Master’s Thesis Strategy
Choosing exploration and exploitation:
How organizational culture affects the firm’s choice to explore and
exploit.
Name: Mark Christiaan Lucas Diekman Student number: 1111157
Date: 23/06/2017
Degree: MSc. in Business Administration - Strategy Track Institution: Amsterdam Business School - University of Amsterdam Supervisor: Dr. Michiel Tempelaar
Statement of originality
This document is written by student Mark Diekman 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.
Table of contents
1. Abstract 3
2. Introduction 4
3. Theory and Hypothesis 8
a. Organizational culture 8
b. Hofstede’s cultural dimensions 9
c. Choosing exploration and exploitation. 11
d. Linking organizational culture with exploration/exploitation 14
e. The moderating role of budget control 18
4. Methods 21
a. Research setting and data collection 21
b. Measures and validation of constructs 22
5. Analysis and results 27
6. Discussion and conclusion 31
7. Limitations and future research 34
8. Appendices 36
1. Abstract
Prior studies have looked at the different antecedents that influence the firm’s choice to either explore or exploit. A number of scholars have specifically looked that the relation between culture and exploration/exploitation. However, our knowledge on the influence of culture on the firm’s inclination towards exploration or exploitation is still limited. Strong cultures can promote exploitation, while weak cultures promotes exploration. However, the different typologies of culture have never been linked with exploration or exploitation. Based on this thought, I have incorporated the dimensions of organizational culture proposed by Geert Hofstede et al. (1990) as an antecedent of exploration and exploitation and I examined how these cultural dimensions influence the firm’s inclination towards exploration and exploitation. Secondly, I have mediated this relation with budget control, a form of formal management control systems. Overall, my findings suggest that influence of organizational culture on exploitative innovation is present, while influence on explorative innovation is either not present or counterintuitive. The mediating role of budget control is also not present or counterintuitive. Through this paper, I contribute to the knowledge of antecedents influencing exploration and exploitation, creating a better understanding for managers why firms pursue exploration and/or exploitation.
2. Introduction
In the last 20 years, many scholars have done research about the balance between exploration and exploitation within firms (Jansen et al. 2009; Andriopoulos & Lewis, 2009; Tushman et al., 2010). Ever since Jim March wrote his seminal work on exploration and exploitation, there have been many studies on the interplay between exploration and exploitation (March, 1991). On the one hand, firms need to generate new knowledge, to create new products and services for new markets. On the other hand, firms need to leverage existing competences and exploit the known products and services (Jansen et al., 2009) However, exploration and exploitation often seem to be add odds with one another. Whereas exploration has been associated with flexibility, decentralization, and loose cultures, exploitation has been related to efficiency, centralization, and tight cultures (Benner and Tushman 2003). They require fundamentally different architectures and competencies, which creates paradoxical challenges for the firm (Jansen et al. 2009)
In a rational model of choice, the balance between exploration and exploitation is discussed in terms of alternative investment opportunities, each having a probability on return (March, 1991). Information is widely available and choices are made rationally. In the real world, this is off course not the case. Choices are made with limited information and under time pressure. Under these conditions, tradeoffs between exploration and exploitation are made (March, 1991). However, firms choose their strategic direction consciously. These conscious choices are influenced by antecedents. For example: Firms that are successful, can fall in a success trap, which leads them to focus more on exploitation. On the other hand, firms that are failing will look frantically for new innovation, creating a failure trap leading to increased exploration (Levinthal and March, 1993). The firm is being surpassed by its competitors, and is
therefore influenced by the competitive intensity, a known antecedent of exploration (Lavie et al. 2010)
Scholars agree that reaching a balance between exploration and exploitation has a positive influence on firm performance (Gibson and Birkinshaw 2004, He and Wong 2004). However, in accordance with March (1991), scholars have researched not the ability to achieve ambidexterity, but the drivers towards either exploration or exploitation (Jansen et al. 2005 & 2006, Sidhu et al, 2004; Cohen & Levinthal, 1990; Tripsas, 2008). I shall build on this literature, by examining the choice of exploration and exploitation by a firm. I shall do this by looking at the antecedents of exploration and exploitation. I consider the following: Firms do not always make a trade-off due to lack of resources or information. Firms often make a deliberate choice for exploration, exploitation or both, based on the antecedents influencing the firm (Lavie et al. 2010). It is the literature on these antecedents that I shall try to expand.
There are a number of antecedents that have a proven influence on exploration and/or exploitation. Environmental dynamism, competitive intensity, absorptive capacity, slack resources, organizational structure, risk aversion and management control systems to name just a few (Lavie et al., 2010). The presence or lack of these antecedents can increase or decrease the inclination of a firm towards exploration or exploitation. Effectively, certain combinations can increase or decrease both exploration and exploitation. Exploration and exploitation is a continuum, not a discrete choice between either of the two (Lavie et al. 2010).
In the literature on antecedents of exploration and exploitation, culture has also been reviewed, but has been defined differently by scholars. It has been considered a set of values that guide employees towards exploration or exploitation (Miles & Snow 1978; Tripsas 2009). Or it has been considered either weak or strong, where a strong culture implies coherence and shared values amongst employees. (Andrews, Basler & Coller 1999; Sorensen 2002) Yet,
scholars researching the exploration/exploitation antecedents have never empirically researched different typologies of culture and their influence on the firm’s investments in explorative or exploitative innovation. Personally, I find this typology of weak and strong cultures is a very limited way of looking at culture. Authors like Hofstede (1980, 1984, 1990), Sheridan (1992) or Schein (1990) have specified culture into specific dimensions, giving a much more detailed and exact measure of culture. Such dimensions have never been used to look at strategy, even though Weick (1985) had long since indicated that strategy and culture overlap. The cultural dimension can also be linked with other antecedents of exploration and exploitation. Like Bozeman & Kingsley (1998) linking culture with risk aversion or Bass & Avolio (1993) that link organizational culture with leadership.
I will conceptualize organizational culture based on the dimensions proposed by Hofstede et al. (1990) and argue that organizational culture plays a crucial role in a firm’s decision to explore or exploit. I shall use the organizational culture dimensions and compare them with the exploration versus exploitation construct that characterizes how firms strategically prioritize their investment in technological innovation with explorative versus exploitative objectives (He and Wong, 2004). Furthermore, I shall moderate the relation between organizational culture and the exploration/exploitation trade-off with budget control, a known antecedent of ambidexterity (McCarthy & Gordon, 2011),
This paper will contribute to the literature on exploration and exploitation. Many antecedents have been researched that influences a firm’s ability or willingness to engage in either exploration, exploitation or ambidexterity (Jansen et al. 2005 & 2006, Sidhu et al, 2004; Cohen & Levinthal, 1990; Tripsas, 2008). This paper will add the six cultural dimensions proposed by Hofstede (1990) to the list of antecedents influencing a firm’s behavior when it comes to the choice to either explore or exploit. Culture will not only be considered weak or
strong, but having set characteristics in each different firm. For each dimension, the influence on the firm’s inclination for exploration and exploitation shall be measured.
Secondly, this paper will test the influence of management control systems - in the form of budget control - on the relation between organizational culture and exploration/exploitation. MCS has been researched as an antecedent of innovation and ambidexterity (Simons, 1990; McCarthy & Gordon 2011). Ueno and Wu (1993) have looked at the relation between culture and budget control practices. This paper looks at the relation between culture - an informal design feature - and exploration/exploitation. This paper will add to the literature by moderating this relation with a formal design feature, MCS.
Lastly, this study will contribute to the literature on organizational culture. Firstly, it will address the gap in the literature on the influence of organizational culture on the choice of exploration/exploitation. This paper will be the first to use Hofstede’s (1990) organizational dimensions to empirically test their influence on a firm’s inclination towards exploration and exploitation. I shall look further than the previous literature on culture and exploration/exploitation, by recognizing organizational culture as having different dimensions and measuring these dimensions and their influences on the firm’s inclination towards exploration/exploitation. Weick already proposed in 1985 that culture and strategy are overlapping constructs (Weick, 1985). Specifying cultural dimensions in a crucial strategic choice to explore and/or exploit is a very valuable contribution to the existing literature of exploration and exploitation.
This paper will try to answer the question: “How does organizational culture influence the firm’s inclination towards exploration or exploitation?”.
The paper unfolds along the following lines. The next section presents the literature review and hypotheses for why preferences for exploration or exploitation should vary across
different organizational cultures and offer hypotheses about the effect of the six cultural dimensions proposed by Hofstede et al. (1990) on the preferred choice of either exploration and exploitation. Following this, I shall present the methodology used to test the hypotheses, the data analysis and the results. I will close with my conclusions and discussions of the implication of this study for managers.
Theory and hypotheses Organizational Culture
Culture has long since been a subject of interest to academic scholars on strategic literature. For corporate strategy, organizational culture has acquired a status almost similar to other pillars in the strategic literature, like structure, strategy and control (Hofstede, 1990). In 1985, Weick even argued that culture and strategy are overlapping constructs (Weick, 1985). And in the last decades, many studies have linked culture not only with aspects of strategic literature (Denison, 1990; Denison & Mishra, 1995) but also with ambidexterity (Miles & Snow, 1978; Tushman & O’Reilly, 1996; Sorensen, 2002; Tripsas, 2009; Andriopoulos & Lewis, 2009; Wang & Rafiq, 2014).
Sorensen (2002) discusses culture as one of the variables that influence the choice between exploration and exploitation. He sees that ‘strong-culture’ organisations are generally ill-suited to exploratory learning, since they have difficulty recognizing the need for change. ‘Weak-culture’ organizations are better at responding to environmental volatility. According to the author, ‘strong-cultures’ prompt exploitation of existing capabilities at the expense of exploration, since consensus on corporate goals and values provides a strong foundation for organizations’ exploitation capabilities (Sorensen, 2002).
Culture is a fairly broad concept, and after further researching the cultural antecedent, different meanings of the concept ‘culture’ come up. Miles & Snow (1978) and later Tripsas (2009) say that the organization’s identity provides its members with a perspective on organizational goals, mission, and dominant logic that guides exploratory and exploitative activities. Alvesson & Willmott (2002) says that organizational culture pertains to the attitudes, experiences, beliefs and values that guide the behavior of organizational members. He claims that ‘strong-cultures’ constrain an organization to stay within the realm of what is known and established. Effectively, the authors propose that firms with a strong culture will focus on exploitation, and will consider exploration as less important. Andriopoulos & Lewis (2009) demonstrate that strong organizational cultures lack the ability to observe and respond to identity-challenging environmental threats. Weak cultures do not have a set of shared beliefs and a set long term vision, which makes them more adaptable in disruptive situations. (Andriopoulos & Lewis, 2009)
However, in these settings, many authors perceive the antecedent ‘culture’ as either strong or weak. A strong firm culture means a shared mission, goals and values, a strong company foundation (Sorensen, 2002; Andriopoulos & Lewis, 2009). It does not dissect the firm’s culture. It does not create a typology which allows us to classify different types of organizational culture. To truly test organizational culture as an antecedent for the trade-off between exploration and exploitation, we must have different typologies for firm culture that are measurable qualitatively and quantitatively. For this, we must turn to Hofstede.
Hofstede’s cultural dimensions
In 1990, Hofstede, together with three co-authors, did a quantitative study on organizational culture in The Netherlands and Denmark (Hofstede et al. 1990). They argued that
organizational cultures are partly predetermined by nationality, industry, and task, partly related to organizational structure and control systems, and partly unique products of idiosyncratic features like the organization's history or the personality of its founder (Hofstede et al., 1990). Before this study, scholars agreed that organizational culture was based on firm rituals, heroes, symbols and values (Deal & Kennedy, 1982). Hofstede (1990) set out to discern organizational practices from the organizational values. He labeled organizational rituals, heroes, symbols as practices, and classified these in different dimensions, similar to his seminal work on national culture (Hofstede, 1980, 1983, 1984). According to Hofstede et al. (1990) organizational values is what the employee of the firm should be, whilst organizational practices is what the employee actually is. This part of culture, the organizational practices, have seldom been linked to organizational choices or results. The study by Hofstede et al. (1990) also showed that values give a very small interorganizational difference, where practices gives a large significant difference between firms.
Hofstede et al. (1990) discerned six cultural dimensions from his quantitative study: Process-Oriented vs. Results-Oriented. Employee-Oriented vs. Job-Oriented. Parochial vs. Professional. Open System vs. Closed System. Loose Control vs. Tight Control. Pragmatic vs. Normative. Each of these dimensions - Hofstede refers to them as P1 to P6 - consists of opposites. Dimension P1 opposes a focus on means (process-orientation) and a focus on goals (results-orientation). P2 considers a concern for people (employee-orientation) on the one hand, and a concern for getting the job done (job-orientation) on the other hand. Dimensions P3 sees parochial as being business units whose people derive their identity from the firm, while professional means people who identify with the job they do. P4 is the opposition of open vs. closed systems. The tight system of dimension P5 looks at the amount of internal structuring. Rules and regulations on behavior and clothing. A loose system is quite the opposite,
employees are free and unbound. Finally, dimension P6 looks at the level of customer orientation. Pragmatic firms are more market-driven, while normative firms believe their job is determined by inviolable rules (Hofstede et al, 1990)
Hofstede’s study showed that being a member of an organizational explains a significant share in variance in culture, compared to firms with a similar national culture (Hofstede, 1990). A premise, later confirmed by several other scholars (Sheridan, 1992; Chatman & Jehn, 1994; Schein, 1990). These authors however, discerned different dimensions for organizational culture. Hofstede et al. (1990) had anticipated on this difference. They noted that due to the differences in national culture, the dimensions of firm culture can be different or incomplete in other countries. Therefore, the dimensions of organizational culture can change between countries. Since this study will look at Dutch firms, consistent with the quantitative research done by Hofstede et al. (1990), the dimensions listed above will be suitable to extrapolate to the responsegroup in this study.
Choosing exploration and exploitation.
“Exploration and exploitation compete for the organization’s scarce resources and entail distinctive sets of skills and capabilities” (Lavie et al., 2010). Since exploration and exploitation compete for the same scarce resources, an optimal balance between the two seems impossible. Authors like Lavie et al. (2010) and March (1991) rule out the possibility of a firm being ambidextrous.
Already in 1991, March came to the same conclusion: “Both exploration and exploitation are essential for organization, but they compete for scarce resources. As a result, organizations make explicit and implicit choices between the two. The explicit choices are found in calculated decisions about alternative investments and competitive strategies. The implicit choices are
buried in many features of organizational forms and customs, for example, in organizational procedures for accumulating and reducing slack, in search rules and practices, in the ways in which targets are set and changed, and in incentive systems” (March, 1991).
However, the trade-off between exploration and exploitation should not be viewed as a choice between discrete options, but as a continuum (Lavie et al., 2010). Firms do not choose between exploration or exploitation, but have a higher or lower inclination towards either one of the two. The position of innovation on the exploration/exploitation continuum is defined by the “degree of relatedness between the knowledge embedded in an new innovation and the organization’s existing knowledge base” (Lavie et al., 2010). It is on this notion of innovation in the exploration/exploitation continuum that I shall further build in this paper.
How can we define exploration and exploitation? There is consensus regarding the fact that exploration refers to learning and innovation. Exploitation seems to be more difficult to pin down. The debate considers the question whether both exploration and exploitation are distinguished by differences in the type of learning (Baum, Li & Usher 2000) or by the presence versus the absence of learning (Rosenkopf & Nerkar, 2001).
That exploration and exploitation both entail a form of learning is best described by Benner & Tushman (2002): “Exploitative innovations involve improvements in existing components and build on the existing technological trajectory, whereas exploratory innovation involves a shift to a different technological trajectory” (Benner & Tushman, 2002). Also March (1991) described both exploration and exploitation as some type of learning. March argued that even when an organization is attempting to do nothing more than replicate past actions, it still accumulates experiences and goes through a learning curve. “The essence of exploitation is the refinement and extension of existing competencies, technologies and paradigms (...) The essence of exploration is experimentation with new alternatives” (March, 1991). He & Wong
(2004) also emphasized the fact that both exploration and exploitation entail an aspect of innovation. They argued that firms develop alongside two generic dimensions: (1) an explorative innovation dimension to denote technological innovation activities aimed at entering new product-market domains and (2) an exploitative innovation dimension to denote technological innovation activities aimed at improving existing product-market positions. These two generic dimensions were referred to as explorative innovation strategy and exploitative innovation strategy in their paper (He & Wong, 2004). I will build on this premises and propose that explorative innovation and exploitative innovation fall in the continuum of a firm’s knowledge base. I will examine to what degree the firm has an inclination towards exploration, exploitation or both, within this continuum.
Now that the definition of exploration and exploitation has been defined, it is now crucial to see what rationale drives a firm to make the trade-off between exploration and exploitation. The effect of exploration and exploitation on a firm differ greatly. Their returns vary not only with respect to their expected values, but also with respect to their variability, their timing, and their distribution with and beyond the organization. (March, 1991) This means that the process for resource allocation between exploration and exploitation will embody intertemporal, interinstitutional, and interpersonal comparisons, as well as risk preferences. (March, 1991)
Beside the arbitrary decisions of resource allocation, exploration and exploitation also portray an opposing nature. This is derived from resource-allocation constraints, organizational inertia, and desirable organizational outcomes. Firms experience bottlenecks when it comes to these three factors. First of all, decisions on resource-allocation are made consciously by firms, therefore, the firm proactively chooses to support either exploration or exploitation. The firm makes a choice between the expected consequences of these activities (Lavie et al., 2010). Another aspect of the choice between exploration and exploitation concerns the choice between
stability and adaptability. Decision makers in a firm have certain risk preferences, where one prefers stability and others prefer flexibility. Flexibility and change are often associated with exploration, while stability and inertia are often associated with exploitation that confines adaptation to things already known. Lastly, exploration and exploitation often produce divergent organizational outcomes, which both can be positive. But while exploration and exploitation are both crucial to the profitability - even survival - of the firm, limited resource availability can compel organizations to favor one type of activity over the other. (Lavie et al., 2010)
However, there are other factors - antecedents - which influence the choice between exploration and exploitation. These include environmental factors such as dynamism (Jansen et al. 2005), exogenous shocks (Benner & Tushman 2002), and competitive intensity (Levinthal & March, 1993;). The tendency to explore or exploit is also affected by an organization’s history captured by its age (Sorensen & Stuart, 2000), size (Rothaermel & Deeds, 2004), slack resources (Voss et al., 2008), absorptive capacity (Cohen & Levinthal, 1990), organizational structure (Jansen et al., 2006; Jansen et al., 2009) and culture (Sorensen, 2002; Andriopoulos & Lewis, 2009). Finally, managerial antecedents like managerial biases (Leonard-Barton, 1992; Levinthal & March, 1993; O’Reilly & Tushman, 2008) may drive the organization towards exploration and/or exploitation. Together, these factors influence an organization’s propensity to explore, exploit, or strive toward balance. This paper will contribute to the literature on antecedents of the exploration/exploitation trade-off by further examining culture as an antecedent.
Linking organizational culture with exploration/exploitation
In order to link the six dimensions of organizational culture proposed by Hofstede et al. (1990) with exploration/exploitation, it is key to identify the behavioral typologies derived from
the two opposite sides of organizational culture that Hofstede et al. (1990) describe. For example, P5, loose vs. tight control was already linked by Hofstede et al. (1990) with management control. An organization with tight control in rules and regulations, with stringent budget control or with strict managerial control systems will behave fundamentally different than a firm with loose control (Hofstede et al, 1990). Formal management control systems have been named a deterrent for creativity and coping with uncertainty. It is a hindering factor when it comes to new product innovation (Abernethy & Stoelwinder, 1991; Amabile, 1998).
Dimension P1 opposed firms that look at the means (process orientation) and firms that look at the goals (results orientation). Hofstede et al. (1990) made a distinction between these two cultures along the lines of organic vs. mechanistic organizations. According to the model by Quinn (1988), organic firms are the clan type or adhocracy type. This means flexibility, creativity and entrepreneurship (Quinn, 1988). Behavioral traits often linked with exploration and innovation. Mechanistic firms (Hierarchy and market type) are described as goal oriented - in line with dimension P1 -. These firms have strict rules, policies and regulations - in line with dimension P5 -. The firms are stable, predictable and look for competitive advantage (Quinn, 1988) Behavioral traits that can be linked with exploitation. It shows that behavioral traits of different cultural dimensions overlap with either exploration or exploitation.
Looking further at exploration/exploitation, scholars have often linked exploration with search, experimentation and variation, while exploitation focusses on enhancing productivity and efficiency through choice, execution and variance reduction (Lavie et al. 2010). Exploration is the pursuit of kew knowledge while exploration is the use and development of things already known (Levinthal & March, 1993). To see what drives exploration, empirical studies indicate that more environmental dynamism, a stronger organization mission, a prospector orientation and larger slack resources are associated with a greater exploration orientation (Sidhu et al., 2004).
These drivers of exploration and exploitation can be linked with the notion of strong and weak culture. Hofstede et al. (1990) already made the connection between strong cultures and firms that are result oriented and job oriented. Looking at the literature on culture and ambidexterity, the main premise is that strong cultures are inclined to stay within the realm of what is known. They choose the exploitation of known knowledge over the exploration of new innovations. It also hinders the ability to adapt in changing environments and limits the reaction to environmental threats (Sorensen, 2002; Andriopoulos & Lewis, 2009). Strong culture firms display all the behavioral traits that mechanistic, incumbent firm have. Based on these traits, I can conclude that strong culture firms are inclined to exploitation. Weak cultures firms are heterogeneous and loosely coupled. They have the tendency to explore new options and are more nimble. They can react quickly to environmental threats. They have an inclination towards exploration (Sorensen, 2002; Andriopoulos & Lewis, 2009).
Hofstede et al. (1990) already made a link between result and job orientation and strong culture. However, in the literature, I have found several articles that link the behavioral traits of the six cultural dimensions proposed by Hofstede et al. (1990) with weak or strong culture. For example; Strong cultures display enhanced coordination and control within the firm and improved goal alignment. This is often shown by an increased employee effort (Sorensen, 2002). The notion of goal alignment has striking similarities with Hofstede’s results-oriented firm. In the cultural dimension of process orientation vs. results orientation, results orientation is described as a mechanistic system, where individual performance is of high importance (Hofstede et al., 1990). Goal orientation and the shared set of values within a firm can also be linked to the nature of professionalism in the firm. Hofstede discerns parochial firms with professional firms, where professional firms look at job competence and are more formal.
Employees are aware of in-firm competition. A connection can be made with professional firms and strong-culture firm.
The second dimension of Hofstede’s (1990) organizational culture is employee vs. job orientation. Job orientation looks at the work that people do, where decisions are made by individuals and centralized at the top (Hofstede, 1990). Looking at the strong goal alignment in strong-culture firms, corporate goals and practices are set top-down, ensuring that employees face less uncertainty about the proper course of action. Goals and behavior are determined centralized. Again, a clear similarity between job orientation as proposed by Hofstede (1990), and strong-culture firms as proposed by Sorensen (2002).
Strong culture also refers to norms that define appropriate attitudes and behaviors for organizational members. Employees are told how to feel and behave. Sorensen (2002) calls strong organizational culture a form of social control. (Sorensen, 2002). Alvesson & Willmott (2002) sees a trend where managers influence the processes of corporate culture. This seems to correlate strongly with Hofstede’s notion of tight control. The fifth cultural dimension described by Hofstede compares firms with loose and with tight control. Firm’s with tight control employ people that are well groomed, cost-conscious and speak serious of their organization and job (Hofstede et al., 1990). It appears that tight control also is an example of a strong-culture firm.
It shows that Hofstede’s dimensions can be linked to weak-culture of strong-culture firms, meaning that they can be linked to an inclination towards exploration or exploitation. Since each dimension of Hofstede consists of opposites, we can assume that if result orientation links to strong-culture firm, process orientation links to weak-culture firm. This means that result orientation should have an inclination towards exploitation, while process orientation should have an inclination towards exploration. The same principle goes for the other dimension. Based on this premise, I can hypothesize the following:
Hypothesis 1. Firms that score high on process-orientation, employee-orientation,
parochiality, having an open system, having loose control or are pragmatic, have a higher
inclination towards explorative innovation.
Hypothesis 2. Firms that score high on results-orientation, job-orientation,
professionality, having a closed system, having tight control or are normative, have a higher
inclination towards exploitative innovation.
Lastly, Hofstede (1990) looks at practices, instead of values. This gives a clear notion of what a firm actually does on a daily bases (Hofstede et al., 1990). Whether this correlates with the firm’s choice between exploration and exploitation, will show the influences of actual practices on the exploration/exploitation tradeoff. For the exploration/exploitation trade-off, this has yet to be researched.
The moderating role of budget control
An important question that Hofstede et al. (1990) asked, was to what extent can measurable differences among the cultures of different organizations be attributed to unique features of the organization in question, such as its history or the personality of its founder? To what extent do they reflect other characteristics of the organization, like its structure and control systems, which in themselves may have been affected by culture? Hofstede et al. (1990) hypothesized that a firm’s organizational culture is partly related to organizational structure and control systems. Several studies have looked at the relation between culture and control systems (Harrison, 1992; Chow et al, 1999; Ueno and Wu, 1993). However, these studies have
all used the cultural dimensions proposed by Hofstede in his 1980 book. The control systems were tested on an intercultural level, between countries. This study will test the influence control systems on the relationship of culture and exploration/exploitation between firms.
Looking at management control systems (MCS), scholars have found a clear link with ambidexterity (McCarthy & Gordon 2011). One of the earlier strategic scholars, Michael Porter, wrote that a cost leadership strategy requires cost control. Cost leadership of course, had a strong emphasis on minimizing production costs and optimizing efficiency (Porter, 1980). MCS have been named as a deterrent for creativity, a hinder for innovation (Abernethy & Stoelwinder, 1991, Amabile, 1998). However, other scholars found that only formal MCS limits innovation, while informal MCS can encourage innovation (Kaplan & Norton, 1996; Wheelwright & Clark, 1992). What interested me, was the role of formal MCS and its effect on culture and innovation. I therefore decided to look at budget control, a form of formal MCS (Merchant, 1981). Merchant (1981) saw that corporate budget control influences the managerial behavior of a firm. Firms with high corporate budget control were more centralized, had a more formal administration. Budget goals came from centralized organizational commitments and had a focus on performance and results (Merchant, 1981). High budget control was linked by Merchant (1981) with the cultural dimensions of the second hypothesis (result orientation, job orientation, etc.). Managers’ attitudes towards budget control was linked with high scores on ‘improving methods of production’ (Merchant, 1981). Based on this, I see that high budget control strengthens the relation between firms that are result oriented, job oriented, are professional, closed, tight control, are normative, and the inclination towards exploitation.
Ueno and Wu (1993) looked at budget control between Japanese and American firms and used Hofstede’s (1980) cultural dimensions. Based on a 6 item scale, Ueno and Wu (1993) tried to determine if a firm portrays high or low levels of budget control. This paper will test if
budget control moderates the relation between organization culture and exploration/exploitation. The survey created by Ueno and Wu (1993) looked at firm specifics like top-down management, budget excess, managerial budget control. High budget control is enforced by high centralization, high amount of budget slack and high amounts of control (Ueno & Wu, 1993). Looking at Hofstede’s dimensions, we can identify centralized decision making (job oriented), cost-consciousness (tight control), changes through managerial degree (job orientated) or management stingy with little things (closed system). Clearly, high budget control can be linked with cultural dimensions that represent a strong-culture firm. Therefore, high budget control can be linked to an inclination towards exploitation, where low budget control will push towards exploration. Hence:
Hypothesis 3. High (low) budget control MCS has a positive (negative) influence on the
relationship between firms that are result-oriented, job-oriented, professional, have a closed
system, tight control, are pragmatic, and exploitative innovation
Hypothesis 4. High (low) budget control MCS has a negative (positive) influence on the
relationship between firms that are process-oriented, employee-oriented, parochial, have an open system, loose control, are normative, and explorative innovation
Methods
Research setting and data collection
To gather respondents, I identified a company sample of about 1700 firms from ORBIS, a source available to the University of Amsterdam. The companies selected through ORBIS had to comply with the following conditions: The firms had to be Dutch, with a Dutch headquarters.
The firms must have more than 25 employees and must be in business more than 3 years. Secondly, people in my private environment had been contacted through direct mail, Facebook and LinkedIn. By addressing companies and employees at random and through a personal network, I ensured that the sample is representative. It also dealt with the bias of undercoverage of certain types of employees. In the group of respondents - although many unknown - starters, associates, managers and directors are present. I ensured that the respondents conscientious, accurate and aware of the fact that their privacy was held in the highest regard, by assuring them of confidentiality.
The sample group has been contacted both via direct mail, random mail and social media. Therefore, a significant percentage of the responsegroup has use the anonymous link. It is unknown if people contacted via mail used the anonymous link of the personal link. An accurate response rate is therefore unreliable. 88 respondents filled out the questionnaire. Of these 88 respondents, 3 failed to fill out the questionnaire completely, 3 others did not have the required size of at least 25 employees. This left me with 82 valid respondents. These respondents represented a wide variety of firms. Most firms are Dutch, other are multinational firms. In those cases, to determine the company size, the Dutch operations of such a firm have been used.
The age of the respondents is unknown. The average size of the firms was 4421.46 (s.d. = 13104.366) full time employees and an age of 31.77 years (s.d. = 33.811). Concerning multinational firms with a Dutch department; the size of these firms only covered the Dutch departments, not the entire organization (e.g. Nestlé N.L.). The firms were operating in a wide range of industries, which covers manufacturing, transportation, consumer goods, financing, services and other.
To test for common method bias, I have performed Harman’s one factor test on all survey questions regarding organizational culture, exploration/exploitation and MCS. These 45 survey items were comprised in a factor analysis to see what the total variance explained was of the highest factor. In this case, the highest factor explained 17.5% of the variance. Therefore, the survey items fall well below the 50% variance explained of the highest factor to indicate that there is common method bias (see appendix).
Measures and validation of constructs
Dependent variable - exploration/exploitation
The dependent variable is the respective level of exploration and exploitation in a firm. For this, prior studies have been followed. I regard exploration and exploitation as two distinct dimensions of learning behavior, rather than as two ends of a unidimensional scale (Gibson & Birkinshaw, 2004; He & Wong, 2004). I adopted eight Likert-scale items to measure how firms divide attention and resources between innovation activities with explorative versus exploitative objectives in the last three years. These eight items have been adopted from the study done by He & Wong (2004). These items were designed to measure how important it is for a firm to carry out innovation projects to enter new product-market domains or to improve existing product-market efficiency (e.g., introduce new generation of products vs. improve existing product quality; open up new markets versus reduce production cost). From this eight item scale, a four item scale has been developed for explorative innovation ( = 0.82), which α captures the essence of “exploration of new possibilities” (He & Wong, 2004). Another four item scale was developed to measure the level of exploitative innovation ( = 0.78) which captures α the essence of “exploitation of old certainties” (He & Wong, 2004).
Both explorative and exploitative innovation have been transformed to a single score based on the results on the likert-scale questions. To establish the relation between explorative and exploitative innovation on the one hand, and organizational culture on the other hand, various analyses have been performed. An exploratory factor analysis was performed to establish the two factor structure of the exploration and exploitation construct. The eight item scale for explorative innovation and exploitative innovation should ideally display a two factor division with significant correlations between the four explorative questions and factor 1 and the four exploitative questions and factor 2. All factor loadings were above 0.5 (7 factor loadings were actually above 0.7) and cross loadings were - with one exception - below 0.2. The KMO measure verifies the sampling adequacy for the analysis (KMO = 0.722). Bartlett’s test of sphericity χ² (28) = 265.471, p < .001, indicates that correlations between items were sufficiently large (see appendix).
Secondly, exploration and exploitation have been linked with each mean score of the different dimensions of organizational culture, to establish which cultural dimension has a significant relation with either exploration or exploitation.
When determining either high levels of exploration or exploitation in a firm, I can link these results with the results from the firm’s organizational culture response, giving me inside in the levels of the six dimensions of organizational culture at each firm. With both results at hand, I can set up an SPSS database and test the significance of the correlation levels between each of the dimensions and the levels of exploration or exploitation.
Independent variable - Organizational culture
The independent variables are the six dimensions of organizational culture proposed by Hofstede in 1990. Hofstede empirically researched firms in both The Netherlands and Denmark
and identified the six cultural dimensions. He noted that due to the differences in national culture, the dimensions of firm culture can be different or incomplete in other countries (Hofstede, 1990). This study however, will look exclusively at Dutch firm, making the six organizational dimensions of Hofstede a perfect measure.
The six dimensions proposed by Hofstede are all displayed as two counterparts. According to Hofstede, a firm is either process-oriented or result-oriented, employee-oriented or job-oriented, parochial or professional, an open system or a closed system, loose control or tight control and is either pragmatic or normative. The dimensions were established by an exploratory factor analysis done by Hofstede on all 31 questions. The 6 factors that were derived from this analysis are the dimensions by which he measured organizational culture. I have adopted these factors (dimensions) and by measuring each dimension based on several questions displayed on a 1 to 5 likert-scale, the results of the firm’s organizational culture will be displayed as a numerical score. A firm is either more process-oriented or more result-oriented, based on the result of the survey. Each cultural dimension is a singular dimension where each orientation (e.g. process vs. result) is at the opposite end of the other. A low score for P1 will mean that the firm is more process-oriented, while a high score means that the firm is more result-oriented. The six dimensions are coded in such a way, that high scores - let’s call this culture group A - are hypothesized to have a positive effect on exploitative innovation. Low scores - culture group B - should have a positive effect on explorative innovation.
For P1, an eight item scale has been developed (α = 0.84)that considers the cultural aspects of the firm based on their process orientation or result orientation. All other dimensions adapted from Hofstede et al. (1990) show high or acceptable reliability. P2 ( = 0.75), P3 ( = α α 0.69), P4 ( =0.71), P5 ( = 0.69), P6 ( =0.69) all fall within range of Cronbach’s Alpha’s α α α
acceptable range. Since the Cronbach’s Alpha of all dimensions shows reliability, there is no reason for me to change the factors by which Hofstede created his 6 dimensions.
An exploratory factor analysis has been done on each dimension of organizational culture, to determine if factor loadings are high enough and if questions have an eigenvalue of more than one. Since all dimensions are orthogonal, and testing the dimensions with an oblique setting gave low component correlations, the EFA has been set for orthogonal, using the varimax setting. For all dimensions, the KMO measure verified that the analysis had adequate sampling (respectively 0.75, 0.80, 0.67, 0.64, 0.61 & 0.74) and Bartlett’s test of sphericity gave a significance of <0.001 for all six dimensions. P1 and P2 had two factors with an eigenvalue of over 1, while the other factors only had one factor with an eigenvalue of >1. For P1 and P2, the EFA was performed while stipulating that 2 components would be extracted. All components had a factor loading higher than 0.6. For the other four dimensions only had one component with an eigenvalue that was higher than one, which makes information on the factor loadings redundant (see appendix).
Moderating variable - Management Control Systems
An important challenge for this study is to determine under which circumstances the relation between organizational culture and explorative and exploitative innovation exists. For this, I shall moderate the relation with budget control, a form of management control systems (Ueno and Wu, 1993). Management control systems are not only a known antecedent of the exploration/exploitation trade-off (McCarthy & Gordon, 2011; Lavie et al, 2010), but also has a proven relation with organizational culture (Harrison, 1992; Chow et al, 1999; Ueno and Wu, 1993). The moderating variable, management control systems, will be measured based on the amount of budget control found in the firm. Based on 6 statements on a 5-point Likert scale,
ranging from either firmly disagree or firmly agree, the amount of budget control for each firm will be determined. The questions have been adapted from the research done by Ueno and Wu (1993) and have shown a high reliability in this study (α = 0.7)
The exploratory factor analysis has given results that prove an adequate sampling of the analysis. Due to a low score on component correlation, I choose to use the orthogonal setting of the EFA. KMO’s test have a score of 0.61, with Bartlett’s test of sphericity χ² (15) = 118.448, p < .001, indicating that correlations between items were sufficiently large. Two components had an eigenvalue of more than 1. All but one components had a factor loading of more than 0.74 (see appendix)
Control Variables
In this study, I have controlled the relation between culture and exploration/exploitation for possible alternative explanations. By adding relevant control variables in the regression analysis, I can eliminate alternative explanations. Firstly, the size of the firm. Since larger firms on the one hand have more resources to explore, but also lack flexibility to explore, I included firm size to the control variables. Secondly, the age of the firm. Incumbent organizations are often more inclined towards exploitation (Gilbert, 2005), therefore I have controlled for age. Third, I looked at environmental factors, such as environmental dynamism. These often affect firms in their exploration/exploitation trade-off (Floyd & Lane, 2000). Lastly, I shall control for structural differentiation. Jansen et al. (2006, 2009) have shown the effect of structural differentiation on a firm’s ambidexterity.
Analysis and results
The analysis of the survey results has been done through several forms of analysis. Firstly, table 1 shows the correlation matrix, including the descriptive statistics and Cronbach Alpha’s. The correlation matrix already shows a number of interesting correlations between variables. The two dependent variables, explorative innovation and exploitative innovation, correlate with a number of independent variables and control variables. All but one independent variables have a correlation with either one of the dependent variables.
Table 2 and 3 shows the results of the hierarchal multiple regression analysis performed on the data for both dependent variables. Both regression analysis had shown an acceptable level of multicollinearity. The tolerance for collinearity was between .89 and .93, which is well above the cutoff point of 0.10. The independent variables have low multicollinearity, therefore there is no disturbance in interpreting the beta’s (Cortina, 1993).
The regression analysis looks at the relation between the dependent variables - explorative innovation and exploitative innovation - and the six dimensions of organizational culture. As indicated earlier, exploration and exploitation have been separated in individual variables. The eight item survey constructed by He & Wong (2004) consisted of 4 questions concerning explorative innovation and 4 questions concerning exploitative innovation. The six cultural dimensions are coded differently. As stated before, a high score on P1 to P6 means the firm is more inclined to culture group A, while a low score means an inclination to culture group B. Hypothesis 1 states that high scores on process, employee, etc (culture group B) will lead to higher explorative innovation. Hypothesis 2 states the opposite, high scores on culture group A will lead to higher exploitative innovation. To confirm the first two hypothesise, there must be a positive significant relation between P1 to P6 and exploitative innovation. A negative significant
relation must be present between P1 to P6 and explorative innovation, since a low score on P1 to P6 means an inclination to culture group B.
The regression analysis performed on explorative innovation shows a positive relation between P1 and P6 and exploitative innovation (β = 0.31, p < 0.01; β = 0.36, p < 0.01), meaning that exploitative innovation will increase if a firm portrays higher levels of result-orientation and normative behavior. The regression analysis also shows a positive relation between P1 and explorative innovation (β = 0.32, P < 0.01). However, if we interpret this relation more closely, it shows that an increase in the score of P1 will also increase the level of explorative innovation. This means that a higher score on result-orientation will lead to an increase in explorative innovation. Looking back at the literature, result-orientation was linked with strong-culture firm and a mechanistic system. Such firms remained in the realm of what was known, favouring exploitation in stead of exploration (O’Reilly & Chatman, 1996). This result is therefore highly counterintuitive. For both explorative and exploitative innovation, there were no other cultural dimensions with a significant relation.
The control variables show one interesting - yet also counterintuitive - result. Jansen et al. (2009) researched the effect of structural differentiation on ambidexterity, and found that this relation was positive and significant. This study shows that structural differentiation has a negative and significant relation with both explorative and exploitative innovation, meaning that the higher a firm scores on structural differentiation, the lower the firm’s inclination to both exploration and exploitation become.
The last step of this study was to moderate the relation between organizational culture and exploration/exploitation with budget control. Andrew F. Hayes’ (2013) Process Analysis found one moderating effect of budget control. The relation between P1 and explorative innovation is significantly moderated by budget control. The regression coefficient for XM is (C3
= 0.4) and is statistically different from zero, t(78) = 2.57, P<0.05. Thus, the effect of a firm’s inclination towards process-orientation or result-orientation on the level of explorative innovation depends on the amount of budget control present in the firm. This model accounts for 19% ( ^2 R = 0.19) of the variance in explorative innovation. As seen on graph 1, the effect of high result orientation on explorative innovation will increase when a firm has high budget control. This effect also happens to be counterintuitive. According to the literature, stringent management control systems have a negative effect on exploration (McCarthy & Gordon, 2011).
Graph 1 - Moderating effect of budget control on result orientation and explorative innovation.
Based on the analysis of the results, I can conclude that the first hypothesis is not supported. Five out of six cultural dimensions did not have a significant relation with explorative innovation. The sixth cultural dimension that was significant, had a counterintuitive result. High scores on process-orientation, employee-orientation, parochiality, having an open system
explorative innovation. The second hypothesis is partially supported. Out of the six cultural dimensions, two had a significant relation which was in accordance with the hypothesis. Result-orientation and being normative leads to a higher inclination towards exploitative innovation. Professionality, a closed system and tight control also lead to higher exploitative innovation, yet these relations were not significant.
This study does not support the third and fourth hypothesis. Budget control systems do not moderate the relation between organizational culture and exploitative innovation. Budget control also displays no effect on the relation between organizational culture and explorative innovation. The only relation that was moderated by MCS was between process vs. result orientation and explorative innovation. However, this moderating effect was the opposite of the hypothesized effect of the moderator.
Discussion & Conclusion
A significant amount of scholars have looked at the phenomenon of exploration and exploitation within firms. The literature of exploration and exploitation - or ambidexterity - covers a wide range of subjects. This paper has been in line with the scholars that considered exploration and exploitation as two individual and detached phenomena (e.g. March, 1991; Benner & Tushman, 2002; Lavie et al. 2010). Like the literature that regards exploration and exploitation to be a trade-off, this study also kept the two phenomena separate. But rather than stipulating that a firm needs to make a trade-off between the two, this study considers firms to have a choice between exploration and exploitation.
This study builds on the previous assertions that culture and exploration/exploitation are linked. But unlike previous scholars like Sorensen (2002), Alvesson & Willmott (2002) or Andriopoulos & Lewis (2009), this study has looked beyond the notion of weak and strong
cultures. It has used the cultural dimensions, specified for organizations, constructed by Hofstede et al. (1990). Strong cultures have a set of shared beliefs and constrain the firm to stay in the realm of what is known. Strong culture firms lack the ability to observe and respond to environmental threats, they are less nimble than weak culture firm, who are more adaptive in such situations (Andriopoulos & Lewis, 2009). This study tried to expand the literature, by specifying exact cultural traits present in each firm, that could leads to a higher inclination towards exploration and exploitation. Of the six cultural dimensions used in this study, two had a significant effect on exploitation, only one had a significant effect on exploration. Unfortunately, this last effect was the opposite of the hypothesized effect. Regarding exploitation, I can therefore argue that the amount of result orientation and normative behavior of a firm influences the inclination towards exploitative innovation. The other four cultural dimensions showed no significant relation. My study contributes to previous studies that researched the relation between culture and exploration/exploitation (e.g. Miles & Snow, 1978; Tripsas, 2009). It has shown that organizational culture can be specified further, with regards to its relation with exploration/exploitation. I suggest that future research builds on this knowledge and expands the research on result orientation and normative behaviour and exploitation. Before result orientation and normative behaviour can be added to the list of antecedents of exploration/exploitation, this research must be repeated and expanded.
Regarding the counterintuitive effect of result orientation on exploration, the literature offers some explanation. The hypothesized effect of P1 (process vs. result orientation) on exploration was that process orientation would have a positive relation on explorative innovation. Process orientation was linked with weak culture, nimble firms and organic systems (Sorensen, 2002). Organizational treats that in turn are linked with exploration (Jansen et al. 2006). However, the study showed that not process, but result orientation increases the
inclination towards exploration. This could be explained by the failure trap. Firms lacking success often focus on a frantic search of new innovations, by which the firm can regain success. The more a firm fails, the more it tends to look for new innovation (Levinthal & March, 1993). A firm that is result oriented, is likely to be more sensitive to success and failure traps, since performance is a main objective. If a high amount of result oriented firms that participated in this study had disappointing results in the last years, it could explain the high inclination towards exploration. Since this study did not control for recent performance, this is a potential study for future scholars.
It is interesting that my study does not support the third and fourth hypothesis, regarding the moderating role of budget control. Scholars had shown the limiting effect of formal MCS on innovation (Abernethy & Stoelwinder, 1991; Amabile, 1998; Miles & Snow, 1978), and following the logic of Merchant (1981), formal MCS should strengthen the relation between culture group B and exploitative innovation. However, the results only showed one significant moderating effect of budget control on the relation between culture and exploration/exploitation. This relation was a positive moderating effect between dimension P1 and explorative innovation, implying that high budget control would strengthen the relation between result-orientation and explorative innovation. This effect is counterintuitive since the positive relation between result-orientation and explorative innovation was not hypothesized (although it was found in the results). Secondly, the hypothesized effect of high budget control on the relation between culture and explorative innovation was negative. Higher formal MCS was - according to the literature - a deterrent for innovation and creativity (Abernethy & Stoelwinder, 1991; Amabile, 1998; Miles & Snow, 1978). These counterintuitive results can be explained when we look at Simons (1990; 1991). He claimed that interactive use of formal and informal MCS could be stimulating for creative ideas and product development. “Management controls become
interactive when business managers use planning and control procedures to actively monitor and intervene in ongoing decision activities of subordinates” (Simons, 1990). Instead of programmed controls, interactive controls use interventions that gives the opportunity to debate the control system. Simons (1990) called this ‘levers for control’. This theory has been tested by Bisbe & Otley (2004), who did a survey based research. Their results partially confirmed Simons’ theory. Only in low innovating firms does interactive control systems enhance innovation. High innovating firms are negatively influenced by interactive control systems (Bisbe & Otley, 2004). Since my study did not control for level of innovativeness at the respondents, nor did it specify if the management control system was interactive or not, these two explanations could explain the counterintuitive results of this study. There is a lot of potential for future research in the moderating effect of MCS on culture and innovation. Future researchers can differentiate between formal and informal, interactive and non-interactive, to see which types of MCS actually influence the relation between culture and innovation.
Regarding the negative effect of structural differentiation on both explorative and exploitative innovation, our study goes against the literature. Jansen et al. (2009) showed the positive effect of structural differentiation on ambidexterity. This study shows the complete opposite. The same goes for environmental dynamism and explorative innovation. The most logical explanation is that this study has used non-validated scales for the two variables, which makes these results unreliable.
Limitations and future research
This study has set a first step in using Hofstede’s dimensions of organizational culture and compared it with exploration and exploitation. Future researchers can pick up Hofstede’s dimensions to further pinpoint the effect of organizational culture on different facets of strategic
literature. This study has tried to expand the list of antecedents that influence the choice of exploration and exploitation. Future researchers can look at the cultural dimensions that had a significant relation with exploration/exploitation and further examine this relation. They can also look at the other dimensions of organizational culture, to see if - under different circumstances - they can show a significant relation. In doing this, future researchers can overcome some limitation that this study has endured. One of them was the lack of negatively-keyed items in the questionnaire. This yea-saying bias can have influenced the respondents in answering questions positively. Some questions have also been too leading. A low score on some questions carried a negative load, which can have influenced the respondents. Off course, future research can enlarge the group of respondents, to increase the significance of the results.
Appendices
Appendix 1. Items and measures
P1 - process vs. results orientation (Hofstede et al. 1990) Employees are told when they did a good job The average worker is a fast worker
The average worker tries to be a pionier
The average worker has a direct way of communicating and dealing with problems The average worker performs at maximum effort
Mistakes are tolerated We are open to outsiders
Managers help good employees make steps in the organization P2 - job vs. employee orientation (Hofstede et al. 1990)
Important decisions are made by individuals
Management is only interested in the work done by employees Decisions are made centralized in the top of the firm
Managers try to retain good employees at their department Changes are implemented by decreet of the manager
Little attention is paid to the personal problems of employees P3 - parochial vs. professional (Hofstede et al. 1990)
Nobody interferes with employees private life
Competence for the job performed is the only hiring criterium Employees are aware of in-firm competition
Cooperation and trust between departments is considered normal P4 - open vs. closed system (Hofstede et al. 1990)
Only special people fit in
Management reacts heavily on little issues
Little attention is paid to the physical work environment New employees need a long time to adjust
P5 - loose vs. tight control (Hofstede et al. 1990) Employees are very cost conscious Employees are very punctual
The average employee is well groomed and dressed
The average employee speaks seriously about the firm and the job P6 - pragmatic vs. normative (Hofstede et al. 1990)
The average employee is pragmatic in ethical matters Little is contributed to society
Emphasis is laid on fulfilling customer needs Result is more important than procedure
Explorative innovation (He & Wong, 2004) Introduction of new generation products Expansion of existing product range Penetrating new markets
Development of new production techniques Exploitative innovation (He & Wong, 2004)
Increase yield / reduce material consumptions Improve production flexibility
Reduce production costs Improve existing product quality
Management control system, budget control (Ueno & Wu, 1993)
Communication and coordination is done top-down by top management Long term planning is very important to the firm
A lot of emphasis is laid on the structure of the budgeting process There is often a budget surplus
Budget is often checked by top management
A lot of emphasis is laid on long term results, instead of short term results
Appendix 3 - Exploratory factor analysis exploration/exploitation
Appendix 4 - Exploratory factor analysis dimension P1
Appendix 5 - Exploratory factor analysis dimension P2
Bibliography
Abernethy, M. A., & Stoelwinder, J. U. (1991). Budget use, task uncertainty, system goal orientation and subunit performance: A test of the ‘fit’hypothesis in not-for-profit hospitals. Accounting,
Organizations and Society, 16(2), 105-120.
Alvesson, M., & Willmott, H. (2002). Identity regulation as organizational control: Producing the appropriate individual. Journal of management studies, 39(5), 619-644.
Amabile, T. M. (1998). How to kill creativity (Vol. 87). Boston, MA: Harvard Business School
Publishing.
Andrews, S. B., Basler, C. R., & Coller, X. (1999). Organizational structures, cultures, and
identities: Overlaps and divergences. Research in the Sociology of Organizations, 16, 213-235.
Andriopoulos, C., & Lewis, M. W. (2009). Exploitation-exploration tensions and organizational
ambidexterity: Managing paradoxes of innovation. Organization Science, 20(4), 696-717.
Bass, B. M., & Avolio, B. J. (1993). Transformational leadership and organizational culture. Public
administration quarterly, 112-121.
Baum, J. A., Li, S. X., & Usher, J. M. (2000). Making the next move: How experiential and vicarious learning shape the locations of chains' acquisitions. Administrative Science Quarterly, 45(4), 766-801.
Benner, M. J., & Tushman, M. L. (2003). Exploitation, exploration, and process management: The productivity dilemma revisited. Academy of management review, 28(2), 238-256.
Bisbe, J., & Otley, D. (2004). The effects of the interactive use of management control systems on product innovation. Accounting, organizations and society, 29(8), 709-737.
Bozeman, B., & Kingsley, G. (1998). Risk culture in public and private organizations. Public
Administration Review, 109-118.
Benner, M. J., & Tushman, M. (2002). Process management and technological innovation: A longitudinal study of the photography and paint industries. Administrative science quarterly, 47(4), 676-707.
