Performance feedback and organizational risk-taking behavior : the moderating effects of strategic orientations

(1)

Performance Feedback and Organizational Risk-taking Behavior: The

Moderating Effects of Strategic Orientations

Master Thesis

Student:

Natasha Hoang / Student № 10711708

MSc. Business Administration, Strategy track

University of Amsterdam, Faculty of Economics and Business

Supervisor: MSc. B. Silveira Barbosa Correia Lima

University of Amsterdam, Amsterdam Business School

(2)

Statement of originality

This document is written by student Natasha Hoang 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)

Abstract

This study examines the effects of performance below aspiration levels on organizational risk-taking behavior among firms with different strategic orientations. I posit that firms with distinctive strategic orientations perceive and react differently, in terms of risk-taking

behavior, to performance feedback when performance is below aspiration levels. Moreover, I postulate that strategic orientations influence aspiration preference of managers. Using data on the research and development spending of U.S. manufacturing firms and Miles and Snow’s STRATEGY measure to account for different strategic orientations, I find that

underperforming Prospector firms and underperforming Defender firms take greater risk when their performance falls further below their aspiration level, but the amount of risk-taking is greater for underperforming Prospect firms than for underperforming Defender firms. Furthermore, I find that the aspiration preference of Prospector managers is different from Defender managers, as such that Prospector managers rely solely on social aspirations, whilst Defender managers attend to both historical and social aspirations to learn from performance feedback. These findings are largely consistent with my predictions and suggest that firms with distinct strategic orientations exert different organizational risk-taking

behavior in response to performance feedback.

Key words: Strategic orientations; Prospectors; Defenders; performance feedback; organizational risk-taking behavior; behavioral theory of the firm

(5)

Introduction

There is a considerable amount of research on how firms can learn from past

experiences to enhance their organizational performance (e.g., Argote McEvily & Reagans, 2003; Argote & Miron-Spektor, 2011; Levinthal & March, 1993; Levitt & March, 1988; March, 1991). According to the literature, one way of doing so is by learning from

performance feedback (e.g., Argote & Miron-Spektor, 2011; Audia & Greve, 2006; Cyert & March, 1963; Desai, 2008; Iyer & Miller, 2008; Jordan & Audia, 2012; Kacperczyk,

Beckman, & Moliterno, 2015; Kim, Finkelstein, & Haleblian, 2015). The underpinnings of this view are based on the behavioral theory of the firm, which employs a learning model that focuses on past performance (Cyert & March, 1963). This perspective suggests that firms set levels of performance based on their past performance and based on the performance of other similar firms (Cyert & March, 1963). According to this perspective, performance below aspiration levels prompts managers to engage in problemistic search to identify and pursue alternative courses of action that can improve organizational performance (Cyert & March, 1963). Conversely, performance congruent to- or above aspiration levels reduces the need for problemistic search and stimulates managers to proceed with their ‘effective’ routines and strategic approaches (Bromiley, Miller, & Rau, 2001; Greve, 2003).

In addition, various studies argue that performance feedback influences organizational risk preferences (e.g., Audia & Greve, 2006; Boyle & Shapira, 2012; Bromiley, 1991; Desai, 2008; Greve, 2003; Miller & Chen, 2004; Palmer & Wiseman, 1999). However, these studies fail to provide consistent evidence regarding what the effects of performance below aspiration levels are on risk-taking behavior of firms. More specifically, several studies found that when firms are facing performance shortfalls relative to aspirations, they tend to take greater risks. On the contrary, there are also studies suggesting that when firms perform below aspirations, they tend to become more risk-averse (Stikin & Pablo, 1992; Wiseman & Bromiley, 1996;

(6)

While other studies showed partial effects arguing that firms may in some cases respond conservatively to performance shortfalls due to different organizational determinants (e.g., Chen & Miller, 2007; Desai, 2008; Iyer & Miller, 2008).

Most of these studies examined the effects of performance feedback on organizational risk-taking behavior, but assumed that all firms are the same and disregarded the matter of heterogeneity among firms. However, firms under different organizational conditions are likely to interpret performance feedback in distinctive ways, which leads to distinct organizational risk-taking behavior (Devers et al., 2007b; Tuggle et al., 2010, in Lim & McCann, 2014; O'Regan & Ghobadian, 2006; Slater, Olson, & Hult, 2006). Recent research has come to recognize that and suggests that a richer understanding of the influence of performance feedback on organizational strategic behavior lies in investigating the

moderating factors that characterize these organizational conditions in this relationship (e.g., Audia & Greve, 2006; Desai, 2008; Greve, 2003; Kim et al., 2015; Lim & McCann, 2014). For example, Audia and Greve (2006) found that small firms lower risk-taking behavior in response to performance below aspirations, while it has no effect or increases risk-taking behavior for large firms. Additionally, Kim and McCann (2014) showed that in a negative attainment discrepancy context, high values of option grants decreases the risk-taking

behavior of CEO, while it enhances the risk-taking propensity of outside directors. Hence, to obtain a better understanding of the relationship between performance feedback and

organizational risk-taking behavior, it is necessary to consider the organizational conditions that influence this relationship.

When considering prior research on performance feedback and organizational risk-taking behavior, a crucial but forgotten organizational factor in the literature is the firm’s strategic orientation, in spite of its great influences on organizational behavior. Strategic orientations convey organizational values, beliefs, motivations and aspirations, and guide the

(7)

strategy formulation and operating manners of firms (Fox-Wolfgramm, Boal, & Hunt, 1998; Gatignon & Xuereb, 1997; O'Regan & Ghobadian, 2006; Shapiro, 1989; Slater, Olson, & Hult, 2006). Therefore, every strategic orientation consists of a set of unique strategic decisions (long-lasting commitments) and tactical decisions (short-term responses to the environment), which govern the way a firm responses to internal- and external pressures (Fox-Wolfgramm et al., 1998; Gatignon & Xuereb, 1997; O'Regan & Ghobadian, 2006; Slater et al., 2006). Consequently, firms with distinctive strategic orientations interpret and react to performance feedback differently when they find themselves in the position in which their performance is below their aspiration levels.

This study attempts to reconcile the paradoxical views regarding what the effects of performance below aspirations are on organizational risk-taking behavior by accounting for differences in firms. To do this, I investigate the effects of performance on organizational risk-taking behavior in firms that are performing below aspiration levels by considering the moderating effect of different strategic orientations. Building on the theories of strategic orientations, I propose that a firm’s strategic orientation influences the firm’s risk preferences, and consequently, its willingness to take greater risks.

Miles and Snow (1978) identified two distinctive strategic orientations that, I believe, affect the level and assertiveness of risk-taking of firms. They refer to these strategic

orientations as Prospector and Defender strategic orientations. Firms that are identified as Prospector firms operate in a dynamic environment that requires them to adopt an innovation strategy, including identifying and pursuing new opportunities, experiencing continuous change, embracing uncertainty, and stimulating risk-taking (Ghoshal, 2003; Miles, Snow, Meyer, & Coleman, 1978). Whereas, Defender firms are firms operating in a stable

environment in which they deliberately enact and maintain its stability (Ghoshal, 2003; Miles et al., 1978). They adopt a cost leadership strategy and strive to maintain organizational

(8)

stability with a narrow and stable product focus to compete on price, service or quality (Ghoshal, 2003; Miles et al., 1978). Hence, due to these differences in the two strategic orientations, firms that adopt a Prospector strategic orientation will interpret and react to performance feedback differently than firms that adopt a Defender strategic orientation.

In addition, recent research has begun to consider the effects of historical and social aspirations separately (Boyle & Shapira, 2012; Kacperczyk et al., 2015; Kim et al., 2015; Shipilov et al., 2011, in Kim et al., 2015). According to these studies historical and social aspirations have different impacts on organizational behavior and produce distinct behavioral responses (Boyle & Shapira, 2012; Kacperczyk et al., 2015; Kim et al., 2015; Shipilov et al., 2011, in Kim et al., 2015). These studies stir up the question: Under what conditions are managers more likely to attend to historical- or social aspiration levels? To come a step closer in answering this question, I believe one should consider a firm’s strategic focus as this guides the information process of the firm and specifies which reference point is appropriate to attend (Fox-Wolfgramm et al., 1998; Gatignon & Xuereb, 1997; O'Regan & Ghobadian, 2006; Shapiro, 1989; Slater et al., 2006). Hence, some firms might pay more attention to historical aspirations, while others might prefer to focus more on social aspirations, depending on their strategic orientation. Thus in line with this logic, I believe that a firm’s strategic orientation influences the aspiration preferences of managers.

I explore these ideas by studying the risk-taking behavior of firms when performance is below aspiration levels. A discrete STRATEGY composite measure is the proxy for the strategic orientations of firms (Miles & Snow, 1978, 2003; Hambrick 1983; Segev 1987; Ittner, Larker, & Rajan 1997; Bentley, Omer & Sharp, 2013, in Higgins, Omer, & Phillips, 2014). Following prior research, I use R&D intensity as a proxy for organizational risk-taking behavior. This is consistent with the studies of Alessandri and Pattit (2014), Coles, Daniel, and Naveen (2006), and Lim & McCann, (2014). R&D intensity is a risky decision because it

(9)

entails a great amount of uncertainty and it may include losses (Palmer & Wiseman, 1999), that is R&D investments require expenditures in the short term, which may negatively affect the firm’s performance, while their payoffs may come much later in the long term or it may not lead to any payoffs at all (David et al., 2001, in Lim & McCann, 2014). Therefore

following the literature, I assume that high investments in R&D generally imply that a firm is engaging in a high-risk return strategy.

This paper makes three contributions to the existing literature. First, it contributes to the behavioral theory of the firm and the strategic orientation field by extending previous work on the behavioral theory of the firm through analyzing the moderating effects of strategic orientations on the relationship between performance feedback and organizational risk-taking behavior. Hence, it contributes to a better understanding of the influences of organizational conditions on the relationship between performance feedback and organizational risk-taking behavior. Second, it sheds light on how strategic orientations influence organizational preferences regarding the attended reference point (historical- or social), and thereby contributing to the development of a richer understanding of the implication of performance feedback on firms with different strategic orientations. Third, it enhances research on risk-taking behavior by incorporating the principal arguments of the behavioral theory of the firm with related research on R&D investments of firms.

(10)

Theory and hypotheses

The following section discusses the main insights of the existing literature on performance feedback and strategic orientations, and presents the hypotheses of this research. First, I introduce the nature of performance feedback and how this mechanism influences the risk-taking behavior of firms. Next, I address the literature gap and present the research question. Then, I apply the fundamental principles of the behavioral theory of the firm to related R&D literature and present the first hypothesis. Subsequently, I outline several strategic orientations typologies and justify my choice for selecting the Miles and Snow’s typology of strategic orientations. Finally, I present the strategic orientations, introduce the aspiration preference of each strategic orientation, and outline the other remaining hypotheses.

Performance feedback and organizational risk-taking behavior

A large body of research demonstrates that performance relative to aspirations (or distance from aspiration or attainment discrepancy) influences organizational behavior, including organizational learning behavior (e.g., Levitt & March, 1988), risk-taking behavior (e.g., Audia & Greve, 2006; Desai, 2008; Jordan & Audia, 2012; Kim et al., 2015; March and Shapira, 1987; Miller and Chen, 2004; Miller and Leiblein, 1996; Singh, 1986), and

innovative behavior (e.g., W. Chen & Miller, 2007; Greve, 2003; Lucas, Knoben, & Meeus, 2015; Parker, Krause, & Covin, 2015). The underpinnings of this research are based on the behavioral theory of the firm. This perspective argues that managers are constrained by bounded rationality and are, therefore, not able to use all available information. In order to overcome this cognitive limitation during the decision making process, managers learn from performance feedback (Audia, Locke, & Smith, 2000; Greve, 1998; Lant, Milliken, & Batra, 1992; Mezias, Chen, & Murphy, 2002; Miller & Chen, 1994, in Jordan & Audia, 2012). The behavioral theory of the firm employs a learning model as it is centered around the believe

(11)

that firms learn from past experience and change their operating practices depending on the discrepancy between their set aspiration level and achieved performance (Cyert & March, 1963). According to Levitt and March (1988) “… organizations are seen as learning by encoding inferences from history into routines that guide behavior” (Levitt & March, 1988, p. 320). Firms set aspiration levels to assess organizational goals and to use it as a benchmark to evaluate their performance (Cyert & March, 1963). An aspiration level is a desired

performance level in a specific organizational outcome by which performance is considered as success or failure (Shinkle, 2011). They arise from comparisons of two reference points, which managers use to assess their current performance (Cyert & March, 1963). A

comparison that includes the firm’s own performance history as reference point is defined as historical comparison, and the aspiration set based on performance history is known as a historical aspiration level (Cyert & March, 1963; Levinthal & March, 1981). Additionally, a comparison that uses the performance of a peer group firms as reference point is referred as social comparison, and the aspiration that is based on peer group performance is called a social aspiration level (Cyert & March, 1963). So, the behavioral theory of the firm suggests that firms learn from performance feedback by assessing their performance through a

systematic process, wherein they set historical and social aspiration levels. Subsequently, they evaluate their performance and take action based on their perception of how they have

performed relative to their aspirations (Cyert & March, 1963; Greve, 2008; Jordan & Audia, 2012).

According to the behavioral theory of the firm performance that is below aspiration levels prompts 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). This means that managers identify and pursue alternative courses of action that may improve the performance of the firm and lead to an outcome that is above the desired

(12)

aspiration level (Greve, 2008). However, the performance feedback theory only induces that firms will engage in problemistic search, but it does not predict what kind of actions or solutions managers will adopt to address the problem (Cyert & March, 1963). Therefore, the performance feedback theory needs to be complemented with additional theories in order to predict what kind of search firms will adopt when facing performance below aspirations. Most studies have incorporated different theories, such as prospect theory, threat of rigidity theory, self-enhancement theory and organizational learning theory, to identify the alternative solutions that managers could implement when performance drops below aspirations (e.g., Audia & Greve, 2006; Bromiley et al., 2001; W. Chen & Miller, 2007; Iyer & Miller, 2008; Jordan & Audia, 2012; Miller & Chen, 2004). Mostly, these studies found that the alternative solutions are linked to adopting new routines and/or strategic approaches, such as investing in new innovations or engaging in acquisitions to accelerate growth and improve performance (e.g., Audia & Greve, 2006; Bromiley, 1991; W. Chen & Miller, 2007; Greve, 2003; Iyer & Miller, 2008; Miller & Chen, 2004).

Conversely, performance that exceeds aspiration levels reduces the need for

problemistic search, as firms do not desire to change their ‘effective’ routines and operating practices (Bromiley et al., 2001; Greve, 2003) and they avoid any risky actions that might result in performance falling below aspiration levels (March & Shapira, 1987). Moreover, continuing with prior effective strategic approaches is regarded as more efficient than exploring new, unproven alternatives because it allows firms to benefit from earlier

investment in the requisite skills and techniques (Audia, Locke & Smith, in Kim et al., 2015). Consequently, managers strive to increase performance above aspiration levels, but they are less willing to take risk to increase performance further when the firm is already performing above its aspiration levels (Fiegenbaum, Hart, & Schendel, 1996; Greve, 2003; Iyer & Miller, 2008; Miller & Chen, 2004).

(13)

Additionally, it is argued that performance below aspiration affects the firm’s risk preference (Audia & Greve, 2006; Boyle & Shapira, 2012; Bromiley, 1991; Desai, 2008; Miller & Chen, 2004; Palmer & Wiseman, 1999). According to March and Shapira (1987) risk can be defined as the ‘variation in the distribution of possible outcomes, their likelihood and their subjective values’ (March & Sharipa, 1987, p. 1404). When the variance of the possible (negative or positive) outcomes increases it becomes less predictable. Accordingly, investments are considered as highly risky when there is a high variance of possible negative or positive outcomes. However, managers and executives only associate negative outcomes with risk. Moreover, they consider uncertainty as a factor in risk, while the magnitudes of possible bad outcomes seemed more salient to them (March & Shapira, 1987). March and Shapira (1987) defined this managerial definition for risk as downside risk. Therefore, downside risk can be described as ‘the likelihood that a decision results in an organizational outcome that is below the target value’. This research adopts the latter definition of risk. Various studies that adopted the performance feedback theory and extended it with theories of organizational risk-taking argue that when a firm’s performance relative to aspirations

declines, the firm’s willingness to take greater risk increases (Audia & Greve, 2006; Boyle & Shapira, 2012; Bromiley, 1991; Desai, 2008; Lim & McCann, 2014; Miller & Chen, 2004; Palmer & Wiseman, 1999). These studies found that when firms are facing performance shortfalls they are more inclined to accept the risks inherent in adopting new routines in attempt to overcome the performance gap. On the contrary, the tendency of managers to choose solutions with greater risk decreases when performance is above aspiration level, since there is no desire to change effective operating practices. Moreover, they rather avoid risky actions that might negatively affect their performance than attempt to increase their

(14)

In sum, past research has shown that underperformance regularly stimulate firms to take action by engaging in problemistic search to remedy the performance gap. This leads to organizational learning, risk-assertiveness and change. Whereas aspiration congruent

performance tends to result in risk-aversive behavior and fuels the persistency of exploiting prior strategic actions that have been proven successful (Greve, 2003a; Bromiley et al., 2001, in Shinkle, 2011).

Literature gap and research question

The performance feedback theory has been often used to explain risk-taking behavior of firms. However, given the considerable research scholars still lack a thorough

understanding of what the effects of changes in performance are on organizational risk-taking behavior, since prior studies on this topic frequently yielded inconsistent empirical findings. More explicitly, different studies found significant effects for risk-taking behavior when performance falls further below aspiration levels. According to these studies, firms tend to take greater risk when their performance falls short of aspirations because they are more willing to accept the risks of adopting new routines in order to address the performance shortfalls (Bromiley, 1991; Palmer & Wiseman, 1999; Wiseman & Bromiley, 1996; Audia & Greve, 2006; Boyle & Shapira, 2012; Miller & Chen, 2004). On the contrary, there are also studies that found a contradicting effect suggesting that when firms perform below aspirations they tend to become more risk averse. These firms convert to conservative behaviors by restricting activities to core businesses and becoming reluctant to changes (Audia & Greve, 2006; Sitkin & Pablo, 1992, in Audia & Greve, 2006; McNamara & Bromiley, 1997, in Greve, 2003a; Miller & Bromiley, 1990; Wiseman & Bromiley, 1996). While other scholars have documented partial effects arguing that firms may in some cases respond conservatively to performance shortfalls due to different organizational determinants (Chen & Miller, 2007;

(15)

Desai, 2008; Iyer & Miller, 2008) or due to the focus on either historical aspirations or social aspirations (Kacperczyk et al., 2015; Kim et al., 2015).

When considering these studies, most of them assessed the relationship between performance feedback and organizational risk-taking behavior, but did not accounted for firm heterogeneity. Whilst firms under different organizational conditions are likely to process information on performance feedback differently, which induces distinctive organizational risk-taking behavior (Devers et al., 2007b; Tuggle et al., 2010, in Lim & McCann, 2014; O'Regan & Ghobadian, 2006; Slater, Olson, & Hult, 2006). Recent studies have come to recognize that a richer understanding of these organization conditions is necessary to further extend our knowledge regarding the effects of performance feedback on organizational risk-taking behavior. For example, Audia and Greve (2006) found that performance below aspirations in small firms decreases organizational risk-taking, whereas in large firms it has no effect or increases organizational risk-taking. Desai (2008) showed that when a firm is experiencing a performance gap relative to its aspirations its risk-taking behavior is dependent on its organizational context (i.e., operating experience, organizational legitimacy and age). Similarly, Kim et al. (2015) discovered that prior performance variability intensifies the relationship between aspirations and risk-taking behavior when firms are performing below aspirations, but weakens this relationship when firms are performing above aspirations. Hence to further extend our understanding of the relationship between performance feedback and organizational risk-taking behavior, it is crucial to consider the organizational conditions that influence this relationship.

An obvious, but overlooked organizational condition, which has great influence on organizational behavior, is a firm’s strategic orientation. A strategic orientation entails the firm’s ideology, which consists of its values, beliefs, and aspirations, and guides it long-lasting commitments and short-term responses to its environment (Fox-Wolfgramm et al.,

(16)

1998; Gatignon & Xuereb, 1997; O'Regan & Ghobadian, 2006; Slater et al., 2006).

Accordingly, the firm’s strategic orientation govern the way a firm responses to internal- and external cues. Consequently, when firms with different strategic orientations are facing performance shortfalls relative to aspirations, it is likely to observe distinctive organizational risk-taking behavior.

Hence, I seek to extend the set of organizational conditions by considering the moderating effects of strategic orientations on the relationship between performance below aspiration levels and organizational risk-taking behavior. Since a firm’s strategic orientation has great influences on the strategic behavior of firms (Fox-Wolfgramm et al., 1998;

Gatignon & Xuereb, 1997; O'Regan & Ghobadian, 2006; Slater et al., 2006), it is reasonable to suggest that a firm’s strategic orientation intensifies or attenuates the effects of

performance below aspirations on organizational risk-taking behavior. The aim of this research is to provide an in-depth understanding of the mechanisms of performance below aspirations on organizational risk-taking behavior, while taking into account the moderating effect of the firm’s strategic orientation. Therefore, the research question that this paper aims to answer is the following:

What is the moderating effect of a firm’s strategic orientation on the relationship between performance below aspiration and organizational risk-taking behavior, in terms of changes made in research and development (R&D) investments?

Performance feedback and R&D investments

Managers invest in R&D activities to identify and explore new innovations, and to strengthen their capabilities so that they can exploit current knowledge to enhance

(17)

Lewellyn & Bao, 2015). However, R&D investments often represent significant expenses and are incurred with a great amount of uncertain outcomes as these outcomes may include losses (Chen, 2008). More specifically, investing in R&D activities require large expenditures in the short term while their payoffs may come much later in the long term or it may not lead to any payoffs at all (David et al., 2001, in Lim & McCann, 2014). Moreover, R&D expenditures are mostly viewed as high risk investments when compared to capital expenditures on plant, equipment and/or property, which are less significant and bring about less uncertain outcomes (e.g., Bhagat and Welch, 1995; Kothari et al., 2001, in Coles, Daniel & Naveen, 2006).

Therefore, R&D investments can be considered as a risky decision for a firm to make. However, as these R&D activities allow for exploring long-term creation of genuinely new innovations it is still recognized as a method to improve organizational performance. So according to the behavioral theory of the firm, when managers find themselves in the situation that organizational performance is below their aspiration levels they resort to R&D

investments that could accelerate growth or safeguard new profits in attempt to ensure performance turnarounds (Antonelli, 1989; Hundley, Jacobson & Park, 1996; Kamien & Schwartz, 1982, in Greve, 2003). So following this reasoning, hypothesis 1a is as follow:

Hypothesis 1a: The effect of performance below aspiration on organizational risk-taking behavior is negative, such that further performance decreases relative to aspirations lead to an increase in R&D investments.

However, considering the contradicting results found on the effects of performance below aspiration on organizational risk-taking behavior, I set up another hypothesis that is competing with hypothesis 1a to help resolve the longstanding debate:

(18)

Hypothesis 1b: The effect of performance below aspiration on organizational risk-taking behavior is positive, such that further performance decreases relative to aspirations lead to a decrease in R&D investments.

The moderating effect of strategic orientations

Every firm reacts differently to its environment even when they are operating in the same industry (O'Regan & Ghobadian, 2006; Slater et al., 2006). The way a firm responses to its internal- and external cues is based on its strategic orientation. More specifically, a

strategic orientation expresses the firm’s philosophy of how to operate in its industry through a deeply rooted set of values, beliefs, motivations and aspirations that guides the strategy formulation and the deployment of operating practices to achieve superior performance (Fox-Wolfgramm et al., 1998; Gatignon & Xuereb, 1997; O'Regan & Ghobadian, 2006; Slater et al., 2006). Consequently, firms in the same industry with distinctive strategic orientations operate and react differently to internal- and external cues.

The strategic management literature provides several typologies of strategic

orientations. Some of the well-known typologies include Miles and Snow’s (1978) framework of alternative strategic orientations, which integrates organizational strategy, structure, and process variables into a theoretical model of co-alignment (Miles et al., 1978), and March’s (1991) exploration and exploitation strategies, which defines strategic orientation based on the firm’s choice to explore and/or exploit their operating practices. Other typologies that

examine strategic orientations are the ones of Porter (1980) who describes strategic

orientations in terms of cost leadership and product differentiation, and Treacy and Wiersema (1995) who determine strategic orientations in terms of operational excellence, product leadership and customer intimacy. Whilst the labels of strategic orientations vary across the different typologies, a common feature of all the proposed strategy classifications is that they

(19)

clearly distinguish between firms that operate at one end or the other end of a strategy continuum (Dent 1990; Langfield-Smith 1997; Seifzadeh 2011, in Bentley, Omer & Sharp, 2013). Furthermore, besides both ends of the strategy continuum, each typology makes an attempt to classify firms that operate under a mixed strategy (March, 1991; Miles et al., 1978; Porter, 1980; Treacy & Wiersema, 1995). Mostly, these mixed strategies contain, to varying degrees, the characteristics of firms at both ends of the strategy continuum (Bentley et al., 2013; March, 1991; Miles et al., 1978; Porter, 1980; Treacy & Wiersema, 1995).

There is no definitive view on which strategic orientation classification is most representable of the different strategic orientations that firms could adopt. However, this paper focuses on Miles and Snow’s typology of strategic orientations, as their strategic types are most aligned with the several typologies mentioned above, therefore, inferences based on Miles and Snow’s typology are likely to align with inferences based on the other typologies (Bentley et al., 2013). More specifically, this research focuses on the two distinct strategic orientations that include the one end or the other end of the strategy continuum. Miles and Snow (1978) refer to the firms that are operating under such strategic orientations as Prospector firms and Defender firms. The characteristics of Prospector firms and Defender firms are similar to the strategic orientations suggested by March (1991) as Exploration and Exploitation, by Porter (1980) as Product Differentiation and Cost Leadership, and by Treacy and Wiersema (1995) as Product Leadership and Operational Excellence. Hence, these commonalities between Miles and Snow’s strategic types and the other strategic orientation typologies allow for a better alignment of inferences across the classifications. Furthermore, another reason for selecting Miles and Snow’s typology of strategic orientation is that this typology can be operationalized using archival data, while other typologies require qualitative data, such as personal interviews and surveys of corporate officers (Bentley et al., 2013). Thus, by applying Miles and Snow’s typology in this paper, it produces a replicable measure

(20)

of strategic orientations that allow for generalization to a broad cross-section of firms. Also, this typology has been debated and supported by many researchers (e.g., Blackmore &

Nesbitt, 2013; Ghoshal, 2003; Hawes & Crittenden, 1984; McDaniel & Kolari, 1987; Snow & Hrebiniak, 1980; Zahra & Pearce, 1990). Additionally, studies have found support for either some or all of the strategic types across a number of research domains (e.g., Atkins, 1994; Hambrick, 1983; Parnell and Wright, 1993; Smith et al., 1986, 1989; Snow and Hrebiniak, 1980; Subramanian et al., 1993; Tan, 1997; Weisenfeld-Schenk, 1994; Zajac and Shortell, 1989, in Blackmore & Nesbitt, 2013; Bentley et al., 2013; Higgins et al., 2015; McDaniel & Kolari, 1987; Nicholas O’Regan & Ghobadian, 2006; Noble, Sinha & Kumar, 2002), which makes this typology academically acceptable and internally consist.

Miles and Snow’s typology of strategic orientations

Miles et al. (1978) believe that firms develop relatively stable patterns of strategic behavior that align them with their external environment (Conant, Mokwa, & Varadarajan, 1990). By examining this ‘adaptive cycle’, which characterizes the strategic behavior of the firms, they determined that this process involves three essential strategic ‘problem and

solution’ sets consisting of: (1) an entrepreneurial problem set focusing on the definition of an organization’s product/market domain; (2) an engineering problem set centering on the choice of technologies and operating processes to be employed for production and distribution; and (3) an administrative problem set concerning the selection, rationalization, and development of organizational structure and policy processes (Conant et al., 1990; Miles et al., 1978). Each problem set has multiple dimensions and by combining these dimensions they created a framework in which a continuum of firms exists (Miles et al., 1978). Miles et al. (1978) define four distinct strategic types that exist along this continuum, which are Prospectors, Analyzers, Defenders and Reactors. The firm’s choice regarding where it will operate on this

(21)

strategy continuum is, to a large extent, based on its prevailing trait. This prevailing trait is the result of the influence of its key decision makers, and its perceived view of its external

environment (Miles et al., 1978; Nicholas O'Regan & Ghobadian, 2006). Appendix 1 provides a summary of the distinctive strategic types along the different dimensions.

The Prospector, Analyzer and Defender strategic types are considered to have a distinct, consistent and repetitive behavior regarding the way they evaluate their operating environment (Parnell & Hershey, 2005), their range of product/market domains (Snow & Hrebiniak, 1980), their use of technology to solve problems (Miles et al., 1978), and their innovativeness (Blackmore & Nesbitt, 2013; Parnell & Hershey, 2005). On the contrary, the Reactor strategic type lacks a consistent strategic approach for solving problems and due to this it is usually considered unviable and is often omitted from studies (Conant et al., 1990; Shortell & Zajac, 1990, in Blackmore & Nesbitt, 2013). As already mentioned before, this study is interested in the two viable strategic orientations that comprise the two endpoints of the strategy continuum, which are the Prospector and Defender strategic orientations.

Prospector firms and Defender firms

Firms that are identified as Prospector firms are firms that continuously search for new products and market opportunities through processes of innovation and development in their products and services (Miles et al., 1978). They operate in a dynamic environment that requires an innovation strategy, which entails identifying and pursuing opportunities, experiencing continuous change, embracing uncertainty, and stimulating risk-taking. This innovation strategy is necessary in order to maintain their reputation as an innovator in product- and market developments (Miles et al., 1978). To serve their dynamic environment and maintain their competitive edge, Prospector firms are regularly the creators of change in their industries, which means they typically experience a great deal of environmental change

(22)

and uncertainty (Miles et al., 1978). Since Prospector firms are continuously facing change, they have to be highly flexible and entrepreneurial (Ghoshal, 2003). Due to their focus on innovation it requires them to develop multiple technologies for a distinct product mix and they avoid long-term commitments to a single technological process by leveraging the knowledge and skills of their employees and by adopting a low degree of mechanization (Bentley et al., 2013; Miles et al., 1978).

In contrast, firms that are identified as Defender firms are firms that are deliberately enacting and maintaining stability in their operating environment (Miles et al., 1978). They often produce narrow related products and services to enable production- and distribution efficiencies. To achieve these efficiencies Defender firms invest in enhancing their current routines, which includes focusing on ‘single core’ cost-efficiency technology and continual improvements that lead to routinization and mechanization (Bentley et al., 2013; Miles et al., 1978). Unlike Prospector firms, Defender firms do not aggressively pursue new product and market opportunities and tend to ignore developments and trends outside their strategic domain, as that would upset their stable environment and limit their efficiency efforts. Instead, they adopt a cost leadership strategy in which they compete on price, service, and/or quality, and grow through market penetration and some limited product developments (Ghoshal, 2003; Miles et al., 1978). Also, to maintain the stability Defender firms minimize their exposure to risk and uncertainty, while lessen their pursue to explore and adopt new routines (Ghoshal, 2003; Miles et al., 1978).

Furthermore, when considering the risk preferences of both Prospector firms and Defender firms, Prospector managers are more able to withstand operations that involve uncertainty and risk than Defender managers (Ghoshal, 2003; Miles et al., 1978; Rajagopalan & Finkelstein, 1992). Moreover, Prospector firms adopt compensation plans, such as bonus plans and stock option plans, that have higher payout rates than the Defender firms’

(23)

compensation payout rates to encourage the risk-taking behavior of managers (Balkin and Gomez-Mejia, 1987, in Chen & Jermias, 2014; Rajagopalan & Finkelstein, 1992). Therefore, I expect Prospector managers to address strategic challenges more aggressively and

assertively, in terms of organizational risk-taking behavior, than Defender managers. Following this reasoning, Prospector managers should interpret performance shortfalls relative to aspirations as a challenge that should be tackled by identifying and exploring alternative innovations to enhance their performance and to recover their competitive edge. Thus, stimulated by the performance shortfall and the inclination to mend the gap, Prospector managers will become more prone to take greater risk by resorting to higher R&D

investments. Whereas, Defender managers should perceive performance shortfalls relative to aspirations as a challenge that should be addressed by operational efficiencies that could be obtained by turning their focus towards alternative product- and distribution proficiencies. Furthermore, since Defenders managers have little capacity for exploring and exploiting new areas of opportunities outside their strategic domain, they will focus on enhancing their current operating practices to boost their efficiencies (Blackmore & Nesbitt, 2013; Ghoshal, 2003; Higgins, Omer, & Phillips, 2015; Miles et al., 1978). Hence, threatened by the

performance shortage, the stability discontinuity and their low risk tolerance, Defender managers will restrict their attention to the enhancement of their operating practices and lessen their risk exposure by reducing R&D investments. Therefore, the next hypothesis is as follow:

Hypothesis 2: Strategic orientation negatively moderates the effect of performance below aspiration and R&D investments, such that the Prospector strategic orientation amplifies the relationship when performance falls further below the aspiration.

(24)

In addition, when considering the aspiration preferences of Prospector managers and Defender managers, it is reasonable to assume that both Prospector managers and Defender managers are likely to attend more to one or another aspiration level. Since Prospector firms are operating in a dynamic and fast-changing environment, which requires them to bring about newer and better innovations than their competitors (Miles et al., 1978), they are more likely to rely heavily on social aspirations than on historical aspirations. Furthermore, in order to stay a head of the competition Prospector managers are expected to have an external focus by assessing their competition and by comparing their performance with the performance of competitors (Miles et al., 1978). Therefore, social aspiration levels would be a more

appropriate reference point for Prospector managers to rely on, than historical aspiration levels. Hence, this leads to the next hypothesis:

Hypothesis 3a: Prospector managers are more likely to attend to social aspirations, than historical aspirations.

While, Defender managers are more focused on internal process efficiencies and less concerned about the developments outside their strategic domain (Ghoshal, 2003; Miles et al., 1978). Since they aim to gain a competitive advantage through operational stability, and production- and distribution efficiencies (Miles et al., 1978), they are more likely to consider historical aspiration levels as a relevant reference point to evaluate their performance.

Furthermore, as Defender managers seek to optimize operational efficiencies, they need to focus on their own past performance to examine the areas of improvements (Ghoshal, 2003; Miles et al., 1978). Therefore, I hypothesize the following:

Hypothesis 3b: Defender managers are more likely to rely on historical aspirations than on social aspiration.

(25)

Methodology

This chapter explains the research approach and design of this paper. First, the sampling strategy will be discussed. Then, the section continues with a detailed

operationalization of the dependent, independent and control variables. Finally, the last section details the models used to analyze the data.

Sampling strategy

This study used secondary data that was gathered from Wharton CRSP – Compustat Merged database. The combined CRSP – Compustat dataset provided annual financial- and industry data from 1979 to 2014. The data was restricted to manufacturing industries with SIC codes from 2000 to 3999 to allow comparisons with previous studies (e.g., Chen & Miller, 2007; Chen, 2008; Iyer & Miller, 2008; Lim & McCann, 2014) and to prevent confounding results due to major differences in the industries’ activities (Brush, 1996; Davis & Thomas, 1993; Hitt & Hoskisson, 1996, in Iyer & Miller, 2008). The manufacturing sample consists of 43,539 firm-year observations. The largest two-digit industry segments are electronic and other electrical equipment and components (SIC 3600), measuring, analyzing and controlling instruments (SIC 3800), and industrial and commercial machinery and computer equipment (SIC 3500). These industries account for 23%, 19%, and 18% of all firms in the sample. Furthermore, the independent and control variables lag the dependent variable by one year. Hence, the independent and control variables ranged from 1979 to 2013, and the dependent variable corresponded to the years 1980 to 2014.

Dependent variable

(26)

This proxy for organizational risk-taking behavior is consistent with a number of prior studies (e.g., Hitt, Hoskisson, Ireland, & Harrison, 1991; Lee & O'neill, 2003; Lim & McCann, 2014). The R&D intensity variable has a relatively high mean of 0.20 and the standard deviation of 3.48. Therefore following past studies (e.g., Chen & Miller, 2007; Lim & McCann, 2014), I focused only on firms with R&D intensity less than or equal to 1.0, and removed 480 firm-years observations from the sample. The resulting sample has a mean of 0.10 and a standard deviation of 0.12, which similar to Chen and Miller’s (2007), and Lim and McCann (2014) results.

Independent variables

Firm Performance. To measure firm performance I used ROA (net income divided

total assets) because it is the primary measure of firm profitability within the manufacturing industries and it has been applied in numerous prior studies (e.g., Audia & Greve, 2006; Bromiley, 1991; Chen & Miller, 2004; Desai, 2008; Greve, 2003; Iyer & Miller, 2008; Lim & McCann, 2014). As firm performance is evaluated against historical- and social aspiration levels, I included a variable that measures the focal firm’s own past performance (historical aspiration level) and a variable that measures the past performance of its peer group firms in the same industry (social aspiration level). The historical aspiration level was calculated as the average ROA of the firm’s past three years. The social aspiration level was calculated as the mean ROA of all companies within each industry (defined at the four-digit SIC level). Actual performance was measured at year t – 1, and each aspiration level was proxied at year t – 2. Subsequently, the firm’s performance feedback was computed. The historical performance feedback was calculated by the firm’s ROA – firm’s average ROA of last three years. The social performance feedback was constructed as firm’s ROA – the mean ROA of all companies within each industry (defined at the four-digit SIC level). Then, spline variables

(27)

were created by splitting the firm’s (historical and social) performance feedback variables into two variables to construct a variable for performance below aspiration and a variable for performance above aspiration (Desai, 2008). The performance below aspiration variable only has observations that are smaller than zero, whereas the performance above aspiration

variable only has observations that are greater than zero.

Moderating variables

Strategic orientations. To assign firms to different strategic orientation types, I

computed a discrete STRATEGY composite measure that proxies for a firm’s strategic orientation. The measure is based on variables that reflect different characteristics of the Miles and Snow (1978) typology of strategic orientations. Following Bentley et al. (2013) and Higgins et al. (2015), I used the following variables to construct the STRATEGY composite measure: (a) the ratio of research and development to sales (RDS5), (b) the ratio of employees to sales (EMPS3), (c) a historical growth measure (one-year percentage change in total sales), using the market-to-book ratio as a proxy for growth (REV3), (d) the ratio of marketing (SG&A) to sales (SGA3), and (e) capital intensity using PPENT scaled by total assets (CAP3). All the variables are computed with the rolling average of the past five years. Each of the five variables is used to capture different elements of a firm’s strategic orientation (Higgins et al., 2015). The variable RD3 is a proxy for a firm’s propensity to seek new products, as Prospector firms engage in greater amounts of innovative operations, they are expected to have higher research and development expenditures than Defenders (Bentley et al., 2013; Higgins et al., 2015; Miles et al., 1978). The variable EMPS3 measures a firm’s ability to produce and distribute its goods and services efficiently. Since defender firms focus on organizational efficiency, they are expected to have fewer employees per dollar of sales. The variable REV3 represents a proxy for a firm’s historical growth, which would be lower

(28)

and steadier for Defender firms than for Prospector firms as Prospector firms grow through product- and market development, which occurs in high spurts (Miles et al., 1978). The variables SGA3 measures a firm’s marketing activities and because Prospector firms focus more on the marketing, and research and development functions, while Defender firms focus more on the finance and production functions, Prospector firms are expected to have greater marketing activities (Bentley et al., 2013; Snow & Hrebiniak, 1980). The last variable of the STRATEGY measure is CAP3 and this variable measures the efficiency and automation of operations as reflected in overall capital intensity. According to Hambrick (1983), Defender firms are more automated and efficient so he concludes that Defender firms are more capital intensive than Prospector firms (Bentley et al., 2013). Also, originally the STRATEGY measure consists of six variables, however I excluded one variable (employee fluctuation) due to too many missing observations in this variable. I believe that the omission of this variable does not affect the validity of measure negatively, as this STRATEGY measure yield similar results when compared to the results and company examples of Bentley et al.’s (2013) and Higgins et al.’s (2015) studies.

To construct the STRATEGY measure, the five variables are ranked by forming quintiles within each two-digit SIC industry-year (Bentley et al., 2013; Higgins et al., 2015). Within each company-year, the observations with variables in the top quintile are give a score of 5, those in the next quintile receive a score of 4, and so on, and those observations with variables in the lowest quintile are given a score of 1. However, the scores of CAP3 have to be revised, so that a score of 1 will be 5, a score of 2 will be 4 and so on. Subsequently, for each firm-year, the scores across the five variables are summed as such that a company can receive a maximum score of 25 (Prospector-type) and a minimum score of 5 (Defender-type) (Bentley et al., 2013). I treated the STRATEGY measure as a continuous variable, which implies that the higher the STRATEGY scores of a firm the more it engages in a Prospector

(29)

strategic orientation, while the lower the STRATEGY scores of a firm the more it engages in a Defender strategic orientation.

Control variables

Firm size. As the size of a firm might increase firm risk-taking (Wright et al. 2007, in

Lim & McCann, 2014), I control for firm size. The measure is constructed as the number of employees, which is an appropriate measure of overall firm size in a given industry (Audia & Greve, 2006). I logged the number of employees because this specification better captures the effect of size on risk-taking (Audia & Greve, 2006). According to Audia and Greve (2006), by logging the number of employees it ensures that an increase of given percentage has the same effect, regardless of firm size.

Organizational slack. Since slack resources may influence organizational risk-taking,

I controlled for organizational slack (e.g., Singh, 1986). Consistent with prior studies, I chose the working capital-to-sales ratio and current ratio (current assets-to-current liabilities ratio) as slack proxies (e.g., Chen & Miller, 2007; Lim & McCann, 2014; Singh, 1986).

Distance from bankruptcy. According to March and Shapira (1992) firms tend to be

more risk averse when facing bankruptcy. Therefore, to control for risk aversion I included the Altman’s Z measure as a control variable in this research. Following past studies (e.g., Chen & Miller, 2007; Iyer & Miller, 2008; Lim & McCann, 2014; Miller & Chen, 2004) I measured the distance from bankruptcy with Altman’s Z-score 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 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.

(30)

Industry R&D intensity. To account for industry effects, I included an industry-level variable at the four-digit SIC level. I accounted for industry R&D intensity to control for differences in R&D intensity across industries. This variable is measured by the average of firm R&D intensity in the industry (Gentry & Shen, 2013).

Statistical model

The cross-section and time-series nature of the data were taken into consideration when choosing the estimation method. To check for individual and time effects, I ran a Breusch-Pagan Lagrange multiplier test. The test rejected the null hypothesis ( (1) = 19529.27, p < 0.01), which demonstrates that there are significant differences across firms in the data and indicates that the simple OLS estimation is not appropriate for this study.

Subsequently, I performed the Hausman’s specification test to determine whether to employ the within (fixed effects) or GLS (random effects) estimator (Mutl & Pfaffermayr, 2011). The results of the test were statically significant ( (7)= 15.74, p < 0.05), which indicates that a GLS (random effects) estimator cannot adequately account for firm effects and time effects in the data. Therefore, I ran all analyses using the within (fixed effects) model to control for both firm effects and time effects in the analysis.

Following Lim et al. (2014), I mean-centered all the variables before creating the interaction variables to mitigate any potential multicollinearity problems. In order test the hypotheses I used STATA statistical package to run the within (fixed effects) models.

(31)

Results

The following section reports the results of this research. First, the descriptive statistics for the variables of the study are presented to provide an overview of the data. Subsequently, a

correlation analysis is performed and the significant correlations are reported. Finally, several multiple regressions were carried out to test the hypotheses that were formulated in the theoretical part of this paper.

Descriptive statistics and correlation analysis

Table 1 provides descriptive statistics and the bivariate correlation analysis of the variables for the final dataset. The relationships were investigated using Pearson product-moment correlation coefficient. First, the correlations between the dependent variable and control variables are presented. Subsequently, the correlations between the dependent variable and the independent variables are outlined. Finally, some of the correlations between the moderating variable and independent variables, and the correlations between the moderating variable and the control variables are discussed.

The sample consisted of 43,539 observations from 1979 to 2014. The major two-digit industry segments were electronic and other electrical equipment and components (23%), measuring, analyzing and controlling instruments (19%), and industrial and commercial machinery and computer equipment (18%). When considering the correlations between the dependent variable and the control variables, R&D intensity was negatively correlated with firm size (r = -0.30). In contrast, there was a positive correlation between R&D intensity and organization slack, which was slightly stronger for absorbed slack (r = 0.28) than for

unabsorbed slack (r = 0.22). With regard to the correlation between R&D intensity and the distance from bankruptcy, it was a very weak negative correlation (r = -0.04). In the case of

(32)

the last control variable, industry R&D intensity, there was also a weak correlation with R&D intensity (r = 0.16).

Regarding the correlation between R&D intensity and STRATEGY, these variables are medium correlated (r = 0.45). In addition, R&D intensity is negatively correlated with performance below historical aspiration (r =- 0.17), whereas it is positively correlated with performance above historical aspiration (r = 0.08). Likewise, R&D intensity is also

negatively correlated with performance below social aspiration (r = -0.23). A very weak correlation was found between R&D intensity and performance above historical aspiration (r = 0.08). Additionally, R&D intensity is merely correlated with performance above social aspiration (r = -0.01). Also, the historical aspiration and social aspiration variables are highly correlated for firms performing below their aspirations (r = 0.84). Therefore following prior studies, separated models are used for historical and social aspiration to avoid distorted parameter estimates (e.g., Chen & Miller, 2007; Iyer & Miller, 2008; Lim & McCann, 2014).

When considering the moderating variable with the independent variables,

STRATEGY is negatively correlated with performance below historical aspiration (r =-0.11), whereas it is positively correlated with performance above historical aspiration (r = 0.15). Also, STRATEGY and performance below social aspiration are negatively correlated (r = -0.20). Similarly, STRATEGY and performance above social aspiration are negatively correlated, but the correlation is weaker (r = -0.05). STRATEGY and firm size are medium correlated (r = 0.41). Furthermore, a very weak positive correlation was found between STRATEGY and absorbed slack (r = 0.05), but a modest positive correlation was found between STRATEGY and unabsorbed slack (r = 0.21). For the full set of correlation, please refer to table 1.

(33)

Table 1 Descriptive statistics

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

N Mean S.D. 1 2 3 4 5 6 7 8 9 10 01. R&D intensity 42,487 0.09 0.12 02. Firm size 43,539 6.87 2.10 -0.30** 03. Absorbed slack 43,318 0.86 17.91 0.28** -0.04** 04. Unabsorbed slack 43,318 3.32 3.34 0.22** -0.30** 0.08** 05. Distance from bankruptcy 40,806 1.46 29.55 -0.04** 0.03** 0.00** 0.02** 06. Industry R&D intensity 43,539 0.28 1.00 0.16** -0.08** 0.11** 0.06** -0.02** 07. STRATEGY 43,539 14.66 4.08 0.45** 0.41** 0.05** 0.21** -0.04** 0.06** 08. Performance below aspirations (historical) 36,105 -0.06 0.27 -0.17** 0.15** 0.00** 0.04** 0.05** -0.03** -0.11** 09. Performance above aspirations (historical) 36,105 0.05 0.21 0.08** -0.15** 0.01** 0.02** -0.02** 0.04** 0.15** 0.06** 10. Performance below aspirations (social) 43,211 -0.08 0.37 -0.23** 0.22** -0.01** 0.06** 0.13** -0.03** -0.20** 0.84** -0.01** 11. Performance above aspirations (social) 43,211 0.12 0.21 0.01** 0.00** 0.00** 0.09** 0.01** 0.18** -0.05** 0.09** 0.48** 0.12**

(34)

Regression analysis

Table 2 and table 3 present a summary of the estimation results. Model 1 shows a base model consisting only of the control variables. The independent variables are added in model 2 and 4, and the moderating effect is shown in model 3 and 5. Model 6 and 8 display the base model with the control variables for the Prospector and Defender sample. Model 7 and 9 contain the base model with the dependent variables for the two samples.

Cautious is needed when interpreting the coefficient for performance below aspiration levels (historical and/or social). Following the interpretation of past studies (Chen & Miller, 2007; Iyer & Miller, 2008; Lim & McCann, 2014), a positive coefficient on firms that are performing below their aspirations means that the further past performance drops below the aspiration level, the lower the R&D intensity, whereas a negative coefficient means that the further the past performance falls below their aspirations, the higher the R&D intensity. Hypothesis 1 postulates that the effect of performance below aspiration on organizational risk-taking behavior is negative or positive. When considering the coefficients in model 2 and model 4, they are significant and negative for the main effect of performance below aspiration level for both historical (t = -9.43 p <0.001), and social aspiration (t = -13.22 p <0.001). This suggests that R&D intensity increases as performance falls further below aspiration. Hence, hypothesis 1a is supported, whereas hypothesis 1b found no support in the overall analysis.

Hypothesis 2 states that a firm’s strategic orientation negatively moderates the

relationship between performance below aspiration levels and organizational risk-taking, such that high values of strategic orientations (Prospector strategic orientation) strengthen the relationship. Model 3 and 5 show negative and significant coefficients for the interaction term between performance below historical (t = -4.75, p < 0.001) and social (t = -8.71, p <0.001) aspiration and STRATEGY. This suggests that the interactive effect leads to higher R&D intensity. Therefore, hypothesis 2 is supported. Figure 1 and figure 2 depicts the differing

(35)

interaction effects of low and high values of STRATEGY with performance below historical and social aspiration for underperforming firms. The lowest STRATEGY score of 5 and the highest STRATEGY score of 25 were used to compute the simple slopes, in which the STRATEGY score of 5 depicts the low values of STRATEGY and the STRATEGY score of 25 represent the high values of STRATEGY. Subsequently, I used -1 and 0 as the values of performance below aspiration, such that when performance below aspiration drops from 0 to -1 it means that the firms are experiencing further performance shortfalls relative to aspiration levels. Moreover since this study focuses on underperforming firms, only values of

performance below aspiration that are below 0 are shown in the plots because values above 0 depict firms that are performing above aspiration. In addition following the approach of Sibley (2008) based on Aiken and West simple slope analysis, a test was performed to assess whether the low and high simple slopes of STRATEGY are different from zero or not. In the case of historical aspiration, the effect of performance below historical aspiration is negative and significant for low values of STRATEGY (t = -9.26, p < 0.001) and high values of STRATEGY (t = -5.87, p < 0.001). Likewise in the case of social aspiration, the effect of performance below social aspiration is negative and significant for low values of STRATEGY (t = -15.06, p < 0.001) and high values of STRATEGY (t = -9.472, p < 0.001). Hence when interpreting the simple slopes in figure 1, high values of STRATEGY intensify the risk-seeking effects of performance below historical aspirations. Similarly, figure 2 also

demonstrates that high values of STRATEGY strengthen the negative effect of performance below social aspiration on R&D investments. Additionally, figure 1 and figure 2 show that low values of STRATEGY also amplify the relationship between performance below (historical and social) aspiration and R&D investments, but the interaction effects are less strong than the interaction effects of high values of STRATEGY.

(36)

For hypothesis 3, I divided the sample into low and high STRATEGY scores to test for differences in aspiration preferences among Defender firms and Prospector firms. I did this by restricting the sample to the lowest quintile of the STRATEGY measure and the highest quintile of the STRATEGY measure. The lowest quintile (0%-25%) represents Defender firms that have scores between 5 and 12, and the highest quintile (75%-100%) denotes Prospector firms that have scores ranging from 17 to 25. Hypothesis 3a proposes that firms with a Prospector strategic orientation are more likely to attend to social aspirations then historical aspirations. Model 7 presents the results for the Prospector sample and shows a negative and significant coefficient for performance below social aspiration (t = -5.27, p <0.001), and a negative and significant coefficient for performance above social aspiration (t = -3.43, p <0.001). The coefficients for performance below and above historical aspiration level are not statistically significant. This indicates that Prospector firms attend to social aspiration levels both when they are performing below or above aspirations. Hypothesis 3b states that firms with a Defender strategic orientation are more likely to rely on historical aspirations than on social aspirations. Model 9 includes the Defender sample and shows a positive and significant coefficient for performance below historical aspiration (t = 2.85, p <0.001) and a negative and significant coefficient for performance below social aspiration (t = -1.92, p = 0.05), which suggests that Defender firms rely on both historical and social aspirations when they perform below aspirations. No significant effects were found for performance above historical and social aspirations. Hence, only hypothesis 3a is supported. Regarding the effects of performance below and above aspiration on R&D intensity when considering each sample, the coefficient for performance below social aspiration for Prospector firms indicates a negative relationship, which suggests that the further

performance drops below social aspiration, the more R&D investments will rise. Whereas, the coefficient for performance above social aspiration implies that an increase in performance

(37)

above social aspiration leads to a decrease in R&D investments. With respect to Defender firms, a positive coefficient of performance below historical aspiration indicates that the further past performance falls below historical aspiration, the lower the R&D investments will be. Whilst, the coefficient for performance below social aspiration denotes that the further past performance drops below social aspiration, the higher the R&D investments.

Figure 1 Moderating effect of strategic orientation on the relationship between performance below historical aspiration and organizational risk-taking

Figure 2 Moderating effect of strategic orientation on the relationship between performance below social aspiration and organizational risk-taking

(38)

Table 2. Within (fixed effects) model for R&D intensity

Historical aspiration Social aspiration

Model 1 Model 2 Model 3 Model 4 Model 5

Variables Control SE Main SE Interaction SE Main SE Interaction SE

Firm size _{-0.005** 0.001} _-0.004** _0.001 _-0.004** _0.001 _{-0.003** 0.001} _{-0.003** 0.000} Absorbed slack 0.003** 0.000 0.004** 0.000 0.004** 0.000 0.003** 0.000 0.003** 0.000

Unabsorbed slack 0.003** 0.000 0.003** 0.000 0.003** 0.000 0.004** 0.000 0.004** 0.000

Distance from bankruptcy 0.000** 0.000 0.000** 0.000 0.000** 0.000 0.000** 0.000 0.000** 0.000

Industry R&D intensity -0.003** 0.001 -0.003** 0.001 -0.003** 0.001 -0.002** 0.001 -0.002** 0.001

STRATEGY 0.002** 0.000 0.002** 0.000 0.003** 0.000 0.003** 0.000

Performance below aspirations -0.025** 0.003 -0.022** 0.003 -0.032** 0.002 -0.021** 0.003

Performance above aspirations _{-0.004†* 0.002} _{-0.003** 0.002} _{-0.007** 0.002} _{-0.010** 0.002} Performance below aspirations x

STRATEGY -0.004** 0.001 -0.006** 0.001

Performance above aspirations x

STRATEGY -0.001** 0.001 -0.003** 0.001

Model F 13.03** 10.31** 10.12** 10.51** 10.37**

R2 0.644** 0.625** 0.629** 0.646** 0.648**

N 37,377*_* 32,235*_* _32,235** _37,083* _37,083**

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

(39)

Table 3. Within (fixed effects) model for R&D intensity with Prospector and Defender sample

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

Prospector Defender

Model 6 Model 7 Model 8 Model 9

Variables Control SE Main SE Control SE Main SE

Firm size -0.008** 0.002 -0.007** 0.002 -0.001** 0.002 -0.002** 0.002

Absorbed slack 0.003** 0.000 0.004** 0.000 -0.089** 0.002 -0.090** 0.002

Unabsorbed slack 0.005** 0.000 0.005** 0.000 0.008** 0.001 0.010** 0.001

Distance from bankruptcy 0.000** 0.000 0.000** 0.000 -0.000** 0.000 -0.002** 0.002

Industry R&D intensity 0.000** 0.000 -0.001** 0.001 -0.006** 0.001 -0.007** 0.001

Performance below historical aspiration _{-0.005†* 0.009} _0.059** _0.021

Performance above historical aspiration -0.007** 0.006 -0.010†* 0.016

Performance below social aspiration -0.044** 0.008 -0.035** 0.018

Performance above social aspiration -0.022** 0.006 0.002** 0.006

Model F _9.97** _9.31** 4.55** _4.24***

R2

0.728** 0.740** 0.406** 0.404**

Performance feedback and organizational risk-taking behavior : the moderating effects of strategic orientations