INTRODUCTION
Over the past two decades, the relationship between entrepreneurial orientation (EO) and firm performance has gained increasing attention in contemporary strategic management and organization research (Davidsson, Delmar, & Wiklund, Reference Davidsson, Delmar and Wiklund2002; Wiklund & Shepherd, Reference Wiklund and Shepherd2005; Tang, Tang, Marino, Zhang, & Li, Reference Tang, Tang, Marino, Zhang and Li2008). Although a high level of EO may enhance firms’ abnormal returns and growth, solely pursuing a strong EO strategy may not be sufficient to improve firm performance (Covin & Slevin, Reference Covin and Slevin1989; Ireland, Hitt, & Sirmon, Reference Ireland, Hitt and Sirmon2003; Stam & Elfring, Reference Stam and Elfring2008). Therefore, several questions are raised: does EO foster a set of processes or activities that improve firm performance? Or is the EO–performance relationship less straightforward? What contextual variables might influence the EO–performance relationship? Based on the prior literature, this study argues that the external environment (specifically, environmental dynamism) and internal resources (specifically, the rareness of resource–capability combinations) may influence the relationship between EO and firm performance.
Regarding the EO–performance relationship, Lumpkin and Dess (Reference Lumpkin and Dess1996) have proposed three alternative models: the independent effect model, the mediating effect model, and the moderating effect model. The independent effect model views EO as an independent variable that directly influences firm performance. The mediating effect model suggests that internal organizations may mediate the association between EO and firm performance. The moderating effect model proposes that the EO–performance relationship varies depending on a number of contingency variables, such as environmental factors and internal organizational factors (Covin & Slevin, Reference Covin and Slevin1991; Zahra, Reference Zahra1993; Lumpkin & Dess, Reference Lumpkin and Dess1996).
The conceptual arguments of Lumpkin and Dess (Reference Lumpkin and Dess1996) have been examined in several empirical studies. However, the findings to date are inconsistent. Some studies report that firms with strong EO perform better than firms without strong EO (Miller, Reference Miller1983; Miller & Friesen, Reference Miller and Friesen1983; Wiklund, Reference Wiklund1999), confirming the independent effect model of the EO–performance relationship. Some empirical studies find an insignificant relationship between EO and firm performance (Covin & Slevin, Reference Covin and Slevin1989; Smart & Conant, Reference Smart and Conant1994). Furthermore, some scholars argue that there are inconsistent interaction effects between EO, performance, and contextual factors. For instance, Miller (Reference Miller1988) finds that firms that have a strong EO and compete in a dynamic environment outperform other firms that have strong EO but that compete in a relatively stable environment. However, Hart (Reference Hart1992) argues that EO may be associated with poor performance in dynamic and complex environments because the innovative and proactive activities of organizational members in such contexts are less controlled by top managers in such situations. Moreover, Lumpkin, and Dess (Reference Lumpkin and Dess2001) and Wiklund and Shepherd (Reference Wiklund and Shepherd2005) present similar findings. Given all of these findings, the association between EO and firm performance remains an open research question. More empirical evidence is needed.
In addition to external factors, internal factors such as the rareness of resource–capability combinations (rareness) may also influence the EO–performance relationship. Rareness refers to a valuable resource that can be possessed by many firms, but this resource is paired with the appropriate capability by only a few firms (Barney, Reference Barney1991; Newbert, Reference Newbert2008). To improve performance, firms with EO would have to be aware of the value of their resources and capabilities, and such awareness may differ from that of these firms’ competitors (Schumpeter, Reference Schumpeter1934; Busenitz & Barney, Reference Busenitz and Barney1997). This argument is consistent with the resource-based view (Barney, Reference Barney1991; Newbert, Reference Newbert2008). Wu, Chang, and Chen (Reference Wu, Chang and Chen2008) argue that when firms have strong EO, the positive influence of human capital (a rare combination of resources and capability) on innovation performance is increased. In other words, the relationship between EO and performance is influenced by resource attributes (Lumpkin & Dess, Reference Lumpkin and Dess1996).
This study thus proposes two different contingency models. First, it proposes a two-way interaction model that views the EO–performance relationship as being influenced by external environment and/or internal resource attributes. However, some scholars argue that neither the external environment nor resources alone adequately explain the relationship between EO and abnormal returns (Dess, Lumpkin, & Covin, Reference Dess, Lumpkin and Covin1997; Wiklund & Shepherd, Reference Wiklund and Shepherd2003, Reference Wiklund and Shepherd2005). Therefore, this study proposes a second model, a three-way interaction model, which states that firm performance is determined by the interactions between EO, environmental dynamism, and rareness (Dess, Lumpkin, & Covin, Reference Dess, Lumpkin and Covin1997).
Therefore, the following research questions are raised: does EO directly influence firm performance? Or is the association between EO and firm performance moderated by environmental dynamism and/or the rareness of the resource–capability combination? By answering these questions, this study expects to make several contributions. First, although previous studies have stated that EO has a positive influence on the performance of firms with specific resources, the empirical evidence is limited and inconclusive. Second, this study constructs several contingency models to investigate explicitly how environmental dynamism and rareness moderate the association between EO and firm performance. It also uses the three-way interaction model to examine the overall associations between EO, environmental dynamism, rareness, and firm performance. Finally, this study is expected to confirm whether EO leads to improved performance in the context of an emerging market economy such as Taiwan because entrepreneurial performance is an important source of Taiwan's economic growth (Valliere & Peterson, Reference Valliere and Peterson2009).
This study uses four dimensions to measure EO: innovation, proactiveness, risk taking, and competitive aggressiveness (Lumpkin & Dess, Reference Lumpkin and Dess1996; Wang, Reference Wang2008). Regarding the resource attributes, this study focuses on the rareness of resource–capability combinations (Newbert, Reference Newbert2008), whereas the environmental factor refers to environmental dynamism. Based on the data collected from a secondary database, the Taiwan Economic Journal, and questionnaire data collected from 237 public firms, this study uses the ordinary least squares hierarchical regression model to empirically test our hypotheses.
THEORETICAL BACKGROUND
EO and performance
With respect to the entrepreneurial strategy-making process, Mintzberg (Reference Mintzberg1973) notes that firms make decisions on the basis of their entrepreneurial propensity and that in developing strategies, they link the environment and entrepreneurial propensity together. Some scholars insist that a firm's entrepreneurial strategy making, which is a strategy-making process, can be viewed as an EO (Lumpkin & Dess, Reference Lumpkin and Dess1996; Dess & Lumpkin, Reference Dess and Lumpkin2005). EO refers to a firm's strategic orientation with respect to the processes, practices, and decision-making activities that lead to new entry; it involves the intentions and actions of a firm that is willing to grasp new market opportunities in a dynamic process (Lumpkin & Dess, Reference Lumpkin and Dess1996)Footnote 1. Miller (Reference Miller1983) highlights the characteristics of entrepreneurial firms and argues that an entrepreneurial firm is willing to engage in the innovation of products and technological processes, to undertake risky ventures, and to provide proactive innovations to pursue first-mover advantages.
Prior studies have measured EO with four dimensions: innovation, proactiveness, risk taking, and competitive aggressiveness (Lumpkin & Dess, Reference Lumpkin and Dess1996; Wang, Reference Wang2008). Innovation refers to a firm's tendency to create resources and capabilities (Drucker, Reference Drucker1985), to support new ideas, novelty, and experimentation (Lumpkin & Dess, Reference Lumpkin and Dess1996), and to introduce new products and services that capitalize on market opportunities (Hage, Reference Hage1980; Miller, Reference Miller1983). When existing markets are disrupted by the discovery of new products, services, and processes, the wealth of firms can be created (Schumpeter, Reference Schumpeter1934; Miller, Reference Miller1983). Second, proactiveness refers to the manner of enterprises in attempting to track changes in customer tastes and technology and to seize new opportunities, implying a forward-looking perspective that may or may not be related to current operations (Miller & Friesen, Reference Miller and Friesen1982; Lumpkin & Dess, Reference Lumpkin and Dess1996, Reference Lumpkin and Dess2001). Third, risk taking refers to the firm's propensity to engage in risky projects and managers’ preferences for bold acts to achieve the firm's objectives (Lumpkin & Dess, Reference Lumpkin and Dess1996). Finally, competitive aggressiveness refers to a firm's effort to outperform competitors, either through exploiting existing resources or creating new resource–capability combinations (Lumpkin & Dess, Reference Lumpkin and Dess2001).
In most conceptual studies, EO is viewed as an independent effect that creates or sustains firm performance (Covin & Slevin, Reference Covin and Slevin1991; Lumpkin & Dess, Reference Lumpkin and Dess1996). Previous empirical studies often report a positive impact of EO on firm performance in various contexts. For instance, by using small Swedish firms as the sample, Wiklund (Reference Wiklund1999) finds that a high level of EO is positively related to firm performance. A study by Tang et al. (Reference Tang, Tang, Marino, Zhang and Li2008) on Chinese firms also reports that the association between EO and performance is positive and significant.
However, different research findings are also reported. For instance, Hart (Reference Hart1992) argues that having an entrepreneurial strategy-making mode may lead to poor performance for firms in certain circumstances. Covin and Slevin (Reference Covin and Slevin1988, Reference Covin and Slevin1989) find that the relationship between EO and performance is insignificant. Smart and Conant (Reference Smart and Conant1994) and Stam and Elfring (Reference Stam and Elfring2008) also report an insignificant EO–performance relationship. These inconsistent empirical findings imply that the relationship between EO and firm performance may be less straightforward. Some factors may moderate the EO–performance relationship.
EO has become an increasingly central concept in the domain of entrepreneurship and has received a substantial amount of theoretical and empirical attention (Covin, Green, & Slevin, Reference Covin, Green and Slevin2006). Within this body of research, the EO–firm performance relationship has emerged as one of the most intriguing topics. Previous studies have examined the direct effect of EO on firm performance, but other studies suggest that the relationship is contingent on various factors such as the external environment (Covin & Slevin, Reference Covin and Slevin1989; Dess, Lumpkin, & Covin, Reference Dess, Lumpkin and Covin1997) and internal-organization factors that include strategy (Covin, Green, & Slevin, Reference Covin, Green and Slevin2006; Wang, Reference Wang2008), financial resources (Wiklund & Shepherd, Reference Wiklund and Shepherd2005), social capital (Lee & Sukoco, Reference Lee and Sukoco2007), family involvement (Casillas & Moreno, Reference Casillas and Moreno2010), managerial characteristics (Richard, Wu, & Chadwick, Reference Richard, Wu and Chadwick2009), and knowledge-based resources (Wiklund & Shepherd, Reference Wiklund and Shepherd2003). However, few studies focus on the effects of resource characteristics on the EO–performance relationship. We examine these effects and also examine whether firms’ environmental dynamism can influence the relationship between EO and the profitability of the firm because previous studies present inconsistent findings regarding this moderating effect (Covin & Slevin, Reference Covin and Slevin1991; Wiklund & Shepherd, Reference Wiklund and Shepherd2005; Kreiser & Davis, Reference Kreiser and Davis2010).
The moderating effect of environmental dynamism
In the strategic management and organization theory literature, the external environment has been viewed as a critical contingency factor (Thompson, Reference Thompson1967; Child, Reference Child1972). Environmental dynamism can influence the EO–firm performance relationship (Covin & Slevin, Reference Covin and Slevin1991; Lumpkin & Dess, Reference Lumpkin and Dess1996). Environmental challenges often refer to the degree of dynamism in the environmental settings that a company faces (Miller & Friesen, Reference Miller and Friesen1982; Lumpkin & Dess, Reference Lumpkin and Dess2001). Miller (Reference Miller1983) argues that environmental dynamism is associated with the unpredictability of customer tests, aggressive competitor actions, product/service shifts, and high rates of change in markets and industry innovation. Miller (Reference Miller1990) further argues that firms with a higher degree of EO are more likely to pursue success when they face the stimulation of environmental dynamism, which is related to enhancing customer satisfaction by providing a premium on innovation and unique services. In fact, firm profits are derived from EO with the following attributes: higher self-efficacy, ambition and achievement, readiness to change, interest in innovation, as well as a forward-looking perspective in future markets. These attributes allow the firm to recognize and grasp opportunities in environmental dynamism (Sadler-Smith, Hampson, Chaston, & Badger, Reference Sadler-Smith, Hampson, Chaston and Badger2003). Hamel (Reference Hamel2000) suggests that the life cycle of products and business models has been shortened in today's competitive and dynamic environment. Firms with EO are thus encouraged to earn profit by introducing novel products and services that provide opportunities to respond to the changes of competitors and customers and to reduce the threat resulting from environmental dynamism. For example, the study by Wiklund and Shepherd (Reference Wiklund and Shepherd2005) emphasizes that EO leads to business performance in the context of environmental dynamism, even though the authors find that EO leads to relatively high performance in a non-dynamic environment.
It is argued that there is a positive relationship between EO and firm performance in a dynamic environment. Some existing studies suggest that environmental dynamism magnifies the link between EO and firm performance. For example, based on empirical results from a sample of 607 Chinese firms, Li, Guo, Liu, and Li (Reference Li, Guo, Liu and Li2008) reveal an important finding that technological turbulence significantly and positively moderates the relationship between EO and firm performance. Frese, Brantjes, and Hoorn (Reference Frese, Brantjes and Hoorn2002) report that in a dynamic and hostile environment, EO is positively related to firm growth. Similarly, in an empirical study by Miller (Reference Miller1988), whose sample is composed of 89 firms from the province of Quebec, innovative strategies in a dynamic environment were found to be associated with higher performance. That is, firms facing a higher level of environmental dynamism are more likely than firms facing a relatively stable environment to make profits from innovation (Miller, Reference Miller1988; Kreiser & Davis, Reference Kreiser and Davis2010), from making risky resource commitments (Kreiser & Davis, Reference Kreiser and Davis2010), and from responding to the changes of competitors and customers (Lumpkin & Dess, Reference Lumpkin and Dess1996).
Therefore, this study argues that when a firm is facing an environment characterized by high dynamism, EO is expected to lead to improved firm performance. That is, firms with strong EO are more likely to seize new market opportunities and respond to customer needs and competitor actions in a dynamic environment than in a stable one. In dynamic environments, firms with high EO are more likely to explore a variety of markets and seize abundant opportunities, causing their performance to improve. Therefore, the following hypothesis is presented:
Hypothesis 1 : Environmental dynamism positively moderates the relationship between EO and firm performance.
The moderating effect of rareness
In addition to environmental dynamism, the internal resources and capabilities of firms are another variable that may also moderate the relationship between EO and performance (Covin & Slevin, Reference Covin and Slevin1991; Lumpkin & Dess, Reference Lumpkin and Dess1996). The resource-based view has been adopted extensively in the strategic management literature and increasingly so in entrepreneurship studies (Barney & Arikan, Reference Barney and Arikan2001; Ireland, Hitt, & Sirmon, Reference Ireland, Hitt and Sirmon2003). However, the influence of rare resource–capabilities combinations on the success of entrepreneurial firms is seldom examined because the focus of traditional resource-based studies is mainly on how specific combinations of resources and capabilities influence EO and firm performance (Wiklund & Shepherd, 2005; Richard, Wu, & Chadwick, Reference Richard, Wu and Chadwick2009).
According to Barney (Reference Barney1991), rareness refers to a valuable resource that can be possessed by a large number of firms, but only a few firms are able to implement strategies to exploit such a resource and to promote performance. If we view EO as a ‘strategic orientation’ (Naldi, Nordqvist, Sjöberg, & Wiklund, Reference Naldi, Nordqvist, Sjöberg and Wiklund2007), although firms with such an orientation possess valuable resources, their abnormal returns may not be improved unless they can effectively implement these resources (Barney, Reference Barney1991; Chandler & Hanks, Reference Chandler and Hanks1994; Chrisman, Bauerschmidt, & Hofer, Reference Chrisman, Bauerschmidt and Hofer1998). It is argued that firms with EO are more likely to create their performance in a situation of devoting internally rare resources and capabilities (Covin & Slevin, Reference Covin and Slevin1991; Zahra, Reference Zahra1993). Brown, Davidsson, and Wiklund (Reference Brown, Davidsson and Wiklund2001) and Stevenson and Gumpert (Reference Stevenson and Gumpert1985) also argue that firms with EO are able to exploit opportunities to maximize returns when they deploy rare resources and capabilities in a multi-stage manner with minimal exposure at each stage. Therefore, providing rare resource–capability combinations can lead to the translation of EO into superior performance.
As mentioned earlier, in addition to research on the positive influence of EO on firm performance (Covin & Slevin, Reference Covin and Slevin1991; Lumpkin & Dess, Reference Lumpkin and Dess1996), a widely held view in the academic literature suggests that a firm's competitive advantage stems from rare resource–capability characteristics (Barney, Reference Barney1991). For this entire chain to yield beneficial results for EO, there must be a fit with the organization's resource–capabilities rareness (Covin & Slevin, Reference Covin and Slevin1991; Zahra, Reference Zahra1993). A study by Wu, Chang, and Chen (Reference Wu, Chang and Chen2008) on 170 Taiwanese firms reports an increased influence of EO on innovation performance when firms have intellectual/human resources. Lee and Sukoco (Reference Lee and Sukoco2007), whose study is based on 152 Taiwanese firms listed in the Taiwanese ‘Top 1000’ firms find that social capital, a tacit and rare resource–capability combination, moderates the EO–performance relationship.
Alvarez and Busenitz (Reference Alvarez and Busenitz2001) argue that entrepreneurial opportunities exist primarily because the agents of some firms have views or beliefs about the heterogeneity of resources that decision makers in other firms do not have when they decide which resources to invest in their production. According to the resource-based model (Barney, Reference Barney1991), the attributes of resources are based on an assumption that resources (capabilities) are heterogeneously distributed among firms. Therefore, firms with rare resources facilitate the development of the EO and its translation into enhanced performance by combining rare resources with the capabilities of employing these resources, including physical, financial, human, intellectual, and organizational resources and capabilities. Based on 53 samples from 51 studies, Rauch, Wiklund, Lumpkin, and Frese (Reference Rauch, Wiklund, Lumpkin and Frese2009) suggest that firms with high EO improve firm performance when they take risks to achieve a situation of rare resources (capabilities). Using a sample of small- and medium-sized Swedish businesses, Wiklund and Shepherd (Reference Wiklund and Shepherd2003) suggest that a firm with EO can obtain profits when employing rare resources and capabilities; furthermore, they also find that the interaction between EO and knowledge-based resources leads to improved performance.
All of these studies indicate that firms with strong EO are likely to enhance their profitability by combining rare resources and capabilities. Therefore, the rareness of resource–capability combinations is proposed to positively moderate the association between EO and firm performance. Therefore, the following is hypothesized:
Hypothesis 2 : The rareness of resource–capability combinations positively moderates the relationship between EO and firm performance.
The interactive effect of environmental dynamism and rareness
So far, we have argued that some contingency factors (specifically, environmental dynamism and rareness) will advantage firms with EO in terms of achieving improved performance. However, environmental dynamism may also disadvantage firms (Dess, Lumpkin, & Covin, Reference Dess, Lumpkin and Covin1997). Firms may need the ability to control limited or rare resources to reduce the environmental dynamism that they face (Jones, Reference Jones2007). In other words, environmental dynamism and internal resources act simultaneously to influence the relationship between EO and performance (Lumpkin & Dess, Reference Lumpkin and Dess1996). To increase profits in dynamic environments, firms often employ an EO strategy to access rare tangible/intangible resource–capability combinations (Stevenson & Gumpert, Reference Stevenson and Gumpert1985; Brown, Davidsson, & Wiklund, Reference Brown, Davidsson and Wiklund2001; Sirmon, Hitt, & Ireland, Reference Sirmon, Hitt and Ireland2007). Therefore, it is important to examine how the three-way interaction of EO, rareness, and environmental dynamism influence firms’ performance.
As mentioned above, the profitability of firms may be enhanced through the alignment of rare resource–capability combinations and dynamic environments. For instance, sustained competitive advantages may result from exploiting rare resources/capabilities via the implementation of a differentiation strategy to respond to opportunities and threats in the industrial environment (Barney, Reference Barney1991; Newbert, Reference Newbert2008). With respect to the definition of the rare resources, a valuable resource possessed by only a few potential competitors may be a source of competitive advantage (Barney, Reference Barney1991). This phenomenon implies that rare resources should involve valuable attributes rather than obsolete attributes. Under a dynamic environment that is rapidly changing, products and services easily become obsolete (Miller & Friesen, Reference Miller and Friesen1982), making it important for firms’ EO to lead them to recognize and grasp rare resources and capabilities. In this way, firm performance can be promoted (Covin & Slevin, Reference Covin and Slevin1989).
Miller and Shamsie (Reference Miller and Shamsie1996) find that in an unpredictable environment, firms that possess knowledge, a resource that is valuable and rare, can achieve superior financial performance; however, the moderating effect of knowledge becomes weak in a stable environment. Bierly and Daly's (Reference Bierly and Daly2007) empirical study on small- and medium-sized manufacturing firms confirms the notion that environmental dynamism plays a moderating role in the relationship between knowledge exploration/exploitation and firm performance. The exploration/exploitation of knowledge has a stronger influence on the effectiveness of firms in a high-tech environment than in a low-tech environment. Wu (Reference Wu2006) reports that some resources, such as reputation and know-how, correspond to rare resource–capability combinations, which in turn improve the performance of firms competing in a dynamic environment.
Overall, although some concerns have been raised about the possibility of underperformance resulting from high resource commitment in a dynamic environment (Miller & Friesen, Reference Miller and Friesen1982, Reference Miller and Friesen1983; Miller, Reference Miller1983), it is reasonable to suggest a three-way interaction relationship. In other words, if firms with EO would like to enhance their performance under dynamic environments, the level of rareness of resource–capability combinations must be considered. Therefore, the following is hypothesized:
Hypothesis 3 : The positive relationship between EO and firm performance will be magnified by the interactions of environmental dynamism and the rareness of resource–capability combinations.
METHODOLOGY
Sample and data
This cross-sectional study uses a sample from Taiwanese public firms for several reasons. First, Taiwan represents an emerging market economy with relatively limited natural production factors, whereas its advanced factors, such as innovativeness and entrepreneurship, play an important role in its economic development (Wu, Chang, & Chen, Reference Wu, Chang and Chen2008). Moreover, Taiwanese firms have a long-held reputation of developing core competencies based on intellectual and human resources. Additionally, Taiwanese firms recently underwent a so-called Financial Tsunami and have been under the pressure of dynamic environments; therefore, several firms have developed practices and business models to cope with the challenges of uncertain environments (Tseng & Goo, Reference Tseng and Goo2005).
In 2009, 716 non-financial-sector Taiwan Stock Exchange (TSE) companies and 531 non-financial-sector Over-The-Counter (OTC) companies provided complete data for analysis. Financial service firms were excluded from the research sample because their accounting practices were incompatible with those of other industries. Two methods were used for data collection. First, data on EO and rareness were obtained via a questionnaire survey with members of top management as the respondents. The CEOs and top management of firms were initially contacted via telephone or personal visits to explain the purpose of this study. We sent a total of 1,247 questionnaires to the non-financial-sector TSE/OTC companies via post mail. Of these questionnaires, 247 were returned, for a response rate of 20%. After eliminating some incomplete questionnaires, the final sample consists of 237 firms.
Second, dependent variables (return on assets, ROA and return on sales, ROS) and some control variables were collected via a secondary database maintained by the Taiwan Economic Journal, a leading credit analysis research agent and the most comprehensive business database in Taiwan, which is subscribed to by several international research agents, such as Datastream, Dialog, Reuters, and Capital International.
Measurements
Firm performance
Firm performance is the dependent variable in this study. Consistent with prior studies, this study uses two indices to measure firm performance: ROA and ROS (Venkatraman & Ramanujam, Reference Venkatraman and Ramanujam1986; Beal & Yasai-Ardekani, Reference Beal and Yasai-Ardekani2000; Fitzsimmons, Douglas, Antoncic, & Hisrich, Reference Fitzsimmons, Douglas, Antoncic and Hisrich2005; Luke, Verreynne, & Kearins, Reference Luke, Verreynne and Kearins2007). First, the averaged annual rate of profit after taxes but before interest on total assets (ROA) during 2005–2009 is used. This is the most appropriate measure for estimating the effectiveness of business operations (Combs & Ketchen, Reference Combs and Ketchen1999) due to the high debt-equity ratio and imperfect capital markets in developing economies (Chang & Choi, Reference Chang and Choi1988). The second index is the averaged ROS during 2005–2009, a measure of profitability that is commonly used in studies of developing countries (Cuervo-Cazurra & Dau, Reference Cuervo-Cazurra and Dau2009).
EO
The independent variable in this study is EO. Four dimensions are used to measure EO, including innovation, proactiveness, risk taking, and competitive aggressiveness (Miller, Reference Miller1983; Lumpkin & Dess, Reference Lumpkin and Dess1996). The four dimensions are measured with 11 questions developed by Miller (Reference Miller1983), Covin and Slevin (Reference Covin and Slevin1988, Reference Covin and Slevin1989), Lumpkin and Dess (Reference Lumpkin and Dess2001), and Lumpkin, Cogliser, and Schneider (Reference Lumpkin, Cogliser and Schneider2009)Footnote 2 (see appendix). Following these studies, a semantic differential method is used in the questionnaire. That is, for each question, two opposite phrases are offeredFootnote 3. A higher score indicates a stronger EO of the firm. The Cronbach's α values of these four dimensions are 0.836, 0.850, 0.883, and 0.743, respectively, with an overall Cronbach's α of 0.79. The fit indexes of the four first-order factors (the four dimensions) plus the second-order factor fell within an acceptable range (χ2/df = 2.57, RMSEA = 0.08, GFI = 0.93, CFI = 0.97, NFI = 0.95), supporting the notion that the four dimensions are distinctive.
Rareness
The rareness of resource–capability combinations is a moderating variable in this study. It is operationalized as a firm's exploitation of a common resource (or capability) with a unique capability (or resource) or a firm's exploitation of unique resource–capability combinations, to reduce costs, to utilize market opportunities, or to withstand competitive threats. Following the studies of Barney (Reference Barney1991, Reference Barney1997), Amit and Schoemaker (Reference Amit and Schoemaker1993), and Galbreath and Galvin (Reference Galbreath and Galvin2006), and the measurement of rareness developed by Newbert (Reference Newbert2008), this study primarily focuses on how the EO–performance link is contingent on rareness. This construct is measured with three questions for each with five items, including financial, human, intellectual, organizational, and physical resources and capabilities (Cronbach's α = 0.918, 0.907, 0.928, 0.898, and 0.899, respectively) (see appendix). Similarly, an averaged score of the questions is then calculated to indicate the overall rareness of a firm's resource–capability combination. A higher score indicates a higher rareness of the firm's resource–capability combination. This construct has an overall Cronbach's α of 0.94.
Environmental dynamism
Environmental dynamism is also a moderating variable in this study. It is measured by using five questions, including extreme changes in marketing practices, a rapid rate of obsolescence in fashion goods/semi-conductors, the unpredictability of competitors, unpredictable demand and tastes of customers, and changes in the modes of production/service. The scales of environmental dynamism developed by Miller and Friesen (Reference Miller and Friesen1982) and a semantic differential method are used in the questionnaire. Each question offers two opposite phrases. The overall Cronbach's α is 0.81.
Control variables
Several variables that may influence firm performance are controlled in the regression models, including firm size, firm age, R&D intensity, and industry affiliation. First, firm size reflects economies and diseconomies of scale, which may form barriers to entry (Bain, Reference Bain1968), and is operationalized as the natural logarithm of the five-year average of total assets. Firm age is controlled because prior studies suggest that established organizations are more bureaucratic, and these attributes may influence performance (Hannan & Freeman, Reference Hannan and Freeman1989). A firm's age is measured as the natural logarithm of a company's age since its establishment. R&D intensity is defined as the five-year averaged ratio of R&D expenditures over sales between 2005 and 2009. This variable is controlled because a firm's R&D investment in innovation may lead to products/services differentiation as well as profitability. Finally, the total sample is classified into four industry categories: biotechnology and chemical, electronic manufacturing, traditional manufacturing, and others (dummy coded as 1 and 0) due to possible profitability differences deriving from different industrial characteristicsFootnote 4.
Table 1 reports the mean, standard deviations, and correlation coefficients of all variables. To attenuate possible multi-collinearity problems in our data, mean-centred variables are used in the interaction terms through subtracting the mean from each value of the variable (Aiken & West, Reference Aiken and West1991). After this procedure, the correlation coefficients among all variables are relatively low. In addition, we assess the variance inflation factors and find that all variance inflation factors values are smaller than 2. This finding implies that no serious multi-collinearity problems exist in our models.
Descriptive statistics and correlation coefficients of study variables (n = 237)
Note. EO = entrepreneurial orientation; RD = R&D intensity; ROA = return on assets; ROS = return on sales.
†p < .1, *p < .05, **p < .01.
This study uses self-reported data collected from CEOs or top managers, so it may be vulnerable to common method variance. Using ex ante preventive methods, we guaranteed anonymity and mailed the questionnaires directly to the managers. To avoid respondents guessing the relationship between variables, we also reduced item ambiguity and separated related items (Podsakoff, MacKenzie, Lee, & Podsakoff, Reference Podsakoff, MacKenzie, Lee and Podsakoff2003). For the ex post testing methods, we used Harman's single-factor test, a widely adopted post hoc remedy, to estimate whether our data have a common method variance problem (Podsakoff & Organ, Reference Podsakoff and Organ1986). The result showed that the first factor accounted for only 10.34% of variance among variables. Furthermore, potential common method variance problems are most likely to occur when data from the same source are used for the independent variables and the dependent variables (Podsakoff et al., Reference Podsakoff, MacKenzie, Lee and Podsakoff2003), but in our study, the dependent variables rely on secondary sources rather than self-reporting performance, so our data do not have a serious common method variance problem.
Analytical methods
Hierarchical linear regression analyses were used to test the hypotheses. Hierarchical linear regressions are appropriate when analyzing multiple terms in the regression equations. Following Cohen and Cohen (Reference Cohen and Cohen1983), three processes were used to test the main-effect models, two-way interaction models, and three-way interaction models. The three-way interaction models were constructed with the interactions of EO, environmental dynamism, and rareness, simultaneously included in the equations. In this study, each interaction term is expected to make a significant contribution to firm performance.
RESULTS
Table 2 shows the results of the hierarchical regression models undertaken to test the hypotheses. Models 1 and 5 are the null models that contain only control variables. Models 2 and 6 are the main-effect models. This study then introduces the combination of EO and environmental dynamism and the combination of EO and rareness into Models 3 and 7, respectively, to construct the two-way interaction models for testing Hypotheses 1 and 2. The influences of the three-way interactions of EO, environmental dynamism, and rareness are examined in Models 4 and 8 to test Hypothesis 3.
Hierarchical regression results of firm performance (n = 237)
Note. Unstandardized regression coefficients are presented. Standard errors are in parentheses.
EO = entrepreneurial orientation; RD = R&D intensity; ROA = return on assets; ROS = return on sales.
†p < .1, *p < .05, **p < .01.
The results of Models 2 and 6 show that EO is positively associated with ROA and ROS. The results are consistent with the findings of Wiklund (Reference Wiklund1999) and Tang et al. (Reference Tang, Tang, Marino, Zhang and Li2008), indicating that high EO is associated with high performance. Hypothesis 1 focuses on the moderating effects of environmental dynamism and predicts that under high environmental dynamism, EO will lead to improved performance. The results of Model 3 show that the interaction of EO and environmental dynamism is not significant for ROA but that their interaction is significant for ROS. Therefore, Hypothesis 1 is partially supported. Hypothesis 2 focuses on the moderating effects of resource rareness and suggests that EO will lead to improved performance when rareness is high. Models 3 and 7 show that the interactions between EO and rareness positively influence ROA and ROS. Therefore, Hypothesis 2 is strongly supported.
Hypothesis 3 focuses on the three-way interaction effects and predicts that the combination of EO, environmental dynamism, and rareness will lead to improved performance. We introduce the three-way interaction terms into Models 4 and 8 and find them to be significant in both ROA and ROS. Hypothesis 3 is thus strongly supported.
Additional Tests on Moderating Effects
To further explain the interaction between EO and environmental dynamism, we plot these interaction effects for two levels of environmental dynamism, where the low level is defined as below the mean (N = 127) and the high level is defined as above the mean (N = 110). Figures 1a and 1b show that EO has a significantly positive relationship with ROA and ROS when environmental dynamism is high. That is, a strong EO is more likely to promote high returns in the context of high environmental dynamism than in the context of low environmental dynamism. To explain the interaction between EO and rareness, we plot these interaction effects for two levels of rareness, where the low level is defined as below the mean (N = 98) and the high level is defined as above the mean (N = 139). Figures 1c and 1d show that EO positively promotes ROA and ROS when the rareness is high. As we expected, a strong EO is more likely to promote high returns in high rareness than in low rareness.
(a) Interaction effects of entrepreneurial orientation (EO) and environmental dynamism on return on assets (ROA). (b) Interaction effects of EO and environmental dynamism on ROS. (c) Interaction effects of EO and rareness on ROA. (d) Interaction effects of EO and rareness on ROS
We then perform a simple slope analysis for each line to examine whether its slope is significantly different from zero. With respect to the interaction of EO and environmental dynamism, the result again confirms that EO has a strong positive effect on ROA (β = 0.18, t = 1.85) and ROS (β = 0.22, t = 2.36) when environmental dynamism is high. However, firms’ high EO is unrelated to performance when environmental dynamism is low. These results provide additional support for Hypothesis 1. With respect to the interaction between EO and rareness, the result again confirms that EO has a strong positive effect on ROA (β = 0.33, t = 4.01) and ROS (β = 0.22, t = 2.59) when rareness is high. However, when rareness is low, firms’ high EO is unrelated to performance. Therefore, these results also support Hypothesis 2.
To advance the interpretation of Hypothesis 3, we plot these interaction effects for a high level of rareness and environmental dynamism (N = 67) and for a low level of rareness and environmental dynamism (N = 55). We then plot these interaction effects, and a simple slope test is again conducted. Figures 2a and 2b show that the EO has a significantly positive relationship with ROA (β = 0.46, t = 4.22) and ROS (β = 0.36, t = 3.11) when environmental dynamism and rareness are both high. All of these results indicate that a high level of rareness generates a positive effect when environmental dynamism is high. Therefore, the results of Figures 2a and 2b support Hypothesis 3.
(a) Interaction effects of entrepreneurial orientation (EO) and environmental dynamism and rareness on return on assets (ROA). (b) Interaction effects of EO and environmental dynamism and rareness on return on sales (ROS)
DISCUSSION AND CONCLUSION
Entrepreneurship has been viewed as a central research topic in strategic management (Meyer, Neck, & Meeks, Reference Meyer, Neck and Meeks2002). Prior literature has noted that a favourable EO–performance relationship is determined by several contingency factors, such as the external environmental challenges faced by firms as well as firms’ internal specific resources. In other words, the associations between EO and these contingency factors will lead to firm effectiveness. This study tries to go beyond existing studies and argues that the three-way interaction effects between EO, environmental dynamism, and the rareness of resource–capability combinations positively influence firm performance.
Based on the data collected from 237 public Taiwanese firms, this study finds that EO has a significant influence on the ROA and ROS of firms. This result provides an opportunity to further examine whether there are complexly contextual variables that may influence the EO–performance relationship. By using the two-way interaction models and the three-way interaction models to detect moderating effects, this study obtains three main findings. First, with respect to the relationships between EO, environmental dynamism, and performance, although environmental dynamism directly leads to negative performance, the interaction between EO and environmental dynamism positively influences firm performance. This finding is surprising and interesting because our study's results are not completely consistent with Wiklund and Shepherd (Reference Wiklund and Shepherd2005) but instead are similar to Dess, Lumpkin, and Covin (Reference Dess, Lumpkin and Covin1997).
Environmental dynamism may have a higher probability of hurting firm performance due to the differences in risk taking between various firms. The study by Galbreath and Galvin (Reference Galbreath and Galvin2006) shows that the industrial environment by itself is insufficient to sustain a firm's sales growth and profitability. However, the advantages of a firm with EO are more likely to be realized when its EO is combined with a highly dynamic environment (see Figures 1a and 1b). To earn profit, firms’ EO should involve higher levels of self-efficacy, higher ambition and achievement, readiness to change, interest in innovation, and competitive aggressiveness because all of these attributes favour the perception of opportunities in an environment characterized by rapid change (Sadler-Smith et al., Reference Sadler-Smith, Hampson, Chaston and Badger2003).
Second, despite the negative relationship between rareness and firm performance, this study confirms that rareness greatly influences the EO–performance relationship. Based on our sample, rareness has a limited, direct, and positive effect on profitability for the possible reason that rareness includes tangible and intangible resources and capabilities. Some scholars have found that intangible resources can sustain profits but that tangible resources cannot explain a significant share of the variation in performance (Galbreath & Galvin, Reference Galbreath and Galvin2006). Our results lend some credence to the hypothesis that a firm with a strong EO is inclined to take business-related risks and to have a forward-looking perspective in facing rare resources and capabilities, thus enhancing the firm's performance (see Figures 1c and 1d). As has been suggested by Wiklund and Shepherd (Reference Wiklund and Shepherd200, EO can challenge and employ rare resources to bring about high performance.
Finally, this study introduces a configurational approach except a contingency approach. In addition to the moderating role of environmental dynamism (Wiklund & Shepherd, Reference Wiklund and Shepherd2005; Hmieleski & Baron, Reference Hmieleski and Baron2009), rareness can be viewed as a supplementary mechanism to create effectiveness (Lumpkin & Dess, Reference Lumpkin and Dess1996; Dess, Lumpkin, & Covin, Reference Dess, Lumpkin and Covin1997). As expected, we find that the combination of high environmental dynamism and high rareness does indeed strengthen the impact of EO on firm profits. One possible reason for this finding may be derived from Schumpeter's (Reference Schumpeter1934) insights. Schumpeter argues that entrepreneurship facilitates unique resource–capability combinations in dynamic and high-risk environments, which in turn make it possible for some firms to distinguish themselves from others by reducing costs or differentiating their products and services. This argument is largely consistent with the central notion of the resource-based view, which asserts that a firm's competitive advantage lies in the rare combination of resources and capabilities, as well as in its response to the opportunities and threats of the external environment (Barney, Reference Barney1991).
Limitations
This study has two limitations. First, although it does suggest that these two variables (i.e., environmental dynamism and rareness of resource–capability combinations) could explain a significant share of the variance in the relationship between EO and firm performance, it is not mean to be an exclusive one. It is possible that other contingency factors, such as institutional transitions, organizational slack, other environmental factors, and valuable resources, may influence this linkage (Lumpkin & Dess, Reference Lumpkin and Dess1996). Future studies should thus examine how the interactions between EO and these contingency factors influence the EO–performance relationship. Second, the data evaluated in this study were obtained from diverse industries in Taiwan. It may not be appropriate to generalize to the situations in other developed or developing countries on the basis of empirical evidence derived from a single country. Future researchers may consider collecting data from various countries to achieve a more generalizable research finding.
Appendix
