CEO Overconfidence and Ambidextrous Innovation

Abstract

While prior research has focused on the effect of CEO overconfidence on innovation, few studies have been conducted to reveal how and whether an overconfident CEO affects ambidextrous innovation, which means the simultaneous and balanced pursuit of both exploratory and exploitative innovation. By observing firms’ patenting behavior, we investigate the effect of CEOs’ psychological attribute of overconfidence on innovation ambidexterity. In addition, we examine how a firm’s governance system moderates the relationship between CEO overconfidence and ambidextrous innovation. The results show that overconfident CEOs are more apt to create or magnify an imbalance in innovation ambidexterity. Furthermore, the results regarding the moderating effects of governance and monitoring mechanisms indicate that an independent board and dedicated institutional ownership mitigate the positive relationship between CEO overconfidence and a firm’s ambidextrous imbalance, while transient institutional ownership enhances this relationship. We also find that analyst following does not effectively monitor an overconfident CEO’s tendency toward an ambidextrous imbalance.

Keywords

CEO overconfidence ambidextrous innovation independent board analyst following institutional ownership

Introduction

Innovation ambidexterity is the important dynamic capability of a firm to simultaneously pursue exploratory and exploitative innovation strategies (e.g., Ancona, Goodman, Lawrence, & Tushman, 2001; Tushman & O’Reilly, 1996). Exploratory innovation refers to when a firm creates new technological knowledge that is novel relative to its existing knowledge stock (Benner & Tushman, 2002). For example, Ericsson, a leader of the technological development of mobile telephony in the 1980s, mainly focuses on exploratory mobile technologies. In contrast, exploitative innovation occurs when a firm builds technological knowledge from its existing knowledge stock (Rosenkopf & Nerkar, 2001). Eastman Kodak, a well-known example of exploitation, devoted the majority of its research and development (R&D) resources to support its original chemical photography technology, whereas digital photography was not taken seriously into consideration (Benner & Tushman, 2002; Rosenkopf & Nerkar, 2001).

Overemphasizing either the exploratory or exploitation approach may not always benefit a firm. Although the exploratory approach may be beneficial to new product or service development, integrating overly unfamiliar knowledge or technologies leads a firm to operation inefficiency and increased costs (Fleming & Sorenson, 2001; Tsai & Huang, 2008). A firm highly focused on the exploratory approach without considering the role of exploitation not only continuously drains a firm’s resources but also fails to generate immediate financial gain to support future innovative investments, thereby eventually reducing a firm’s reliability in the industry (Levinthal & March, 1993). The case of Ericsson overinvesting in exploratory activities forced the firm to decrease its ability at adaptability in relation to increasingly commoditized markets and thus laid off around 6,000 employees and closed most of its technology centers in the 1990s (Birkinshaw & Gibson, 2004).

As exploitation is related to the reuse of existing knowledge (Katila & Ahuja, 2002), the returns associated with exploitation are more invariable and reliable than those associated with the exploratory approach. However, a strong focus on exploitative innovation limits the improvement of a firm’s knowledge trajectory, which not only diminishes the returns associated with the same knowledge but also causes a firm to collapse into a “competence trap” that constrains exploratory capabilities and long-term competitiveness (March, 1991; McGrath, 2001). The failure of Eastman Kodak shows that spending heavily R&D on exploitation but little on explorative activities loses a firm’s competitiveness. Through the channel of innovation ambidexterity, a firm would avoid falling into a dilemma when it centers extremely on either exploratory or exploitative innovation.

Achieving innovation ambidexterity by pursuing both short-term financial resources and long-term innovative abilities enables a firm to enhance its performance and competitiveness (Cao, Gedajlovic, & Zhang, 2009). While a firm would attain ambidexterity through organizational structures (O’Reilly & Tushman, 2013), established culture (Wang & Rafiq, 2014) or top management team arrangement (Lubatkin, Simsek, Ling, & Veiga, 2006), prior literature suggests that such a firm is still likely to experience innovative imbalance (He & Wong, 2004; March, 1991; Uotila, Maula, Keil, & Zahra, 2009). Current research has attributed imbalances in innovation ambidexterity to organizational or managerial antecedents, such as the leadership of the top management team, resources, ownership and organizational structure, and competition (Cao et al., 2009; Cao, Simsek, & Zhang, 2010; Gedajlovic, Cao, & Zhang, 2012). However, CEO overconfidence, a key psychological attribute of CEOs, has been identified as an important characteristic that may significantly affect a firm’s innovation decisions (Galasso & Simcore, 2011; Hirshleifer, Low, & Teoh, 2012; Simon & Houghton, 2003). Few studies have investigated the role that CEO overconfidence plays on the balance of a firm’s innovation ambidexterity. This study complements the lines of research regarding organizational ambidexterity as well as CEO overconfidence by studying the effect of CEO overconfidence on a firm’s innovation ambidexterity, which is a critical issue for a firm’s current and long-term success (He & Wong, 2004).

Upper echelons theory argues that CEOs assess information based on their experiences, values, and personalities (Hambrick & Mason, 1984). The psychological attributes of CEOs, such as overconfidence, are thus manifested in a firm’s strategic choices and organizational outcomes (Hambrick, 2007). However, bounded rationality limits a CEO’s cognitive capabilities in the interpretation of events (Hambrick, 2007; Hambrick & Mason, 1984). CEOs’ selective managerial perceptions therefore largely distort their strategic choices and thus affect a firm’s innovation strategy (Campbell, Gallmeyer, Johnson, Rutherford, & Stanley, 2011; Hiller & Hambrick, 2005; Hambrick & Mason, 1984).

CEO overconfidence refers to a CEO whose decision-making process suffers from overplacement, overprecision, and overestimation (Moore & Healy, 2008). Overplacement is defined as when a CEO perceives that he or she is better than others. Overprecision leads CEOs to be inclined to overstate both the precision of their forecasts as well as the information they possess. Overestimation is when a CEO overestimates their ability, performance, level of control, and probability of success (Bollaert & Petit, 2010; Malmendier & Tate, 2005; Moore & Healy, 2008; Picone, Dagnino, & Minà, 2014). Together, these characteristics lead CEOs to be inclined toward a carelessly quick, centralized, as well as faultily persistent decision-making process, and to undertake excessive risk or exaggerated initiatives (Hiller & Hambrick, 2005; Picone et al., 2014). When CEOs are overconfident, the decisions they make are thus potentially biased by the limitations that overplacement, overprecision, and overestimation place on their cognitive capabilities. Such a bias could lead them to pursue innovation strategies that lean toward the extreme of either exploration or exploitation.

Overinvestment in exploratory or exploitative innovation by an overconfident CEO imposing inferior firm performance on shareholders implies that innovation projects are not always in line with the interests of shareholders (Polk & Sapienza, 2009; Titman, Wei, & Xie, 2004). When overconfident CEOs discretionarily overinvest capital funds in innovations to maximize their personal hubristic utility at the expense of the shareholders, the agency conflict between CEOs and shareholders occurs (Jensen, 1986; Richardson, 2006). Agency theorists argue that external governance mechanisms, such as an independent board, analyst following, and institutional ownership, can help mitigate such conflict (Fama & Jensen, 1983). For example, independent directors play a monitoring and advisory role in corporate strategy and resource allocation decisions (Solomon, 2007), analysts who collect and evaluate specific information regarding a firm’s important projects have the capability to monitor overconfident CEOs and their innovation investments (D’Mello & Ferris, 2000), and institutional investors can press overconfident CEOs to carefully evaluate innovation strategies due to their significant level of ownership (Wright, Kroll, & Elenkov, 2002). In this study, we therefore also investigate whether and how these external monitoring functions play important roles in leading overconfident CEOs toward an ambidextrous balance in their exploratory and exploitative innovation strategies.

There are three main contributions of this study. First, although recent literature complemented the extant debate on overconfidence and decisions regarding innovations, the results of these studies have been inconclusive. Simon and Houghton (2003) found that overconfident CEOs embrace pioneering but are less likely to be successful in product introduction, while Tang, Li, and Yang (2015), and Hirshleifer et al. (2012) showed that overconfident executives contribute positively to firm innovation and performance. Moreover, previous research focused on how an overconfident CEO affects innovation as a whole rather than whether overconfident CEOs guide a firm toward achieving a balance between exploitation and exploration. For example, Galasso and Simcore (2011), Tang et al. (2015), and Hirshleifer et al. (2012) focused on the success in obtaining more patents and patent citations. However, the investment in patents may not always reflect a firm’s long-term success (Artz, Norman, Hatfield, & Cardinal, 2010). These studies did not discuss whether overconfident CEOs pursue the dynamic capabilities of innovation ambidexterity and its impact on firm performance. Our study also differs from the research of Simon and Houghton (2003), which documented the inclination of overconfident CEO toward pioneering new products. We argued that the propensity of an overconfident CEO to either expletory or exploitative innovations may be due to his or her persistency. By applying the measures suggested by Katila and Ahuja (2002) to capture expletory or exploitation factors, this study complements upper echelon theory as well as extends the research regarding overconfidence and innovation by further revealing how an overconfident CEO directs a firm’s innovation toward ambidextrous imbalance, thereby influencing firm performance.

Second, both Chatterjee and Hambrick (2007) and Wales, Pankaj, and Lumpkin (2013) noted that overconfident CEOs often experience extreme organizational performance. Our study complements this line of research, finding that an overconfident CEO with an ambidextrous imbalance suffers lower firm performance, as a deviation toward either exploratory or exploitative innovation results in suboptimal innovation strategies for a firm. This is because, on the one hand, a firm which inclines to exploitative innovation, fails to generate stable and sufficient earnings from exploitation, and is thus less likely to finance long-term exploratory innovation. On the other hand, the propensity to exploitation limits a firm in improving its knowledge trajectory by pursuing exploration, and subsequently constrains a firm’s competitive advantages (Leonard-Barton, 1992; Uotila et al., 2009).

Third, while rational agents who are self-interested as well as opportunistically maximize their own utility (Jensen & Meckling, 1976), overconfident CEOs also have the propensity to pursue and maximize individual interest and utility such as investment in exploratory or exploitative innovation to represent personal talent and capabilities (Richardson, 2006; Tang et al., 2015). Previous studies regarding corporate governance have suggested internal and external mechanisms that resolve agency problems between a firm’s agents and its shareholders (Bushee, 2004; Frankel & Li, 2004; Peng, 2004). However, because of overestimating the probability of successful performance regarding the innovative investments, overconfident CEOs are inclined to withhold options when prices are already sufficient high in order to maximize their wealth. Internal mechanisms, such as CEO option–based compensation, may not work effectively in solving conflicts between shareholders and an overconfident CEO who is intent on following a strategy of innovation that leads to an ambidextrous imbalance.

In addition, previous studies argued that governance attributes improve agency problems. Nevertheless, literature regarding overconfidence has overlooked the role governance structure plays in the context of an overconfident CEO (Picone et al., 2014). Even though some research has investigated the moderating role of managerial discretion, market or environmental factors, board structure and the overall institution in the relationship between CEO overconfidence and the inclinations of firm risk taking, acquisition decisions, accounting conservation, financial misreporting as well as firm performance (Ahmed & Duellman, 2013; Eranova & Prashantham, 2016; Kolasinski & Li, 2013; Li & Tang, 2010; Picone et al., 2014; Schrand & Zechman, 2012; Tang et al., 2015), it is still unclear which governance mechanisms help mitigate the problem of ambidextrous imbalance that arise due to CEO overconfidence. Essentially, different governance mechanisms have different effects on an overconfident CEO who is discretionary in innovative decision making. In this study, based on the context of ambidextrous innovation, we complement prior studies regarding CEO overconfidence and corporate governance to specifically determine whether an overconfident CEO can be monitored through external control mechanisms such as an independent board, analyst following, and dedicated or transient institutions to reduce the propensity of overinvestment in either exploration or exploitation.

This article proceeds as follows. Section 2 discusses the theoretical background. Section 3 describes the sample selection and methodology. The empirical results are presented in Section 4, while Section 5 offers the discussion and conclusions.

Background and Hypotheses

CEO Overconfidence and an Imbalance in Innovation Ambidexterity

Upper echelons theory suggests that CEO cognition is affected by a CEO’s psychological attributes. Holding the authority of management and directing the firms’ strategies, CEOs are crucially involved in making important firm decisions (Goel & Thakor, 2008). Embedded with the attributes of overestimation, overprecision, and overplacement, an overconfident CEO’s preferences and biases therefore highly influence a firm’s decision-making process and thus its innovation ambidexterity (Miller, Kets De Vries, & Toulouse, 1982). This can potentially lead a firm toward either an exploratory or exploitative imbalance.

The reasons that an overconfident CEO may deviate toward an exploitative imbalance is that, first, overconfident CEOs highly credit themselves in their ability to accurately and correctly assess situations and environments (Klayman, Aoll, González-Vallejo, & Barlas, 1999). This leads an overconfident CEO to be apt to strongly adhere to a fixed goal by keeping a certain route to pursue the firm’s strategy (Picone et al., 2014). Overconfident CEOs who tend to miscalculate may also be more likely to attribute failure to bad luck, rather than their own poor decisions. In the same way, they are more apt to associate successes with their own abilities. This leads them to incorporate less information from previous decisions into their future investments, and to resist correcting their errors over time (Chen, Crossland, & Luo, 2015). Moreover, overconfident CEOs who are certain that they can successfully perform a task do not see the need for others to confirm their decisions (Hiller & Hambrick, 2005; Malmendier & Tate, 2005). In particular, when confronted with conflicting signals between current and future success or environmental changes, overconfident CEOs are stubbornly persistent in the pursuit of certain strategies (Hiller & Hambrick, 2005). Therefore, a firm may experience an exploitative imbalance when an overconfident CEO is persistent in his or her pursuit of innovation strategies.

Second, the attribute of overestimating their personal capabilities and performance leads CEOs to believe that they hold the recipe for success, and favor the repetition of a certain manner across different types of decisions (Hiller & Hambrick, 2005). In addition, overconfident CEOs generally prefer to undergo a quick decision-making process (Hambrick, Cho, & Chen, 1996) based on their already accumulated successful experiences and their belief that they know the correct answers and have a clear understanding of the opportunities underlying a specific strategy (Hiller & Hambrick, 2005; Picone et al., 2014). However, an overconfident CEO who prefers a quick decision-making process and mechanically repeats actions often pay little attention to strategy formulation. This may prevent such a CEO from seeing the need for change based on the competitive environment (Picone et al., 2014). In sum, when a CEO is overconfident, he or she may execute a noncomprehensive and fast decision-making process that can lead a firm toward an exploitative imbalance due to their previous successful experiences within exploitative innovation, without considering the dynamic and competitive environment.

Third, overconfident CEOs often believe that they are better than their peers and possess all necessary insights and skills. They also perceive that they can lead the organization to achieve superior performance. Such CEOs generally favor a centralized decision-making process, and put a firm’s responsibilities on their own shoulders (Hiller & Hambrick, 2005; Picone et al., 2014). This propensity of gut feeling generally ignores the potential contribution of others, and prevents overconfident CEOs from accepting relevant information and intelligence which does not support their intuitive judgments (Owen, 2012). While overconfident CEOs incline to defend against feedback from other organizational members and therefore exclude other individuals from strategy decisions, experiences from subordinates are critical to efficaciously solving problem within the leaders’ decision-making process (Picone et al., 2014; Tetlock, 2000; Wakefield, 2003).

Furthermore, subordinates are likely to conform to the projects performed by the overconfident CEO even though those decisions have destructive influences, as people have the tendency to comply with authorities and group norms (Padilla, Hogan, & Kaiser 2007). As a result, overconfident CEOs who prefer centralization potentially lead subordinates to provide the CEO with innovative information that fits the CEO’s known acceptance zone (Hambrick & Fukutomi, 1991; Tetlock, 2000). These issues inhibit CEOs from being able to see the potential threats inherent in their innovative initiatives (Picone et al., 2014). Thus, we may expect that without challenges from other organizational members, overconfident CEOs with past successful exploitative experiences will incline to allocate resources to exploitative innovation regardless of the threats inherent in such an ambidextrous imbalance.

On the other hand, overconfident CEOs may lead a firm toward an exploratory imbalance. First, as discussed above, overconfident CEOs who possess the attributes of overestimation, overprecision, and overplacement tend to guide a firm’s innovation strategy under a persistent, quick and centralized decision-making process. Therefore, we may expect that a firm deviates toward an exploratory imbalance if the CEO holds past successful experience that is related to exploratory innovation, despite the fact that exploratory innovation is embedded with uncertainty and risk to a greater extent than exploitative innovation activities.

Second, because overconfident CEOs overestimate their personal ability to control outcomes and solve problems that may arise after investment (Bollaert & Petit, 2010; Picone et al., 2014), they often underestimate the uncertainties of a project and the probability of failure (Malmendier & Tate, 2005). Additionally, overconfident CEOs perceive there is less risk in general, as they overvalue the successful opportunities and the potential gains of investment plans (Bollaert & Petit, 2010; Hiller & Hambrick, 2005). As a result, overconfident CEOs are more likely to undertake a large number of risky investments (Goel & Thakor, 2008; Hiller & Hambrick, 2005). Moreover, overconfident CEOs are particularly prone to pursue more venturesome enterprises, which can be attributed to their predominant belief that they are not only able to successfully perform difficult tasks better than others but also to their self-confident attitude that the firm is in their capable management (Bass, 1990). Exploratory innovation not only involves the danger of squandering precious resources for little payback in the short term but also is perceived to be difficult and risky (Uotila et al., 2009). Success in such innovation represents an indicator in favor of the CEO’s strong managerial capabilities (Tang et al., 2015). As such, it is common that overconfident CEOs who are negligent in risk incline to engage in large-stake exploratory innovations (Simon & Houghton, 2003).

From the aforementioned discussions, we may expect that when overconfident CEOs allocate firm resources to innovation strategies, they are more likely to invest in innovation projects that are consistent with their cognitive bias. They will thus lead a firm toward either an exploratory or exploitative imbalance.

Hypothesis 1: CEO overconfidence is positively associated with a firm’s ambidextrous imbalance in exploratory versus exploitative innovation.

The Moderating Effect of Independent Directors

Corporate governance research suggests that a firm’s board of directors has a critical role in mitigating the conflict of interest that arises between managers and shareholders, because the structure of the board may influence the strategic choices of a firm (Zahra & Pearce, 1989). Independent directors are members of the board who do not have current or past professional or personal associations with the firm (Johnson, Hoskisson, & Hitt, 1993). A strong board of directors, which typically has a majority of independent directors, is less likely to be dominated by the firm’s CEO (Fama & Jensen, 1983; Johnson et al., 1993). Moreover, in order to protect human capital, independent directors also have a higher incentive to fulfill their fiduciary duties to monitor managers on the behalf of shareholders (Wright et al., 2002; Zahra, Neubaum, & Huse, 2000). Thus, independent directors can play a central role in preventing the misuse of power and authority by CEOs in the allocation of a firm’s resources for innovation (Conyon & Peck, 1998).

In addition to their monitoring function, an independent board with relevant expertise, experience, knowledge, and skills may not only have the ability to give advice and provide counsel regarding CEOs’ discretional investments but also possess the means to resolve problems during the process of innovation (Nielsen & Huse, 2010). This assistance may be in the form of helping a firm to search for appropriate exploratory as well as exploitative innovation strategies, and to avoid evaluating innovation strategies solely from the individual view of the CEO (Hillman & Dalziel, 2003). Prior studies have shown that the more independent the director, the more the director is likely to be involved in a firm’s strategic decision-making process, which leads the director not only to become more familiar with the firm’s innovation activities but also to gain valuable knowledge that may further broaden the board’s perspective and push executives toward achieving an ambidextrous balance (Johnson et al., 1993; Zahra et al., 2000). Brickley, Coles, and Terry (1994) suggested that a higher percentage of independent directors may be associated with greater firm value. Peng (2004) also found that the presence of independent directors is positively associated with firm performance.

Based on the argument above, we can therefore expect that when a firm has an overconfident CEO who may potentially misallocate a firm’s resources, favoring either exploratory or exploitative innovation projects that would lead to an ambidextrous imbalance, independent directors can play a critical role in mitigating such overinvestments.

Hypothesis 2: An independent board mitigates the positive relationship between an overconfident CEO and a firm’s ambidextrous imbalance in exploratory versus exploitative innovation.

The Moderating Effect of Analyst Following

Innovation projects are associated with a high level of information asymmetry since corporate insiders, such as CEOs, often have greater access to firm information than shareholders (Aboody & Lev, 2000).¹ Information asymmetry concerning innovation exacerbates agency problems because shareholders do not have adequate information to judge the decisions performed by self-serving CEOs (Comment & Jarrell, 1991).

Analysts who are not only able to request but also analyze specialized resources and detailed knowledge of firm-specific information can reduce agency conflict, and thus serve as an external monitor of managerial activity (D’Mello & Ferris, 2000). Moreover, as they are industry experts, analysts help evaluate a firm’s innovation decisions and provide shareholders with sound analyses concerning innovation projects. Based on the professional information provided by analysts, investors react to executives’ innovation decisions by trading securities (Wright et al., 2002). In this way, analysts play a monitoring role by helping reduce information asymmetry regarding innovation activities.

With an analyst following, we can expect that overconfident CEOs will carefully evaluate both exploratory and exploitative innovation investments toward an ambidextrous balance, allowing them to pursue better performance in order to meet the expectations of analysts. Chung and Jo (1996) suggested a positive relation between firm quality and analyst following, and provided evidence consistent with analysts serving as independent monitors. Frankel and Li (2004) found that firms with greater analyst following face less information asymmetry between insiders and investors.

Hypothesis 3: Analyst following mitigates the positive relationship between an overconfident CEO and a firm’s ambidextrous imbalance in exploratory versus exploitative innovation.

The Moderating Effect of Institutional Investors

Prior studies have suggested that institutional investors with block shareholdings as well as informed oversight skills have both the incentive and the means to prevent a CEO from engaging in self-serving behavior. Thus, institutional investors help mitigate potential conflicts to ensure that managerial decisions are in line with other shareholders’ interests (Aguilera, 2005). Previous studies have indeed found institutional investor activism to be a source of external monitoring (Wright et al., 2002; Zahra, 1996). However, the monitoring function of institutional ownership is contingent on the type of institutional investor, for example, dedicated versus transient institutional investors (Bushee, 1998; Tihanyi, Johnson, Hoskisson, & Hill, 2003). This is because different institutional investors have different investment objectives (David, Kochhar, & Levitas, 1998; Kochhar & David, 1996). These distinct objectives lead to different preferences in terms of investment horizons, which influence the institutions’ willingness to use their power to monitor managers (Zahra et al., 2000).

Dedicated institutions, such as pension funds, with direct long-term interests in regard to their portfolios, can access private information better in order to pursue investor-oriented policies by actively participating in a firm’s strategic decision-making process (Porter, 1992; Tihanyi et al., 2003). The reason is that such investors generally hold block equity and extend the investment horizons of a firm, which allows them to understand richer and more complex information about the firms in which they invest (Bushee, 2004; Bushee & Noe, 2000; Porter, 1992). Thus, dedicated institutions not only have the incentive but also the capabilities to develop and provide expertise in evaluating the potential strategic actions of a firm (Bushee, 2001), which in turn facilitates advice seeking executives of the firms that the dedicated institutions invest in to be willing to augment a firm’s private information with the dedicated institutional investors (Pollock, Rindova, & Maggitti, 2008), and help a firm focus on competitive actions that produce long-term value. For example, Edward Lampert, the founder of ESL Investments Inc. (an employee-owned hedge fund) has a reputation for working with the firm he invested in to help improve the firm’s performance (Connelly, Tihanyi, Certo, & Hitt, 2010). In addition, these investors face pressure from their members to maximize the long-term value of their holdings (Blair, 1995). Such institutions would be highly motivated to exercise their voting power through direct negotiations, public announcements, and shareholder proposals on decisions performed by CEOs (David, Hitt, & Gimeno, 2001). Furthermore, by directly affecting a firm’s corporate governance structure, dedicated institutions are able to control managerial opportunism through their concentrated ownership (Tihanyi et al., 2003).

Exploratory innovation contributes to a firm’s long-term success with little payback in the short term (Uotila et al., 2009). Exploitative innovation, on the other hand, may increase efficiency and lead to positive short-term performance but sacrifice long-term competitive advantage through the exploration of new competencies (Leonard-Barton, 1992). However, firms need an ambidextrous balance to achieve optimal performance (Gupta, Smith, & Shalley, 2006). In order to maximize the overall value of their portfolios through ambidextrous balance, dedicated institutions should monitor both exploratory innovation activities to safeguard a firm’s long-term performance (O’Brien, 2007), and exploitative innovation to ensure sufficient funding for the exploratory projects (Kochhar & David, 1996). Based on the above, we can expect that dedicated investors with voting power and professional knowledge will influence overconfident CEOs to rearrange their overinvestment in either exploratory or exploitative innovation to achieve an ambidextrous balance.

Hypothesis 4a: Dedicated institutional ownership mitigates the positive relationship between an overconfident CEO and a firm’s ambidextrous imbalance in exploratory versus exploitative innovation.

However, transient investors such as mutual funds may not intend to play an active monitoring role relative to managerial decisions because they generally invest in a firm for a short period (Bushee, 1998, 2004; Zahra, 1996). Due to their short-term investment behavior, transients are less likely to own a stake in a firm long enough to realize the gains associated with strategic competitive actions (Abarbanell, Bushee, & Raedy, 2003; Bushee, 1998, 2001). Therefore, transient institutions tend to sell their shareholdings rather than challenge managerial decisions when conflicts arise between the managers and themselves (Jacobs, 1991). Thus, transient investors may be reluctant to influence managerial decisions by the exercise of voting power through direct negotiations, public announcements, and shareholder proposals, that is, they may not actively execute their monitoring role. For example, Jennison Associates LLC and MFS Investment Management acquired equity stocks of Home Depot but only held those shares for less than 1 year before selling them (Connelly et al., 2010).

Given the nature of their inactive monitoring role, we can expect that transient institutions may sell their holdings when overconfident CEOs suffer from ambidextrous imbalance. As a result, with the lack of active monitoring from transient investors, an overconfident CEO can be expected to have even greater discretionary power to overinvest in their preferred innovation activities.

Hypothesis 4b: Transient institutional ownership enhances the positive relationship between an overconfident CEO and a firm’s ambidextrous imbalance in exploratory versus exploitative innovation.

Methodology

Sample Selection

In this study, a data set regarding publicly traded U.S. firms from 2003 to 2010 was used to test how CEO overconfidence affects a firm’s ambidextrous imbalance. The sample set included all companies from both traditional manufacturing as well as high-tech industries for which the measures of exploratory and exploitative innovation activities are particularly relevant. Specifically, patenting is crucial to the success and survival of manufacturing and high-tech-based firms (Allred & Park, 2007; Lee, Wu, & Pao, 2014), because patents can protect and generate enormous profits. In addition to the function of profit, patent data offer us a lens to observe a firm’s innovative behavior by the citation information (Katila & Ahuja, 2002). For this reason, firms in manufacturing and high-tech sectors provide a suitable research context.

We first used the firm identifier (GVKEY) to merge a panel of publicly listed firms to the patent data obtained from the U.S. Patent and Trademark Office. Second, the set of U.S. patent data with the firm identifier was further matched to the ExecuComp database. All ExecuComp firms that operate in the same four-digit SIC industries as the firms in the patent database were included. Finally, financial data from Compustat as well as stock prices from CRSP were collected. Regarding the treatment of missing data, the majority of missing data come from the dependent variable, which is not suggested to be replaced by other means (Allison, 2001). We therefore adopted the most common method to drop the sample firms which do not have complete data. All the independent as well as control variables were lagged by 1 year to a firm’s tendency of ambidextrous imbalance in regard to its exploratory versus exploitative innovation activities. Following these procedures, the final sample comprised 1,476 firm-year observations from 297 firms.

Dependent Variable Measure

In this study, to test how CEO overconfidence affects a firm’s tendency toward an ambidextrous imbalance in exploratory versus exploitative innovation, we focused on patent data that show how a firm draws on elements of knowledge (patent citations) it has previously used, which reflects the firm’s practice of searching and exploiting its extant knowledge stock (Phelps, 2010). Patent data used to measure exploratory and exploitative innovation is appropriate for this study because patents are not only measures of novel inventions but also valid and robust indicators because of the patent examination process (Griliches, 1990). By extending the prior studies that used the multiplication of exploration and exploitation to measure a firm’s ambidextrous balance (e.g., Katila & Ahuja, 2002; Lee & Huang, 2012), ambidextrous imbalance was computed as the reciprocal of the product of exploratory innovations times exploitative innovations. The higher value indicates that firms behave in a more balanced innovation policy.

Following prior studies, exploratory and exploitative innovation were measured by U.S. patent citations (e.g., Benner & Tushman, 2002; Katila & Ahuja, 2002; Rosenkopf & Nerkar, 2001). Considering the perishable nature of technological knowledge, this study takes Argote’s (1999) suggestion to adopt a firm’s patents issued in the past 5 years prior to the focal year. Exploratory innovation was measured by the share of citations in the focal year’s citations that are not found in the previous 5 years’ list of patent citations (i.e. new citationsi,t /total citationsi.t $n e w c i t a t i o n s_{i, t} / t o t a l c i t a t i o n s_{i . t}$ ). Exploitative innovation was measured by the average number of times that each citation in the focal year has been repeatedly used during the past 5 years (i.e. $\sum_{y = t - 5}^{y = t - 1} r e p e t i t i o n c o u n t_{i, y} / t o t o a l c i t a t i o n s_{i, t}$ ). Information on U.S. patents was obtained from the U.S. Patent and Trademark Office, which provided information on U.S. patents during our sample period, including application and grant years, citations to other patents, and assignee codes that can be used to identify the owners.

Independent Variable Measures

Measure of CEO Overconfidence

To test the hypotheses, following the method of Malmendier and Tate (2005, 2008), and Campbell et al. (2011), CEO overconfidence was measured by the variable of option-based CEO overconfidence, which suggests that executive options give the CEOs the right to purchase company stock at a certain stock price. On exercise, the CEOs receive the company’s shares, and almost immediately resells them (Malmendier & Tate, 2005).

This measure is used for the following reasons. CEOs are typically overexposed to idiosyncratic risk in regard to their firm because a large part of their compensation comes from grants of stock and options. The CEOs’ personal wealth may fluctuate dramatically with the stock price of their firms. In addition, since their human capital is embedded with the firms, negative outcomes not only affect their personal wealth but also influence their future employment opportunities. Unlike perfectly hedged outside investors, CEOs have to trade off their option value against the costs of underdiversification. To maximize personal wealth as well as to avoid idiosyncratic risk and undiversified stock portfolios, CEOs should exercise their options early when the stock price is sufficiently high, as executive options are nontradable, and CEOs cannot hedge the risk of their holdings by short selling company stock (Hall & Murphy, 2002; Malmendier & Tate, 2005). Prior research has indeed suggested that CEOs who are risk averse and underdiversified exercise their options early or intend to minimize their shareholdings (e.g., Malmendier & Tate, 2005). However, overconfidence may lead CEOs to overestimate the future returns of their investment projects. As a result, overconfident CEOs may believe that the stock price will continue to rise under their leadership, and thus may postpone exercising their options, or buy additional company stock to increase personal wealth from the expected future gains.

As risk averse and underdiversfied CEOs should exercise options at the first opportunity in the vesting period, that is, if the option value in the money is beyond a rational benchmark, this study defined CEOs as overconfident if they did not exercise stock options when the stock price was higher than the exercise price by more than 67% during the sample period (Campbell et al., 2011; Malmendier & Tate, 2005, 2008). This study applied a cutoff across the full sample the first time the CEO exhibited such behavior. A CEO was thus classified as overconfident if he or she appeared to exhibit such holding behavior once over the past 5 years before the year that the firm’s tendency toward an ambidextrous imbalance in exploratory versus exploitative innovation.

Because the stock option holding and exercise data in the IBES database that we used are not as detailed as the proprietary data that Malmendier and Tate (2005, 2008) used in their studies, we cannot acquire the realized returns. In addition, the degree of option moneyness for overconfident CEOs mainly focuses on the exercisable option, but CEOs chose to hold rather than to sell as such CEOs have a propensity to postpone exercising their exercisable options due to their overconfidence syndrome. Following the approach of Campbell et al. (2011) and Hirshleifer et al. (2012), this study used an approximation method to estimate the average exercise price of the options, which is computed as the stock price at the fiscal year end minus the per option realizable value. This method has have been widely adopted by management (e.g., Dutta, Malhotra, & Zhu, 2016; Engelen & Neumann, 2015; Galasso & Simcore, 2011) as well as finance and accounting literature (e.g., Campbell et al., 2011; Hirshleifer et al., 2012; Kim, Wang, & Zhang, 2016). The realizable value per option is computed as the total realizable value of the exercisable options which CEOs choose to hold divided by the number of exercisable options which CEOs choose to hold (both of the data collected from IBES database). The average percentage of moneyness of the options is then computed as the per option realizable value divided by the estimated average exercise price. The dummy variable of option-based CEO overconfidence equaled one for CEOs that did not exercise stock options when the option value was more than 67% in the money, and zero otherwise (Campbell et al., 2011; Malmendier & Tate, 2005). Option data were obtained from the IBES database in Wharton Research Data Services.

Independent Board

Independent board was measured by dividing the number of independent directors by the total number of directors on a company’s board. Independent directors were defined as directors without current or past professional or personal associations with the firm, such as business connections with the company (Johnson et al., 1993; Zahra et al., 2000). Data regarding independent directors were collected from the RiskMetrics database in Wharton Research Data Service.

Analyst Following

In this study, two types of analyst activity data were collected first: the number of analysts and the intensity of analyst following. According to Wright et al. (2002), the number of analysts is defined as the number of analysts that followed a firm, while the intensity of analyst following was measured by adjusting for firm size. Specifically, the intensity of analyst following was computed as the difference between the actual number of analysts following the firm and the value predicted by the regression model. The expected intensity of analyst following was estimated by the regression model: “Number of analysts following firm= a₀ + a₁ firm size_i,t + e_i,t.” Sales were used as the proxy for firm size. The residual values were further employed as the proxy for the intensity of analyst following. Data concerning the nature of analyst activity were obtained from the IBES database in Wharton Research Data Services.

After the data were collected, a principal component analysis with varimax rotation was conducted to reduce multicollinearity prior to applying linear regression. The principal component analysis produced a single factor explaining 97.32% of the variance. The factor loadings for the number of analysts and the intensity of analyst following were both 0.97; the eigenvalue was 1.95. These results suggest that the number of analysts and the intensity of analyst following can be grouped into a composite indicator of analyst following. Accordingly, we calculated a single analyst following variable as the standardized factor score for this common factor.

Institutional Ownership

Bushee and Noe (2000) suggested that institutional investors could be classified into transient and dedicated institutions based on the turnover rate, trading sensitivity, and the diversification of their portfolios. Dedicated investors tend to be focused and have longer term views in regard to managerial decisions while transient institutions tend to hold fragment stakes in the firms they invest in for short periods and prefer quick stock value gains (Porter, 1992).

To capture the governance effect of institutional investors, in this study, we first categorized institutions based on Bushee’s (1998, 2004) coding scheme. Bushee’s coding uses factor and cluster analysis to parsimoniously classify institutional owners (Abarbanell et al., 2003; Bushee & Noe, 2000). This approach first condenses variables which describe the past investment behavior of institutions into three factors related to portfolio diversification, portfolio turnover, and trading sensitivity (for a review of the variables used for factor analysis, please see Bushee and Noe, 2000). Second, based on the three factor scores conducted by factor analysis, institutional investors are then clustered into dedicated, transient, and quasi-indexer investor by adopting cluster analysis. As suggested by Connelly et al. (2010) that quasi-indexers are not differentiated from either dedicated or transient institutions, this study explored the effect of most differential institutions (dedicated or transient investors) on the relationship between overconfidence and ambidextrous imbalance. Transient investors have high diversification, portfolio turnover, and high trading sensitivity to current earnings, while dedicated investors exhibit low diversification, portfolio turnover, and trading sensitivity. The variables Transient institutional ownership and Dedicated institutional ownership are then calculated as the sum of the percentage equity ownership held by either transient or dedicated investors in the focal firms, respectively. Data regarding percentage of stock held by the two classes of institutional investors were collected from the Thomson Reuters/Institutional (13f) Holdings in Wharton Research Data Service.

Control Variables

Several organizational characteristics have been proposed or approved to influence a firm’s ambidextrous innovation tendencies, such as firm size, firm age, R&D intensity, patent stock, financial leverage, and firm growth. First, we controlled for the effect of firm size, measured by the natural logarithm of the number of year-round employees, because prior studies have considered the influence of firm size on ambidextrous innovation (e.g., Raisch & Birkinshaw, 2008). Given that organizational inertia has been proposed as influencing a firm’s achievement of ambidextrous innovation (O’Reilly, Harreld, & Tushman, 2009), it was necessary to control for the potential effect of organizational inertia. The organizational inertia was proxied by firm age, calculated as the number of years since its establishment. R&D intensity (R&D expenditure/sales), which is often seen as a sign of absorptive capacity (Cohen & Levinthal, 1990), has an influence on a firm’s strategic choice. Rothaermel and Alexandre (2009) also empirically found the effect of R&D on ambidexterity. The possible effect of R&D intensity on ambidextrous innovation needs to be controlled. Lee and Huang (2012) found that patent stock has an impact on ambidextrous innovation since it serves as the foundation for creating new technological solutions. Patent stock is used as the natural logarithm of a firm’s granted patents in the past 5 years (t−4 to t). A firm’s ambidextrous innovation may also be influenced by its financial leverage and growth opportunity because higher financial leverage may deter managers from engaging in risky business ventures (Bierly, Damanpour, & Santoro, 2009), while greater firm growth opportunities may induce decision makers to undertake various forms of risk-taking behavior (Thakur, 1999). We measured financial leverage computed as total debt divided by total assets. Firm growth in our study was calculated as the growth rate of total assets. Financial data were collected from the COMPUSTAT database in Wharton Research Data Service.

Empirical Results

Descriptive statistics and correlations for the variables of this study are reported in Table 1. The correlation coefficients of the main variables of interest are modest, suggesting that the multicollinearity problem of variables should be not an issue. The data structure of this study is a cross-sectional time series design. We therefore followed the method employed in prior studies to obtain the benefits of panel data (e.g., Samila & Sorenson, 2010). Each firm has its own individual characteristics which may or may not influence the predictor variables. It is thus necessary to control these hidden factors that may affect or bias the predictor or outcome. Before applying panel data regression models estimation, we provide a test exploring which panel regression method is more effective. A Hausman test was executed to compare random- and fixed-effects estimations of the model. This test rejected the appropriateness of using random-effects estimation (test value, 100.02 with p > χ² = .0000). Before conducting panel data regression analysis, we also checked the skewness and kurtosis of the variables by the Shapiro–Wilk test. The results reveal that some of our variables were not normally distributed. We adopted Blom’s (1958) instrument to transform the data into normal scores. However, the results from the analysis of the transformed and untransformed data were similar; therefore, all the results presented are based on nontransformed data. Table 2 thus provides the regression results of our fixed effects models with levels of significance reported for two-tailed tests.

Table 1.

Descriptive Statistics and Correlation Matrix.

		M	SD	1	2	3	4	5	6	7	8	9	10	11
1	Firm size (FS)	8.21	1.83
2	Firm age (FA)	23.69	18.87	0.57
3	R&D (RD)	0.05	0.12	0.63	0.54
4	Patent stock (PS)	4.42	1.90	0.57	0.38	0.47
5	Financial leverage (FL)	0.45	0.28	0.36	0.22	0.24	0.10
6	Firm growth (FG)	−0.07	0.23	−0.02	0.05	0.00	0.11	−0.31
7	Overconfidence (OC)	0.27	0.44	0.19	0.17	0.13	0.05	0.05	−0.05
8	Independent board (IB)	0.58	0.33	0.31	0.28	0.20	0.21	0.06	0.03	0.23
9	Analyst following (AF)	0.00	1.00	0.17	0.03	0.21	0.26	0.02	−0.04	0.15	0.14
10	Dedicated institutional ownership (DIO)	0.04	0.06	0.05	0.01	−0.01	0.05	0.03	0.06	0.00	0.00	−0.02
11	Transient institutional ownership (TIO)	0.19	0.12	−0.22	−0.18	−0.21	−0.20	−0.01	−0.12	−0.04	−0.03	−0.07	0.19
12	Ambidextrous imbalance	1.18	0.73	−0.16	−0.13	−0.09	−0.36	−0.03	−0.08	0.19	-0.12	0.01	−0.02	0.07

Note. The bold values represent statistically significant (p < .05). The largest and average of the variance inflation factor were separately 2.508 and 1.443.

Table 2.

Fixed-Effect Estimates of Imbalanced Ambidextrous Innovation (Option-Based OC).

		Model 1	Model 2	Model 3	Model 4	Model 5	Model 6
Variable
FS		0.066*	0.062*	0.055	0.061	0.059*	0.068*
		(0.030)	(0.030)	(0.029)	(0.030)	(0.030)	(0.030)
FA		0.003	0.005	0.007	0.001	0.004	0.007
		(0.007)	(0.007)	(0.007)	(0.007)	(0.008)	(0.007)
RD		0.605	0.648	1.029^†	0.621	0.595	0.720
		(0.585)	(0.580)	(0.556)	(0.581)	(0.581)	(0.578)
PS		−0.379***	−0.382***	−0.346***	−0.381***	−0.379***	−0.382***
		(0.040)	(0.039)	(0.038)	(0.039)	(0.039)	(0.039)
FL		−0.005	0.003	−0.024	0.017	0.002	−0.017
		(0.092)	(0.091)	(0.088)	(0.091)	(0.091)	(0.090)
FG		−0.022	−0.010	−0.040	−0.006	−0.015	−0.016
		(0.081)	(0.080)	(0.078)	(0.080)	(0.081)	(0.080)
OC	(H1)		0.207***	1.067***	0.219***	0.209***	0.213***
			(0.045)	(0.113)	(0.057)	(0.045)	(0.045)
IB				0.030
				(0.078)
AF					0.083*
					(0.004)
DIO						−0.052
						(0.279)
TIO							0.122
							(0.153)
OC × IB	(H2)			−1.184***
				(0.146)
OC × AF	(H3)				−0.077*
					(0.035)
OC × DIO	(H4a)					−0.887*
						(0.460)
OC × TIO	(H4b)						1.254***
							(0.316)
Constant		2.239***	2.187	1.995***	2.241***	2.215***	2.084***
		(0.284)	(0.282)	(0.275)	(0.272)	(0.299)	(0.286)
Number of observations		1,476	1,476	1,476	1,476	1,476	1,476
Number of firms		297	297	297	297	297	297
F		15.49***	16.41***	20.73***	15.50***	13.53***	14.94***
R ²		.133	.147	.192	.168	.153	.143

Note. FS = firm size; FA = firm age; RD = research and development; PS = patent stock; FL = financial leverage; FG = firm growth; OC = overconfidence; IB = independent board; AF = analyst following; DIO = dedicated institutional ownership; TIO = transient institutional ownership; H = hypothesis. The numbers in the parentheses are the standard errors.

†

p < .1. *p < .05. **p < .01. ***p < .001.

Model 1 includes only the control variables to provide a baseline for the subsequent analyses. The results of Model 2 indicate a positive and significant relationship between CEO overconfidence and Ambidextrous imbalance (p < .001), thus confirming Hypothesis 1. Hypothesis 2 proposed that an independent board mitigates the positive effect of CEO overconfidence on an ambidextrous imbalance. As the results show in Model 3, the interaction effect of CEO overconfidence and independent board has a significant and negative effect on ambidextrous imbalance. Moreover, following the method of Aiken and West (1991), we plotted the figure to further substantiate and validate the moderating effect of an independent board. Figure 1 illustrates that higher board independence mitigates the positive effect of CEO overconfidence on an ambidextrous imbalance, further lending support to Hypothesis 2.

Figure 1.

The moderating effect of an independent board.

Model 4 reveals the moderating effect of analyst following. The results in Model 4 show that analyst following has a significant moderating effect on the relationship between CEO overconfidence and ambidextrous imbalance (p < .05). By comparing the moderating effects of high and low analyst following, we found that high levels of analyst following have lower slopes in Figure 2. Hypothesis 3 is thus supported.

Figure 2.

The moderating effect of analyst following.

Models 5 and 6 show the results of the moderating effects of dedicated and transient institutional ownership, respectively. As can be seen by the results presented in Model 5, the interaction term of CEO overconfidence and dedicated institutional ownership is found to have a significant and negative effect on ambidextrous imbalance (p < .05). Figure 3 likewise shows that higher levels of dedicated institutional ownership have lower slopes. In Model 6, transient institutional ownership was found to have a significant and positive moderating effect (p < .001). This finding can also be seen in Figure 4. Hypotheses 4a and 4b are thus both supported.

Figure 3.

The moderating effect of dedicated institutional ownership.

Figure 4.

The moderating effect of transient institutional ownership.

To confirm the stability and robustness of the above results, four more robustness tests were conducted. First, following prior research, CEO overconfidence was further measured based on the information collected from news media. This proxy provides direct insight into a CEO as classified as overconfident based on outside information (Malmendier & Tate, 2008). The CEO was categorized as overconfident when the number of articles related to the overconfident category in regard to a CEO exceeded the number of articles related to the nonoverconfident category 1 year before the measure of ambidextrous imbalance.² The dummy variable of media-based CEO overconfidence equaled one if a CEO was classified as overconfident, and zero otherwise. News articles were collected from FACTIVA news database. This study replicated the aforementioned analyses in Models 2 to 6 using media-based CEO overconfidence as the independent variable. As the results show in Table 3, Hypotheses 1, 2, 4a, and 4b were further confirmed. The findings of these hypotheses are thus robust. On the other hand, Model 9 shows that the interaction terms of analyst following insignificantly influence the positive relationship between media-based CEO overconfidence and ambidextrous imbalance, and thus, Hypothesis 3 is not supported. This result reflects the controversial role of stock analysts in corporate governance as suggested in the studies of Bower and Gilbert (2007) and Zhang and Gimeno (2010).

Table 3.

Fixed-Effect Estimates of Imbalanced Ambidextrous Innovation (Media-Based OC).

		Model 7	Model 8	Model 9	Model 10	Model 11
Variable
FS		0.062*	0.062*	0.055	0.058	0.065*
		(0.030)	(0.030)	(0.030)	(0.030)	(0.030)
FA		0.007	0.008	0.003	0.005	0.010
		(0.007)	(0.007)	(0.007)	(0.008)	(0.007)
RD		0.625	0.542	0.574	0.654	0.515
		(0.584)	(0.583)	(0.586)	(0.584)	(0.578)
PS		−0.365***	−0.356***	−0.366***	−0.362	−0.359***
		(0.040)	(0.040)	(0.040)	(0.040)	(0.040)
FL		0.011	0.012	0.024	0.011	0.018
		(0.092)	(0.091)	(0.092)	(0.092)	(0.090)
FG		−0.004	−0.009	0.004	−0.007	0.012
		(0.081)	(0.081)	(0.081)	(0.081)	(0.080)
OC	(H1)	0.113**	0.259***	0.115**	0.118**	0.095*
		(0.041)	(0.072)	(0.041)	(0.041)	(0.040)
IB			0.067
			(0.093)
AF				0.067
				(0.041)
DIO					−0.102
					(0.287)
TIO						0.039
						(0.154)
OC × IB	(H2)		−0.256*
			(0.103)
OC × AF	(H3)			0.043
				(0.039)
OC × DIO	(H4a)				−0.684***
					(0.437)
OC × TIO	(H4b)					1.471***
						(0.271)
Constant		2.085***	1.994***	2.181***	2.139***	1.960***
		(0.289)	(0.290)	(0.281)	(0.305)	(0.291)
Number of observations		1,476	1,476	1,476	1,476	1,476
Number of firms		297	297	297	297	297
F		14.59***	12.40***	13.46***	11.98***	14.95***
R ²		.130	.134	.152	.138	.142

p < .05. **p < .01. ***p < .001.

Second, following the approach of Campbell et al. (2011) that the 100% cutoff should identify the set of CEOs who are even more optimistic, we also extended the 67% cutoff to identify overconfidence by the 100% cutoff, and redid the regression analysis. We found that the results are similar and confirmed the relationship between CEO overconfidence and ambidextrous imbalance on the 67% cutoff. Third, in addition to considering a different cutoff points (i.e., dummy variable), we also adopted Schrand and Zechman’s (2012) method of using a continuous variable of option-based CEO overconfidence (i.e., untransformed data) to check the robustness of our results. The results remain similar. Finally, prior research has suggested that imbalanced innovation activities lead to a poorer performance (e.g., Uotila et al., 2009) because exploiting existing innovation tends to generate short-term profits at the expense of long-term performance that could be achieved through the exploration of new competencies (Leonard-Barton, 1992). On the other hand, investing in exploratory innovation activities is uncertain, and the rewards tend to be limited in the short run (Ancona et al., 2001; Uotila et al., 2009). This study therefore further investigated the performance implication of an ambidextrous imbalance to strengthen our argument. We did this by testing whether imbalanced innovation activities that resulted from an overconfident CEO led to a worse performance (Uotila et al., 2009). By regressing a comprehensive performance index (i.e., averaged ROA_{t+1 to t+3}) on ambidextrous imbalance, we confirmed that imbalanced innovation behavior of an overconfident CEO leads to poorer performance (p < .05), as suggested by the line of research on ambidextrous imbalances. These robustness tests are robust and consistent.

Discussion and Conclusion

This study investigated how overconfident CEOs affect the behavior of a firm’s ambidextrous balance within innovation activities. In addition, we further examined whether the mechanisms of corporate governance moderates the above relationship. Our results support the notion that overconfident CEOs are inclined to follow ambidextrously imbalanced innovation strategies, tending toward either exploratory or exploitative innovation activities. This finding is consistent with that of March (1991), who suggested that corporations often deviate from an ambidextrous balance. While our findings are also consistent with those of Galasso and Simcore (2011), Hirshleifer et al. (2012), and Tang et al. (2015), who stated that overconfident CEOs are more likely to pursue innovation, we further determined the effect of CEO overconfidence on the deviation of a firm’s ambidextrous balance of innovation by disentangling innovation strategies into exploratory and exploitative activities. Finally, we found that firms with an overconfident CEO who leads the firm toward an innovation imbalance experience lower firm performance. This result complements upper echelons theory and implies that a CEO’s psychological attribute of overconfidence has an important effect on not only the volume but also the content of innovation activities, and thus firm performance.

Our study further supports the notion that an independent board mitigates the tendency of overconfident CEOs to lead a firm toward an ambidextrous imbalance. This is consistent with the results of Hillman and Dalziel (2003) and Peng (2004) that an independent board is effective in execution of its monitoring and service roles. Specifically, our results imply that overconfident CEOs’ behavior and decisions can be adequately monitored by independent boards, thereby reducing the negative effects of CEO overconfidence, leading the firm toward an ambidextrous balance in innovation strategies.

Our results are also consistent with the findings of O’Brien (2007) and Tihanyi et al. (2003), who suggested that dedicated investors have a greater incentive to execute their monitoring role. We found that transient investors, on the other hand, are unwilling to exercise their ownership power to monitor an overconfident CEO. This is consistent with the results of Bushee (1998) and Cox, Brammer, and Millington (2004), who stated that the monitoring role executed by transient investors is lax, enabling CEOs to be discretionary in their innovation strategies. Our results thus suggest that institutional investors execute their monitoring role only when they have incentive to actively participate in a firm’s strategic decision-making process (Porter, 1992; Tihanyi et al., 2003).

Though the results regarding the effectiveness of analysts’ monitoring role are not as strong as expected, they are consistent with the results of Bower and Gilbert (2007) and Zhang and Gimeno (2010). This may be because analysts are often more attentive to a firm’s short-term profit generated by existing technology, thus inducing a firm to neglect its long-term development. This would thereby lead a firm toward an ambidextrous imbalance (Benner, 2010). In fact, more and more theorists have raised doubts regarding the effectiveness of the governance mechanism of stock analysts (e.g., Benner, 2010; Zhang & Gimeno, 2010), and have pointed out that firms sacrifice some long-term benefits to achieve the short-term earnings expectations made by stock analysts (McKinsey & Co., 2006).

As with all studies, ours faces certain limitations. First, this study focused on the ambidextrous imbalance of innovation. Future research could study how CEO overconfidence affects overall organizational ambidexterity. Second, our study only investigated the moderating role of the governance mechanism. Future research could investigate whether other factors such as within-industry competition or corporate efficiency could moderate the effect of an overconfident CEO on innovation ambidexterity. Third, our study revealed overall that an independent board and dedicated investors play an important role in mitigating CEO overconfidence. Future research may investigate through which channel such external governance mechanisms reduce the effect of CEO overconfidence. Fourth, we measured CEO overconfidence based on quantitative measures. Such indirect measures have potential weaknesses when it comes to determining the direct impact of overconfidence. Future research could revisit this issue by applying qualitative analysis methods or a questionnaire to measure CEO overconfidence. Finally, overconfident CEOs who favor centralization are not only inclined to exclude other organizational members from making innovative choices but also to be less responsive to corrective feedback when the suggestions do not support their intuitive judgments (Chen et al., 2015; Tetlock, 2000). However, subordinates are important for leaders to generate effective problem solutions (Shipman & Mumford, 2011). Future research can revisit this issue by using qualitative research to further investigate how the leadership of an overconfident CEO affects innovative behaviors of the subordinate.

Footnotes

Acknowledgements

The authors are grateful to acknowledge financial support from the Ministry of Science and Technology of Taiwan (MOST 103-2410-H-151-022).

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: The study was support by the Ministry of Science and Technology of Taiwan (MOST 103-2410-H-151-022).

Notes

Author Biographies

Ying-Jiuan Wong is an assistant professor at the Department of Business Administration, National Kaohsiung University of Applied Sciences. She received her PhD from the Department of Business Administration at National Cheng Kung University in Taiwan. Her main research interests are innovation, family business, corporate governance, and corporate finance. Her work has been published in the British Journal of Management, Corporate Governance: An International Review, and so on.

Cheng-Yu Lee is an associate professor at the Department of Industrial Management and Information, Southern Taiwan University of Science and Technology, Taiwan. He received his doctorate in Business Administration from National Cheng Kung University. Currently an educator and a researcher in technology and innovation management field, he is interested in innovation strategy. His publications have appeared in the Asia Pacific Journal of Management, European Management Journal, IEEE Transactions on Engineering Management, Journal of Engineering and Technology Management, Technovation, and other journals.

Shao-Chi Chang is a professor in the Institute of International Business, National Cheng Kung University, Taiwan. He received his doctorate from the State University of New York at Buffalo. He has published papers in Journal of Empirical Finance, Journal of Banking & Finance, Financial Management, Journal of Business, Finance and Accounting, British Journal of Management, Journal of World Business, Corporate Governance: An International Review, and some other journals.

References

Abarbanell

J. S.

Bushee

B. J.

Raedy

J. S.

(2003). Institutional investor preferences and price pressure: The case of corporate spin-offs. Journal of Business, 76, 233-261.

Aboody

Lev

(2000). Information asymmetry, R&D, and insider gains. Journal of Finance, 55, 2747-2766.

Aguilera

R. V.

(2005). Corporate governance and director accountability: An institutional comparative perspective. British Journal of Management, 16, 39-53.

Ahmed

A. S.

Duellman

(2013). Managerial overconfidence and accounting conservatism. Journal of Accounting Research, 51, 1-30.

Aiken

L. S.

West

S. G.

(1991). Multiple regression: Testing and interpreting interaction. London: Sage.

Allison

P. D.

(2001). Missing data (Sage University Papers Series on Quantitative Applications in the Social Sciences). Thousand Oaks, CA: Sage.

Allred

B. B.

Park

W. G.

(2007). The influence of patent protection on firm innovation investment in manufacturing industries. Journal of International Management, 13, 91-109.

Ancona

D. G.

Goodman

P. S.

Lawrence

B. S.

Tushman

M. L.

(2001). Time: A new research lens. Academy of Management Review, 26, 645-663.

Argote

(1999). Organizations learning: Creating, retaining, and transferring knowledge. Boston, MA: Kluwer Academic.

10.

Artz

K. W.

Norman

P. M.

Hatfield

D. E.

Cardinal

L. B.

(2010). A longitudinal study of the impact of R&D, patents, and product innovation on firm performance. Journal of Product Innovation Management, 27, 725-740.

11.

Bass

B. M.

(1990). Bass and Stogdill’s handbook of leadership: A survey of theory and research. New York, NY: Free Press.

12.

Benner

M. J.

(2010). Securities analysts and incumbent response to radical technological change: Evidence from digital photography and Internet telephony. Organization Science, 21, 42-62.

13.

Benner

M. J.

Tushman

(2002). Process management and technological innovation: A longitudinal study of the photography and paint industries. Administrative Science Quarterly, 47, 676-706.

14.

Bierly

P. E.

Damanpour

Santoro

M. D.

(2009). The application of external knowledge: Organizational conditions for exploration and exploitation. Journal of Management Studies, 46, 481-509.

15.

Birkinshaw

Gibson

(2004). Building ambidexterity into an organization. MIT Sloan Management Review, 45, 46-55.

16.

Blair

M. M.

(1995). Rethinking assumptions behind corporate governance. Challenge, 38, 12-17.

17.

Blom

(1958). Statistical estimates and transformed beta variables. New York: Wiley.

18.

Bollaert

Petit

(2010). Beyond the dark side of executive psychology: Current research and new directions. European Management Journal, 28, 362-376.

19.

Bower

J. L.

Gilbert

C. G.

(2007). How managers’ everyday decisions create—or destroy—your company’s strategy. Harvard Business Review, 85, 72-79.

20.

Brickley

J. A.

Coles

J. A.

Terry

R. L.

(1994). Outside directors and the adoption of poison pills. Journal of Financial Economics, 32, 371-390.

21.

Bushee

B. J.

(1998). The influence of institutional investors on myopic R&D investment behavior. Journal of Accounting Research, 73, 305-333.

22.

Bushee

B. J.

(2001). Do institutional investors prefer near-term earnings over long-run value? Contemporary Accounting Research, 18, 207-246.

23.

Bushee

B. J.

(2004). Identifying and attracting the “right” investors: Evidence on the behavior of institutional investors. Journal of Accounting Research, 38, 171-202.

24.

Bushee

B. J.

Noe

C. F.

(2000). Corporate disclosure practices, institutional investors, and stock return volatility. Journal of Accounting Research, 38, 171-202.

25.

Campbell

T. C.

Gallmeyer

Johnson

S. A.

Rutherford

Stanley

B. W.

(2011). CEO optimism and forced turnover. Journal of Financial Economics, 101, 695-712.

26.

Cao

Gedajlovic

E. R.

Zhang

(2009). Unpacking organizational ambidexterity: Dimensions, contingencies, and synergistic effects. Organization Science, 20, 781-796.

27.

Cao

Simsek

Zhang

(2010). Modeling the joint impact of the CEO and the TMT on organizational ambidexterity. Journal of Management Studies, 47, 1272-1296.

28.

Chatterjee

Hambrick

D. C.

(2007). It’s all about me: Narcissistic chief executive officers and their effects on company strategy and performance. Administrative Science Quarterly, 52, 351-386.

29.

Chen

Crossland

Luo

(2015). Making the same mistake all over again: CEO overconfidence and corporate resistance to corrective feedback. Strategic Management Journal, 36, 1513-1535.

30.

Chung

K. H.

(1996). The impact of security analysts’ monitoring and marketing functions on the market value of firms. Journal of Financial and Quantitative Analysis, 31, 493-512.

31.

Cohen

W. M.

Levinthal

D. A.

(1990). Absorptive capacity: A new perspective on learning and innovation. Administrative Science Quarterly, 35, 128-152.

32.

Comment

Jarrell

(1991). The relative signaling power of Dutch-auction and fixed-price self-tender offers and open-market share repurchases. Journal of Finance, 46, 1243-1271.

33.

Connelly

B. L.

Tihanyi

Certo

S. T.

Hitt

M.A.

(2010). Marching to the beat of different drummers: The influence of institutional owners on competitive actions. Academy of Management Journal, 53, 723-742.

34.

Conyon

M. J.

Peck

S. I.

(1998). Board control, remuneration committees, and top management compensation. Academy of Management Journal, 41, 146-157.

35.

Cox

Brammer

S. B.

Millington

(2004). An empirical examination of institutional investor preferences for corporate social performance. Journal of Business Ethics, 52, 27-43.

36.

David

Hitt

M. A.

Gimeno

(2001). The influence of activism by institutional investors on R&D. Academy Management Journal, 44, 144-157.

37.

David

Kochhar

Levitas

(1998). The effect of institutional investors on the level and mix of CEO compensation. Academy of Management Journal, 41, 200-208.

38.

D’Mello

Ferris

S. P.

(2000). The information effects of analyst activity at the announcement of new equity issues. Financial Management, 29, 78-95.

39.

Dutta

D. K.

Malhotra

Zhu

P. C.

(2016). Internationalization process, impact of slack resources, and role of the CEO: The duality of structure and agency in evolution of cross-border acquisition decisions. Journal of World Business, 51, 212-225.

40.

Engelen

Neumann

(2015). “Of course I can”: The effect of CEO overconfidence on entrepreneurially oriented firms. Entrepreneurship Theory and Practice, 39, 1137-1160.

41.

Eranova

Prashantham

(2016). Decision making and paradox: Why study China? European Management Journal, 34, 193-201.

42.

Fama

E. F.

Jensen

M. C.

(1983). Separation of ownership and control. Journal of Law & Economics, 26, 301-325.

43.

Fleming

Sorenson

(2001). Technology as a complex adaptive system: Evidence from patent data. Research Policy, 30, 1019-1039.

44.

Frankel

(2004). Characteristics of a firm’s information environment and the information asymmetry between insiders and outsiders. Journal of Accounting and Economics, 37, 229-259.

45.

Galasso

Simcore

T. S.

(2011). CEO overconfidence and innovation. Management Science, 57, 1469-1484.

46.

Gedajlovic

Cao

Zhang

(2012). Corporate shareholdings and organizational ambidexterity in high-tech SMEs: Evidence from a transitional economy. Journal of Business Venturing, 27, 652-665.

47.

Goel

A. M.

Thakor

A. V.

(2008). Overconfidence, CEO selection, and corporate governance. Journal of Finance, 6, 2737-2784.

48.

Griliches

(1990). Patent statistics as economic indicators: A survey. Journal of Economic Literature, 28, 1661-1797.

49.

Gupta

A. K.

Smith

K. G.

Shalley

C. E.

(2006). The interplay between exploration and exploitation. Academy of Management Journal, 49, 693-706.

50.

Hall

B. J.

Murphy

K. J.

(2002). Stock options for undiversified executives. Journal of Accounting and Economics, 33, 3-42.

51.

Hambrick

D. C.

(2007). Upper echelons theory: An update. Academy of Management Review, 32, 334-343.

52.

Hambrick

D. C.

Cho

T. S.

Chen

(1996). The influence of top management team heterogeneity on firms’ competitive moves. Administrative Science Quarterly, 41, 659-684.

53.

Hambrick

D. C.

Fukutomi

G. D. S.

(1991). The seasons of a CEO’s tenure. Academy of Management Review, 16, 719-742.

54.

Hambrick

D. C.

Mason

P. A.

(1984). Upper echelons: The organization as a reflection of its top managers. Academy of Management Review, 9, 193-206.

55.

Wong

. (2004). Exploration vs. exploitation: An empirical test of the ambidexterity hypothesis. Organization Science, 15, 481-495.

56.

Hiller

N. J.

Hambrick

(2005). Conceptualizing executive hubris: The role of (hyper-)core self-evaluations in strategic decision-making. Strategic Management Journal, 26, 297-319.

57.

Hillman

A. J.

Dalziel

(2003). Boards of directors and firm performance: Integrating agency and resource dependence perspectives. Academy of Management Review, 28, 383-396.

58.

Hirshleifer

Low

Teoh

S. H.

(2012). Are overconfident CEOs better innovators? Journal of Finance, 4, 1457-1498.

59.

Jacobs

M. T.

(1991). Short-term America: The causes and cures of our business myopia. Cambridge, MA: Harvard Business School Press.

60.

Jensen

M. C.

(1986). Agency costs and free cash flow, corporate finance and takeovers. American Economic Review, 76, 323-329.

61.

Jensen

M. C.

Meckling

W. H.

(1976). Theory of the firm: managerial behavior, agency costs and ownership structure. Journal of Financial Economics, 3, 305-360.

62.

Johnson

R. A.

Hoskisson

R. E.

Hitt

M. A.

(1993). Board of director involvement in restructuring: The effects of board versus managerial controls and characteristics. Strategic Management Journal, 14, 33-50.

63.

Katila

Ahuja

(2002). Something old, something new: A longitudinal study of search behavior and new product introduction. Academy of Management Journal, 45, 1183-1194.

64.

Kim

J. B.

Wang

Zhang

(2016). CEO overconfidence and stock price crash risk. Contemporary Accounting Research, 33, 1720-1749. doi:10.1111/1911-3846.12217

65.

Klayman

Soll

J. B.

González-Vallejo

Barlas

(1999). Overconfidence: It depends on how, what, and whom you ask. Organizational Behavior and Human Decision Processes, 79, 216-247.

66.

Kochhar

David

(1996). Institutional investors and firm innovation: A test of competing hypotheses. Strategic Management Journal, 17, 73-84.

67.

Kolasinski

A. C.

(2013). Can strong boards and trading their own firm’s stock help CEOSs make better decisions? Evidence from acquisitions by overconfident CEOs. Journal of Financial and Quantitative Analysis, 48, 1173-1206.

68.

Lee

C. Y.

Huang

Y. C.

(2012). Knowledge stock, ambidextrous learning, and firm performance: Evidence from technologically intensive industries. Management Decision, 50, 1096-1116.

69.

Lee

C. Y.

H. L.

Pao

H. W.

(2014). How does R&D intensity influence firm explorativeness? Evidence of R&D active firms in four advanced countries. Technovation, 34, 582-593.

70.

Leonard-Barton

(1992). Core capabilities and core rigidities: A paradox in managing new product development. Strategic Management Journal, 13, 111-125.

71.

Levinthal

D. A.

March

J. G

. (1993). The myopia of learning. Strategic Management Journal, 14, 95-112.

72.

Tang

(2010). CEO hubris and firm risk taking in China: The moderating role of managerial discretion. Academy of Management Journal, 53, 45-68.

73.

Lubatkin

M. H.

Simsek

Ling

Veiga

J. F.

(2006). Ambidexterity and performance in small-to medium-sized firms: The pivotal role of top management team behavioral integration. Journal of Management, 32, 646-672.

74.

Malmendier

Tate

(2005). CEO optimism and corporate investment. Journal of Finance, 60, 2661-2700.

75.

Malmendier

Tate

(2008). Who makes acquisitions? CEO overconfidence and the market’s reaction. Journal of Financial Economics, 89, 20-43.

76.

March

J. G.

(1991). Exploration and exploitation in organizational learning. Organization Science, 2, 71-87.

77.

McGrath

R. G.

(2001). Exploratory learning, innovative capacity, and managerial oversight. Academy of Management Journal, 44, 118-131.

78.

McKinsey & Co. (2006). Weighing the pros and cons of earnings guidance. New York, NY: Author.

79.

Miller

Kets

Vries

M. F. R.

Toulouse

J. M.

(1982). Top executive locus of control and its relationship to strategy making, structure, and environment. Academy of Management Journal, 25, 237-253.

80.

Moore

D. A.

Healy

P. J.

(2008). The trouble with overconfidence. Psychological Review, 115, 502-517.

81.

Nielsen

Huse

(2010). Women directors’ contribution to board decision-making and strategic involvement: The role of equality perception. European Management Review, 7, 16-29.

82.

O’Brien

(2007). Caution regarding rules of thumb for variance inflation factors. Quality & Quantity, 41, 673-690.

83.

O’Reilly

C. A.

Harreld

J. B.

Tushman

M. L.

(2009). Organizational ambidexterity: IBM and emerging business opportunities. California Management Review, 51, 75-99.

84.

O’Reilly

C. A.

Tushman

M. L.

(2013). Organizational ambidexterity: Past, present, and future. Academy of Management Perspectives, 27, 324-338.

85.

Owen

. (2012). The hubris syndrome: Bush, Blair and the intoxication of power. London, England: Politico.

86.

Padilla

Hogan

Kaiser

R. B.

(2007). The toxic triangle: Destructive leaders, susceptible followers, and conducive environments. Leadership Quarterly, 18, 176-194.

87.

Peng

M. W.

(2004). Outside directors and firm performance during institutional transitions. Strategic Management Journal, 25, 435-471.

88.

Phelps

C. C.

(2010). A longitudinal study of the influence of alliance network structure and composition on firm exploratory innovation. Academy of Management Journal, 53, 890-913.

89.

Picone

P. M.

Dagnino

G. B.

Minà

(2014). The origin of failure: A multidisciplinary appraisal of the hubris hypothesis and proposed research agenda. Academy of Management Perspectives, 28, 447-468.

90.

Polk

Sapienza

(2009). The stock market and corporate investment: A test of catering theory. Review of Financial Studies, 22, 187-217.

91.

Pollock

T. G.

Rindova

V. P.

Maggitti

P. G.

(2008). Market watch: Information and availability cascades among the media and investors in the US IPO market. Academy of Management Journal, 51, 335-358.

92.

Porter

M. E.

(1992). Capital choices: Changing the way America invests in industry. Boston, MA: Council on Competitiveness at Harvard Business School.

93.

Raisch

Birkinshaw

(2008). Organizational ambidexterity: Antecedents, outcomes, and moderators. Journal of Management, 34, 375-409.

94.

Richardson

(2006). Over-investment of free cash flow. Review of Accounting Studies, 11, 159-189.

95.

Rosenkopf

Nerkar

(2001). Beyond local search: Boundary-spanning, exploration, and impact in the optical disk industry. Strategic Management Journal, 22, 287-306.

96.

Rothaermel

F. T.

Alexandre

M. T.

(2009). Ambidexterity in technology sourcing: The moderating role of absorptive capacity. Organization Science, 20, 759-780.

97.

Samila

Sorenson

(2010). Venture capital as a catalyst to commercialization. Research Policy, 39, 1348-1360.

98.

Schrand

C. M.

Zechman

S. L. C.

(2012). Executive overconfidence and the slippery slope to financial misreporting. Journal of Accounting and Economics, 53, 311-329.

99.

Shipman

A. S.

Mumford

M. D.

(2011). When confidence is detrimental: Influence of overconfidence on leadership effectiveness. The Leadership Quarterly, 22, 649-665.

100.

Simon

Houghton

S. M.

(2003). The relationship between overconfidence and the introduction of risky products: Evidence from a field study. Academy of Management Journal, 46, 139-149.

101.

Solomon

(2007). Corporate governance and accountability. New York, NY: John Wiley.

102.

Tang

Yang

(2015). What I see, what I do: How executive hubris affects firm innovation. Journal of Management, 41, 1698-1723.

103.

Tetlock

P. E.

(2000). Cognitive biases and organizational correctives: Do both disease and cure depend on the ideological beholder? Administrative Science Quarterly, 45, 293-326.

104.

Thakur

S. P.

(1999). Size of investment, opportunity choice and human resources in new venture growth: Some typologies. Journal of Business Venturing, 14, 283-309.

105.

Tihanyi

Johnson

R. A.

Hoskisson

R. E.

Hill

M. A.

(2003). Institutional ownership differences and international diversification: The effects of boards of directors and technological opportunity. Academy of Management Review, 46, 195-211.

106.

Titman

Wei

K. C. J.

Xie

(2004). Capital investments and stock returns. Journal of Financial and Quantitative Analysis, 39, 677-700.

107.

Tsai

M. T.

Huang

Y. C.

(2008). Exploratory learning and new product performance: The moderating role of cognitive skills and environmental uncertainty. Journal of High Technology Management Research, 19, 83-93.

108.

Tushman

M. L.

O’Reilly

(1996). Ambidextrous organizations: Managing evolutionary and revolutionary change. California Management Review, 38, 8-30.

109.

Uotila

Maula

Keil

Zahra

(2009). Exploration, exploitation, and financial performance: Analysis of S&P 500 corporations. Strategic Management Journal, 30, 221-231.

110.

Wakefield

(2003). The development of creative thinking and critical reflection: Lessons from everyday problem solving. In Runco

M. A.

(Ed.), Critical creative processes (pp. 253-272). Cresskill, NJ: Hampton.

111.

Wales

W. J.

Pankaj

C. P.

Lumpkin

G. T.

(2013). In pursuit of greatness: CEO narcissism, entrepreneurial orientation, and firm performance variance. Management Studies, 50, 1041-1069.

112.

Wang

C. L.

Rafiq

(2014). Ambidextrous organizational culture, contextual ambidexterity and new product innovation: A comparative study of UK and Chinese high-tech firms. British Journal of Management, 25, 58-76.

113.

Wright

Kroll

Elenkov

(2002). Acquisition returns, increase in firm size, and chief executive officer compensation: The moderating role of monitoring. Academy of Management Journal, 45, 599-608.

114.

Zahra

S. A.

Neubaum

D. O.

Huse

(2000). Entrepreneurship in medium-size companies: Exploring the effects of ownership and governance system. Journal of Management, 26, 947-976.

115.

Zahra

S. A.

Pearce

J. A

. (1989). Boards of directors and corporate financial performance: A review and integrative model. Journal of Management, 15, 291-334.

116.

Zhang

Gimeno

(2010). Earnings pressure and competitive behavior: Evidence from the U.S. electricity industry. Academy of Management Journal, 53, 743-768.