictability of change in the external environment of an organization (Dess & Beard, 1984). Dynamic environments are characterized by changes in technologies, variations in customer preferences and fluctuations in product supply or demand of materials (Jansen et al., 2006). According to Jansen et al. (2006) the level of environmental dynamism can be measured through four dimensions.
These dimensions are: changes in the environment of the local market, changes in client demands, changes in the local market and changes in the volume of products and services to be delivered (Jansen et al., 2006). The level of environmental dynamism might influence the effects between leadership, organizational context and ambidexterity in two ways.
Leadership, organizational context and environmental dynamism
Firstly, the level of environmental dynamism may affect the relation between balancing high levels of transformational and transactional leadership and a supportive organizational context for ambidexterity. Different effects of environmental dynamism are studied, however how environmental dynamism relates to different contexts and leadership characteristics has not been investigated yet (Raisch & Birkinshaw, 2008). Dynamic capabilities are routines by which managers acquire, integrate and recombine resources to adapt to and align with the environment (Eisenhardt & Martin, 2000). According to Eisenhardt and Martin (2000), in dynamic environments there is a higher demand for more simple routines that are based on new knowledge and selection. In contrast, relatively stable markets ask for more extensive routines and variation that rely on existing knowledge (Eisenhardt & Martin, 2000). As a result of combining transformational and transactional leadership as paradoxical leadership styles, a broad repertoire of solutions for various situations is obtained (Eisenhardt & Martin, 2000). These dynamic capabilities enable managers to meet rising and contradictory demands (Eisenhardt & Martin, 2000; Lavine, 2014; Lewis et al., 2014). In creating a supportive organizational context for ambidexterity, the performance management context and the social context should not drive out one or another (Gibson & Birkinshaw, 2004). Managers should therefore create an organizational context that is in balance and able to meet the requirements of the environment (Gibson & Birkinshaw, 2004; Dutta, 2013). In relatively stable
environments these environmental demands are predictable but in dynamic environments these demands are continuously changing (Dess & Beard, 1984). In dynamic environments it is therefore of greater value to possess dynamic capabilities to adapt to and align with the environment (Eisenhardt & Martin, 2000). The external environment thus determines which management capabilities should be used in order to create a supportive organizational context that meets the requirements of the environment. In other words, environmental dynamism functions as an integrations mechanism when it comes to matching dynamic capabilities with the requirements of the environment in establishing a supportive organizational context for ambidexterity. In addition organizational survival depends on paradoxical leadership in highly dynamic environments, because by doing so managers are able to meet rising and contradictory demands (Lavine, 2014; Lewis et al., 2014). Given these findings, the following effect of environmental dynamism as a moderator on balancing high levels of both leadership styles and organizational context is expected:
Hypothesis 4: Environmental dynamism moderates the relationship between balancing high levels of transformational and transactional leadership and a supportive organizational context for ambidexterity, so that this relationship is stronger for higher levels of environmental dynamism.
Organizational context, ambidexterity and environmental dynamism
Secondly, the level of environmental dynamism can influence the relation between the organizational context and ambidexterity. However, different factors in the external environment of the organization can influence this balance (Kim & Huh, 2015) and it depends on the nature of the external environment what type of innovation is effective
(Zahra, 1996; Zahra & Bogner, 1999). Therefore, in order to achieve ambidexterity, organizations must build on an appropriate organizational context that meets the requirements of the environment (Dutta, 2013). Several studies found that organizations are more likely to become ambidextrous when their environment is highly dynamic (Jansen, Volberda & van den Bosch, 2005; Raisch & Birkinshaw, 2008). A supportive organizational context is created when organizations score high on establishing both performance management and social support (Birkinshaw & Gibson, 2004). This is called a high-performance organizational context and enables ambidexterity (Birkinshaw & Gibson, 2004). From a dynamic capabilities point of view high-performance work systems create a more flexible and responsive organizational context in order to align with and adapt to changes (Eisenhardt & Martin, 2000; Wei & Lau, 2010). This is the case, because high-performance work systems provide a learning platform, which enables organizations to respond to market changes and engage in innovations (Wei & Lau, 2010). In dynamic environments organizations perform better when they are cautious about external changes and are able to exploit unexpected opportunities (Aldrich, 1979; Bhattacharya, Krishnan & Mahajan, 1998), and when high-performance work systems are able to meet these demands (Eisenhardt & Martin, 2000; Birkinshaw & Gibson, 2004; Wei & Lau, 2010). Therefore it is expected that when organizations with a supportive organizational context for ambidexterity face a high level of environmental dynamism, these organizations are more likely to become ambidextrous.
Hypothesis 5: Environmental dynamism moderates the relationship between a supportive organizational context and ambidexterity, so that this relationship is stronger for higher levels of environmental dynamism.
Data and method
Research setting and data collection
The research was conducted in different branches in the Netherlands, operating in different industries. To measure the constructs from the literature at a firm level, the respondents had to be in the senior management team of the company. Since it otherwise was not possible to report on the companies’ leadership and innovation approach and the supportiveness of the organizational context regarding these approaches on a firm level. Thus, to measure top executives as antecedents of and their effects on ambidexterity, the unit of analysis was set at the senior management team level. It was ensured that the participants were professionally interested, conscientious and committed to provide accurate data by guaranteeing the privacy of the respondents and by offering them a report of the research findings.
This explanatory study was operated through a survey design, because many respondents could be reached in a short period of time by doing so (Saunders & Lewis, 2012). The sample used for the research consisted of two mailing lists, one with recruiter contacts provided by ‘Integrand Nederland’ and one with general business mail addresses. To reach senior managers, the receivers of the emails were asked to forward the email that contained a link to the questionnaire to senior managers of their organization or, in case they were a senior manager, to fill in the questionnaire. In total 2502 emails were sent to the recruiters of the companies that cooperate with Integrand Nederland. In addition, 3219 emails were sent to a database obtained from a Dutch commercial provider, the ‘Kamer van Koophandel’. In the end, 186 of the 5721 contacts completed the questionnaire, this makes a response rate of 3.25%.
The sample contained organizations with an average age between 11 and 25 years, an average firm size between 10 and 49 full-time employees and an average senior management
team size between 5 and 10 members. The organizations in the sample are active in different industries, namely: the primary sector (6%), the manufacturing industry (16%), retail (14%), financial service industries (11%), other service industries (43%) and government organizations (9%).
Measurement and validation of constructs
In the survey, in the form of an online questionnaire, existing multi-item scales were used (see Appendix). In the following section the measurement and the validation of the dependent, the independent, the mediating, the moderating and the control variables are discussed.
Dependent variable
Exploratory and exploitative innovation were used to measure ambidexterity as dependent variable, since they are orthogonal (Gibson & Birkinshaw 2004; He & Wong, 2004). Exploratory factor analysis clearly replicated the intended two-factor structure with each exploratory and exploitative innovation loading on their intended factors. All items had factor loadings above 0.5, however one item did not have a cross-loading below 0.4. After deleting this item all factor loadings were above .55 with cross-loadings below 0.31 and both factors having eigenvalues greater than one. Both the measures for exploratory and exploitative innovation were adapted from Jansen et al (2006). Exploratory innovation (α = 0.76) consisted of a four-item scale and exploitative innovation (α = 0.71) consisted of a three-item scale after deleting one item, the items were measured on a seven-point Likert-scale, anchored by 1 = strongly disagree and 7 = strongly agree. Exploratory innovation captured to what extent organizations depart from existing knowledge and pursue radical innovations for
new opportunities. Exploitative innovation captured to what extent organizations build on existing knowledge and pursue incremental innovations. The measure for ambidexterity is constructed by multiplicative interaction between exploratory and exploitative innovation, because these two constructs are non-substitutable and interdependent (Gibson & Birkinshaw, 2004).
Independent, mediating and moderating variables
Balancing high levels of transformational and transactional leadership was measured by the creation of a new variable in which the average scores on both leadership styles were combined. Both constructs were first measured separately on a seven-point Likert-scale, anchored by 1 = strongly disagree and 7 = strongly agree. The five-item scale for transformational leadership (α = 0.90) and the four-item scale for transactional leadership (α = 0.77) were adapted from Hoch, Pearce and Welzel (2010). Transformational leadership captured to what extent a top executive inspires, energizes and intellectually stimulates his/her employees. Transactional leadership examined to what extent a top executive is directive and task-oriented based on transactions with his/her employees. The factor loadings of all factors were above .75 and each leadership style loaded on a single factor with an eigenvalue of more than one. The balance between high levels of both transformational and transactional was constructed as a binary variable. Respondents that perceived transformational leadership higher then 5.28 and transactional leadership higher then 4.91 were coded as ‘1’, all other respondents were coded as ‘0’. By doing so the ‘high leadership balance’ variable, in which top executives practise high levels of both transformational and transactional leadership, was created.
In constructing the measure for organizational context, multiplicative interaction of the performance management context and the social context was used to create an interaction
term. Organizational context is constructed this way, because the performance management context and the social context should be considered holistically and are non-substitutable (Gibson & Birkinshaw, 2004). The performance management context (α = 0.74) consisted of four items that all loaded on a single factor, having an eigenvalue of 2.26 and accounting for 56.5% of the variance. Performance management context captured to what extent a top executive stimulates his/her employees to deliver high-quality results and holds employees accountable for their performance. To capture the social context (α = 0.83) a five-item scale was used. Like the performance management context items, the social context items were adapted from Birkinshaw and Gibson (2004) and measured on a seven-point Likert-scale, anchored by 1 = not at all and 7 = to a very great extent. The social context captured to what extent a top executive provides his/her employees with the security and latitude they need to perform. The five items loaded on a single factor, having an eigenvalue of 2.97 and accounting for 59.3% of the variance.
The four-item scale of environmental dynamism (α = 0.77), adapted from Jansen et al. (2006), tapped into the rate of change and the instability of the external environment of the respondents. The construct was measured on a seven-point Likert-scale, anchored by 1 = strongly disagree and 7 = strongly agree. All four items loaded on a single factor with an eigenvalue of 2.43.
Control variables
By including relevant control variables, possible alternative explanations were checked. The included control variables are: firm size, firm age, senior management team size and industry. Firm size was chosen, because larger firms may have more resources but may also be less flexible to achieve ambidexterity. Therefore, a natural logarithm of the number of employees within the organization is included to account for firm size. Firm age is measured as well
with a natural logarithm of the number of years the firm exists. Firm age was included, because incumbent firms are more likely to rely on exploitative efforts in comparison with relatively ‘new’ firms in an industry (Gilbert, 2005). Senior management team size may affect the heterogeneity of the senior management team and thus the achievement of ambidexterity, this variable is measured by the number of senior executives in the organization. Firm industry controls to what extent organizations are more likely to pursue exploratory and exploitative innovation, because this may depend on the context or industry of an organization (He &Wong, 2004). There are six industry dummies identified based on the answers on the industry question in the questionnaire: the primary sector, manufacturing, retail, financial services, other services and government organizations.
Analyses and results
Table 1 represents the means, standard deviations and correlations between the constructs that are used in the analyses to test the hypotheses. In order to examine multicollinearity, the variance inflation factors (VIFs) for each of the regression equations were calculated. The maximum VIF within the models was 1.11, which is well below the rule-of-thumb cutoff of 10 (Neter, Wasserman & Kutner, 1990). To test the hypotheses, different analyses were executed. At first, hypothesis one and two were examined through a hierarchical regression analysis. Secondly, the third, fourth and fifth hypotheses were tested by both a hierarchical regression analysis and a simple mediation and moderation model provided by Andrew F. Hayes. Thirdly, in the robustness paragraph, the results were tested for the common method bias through Harman’s one-factor test. Fourthly, post hoc analyses on the results in this paragraph were executed and will also be provided in the robustness paragraph.
Hierarchical regression was performed to investigate the ability of balancing high levels of both transformational and transactional leadership to predict the supportiveness of the organizational context and ambidexterity, after controlling for firm age, firm size, top management team size and industry. When testing for the effect of balancing high rates of transformational and transactional leadership on ambidexterity (table 2), the control variables were entered in model 1. This model was statistically significant F (9, 173) = 3.90; p < 0.001 and explained 16.9% of the variance in ambidexterity. In model 2 the balance between high rates of both leadership styles were entered, the total variance explained by the model as a whole was 27.2% F (10, 172) = 24.43; p < 0.001. As shown in model 2, the coefficient for balancing high levels of both leadership styles was positive and significant (β = 0.33, p < 0.001). This supports the first hypothesis, namely: balancing high levels of transformational and transactional leadership relates positively to ambidexterity.
The relationship of balancing high levels of transformational and transactional leadership on organizational context was tested through a hierarchical regression analysis as well (table 2).
Model 7 includes the control variables (F (9, 173) = 0.79; p > 0.05) and model 8 includes the balance between high levels of transformational and transactional leadership (β = 0.09, p > 0.05). However, none of these models are significant. From this, it can be concluded that the second hypothesis, balancing high levels of transformational and transactional leadership relates positively to a supportive organizational context for ambidexterity, is not supported.
For the third hypothesis, model 3 (table 2) partly provides insight in this. When adding organizational context as a mediator in model 3, this variable is not significant in relation to the other variables (β = -0.01, p > 0.05). When organizational context was mediating the leadership effect on ambidexterity, the high leadership balance variable should become less or not significant, which is not the case in model 3 (β = 0.33, p < 0.001). This is supported by the simple mediation model of Andrew F. Hayes, only the direct effect of balancing high levels of transformational and transactional leadership on ambidexterity is significant (c’ = 6.24, p = 0.000, CI: 3.77 to 8.71). For balancing high levels of both leadership styles on organizational context (a = 0.24, p = 0.213) and organizational context on ambidexterity (b = -0.05, p = 0.912) no significant effects were found. Thus the third hypothesis, organizational context mediates the relationship between balancing high levels of transformational and transactional leadership and ambidexterity, is not supported.
To test the fourth hypothesis, the moderating effect of environmental dynamism was tested between balancing high levels of transformational and transactional leadership and organizational context. The simple moderation model of Andrew F. Hayes, however, was not significant, t (183) = 0.86, p > 0.05. It is not possible to continue the analysis when the model is not significant, therefore no moderating effect was found between balancing high levels of both leadership styles and organizational context. In model 9 and 10 (table 2) the direct
effects and the interaction effect of balancing high levels of transformational and transactional leadership and environmental dynamism on organizational context are described. Neither balancing high levels of both leadership styles (β = 0.09, p > 0.05) nor environmental dynamism (β = 0.00, p > 0.05) does relate significantly to a supportive organizational context. Thereby, no interaction effect between balancing high levels of both leadership styles and environmental dynamism on organizational context was found (β = 0.10, p > 0.05). Concluding the fourth hypothesis, environmental dynamism moderates the relationship between balancing high levels of transformational and transactional leadership and a supportive organizational context for ambidexterity, so that this relationship is stronger for higher levels of environmental dynamism, is not supported.
Examining the moderating effect of environmental dynamism between a supportive organizational context and ambidexterity tested the fifth hypothesis. The simple moderation model of Andrew F. Hayes was significant, t (184) = 8.68, p < 0.001. Additionally, the regression coefficient for the interaction effect = 1.03 and is statistically different from zero, t (184) = 2.30, p < 0.05. Thus, the effect of a supportive organizational context on the creation of ambidexterity depends on the level of environmental dynamism. Moreover, this interaction effect accounts for 2.0% of variance in the creation of ambidexterity. In model 4, 5 and 6 (table 2) the direct effects and the interaction effect of organizational context and environmental dynamism on ambidexterity are described. Organizational context (β = 0.02, p > 0.05) does not significantly relate to ambidexterity, environmental dynamism however, does (β = 0.42, p < 0.001). Thereby a significant interaction effect between organizational context and environmental dynamism on ambidexterity was found (β = 0.15, p < 0.05). The moderating effect of environmental dynamism on the relation between the organizational context and ambidexterity is provided in figure 2. These findings support the fifth hypothesis: environmental dynamism moderates the relationship between a supportive organizational
context and ambidexterity, so that this relationship is stronger for higher levels of environmental dynamism.
Figure 2: The moderating effect of environmental dynamism
Robustness
In this paragraph, first the concepts used in the analyses are tested for the common method bias. After that, the results provided in the previous section will be discussed in detail through several post hoc analyses. Post hoc analyses will be provided for the results used to test the first, second, third, fourth and fifth hypotheses in order to obtain more detailed insights in the relations between the concepts.
To test for the common method bias, all the Likert-scale variables in the survey were examined by Harman’s one-factor test. Unrotated principal axis factor analysis revealed that the first factor explained 29.2% of the variance, thus the first factor did not account for the majority of the variance. In addition, all factors with eigenvalues greater than 1.0 together account for 53.9% of the total variance. Since there was not a single factor accounting for the
majority of covariance and more than one factor emerged from the factor analysis, the common method bias is not present (Podsakoff & Organ, 1986). This means that in shaping the pattern of findings of the survey the common method bias was not a serious factor (Heavey & Simsek, 2014).
Balancing high levels of transformational and transactional leadership relates positively to ambidexterity. Through a hierarchical regression the separate effects between both transformational and transactional leadership were examined. After entering the control variables, transformational leadership did have a significant effect on exploratory innovation (F (10, 173) = 9.22; p < 0.01). The coefficient for transformational leadership was positive and significant in relation to exploratory innovation (β = 0.22, p < 0.01). In addition transactional leadership did have a significant effect on exploitative innovation (F (10, 173) = 22.76; p < 0.001). Also, the coefficient for transactional leadership (β = 0.32, p < 0.001) relates positively to exploitation.
When testing for the relation between balancing high levels of transformational and transactional leadership and a supportive organization context for ambidexterity, no significant effects were found. But when testing for the relation of balancing high levels of both transformational and transactional leadership on the two dimensions of organizational context separately, other results were obtained. Balancing high levels of both leadership styles does positively relate to the performance management context dimension (β = 0.42, p < 0.001) and to the social context dimension (β = 0.40, p < 0.001).
Thereby, the results in the previous paragraph did not support the expected mediating effect of organizational context on the balance of high levels of transformational and transactional leadership and ambidexterity. However, since the high leadership balance did have a positive effect on the performance management context dimension and the social context dimension of organizational context, the mediating effects of these dimensions were
tested as well. Firstly, performance management context does relate positively to ambidexterity (β = 0.32, p < 0.001). Also, the direct effect of balancing high levels of transformational and transactional leadership still relates positively to ambidexterity when adding the performance management context (β = 0.19, p < 0.01), but is a somewhat less significant than without the performance management context variable (β = 0.33, p < 0.001). When testing with Andrew F. Hayes’ mediation model, the model turns out significant t (183) = 7.21, p < 0.001. For the direct effect between balancing high levels of transformational and transactional leadership and ambidexterity t (183) = 2.80, p < 0.01 (coefficient = 3.68, SE = 1.32, CI: 0.91 to 6.19). The values for the effect of balancing high levels of transformational and transactional leadership on performance management context are t (183) = 6.46, p < 0.001 (coefficient = 0.85, SE = 0.13, CI: 0.59 to 1.11). And for the effect of performance management context on ambidexterity t (183) = 4.39, p < 0.001 (coefficient = 3.00, SE = 0.68, CI: 1.65 to 4.36).
In addition, social context does relate positively to ambidexterity (β = 0.17, p < 0.05). The direct effect of the high leadership balance still relates positively to ambidexterity when adding the social context in the regression (β = 0.26, p < 0.01), but is as well somewhat less significant than without the social context variable (β = 0.33, p < 0.001). When testing with Andrew F. Hayes’ mediation model, the model turns out significant t (183) = 9.03, p < 0.001. For the direct effect between balancing high levels of both leadership styles and ambidexterity t (183) = 3.55, p < 0.001 (coefficient = 4.85, SE = 1.36, CI: 2.15 to 7.55). The values for the effect of balancing high levels of transformational and transactional leadership on social context are t (183) = 6.47, p < 0.001 (coefficient = 0.84, SE = 0.13, CI: 0.58 to 1.10). And for the effect of social context on ambidexterity t (183) = 2.28, p < 0.05 (coefficient = 1.64, SE = 0.72, CI: 0.22 to 3.06).
Although the interaction model to test the moderation effect of environmental dynamism on the relation between high levels of transformational and transactional leadership was not significant, the conditional effect of environmental dynamism was tested through the Johnson-Neyman test. However, the Johnson-Neyman test did not provide any results, which means irrespectively of the rate of environmental dynamism there were no interaction effects with balancing high levels of transformational and transactional leadership. To obtain further insights, the separate effects of transformational and transactional leadership on organizational context when interacting with environmental dynamism were tested. According to the simple moderation model of Andrew F. Hayes, although the overall model is not significant (t (184) = 1.11, p > 0.05), transformational leadership does have a significant interaction effect with environmental dynamism (t (184) = 2.35, p < 0.05 (coefficient = 0.20, SE = 0.08, CI: 0.03 to 0.36)). According to the Johnson-Neyman test the conditional effects of environmental dynamism indicate that transformational leadership interacts negatively with low levels (effect = between -0.62 and -0.24, SE = between 0.26 and 0.12, CI: between -1.13 and -0.47 to between -0.11 and 0.00) and interacts positively with high levels (effect = 0.31, SE = 0.18, CI: 0.00 to 0.66) of environmental dynamism. Transactional leadership does not have a significant interaction effect with environmental dynamism on the organizational context (t (184) = 1.13, p > 0.05). Although not significant, the conditional effects were examined through the Johnson-Neyman test. However, no conditional effects were provided. This indicates that for both high and low levels of environmental dynamism there is no interaction effect with transactional leadership when it comes to the creation of a supportive organizational context for ambidexterity.
The findings support the moderating role of environmental dynamism on the relationship between a supportive organizational context and ambidexterity. A closer inspection of the conditional effects through the Johnson-Neyman test indicates the
relationship between a supportive organizational context and ambidexterity as significant for low (effect = between 2.91 and 2.12, SE = between 1.39 and 1.07, CI: between 5.66 and -4.24 to between -0.16 and 0.00) and high (effect = between 1.45 and 1.97, SE = between 0.73 and 0.92, CI: between 0.00 and 0.15 to between 2.89 and 3.79) levels of environmental dynamism. As it can be seen from probing the interactions, the slope linking organizational context and ambidexterity is negative for a low level of environmental dynamism and positive for a high level of environmental dynamism. In other words, a lower level of environmental dynamism has a negative influence on the effect between organizational context and ambidexterity. However, a high level of environmental dynamism has a positive direction on this effect.
Though it is not included in the research question, the moderating effect of environmental dynamism was tested on the relation between balancing high levels of transformational and transactional leadership and ambidexterity. The simple moderation model of Andrew F. Hayes is significant, t (183) = 9.64, p < 0.001. The interaction effect between balancing high levels of both leadership styles is not significant (t (183) = 0.49, p > 0.05), nevertheless, Johnson-Neyman’s conditional effects are provided. The conditional effects of environmental dynamism suggest that the relation between balancing high levels of transformational and transactional leadership and ambidexterity is significant for low, medium and high levels of environmental dynamism (effect = between 4.10 and 5.91, SE = between 2.08 and 2.25, CI: between 0.00 and 1.47 to between 8.20 and 10.34). Only in environments with very low dynamism or no dynamism the effect between balancing high levels of both leadership styles and ambidexterity is not significant.
Discussion
Discussion of results
As can be concluded from the results described in the previous chapter, the first and fifth hypotheses were tested as significant. This entails, firstly, that balancing high levels of transformational leadership relates positively to the establishment ambidexterity. This is supported by the fact that transformational leadership relates positively to exploratory innovation and transactional leadership to exploitative innovation (Jansen et al., 2009b), the previous was also found in this research. In addition, the higher the level of transformational and transactional leadership, the more positively these leadership styles relate to exploration and exploitation. Thus, by balancing high levels of both transformational and transactional leadership, leaders are able to manage the contradictory demands between exploratory and exploitative activities simultaneously and therefore relate positively to the establishment of ambidexterity (Andriopoulus & Lewis, 2009; Raisch et al., 2009; Lavine, 2014; Lewis et al., 2014).
Additionally, the findings entail that environmental dynamism moderates the relation between a supportive organizational context and ambidexterity. This is supported by the dynamic view which states that performance work systems, and thus a high-performance organizational context, are able to create a more flexible and responsive context to align with and adapt to changes (Eisenhardt & Martin, 2000; Wei & Lau, 2010). By doing so, the high-performance organizational context enables organizations to respond to market changes and to engage in innovations (Wei & Lau, 2010). However, given the results, low environmental dynamism moderates the relationship between a supportive organizational context and ambidexterity, so that this relationship is negative for low levels of environmental dynamism. This is in line with the dynamic view as well. High-performance
organizational contexts are more flexible and responsive and will work best in dynamic and unstable environments (Arogyaswamy & Byles, 1987; Chandler, Keller & Lyon, 2000). An organization in a stable environment does not demand the dynamic capabilities of a high-performance organizational context in order to establish ambidexterity. Therefore a negative interaction between a supportive organizational context and low rates of environmental dynamism is obtained when it comes to the establishment of ambidexterity.
However, not all hypotheses were supported by the findings of this research, the second, third and fourth hypotheses turned out not significant. First of all, balancing high levels of transformational and transactional leadership does not affect a supportive organizational context for ambidexterity. Although balancing high levels of both leadership styles does relate positively to the establishment of the performance management context and the social context separately, the interaction between the two organizational context dimensions does not lead to a supportive or, in other words, a high-performance organizational context. An explanation for these findings is that the performance management context and the social context are of equal importance and mutually reinforcing (Birkinshaw & Gibson, 2004). Only when both are strongly present, this will create a high-performance organizational context (Birkinshaw & Gibson, 2004). Thus, when there is an imbalance between the performance management context and the social context, or a lack of both, a supportive organizational context for ambidexterity will never be created (Birkinshaw & Gibson, 2004). Balancing high levels of transformational and transactional leadership does relate positively to the two dimensions of the organizational context, but not strongly enough that the dimensions reinforce each other with the creation of a high-performance context as result. A possible explanation is that organizational context is a combination of structure, culture and climate (Gibson & Birkinshaw, 2004). Transformational and transactional leadership, however, are focused on the development of individuals in an organization
(Eisenbach, Watson & Pillai, 1999; Berson, Nemanich, Waldman, Galvin & Keller, 2006). As a result, transformational and transactional leadership do relate positively to the establishment of the culture and climate dimension of the organizational context (Bass & Avolio, 1993; Eisenbach et al., 1999; Porter & McLaughlin, 2006). But a combination of both leadership styles does not relate to the structure dimension of the organizational context, which is focused on how elements in the organization are structured, instead of the individual behavioral approach (Porter & McLaughlin, 2006).
Secondly, the organizational context does not mediate the effect between balancing high levels of transformational and transactional leadership and ambidexterity. This logically follows from the results of the relation between balancing high levels of transformational and transactional leadership and the creation of a supportive organizational context for ambidexterity. Only the strong presence of both the performance management context and the social context will create a high-performance organizational context that enables a truly ambidextrous organization (Birkinshaw & Gibson, 2004). Thus, when the balance between high levels of both leadership styles does not create a high-performance organizational context, this context is not supportive for ambidexterity. And therefore, the created organizational context in this research will not relate positively to ambidexterity. Thereby, the external environment was not taken into account when testing the mediation effect. While it depends on the external environment whether the organizational context meets the requirements of the environment (Dutta, 2013). Again, when testing for the separate effects of performance management context and social context, other results were found. It turns out that the performance management context positively mediates the effect between balancing high levels of transformational and transactional leadership and ambidexterity. The same result is found for the social context. Likewise, the explanation for the second hypothesis is applicable, namely that transformational and transactional leadership are focused more on
