Too busy to balance? A longitudinal analysis of board of director busyness and firms’ ambidextrous orientation

Abstract

Studies commonly highlight the informational upside of a board of directors’ connections to its external environment. Through their seats on multiple outside boards, directors are positioned to bring valuable informational resources to complex internal tasks on a focal firm. Crafting an ambidextrous strategic orientation is such a task, requiring great informational resourcing from a board to reconcile contradictions of exploration and exploitation. Yet, we assign an important boundary condition to this expectation by unpacking the idea of “busyness” as an important consideration in a board’s (in)ability to apply their informational resources. We complement Resource Dependence Theory with insights from bounded rationality and bounded reliability, to challenge the “more is better” assumption of the benefits of outside board seats. We develop corresponding hypotheses on the extent to which busyness of different director types (exemplified here via the busyness of non-executives, executives, and women directors) is related to the ambidextrous strategic orientation of a firm. Our results from a robust longitudinal panel analysis of publicly listed UK firms uncover complex patterns and provide evidence that boards with busy non-executives have a negative influence on the ambidextrous strategic orientation of firms, whereas boards with busy executive directors do not seem to exert an influence. We further find that boards with busy women directors show an inverted U-shaped relation with ambidextrous strategic orientation. We discuss implications for theory and practice.

1. Introduction

Ambidextrously oriented firms thrive by exploiting existing strengths, while concurrently exploring new opportunities for firm growth, renewal, and survival (Junni et al., 2013; Papachroni et al., 2016; Fourné et al., 2019). However, reconciling exploration and exploitation is cognitively demanding on those entrusted to craft a firm’s strategy (Boyd, 1990; Smith, 2014), given a need to reconcile complex contradictions in activities with different time-horizons, payoff structures, and resourcing demands (Fourné et al., 2019; Matthews et al., 2022). As the highest authority in the firm (Forbes and Milliken, 1999; Rindova, 1999; Stiles, 2001), the board of directors plays an important role by contributing external information accrued from serving on other boards in the corporate network to address the complex problems of a focal firm (Davis and Greve, 1997; Drees and Heugens, 2013). Given that the board functions as a key informational interface between the organization and its external environment (Georgakakis et al., 2022; Huynh et al., 2022; Van Doorn et al., 2022), a well-connected board is expected to be a prized asset for a firm (Cook and Wang, 2011; Zona et al., 2018; Li, 2021). Yet, while studies on the board-strategy link have shown that boards influence exploratory (e.g., Tuggle et al., 2010a; Diestre et al., 2015; Li, 2019) or exploitative strategic orientations (e.g., Hoskisson et al., 1994; Yawson, 2006), few have specifically considered how boards shape an ambidextrous strategic orientation of a firm (i.e., combining exploration and exploration). In particular, is a well-connected board always beneficial for an ambidextrous strategic orientation?

In this study, we bridge Resource Dependence Theory (RDT), bounded rationality/reliability (Van Ees et al., 2009; Georgakakis et al., 2023), and ambidexterity literatures to examine the limits of the board of directors’ external connections on the ambidextrous strategic orientation of a firm (Randhawa et al., 2021). We argue that while informational resources provided by the board are often gathered through directors sitting on additional outside boards, it may come at a hidden cost—busyness. By engaging with the notion of “busy boards” (i.e., boards with a given proportion of directors being active on multiple boards; Ferris et al., 2003; Falato et al., 2014; Hauser, 2018), we challenge a long-standing assumption in RDT that more connections to firms in the corporate network, by virtue of board seats, are always beneficial (Davis and Greve, 1997; Li, 2021). We particularly organize our theoretical rationale along the bounded rationality and bounded reliability implications of busy boards for a focal firm’s strategy. Accordingly, we examine the plausibility that well-connected boards may not always deliver on their purported benefits, as the external workload of directors may affect their internal task performance, notably when board tasks are cognitively demanding (e.g., crafting ambidextrous strategic orientation).

With our approach, we respond to a call by Wilden et al. (2018) for more research on the board of directors as an underemphasized driver of firms’ ambidextrous orientation (AO). Notably, we draw attention to outside board seats, as a distinctive way in which boards access informational inputs that can be beneficial to strategy-making (Boyd et al., 2010; Ma et al., 2021). We theorize that while being well-connected is assumed to be beneficial, director busyness exposes the firm to heightened bounded rationality (e.g., cognitive overload) and bounded reliability (e.g., time constraints) stemming from the highest authority in the firm (Van Ees et al., 2009; Kano and Verbeke, 2015; Georgakakis et al., 2023). Theoretically, we thus highlight that busyness can be a crucial boundary condition for the information processing capacity of boards (Forbes and Milliken, 1999; Rindova, 1999), particularly for directors prized for their access to external informational resources.

We empirically test corresponding hypotheses on a longitudinal sample of UK firms from 2010 to 2015 and make several theoretical contributions, which we expand upon in the discussion. Notably, we highlight implications and contributions to (i) research on the board of directors—strategy link, focusing on a particularly cognitively taxing aspect (crafting ambidextrous strategic orientation that reconciles the contradictory demands of exploration and exploitation); (ii) director busyness as a boundary condition to a board’s information processing capacity, through which we challenge the “more is better” assumption about the upsides of well-connected boards; (iii) how proportion of busyness of different director types affects the board as a collective (illustrated here by distinguishing between non-executive directors, executive directors, and women directors); (iv) potential conditions heightening bounded rationality and bounded reliability; (v) board composition and how busyness affects underrepresented directors types (e.g., women directors); and (vi) practical and policy implications. Overall, our study and approach provide new insights into why seemingly “high quality” boards may (under)deliver in crafting an ambidextrous strategic orientation.

2. Conceptual background

A firm’s strategic orientation refers to its overall long-term direction and “defines the broad outlines for the firm’s strategy while leaving the details of strategy content and strategy implementation to be completed” (Slater et al., 2006: 1224). The board of directors is uniquely responsible for shaping the overall strategic orientation of the firm (Stiles, 2001), keeping in mind the firm’s ability to sustain or regenerate competitive advantage by ratifying associated resource allocation, structural, and systems choices (Schmidt and Brauer, 2006). March (1991) influentially suggested that organizations face tensions between exploration and exploitation in their strategic orientation: while exploration is premised on search, flexibility, and risk-taking, exploitation is characterized by efficiency, routines, and incremental change (Wilden et al., 2018). Ambidexterity has been conceptualized as a way of managing such tensions (Andriopoulos and Lewis, 2009; Knight and Paroutis, 2017; Matthews et al., 2022) through integrative strategies and providing relevant resources, as well as creating structural arrangements and deploying organizational systems that enable a simultaneous pursuit of exploration and exploitation (Dixon et al., 2007; Fourné et al., 2019). Accordingly, understanding the extent to which exploration and exploitation are jointly considered at the highest level of strategy—as shaped by the board of directors—is crucial.

As noted, some studies on the board-strategy link usually show how boards influence either exploratory (e.g., Tuggle et al., 2010a; Diestre et al., 2015; Li, 2019) or exploitative (e.g., Hoskisson et al., 1994; Yawson, 2006) strategic direction of a focal firm. For instance, boards have been linked to exploitation-type actions and outcomes such as technical efficiency (Bozec and Dia, 2007), downsizing (Yawson, 2006), and acquisition of related businesses (Hoskisson et al., 2004). In turn, related to exploratory-type actions and outcomes, boards have been shown to influence new product market entry (Diestre et al., 2015), entrepreneurial focus (Tuggle et al., 2010a), and innovation strategy (Kor, 2006; Makkonen, 2022). While insightful, to the extent that firms pursuing strategies that embody both exploration and exploitation tend to thrive (Junni et al., 2013), studies that only focus on either exploration or exploration provide only a one-sided view on how boards matter for the overall wellbeing of the corporation.

Perhaps surprisingly, while the competitive edge goes to firms that are able to simultaneously explore and exploit, the question of how boards contribute to ambidexterity is fairly underexplored. In fact, only a handful of studies have sought to tackle this specific question of how boards are related to ambidexterity (as opposed to either exploration or exploration outcomes). Notably, applying a simulation study, Walrave et al. (2011: 1727) developed an iterative process model to explain how boards influence managers to break away from exploitation. They found the board to be a mechanism for counterbalancing firm’s preferences to exploit. In terms of structure, Heyden et al. (2015) conducted a cross-national analysis between UK and German pharmaceutical companies and showed that the one-tier board model (i.e., adopted by companies in the UK) better enabled non-executive directors to influence the relative focus on exploration-exploitation of the firm. Finally, investigating board composition in terms of knowledge, Oehmichen et al. (2017a) documented a U-shape relationship between the board and AO, suggesting that the latter was only fostered under conditions of very high knowledge heterogeneity among non-executive directors.

The dearth of studies on the board-ambidexterity link could be indicative that AO remains a particularly complex issue for boards, thus warranting further examination. Indeed, while the allied activities, timelines, and organizational arrangements for either exploration or exploitation may be internally consistent when kept separately (e.g., innovation decisions), in conjunction, they can seem contradictory if not absurd (e.g., innovating while lowering costs). Tackling the complexities of simultaneous reconciliation of contradictory elements can thus be expected to pose great informational demands from decision-makers. Indeed, studies have more broadly suggested that fostering ambidexterity requires greater information processing from decision makers (Ou et al., 2018), because of the need to creatively integrate contradictory activities in the firm’s strategy (e.g., long- and short-term considerations), sensibly allocate resources, design complex structural arrangements, and redefine support systems (Fourné et al., 2019).

Based on the above, we can expect that crafting an ambidextrous strategic orientation requires substantial informational resourcing from a board. We draw attention to a distinctive way in which boards are expected to bring informational resources to fuel AO, notably through their access to unique insights by virtue of their memberships on other corporate boards (Davis, 1996; Drees and Heugens, 2013; Li, 2021). We build on this premise to understand how boards can contribute to ambidexterity, by first theorizing how directors’ external connections (via their outside board seats) can boost the board’s information processing capacity—which we have argued is an important prerequisite for reconciling exploration and exploitation in strategic orientation—and then assigning some key boundary conditions.

2.1 Boards as information processing groups

We have aligned with a conceptualization of boards as an information processing group entrusted with a firm’s strategic orientation (Forbes and Milliken, 1999; Rindova, 1999), where information processing entails the exchange and integration of information in decision-making. RDT has long highlighted the purported upsides of having well-connected directors, by virtue of their seats on the board of other companies (Boyd, 1990). Cook and Wang (2011) argue that multiple directorships often signal the exceptional ability of the director, which (Ferris et al., 2003) found to be positively correlated with firm performance (Li, 2019). Well-connected directors may be more confident in dealing with uncertainty and complexity (Fattobene and Caiffa, 2016; Brahma et al., 2023), as their exposure to insights from the broader corporate network improves their informational role (Biddle, 1979), through their familiarity with best practices accrued from other boards. By learning from other companies, directors can recognize problems faster, provide unique insights, and enhance performance in important corporate decisions (see also Harris and Shimizu, 2004; Field et al., 2013). The external labor market acknowledges directors’ relational skills, and the number of external directorships has long served as a proxy for the director’s reputation (Masulis and Mobbs, 2011), which in turn may also help the firm in recruiting or resource access (Westphal and Milton, 2000; Field et al., 2013). Together, these insights may allude to a “more is better” thinking, encouraging the appointment of directors with multiple board seats elsewhere to benefit a focal firm. Thus, due to information gained from sitting on outside boards that can be leveraged to craft superior strategies for a focal firm, well-connected directors are expected to be a key asset for the firm (Johnson et al., 2013).

Despite these purported benefits, we add a cautionary tale to this notion, as serving on several boards may come at a hidden cost: busyness. Boards require high information processing capacity to fulfill their roles (Rindova, 1999), where we have argued that crafting a strategic orientation that simultaneously consolidates both exploration and exploitation elements is one such particularly complex task. Research has shown that the collective information processing capacity of groups underpins novel idea generation and creative problem-solving (Dahlin et al., 2005; Chae and Choi, 2019). By exchanging and integrating informational resources obtained from a director’s outside board memberships during strategy-making (Heyden et al., 2015), directors can help shape strategic orientations (Pugliese et al., 2009). Yet, busy directors may not always be able to commit informational resources to realize the benefits for a focal firm.

2.2 Bounded rationality and bounded reliability of busy directors

To understand a board’s enactment of this information provision role, we build on the idea that while “outside commitments provide important learning and networking opportunities, they also contribute to cognitive overload and limit the time that directors spend assessing strategy and risk” (Kress, 2018: 878; emphases added). As such, accounting for director busyness, often considered as directors with three or more outside seats (Fich and Shivdasani, 2006), is important to understand their actual contributions to a focal firm. Theoretically, this premise is consistent with notions of bounded rationality and bounded reliability (Van Ees et al., 2009; Georgakakis et al., 2023). We build on this premise to question this “more is better” assumption on directors’ outside seats, as it may also expose the focal board’s information processing capacity to increased bounded rationality and bounded reliability (Forbes and Milliken, 1999; Rindova, 1999). This is an important angle to our work here, as while RDT assumes that the information access of boards (by virtue of directors’ outside board memberships) translates to benefits in strategic decisions in a focal firm, we add to the literature that busyness may restrict the information processing capacity of the board by virtue of exposing the firm to heightened bounded rationality and bounded reliability. We briefly revisit notions of bounded rationality and bounded reliability here as they are particularly relevant in the context of complex strategic orientations.

2.2.1 Bounded rationality

Addressing complex strategic issues requires great cognitive effort (Greve, 2003; Van Ees et al., 2009). Cognitive load affects the working memory of actors, which may affect their ability to comprehensively engage in complex decision-making (Laureiro-Martinez et al., 2019; Maghzi et al., 2023). When dealing with complex strategic orientations, boards may “cope with uncertainty by complexity reduction and by routinely simplifying and structuring information through their perceptual filters and pre-existent knowledge structures” (Van Ees et al., 2009: 313). In essence, bounded rationality delimits the board from understanding and considering the full scope of alternatives available (and possible unusual combinations of exploration and exploitation) (Simon, 1979; Puranam et al., 2015). Given boards’ limited ability to process all information, they may rely on cognitive shortcuts and may be more susceptible to bias, which tend to result in suboptimal choices (e.g., Foss and Weber, 2016). Given limited information processing capacity to expend on reconciling tensions and solving integrative complexities, boards tend to engage in limited search for new creative solutions and focus on more familiar options (Walrave et al., 2011) or “good enough” (Van Ees et al., 2009).

By being cognitively overloaded, busy directors may further reduce the amount of critical and new information elaborated upon during task discussions. In addition, pressing issues (e.g., short-term market pressure) will command the highest demand on busy director attention (Demirag, 1998). Prioritizing pressing issues lowers the likelihood that subsequent issues, especially those that are more long-term in nature, will receive commensurate attention to reconcile them (Tuggle et al., 2010a). Busy directors may particularly struggle to switch between complex tasks when cognitively overloaded (Leroy, 2009), and thus may tend to focus on discrete issues in a sequential rather than holistic fashion (Tempelaar and Rosenkranz, 2019). Accordingly, cognitive overload due to busyness can be interpreted as a manifestation of heightened bounded rationality that may restrict directors’ ability to simultaneously consider exploratory and exploitative issues and their integration. Absence of critical discussions, tensions, and assumptions that require reconciliation are less likely to emerge, become salient, and become a focal topic in board discussions (Tuggle et al., 2010a). This cognitive overload may also translate into directors being less inclined to engage critically in discussions or engage in perspective-taking (Hoever et al., 2012), as doing so may expose their lack of depth of understanding of specific issues. Thus, busyness may hamper information processing of the board by reducing critical insights that need to be reconciled in strategy-making.

2.2.2 Bounded reliability

A second reason why busyness affects directors’ ability to fully commit their informational resources obtained from the corporate network to a focal board is bounded reliability. Bounded reliability is the idea that actors may imperfectly make good on open-ended commitments (Kano and Verbeke, 2015). Notably, while boards are entrusted with the long-term wellbeing of the firm, busy directors may be less consistent in executing their roles; as they are faced with time constraints, they may shirk and prioritize near-term challenges that arise on the boards they serve leaving little time and energy to prepare for and engage in detail in discussions about strategic orientations. This premise is consistent with findings from the attention-based view, which suggests that attention is a finite resource (Laureiro-Martínez et al., 2015) and that directors fluctuate in the attention devoted to their board tasks (Tuggle et al., 2010b).

Importantly, these busy directors are particularly concerned about their reputation (Dewally and Peck, 2010) and may shy away from complex or possibly controversial decisions, or may succumb to pressure to focus on near-term earnings related actions (Hoitash and Mkrtchyan, 2022). Furthermore, given time constraints, busy directors tend to be more absent and underprepared for board meetings (Harris and Shimizu, 2004; Whitler and Puto, 2020), and are thus less able to exchange their unique knowledge with the board. Evidence suggests that board attendance is positively related to performance, as it enables directors to better develop common understandings to contribute to collective tasks (Chou et al., 2013). The social psychology literature shows that when group members are underprepared for tasks, people focus discussions on elaborating and recirculating commonly shared perspectives and interpretations instead of introducing new insights or solutions (van Ginkel and van Knippenberg, 2008), which may be reflected in the higher opinion conformity of the group (Oehmichen et al., 2017a). As such, one of the constraints to information processing stems from the bounded reliability of busy directors; due to lower attendance and lack of preparedness, busyness may limit information exchanged and novel perspectives introduced that could aid in crafting an ambidextrous strategic orientation.

The aforementioned discussion theoretically highlights how director busyness introduces some constraints to the board’s information processing capacity, by accentuating issues of bounded rationality and bounded reliability. By demarcating the board’s information processing capacity, the presence of busy directors may impede the group’s ability to reconcile tensions and find creative solutions to craft an AO. To exemplify this idea, we draw attention to three types of directors that we can expect to intuitively bring different information to shape ambidextrous strategic orientations, setting the stage for further examinations of other attributes of boards. We proceed to exemplify this by hypothesizing the influence of three notable director types, which we have suggested previously may be in particularly high demand: Non-executive directors (H1), executive directors (H2), and women directors (H3).

3. Hypotheses

The representation of different director types on a board is expected to accentuate different informational inputs available to inform an AO. We highlight three director types to exemplify how busyness of different director types may affect a board’s information processing capacity. First, the push for board independence has created higher demand for non-executive directors. Young (2000), for instance, finds evidence of sharp increase in the number of non-executive directorships as a result of increased demand for independent directors following the Cadbury (1992) report in the UK (see also Upadhyay & Triana 2021). Mirroring this dynamic, given the demand for capable non-executive directors in the labor market, executive directors of a focal firm may have a career incentive to take on outside directorships. Guest (2008) reports a steady increase in the number of outsiders on boards and a decline in the number of insiders and overall board size, which has made it less likely to find directors on their “home” boards, rather incentivizing them to seek outside board seats (see also Knyazeva et al., 2013). Finally, regulatory pushes for equity and representation in corporate leadership ranks (e.g., quotas and targets) have created increased demand for directors from underrepresented demographic groups, most notably women directors (Terjesen et al., 2015; Gould et al., 2018; Cook et al., 2019). This non-exhaustive distinction of director types provides us with an intuitive starting point for examining the extent to which the busyness of different directors can be expected to affect the ambidextrous strategic orientation of a firm.

3.1 Busy non-executive directors and AO

Non-executive directors, or “outsiders,” are expected to present the firm with critical and more “objective” points of view about the environment. By being independent from the firm’s managers, their vested interests, and commitment to previous courses of action, non-executive directors are expected to counterbalance decisions that are biased toward the familiar (Masulis and Mobbs, 2014). Heyden et al. (2015: 156) suggest that these directors “inform firm strategy with insights about opportunities and threats residing in blind spots (e.g., changing consumer preferences), assist in identifying weak signals in the environment (e.g., emerging technologies), act as early-warning system for imminent changes (e.g., regulatory), and provide assessments and judgments of best practices (e.g., new ways of working).” Cavaco et al. (2017) show that non-executive directors tend to have high individual ability; however, they also suffer from informational deficit about firm-specificities and thus may oversimplify problems or fail to link their insights to the focal firms’ ambidexterity-related challenges. Thus, we argue that busy non-executives may be less able to critically reflect on the extent to which their knowledge is specifically applicable to the focal firm, to vet outside information, and to integrate it to address the opposing demands of exploration and exploitation. Such directors will struggle to find the complementarities among exploration and exploitation that can be leveraged, given the idiosyncratic capabilities of a focal firm.

This is amplified as busy non-executive directors are more frequently absent (Chou et al., 2013), accentuating their bounded reliability. Given this higher absence rate they tend to be underprepared for idiosyncratic firm-specific issues (Harris and Shimizu, 2004), and are thus less able to exchange their unique knowledge with others, ultimately struggling to identify the roots of tensions or to dissect the integrative complexities inherent in an AO (King and Zeithaml, 2001). Lack of firm-specific knowledge reduces the ability to provide and integrate information gathered outside the organization and to challenge and debate other directors’ proposals. As such, busy non-executive directors may struggle in enabling the board to economize on bounded rationality. Further, board attendance is positively related to company performance, as it enables directors to better develop common understandings to contribute to collective tasks (Chou et al., 2013). When group members are underprepared for tasks, individuals focus discussions on elaborating and recirculating commonly shared perspectives and interpretations instead of introducing new insights or solutions (van Ginkel and van Knippenberg, 2008). As such, due to lower attendance and lack of preparedness, busyness of non-executive directors may limit the boards’ knowledge exchange and integration of novel perspectives.

Non-executive directors are expected to provide more extensive external connections and are thus expected to bring unique knowledge from their experience in other firms and relations to other directors. Non-executive directors may be skillful in managing relations to a variety of stakeholders and have the potential to serve as information brokers, yet the busier a director, the more relationships require nurturing and attending to reciprocal demands (and thus less time can be devoted to individual boards). Busy non-executive directors may focus on satisficing solutions to maintain their relationships but are unlikely to go above and beyond expectations of stakeholders in the focal firm. Drained from managing their network, busy non-executive directors may optimize cognitive effort, and information processing is more likely to focus on satisfying solutions and are thus unable to help the board economize on bounded rationality. Essentially, the benefits of busy non-executive directors’ information brokerage may not be realized in the context of boards’ decision-making regarding AOs.

Finally, when making decisions, non-executive directors often try to maintain their reputation (Dewally and Peck, 2010), typically reflected in higher opinion conformity (Oehmichen et al., 2017a) as agreeableness is often (mis)interpreted as a sign of ability in groups (Park et al., 2011). Thus, busyness may further decrease non-executive directors’ information provision to and processing by reducing critical insights that need to be considered when developing strategic orientations. That is, busy non-executive directors may fulfill their informational role less; they rather advocate “proven” solutions as well as focus on meeting short-term targets that signal their immediate added-value instead of developing novel firm-specific solutions when engaging in task discussions. As such, busy non-executive directors are prone to diffuse more homogenous interpretations of the external environments by emphasizing practices that are in fashion, by prioritizing near-term issues, or by endorsing generic templates of best practices observed from a few exemplar firms (Abrahamson and Fairchild, 1999; Nicolai et al., 2010). This discussion indicates that busy non-executive directors are likely limited in their information provision and processing related to facilitating and supporting decisions regarding an ambidextrous strategic orientation.

Thus, we argue that busyness of the board driven by a higher proportion of busy non-executive directors reduces the benefits that boards can extract from non-executive directors. In sum, we argue that:  

3.2 Busy executive directors and AO

Executive directors have a deeper and more contextualized understanding of the capabilities of their organization, which are often enmeshed in socially complex and path-dependent relationships that are not visible to outsider observers (Mom et al., 2015; Van Doorn et al., 2022). By serving on multiple outside boards while having a rather stable reference point (i.e., their own firm) through which to interpret new knowledge, insiders can develop ideas on how to the focal organization’s own capabilities can be refined, while acquiring information on new opportunities that can be explored and may be suitable for integration in their firm’s strategic orientation (Rothaermel and Alexandre, 2009). By combining their firm-specific insights with their embeddedness in the external environment through outside appointments, executive directors can provide their home firms with actionable information, especially on feasible avenues for growth (Geletkanycz and Boyd, 2011) and for exploration-oriented initiatives and partnerships (Ni Sullivan and Tang, 2013). As such, busy executive directors may contribute positively to the information processing capacity of the board by virtue of higher knowledge sharing and higher knowledge integration, which is aimed at reconciling short- and long-term demands of both exploitation and exploration.

Busy executive directors increase the board’s ability to creatively tackle tensions and to find ways to harness (some of the) complementary benefits among exploration and exploitation, and to communicate them convincingly to external and internal stakeholders or “insiders.” As busy executive directors combine their board role with duties that primarily revolve around the day-to-day organizational demands of strategy execution in their focal firm, they play an important boundary spanning role: busy executive directors can support the integration of knowledge and purpose fit communication of information across the focal firm to clarify how a simultaneous pursuit of exploration and exploitation may be feasible and why it is a strategic imperative. As their primary affiliation is with the focal firm, executive directors are more reliable and more likely to attend meetings and be better informed about agenda items, and thus ready to engage in-depth in decision-making regarding strategic orientations (Chou et al., 2013) and limiting potential issues associated with bounded reliability. Executive directors have the confidence to speak up and debate critically, as they possess the experience, knowledge, and external legitimacy to gain internal promotions (Connelly et al., 2014; Georgakakis et al., 2022). By being able to critically debate with all other board members and by having greater insights into the internal workings of the firm, information provision to and processing of a board that comprises busy executive directors may facilitate an objective evaluation of how to integrate exploration and exploitation in the focal firms’ strategic orientation.

Furthermore, the labor market incentivizes directors to perform well within the “home” company, as poor performance may cease access to additional directorships and could harm the director’s reputation and career progress (Levit and Malenko, 2016). As such, busy executive directors play an important decisional role in their focal firm as they use their home company to develop deeper understandings of core markets and the capabilities needed to exploit in these settings, while gaining insights regarding new opportunities in their environment from their outside directorships. This can incentivize input into how the focal firm can regenerate competitive advantages by balancing exploration and exploitation. It is in the interest of the busy executive director to push the board to economize on bounded rationality, to avoid satisfying decisions or focusing on the status quo, but rather to strive for a more balanced strategic orientation that showcases their firm’s ability to combine exploitation for near-term, easy to measure success with exploratory initiatives for sustained competitiveness and technological leadership. All else being equal, we can expect that:  

3.3 Busy women directors and AO

Given their historical underrepresentation, women directors are particularly expected to bring different insights, as they are often new to the corporate elites and may be particularly valuable in bringing fresh and unconventional ideas to bear on strategy, thus reducing incumbent cognitive biases (Singh et al., 2008; Miller and Del Carmen Triana, 2009). They also often contribute richer, tacit knowledge to the board (Gould et al., 2018). This is because the experience of women directors is likely to be with more precarious positions (Ryan and Haslam, 2007), such as poorly performing and smaller firms (Westphal and Milton, 2000), as well as often reaching senior corporate positions on the back of experience in non-traditional contexts, such as (resource scarce) non-profits (Bear et al., 2010). These career paths can offer unique learning opportunities, as the unique knowledge accrued in these contexts may equip women directors with fresh perspectives and creative solutions not typically considered by the incumbent directors (Westphal and Milton, 2000; Carpenter and Westphal, 2001). Accordingly, women directors’ contributions to making complex decisions may be particularly valuable, especially early in strategic decision processes (Müller-Horn et al., 2024). In sum, insights from women directors can particularly be beneficial for creative solutions geared at reconciling exploration and exploitation tensions and for considering how to combine strategic initiatives with different time horizons.

In terms of how busyness affects bounded reliability of women directors, women directors are not only more likely to attend meetings, but also more likely to invest time to attend well-prepared (Adams and Ferreira, 2012; Baghdadi et al., 2023), allowing for more comprehensiveness in the evaluation of complex strategic orientations that affect a variety of stakeholders (Baghdadi et al., 2023). This is consistent with the expectation of women in organizations to go the “extra mile” (Heilman and Chen, 2005), even when it levies a personal toll on them (Bolino and Turnley, 2005). Given their diligence in preparation and attendance, paired with knowledge from unconventional domains, busy women directors will tend to particularly enrich information processing of the board. They do so through the exchange of richer knowledge and through directing attention to both short- and long-term projects. This enriched information processing can result in decreasing problems usually associated with bounded rationality when making complex decisions. Furthermore, well-prepared busy women attendees of board meetings may improve the board’s decision-making process, for instance, by proposing a balanced set of decision-criteria that enable a fair evaluation of both near- and long-term prospects of a strategic orientation. Importantly, given that women directors often apply a longer time horizon and have been linked to greater innovation (Griffin et al., 2021), the likelihood of excessive prioritization of near-term issues (such as earnings concerns and short-term budgets) may be reduced. As such, busy women directors may ensure that sufficient attention of the board will be devoted considering the merits of combining exploitation and exploration initiatives in an AO, and that those merits will not be discounted strongly because of a multiple or overall longer time-horizons (to yield results).

Despite the benefits of busy women director participation in the board, it is likely that such busyness may become too great and come at a cost. To the extent that busy women directors are expected to bring new perspectives from different contexts, oversaturation of viable ideas introduced to strategy making could be a by-product that poses additional cognitive challenges for boards. Integrating knowledge from multiple and diverse external contexts (that busy women directors have access to) increases coordination costs that may strain other board members’ capacity to exchange and integrate knowledge (Singh, 2008)—and may to some extent cause “paralysis by analysis,” exacerbate bounded rationality problems, and could be followed by a regression to the industry mean risk taking levels or even lower to focus on exploitation. Thus, the richness gained from busy women directors’ exposure to diverse set of organizational contexts (given the high demand for them across industries) may make it harder to fully integrate their unique input to board discussions. To ensure some of their valuable insights are included, women directors may have to establish workable understandings by simplifying their inputs to accommodate and persuade others of the validity of their arguments to incumbent members (van Ginkel and van Knippenberg, 2008; Gino et al., 2009). This takes effort and time and likely reduces the range of very busy women directors’ perspectives for reconciling tensions between exploration and exploitation.

In sum, we argue that busy women directors will contribute positively to the board instilling an AO in the firm, but only up to a point:  

4. Data and methods

4.1 Sample and data

We test our hypotheses on a longitudinal sample of UK publicly listed companies from the FTSE 350 index from 2010 to 2015. The choice of FTSE 350 companies is justified by their relatively rich endowment of resources and capabilities for exploration and exploitation activities than smaller, more resource constrained firms (Cao et al., 2009). Further, studies on board influences on strategy have recently shown that the effect of boards on exploration and exploitation can be expected to be most pronounced in one-tier board models, such as those of companies in UK (Heyden et al., 2015). The list of FTSE 350 companies was retrieved from the Bloomberg database for the 31st of December of each year during the period of 2010–2015. The choice of this study period is intended at mitigating the effects of the 2008–2009 global financial crisis, as well as a period where codes of governance have called for increased board independence and personal accountability of directors, an empirical setting that is thus meaningfully aligned with our theoretical assertions of why boards are becoming busier (Financial_Reporting_Council, 2012). In addition, this time period and setting also capture increasing demand for women directors where the recommendation for increased representation of women on boards was passed in 2010, with first compliance audit date in 2015 (Vinnicombe et al., 2015). In this period and setting we can expect to observe sufficient variation in busyness of women directors. The total sample comprised 3172 individual directors.

Financial and market data were collected from Thomson Reuters Datastream, whereas corporate governance and directors’ data are from BoardEx. To account for survivorship bias, all companies for which information was available from these sources were included in our sample. After dropping observations for which complete data were not available, we were left with 252 firms in our final sample, yielding 1162 firm-year observations for the 6-year period from 2010 through 2015. Table A1 reports sample distribution by industry (Panel A) and by year (Panel B).

4.2 Variable operationalizations

4.2.1 Independent variables

We use several measures of board busyness. We first constructed “overall” board busyness (Busy board) by computing a proportion of busy directors following Fich and Shivdasani (2006). We code a director as busy if their total workload consists of appointments on boards of three or more listed companies in the observation year (Ferris et al., 2003), a measure which has been shown to be parsimonious and robust (Cashman et al., 2012). Our measure for H1 is based on the proportion of busy non-executive directors (Busy NED), those directors on the boards with no formal executive, staff, or operational mandates in the focal firm. Our measure for H2 is based on the proportion of busy executive directors (Busy ED), those directors serving on the board who also have insider positions (e.g., C-level and [senior]vice presidents). Following the same logic, our measure for H3 accounts for the proportion of busy female directors on the board, based on binary sex classification indicated in BoardEx (i.e., male or female). Table A2 provides an overview of how the different busy director types over the years of our sample and across the industries are represented.

4.2.2 Dependent variable

A computer-aided text analysis (CATA) approach is used to measure exploratory and exploitative markers representing the firm’s strategic orientation, which are subsequently combined to create a measure of AO (see Matthews et al., 2022). CATA “is a form of content analysis that enables the measurement of constructs by processing text into quantitative data based on the frequency of words” (McKenny et al., 2018: 2910). It assumes that frequency of theoretically meaningful markers in a corpus of text can capture underlying themes and patterns (Duriau et al., 2007). By combining the strengths of human judgment and computer reliability, CATA allows us to capture exploration and exploitation orientation reliably between firms and over time (Uotila et al., 2009; Heyden et al., 2015; Matthews et al., 2022), for rigorous quantitative testing and in different empirical settings (see also Belderbos et al., 2017; Gaur and Kumar, 2017). Gaur and Kumar (2017) noted that “a good example of auto coding using CATA software” can be found in Heyden et al., (2015), which was also recently applied in Oehmichen et al. (2017b). Following this approach, we used NVivo and took the validated dictionaries of Uotila et al. (2009) and Heyden et al. (2015), as our starting point, while making some key improvements to received search dictionaries following recent best practice recommendations by Belderbos et al. (2017). Further in line with these studies, our corresponding corpus of text was based on annual reports. Although the pros and cons of relying on corporate communications have been debated (Crawford 2003), their validated utility for capturing exploration and exploitation orientation have been repeatedly demonstrated (Matthews et al., 2022).

4.2.3 Search dictionary

We made several modifications to previously adopted dictionaries on exploration and exploitation. Belderbos et al. (2017) describe two approaches that are generally used in selecting keywords for CATA: deductive and inductive. The deductive approach starts from theoretical definitions and uses critical keywords to reflect the concept in question while the inductive approach uses the body of text under analysis to develop keywords. The keywords in the published dictionaries of previous studies (Uotila et al., 2009; Heyden et al., 2015) have covered the deductive approach comprehensively and the inductive approach to some extent.

Heyden et al. (2015), built on the original keywords of Uotila et al. (2009), developed further 66 and 75 words reflecting exploration and exploitation, respectively. As Heyden et al. (2015) tailored their dictionary for the pharmaceutical industry, we generalized and improved their keywords (see also Matthews et al., 2022, for a recent update). We adopted and restructured in line with the underlying thematic notions of exploration and exploitation as presented by March (1991). The thematic notions recognized for exploration are discovery, experimentation, risk-taking, flexibility, innovation, and variation, while those recognized for exploitation are efficiency, refinement, implementation, production/operation, and selection. In the spirit of March (1991), we further extended the wordlists by adding 25 general keywords for exploration and 59 for exploitation. At this stage, we performed a face validity check on all keywords (original and new) in consultation with three international experts on exploration and exploitation and adjusted the dictionary accordingly.¹

In addition to the face validity checks in developing the dictionary, we checked the reliability of the new dictionary by examining keywords-in-context (KWIC) (Krippendorff, 2004; Belderbos et al., 2017) and validity of the overall dictionary in terms of accurately demonstrating the underlying phenomena (Belderbos et al., 2017; McKenny et al., 2018). We did so by running a preliminary text-search query for each word in the dictionary independently, and a minimum of 10 instances of the results were manually checked by two members of the research team. This phase of validation was conducted on multiple iterative stages to ascertain reliability consistent with previous studies (Krippendorff, 2004; Uotila et al., 2009; Heyden et al., 2015; Belderbos et al., 2017). Then, we applied the search dictionary to a random sample of 350 annual reports using NVivo (30% of the total annual reports under study). For each keyword, we examined a random selection of 20 instances. Keywords that resulted in fewer than 20 extractions were removed from the dictionary, as we did not consider them as capturing exploration and exploitation in sufficiently common terms to be comparable across firms We then critically evaluated the keywords extracted and coded each instance as either exploratory or exploitative, based on mutual agreement and expert judgment of two independent coders. When the majority of the 20 KWIC were considered to correspond reliably to either exploration or exploitation (≥60%; cf. Belderbos et al. (2017), whose lowest inductive threshold was 67%), we kept them in the dictionary. This provided us with the confidence that our refined search dictionary was performing reliably.

Finally, as a final check for accuracy, we ran the overall complete dictionary for each theme, which was retrieved in a text-search query for the same random sample of 350 annual reports. The retrieved outcome was closely examined to validate the accuracy of the overall dictionary in capturing the themes of exploration and exploitation. After a number of iterations, this phase resulted in no words being removed; suggesting a degree of saturation. After concluding all alterations, additions, and content validity checks, a total of 61 and 110 keywords for exploration and exploitation nodes, respectively, were included in our final search dictionary used for the CATA procedure.

A total of 1162 annual reports were retrieved and text analyzed with the NVivo software for wordlists of each node independently. NVivo generated word frequencies and coverage percentages (i.e., extracted words expressed as a percentage of the total text) for each report. The corresponding AO score is obtained by square-rooting the product of exploration and exploitation coverage percentages (which allows us to normalize the words extracted by the length of the annual report). The square-rooting is conducted to transform the functional form of AO back to a percentile scale for normality purpose. This operationalization of AO is consistent with the combined dimension in the literature (i.e., higher scores reflect higher coexistence of exploration and exploitation). We also use a conventional measure of AO and compute it as a product of exploration and exploitation coverage percentages, which we used for sensitivity tests.

4.3 Control variables

Table A3 provides an overview of the comprehensive suite of predictor variables included in this study, including controls. To control for general board effects, we use board independence as it might influence group dynamic and innovative performance (Goodstein et al., 1994). We also include board gender diversity and board age (Miller and Del Carmen Triana, 2009). As directors’ understanding of the resources and capabilities of organization is closely related to time directors spent on board, we control for this by including board tenure. To further account for the human and social capital of directors, we controlled for whether directors’ outsider directorships were involved in the same industry, as extra-industry experience connects directors to different networks can be expected to help directors bring more different perspectives. Similarly, we also accounted for board nationality, as foreign directors may instill cultural variety and different perspectives in board discussions (Estélyi and Nisar, 2016). In addition, we account for board functional background by distinguishing between background in service roles, production roles, and support roles, as well for educational qualifications of the board (Richard et al., 2019).

Given the importance of the CEO, we also account for CEO age to consider risk propensity and learning dispositions inherent in the life stage of CEOs (Hambrick and Fukutomi, 1991; Henderson et al., 2006) as well as CEO tenure, which have been shown to influence ambidexterity (Fernández-Mesa et al., 2013). We account for CEO nationality (CEO-British) as national culture was found important for firms’ innovative capabilities (van Everdingen and Waarts, 2003; Georgakakis et al., 2016). We include CEO–chairman duality (CEO Duality) to account for CEO power (Haynes and Hillman, 2010). At the firm level, we included firm size, firm age, performance, leverage, and R&D expenditures. Year dummies are included to control for unobserved macroeconomic influences, and Herfindahl-Hirschman Index (HHI) is used as a proxy for industry competition. The HHI is calculated as the sum of squared market shares as follow: |$HH{I_{j,t}} = \mathop \sum \limits_{i = 1}^{{N_J}} S_{i,j,t}^2$|⁠, where S_i,_j,t is the market share of firm i in industry j in year t. Market share is calculated using firm sales. We estimate industry competition for each of the industry classifications. High values of HHI indicate weaker industry competition and vice versa.

5. Analysis and results

We analyze our longitudinal data using a fixed-effects model to account for unobserved firm-specific characteristics. The fixed effects, or within estimator technique, is based on a deviation from a companies’ mean transformation (firm’s mean for the sample time period is subtracted from each observation) and estimates all coefficients without estimating individual effects. Since we are interested only in slope coefficients, this transformation is a convenient and appropriate one (Baltagi, 2008). An important issue when dealing with panel data sets is the estimation of robust standard errors to mitigate serial correlation patterns (Petersen, 2009). To account for this, we run all our fixed effects models with robust standard errors clustered at the firm level (Wooldridge, 2010).

5.1 Univariate and bivariate statistics

Average board busyness in the sample is 24% ranging from absence (zero) busy directors (for 55 cases) to a maximum of 86% of busy directors. The average number of busy female directors is 4.4%. Table A4 displays summary statistics and correlations for the variables used in this study.

5.2 Multivariate results

Table A5 presents the results of the fixed-effects models we used to analyze our panel data and test our hypotheses. We checked the variance inflation factors (VIF) against possible multicollinearity issues. The highest VIF value in our regression models was 5.84, which is well below the suggested cut-off point of 10 (Kutner et al., 2004), indicating that multicollinearity is not a concern in our analyses. Hence, we proceeded with our multivariate analyses and in Model 1 we included Busy board and Gender (diversity) to establish a baseline expectation. We find that board busyness is negative and board gender diversity has a positive association, but neither is statistically significant. Model 2 was used for testing our first two hypotheses. Using Busy NED as a proxy for the presence of busy non-executives for H1, we expected that these busy directors have a negative influence on AO. The coefficient estimates are negative and significant (P < 0.05), corroborating H1. Using Busy ED to test H2, we expected that busy executive directors positively relate with AO of the firm. The coefficient for busy executives has a positive sign but this variable is not statistically significant; as such, H2 is not supported. In the final model, we used the proportion of busy female directors and the quadratic term of this proportion to test H3. The results from these models provide support for H3, as the coefficient on linear term of Busy female variable is positive and significant (P < 0.05), while its quadratic term is negative and also statistically significant (P < 0.05). We exclude Busy NED and Busy ED from these models, as we could not distinguish whether Busy female directors were executives or non-executives due to multicollinearity issues. Female directors’ busyness thus seems to influence AO in a non-linear manner.

To further interpret the results of H3, we provide a visual representation of the non-linear association in the predictive margin plot in Figure 1 (see Mueller et al., 2021, for a recent example). This graph shows clearly the inverted-U relationship between busy female directors and organizational ambidexterity with clear indication to the turning point. Consistent with the recommendations of Haans et al. (2016), we control for the cubic term of busy female directors to ascertain that it has an inverted U-shape rather than S-shaped effect. Unreported results confirm the persistence of an inverted U-shape effect given the insignificance of the cubic term coupled with the significance of the linear and quadratic terms at 5% level. However, we note that adding in the cubic term does not improve the model fit and, thus, further support our finding of the inverted U-shape association. Based on the results of Model 3 (Table A5), the maximum turning point for busy female directors (⁠|$\frac{{{{\beta \,Busy\,female}}}}{{ - 2{{*\beta \,Busy\,female}}2{\rm{\,}}}}$|⁠) is estimated to be at 16.1%; meaning that busy female directors beyond such threshold will result in lower strategic ambidexterity.

5.3 Robustness, endogeneity, and sensitivity tests

To account for the immediate effect of board busyness, corporate governance, and firm characteristics on organizations’ AO, current period values of the independent variables are included. Although theoretically we view boards as antecedents, consistent with prior literature, to account for the possible reverse causality issues, we re-ran all models with lagged independent variables. The results from the lag-transformed models are qualitatively similar to results from models with the contemporaneous independent variables. However, corresponding coefficients in lag-transformed models are marginally low; also, R^2, which measures goodness of fit of the model, is lower than the corresponding R² from the level model, suggesting that lag-transformed models have slightly lower explanatory power.

To further rule out the issues of causality, we also examine the extent to which changes in directors’ business affect AO. The identification logic behind this is as follows: if busyness increases (decreases) from the previous years’ busyness, it can broaden (narrow) a board’s (collective) information capacity. Thus, changes in director type busyness from the previous year to the current should be reflected in corresponding patterns of AO. We test the effect of first difference in boards’, executive directors’, non-executive directors’, and female board members’ busyness on AO to evaluate how changes in director busyness translate into higher/lower AO. When we consider the change in the level of busyness, changes in overall level of board busyness and executive directors’ busyness are not significant, and do not affect AO. Change in non-executive directors’ busyness remains negative and statistically significant; if busyness of non-executive directors increases, AO declines. Change in female directors’ busyness and its quadratic term retain their signs and significance. The coefficients of female busyness difference and its quadratic term are marginally higher than the coefficient of the level of the same variables. This indicates that marginal increases in female director busyness will have a particularly pronounced non-linear association with AO.

We also ran tests to differentiate between busy executive and non-executive female directors in a supplementary analysis, as female directors are also more likely to be non-executives (e.g., given the small pool of female executives; Fairfax, 2006; Helfat et al., 2006). We ran a new model, in which we include our new variables Busy female NED and Busy female ED as independent variables. The new variables are positive but not statistically significant. Thus, separating them may mask the reality that there is an effect of busy female directors. Given the low representation of women on boards more generally, further sub-splitting this type of director may mask this effect due to a lack of statistical power, and thus produce a false-negative. This implies that although we tentatively accept this hypothesis at this stage, we recommend that this hypothesis be retested in the future as the population of women directors increases and distinguishing between their roles becomes possible. We further conducted a series of robustness and sensitivity tests to check for the consistency of our findings from the main models. We tested our models on (i) sub-sample of R&D intensive companies; (ii) sub-sample with majority of British directors; (iii) sub-sample with British CEOs; (iv) models with R&D and Busy board interaction variables; and (v) replicated the findings using a simple multiplicative measure of exploration × exploitation. These results produced qualitatively the same conclusions for the main predictor variables and all results are available from the authors upon request.

In addition, since board gender diversity and proportion of busy female directors are correlated only at r = 0.08—which means that these two aspects can be treated as independent, we checked the moderating role of gender diversity, examining whether overall board gender diversity may influence the strength or direction of the relationship between the proportion of busy female directors and AO. We included the interaction effect Busy female × Gender diversity in our analysis and re-ran the regressions. The coefficient of the Busy female × Gender diversity interaction is positive but is not statistically significant at any meaningful level. This result (also available from the authors upon request) suggests that busy female directors can independently influence AO regardless of the level of board gender diversity.

Finally, we mention that our evidence satisfies the conditions confirming an inverted U-shape, consistent with Lind and Mehlum (2010) and Haans et al. (2016). The negative coefficient of DT2 is significant, the slopes between the lowest and highest points of busy female directors (⁠|${{{\beta }}_1} + 2{{{\beta }}_2}{\rm{Busy\,femal}}{{\rm{e}}_{\rm{L}}} \gt 0$| and |${{{\beta }}_1} + 2{{{\beta }}_2}{\rm{Busy\,femal}}{{\rm{e}}_{\rm{H}}} \lt 0$|⁠) are both significant, and |$\frac{{{{\beta \,Busy\,female}}}}{{ - 2{{*\beta \,Busy\,female}}2{\rm{\,}}}}$| has a 95% confidence interval of (2.33% to 19.27%), which is well within the data range.

6. Discussion

Equipping the board with well-connected directors is expected to be a key asset for firms—both substantively in terms of information provision influencing the strategic orientation of the firm, as well as symbolically by signaling quality and legitimacy to outside stakeholders. However, today’s directors face greater expectations for involvement in and accountability for the firm’s strategy than ever before. Accordingly, even well-connected boards may underdeliver on their contributions to complex strategic tasks in a focal firm—such as crafting a strategic orientation that is characterized by both exploration and exploitation elements (i.e., an ambidextrous strategic orientation). In examining this idea, we have theoretically grounded our study in the premise that while well-connected directors are expected to increase the information processing capacity of a board, director busyness could actually expose the board to heightened bounded rationality and bounded reliability. We tested corresponding hypotheses on a sample of large UK corporations, unearthing several implications, as discussed below.

6.1 Contributions & implications

6.1.1 Board of directors and AO

Against an appreciation of the influence of directors on strategy, understanding how director’s external workload influences their contributions and the boards’ information processing emerge as an important topic for understanding why firms vary in their ambidextrous strategic orientations. Ironically, directors who are in highest demand, the most reputable, and have superior knowledge access may harm the focal firm (Falato et al., 2014). Thus, despite the intuition that the “best” directors (e.g., by virtue of being well-connected) are crucial for firm effectiveness (Boyd, 1990; Drees and Heugens, 2013), having busy directors may not always deliver expected results for firms. Overall, we question whether “more is better” when it comes to the expected contributions of well-connected corporate directors.

Our study complements macro examinations of ambidexterity at industry, firm, and inter-firm levels, as well as the increasing emphasis on individuals and teams (Venugopal et al., 2020), by drawing attention to the board of directors as a crucially underemphasized factor fostering or constraining the ambidextrous strategic orientation of an organization. While research on boards has shown how boards relate to exploratory (e.g., Tuggle et al., 2010a; Diestre et al., 2015; Li, 2019) or exploitative (e.g., Hoskisson et al., 1994; Yawson, 2006) initiatives, only recently has the issue been raised of how boards can influence ambidexterity (Oehmichen et al., 2017b; Wilden et al., 2018). While focusing on either exploration or exploitation is not without its complexities, their considerations are more internally consistent than the paradoxical complexity of finding ways of reconciling both exploration and exploitation (Fourné et al., 2019). Indeed, extant research provides a rather one-sided interpretation of the board-strategy link, omitting the extent to which boards may or may not be contributing to an ambidextrous strategic orientation—which is expected to be crucial for long-term performance (Junni et al., 2013). Our study is among the first to address this omission by pointing to challenges that different board members face in balancing the competing thrusts underlying an ambidextrous strategic orientation.

6.1.2 Busyness as boundary condition to board information processing capacity

The demand for competent and well-connected directors has increased over the last decades (Young, 2000; Guest, 2008; Rigolini and Huse, 2021)—yet, we draw attention to busyness as a potential allied hidden cost. Theoretically, we draw attention to how busyness may accentuate bounded rationality and bounded reliability to information processing. We develop theory exposing the benefits and downsides of directors’ participation in the broader corporate network and threshold conditions when external board membership can be interpreted as making directors “busy,” thus potentially affecting their contributions to focal boards’ complex tasks; exemplified here through board’s influence on shaping their firms’ ambidextrous strategic orientation. By conceptualizing director busyness in terms of its implications for the collective information processing capacity of a board (Chae and Choi, 2019), we have connected this potential decrease in information processing capacity to a board’s reduced ability to reconcile and integrate the contradictory demands of exploration and exploitation, thus negatively affecting the AO of a focal firm. Specifically, we highlight that high well-connected boards may underdeliver in relation to crafting ambidextrous strategies, as despite “on paper” having all the right skills and informational resources to inform strategy (Oehmichen et al., 2017a), their externally well-connected members may be too busy to share and integrate their knowledge, thus affecting the information processing capacity of the board needed for ambidextrous strategies.

The extent to which the external workload of directors has benefits and costs to their internal task performance remains debated (Oehmichen et al., 2017a). Our study also informs unresolved debates on the pros and cons of busy boards (Field et al., 2013; Ferris et al., 2018; Hauser, 2018). While Cook and Wang (2011) argue that multiple directorships signal an exceptional ability of the director, we highlight that some of these directors may be too busy to realize the potential benefits (Falato et al., 2014). Directors who overstretch themselves and accept additional seats tend to spend less time on each individual board may compromise their preparation, limit contributions, and may neglect (some of) their duties. We contribute to research board busyness as an important boundary condition of the effects of boards on firm strategy. So far, findings on the implications of busy boards are rather scarce and inconclusive, having mostly focused on distal outcomes such as financial performance, but less so on strategic outcomes or other intermediary steps such as strategic orientation that may link board enactment of their roles to ultimate firm performance.

6.1.3 Different implications of busyness across director types

Overall, we developed theory and assign boundary conditions to connect the proportion of different types of busy directors on a board to a firm’s ambidextrous strategic orientation. While it is plausible that board busyness may reduce the board’s collective information processing capacity, by virtue of accentuating bounded rationality and bounded reliability limitations, we add nuance to this idea by highlighting that this varies depending on the busyness of different types of directors. Our theory and results have implications for boards as information processing bodies that rely on directors’ contributions when dealing with complex decisions and strategic considerations.

By distinguishing between different director types, we find that busyness limits the contributions of non-executive directors to firms’ AOs, whereas busy executive directors seem less affected by their external and internal workloads and draw on their firm-specific experience in ways that enables positive contributions to AOs. From our theory and findings, we can speculate that dealing with the informational challenges associated with complex strategic problems (such as crafting an ambidextrous strategic orientation) requires integration of outside information with firm-specific knowledge, preparation, and motivation to participate actively—which might be more characteristic of executive directors. It would be worthwhile to measure information sharing and processing more directly in future research about board members’ contributions—to assess more precisely how busyness, preparation, firm- (or even industry-) specific knowledge influence board decision-making and to understand which kinds of contributions (in terms of information sharing, interpretation, debate, etc.) allow for mitigating bounded rationality and bounded reliability and thereby facilitate making complex decisions. It may seem ironic that the directors best equipped with information from the business environment, other firms, even competitors, may also be the directors with least time and energy to draw on these externally derived insights to push a focal firm to a more balanced approach to exploration and exploitation.

6.1.4 Heightened bounded rationality and bounded reliability of boards

Our study also has implications for board contributions to strategy in increasingly complex firms, as an appreciation of conditions and characteristics under which the bounded reliability and rationality of boards are heightened (or reduced) has important implications. For instance, as multinational enterprises (MNEs) expand their footprint in foreign markets, the demands on the board become greater, while the operational realities may become more distant. Busyness may be particularly problematic for firms facing these specific challenges. Our results suggest that the ability of the board to take into account bounded reliability and bounded rationality of different directors is important to foster an AO. This may be particularly observable in internationally dispersed organizations as distance (e.g., geographic, cultural, political) may compound the challenges boards face with bounded reliability and bounded rationality. Monitoring, coordination, and contextual understanding challenges may be compounded with distance and add to bounded reliability and bounded rationality of board members, potentially reducing their ability to foster an ambidextrous strategic orientation. As such, we recommend future research to examine the role of bounded reliability and bounded rationality in more (and less) complex and internationally dispersed organizations to understand how bounded reliability and bounded rationality manifest at the board level and are connected to these organizations’ AOs. In the spirit of your suggestion, we recommend considering situations where busy boards face additional challenges (e.g., monitoring, coordination, or contextual understanding) and how this affects their decisions regarding ambidextrous (or other) strategic orientations at the firm level and at subsidiary levels.

6.1.5 Board composition and women’s representation

Our study also adds to the debate on board composition in terms of diversity (Baghdadi et al., 2023), exemplified here via women’s representation on corporate board (Post and Byron, 2015). The general thesis regarding board facilitation of innovation is rooted in the notion that more diverse boards are better at providing informational resources that can underpin new ideas, reduce narrow-mindedness, challenge myopic tendencies, improve understanding markets, help find novel opportunities, and ensure more balanced resource allocation (see Makkonen, 2022, for a recent meta-analysis). More broadly, while women’s board participation has been found to positively contribute to innovation-related outcomes (i.e., exploration), such as R&D intensity (Miller and Del Carmen Triana, 2009) and innovative initiatives in general (Makkonen, 2022), we note that busyness of women directors may act as a particularly complex (non-linear) boundary condition for such findings (Liu et al., 2020). Accounting for such busyness is of practical relevance as the push for women on boards has not led to a commensurate increase in the actual pool of women directors and, in some instances, may have created “an unforeseen emerging elite of women who hold multiple directorships” (Seierstad, 2010), making this group of directors particularly busy. Accordingly, this sub-group of directors may face additional challenges associated with busyness.

6.1.6 Managerial implications

Composing a board with experienced and well-connected directors remains an objective for many firms. Yet, sought-after directors increasingly serve on globally dispersed boards (Oehmichen et al., 2017a), face heightened expectations, and are accountable for active involvement in strategy-making (Stiles, 2001; Roberts et al., 2005). They have to comply with increasingly complex regulatory requirements (Aguilera, 2005) and carefully manage their reputation (Dewally and Peck, 2010). Directors themselves may perpetuate this issue, as they can increase their status, income, and career opportunities by pursuing additional directorships (Perry and Peyer, 2005). The board sets priorities on behalf of shareholders and serves in informational, relational, and decisional roles to support senior executives in developing a firm’s strategic orientation (Georgakakis et al., 2022; Srivastava et al., 2023). To the extent that busyness may well be a boundary condition explaining variation in directors ability to contribute their information resources to a focal firm, theory of busy directors resonates with emerging examinations on information brokerage research and the (in)ability to realize the benefits of being an information broker (cf. Glaser et al., 2021). Our study opens the possibility for recognizing that bounded rationality and bounded reliability could vary over time and may affect in particular the more challenging board tasks. Thus, timing of board meetings is not a simple logistical exercise, but may have more enduring strategic implications than previously thought. Accordingly, scheduling of board interactions may need to account for the distribution of heightened busyness. Furthermore, from the different results for busy non-executive and busy executive directors, it may be implied that the motivation to serve a focal firm may also play a role. The question which firms do busy directors prioritize or put most effort in for deserves further investigation.

6.2 Future research and limitations

Our study is prone to several limitations, which also open up opportunities for future research. First, boards balance collaboration and control (Adams and Ferreira, 2007; Zona et al., 2018; Baghdadi et al., 2023). Although our study has been particularly informed by the collaborative approach, this is only a part of the complex mixture of informational, relational, and decisional roles that members of the board of directors may play (Zattoni et al., 2022). To the extent that actors play a role in both formulation and implementation of ambidextrous strategic orientations, our study raises the important question of how boards play their controlling role, as usually informed by agency theory (Hillman and Dalziel, 2003). While our focus is on the information processing capacity underpinning an AO (Forbes and Milliken, 1999; Rindova, 1999), a focus on the controlling or monitoring roles of board members may be insightful when investigating the implementation of such a strategic orientation; and which decisions they advise to make or will support (Boyd et al., 2010; Ma et al., 2021).

Second, there may be personal differences between those who actively seek out multiple directorships, such as differential intrapersonal abilities to handle complexity and workloads, or personality traits such as greed, narcissism, and humility, which may offer complementary insights into the willingness and ability to exchange and integrate knowledge in their resource provisioning tasks. While we have sought to establish a generalizable pattern of how proportion of different types of busy directors on a board may influence complex strategic orientations, such as those characterized by both exploration and exploitation simultaneously, we do not claim that all busy directors are the same and a natural next step is to examine variation in how different types of busy directors cope with their board membership demands. Here, an experience sampling methodology of directors could be insightful. Our study provides an important platform for spearheading research in this area and for recognizing the costs and benefits of busyness.

Third, the link between board composition and some innovation outcomes may depend on other aspects, such as CEO power [e.g., the impact of board configurations to support environmental innovation varies depending on CEO power (Shui et al., 2022) and the positive association between board capital and R&D investment is positively moderated by CEO power (Haynes and Hillman, 2010; Chen, 2014)]. Hence, we recommend for future research to develop a more comprehensive understanding of the board-CEO interface (Georgakakis et al., 2022; Van Doorn et al., 2022). The role of the CEO and the board-CEO interface are indeed a relevant vantage point for future research on board level strategic decision-making (Huynh et al., 2022). As such, CEO power may accentuate or reduce the effects of well-connected directors and gender diversity. To bridge our efforts with the latest advances in upper echelons theory, we particularly encourage the examination of concepts that may further accentuate bounded rationality and bounded reliability, such as cognitive complexity (Graf-Vlachy et al., 2020) or political ideology (Kalogeraki and Georgakakis, 2021).

Finally, we encourage more research on women directors and their influence on firm strategy and decision-making (Baghdadi et al., 2023). In our study, we focus on gender as a salient aspect of minority representation on corporate boards, but we caution against homogenizing the experience and attributes of women on boards. Women directors may also embody other minority status criteria (such as ethnicity), and thus may experience additional challenges which can be tackled through an intersectional lens. Furthermore, we note that whereas women directors particularly contribute through experience and networks accrued from less common career trajectories (Ryan and Haslam, 2007), this could also be the case for some male directors and/or those expressing different gender identities. We recommend future research to tease out unique career paths of directors that may have prepared them for coping better with busyness in terms of preparing for board meetings, leveraging their superior access to informational resources, and being motivated to contribute to the complex decisions that boards tackle. Altogether, our study provides an exciting launching pad for research on the contributions and limitations of (potentially) high caliber boards.

Acknowledgments

Mariano L. M. Heyden acknowledges support from the Australian Research Council Discovery Early Career Award (DE170100381). Part of the work was also conducted while the author was a Visiting Lazaridis Professor at Wilfrid Laurier University.

Sebastian P. L. Fourné acknowledges support from The Canadian Social Sciences & Humanities Research Council (SSHRC) Insight Grant 435-2022-0687.

Footnotes

References

Appendix