Performance feedback and organizational risk-taking behavior : the moderating effect of organizational culture.

(1)

Master Thesis

Performance Feedback and Organizational Risk-Taking Behavior.

The Moderating Effect of Organizational Culture.

Student: Rosanne Claire Maria Somers / Student No 11423129

MSc. Business Administration, Strategy track University of Amsterdam, Faculty

of Economics and Business

Supervisor: Silveira Barbosa Correia Lima, Bernardo

(2)

1

Statement of originality

This document is written by Student Rosanne Somers who declares to take full responsibility for the contents of this document.

I declare that the text and the work presented in this document is original and that no sources other than those mentioned in the text and its references have been used in creating it.

The Faculty of Economics and Business is responsible solely for the supervision of completion of the work, not for the contents.

(3)

2

Abstract

Poor performance indicates that the routines and processes of the organizations are not suitable for its environment and triggers decision-makers to search for solutions. However, the results of how this process influences organizational risk-taking behavior conflict. Empirical evidence suggests that some managers tend to take more risk when they face performance shortfalls, whereas others, become more risk-averse. These contradicting results are interesting, since a certain level of risk-taking is necessary to enhance performance, to adjust to the environment and to stay competitive. This study examines organizational culture that moderates risk-taking induced by performance shortfalls, because organizations and managers tend to react to challenges and changes by intensifying their core cultural values, which impacts decision-making. The moderating effect is based on two dimensions of organizational culture: focus (external vs. internal orientation) and structure (flexible vs. control orientation). First, this study argues that external-oriented firms are more risk-taking than internal-oriented firms. Second, this study proposes that flexible-oriented firms are more risk taking than control-oriented firms. Empirical results, of panel data from 1979 to 2004 on the R&D spending of U.S. manufacturing firms based on two-way fixed linear regression, show that external-oriented firms decrease risk-taking behavior when performing below organizational aspirations, whereas internal-oriented firms increase risk-taking behavior when underperforming their aspirations. Flexible-oriented and control-oriented firms both take less risks when performing below aspiration levels. However, control-oriented firms restrict their activities more than that flexible-oriented firms do. These findings largely contradict the predictions and suggest that distinctive organizational cultures employ different risk-taking behaviors in response to performance shortfalls than expected. The

(5)

4 results contribute to the theories of organizational learning, the behavioral theory of the firm and theories of organizational culture.

Key words: Organizational Culture, Focus, Structure, Performance Feedback, Organizational Risk-Taking, Behavioral Theory

(6)

5

Introduction

Organizational learning and performance feedback are guided by the behavioral theory of the firm (Cyert & March, 1963; March & Shapira, 1987, 1992). The main argument of the behavioral perspective is that companies or decision-makers set organizational aspiration levels to evaluate performance, which in turn enables them to adjust their behavior with the aim to improve their performance (Cyert & March, 1963; Kim, Finkelstein, & Haleblian, 2015; Levinthal & March, 1993; Levitt & March, 1988). However, this relation is to this day still not completely explained by the literature.

Companies set their aspiration levels of performance in comparison to own historical performance or industry peers (Cyert & March, 1963). When firms perform below their organizational aspiration levels, managers are triggered to engage in problemistic search to identify alternative courses that may improve firm performance (Cyert & March, 1963). On the other hand, firms that perform congruent or above their organizational aspiration levels reduce their problemistic search activities. In this situation managers will proceed with their current and already to be proven effective routines and strategies (Bromiley, Miller, & Rau, 2001; Greve, 2003). This is interesting, because individuals and firms learn from experience and change their behavior accordingly, in order to adapt to their environment, to improve performance and to remain competitive (Desai, 2008; Greve, 2003; Wiseman & Bromiley, 1996).

Some authors have argued that performance feedback also influences risk taking behavior (Audia & Greve, 2006; Desai, 2008; Jordan & Audia, 2012; Kim et al., 2015; March & Shapira, 1987; Miller & Chen, 2004; Singh, 1986). However, the findings on what the effects of performance below organizational aspiration levels on risk-taking behavior are inconsistent, leading to a variability in risk-taking behavior. (Kacperczyk, Beckman, & Moliterno, 2015; Kim

(7)

6 et al., 2015). Some studies state that when firms, performing below their organizational aspiration levels, tend to take more risks, whereas others, argue that they tend to become more risk-averse (Audia & Greve, 2006; Greve, 2003; Iyer & Miller, 2008; Wiseman & Bromiley, 1996). Moreover, studies found that firms have contradictory behavior to performance below aspirations because of distinctive organizational determinants (Audia & Greve, 2006; Chen & Miller, 2007; Desai, 2008; Iyer & Miller, 2008; Kim et al., 2015; Vissa, Greve, & Chen, 2009). Thus, the painting of the total landscape of the variability of risk-taking behavior determined by performance below aspirations is still unfinished. Hence, in this research the brush is picked up again to continue the painting.

Recent research began to recognize the role of moderators that influence performance feedback on risk-taking behavior to create a richer understanding of the phenomena (Lim & McCann, 2014). The reason behind this comes from the fact that risk-taking is sensitive to the risk-taker’s fortune (March & Shapira, 1992). In other words, decision-makers perceive negative performance and the need for risk-taking actions, based on whether the gap is repairable or seen as a threat to the survival of the firm. (Audia & Greve, 2006). Therefore, research started to focus on these boundary conditions. For instance, Audia and Greve (2006) showed that performing below aspirations lowers risk-taking behavior in small firms, but had no significant effect or increases risk-taking behavior in large firms. Desai (2008) found that when firms experience a discrepancy between their set aspiration level and their actual performance, its risk-taking behavior depends on its organizational context (i.e., operating experience, organizational legitimacy and age). Vissa et al. (2009) discovered that organizational form (affiliated vs. unaffiliated business group firms) influence performance feedback and the type of search. Kim et al. (2015) found that prior performance variability affects the relationship between organizational

(8)

7 aspiration levels and risk-taking behavior. Performing below aspiration levels strengthens the relationship, while performing above aspiration levels weakens the relationship. These studies propose that gaining more insight into the relationship between performance feedback and organizational risk-taking behavior relies on researching the moderating factors affecting this relationship (Lim & McCann, 2014). Therefore, it is essential to consider the organizational conditions that influence the relationship. By introducing organizational culture, this research zooms in on a new moderator to explain the variability of firm risk-taking behavior.

Organizational culture has not been considered by previous research on performance feedback, even though it is an important factor for a firm’s competitive advantage, since organizational culture has a significant impact on decision-making (Braunscheidel, Suresh, & Boisnier, 2010). Also, organizations and managers tend to react to challenges and changes by intensifying their core cultural values and giving more prominence to their organizational culture (Cameron, 2008). An organization’s culture reflects core values, leadership and management style, goals, procedures, routines and governs how firms behave unconsciously (Denison, 1990; Schein, 1996). Organizational culture is taken-for-granted assumptions and influences how it perceives, thinks about and reacts to internal and external environmental cues (Schein, 1996).

In this research, organizational culture is conceptualized based on the Competing Values Framework (CVF) originally developed by Quin and Rohrbaugh (1983) to identify the key factors of organizational effectiveness. Cameron and Quinn (1999), however, refined the model to measure the organizational culture construct (Braunscheidel et al., 2010; Eisend, Evanschitzky, & Gilliland, 2016; Naranjo-Valencia, Jiménez-Jiménez, & Sanz-Valle, 2011; Porcu, Barrio-García, Alcántara-Pilar, & Crespo-Almendros, 2017; Reis, Trullen, & Story, 2016). The model describes organizational culture along two dimensions: structure and focus (Quinn & Rohrbaugh,

(9)

8 1983). Structure indicates whether firms are flexible and discrete or stable and controlled. Focus looks whether firms are focusing inwards or outwards. Based on these dimensions, four distinctive and opposing types of organizational cultures are defined: market, adhocracy, clan and hierarchy (Braunscheidel et al., 2010; Eisend et al., 2016; Naranjo-Valencia et al., 2011; Porcu et al., 2017).

This research focuses on the moderating effect of organizational culture, based on the two dimensions of the CVF. This research proposes that managers from firms with a flexibility- oriented structure and external-oriented focus will be willing to take more risk by investing more in R&D and explore new alternatives, when performing below aspiration levels. Managers from firms with a stability and control-oriented structure and internal-oriented focus will experience the performance shortfall more as a challenge and focus on improving current operational practices.

Thus, each organizational culture has its own organizational conditions, unique way of doing business, strategic focus and decision-making tactics that affect how firms react to internal and external cues (Braunscheidel et al., 2010; Eisend et al., 2016; Naranjo-Valencia et al., 2011; Porcu et al., 2017). When facing challenges organizations turn to their core values and emphasize their organizational culture more (Cameron, 2008). This results in different interpretation of performance causing distinctive organizational risk-taking behaviors when performing below organizational aspiration levels (Lim & McCann, 2014). However, this has not been examined in the literature so far. Therefore, this study will provide new insights in the conflicting views of the effects of performance below organizational aspiration levels on risk-taking behavior moderated by heterogeneous organizational cultures.

(10)

9 This research makes four contributions to the existing literature. First, it contributes to the behavioral theory of the firm that assumed that organizational determinants did not affect positive or negative differences in performance. The theorizing and empirical findings of this research confirm, however, that organizational determinants do matter, specifically that these effects depend on the dimensions focus and structure of organizational culture. This research thus enhances our understanding about the applicability of the scope of the behavioral theory of the firm.

Second, this study contributes to a richer understanding of the implications of performance feedback, demonstrating an interaction of organizational culture (focus and structure) with performance feedback. The joint effect illustrates a better understanding of how organizational culture matter and how this determinant effect firm strategic risk by interacting with performance feedback.

Third, this research contributes to the literature stream of organizational risk-taking behavior by addressing the impact of organizational culture (focus and structure) on a firm’s risk-taking. It adds value by the development of a richer understanding how focus and structure influence the decision-making process in addition to other organizational determinants.

The fourth contribution is to the research stream of organizational culture, that has begun to address the impact of organizational culture on performance, new product development, innovation and supply chain integration (Braunscheidel et al., 2010; Eisend et al., 2016; Naranjo-Valencia et al., 2011; Sugita & Takahashi, 2015). This research contributes to a richer understanding of the role of organizational culture in the decision-making process and its influence on risk-taking behavior by operationalizing organizational culture and to interact it with performance feedback.

(11)

10

Theory and Hypothesis

The first section introduces the phenomena of performance feedback and how it influences organizational risk-taking behavior. The second section outlines the conflicting theory regarding risk taking behavior caused by performance below organizational aspiration levels. Subsequently, the different determinants that moderate the effect of firm performance on risk-taking behavior, and how organizational culture is contributing to the set of moderators are highlighted. The third section is concerned with the effect of performance feedback on R&D and how this relation is affected by the moderating effect of organizational culture.

Organizational Risk Taking Behavior and Performance Feedback

Significant research has been done on performance feedback and the effects it has on organizational behavior, including organizational learning behavior (Levitt & March, 1988), innovative behavior (Chen & Miller, 2007; Greve, 2003; Lucas, Knoben, & Meeus, 2018; Parker, Krause, & Covin, 2017) and risk-taking behavior (Audia & Greve, 2006; Desai, 2008; Jordan & Audia, 2012; Kim et al., 2015; March & Shapira, 1987; Miller & Chen, 2004; Singh, 1986). Therefore, the behavioral theory of the firm is the underlying concept of this research. This theory states that managers are boundedly rational and, therefore, have incomplete information. Decision-makers attempt to simplify decision heuristics of performance to the discrete measure of failure and success to save the amount of information processing needed to come to conclusions and make decisions (Greve, 2003; Mezias, Chen, & Murphy, 2002). To manage this limitation in the decision-making process, managers try to assess the effectiveness of their past actions and decisions (i.e. learn from their experience) (Greve, 1998; Jordan & Audia, 2012; Mezias et al., 2002; Miller & Chen, 2004). Accordingly, they need to decide if their performance

(12)

11 was sufficient. To do this, managers compare their aspiration targets to the actual performance (Cyert & March, 1963).

Accordingly, firms learn from past performance and adapt their activities depending on the discrepancy between their set performance aspiration and their actual performance (Cyert & March, 1963). This process is at the same time complicated by the simultaneously adapting behavior of other organizations within the industry, and by endogenous changing environments (Levitt & March, 1988). The reason firms set aspiration levels is to evaluate organizational goals, targets and their performance (Cyert & March, 1963). Organizational aspiration level can be defined as a desired performance level of specific organizational outcomes. It determines whether the achieved performance is considered a success or a failure (Shinkle, 2012). One can also refer to them as goals or reference points (Shinkle, 2012). Thus, aspiration level refers to “the smallest outcome that would be deemed satisfactory by the decision maker” (Schneider, 1992, p.1053).

Organizational aspiration levels can be determined by two forms of reference points that both could evaluate achieved firm performance; historical and social aspiration level (Cyert & March, 1963). Historical aspiration levels are based on the comparison of the firm’s own performance history, also referred to as historical comparison (Cyert & March, 1963; Levinthal & March, 1993). Social aspiration levels are based on the comparison of the firm’s performance regarding peer group firms, also known as social comparison (Cyert & March, 1963). Thus, firms learn from performance feedback by evaluating their performance based on the methods wherein historical and social aspiration levels are set. Subsequently, firms compare the achieved performance with their organizational aspiration level and determine whether they need to take

(13)

12 action depending on the discrepancy between the actual performance and its aspiration level (Cyert & March, 1963; Greve, 2008; Jordan & Audia, 2012).

Failure to achieve the organizational aspiration levels triggers managers to engage in problemistic search, that is, “search that is stimulated by a problem … and is directed toward finding a solution to that problem” (Cyert & March, 1963, p.121). According to the behavioral theory of the firm, there are three underlying assumptions of problemistic search: (1) problem solving is encouraged by managers in areas that are important for them, (2) problem solving is formed by the information available to managers to make decisions and their experience, and (3) solutions to problems are sought near the symptoms (Gaba & Joseph, 2012). These assumptions partly explain the distinction between firm responses to performance problems, but not completely (Gaba & Joseph, 2012). Thus, managers pursuing problemistic search to identify an alternative course for the firm may improve firm performance and arrive at an outcome that is above the organizational aspiration level (Greve, 2008). However, it is unclear what solutions managers should pursue and implement to address the problem (Cyert & March, 1963). Adopting new routines or strategies like engaging in acquisition or invest in new product innovations to increase growth and improve performance are possible alternatives, according to organizational learning theory, self-enhancement theory, prospect theory and threat of rigidity theory (Audia & Greve, 2006; Bromiley et al., 2001; Chen & Miller, 2007; Iyer & Miller, 2008; Jordan & Audia, 2012; Miller & Chen, 2004). Additional research, however, could be conducted to complement these theories and find alternative solutions.

Performance above organizational aspiration level, on the other hand, decreases problemistic search, because firms do not wish to change their effective strategic routines and operations (Bromiley et al., 2001; Greve, 2003). In addition, firms avoid activities that might

(14)

13 result in decreasing performance that may fall below aspiration levels (March & Shapira, 1987). Also, pursuing strategies, routines and operations that already have been proven to be effective is more efficient than exploring new alternatives. This is because firms can trust and benefit from investments in prior skills, routines and techniques (Kim et al., 2015). Thus, managers attempt to increase performance above organizational aspiration levels, however, they take less risks if the firm is already performing above its aspiration level (Fiegenbaum, Hart, & Schendel, 1996; Greve, 2003; Iyer & Miller, 2008; Miller & Chen, 2004).

Furthermore, some authors argued that performance feedback influences organizational risk-taking behavior (Audia & Greve, 2006; Desai, 2008; Jordan & Audia, 2012; Kim et al., 2015; March & Shapira, 1987; Miller & Chen, 2004; Singh, 1986). Risk is defined as the “variation in the distribution of possible outcomes, their likelihood and their subjective values” (March & Shapira, 1987, p.1404). Risk is also measured as the variance of the probability distribution of possible gains and losses in comparison to an alternative (Arrow, 1965; Pratt, 1964). Variability in outcomes is one of the main elements of risk (Sanders & Hambrick, 2007). New product development could lead to risk for the firm (Greve, 2003), whereas other changes might not bear that much risk, because they involve incremental adaptations or imitation of well-established practices and strategies (David J. Ketchen & Palmer, 1999; Massini, Lewin, & Greve, 2005; Schwab, 2007).

According to decision making theory, choice involves a trade-off between expected returns and risk (March & Shapira, 1987). Managers prefer larger expected returns to smaller ones, other factors (e.g. risk) being constant (Lindley, 1971 in Kacperczyk, Beckman, & Moliterno, 2015). In addition, managers favor smaller risks over larger ones, other factors (e.g. expected value) being constant (Arrow, 1965). Thus, it is assumed that expected value is

(15)

14 positively and risk is negatively related to the attractiveness of the alternative (March & Shapira, 1987).

However, managers perceive risk differently and less precisely than outlined in the decision-making theory, because: (1) uncertainty about positive outcomes is not threatened as an important aspect of risk, (2) risk is not primarily a probability concept and uncertainty is a factor in risk, and (3) managers seek precision in estimating risk rather than using quantities (March & Shapira, 1987). Thus, risk is reflected as downside risk, meaning the probability that a decision would result in an outcome below the aspiration level (March & Shapira, 1987).

Many researchers combined performance feedback theory with organizational risk-taking behavior theory and decision-making theory and argue that performance feedback will affect firm behavior. Performance below aspiration level triggers problemistic search and increases risk-taking behavior, whereas performance above aspiration level decreases risk-risk-taking behavior (Audia & Greve, 2006; Boyle & Shapira, 2011; Bromiley, 1991; Lim & McCann, 2014; Miller & Chen, 2004). However, the results on what the effects of performance below organizational aspiration levels are on taking behavior are inconsistent, leading to a variability in risk-taking behavior.

Conflicting Theory Regarding Risk Taking Behavior Following Poor Performance

Although extensive research is already present about the relationship between performance feedback and risk-taking behavior, researchers still lack a complete understanding of this relationship, since they created conflicting predictions and received different empirical findings of these two perspectives. On the one hand, research showed that firms take more risk when they perform below their organizational aspiration levels in order to overcome the

(16)

15 performance gap (Audia & Greve, 2006; Boyle & Shapira, 2011; Bromiley, 1991; Miller & Chen, 2004; Wiseman & Bromiley, 1996). On the other hand, research also found that when firms perform below their organizational aspiration levels they become more risk-averse. These firms become reluctant to change and restrict activities (Audia & Greve, 2006; Wiseman & Bromiley, 1996). However, other research showed partial results due to different moderating factors of organizational elements (Chen & Miller, 2007; Desai, 2008; Iyer & Miller, 2008) or the focus on historical and social aspirations (Kacperczyk et al., 2015; Kim et al., 2015). These conflicting results are largely based on whether decision-makers perceive negative performance as a repairable gap or as a threat to firm survival. This is interesting because firms need some level of risk-taking to adapt to their environments, improve performance, and stay competitive, although the relationship varies across studies (Desai, 2008; Greve, 2003; Wiseman & Bromiley, 1996). According to March and Shapira (1992), managers facing organizational shortfalls may focus on survival and become more conservative in their risk-taking behavior as the level of survival is approached, since risk could also lead to failure during performance repair.

To explain the conflicting results regarding risk-taking behavior, recent research began to recognize the importance of the effect of the moderators that influence performance feedback on risk-taking behavior to create a richer understanding of the phenomena (Lim & McCann, 2014) and to determine the risk-taker’s responsiveness to performance below organizational aspiration levels (March & Shapira, 1992). This means that if an organization’s performance drops below its aspiration level, it will either become more risk-averse or risk-taking. For instance, Audia and Greve (2006) showed that performing below aspiration level lowers risk-taking behavior in small firms, but it has no significant effect or increases risk-taking in large firms. Desai (2008) found that when firms experience a discrepancy between their set aspiration level and their actual

(17)

16 performance, its risk-taking behavior depends on its organizational context (i.e., operating experience, organizational legitimacy and age). Vissa et al. (2009) discovered that organizational form (affiliated vs. unaffiliated business group firms) influence performance feedback and the type of search. Kim et al. (2015) found that prior performance variability affects the relationship between organizational aspiration levels and risk-taking behavior. Performing below aspiration levels strengthens the relationship while performing above aspiration level weakens the relationship.

Thus, prior research found that organizational contexts can partly explain the inconsistent findings, however, additional work is needed to examine whether and how other organizational characteristics affect risk-taking across organizations to explain the risk-taking variability among firms (Audia & Greve, 2006; Desai, 2008; Greve, 2003). Therefore, to further enrich the landscape of organizational learning theory, especially that of performance feedback theory in relation to risk-taking behavior theory, one needs to consider other organizational determinants that influence the relationship between performance feedback and risk-taking behavior. In this next section, it will be examined how organizational culture, an extensively studied concept in other domains, can influence attention focus and differentiate between risk taking and risk aversion as a response to performing below organizational aspiration levels. Thus, the research question of this study is:

What is the moderating effect of organizational culture on the relationship between performance below organizational aspiration level and risk-taking behavior, in terms of changes made in research and development (R&D) investment?

(18)

17 Figure 1. Theory Through a Path Model

Moderating effect of Organizational Culture

To date, research has not accounted for intangible firm heterogeneity. Even more so, it has not taken into account differences in organizational culture that influences the relationship between performance feedback and organizational risk-taking behavior. Whereas an organization’s culture is an important determinant of a firm’s competitive advantage and it has a significant effect on decision-making (Braunscheidel et al., 2010). According to Cameron and Quin (1999): “the reason organizational culture was ignored as an important factor in accounting for organizational performance is that it encompasses the taken-for-granted values, underlying assumptions, expectations, collective memories, and definitions present in an organization. It represents ‘how things are around here.’ It reflects the prevailing ideology that people carry inside their heads. It conveys a sense of identity to employees, provides unwritten and often unspoken guidelines for how to get along in the organization, and it enhances the stability of the social system that they experience.” Thus, organizational culture is taken-for-granted expectations and influences how it perceives, thinks about and reacts to internal and external environmental cues (Schein, 1996), and thus effects organizational risk-taking behavior unconsciously.

(19)

18 The literature defines organizational culture in many ways. However, one of the most influential definitions is based on the Competing Values Framework (CVF), originally developed by Quinn and Rohrbaugh (1983) to identify the key factors of organizational effectiveness. Cameron and Quinn (1999), however, refined the model to measure the organizational culture construct (Braunscheidel et al., 2010; Eisend et al., 2016; Naranjo-Valencia et al., 2011; Porcu et al., 2017; Reis et al., 2016). The reason for choosing this framework for the analysis is that it provides a systematic comparison across organizations (Howard, 1998). Furthermore, it has already been applied in multiple studies such as health care (Gregory, Harris, Armenakis, & Shook, 2009), libraries (Currie & Shepstone, 2008), military (Yardley & Neal, 2007) and manufacturing (McDermott & Stock, 1999). This theoretical framework has both face and empirical validity and integrates many of the dimensions of organizational culture proposed by various scholars (Braunscheidel et al., 2010; Eisend et al., 2016; Naranjo-Valencia et al., 2011; Porcu et al., 2017).

The CVF suggests that organizations have multiple tasks and outcomes that compete with one another (Eisend et al., 2016). The model describes organizational culture along two dimensions: structure and focus (Quinn & Rohrbaugh, 1983). Structure outlines one differentiating criteria that emphasizes flexibility from the ones that emphasize stability. Focus outlines one differentiating criteria that emphasizes an internal focus from the ones that emphasize an external focus. These dimensions cause four quadrants and each quadrant is labeled to distinguish its characteristics: clan, adhocracy, market and hierarchy cultures. Cameron and Quin (1999, p. 36) observed that “the four quadrants match precisely the main organizational forms that have developed in organizational science” and “also match key management theories about organizational success, approaches to organizational quality, leadership roles, and

(20)

19 management skills”. These types of organizational culture are leading to their own unique way of doing business. More specifically, each organizational culture has its own organizational conditions, strategic focus and decision-making tactics affecting how firms react to internal and external cues (Braunscheidel et al., 2010; Eisend et al., 2016; Naranjo-Valencia et al., 2011; Porcu et al., 2017). Thus, organizational culture affects the way a firm operates in countless ways and therefore affects the firm in many facets (McDermott & Stock, 1999). This results in different interpretations of performance causing distinctive organizational risk-taking behavior when the organization is performing below its organizational aspiration levels (Lim & McCann, 2014).

To determine the influence of organizational culture on the relationship between performing below organizational aspiration level and risk-taking behavior, this study focuses on the value dimensions focus and structure. These two dimensions represent distinctive organizational and individual factors, identifying the criteria of organizational effectiveness, and the managerial competencies that are most effective (Cameron, 2009). In addition, these four values represent opposite or competing assumptions. This means that each dimension emphasizes one core value that is opposite from the value on the other end of the continuum (Cameron, 2009).

Firms that have an internal focus are concerned with the well-being and the development of the people within the organization (Quinn & Rohrbaugh, 1983). Also, these organizations emphasize on improving internal processes and systems. The priority of the internal-oriented organization lies with maintenance and improvement of the existing organization (McDermott & Stock, 1999). As a result, these organizations utilize resources to optimize existing operational practices and focus on productivity and efficiency (McDermott & Stock, 1999). In this view,

(21)

20 managers and organizations are seen effective if internal relations and processes are harmonious (Cameron, 2009). By reducing variety and increasing efficiency in current operational practices, the internal-oriented organization emphasizes exploitation of its activities and capabilities (Uotila, Maula, Keil, & Zahra, 2009). Exploitation activities include “such things as refinement, choice, production, efficiency, selection, implementation, execution” (March, 1991, p.71). Firms focusing on exploitation use the same knowledge elements continually, which reduces the likelihood of errors and help to develop routines (Katila & Ahuja, 2002). In addition, search becomes more predictable, since the direction of the search is well-known. This results in the fact that product development can be decomposed into sub-problems and firm activities can be organized in a more efficient order, thus avoiding unnecessary steps (Katila & Ahuja, 2002). Furthermore, the firm has a deeper understanding of the knowledge by which they can identify and combine the valuable knowledge elements better than less experienced firms with these concepts (Katila & Ahuja, 2002).

Firms that have an external focus, on the other hand, are concerned with the well-being and development of the organization itself (Quinn & Rohrbaugh, 1983). These organizations emphasize innovation, adaptation, competition and interaction with the external environment (McDermott & Stock, 1999). The environment is a source of opportunities, threats and resources for these firms (Cooper & Quinn, 1993). Organizations with an external orientation evaluate their competitive landscape in order to determine their relative strengths and weaknesses with regard to their consumers and competitors, and strive to change accordingly (McDermott & Stock, 1999). Also, external-oriented organizations highlight differentiation and entrepreneurism so that they can compete with other firms (Kaarst-Brown, Nicholson, Von Dran, & Stanton, 2004). In this view, managers and organizations are seen effective if they successfully compete against

(22)

21 rivals and establish a market niche (Cameron, 2009). By adapting to the changes of the environment, the external-oriented organization increases its variety and emphasizes exploration of its activities by developing new knowledge and capabilities (Uotila et al., 2009). Exploration activities include “things captured by terms such as search, variation, risk-taking, experimentation, play, flexibility, discovery, innovation” (March, 1991, p.71). Firms focusing on exploration enrich their knowledge pool by accumulating it with new variations. These new variations help organizations to solve problems. In addition, they stimulate differentiation through re-combinatory search. Because there is a limit of possible new ideas with existing knowledge, an increase in scope could improve the firm’s possibilities for finding new combinations (Katila & Ahuja, 2002). Thus, the focus determinant is a continuum that ranges from an internal to an external orientation.

Organizations and managers tend to react to challenges and changes in the environment by intensifying their core cultural values. As competition changes and intensifies, organizational culture is given more prominence within the organization (Cameron, 2008). This means that in the case of experiencing organizational challenges, like performing below organizational aspiration levels, the organizational culture will help to determine how the firm will respond to these challenges. Therefore, when performing below aspiration levels, decision-makers in firms with an external orientation will focus on the aspiration level and interpret the performance gap as repairable which increases their risk-taking. Accordingly, the performance gap should be handled by finding new solutions and innovations to improve performance and restore the firm’s competitive advantage. However, firms with an internal orientation interpret performance shortfalls as a threat to the organization and the decisions-makers. Hence, the firm focuses on

(23)

22 survival by improving their current operations and becomes more risk-averse. Therefore, the hypothesis states:

Hypothesis 1: When performance is below the organizational aspiration level, performance decreases lead to more risk-taking behavior among external-oriented organizations than among internal-oriented organizations.

Furthermore, it is expected that structure will influence organizational risk-taking behavior. Firms that emphasize a flexible orientation allow spontaneity and flexibility within their organizations (McDermott & Stock, 1999). Also, innovation and change are the viewpoints that are at the heart of the organization (Quinn & Rohrbaugh, 1983). This is because flexibility, the lack of formality and organic structures, lead to autonomy and freedom in the organization that encourages creativity, which is the key to developing innovations and firm growth (Naranjo-Valencia et al., 2011). Creativity within a firm can be seen as the capability to perform in original and suitable way, i.e. useful (McLean, 2005) or to produce new and valuable ideas (Amabile, 1998). This creativity is also a key feature of an innovative culture and an element of the innovator character of the firm (Sugita & Takahashi, 2015). In addition, creative acts can influence knowledge processes, which lead to distinctive organizational learning effects (Ford & Ogilvie, 1996). Similarly, organizational learning processes that consist of constant creativity, are more innovation oriented (Baker & Sinkula, 2007). Furthermore, the firm openness in flexible-oriented organizations encourages employees to use and share knowledge (Alas, Ubius, & Gaal, 2012; Eisend et al., 2016; Naranjo-Valencia et al., 2011; Reis et al., 2016) and improve their

(24)

23 innovation abilities (Lin, 2007). In this view, managers and organizations are seen effective if they are adaptive, changing, organic and transformational (Cameron, 2009).

Firms that have a control orientation, however, focus on stability, control and order in their organizations (McDermott & Stock, 1999). Rules, regulations and excessive authority are characteristics of the control orientation (Naranjo-Valencia et al., 2011). These characteristics together with the poor participation of the employees (i.e. centralization) limits the capacity of the organization to accept the risks of innovation (Child, 1973). The organization has therefore, adverse circumstances for searching and finding new managerial opportunities (Sugita & Takahashi, 2015). Bureaucratic control, through several mechanisms like rules and procedures and hierarchy of authority, standardizes outputs to realize efficiency (Ouchi, 1979). Thus, the efficient-bureaucratic firm realizes organizational efficiency by striving for certainty in its internal operations and processes (Covin & Slevin, 1989). In this view, managers and organizations are seen effective if they are consistent, stable, mechanistic and predictable (Cameron, 2009). Thus, the structure determinant is a continuum that ranges from flexibility to control.

An organization whose culture is characterized by flexibility is likely to deal better with uncertainty and challenges than an organization that has a control orientation (McDermott & Stock, 1999). Therefore, this research expects that regarding risk-taking behavior, managers working in an organization with a flexible orientation will respond more forceful and aggressively toward performance challenges than managers working in an organization with a control orientation. Flexible-oriented organizations favor a more innovative orientation, whereas stability and control-oriented organizations will hinder it (Naranjo-Valencia et al., 2011). Following this reasoning, managers that come from a flexible orientation background should

(25)

24 experience performing below organizational aspiration levels more as a challenge that can be tackled by encouraging creativity, adaptability and developing pioneering innovations. Thus, managers working in flexible-oriented organizations will take greater risks when performing below their aspiration levels to repair the performance gap. Managers from a stability and control-oriented organizations will experience performing below organizational aspiration levels as a challenge that should be handled by controlling the current operational practices. Consequently, managers working in stability and control-oriented organizations will become more risk-averse when performing below their aspiration levels and focus their attention on improving their current operations. Therefore, the hypothesis states:

Hypothesis 2: When performance is below the organizational aspiration level, performance decreases lead to more risk-taking behavior among flexibility-oriented organizations than among stability and control-oriented organizations.

Methods

Samples and Data

To date, organizational culture has not been measured using secondary data. Past research has performed surveys to measure the construct of organizational culture. An overview of some past research on organizational culture can be found in Appendix 1. In these studies, however, the independent variable of performance feedback has not been included and measured. To determine the influence of past performance, data accumulated over multiple years is needed. As a result, in prior research the relationship between performance feedback and risk-taking behavior is measured through secondary data (Audia & Greve, 2006; Bromiley, 1991; Desai, 2008; Greve,

(26)

25 2003; Iyer & Miller, 2008; Lim & McCann, 2014; Miller & Chen, 2004). Consequently, the same data design is used in this research in combination with the operationalization of the construct organizational culture.

To test the hypothesis, this research used secondary data to examine the moderating effect of organizational culture on the relation between performance below organizational aspiration level and risk-taking behavior in terms of changes made in research and development (R&D) investment. Secondary data from the Wharton CRSP – Compustat Merged database was used and combined with patent data. The CRSP – Compustat database is a database of U.S. fundamental and market information that provides annual and quarterly Income Statement, Balance Sheet, Statement of Cash Flows, and supplemental data, like industry items. To keep the industry backgrounds comparable and to take into account that patent data was used, data from the U.S. manufacturing industries with the SIC codes from 28 (chemicals and allied products), 35 (industrial and commercial machinery and computer equipment) and 36 (electronic, electronical equipment and components, except computer equipment) were used. These industries are known for their patent use and display the highest number of citations placed (Hall, Jaffe, & Trajtenberg, 2001). This also provides the opportunity to compare results with previous studies (Chen & Miller, 2007; Iyer & Miller, 2008; Lim & McCann, 2014; Palmer & Wiseman, 1999). The independent variable was lagged at time t - 1. Accordingly, the independent variables ranged from the years 1979 to 2003, and the dependent and control variables corresponded to the years 1980 to 2004.

(27)

26 Measures

Dependent Variable. The dependent variable (DV) in this research is risk-taking behavior. Risk is examined in two ways: as organizational risk or as managerial risk taking. Organizational risk is measured through income stream variation, whereas managerial risk-taking behavior is measured through resource allocations toward uncertain activities (Audia & Greve, 2006; Desai, 2008). This study focuses on managerial risk-taking, because this complies more with managers’ actual risk-taking behaviors (Palmer & Wiseman, 1999). More specifically, this study examined organizational risk-taking behavior as R&D expenditures in millions of U.S. dollar divided by sales. This has been an empirically proven measure for risk-taking behavior and is consistent with prior studies (e.g.Chen & Miller, 2007; Lim & McCann, 2014). Also, following past studies, this research focused on firms with R&D intensity less than or equal to 1.0, because firms with R&D intensity above 1.0 are perceived as R&D specialists and they could manipulate the findings. As a result, 7344 firms were removed from the sample and the resulting sample has a mean of 0.106 and a standard deviation of 0.151.

Independent Variable. The independent variable (IV) in this study is firm performance and measured by return on assets (ROA) (net income divided by total assets). This measurement was incorporated because ROA is the primary measure for firm profitability in the manufacturing industry and it is relevant to asset growth (Desai, 2008; Greve, 2003). Furthermore this measurement is employed in previous studies (Audia & Greve, 2006; Bromiley, 1991; Chen & Miller, 2007; Desai, 2008; Greve, 2003; Iyer & Miller, 2008; Lim & McCann, 2014). Firm performance was measured by historical and social aspiration levels, therefore this study incorporated a variable that measured the focal firm’s own past performance (historical aspirations) and one that measured the past performance compared to peer group firms in the

(28)

27 same industry (social aspirations). Historical aspiration level was measured as the average firm’s ROA of the last three years. Social aspiration level was measured as the mean ROA of all firms within each industry. The actual firm performance was calculated at year t - 0, and the aspiration levels were proxied at year t - 1. As a result, firm performance feedback was calculated. Historical performance feedback was calculated by the firm’s average ROA of the last three years. Social performance feedback was calculated by the mean ROA of all organizations within each industry. Subsequently, spline variables were computed by dividing both historical and social performance feedback variables into variables that measure performance below aspiration and performance above aspiration (Desai, 2008). The variable performance below aspiration level consisted of observations that are smaller than zero, whereas the variable performance above aspiration level consisted of observations that are greater than zero.

Moderating Variable. The moderating variable (MV) in this study is organizational culture. The conceptualization of organizational culture was based on the Competing Values Framework (CVF). The model describes organizational culture along two dimensions: structure and focus (Quinn & Rohrbaugh, 1983). Structure indicates whether firms are flexible and discrete or stable and controlled. Focus looks whether firms are in- or outwards focused (Quinn & Rohrbaugh, 1983). As mentioned above, the reason for choosing this framework for the analysis was that it provides a systematic comparison across organizations (Howard, 1998). Furthermore, it has been applied in multiple studies such as health care (Gregory et al., 2009), libraries (Currie & Shepstone, 2008), military (Yardley & Neal, 2007) and manufacturing (McDermott & Stock, 1999). Also, this theoretical framework has both face and empirical validity and integrates many of the dimensions of organizational culture proposed by various scholars (Braunscheidel et al.,

(29)

28 2010; Eisend et al., 2016; Naranjo-Valencia et al., 2011; Porcu et al., 2017). Below an overview of the proxies of organizational culture is provided in the dimensions: structure and focus.

Focus: Internal-oriented organizations focus on activities within the organization and emphasizes maintenance and optimization of the existing organization. The organization expend resources to improve the current operational practices (McDermott & Stock, 1999). By reducing variety and increasing efficiency in current operational practices, the internal-oriented organization emphasizes exploitation of its activities and capabilities (Uotila et al., 2009). External-oriented organizations focus on activities outside the organization and emphasize adaptation and competition with the external environment. Therefore, the organization scans the competitive environment and makes changes to the firm accordingly (McDermott & Stock, 1999). By adapting to the changes of the environment, the external-oriented organization increases its variety and emphasizes exploration of its activities by developing and exploring new knowledge and capabilities (Uotila et al., 2009). The relative amount of exploitation vs. exploration in the company’s business activities was measured by the depth and breadth of search activities (Katila & Ahuja, 2002).

In line with Katila & Ahuja (2002), the variable search depth is the search of experience with similar knowledge elements. As argued above, the more a firm exploit knowledge, the more in depth that firm knows it and the more the firm has an internal focus. “Thus, search depth was measured as the average number of times a firm repeatedly used the citations in the patents it applied for. The depth variable was calculated by the number of times that, on the average, each citation in year t - 1 was repeatedly used during the past five year” (Katila & Ahuja, 2002). The variable search scope, is in line with the theoretical idea of exploration of new knowledge. This measurement corresponds with the quantity of formerly unused citations in the firm’s focal year’s

(30)

29 list of citations. “The share of citations was assessed in a focal year's citations that could not be found in the previous five years' list of patents and citations by that firm. Values for this variable, which was calculated as follows, range from 0 to 1” (Katila & Ahuja, 2002). However, following Katila & Ahuja (2002) in threatening search depth and scope or in other words exploitation and exploration as two separate construct, this research followed Uotila et al., (2009) by combining the two constructs in one measurement in a continuum. The variable focus, the relative amount of exploratory activities of an organization, was therefore constructed as the amount of exploratory activities (search scope) divided by the sum of exploratory (search scope) and exploitative activities (search depth) (Uotila et al., 2009). Yet, before computing the variable relative amount of exploratory activities, both search depth and scope were standardized. Higher focus scores represent external-oriented organizations and lower scores represent internal-oriented organizations.

Structure: Flexible-oriented organizations focus on innovation and adaptability, whereas control-oriented organizations emphasize stability, productivity and efficiency (Braunscheidel et al., 2010; Porcu et al., 2017; Quinn & Rohrbaugh, 1983). Therefore, to measure structure a composite measurement was constructed. Higher structure scores represent flexible-oriented organizations and lower scores represent control-oriented organizations. The composite measure consists of: (1) ratio of research and development to sales (RDS5), (2) ratio of employees to sales (EMPNS5), (3) historical growth (REV5) as the one-year percentage change in total sales and (4) market share (MKS5) market dominance as company market share relative to industry average market share (Bentley, Omer, & Sharp, 2013; Higgins, Omer, & Phillips, 2015; Ittner, Larcker, & Rajan, 1997). All variables were measured over a period of five years.

(31)

30 The ratio research and development to sales (RDS5) serves as a measure of a firm’s propensity to innovate and seek new products (Higgins et al., 2015). Because flexible-oriented organizations engage more in innovative activities, it is expected that these type of organizations will have higher research and development costs than control-oriented organizations (Braunscheidel et al., 2010; Porcu et al., 2017; Quinn & Rohrbaugh, 1983). The ratio employees to sales (EMPNS5) is a proxy for a firm’s ability to produce efficiently (Higgins et al., 2015). Because the control-oriented organization focuses on improving operational practices and emphasizes productivity and efficiency, it is expected that these organizations will have fewer employees per dollar of sales (Braunscheidel et al., 2010; Porcu et al., 2017; Quinn & Rohrbaugh, 1983; Zammuto & O’Connor, 1992). Furthermore, it is to be expected that flexible-oriented organizations have greater growth opportunities (REV5) than control-flexible-oriented organizations, because the flexible-oriented organizations emphasizes growth and acquires resources through flexibility and readiness (Braunscheidel et al., 2010; Eisend et al., 2016). Finally, it is also to be expected that company market dominance as market share (MKS5) relative to the industry average will be higher for control-oriented organizations than flexible-oriented organizations, since control-flexible-oriented organizations strive for market superiority and success is defined in terms of market share and penetration (Braunscheidel et al., 2010; Sugita & Takahashi, 2015) .

To measure structure, the four variables were ranked by quintiles within each year (Bentley et al., 2013; Higgins et al., 2015). Each year the observations were ranked with a 5 if they belong to the top quintile, and so on, until the variables with the lowest quintile, they were given a score of 1. The scores of Market Dominance was revised, so that score 1 is 5 and score 5 is 1 and so on. Subsequently, the scores over the years are summed up. The maximum score a

(32)

31 company could get was 20 (flexible-oriented) and the minimum score was 5 (control-oriented). Structure was measured as a continuous variable, meaning the higher the score the more it involves a flexible-oriented structure and the lower the score the more it involves a control-oriented structure.

Control variables (CV) in this research are firm size, distance to bankruptcy, and organizational slack.

Firm size. The reason for incorporating firm size was because organizational risk-taking might increase when firm size increases (Lim & McCann, 2014). Firm size was measured by the number of employees. The number of employees was logged, because this better captures the effect size of firm size on organizational risk-taking (Audia & Greve, 2006). Also, it assured that, regardless of firm size, an increase of a given percentage has the same effect (Audia & Greve, 2006).

Organizational slack. Slack resources may influence organizational risk taking (Singh, 1986), therefore this control variable was incorporated in this research. In line with previous studies, this study had chosen working capital-to-sales ratio (absorbed slack) and current ratio, calculated by current assets-to-current liabilities ratio (unabsorbed slack) as slack proxies (Chen & Miller, 2007; Lim & McCann, 2014; Singh, 1986).

Distance from bankruptcy. Distance from bankruptcy as a control variable was incorporated, because firms are tending to be more risk-averse when facing bankruptcy (March & Shapira, 1992). To measure distance to bankruptcy this research followed past studies by using Altman’s Z-score (Chen & Miller, 2007; Iyer & Miller, 2008; Lim & McCann, 2014; Miller & Chen, 2004). Altman’s Z-score is measured as: “(1.2 x working capital divided by total assets) + (1.4 x retained earnings divided by total assets) + (3.3 x income before interest expense and taxes

(33)

32 divided by total assets) + (0.6 x market value of equity divided by total liability) + (1.0 x sales divided by total assets).” A higher value of this measure indicates a lower risk of bankruptcy.

Analysis

To analyze the influence of organizational culture on the relationship between organizational performance and risk-taking behavior, the cross-section and time-series differences within the data were taken into account in the estimation method. The Breusch-Pagan Lagrange multiplier test was performed to test for individual specific effects. The null hypothesis states that there are no individual specific effects. The test rejected the null hypothesis (χ² (4) = 5122.62, p < 0.00), which indicates that there are significant differences across firms in the data. This means that the simple OLS estimation is not appropriate for this study.

Subsequently, the Hausman’s specification test was executed to determine to run the within (fixed effects) or GLS (random effects) estimator (Mutl & Pfaffermayr, 2011). The null hypothesis states that difference in coefficients are not systematic. The test rejected the null hypothesis (χ² (4)= 243.66, p < 0.00), meaning that the random effect model is not appropriate and that the GLS estimator cannot account for firm and time effects in the data. Thus, this study performed all analyses using the two way fixed effects model to control for firm and time effects. Also, to take into account possible heteroscedasticity-consistent standard errors in the data, the regression analyses were performed with a robust variance estimator based on a varlist of equation-level scores and a covariance matrix. Finally, all the independent and moderating variables were mean-centered before creating the interaction variables in order to mitigate potential multicollinearity problems (Aiken, West, & Reno, 1991; Lim & McCann, 2014).

(34)

33

Results

Descriptive statistics and correlations are shown in Table 1. The relationships were analyzed using the list-wise correlation coefficient. First, the correlations between the dependent variable and the control variable are outlined. Second, the correlations between the dependent variable and the independent and moderating variables are discussed. Third, the correlations between the moderating variables and the independent variables, and the correlations between the moderating variables and the control variables are presented.

The sample consisted of 16.800 observations from 1979 to 2004. The three two-digit industries that were analyzed were (1) chemicals and allied products, (2) industrial and commercial machinery and computer equipment and (3) electronic, electronical equipment and components, except computer equipment. Looking at the correlations between the dependent variable and the control variables, R&D intensity was negatively correlated with firm size (r=-0.25). Organizational slack is, however, positively correlated with R&D intensity. This relation was slightly stronger for absorbed slack (r=0.36) than for unabsorbed slack (r=0.23). Also, the correlation between R&D intensity and distance from bankruptcy was positive (r=0.11).

When considering the correlations between the dependent variable and the moderating variables, R&D intensity had not a significant correlation with focus, whereas had a very strong positive correlation with structure (r=0.41). Furthermore, analyzing the correlations between the dependent and the independent variables, R&D intensity was negatively correlated with performance below historical aspirations (r=-0.16), whereas it is positively correlated with performance above historical aspirations (r=0.14). Moreover, R&D intensity is negatively correlated with performance below social aspirations (r=-0.21). Performance above social aspirations, however, had a weak positive correlation with R&D intensity (r=0.02). Following

(35)

34 Table 1. Means, Standard Deviations, and Correlationsᵃ

_{Mean S.D.} ₁ ₂ ₃ ₄ ₅ ₆ ₇ ₈ ₉ ₁₀ 1. R&D Intensity 0,11 0,15 2. Firm size² -0,29 2,08 -0,25 ** 3. Absorbed slack 0,52 1,40 0,36 ** -0,19 ** 4. Unabsorbed slack 3,43 3,87 0,23 ** -0,27 ** 0,59 ** 5. Distance from bankruptcy 6,31 15,38 0,11 ** -0,08 ** 0,36 ** 0,51 ** 6. Focus 0,78 3,01 -0,01 -0,01 -0,01 -0,003 -0,003 7. Structure 11,41 3,45 0,41 ** -0,51 ** 0,21 ** 0,21 ** 0,11 ** 0,00 8. Performance below aspirations (historical) -0,07 0,17 -0,16 ** 0,18 ** 0,03 ** 0,03 ** 0,11 ** 0,00 -0,12 ** 9. Performance above aspriations (historical) 0,05 0,16 0,14 ** -0,19 ** 0,05 ** 0,04 ** 0,04 ** -0,005 0,20 ** 0,14 ** 10. Performance below aspriations (social) -0,06 0,20 -0,21 ** 0,24 ** 0,06 ** 0,07 ** 0,12 ** 0,00 -0,22 ** 0,8 ** -0,01 11. Performance above aspirations (social) 0,14 0,16 0,02 ** 0,09 ** 0,06 ** 0,13 ** 0,15 ** -0,003 -0,06 ** 0,21 ** 0,10 ** 0,23 ** ᵃ N = 16.800

ᵇ log number of employees

**. Correlation is significant at the 0.01 level (2-tailed) *. Correlation is significant at the 0.05 level (2-tailed)

(36)

35 prior studies, separated models were used for historical and social aspirations to circumvent distorted parameter estimates (Chen & Miller, 2007; Iyer & Miller, 2008; Lim & McCann, 2014). Regarding the correlations between the moderating variables and the independent variables, focus was not correlated with above and below aspirations (historical and social). In addition, the moderating variable structure was negatively correlated with below historical aspirations (r=-0.12), but was positively correlated with above historical aspirations (r=0.20). Furthermore, structure was both negatively correlated with below social aspirations (r=-0.22) and above social aspirations (r=-0.06), although this correlation was very weak for the latter. Focus had a non-significant correlation with all the control variables. Structure, on the other hand, had a strong negative correlation with firm size (r=-0.51). In contrast, structure had a positive correlation with absorbed slack (r=0.21), unabsorbed slack (r=0.21) and distance from bankruptcy (r=0.11). For the complete overview of all the correlations, please refer to Table 1.

Interaction Effect of Organizational Culture and Performance Below Organizational Aspirations on Firm Risk Taking

The Models in Table 2 examine the effect of performance below aspirations on R&D intensity moderated by focus and controlled by firm size, absorbed slack, unabsorbed slack and distance to bankruptcy. The Models in Table 3 are similarly constructed but examine the moderating effect of structure in the relationship. Model 1 in both Table 2 and 3 are similar and is the base model consisting of the control variables. In Models 2 and 4 in Table 2 and Models 6 and 8 in Table 3 the independent variables are added and show the main effect. Models 3 and 5 in Table 2 show the interaction effect of the moderating variable focus on the dependent variables. Similar is the case with Models 7 and 9 in Table 3, however here the interaction effect of the

(37)

36 Table 2. Fixed-Effects Models of R&D Intensity and Moderating Effect of Focus

Historical aspirations Social aspirations Model 1 Model 2 Model 3 Model 4 Model 5 Variable Coefficient S.E. Coefficient S.E. Coefficient S.E. Coefficient S.E. Coefficient S.E. Firm size -0,013 ** 0,002 -0,012 ** 0,002 -0,011 ** 0,002 -0,010 ** 0,002 -0,010 ** 0,002 Absorbed slack 0,170 ** 0,004 0,030 ** 0,008 0,030 ** 0,008 0,029 ** 0,006 0,029 ** 0,006 Unabsorbed slack -0,001 0,001 -0,001 0,001 -0,001 0,001 -0,001 0,001 -0,001 0,001 Distance from bankruptcy -0,0004 † 0,000 -0,000 † 0,000 -0,000 † 0,000 -0,000 * 0,000 -0,0005 * 0,000 Focus -0,000 † 0,000 -0,002 * 0,001 -0,000 0,000 -0,001 ** 0,004 Performance below aspirations -0,019 † 0,01 -0,016 0,011 -0,018 † 0,010 -0,017 † 0,010 Performance above aspirations -0,019 0,012 -0,027 ** 0,010 -0,026 ** 0,006 -0,027 ** 0,006 Performance below aspirations x focus 0,025 ** 0,021 † 0,0127 Performance above aspirations x focus -0,029 † 0,003 0,009 Model F 7,91 5,11 7,46 6,89 7,29 R² 0,156 0,170 0,168 0,193 0,0193 N = 16.800

**. Correlation is significant at the 0.01 level (2-tailed). *. Correlation is significant at the 0.05 level (2-tailed). †. Correlations significant at the 0.10 level (2-tailed).

(38)

37 Table 3. Fixed-Effects Models of R&D Intensity and Moderating Effect of Structure

Historical aspirations Social aspirations

Model 1 Model 6 Model 7 Model 8 Model 9 Variable Coefficient S.E. Coefficient S.E. Coefficient S.E. Coefficient S.E. Coefficient S.E. Firm size -0,013 ** 0,002 -0,011 ** 0,002 -0,011 ** 0,002 -0,010 ** 0,002 -0,010 ** 0,002 Absorbed slack 0,17 ** 0,004 0,029 ** 0,008 0,03 ** 0,008 0,029 ** 0,006 0,029 ** 0,006 Unabsorbed slack -0,001 0,001 -0,001 0,001 -0,001 0,001 -0,001 0,001 -0,001 0,001 Distance from bankruptcy -0,0004 † 0,000 -0,0004 * 0,000 -0,0004 * 0,000 -0,001 * 0,000 -0,001 * 0,000 Structure 0,003 ** 0,001 0,004 ** 0,001 0,003 ** 0,001 0,003 ** 0,001 Performance below aspirations -0,020 * 0,010 -0,030 ** 0,008 -0,020 * 0,010 -0,020 * 0,009 Performance above aspirations -0,019 0,012 -0,007 0,012 -0,023 ** 0,006 -0,037 ** 0,007 Performance below aspirations x focus 0,005 † 0,003 0,002 0,002 Performance above aspirations x focus -0,005 † 0,003 -0,01 ** 0,002 Model F 7,91 5,49 7,29 7,57 R² 0,156 0,224 0,246 0,224 N = 16.800

**. Correlation is significant at the 0.01 level (2-tailed). *. Correlation is significant at the 0.05 level (2-tailed). †. Correlations significant at the 0.10 level (2-tailed).

Performance feedback and organizational risk-taking behavior : the moderating effect of organizational culture.

Master Thesis