The effect of firm reputation and
CEO celebrity on the innovative
behavior of a firm
Olivier Steenvoorden 10678433 23-06-2017 2016/2017 Supervisor: Hesam Fasaei University of Amsterdam Master Thesis MSc. in Business Administration-Strategy Track.
Statement of originality
This document is written by Olivier Steenvoorden who declares to take full responsibility for the contents of this document.
I declare that the text and the work presented in this document is original and that no sources other than those mentioned in the text and its references have been used in creating it. The Faculty of Economics and Business is responsible solely for the supervision of completion of the work, not for the contents.
Table of contents Abstract ... 5 Introduction ... 6 Literature review ... 8 Firm reputation ... 8 CEO celebrity ... 11 Hypotheses ... 15 Firm reputation ... 15 CEO celebrity ... 21 Conceptual model ... 24 Research design ... 25 Description of sample ... 25
Method and data collection ... 25
Variables and measures ... 26
Dependent variables ... 26 Independent variables ... 26 Control variables ... 28 Results ... 30 Correlations... 30 Firm reputation ... 31 CEO celebrity ... 34
5
Abstract
The reputation of a firm is an important factor for firms as they provide a variety of benefits. A positive reputation can namely create economic value, increase the performance and increase the capacity to secure resources. Another factor that can have a profound influence on a firm is CEO celebrity. These celebrity CEOs have become more popular and visible, as a result of mass and instant media. Currently, the understanding of these two social evaluations and their effect on the innovative behavior of firms is rather limited.Existing research is primarily focused on the benefits and disadvantages of social evaluations of the firm and on outcomes such as performance, behavior of stakeholders, bargaining power, attitudes and beliefs. As a result, the effect that firm reputation and CEO celebrity have on the decision-making of firms has received little research attention. This empirical study will contribute to the existing literature by examining the effect of firm reputation and CEO celebrity on the innovative behavior of firms. Data on firm reputation, CEO celebrity and innovation were drawn from databases and empirically tested. This study found that neither firm reputation nor CEO celebrity have a significant positive effect on the innovative behaviour of firms. It is relevant for theory and practice to understand how both firm reputation and CEO celebrity affect the innovative behavior of firms. With this information, firms and the board of directors are more informed to make changes to organizational routines or to organizational roles when there are undesirable levels of innovative behavior.
6
Introduction
The reputation of a firm is essential to its survival. The confidence and trust the consumers of firms have has the potential to substantially and directly affect the firm's net income (Bracey, 2012). According to Flanagan and O'Shaughnessy (2005), one of the most important strategic resources of a firm is its reputation. When a firm has a strong reputation, this key resource provides firms with strategic advantages at market and asset levels and it has a significant influence on the performance and ultimate survival of a firm (Deephouse, 2000). In addition, Gray and Balmer (1998) argued that building and then promoting a firm reputation is a modern business requirement.
Another factor that can have a profound influence on the decision making of a firm is CEO celebrity. Edwards, Ketchen, Short, and Try (2014) argued that celebrity CEOs have become more popular and visible, as a result of mass and instant media. They also argued that CEO celebrity has the potential to be an intangible asset for the CEO's firm and to increase the opportunities of the firm. However, Edwards et al. (2014) argued that hiring a celebrity CEO also has the potential to increase risk as a result of the increased attention on the performance and ethical behavior of the firm. Hiring a celebrity CEO increases the attention of stakeholders and media on the firm and consequently heightens their expectations about future performance (Wade, Porac, Pollock, & Graffin, 2008). This increased attention on the firm via the celebrity CEO means that if there is a gap between actual and expected firm performance, then this gap is magnified. Also, if a celebrity CEO behaves unethical or takes illegal actions, then it is likely that the CEO’s firm will receive more media attention than firms which act unethical/illegal but who do not have a celebrity CEO (Ranft, Zinko, Ferris, & Buckley, 2006).
Although much research has been done on firm reputation and CEO celebrity,
7 on the behavior and decision-making of the firm itself and in particular on the innovative behavior of firms. Existing research has shown that CEOs have an influence on the decision- making of firm practices (Bertrand & Schoar, 2003). Existing research has also indicated that if firms value their reputation, the desire to protect their reputation influences how they behave, how they make decisions and it also influences their strategic response towards environmental events (Fombrun & Shanley, 1990).
These two facts show that both CEOs and the reputation of a firm influence the decision-making on a wide variety of firm practices. That is why it is important to examine how CEO celebrity and firm reputation affect the decision-making of the firm itself and in particular the innovative behavior of firms. It is namely especially important to examine the effect of firm reputation and CEO celebrity on the innovative behavior of a firm, because innovation is highly important for a firm. Haour (2003) argued that innovation is central to the health and growth of commercial companies and to the wellbeing of societies. Innovation is important for all kind of firms whether they are new or well established (Cefis & Marsili, 2005). For new firms, innovation is important in order to successfully enter the market and weaken established firms (Schumpeter, 1942). For well established firms, innovation is important in order to protect their competitive position and to deal with changes in the external environment (Christensen, 1997).
This topic is also important to study because it has a number of contributions to theory and practice. This study will contribute to practice, because when firms fully understand the possible consequences and implications of firm reputation and CEO celebrity on the
innovative behavior of the firm, the firms itself and the board of directors is more informed to make organizational changes, change organizational routines and change organizational roles when firm reputation and CEO celebrity has an undesirable effect on the innovative behavior of the firm (Financial Reporting Council, 2011).
8 This study will also contribute to theory in a number of ways. Firstly, this study will contribute to the behavioral theory of the firm, because it provides insight into the
strategic/behavioral effects of firm reputation and CEO celebrity on firms and in particular on the innovative behavior of firms. Currently, the understanding of these two social evaluations and their effect on the innovative behavior of firms is rather limited. Previous research is namely focused on the benefits and disadvantages of social evaluations of the firm and on outcomes such as performance, behavior of stakeholders, bargaining power, attitudes and beliefs (Brown & Perry, 1994; Deephouse, 2000; Brown & Dacin, 1997; Kreps & Wilson, 1982; Fombrun, 1996). Previous research is not really focused on the impact of social evaluations on the behavior and decision-making with regard to the innovative behavior of firms. Secondly, this study will also contribute to social evaluation research by solely examining the effect of firm reputation and CEO celebrity on the innovative behavior of a firm empirically, which has not been done in previous research.
In this study, the old, current and important literature on firm reputation and CEO celebrity together with the research question are discussed first. Then, the hypotheses are developed. Subsequently, the research design is discussed. Then, the results are discussed based on the empirical evidence. Finally, an answer on the research question is given, the findings and implications are discussed, limitations of this study are provided and
recommendations for future research are given.
Literature review
Firm reputation
In general, all the different types of social evaluations of a firm and its CEO, such as reputation, celebrity, status, legitimacy and stigma have received much attention by researchers. Researchers have studied what the different types of social evaluations of the firms are, what their meaning is, how they are different and how they need to be
9 conceptualized (Devers, Dewett, Mishina, & Belsito, 2009; Mishina & Devers, 2012). Lange, Lee, and Dai (2011) focused only on organizational reputation. They wanted to clarify what organizational reputation is. Three dominant conceptualizations were identified, namely that reputation consists of familiarity with the organization, beliefs about what to expect from the organization in the future and impressions about the organization’s favorability (Lange et al., 2001).
In contrast, not all researchers only wanted to conceptualize and define social evaluations, a number of researchers also examined the implications and consequences of social evaluations of the firm besides conceptualizing and defining these social evaluations. Of all the social evaluations, reputation and celebrity have attracted considerably the most interest in research. For example, Philippe and Durand (2011) argued that reputation refers to ''the perceptions by a firm's audience about the firm's ability to provide value compared with its peers and rivals'' (p. 972). Firms send signals to its observers trough its strategic choices and these observers use these signals to form impressions of these firms. Observers namely explore and interpret the actions of firms and consequently evaluate the firm's underlying but unobserved key characteristics based on these actions (Basdeo, Smith, Grimm, Rindova, & Derfus, 2006).
In addition, Devers et al., (2009) defined reputation as a firm-specific evaluation used by a firm's audience as a signal of quality, in terms of product/service quality and financial performance, and as a signal of expected behavior. Furthermore, they argued that reputation differs from other social evaluations in terms of definition, foundational theoretical literature, whether the evaluation is individuating, whether an affective response is required, subsequent outcomes and underlying social basis. They also have examined some cognitive and
behavioral consequences of social evaluations, but they focused on the behavior of
10 found that stakeholders dissociate and negatively alter the quantity and quality with
stigmatized organizations (Devers et al., 2009).
Moreover, Pfarrer, Pollock, and Rindova (2010) have also defined reputation and celebrity, but they also examined how reputation and celebrity are gained. Furthermore, they examined what the effects of firm reputation and celebrity are on the likelihood that firm announces a positive/negative earnings surprise and on the investors' reactions to these surprises. They found that firms with high levels of reputation are less likely, and firms that have achieved celebrity are more likely to announce positive surprises than firms without these assets. These firms also experience greater market rewards for positive surprises and smaller market penalties for negative surprises than other firms (Pfarrer et al., 2010). Similar to Devers et al. (2009), Pfarrer et al. (2010) only focused on the effects of the two social evaluations on stakeholders and not on the effect of firm reputation and celebrity on the behavior and decision-making of the firm itself.
Barnett and Pollock (2012) focused on corporate reputation by examining what it means to have a good or bad reputation, how it creates value and how to manage corporate reputation. They found that reputation has different conceptualizations and that reputation facilities economic transactions by providing incentives to firms to behave in a certain predictable way.
Having a good reputation provides a variety of benefits for a firm. For example, customers prefer to transact with firms with a high reputation. As earlier mentioned,
reputation is among other things a signal of the underlying quality of products/services of a firm. As a result, customers may be more willing to pay a premium for the offerings of a high-reputation firm. Furthermore, employees are more willing to work harder for high-high-reputation firms and suppliers are less worried about contractual hazards when they work with high-reputation firms and this also leads to lower costs of contracting and monitoring (Roberts &
11 Dowling, 2002).
Moreover, organizational reputation is an important aspect for organizations. Hall (1993), for example, found that positive reputations can create economic value, but he did not explain the effects of these positive reputations on the behavior and decision-making within the firm itself. Numerous other studies have also empirically linked firm reputation to its financial performance (Brown & Perry, 1994; Deephouse, 2000; Fombrun & Shanley, 1990). These studies found a positive relationship between reputation and performance. In addition, a firm's reputation can impact individual beliefs and attitudes regarding the firm (Brown & Dacin, 1997). It has also the potential to influence the organizational identification of employees (Kreps & Wilson, 1982). Moreover, Fombrun (1996) argued that positive reputations improves the bargaining power of an organization and the capacity to secure resources. Despite the fact that all previously mentioned studies made a valuable contribution to the literature, they all have in common that they did not examine the effect of these positive reputations on the behavior and decision-making within the firm itself and in particular on the innovative behavior of firms.
CEO celebrity
With regard to the social evaluation ''celebrity'', it can be examined on different levels of analysis, namely on the organizational-level and the individual-level. Hayward, Rindova, and Pollock (2004) have examined celebrity on the individual-level by examining what CEO celebrity is, how celebrity CEOs are created and what the consequences are of celebrity. They argued that celebrity arises when journalists announce that the positive performance of a firm has been caused by the actions of its CEO. This definition has three core aspects. Firstly, journalists announce such attributions via mass media. Secondly, the attribution involves the causes of the actions of a firm that lead to the positive performance of this firm. Thirdly, the firm's actions and performance are attributed to only the CEO. This means that celebrity
12 CEOs only arise when the performance can be attributed to only the CEO and when it does not involve attributions to other factors such as luck, environmental factors or the actions of other persons (Hayward et al., 2004).
The consequences of celebrity explained by Hayward et al. (2004) are primarily related to the CEO himself/herself, because they argued that CEOs who internalize celebrity will tend to believe this over-attribution and become overconfident about his/her own abilities. Hayward et al. (2004) argued that this can lead to problematic firm decisions.
However, they did not explain the full consequences and specific actions celebrity CEOs take with regard to innovation and how this affects the innovative behavior of a firm.
Hayward et al. (2004) also argued that CEOs at larger firms have a higher chance of becoming celebrities than CEOs at smaller firms. The reason for this is that the public is more interested in larger firms and they receive more media coverage. However, not only the size matters, also firms that differ positively and significantly from the norms of its industry are also more likely to receive more attention from journalists (Hayward et al., 2004).
Celebrity CEOs are thus those individuals who are highly visible for the public audience and often receive significant public attention (Chen & Meindl, 1991). As a firm, having a celebrity CEO has some implications and positive consequences. Celebrity CEOs often gain significant media attention, increase the confidence of outside investors, have the potential to attract important resources such as employees with good qualities, signals good firm perspectives and may also have a positive influence on the stock market performance (Sinha, Inkson, & Barker, 2012). Moreover, celebrity CEOs also help to obtain good relationships with suppliers and financial capital (Fombrun, 1996).
However, there is little evidence that in fact organizational leaders and thus CEOs significantly affect the actual performance of a firm (Fanelli & Grasselli, 2006; Wade, Porac, Pollock, & Graffin, 2006). Rindova, Pollock, and Hayward (2006) agreed and they argued
13 that celebrity CEOs often receive high compensation without proportionately high levels of firm performance. This shows that celebrity CEOs do not always have a significant effect on the performance of the firm, but they do have an influence on the actions of the firm.
Cho, Arthurs, Townsend, Miller, and Barden (2016) found that the internalization of celebrity by CEOs has a strong influence on the decision making of CEOs. Although they only focused at strategic actions of celebrity CEOs with regard to acquisitions price
premiums, they expect that if celebrity CEOs have strong identity-control motives, that this celebrity status also has a notable impact on their strategic decisions in multiple other areas. This shows that CEO celebrity does not only have a impact on the behavior of stakeholders, such as suppliers, employees and consumers and stock performance, it also has an impact on the decision making within the firm on multiple areas and thus for example on the innovative behavior. This makes it interesting to examine how CEO celebrity affects the innovative behavior of firms.
Moreover, existing empirical studies often focused on firm-level, industry-level, or market-level characteristics to explain the behavior of firms and their performance. These studies do not focus enough on the role individual managers have in the behavior and performance of firms (Bertrand & Schoar, 2003). Celebrity CEOs thus do not only have an influence on the behavior of stakeholders, bargaining power, attitudes and beliefs, but Bertrand and Schoar (2003) argued that CEOs and top executives are also key factors in making decisions about firm practices. They suggested that managers have their own style when they make strategic decisions and as a result they have a influence on the companies they manage.
With regard to innovation, Bertrand and Schoar (2003) found that that executives have an effect on the organizational strategy variables, one of those variables is R&D expenditure which relates to innovation. They also found that CEOs have a larger effect on the
14 organizational strategy variables and thus also R&D expenditure than CFOs. This shows that there indeed exists a relationship between CEOs and the innovative behavior of firms and that it is interesting to examine how CEOs with a celebrity status affect the innovative behavior of firms.
Moreover, Abril and Olazábal (2009) have also focused on CEO celebrity. They examined what influenced the rise of the celebrity CEO. They found that the democratization of stock ownership, the fact that there is more detailed personal information available and the high-profile corporate issues of the day have encouraged the enormous interest in and scrutiny of the personal lives of CEOs. Abril and Olazábal (2009) did also not examine anything about the effects of CEO celebrity on the innovative behavior of a firm.
In addition, Shemesh (2016) focused on changes in CEO status and whether this affects risk-related business decisions. He found that firms who have CEOs who won awards decrease their idiosyncratic volatility, increase investment in fixed assets and reduce their spending on research and development. This research thus found some effects of CEO
celebrity on the innovative behavior of the firm, but this outcome was part of a larger research and was not of primarily importance for the researcher. This research was more concerned in examining how CEO status affect the risk-related business decisions of firms in general, rather than the innovative behavior of firms (Shemesh, 2016).
In conclusion, previous research has inadequately examined the effect of firm reputation and CEO celebrity on the behavior and decision-making of the firm itself and in particular the innovative behavior of firms. This provides the opportunity to further examine the effect of firm reputation and CEO celebrity on the innovative behavior of firms. It is especially interesting to examine the effect on the innovative behavior of a firm, because innovation is highly important for a firm. Haour (2003) argued that innovation is central to the health and growth of commercial companies and to the wellbeing of societies. He also
15 suggested that it has a great influence on creating economic value (Haour, 2003). Henard and Dacin (2010) indicated that organizations spend altogether billions of dollars on research and development activities, because these organizations have a drive for innovation and product development.
Furthermore, Haour (2003) argued that innovation can be seen as an invention converted into a product, an industrial process or a service for the marketplace or it may be a new way of doing business, such as easyJet in the airline industry. The innovative behavior of firms can also be seen in for example the number of patents filed and the amount of dollars spent on research and development (Henard & Dacin, 2010).
Moreover, innovation is important for all kind of firms whether they are new or well established (Cefis & Marsili, 2005). For new firms, innovation is important in order to be able to successfully enter the market and weaken the established firms (Schumpeter, 1942). For well established firms, innovation is important in order to protect their competitive position when dealing with new, emerging and disruptive technologies and in order to have the firm-specific capabilities to deal with changes in the external environment (Christensen, 1997). Innovation matters also for the existence of a firm due to the nature of technology, which affects the general conditions for survival of the firm in the market in which they are active (Agarwal & Audretsch, 2001).
Innovation is thus an important aspect of a firm and is it therefore interesting to study how firm reputation and CEO celebrity have an effect on the innovative behavior of a firm. Therefore the research question of this study is: how does firm reputation and CEO celebrity affect the innovative behavior of a firm?
Hypotheses
Firm reputation
16 experiences with the firm as the reference point of the firm's audience when they make
judgments about the reputation of a firm (Helm, Garnefeld, & Tolsdorf, 2009). Helm et al. (2009) also argued that individuals tend to make judgments about the reputation of a firm based on their impression of a limited number of attributes of the firm, without even having tangible information about all other facets of corporate reputation. The satisfaction of
individuals with products or services serves to indicate the overall firm behavior. This shows that having products with high quality is necessary to achieve a good firm reputation (Helm et al., 2009). As a result, firms with high reputation will likely invest in high quality
products/services to sustain a high firm reputation, because Helm et al. (2009) argued that this investing in high quality products/services is essential for the reputation of a firm.
In addition to Helm et al. (2009), Chun (2005) argued that the individuals also make judgments about a firm's reputation based on its innovativeness, amongst other things such as financial soundness, use of corporate assets and quality of company's management. Moreover, Ang and Wight (2009) argued that firms see sustaining and defending their obtained
reputation as a vital strategic concern. Focusing on their innovativeness is thus important for firms with a high reputation in order to sustain and defend their obtained reputation.
Moreover, Adner and Levinthal (2001) argued that when firms focus on innovation, they mainly focus on product innovation as this is the dominant mode of innovation. Product innovation focuses primarily at improving the performance of the products. When product innovations are successful, they have an optimal product configuration which is called a dominant design (Adner & Levinthal, 2001). Also, Osborn and Morley (2016) argued that a focus on innovation can lead to products or services that are new or significantly improved. So, when firms focus on innovation this may lead the firm's audience to value the products and services even more. As earlier mentioned, when the firm's audience makes judgments about the reputation of a firm they mainly focus on the products and services they sell (Helm
17 et al., 2009). By focusing on innovation, which has the potential to lead to new and high quality products and services, the high reputation of the firm is conserved and sustained.
When a firm has a high reputation its stakeholders have high expectations for them (Petkova, Wadhwa, Yao, & Jain, 2014). To meet these expectations maintaining consistent performance may not be enough (Mishina, Dykes, Block, & Pollock, 2010). In order to increase their performance, reputable firms may stop following the performance-maintenance path and instead search for new and riskier opportunities (Petkova et al., 2014). Examples of such bold and risky strategic moves of high reputable firms are Google’s development of driverless cars and the introduction of mobile and entertainment devices of Apple (Petkova et al., 2014). These are examples of innovative behavior. Mansfield (1961) also argued that a form of risk taking is having high levels of innovation. Therefore, it can be argued that firms with a high reputation will demonstrate high levels of innovative behavior in order to meet the high expectations of their stakeholders.
The expectations that stakeholders have for firms with a high reputations also raises the aspirations level of these firms themselves (Petkova et al., 2014). These firms often understand that the strategies they are currently employing will not allow them to achieve their aspiration levels. To achieve these aspiration levels, firms with a high reputation instead will employ new and risky strategic moves (Bromiley, 1991). Greve (1998) argued that an example of a risky strategic move that is triggered by high aspirations are entering new markets. For entering new markets, innovation is important as it helps firms to enter new markets fast and deep (Lafley & Charan, 2010). Innovation in itself is also a risky strategic move, MacAlister (2016) namely argued that the variability of outcomes is high and that the impact on stakeholders is not completely known. However, having high levels of innovative behavior can help firms to achieve their aspirations levels, MacAlister (2016) namely argued that the expected new return on investment as a result of innovation is higher than the current
18 or non-innovation return on investment. Therefore, it can be argued that firms with a high reputation will demonstrate high levels of innovative behavior in order to achieve their own aspiration levels.
The high reputation of a firm also increases the expectations of customers towards the quality of the products (Rhee, 2009). As a result, these expectations put pressure on these firms with a high reputation to raise their aspiration for future performance (Mishina et al., 2010). One way firms can meet the expectations with regard to product quality is by being innovative. Osborn and Morley (2016) namely argued that a focus on innovation can to lead products or service that are new or significantly improved. It therefore can be argued that firms with a high reputation will have high levels of innovative behavior in order to produce new and significantly improved products which helps them to meet the expectations of their customers towards the quality of the products.
Furthermore, Stuart (2000) argued that a firm's reputation affects its ability to recruit human resources of high-quality and to obtain the funds and market positions which are necessary to launch innovative projects of major proportions. It therefore can be argued that having a high-firm reputation provides incentives to demonstrate more innovative behavior, because investing in innovation will likely be valuable as result of the high-quality human resources and favorable market positions which support the investments in innovation.
Del Canto and Gonzalez (1999) argued that the probability of a firm carrying out internal R&D is mainly determined by its intangible factors, of which one factor is the
reputation or image of a firm. The intangible commercial resources have often the potential to become the complementary resources necessary for appropriately exploiting the innovations that arise as a result of the internally developed R&D. Del Canto and Gonzalez (1999) found that the intangible resources are positively associated with the carrying out of R&D activities internally. This can indicate that firms with a high reputation are more likely to carry out
19 R&D activities internally and thus have more innovative behavior.
Dowling (2001) argued that having a good reputation supports the introduction of new products. This can incentivize firms with a high reputation to demonstrate more innovative behavior, because outcomes of this innovative behavior such as new product introductions will likely be supported by their reputation and as a result have a higher chance of being a success. In some firms, their firm reputation is directly associated with product innovation, because these firms are primarily focused on innovation and product development (Henard & Dacin, 2010). It therefore can be argued that firms with a high reputation for product
innovation will logically have high levels of innovative behavior.
Firm reputation is also likely to have an effect on the innovative behavior of firms via its employees. Newburry (2010) namely argued that individuals who are working for a firm which they perceive as having a high firm reputation may be more likely to pursue company-supportive behavior. The reason for this is that those individuals make positive judgments about the firm and as a result are more likely to perceive a fit between themselves and these firms (Newburry, 2010 ). Gray and Balmer (1998) also argued that favorable judgments about the reputation of a firm of employees can have an effect on how these employees behave. They argued that these favorable judgments of the firm lead to high morale and productivity of employees. Moreover, Helm et al. (2009) argued that investing in high quality
products/services is essential for the reputation of a firm. Therefore, it can be argued that the individuals working for these high-reputation firms will support this type of innovative behavior, which results in increased attention on innovation.
Although no research has been done on solely the effect of firm reputation on the innovative behavior of the firm, Galende and De La Fuente (2003) have included firm reputation in their research as part of a larger construct called ''commercial resources''. They wanted to examine which internal factors determined a firm's innovative behavior. In their
20 research they used internal factors such as size, human resources, organizational resources, diversification and commercial resources. The construct ''commercial resources'' consisted of image, complementary resources, information and firm reputation. Galende and De La Fuente (2003) found that the possession of many commercial resources is strongly related to an intense development of product innovations. This shows that commercial resources impact the intense generation of product innovations. Firms carry out product innovations to differentiate itself from competitors and in order to achieve a competitive advantage (Galende & De La Fuente, 2003). The effect of commercial resources on the intense generation of product
innovations was so strong that isolating the commercial resource called ''firm reputation'' from the other commercial resources such as image, complementary resources and information will likely show that firm reputation in itself also affects the innovative behavior of the firm.
In short, firms with a high reputation will focus on innovation in order to sustain and defend their obtained reputation, which is a vital strategic concern. Furthermore, firms with a high reputation will demonstrate high levels of innovative behavior in order to meet the high expectations of their stakeholders with regard to the performance of the firm and with regard to the quality of the products. Firms with a high reputation will also demonstrate high levels of innovative behavior as it can help to achieve their own high aspiration levels. Furthermore, firms with a high reputation are incentivized to demonstrate high levels of innovative
behavior, because investing in innovation will likely be valuable and beneficial as a result of the possession of high-quality human resources, the support of new product introductions and the obtained favorable market positions which all support the investments in innovation. Also, individuals working for high-reputation firms will pursue company-supportive behavior and thus as a result will support innovative behavior, which results in increased attention on innovation. Hence,
21 Hypothesis 1: Firms with a higher reputation will demonstrate more innovative behavior than firms with a lower reputation.
CEO celebrity
A celebrity is a social actor whose name is well-known, who often receives significant public attention and who has profit-generating value (Gamson, 1994). As earlier mentioned, CEO celebrity arises when journalists announce that the positive performance of a firm has been caused by the actions of its CEO (Hayward et al., 2004). Individuals and thus also CEOs are influenced by others and they rely on the perceptions and actions of other individuals to determine what their own attitudes and beliefs are. This means that when CEOs become conscious of the fact that he/she is a celebrity, the attributions underlying their celebrity become more available to the CEO as an explanation for firm performance (Hayward et al., 2004).
Furthermore, individuals and thus also CEOS are more likely to believe certain information is true when they are frequently exposed to this information (Hawkins & Hoch, 1992). These two aspects lead to the fact that when a CEO attracts celebrity, it is difficult for him/her to believe that he/she does not have a significant influence on the performance of a firm (Hayward et al., 2004). CEOs also embrace their celebrity, because being a celebrity CEO can lead to immense financial rewards (Rosen, 1981). This in turn again increases the believe of the CEO that he/she is responsible for the firm performance (Hayward &
Hambrick, 1997).
Hayward et al. (2004) argued that all these arguments show that celebrity CEOs are less likely to ignore or reject their celebrity status and more likely to breed and internalize celebrity. Furthermore, they argued that the more celebrity a CEO attracts, the more the CEO perceives that he/she has control over the actions and performance of the firm. In this process of attracting celebrity, internalizing celebrity and perceiving control over firm actions and
22 performance, CEOs are likely to become overconfident about their own abilities and more devoted to the strategies which helped them to become a celebrity CEO (Hayward et al., 2004).
It is this overconfidence and devotion to their own strategies of celebrity CEOs which can influence the actions of a firm with regard to innovation. Simon and Houghton (2003), namely argue that overconfidence can lead to the development and introduction of riskier products. Camerer and Lovallo (1999) argued that celebrity CEO's overestimate the benefits from investing in new products or entering new markets.
Galasso and Simcoe (2011) found that overconfident CEOs underestimate the chance of failure and they are more likely to pursue innovation and to take their firm in a new
technological direction. Galasso and Simcoe (2011) argued that this pursuit of innovation can be seen in terms of patents counts and research and development expenditures. This
significant investment in research and development by overconfident CEOs is also found by Shemesh (2016). He argues that when CEOs become more overconfident as a result of having won an award, it is expected that these CEOs would invest more heavily in research and development. Also, Hirshleifer, Low, and Teoh (2012) found that firms with overconfident CEOs would invest more in innovation. Overconfident CEO's will also underestimate the impact of situational factors in their industry and business environment (Hayward et al., 2004). They will for example underestimate changing governmental regulations and technology, which lowers the chance of success with regard to actions such as new product/service introductions (Camerer & Lovallo, 1999).
Furthermore, when a firm has weaker performance celebrity CEOs may expand their devotion to their own strategies out of proportion. The reason for this is that the overconfident CEOs think they can change the weaker performance of the firm into better performance by assigning a lot more resources to their chosen strategy. In general, celebrity CEOs tend to
23 reproduce the actions that led to their celebrity status (Hayward et al., 2004). The actions of CEOs which led to their celebrity status are assessed by journalists as highly distinctive. This distinctiveness relates to the fact some actions CEO's take are distinctive from the action of peer firms exhibit in similar industry environments and this distinctiveness can also relate to the fact that the action of a firm under the guidance of a new CEO may be different than the past actions displayed by the former CEO (Hayward et al., 2004). This shows that the actions of the celebrity CEO are innovative and unique, which indicates that that CEO celebrity has a positive influence on the innovative behavior of firms.
Also, Cho et al. (2016) suggested that the internalization of celebrity by CEOs has a strong influence on the decision-making of CEOs. They argued that the celebrity status of a CEO has an impact on the risk-taking propensity of these CEOs. They suggested that the reason for this is that celebrity CEOs look at certain strategic situations and make decisions in a way that is coherent with the celebrity status in the eyes of themselves and other individuals (Cho et al., 2016). Furthermore, Cho et al. (2016) argued that under normal conditions
celebrity CEOs are less likely to take extreme risk. Normal conditions are situations where prior firm performance does not deviate from aspirations and industry average performance. Celebrity CEOs prefer this, because taking extreme risk can harm their celebrity status and they may lose the benefits of the celebrity status. Rindova et al. (2016) argued that one of those benefits of CEOs with celebrity status is receiving high-value compensation packages.
However, when situations are not normal, for example if the stock market performance and firm performance are not positive, celebrity CEOs will take bold and risky actions in order to create consistency between their status and the performance of the firm, which helps them to maintain their status as celebrity. In these situations, celebrity CEOs will likely make decisions which portrays and reaffirms their celebrity status which is needed for celebrity CEOs to maintain their status as a celebrity. This means that celebrity CEOs may become
24 risk-seeking in order to protect their celebrity status (Cho et al., 2016). Another example which influences celebrity CEOs to take risky decisions is hubris (Chatterjee & Hambrick, 2011). This means that when firm performance is extremely high, CEO's become
overconfident about their ability to control strategic situations and as a result take bold strategic actions (Cho et al., 2016).
In short, celebrity CEOs have a tendency to embrace, breed and internalize celebrity, which means that celebrity CEOs willingly and enthusiastically accept that they indeed are celebrities and that by accepting and absorbing this idea it becomes part of their character. Furthermore, the more celebrity a CEO attracts, the more the CEO perceives that he/she has control over the actions and performance of the firm. As result, CEOs are likely to become overconfident about their own abilities and more devoted to the strategies which helped them to become a celebrity CEO.
This overconfidence will lead celebrity CEOs to take risky actions towards innovation and as result the firms will demonstrate more innovative behavior. As earlier mentioned, Galasso and Simcoe (2011) namely argued that overconfident CEOs underestimate chance of failure and they are likely to pursue innovation. Shemesh (2016) suggested that overconfident CEOs are expected to invest more heavily in research and development. Hirshleifer et al. (2012) also found that overconfident CEOs would invest more in innovation. Besides overconfidence, also in situations where firm performance is not consistent with their
celebrity status, celebrity CEOs will take bold and risky strategic actions which influences the firm to demonstrate more innovative behavior. Hence,
Hypothesis 2: Firms with celebrity CEOs will demonstrate more innovative behavior than firms without celebrity CEOs. Conceptual model
25 Figure 1. Conceptual model.
Research design
Description of sample
The overall sample population is based on large international firms with high/low reputation and with/without celebrity CEOs over the period 2013-2015. All firms in all countries and industries have some sort of reputation, this can be high, low or somewhere in between. Although leading research/advisory/consulting firms with regard to reputation such as the Reputation Institute, Korn Ferry Hay Group or Harris Interactive have examined the reputation of many firms in different countries and industries, they have not examined the reputation of every firm. Also, not every firm whose reputation was measured did report their expenditures on research and development So, a panel data set was constructed and due to data availability on reputation scores of firms and data on research and development
expenditure, the final sample of this study consisted of 91 firms and 255 firm-years.
Method and data collection
The research strategy that was used in this study is archival research and in particular database research where data is drawn from databases and empirically tested. This study thus used secondary data, which is data that were originally collected for some other purposes and
26 which were already available in public domain, which is the case for data on the innovative behavior of firms. This is the case, because most firms provide annual reports and
administrative records where proxy's for innovation/innovative behavior can be found, such as the amount of dollars spent on research and development, the number of patents filed and the amount of new product introductions. These data are also available in large databases such as
COMPUSTAT and Orbis.
Variables and measures Dependent variables. To measure the overall innovative behavior of a firm, three
variables are used in this study. Innovation inputs was measured by research and development expenditures (COMPUSTAT item 46) (Galasso & Simcoe, 2011; Artz, Norman, Hatfield, & Cardinal, 2010). Research and development intensity was measured as research and
development expenditures (COMPUSTAT item 46) divided by sales (COMPUSTAT item 12) (Greve, 2003; Heyden, Reimer, & van Doorn, 2015). This variable was used, because it shows the relative importance of research and development in the strategy of the firms (Heyden et al., 2015). Research and development concentration was measured as the ratio of research and development expenditures (COMPUSTAT item 46) over lagged total assets (COMPUSTAT item 6) (Bertrand & Schoar, 2003).
The data for the three dependent variables were drawn from COMPUSTAT. COMPUSTAT is a major data source that reports the financial variables of around 7500 firms. The data which are available in this data source were drawn from annual reports, 10-K/10-Q fillings and large companies with public ownership (Bertrand & Schoar, 2003). The data drawn from COMPUSTAT was double checked from annual reports.
Independent variables. Firm reputation was measured by ratings on firm reputation
drawn from the Reputation Institute's Reputation Quotient (Fombrun, Gardberg, & Sever, 2000; Helm, 2005). Reputation Institute is a leading reputation-based advisory firm which
27 annually measures the reputation of the most highly regarded companies. The Reputation Institute conducts around 55000 interviews based on 7 criteria. Examples of these seven criteria are performance, products/services, innovation, leadership (Reputation Institute, 2013).
The Reputation Institute measures the scores of the most reputable firms, so in order to be able to also measure the scores of firms with lower reputation, the Harris Poll Reputation Quotient was used (Gomes-Casseres, 2005). The Harris Poll Reputation Quotient measures the opinion on the most recognizable companies based on 6 criteria's. Examples of these criteria are vision and leadership, products and services, financial performance (Sauter & Frohlich, 2015).
For the reputation measure, a score of less than 40 is perceived as a poor reputation, a score between 40-59 is perceived as a weak reputation, a score between 60-69 is perceived as an average reputation, a score between 70-79 is perceived as strong reputation and a score of more than 80 is perceived as an excellent reputation (Reputation Institute, 2014). Multiple sources were used in this study to measure firm reputation to increase data availability and the reliability and validity of the measure.
CEO celebrity was measured as a dummy variable (Hengartner, 2007). The dummy
variable takes the value of 1 if the CEO of a particular firm had achieved a celebrity status and 0 if the CEO of a particular firm had not achieved a celebrity status. CEOs are believed to have a celebrity status when the name of the individual is well-known, when they often receive significant public attention and when they have profit-generating value (Gamson, 1994). Celebrity CEOs were identified based on whether they have won high-profile awards at prestigious contests organized by large international business publications such as Business Week, Financial World, Chief Executive, Forbes, Fortune, Time and also by highly regarded accounting firm Ernst & Young (Koh, 2011). CEO celebrities were also identified based on
28 insights of media and existing studies (Forbes, 2013; Koh, 2011; Hansen, Ibarra, & Peyer, 2013).
All the independent variables were lagged by one year in the panel data set structure, this was done because this study is focused on how firm reputation and CEO celebrity affects the innovative behavior. By lagging variables, it is ensured that the cause precedes the effects (Mills, 2011). Moreover, the effect of firm reputation and CEO celebrity will likely not happen immediately, so lagged variables are appropriate. Lagging the independent variables allows for temporal manifestations of the theorized effects (Heyden et al., 2015). Lags with regard to R&D spending, innovation activities are normal in studies which focus on
innovation (Artz et al., 2010).
Control variables. A number of control variables are used which account for majors
factors identified in the literature to affect firm innovation. A factor that was used as control variable is firm size. Firm size was measured by number of employees (Martin, Swaminathan, & Mitchell, 1998; Georgakakis & Ruigrok, 2017). Data on number of employees were drawn from Orbis database (2017) and double checked from annual reports. The reason why firm size was used as control variable is the fact that firm size plays an important role in the innovative behavior of firms. Cohen and Klepper (1992) argued that larger firms can spread the costs of innovation over a larger output and thus would invest more in innovation than smaller firms. Furthermore, Arora and Gambardella (1994) argued that smaller firms would be less likely to invest in innovation as a result of difficulties in acquiring the necessary assets and as a result of the large fixed costs of creating a knowledge base to innovation. Moreover, also Heyden et al. (2015) argued that larger firms can be expected to have more established R&D programs.
Another factor that was used as control variable is profitability. Profitability was measured by net income (Nachum & Zaheer, 2005). Data on net income were drawn from
29 Orbis database (2017) and double checked from annual reports. The reason why profitability was used as control variable is the fact that profitability plays an important role on the innovative behavior of firms. The more money a firm makes, the more they are going to invest in research and development and profitability has thus an impact on the innovativeness of firms. This relationship is even stronger when these firms believe that investing in research and development has helped them they make their initial profit (National Research Council, 2003).
Another factor that was used as control variable is the company's industry. Companies in different industries tend to spend different amounts on research and development
(Jaruzelski & Hirsh, 2015). For example, pharmaceuticals, semiconductor and
software/technology companies tend to spend the most on research and development in general (Investopedia, 2017). Industry effects were measured by dummy variables for each firm's main sector as indicated by their industry code (NACE code) (Pérez-Luño, Wiklund, & Cabrera, 2011). Data on NACE codes were drawn from Orbis database (2017).
The final factor that was used as control variable is environmental dynamism. Environmental dynamism describes the rate and the unpredictability of changes in the external environment of a firm (Dess & Beard, 1984). Zahra and Covin (1995) argued that business in firms in turbulent environments are required to renew their products or services continuously in order to respond to changes in the environment. Tushman and Romanelli (1985) suggested that top managers of firms in environments where there is significant legal, social and technological uncertainty are more likely to take radical changes. Moreover, Brown and Eisenhardt (1997) argued that firms in high-velocity environments are required to
innovative quickly and continuously in order to survive. These arguments confirm that environmental dynamism has an effect on the innovative behavior of firms and therefore affirms the importance of including environmental dynamism as a control variable.
30 Environmental dynamism was measured by regressing time against industry sales for the five year period of 2007-2011. The data for total industry sales were obtained from Orbis database (2017) and double checked from annual reports. The total industry sales were summed for each industry category across the five year period by groupings based on the NACE code. In order to conduct the regression, time (period 2007-2011) was used as an independent variable and annual industry sales was used as a dependent variable for every industry group based on the NACE code. Finally, the standard error of the regression coefficient was divided by the mean sales values of the five years (Baron & Tang, 2011).
Results
Correlations
Before the hypotheses were tested, a correlation analysis was done to examine the relationships and thus the correlation between all the variables. This correlation analysis shows the strength and direction of the relationship. Some variables had a Pearson correlation coefficient of larger than r = 0.3 and a significance value of p < 0.01. According to Cohen (1988) a Pearson correlation coefficient of r>0.3 is considered as a medium effect between two variables. All the coefficients and significance levels, together with the mean and
standard deviation of the variables can be found in table 1. The control variable firm industry is not included in table 1 as it is a categorical variable with more than two categories. This means that there is no intrinsic ordering to the categories and the categories have no
mathematical meaning. As a result, the mean, standard deviation and correlations cannot be analyzed, because these statistics provide no meaning when used with a categorical variable with more than two categories (Statistics Solutions, 2017).
Table 1. Means, Standard deviations and Correlations
Variables M SD 1 2 3 4 5 6 7 8 1. Firm reputation 70.24 4.48 -
31 2. CEO celebrity 0.47 0.50 .11 - 3. Innovation inputs 306.97 323.37 .34** .26** - 4. R&D intensity 0.06 0.06 .11 .04 .49** - 5. R&D concentration 0.04 0.04 .12 .04 .48** .77** - 6. Profitability 508.13 771.89 .20** .26** .53** .08 .05 - 7. Firm size 126.88 106.47 .20** .24** .39** -.19** .15* .25** - 8. Environmental dynamism 0.10 0.05 .14* .02 .03 .02 .08 .04 .11 -
Note. M and SD of innovation inputs and profitability x 10,000,000. M and SD of firm size x 1,000.
**. Correlation is significant at the 0.01 level (2-tailed). *. Correlation is significant at the 0.05 level (2-tailed).
Although correlation can show how two variables are correlated, it does not imply anything about causality. When two variables both increase or decrease, it can be a result of another factor or factors causing both variables to increase or decrease. Causality describes the cause and effect relationship between two variables. So, in order to test the hypotheses a regression analysis and a One-Way MANCOVA test was done that include the control variables in order to test the cause and effect relationships between the variables in the hypothesis.
Firm reputation To test the first hypothesis, three different hierarchical multiple regression were performed for the three different dependent variables: innovation inputs, R&D intensity and R&D
concentration. The first hierarchical multiple regression was performed in order to examine the ability of firm reputation to predict the innovation inputs of firms, after controlling for profitability, firm size, industry and environmental dynamism. In the first step of the hierarchical multiple regression, four different predictors were added namely profitability, firm size, industry and environmental dynamism. This first model was statistically significant F (4,218) = 51.30; p < 0.001 and it explained 48.5% of variance in innovation inputs.
32 explained by this model was 50.9% F (5,217) = 44.98; p < 0.001. This showed that the
entering of firm reputation explained additional 2.4% of variance in innovation inputs, after controlling for profitability, firm size, industry and environmental dynamism (R2 Change = .02; F (1,217) = 10.63; p < 0.01. In the final model four out of five predictor variables were statistically significant, namely profitability, firm size, industry and firm reputation with firm reputation having a high Beta value (β = .16, p < 0.01). This means that if the reputation of a firm increases for one, their innovation inputs will increase for 0.16. All the relevant statistics can be found in table 2.
Table 2. Hierarchical regression model of innovation inputs.
Note. Environmental dynamism x 1000.
Statistical significance: * p < 0.05; ** p < 0.01; *** p < 0.001.
The second hierarchical multiple regression was performed in order to examine the ability of firm reputation to predict the R&D intensity of firms, after controlling for profitability, firm size, industry and environmental dynamism. In the first step of the hierarchical multiple regression, four different predictors were added namely profitability, firm size, industry and environmental dynamism. This first model was statistically significant F (4,218) = 14.28; p < 0.001 and it explained 20.8% of variance in R&D intensity.
In the second step, firm reputation was added and as a result the total variance explained by this model was 20.9% F (5,217) = 11.50; p < 0.001. This showed that the
Step Variables R R² R² Change B SE β T 1. Profitability .70 .49*** .23 .02 .50*** 9.68 Firm size 6.58 1.57 .22*** 4.20 Industry 641.72 126.40 .25*** 5.08 Environmental dynamism -1158.07 2977.74 -.02 -.39 2. Profitability .71 .51*** .02** .22 .48 .48*** 9.38 Firm size 5.97 1.55 .20*** 3.86 Industry 580.85 125.10 .23*** 4.64 Environmental dynamism -2234.18 2932.72 -.04 -.76 Firm reputation 119928.80 36787.11 .16** 3.26
33 entering of firm reputation explained additional 0.2% of variance in R&D intensity, after controlling for profitability, firm size, industry and environmental dynamism (R2 Change = .002; F (1,217) = .50; p = 0.48. In the final model three out of five predictor variables were statistically significant, namely profitability, firm size and industry. Firm reputation had a Beta value (β = .05, p = 0.48). This means that if the reputation of a firm increases for one, their R&D intensity will not increase significantly. All the relevant statistic can be found in
table 3. Table 3. Hierarchical regression model of R&D intensity.
Step Variables R R² R² Change B SE β T 1. Profitability .46 .21*** 1.543E-9 .00 .22** 3.37
Firm size -1.152E-7 .00 .24*** -3.80 Industry 1.467E-5 .00 .37*** 6.00 Environmental dynamism .02 .06 -.02 -.35 2. Profitability .46 .21*** .00 1.500E-9 .00 .21** 3.24
Firm size -1.179E-7 .00 -.25*** -3.86 Industry 1.441E-5 .00 .37*** 5.82 Environmental dynamism -.03 .06 -.03 -.43 Firm reputation .00 .00 .05 .71
Note. Statistical significance: * p < 0.05; ** p < 0.01; *** p < 0.001.
The third hierarchical multiple regression was performed in order to examine the ability of firm reputation to predict the R&D concentration of firms, after controlling for profitability, firm size, industry and environmental dynamism. In the first step of the hierarchical multiple regression, four different predictors were added namely profitability, firm size, industry and environmental dynamism. This first model was statistically significant F (4,217) = 7.11; p < 0.001 and it explained 11.6% of variance in R&D concentration.
In the second step, firm reputation was added and as a result the total variance
explained by this model was 12.1% F (5,216) = 5.93; p < 0.001. This showed that the entering of firm reputation explained additional 0.5% of variance in R&D concentration, after
34 .005; F (1,216) = 1.20; p = 0.28. In the final model two out of five predictor variables were statistically significant, namely firm size and industry. Firm reputation had a Beta value (β = .07, p = 0.28). This means that if the reputation of a firm increases for one, their R&D
concentration will not increase significantly. All the relevant statistics can be found in table 4.
Table 4. Hierarchical regression model of R&D concentration.
Step Variables R R² R² Change B SE β T 1. Profitability .34 .12*** 1.047E-10 .00 .02 .29
Firm size -5.001E-8 .00 -.14* -2.12 Industry 9.041E-6 .00 .31*** 4.75 Environmental dynamism .06 .05 .08 1.22 2. Profitability .35 .12*** .01 5.321E-11 .00 .01 .15
Firm size -5.325E-8 .00 -.15* -2.24 Industry 8.727E-6 .00 .30*** 4.53 Environmental dynamism .05 .05 .07 1.09 Firm reputation .00 .00 .07 1.10
Note. Statistical significance: * p < 0.05; ** p < 0.01; *** p < 0.001.
Based on the empirical evidence, hypothesis 1 will not be supported. Although firm
reputation, after controlling for profitability, firm size, industry and environmental dynamism had a positive effect on innovation inputs, it did not have a significant effect on the R&D intensity and R&D concentration of the firm. So, based on these findings it can be assumed that firms with a higher reputation will not demonstrate more innovative behavior than firms with a lower reputation.
CEO celebrity
To test the second hypothesis, a One-Way MANCOVA was performed in order to examine the ability of CEO celebrity to predict the innovative behaviour of firms, after controlling for profitability, firm size, industry and environmental dynamism. Performing a One-Way MANCOVA showed that there was no statistically significant effect of CEO celebrity on the innovative behaviour of firms after controlling for profitability, firm size, industry and environmental dynamism, F(3,242) = 0.32 , p = .81. Looking at the different dependent
35 variables separately confirmed these findings. The effect of CEO celebrity on innovation inputs was namely not significant with p = .83 and F = .05. The effect of CEO celebrity on R&D intensity was namely also not significant with p = .37 and F = .80. The effect of CEO celebrity on R&D concentration was namely also not significant with p = .58 and F = .31. This means that there was no significant difference between celebrity CEOs and not celebrity CEOs with regard to the innovative behaviour of the firm. Based on the empirical evidence, hypothesis 2 will not be supported. So, based on these findings it can be assumed that firms with celebrity CEOs will not demonstrate more innovative behavior than firms without celebrity CEOs.
Discussion & Conclusion
The research question of this study was: how does firm reputation and CEO celebrity affect the innovative behavior of a firm? The empirical evidence of this study showed that neither firm reputation nor CEO celebrity had a significant positive effect on the innovative
behaviour of firms. This means that if the reputation of a firm is high it does not imply that it has a strategic effect on having high levels of innovative behaviour and vice versa. This study therefore does not confirm the strategic effect of firm reputation on the innovative behaviour of firms that was expected based on the existing literature. Also, celebrity CEOs do not significantly positively affect the innovative behaviour of firms, which means that the strategic effects of celebrity CEOs with regard to the innovative behaviour are more limited than existing literature predicted. The existing literature based this prediction on celebrity CEOs being overconfident and devoted to their owns strategies which can lead them to demonstrating higher levels of innovative behaviour, but this effect of celebrity CEO on firm
innovative behaviour was not found in this study. There are a number of explanations which may explain why this study did not find
36 answered satisfactorily. One possible explanation of celebrity CEOs not having a significant positive effect on the innovative behaviour of the firm, is that they might rely on and be influenced by other top managers who provide input with regard to operational decisions and planning for the long-term (Parnell, 2013). These top managers can thus diminish the
overconfidence and devotion to own strategies of celebrity CEOs which was expected to lead to demonstrating more innovative behaviour.
Existing literature argued that the celebrity status of a CEO has an impact on their strategic decisions in multiple areas (Cho et al., 2016). The fact that this study did not find empirical evidence on the positive effect of CEO celebrity on firm innovative behaviour which was expected based on existing literature raises some issues about this existing
literature. Further research is therefore required to support the claim that the celebrity status of a CEO does have an impact on their strategic decisions and the firms they manage. To resolve this issue, future research can for example examine the impact of the celebrity status of the CEO in areas such as mergers and acquisitions, foreign direct investment or diversification.
One possible explanation of firm reputation not having a significant positive effect on the innovative behaviour of the firm, is that firms with a high reputation may focus on
delivering consistent performance over time to conserve the already established reputation and this may lead to greater use of risk reduction strategies (Petkova et al., 2014). This may lead firms to not demonstrating high levels of innovative behaviour. Perrin (2002) namely argued that innovation is risky and unpredictable in terms of which particular activity works, who will benefit and when these benefits will occur.
Based on existing literature it was expected that firms with a high reputation will demonstrate high levels of innovative behaviour. As earlier mentioned, this innovative
behaviour can be considered as risky (Perrin, 2002). Petkova et al. (2014) had a different view and they argued that a firm with a high reputation may want to conserve its established
37 reputation by using risk reduction strategies. This may lead to demonstrating lower levels of innovative behaviour. The fact that this study did not find empirical evidence on neither the positive effect nor negative effect of firm reputation on the innovative behaviour of firms raises some issues about which claim prevails. Further research is therefore required to resolve this issue. Future research can for example also examine the strategic effect of firm reputation in areas such as mergers and acquisitions, foreign direct investment or
diversification.
Understanding that both the reputation of a firm and having a celebrity CEO does not affect the innovative behaviour of the firm has some implications for real-world practice. When a firm has undesirable levels of innovative behaviour, they are informed and aware that both the reputation of the firm and the celebrity CEO do not cause this undesirable state. The firm itself and the board of directors are thus more informed about the effect of their firm reputation and its CEO celebrity and they understand that there is no need to change the CEO, to change organizational routines or to change organizational roles to reduce the effect of the two social evaluations, because in fact they do not have an effect on the innovative behaviour of the firm. As a result, they can focus on other factors which influence the innovative
behaviour of firms in order to reach desirable levels of innovative behaviour.
This study contributed to the behavioral theory of the firm, because it provided insight into the strategic effects of firm reputation and CEO celebrity on the innovative behavior of firms. The strategic effects of these social evaluations were not examined previously and it provided new knowledge on whether these social evaluations have an impact on what happens inside the firm and on the way decisions are made regarding innovation. This study also contributed to social evaluation research by solely examining the effect of firm reputation and CEO celebrity on the innovative behavior of a firm empirically, which has not been done in previous research. Previous research on these social evaluations was mainly focused on
38 benefits, disadvantages and outcomes such as performance, bargaining power, attitudes and beliefs. This study thus provided new knowledge on the social evaluations as it examined the strategic effects of the two social evaluations on innovation, which increases the overall knowledge on social evaluations of the firm.
There were some strengths and limitations regarding this study. One of the strengths of this study is that it relies predominately on data which is verifiably and objective, because the data were drawn from databases, annual reports and large research/advisory/consulting firms. Other research strategies, such as interviews and surveys have the potential to include
interviewer bias and response bias. These biases influence the responses of the respondents and as a result the responses and thus data are less accurate and truthful and more skewed and biased (Morrel-Samuels, 2002; Lottie, 2011).
Another strength of this study is that it measures firm reputation based on data from the Reputation Institute and Harris Interactive and not on data from the Fortune's publication of the World's Most Admired Companies, which is done by multiple previous studies (Bear, Rahman, & Post, 2010; Chun, 2005; Fombrun & Shanley, 1990; Hammond & Slocum, 1996). Previous studies have mentioned some weaknesses with regard to the Fortune measure
(Fombrun & Shanley, 1990; Brown & Perry, 1994; Fryxell & Wang, 1994; Gardberg & Fombrun, 2002). They argued that the Fortune measure is too strongly associated with
financial performance, too strongly correlated between subscales of reputation, too focused on large and visible firms and the respondents of the surveys are not representative for the
population. These arguments provide sufficient reason to not use the Fortune's publication of the World's Most Admired Companies to measure firm reputation.
One of the limitations of this study is that there were some reporting problems with regard to expenditures on research and development in COMPUSTAT, which is similar in previous studies (Artz et al., 2010; Galasso & Simcoe, 2011). Some firms could therefore not
