64 Connecting Governance to CEO Replacement and Organizational Recovery Meredith Downes, Ph.D. Illinois State University Manuscript Type: Conceptual/Theoretical Research Question/Issue: Why do CEOs overstay and, in some cases, long overstay their welcomes? Social Network Theory may offer some explanation for delays in dismissing CEOs that, in the eyes of shareholders, should have been long gone. Social Network Theory also suggests that delaying the dismissal of a CEO will in turn delay firm performance recovery. Social Network Theory is reviewed in the context of board monitoring and its effects on CEO dismissal and subsequent recovery. It is ultimately suggested that certain board attributes will have an indirect effect on recovery time, as mediated by CEO dismissal time. A model and propositions are laid out and potential next steps are outlined for pursuing this line of inquiry. Research Insights: Based on the literature review, it seems likely that board composition will have an indirect effect on how long it takes for organizations to recover from poor performance, after replacing an underperforming CEO. Theoretical Implications: This paper makes important contributions to the corporate governance literature. First and foremost, it extends the research agenda on board composition, CEO turnover, and performance, to include the element of time. More specifically, it suggests that certain characteristics of corporate boards are more likely to inhibit the types of governance necessary to remove underperforming CEOs, and this, in turn, will impact the time it takes for organizational performance to recover. The paper also deepens the application of Social Network Theory to the study of corporate governance, addressing several elements of social networks that are found in non- independent boards and with “overboarded” directors. Practitioner Implications: Risks of poor performance due to inadequate governance are far- reaching, and shareholders and D&O (Directors and Officers) liability insurers are particularly vulnerable. Even with new management in place, shareholders might lose faith in the board if important strategic actions were delayed and especially if they negatively impact their investments. The anticipated effects of recovery can also discourage future investment in the company. Vigilant boards, composed of independent directors with optimal bandwidth, would be more likely than dense and embedded boards to replace the CEO when it is warranted. It would, therefore, behoove shareholders to participate in the election of directors, rather than turning that privilege over to the very board that took too long to replace the CEO due to its strong network structure. Keywords: Corporate Governance; Social Network Theory; Board Composition; CEO Dismissal http://journals.sfu.ca/abr ADVANCES IN BUSINESS RESEARCH 2019, Volume 9, pages 46-59 65 Introduction Along with stock ownership concentration, executive compensation, and the market for corporate control, boards of directors can be powerful mechanisms for governing today’s corporations (Barney & Ouchi, 1986; Dyl, 1988). The literature on corporate governance emphasizes the abuse of managerial power in the absence of robust boards (Burrough & Helyar, 1990; Hwang, 1993; Lublin, 1991). We see evidence of this abuse in the lofty compensation packages afforded to CEOs even when firms are underperforming. According to management guru Peter Drucker, “Whenever an institution malfunctions as consistently as boards of directors have in nearly every major fiasco of the last 40 or 50 years, it is futile to blame men. It is the institution that malfunctions” (Drucker, 1981). Therefore, allowing CEOs to receive such large sums in compensation for mediocre or even poor performance is evidence of board ineffectuality. In fact, allowing the CEO to continue to reign over the organization may indicate board deficiency. These and other practices have certain shareholders, particularly large-block institutional investors, up in arms -- and they are not blaming management. They are blaming the boards (Pulliam & Pereira, 1995; Smith, 1996). The board of directors is charged with oversight responsibilities to ensure that management actions are in the shareholders’ best interests (Monks & Minow, 1996). It is only in the last two decades that the board’s role as an “ornament on a Christmas tree” has been vehemently denounced. Instead, they have been “granted authority” by the example set by a few boards (e.g. IBM, General Motors), to replace the current CEO. Subsequently, CEOs at Westinghouse, American Express, Eastman Kodak, Scott Paper, and Borden were all pressured to resign. However, these companies suffered years of decline before the CEOs were ousted. This begs the issue of why some firms wait too long for proof of managerial incompetence before making a move, and thus board effectiveness comes into question. For instance, it may be that, true to the “old boy’s club” accusation, there is a collegiate culture in the boardroom that fosters an unwillingness to speak out against a prevailing view. Board members who have served together for years, or those who are somehow beholden to top management, are particularly likely to favor the status quo. If the agency relationship between owners and managers explains the need for governance mechanisms such as corporate boards, what explains the fact that boards may not always be prompt in addressing the agency problem? Although charged with the task of protecting owners (and, more recently, other stakeholder groups as well), from managerial opportunism, there is no guarantee that they will fulfill this responsibility. Research has given some weight to the role that board composition plays in its ability to monitor effectively (i.e. by removing an underperforming CEO). The extant research suggests that outside-dominated, independent, diverse boards, with limited director bandwidth and no CEO duality, are ideally composed and thus the most likely boards to pursue CEO dismissal when it is warranted. Some of this research has been empirical (e.g. Triana, Miller, & Trzebiaowski, 2014; Vafeas, 2003), although most are suggestive, based on agency and social network theory (e.g. Hambrick, Misangyi, & Park, 2015; Libit & Freier, 2015a). The current paper adds to this research agenda by addressing a related yet unexamined phenomenon – that of delays in CEO dismissal – in order to capture the dynamic elements of the decision to dismiss. Another addition of this paper is in the way that subsequent performance is addressed, also dynamically, in terms of how long it takes for organizational performance to recover following CEO dismissal. Specifically, in modeling the relationship between board composition and organizational performance, the delay in CEO dismissal is incorporated as a mediator. It is suggested that certain characteristics of corporate boards are more likely to inhibit the types of governance necessary to promptly remove underperforming CEOs and that this, in turn, will impact the time it takes for organizational performance to recover. Direct and indirect effects of specific board characteristics on recovery will be addressed. Figure 1 shows the proposed relationships among the relevant constructs, each of which is developed below. 66 Figure 1. Theoretical Model and Proposed Relationships Literature Review and Proposed Model CEO dismissal can be defined as a situation in which the CEO’s departure is against his or her will and is not part of a mandatory retirement policy. Whether due to the need for a scapegoat (Gamson & Scotch, 1964; Nystrom & Starbuck, 1984), the desire to shift organizational direction, or CEO wrongdoing, dismissal represents a major organizational event typically initiated by the board of directors. The mere threat of dismissal acts as a constraint on managerial actions, with poor performing CEOs more likely to be replaced if internal governance is strong. Although poor corporate performance has been consistently associated with high rates of CEO turnover (Giambatista, Rowe, & Ruiz, 2005), many CEOs have been dismissed even when their firms were performing well. In addition, some findings indicate a direct performance-dismissal link (Giambatista et al., 2005; Haleblian & Rajagopalan, 2006), whereas others show a more modest association between actual performance and dismissal (Phan & Lee, 1995; Triana et al., 2014). This calls the causal nature of the relationship into question and suggests that the link may be mediated by other forces. Therefore, this paper explores the role of board composition on its decision to dismiss an underperforming CEO and how this, in turn, impacts future performance. Social Network Theory is used to develop the ideas for this paper and is therefore reviewed in general terms below, followed by more deliberate application of the theory in order to support the arguments regarding board composition, CEO dismissal, and subsequent recovery. Social Networking Theory Networked structures are those systems characterized in terms of nodes (individual actors, people, or things within the network) and the ties or edges (relationships or interactions) that connect them. There are three types of social networks. Ego-centric networks are connected through the ties that members have with a single node or individual, socio-centric networks are closed systems with connections across multiple nodes and where the boundaries of the network are formal and defined, and open-system networks are those where the boundaries are not clear since they cross over formal 67 structures. The structural patterning of a network depends on the combination of characteristics that it takes on, with most attention typically given to its density and its embeddedness. Highly embedded networks are composed of actors who show a preference for transacting with other actors inside rather than outside the network (Granovetter, 1985; Uzzi, 1996). Embeddedness emphasizes not only the social ties between certain actors but also the absence of ties between others (Wellman, 1999). Density represents the cohesion of the network. Dense networks are those with a high percentage of actual connections relative to the possible number of connections in the network. Networks can also be described in terms of their range (or size) and their diversity, with homophily referring to the extent to which actors form ties with similar as opposed to dissimilar actors in the network. Finally, the multiplexity of a network refers to the number of ways in which actors within the network are connected to each other. Some may only be co-workers, for example, whereas others may work together but also attend the same church. While centrality has been used to characterize networks as well, it is more individualistic in nature, referring to the importance of a particular node to the entire network. This relates to the length of the path, such as the degrees of separation between nodes, with some nodes holding considerable power as bridges to other nodes in the network. Network characteristics are somewhat elusive, in that they are very difficult to measure directly. However, boards of directors, by virtue of their compositions, enable us to capture many of these characteristics. Corporate boards most closely resemble socio-centric networks, and the group of firms and directors connected through board interlocks form open-systems networks. Table 1 allows us to see both types and will be referenced throughout the paper. Board multiplexity can be captured by the number of interlocks among directors, and embeddedness is reflected in the proportion of insiders or non-independent directors on the board. In addition to capturing aggregate board characteristics, Hambrick et al. (2015) assert that even the most ideally composed boards may not include a single director with all of the necessary qualities to be an effective monitor. This is analogous to a movie with a decent cast but which, without at least one stellar acting performance, is not likely to be nominated for an Academy Award. The Hambrick et al. (2015) model calls for at least one but preferably two directors with a requisite four attributes (independence, domain expertise, bandwidth, and motivation) in order to render the board effective. Their “quad model” also stresses the need to address these attributes conjointly, rather than as mutually exclusive characteristics. The current paper addresses many issues raised by previous studies, thoughtfully grouping board characteristics to be considered conjointly. It also makes a unique contribution to the literature, by (i) adding a dynamic element to the heretofore static conceptualizations of board monitoring by addressing delays in CEO dismissal and delays in firm recovery, and (ii) solidly grounding all of the above in theory. 68 Propositions Most of the research on CEO dismissal takes a static approach, treating dismissals as successful board events when they coincide with poor firm performance. This ignores the damage to the firm in the time leading up to a dismissal. There have in fact been cases, such as the United Way, the Jewish Center of Washington, DC, and Food for the Poor (Glazer, 1994; Kay, 1994; Sinclair, 2000), where CEO misdeeds went on for years without detection. This indicates a governance problem. Although the CEOs in each of these cases perpetuated the wrongdoing, it was the board that failed to institute accountability measures and therefore failed in its oversight responsibilities. Phan and Lee (1995) examined firm performance for the three years prior to dismissal, hinting at the dynamic nature of such an event. Haleblian and Rajagopalan (2006) also included the element of time in their study of board cognitions and the decision to dismiss, considering the board’s perception of the firm’s performance, its attributions of that performance, and its assessment of the efficacy of the CEO. This suggests that the dismissal is not a one-time event but rather a process of gathering information and making a series of assessments that may or may not lead to CEO dismissal. Therefore, the logic suggested by the extant literature is extended here to include the length of time over which organizations are declining before action is taken to replace the CEO. Since monitoring is an on- going process, it would make sense that a board that is diligent in its monitoring role would detect inadequacy when it emerges and address it in a timely manner. The literature on board of directors’ best practices tells us that certain board characteristics should facilitate such diligence and that opposite characteristics are likely to inhibit it. Several of these are discussed below, applying Social Network Theory to the explanation for why some boards are more effective monitors than others. Independence Although often treated as separate phenomena, the board characteristics discussed in this section all describe the ability or willingness of board members to be objective where management is concerned and therefore reflect director independence. Insiders/Outsiders Agency Theory questions the appropriateness of inside directors (those who are employed by the firm) due to the potential for ineffective monitoring and conflicts of interest. For a board composed primarily of insiders, directors would essentially be charged with monitoring themselves, only to be exacerbated when the CEO is also chairman of the board. With each director representing a node in the board network, those who are insiders share more than one connection with the CEO, in that they serve together on the board and also work in the same company, thereby increasing their multiplexity. They are also likely to have more connections within than across networks, since insiders spend more time at the same firm, further embedding the network. It may even be the case that insiders serve on the board out of reciprocity, loyalty, or fear, and thus may not play a monitoring role at all. With their careers potentially tied to their board service (Mobbs, 2013), insiders have little power vís-a-vís their CEOs, and this increases the CEO’s centrality within the network. Finally, according to (Haleblian & Rajagopalan, 2006), the board makes attributions about its company’s performance, in which case insiders are less likely to attribute poor performance to internal factors such as their CEO. It stands to reason, then, that with more insiders on the board, it will take more time to convince them that ousting the CEO is the appropriate course of action. Outside directors, on the other hand, and especially those without any other affiliation to the firm, will be more effective in performing a monitoring role based on their increased objectivity (Zahra & Pearce, 1989). For outsiders, the backlash for speaking up may not be as imminent and may have more of a social rather than a career impact, and thus they may be more inclined to voice their concerns about the CEO or other top managers. 69 Reciprocal Interlocks Board interlocks (also referred to as interlocking directorates) take on many of the characteristics of social networks (e.g. Burt, 1980; Mizruchi, 1996; Palmer, 1983; Palmer, Friedland, & Singh 1986), increasing the multiplexity and reciprocity within the board. This situation is present if directors serve together on multiple boards, as well as if a director on the board of Firm A is the CEO of Firm B, and the CEO of Firm A sits on the board of Firm B. This special kind of interlock speaks to the independence of board members, creating a situation of “you scratch my back, I’ll scratch yours” (see Table 1), and thus it becomes very unlikely that either would call for the other’s ouster. Interlocks embed the board, creating social circles that entrench the CEO and interlocked directors in the social environment, making it more difficult to dislodge any of the actors because social systems favor stability over change (Zucker, 1977). CEO Involvement in Selection Through their involvement in director selection, CEOs may actually be in control of whether or not they are dismissed by the board. Although shareholders vote on board membership, selection can be heavily influenced by the CEO. Without a nominating committee, for example, all board members including the CEO participate in the nomination process. If the board does have a separate nominating body, it is possible that input is sought from the CEO or even that he or she is on the committee. In a study by Shivdasani and Yermack (1999), greater CEO involvement leads to the nomination of less independent directors. This includes not only insider nominees but also “gray” outsiders who have conflicts of interests (e.g. are relatives of, or have interlocks with, the CEO). With heavy CEO involvement in selecting directors, board embeddedness is enhanced as CEOs opt for new or returning directors with whom they already have ties, and board density is strengthened through CEO centrality, thereby increasing the proportion of directors who are connected to each other. In addition to influencing board nominations, the CEO can affect the outcome of the shareholder vote itself. In situations where the CEO owns significant shares in the company, we can again expect more non-independent directors to ultimately be selected. According to the similarity- attraction theory, CEOs with influence in the selection process prefer to interact with others who have similar values and attitudes (Byrne, 1997; Condon & Crano, 1988; Montoya & Horton, 2004), which serves to reinforce their own opinions and beliefs. Similarity also reduces the uncertainty of interpersonal interaction and helps people to communicate more easily with one another (Zhu & Chen, 2015). Therefore, even if the CEO does not have ties to board nominees prior to selection, their like-mindedness is likely to facilitate these ties over time. This groupthink board mentality is an obstacle to change and thus it will take longer to convince members that any of the actors should be dismissed. Tenure While the average tenure of directors at S&P 500 companies has dropped considerably since the passing of Sarbanes-Oxley, there is still concern over lengthy board stays. This concern exists among non-U.S. firms as well, as the European Commission has recommended that EU-based companies limit director tenure to 12 years (Libit & Freier, 2015b) limits. The 2014 Spencer Stuart Board Index shows that only three percent of the S&P 500 boards have term limits for directors, ranging from 10 to 30 years, but that many companies have mandatory retirement ages for directors. There are arguments both for and against board tenure as an effective governance mechanism. Although long-standing directors are experienced and knowledgeable, there are concerns that once a director reaches a particular length in tenure, his or her independence from management may become compromised. Directors may start to think like insiders, having been exposed to the viewpoints of management, especially if serving among other long-standing directors. Seasoned directors are more likely to befriend and less likely to monitor managers. They may even be co-opted by management as they become less mobile and less employable. Senior board members are frequently engaged in “grey” 70 professions, in that they may be hired by management to render services to the company, managerial business, and directors with twenty or more years of service are almost twice as likely to be classified as grey (Vafeas, 2003). According to Daily, Johnson, and Dalton (1999), such directors are by no means independent, since independence would require no previous connection to the firm or the CEO for three years prior to serving. Lengthy board tenure serves to further embed boards as ties are formed over time, and if several board members serve together for years, the mulitplexity of ties among members is likely to increase. It becomes more difficult to remove actors from the network since dismissal is likely to cost the board more than just its company’s CEO – it may cost friendships, sever bonds, and cause factions within the board. This leads to CEO entrenchment, whereby he/she is more valuable if retained rather than dismissed (Gibelman & Gelman, 2002). There is also the potential that long-tenured board members may be “zombie directors” (Libit & Freier, 2015a), with no initiative or enthusiasm, just going through the motions of serving on a board where nothing really changes. Finally, extended tenure also tends to reduce intra-group communication and isolates groups from key information sources (Katz, 1982). Libit and Freier (2015a), provide several company cases where “board refreshment” is preferred but also state that best practices may depend on a number of factors, including the relationship that a firm has with its shareholders. Director turnover increases the opportunity for diversity as well as the chance to bring fresh ideas into the boardroom. New directors may come with updated industry expertise and, if they are outsiders and without ties to the CEO, they are more likely to embrace changes when necessary. In a recent study, Clements et al. (2018), found a positive relationship between director quality and low/intermediate board tenure, and a negative relationship between director quality and lengthy tenure, although these negative effects varied based on a number of company characteristics. With the research to date showing more preference for limited board tenure, Social Network Theory leads us to believe that non-independence is at the forefront of this concern. Therefore, even when the firm is underperforming, it is expected that boards with long-standing directors would be more likely to postpone the inevitable task of dismissing the CEO. Diversity There is reason to believe that diversity among board members impacts the speed and quality of its decisions since most attributes of social networks indicate homogeneity and thus any deviations are likely to affect the network structure. Research suggests that this is a double-edged sword. On the one hand, heterogeneous boards have better access to information (Adams & Ferreira, 2007; Bebchuk & Weisbach, 2010), which creates a culture of communication and questioning (Van Knippenberg, De Dreu, & Homan, 2004). The board becomes aware of and more sensitive to a wider range of stakeholder concerns. In addition, heterogeneity increases the likelihood that some board members will not belong to the social network. Such boards will be less dense and less embedded, with many directors free to remove underperforming CEOs without the professional or social repercussions felt by directors with stronger ties. However, when competing claims vary widely, the board may reach a stalemate in the decision-making process. In addition, since cognitions may vary as a function of demographics (Milliken & Martins, 1996; Robinson & Dechant, 1997), diverse boards are less cohesive than homogeneous ones, also stalling the process. It is difficult to separate board diversity from director tenure and insider-dominance. A mixture of new and long-standing members, for example, creates heterogeneity, as does a combination of inside and outside directors. Outside directors can offer a fresh perspective, may represent or be connected (through their firms or industries) to important stakeholder groups (Wiley, 1995), and may be more sensitive to a wider range of stakeholder concerns. Wahid (2012) found that diversity in terms of director tenure, as well as rank, enhances CEO performance-turnover sensitivity. The effects of gender diversity on CEO turnover has also been addressed in the literature, although with inconsistent findings. While Wahid (2012) found no significant impact on turnover, Triana et al. (2014) found gender diversity to be important when firm performance is poor, as the potential for conflict among competing claims is 71 high (Daily, Dalton, & Cannella, 2003; Westphal & Bednar, 2005). In times of crises, rather than amplify the variety of alternatives that can lead to a stalemate, diverse teams tend to reduce information processing by narrowing it down to what is absolutely necessary to reach a decision (Gladstein & Reilly, 1985). Based on this discussion, it is expected that diverse boards are more independent than homogenous ones, so they are less likely to succumb to management pressure (Carter, Simkins, & Simpson, 2003) and are more likely to produce devil’s advocates and engage in focused debate that leads to prompt decision-making. To summarize, independent boards are those which are diverse and outsider-dominated, with limited director tenure and limited CEO involvement in director selection, and with no reciprocal directorships. Therefore, the majority of directors on independent boards are not beholden to the CEO for their selection to the board, they do not work for the CEO, and they are more likely to question decisions rather than to protect the CEO as a member of their social network. Alternatively, non-independent boards are less likely to blame the CEO for poor performance, protecting him or her for as long as possible. It is only when the firm is in crisis mode that non-independent directors will succumb to pressure to remove the CEO. P-Independence. Independent boards will take less time to respond to organizational underperformance by dismissing the CEO. Bandwidth Research on the effects of multiple directorships (referred to as “overboarding”) has revealed two competing theories, as synthesized by Mobbs (2013). The attentiveness theory (Fich & Shivdasani, 2006; Perry & Peyer, 2005) suggests that when directors serve on three or more boards, their value as a director diminishes because they become too busy. Too many directorships can reduce the time and energy a director has to invest in any one firm where s/he serves (Jiraporn & Gleason, 2007), whether it be to monitor CEO activity or to vie for the CEO position him-/herself. Conversely, those with few directorships are more dedicated and hence devote more time to their monitoring roles. Mobbs (2013) states that it is the routine type of monitoring that is best served by directors who serve on fewer boards (Mobbs, 2013). The expertise theory, on the other hand, argues that individuals holding more outside directorships are in high demand in the market because their experience increases their human capital. Harris and Shimizu (2004) find that outside directors with multiple directorships are important sources of knowledge during acquisitions, for example, and are therefore more valuable for monitoring under special circumstances. Consistent with the attentive perspective addressed above, companies and even the stock exchanges on which they are listed, have imposed limits on the number of directorships. Seventy-four percent of S&P 500 companies limit other corporate directorships for their board members. One example is Nuevo Energy Corp., which restricts its CEO to two outside board seats. In addition, NASDAQ and the NYSE require outside directors to meet in executive session, which commits more of their time and in effect makes it more difficult to serve on multiple boards. Direct and Indirect Interlocks Interlocking directorates also indicate multiple directorships, and these open-systems networks are created as companies connect through their board memberships, as shown in Table 1. These become social circles, and the directors who belong to these networks carry a certain cachet, bringing with them the reputation of the companies where they are (or were) employed. To each board on which they sit, they bring with them the reputation of their own companies if they work for one. Also, their behavior on these boards is a reflection of the companies for which they work and thus facilitates good governance. So, in addition to the expertise that interlocks bring to each of the connected boards, it also incentivizes directors to uphold their own company’s reputation by monitoring in a timely and effective manner. 72 What we can surmise from the scenario presented results thus far is that director expertise derived from extensive bandwidth facilitates monitoring to a point, beyond which they are simply too busy to participate in the routine monitoring which is necessary for acting in a timely manner. Based on the above discussion, two directorships are considered optimal bandwidth for a board member, as attentiveness and interlocks will both be possible. Specifically, directors who have the time to be more attentive to the company will be more sensitive to changes in performance, as well as to whether any decline should be attributed to the CEO. For an otherwise busy and less attentive director, the performance of any one firm on whose board he sits may not be as consequential to him. He is, therefore, less likely to invest the time it takes to initiate or support the dismissal of the CEO. However, if the board is vigilant in its oversight and evaluation of the CEO, early identification of problems is likely. In this event, there is time to work with the CEO to remedy the performance deficits. But this is a time-consuming process that requires the attentiveness of directors who are not overboarded. In addition, Kim, Kim, and Miller (2006) suggest that the anticipated negative consequences of turnover are the very reasons that keep CEOs from getting fired. With extensive interlocks (which include those connections through the CEO), dismissal of an entrenched CEO can translate into significant losses of the relationships that he or she facilitates. Therefore, we expect timely CEO dismissal from boards with enough bandwidth to be concerned about their reputations but with not so many interlocks that the CEO is entrenched. P-Bandwidth. Boards with optimal average bandwidth among outside directors will take less time to respond to organizational underperformance by dismissing the CEO. Recovery As evidenced above, empirical research on the speed of CEO dismissals has been limited. However, there is reason to believe that speed of dismissal and speed of recovery are linked (Kim et al., 2006), although there is a lack of consistent results regarding this relationship (Finkelstein et al., 2008). This may be due to the varying ways in which post-succession consequences have been measured, which include not only the strategic changes that ensue but also the firm’s accounting and market performance (Kim, 2011). According to Kim (2011), as well as to Ertugrul and Krishnan (2011), dismissals which are too early or too late may have negative effects on firm performance, suggesting that a moderate speed to dismissal would be ideal. But recent research has found a more linear relationship between speed of dismissal and speed of recovery. According to Hazarika, Karpoff, and Nahata (2012), timely recognition of performance losses improves the board’s ability to use both accounting and market performance information effectively when evaluating CEO performance and making CEO replacement decisions, and Hu, Kim, and Lin (2015), have found that such timeliness leads to greater CEO turnover-performance sensitivity. Similarly, in a study by Ertugrul and Krishnan (2011), firms that acted early in dismissing underperforming CEOs experienced short-lived declines in operating performance close to the time of dismissal but recovered immediately afterward. On the other hand, when boards are not vigilant, issues can go undetected and thus remedies such as CEO dismissal will be delayed (Gibelman & Gelman, 2002). When the need for a replacement finally becomes urgent, the firm often finds itself in crisis mode, scrambling for a replacement rather than planning for a smooth transition in leadership. The delay also signals to shareholders that their needs were ignored, damaging board reputation and institutional trust (Haleblian & Rajagopala, 2006). As a result, future investment may be slow in coming. Gibelman and Gelman (2002) attribute the effects of CEO turnover to the circumstances surrounding the departure, and the inconsistent findings regarding the relationship between speed of dismissal and speed of recovery may also be explained this way. For example, Kim (2011) found type of performance feedback information to affect CEO dismissal speed, which in turn showed a U-curve relationship with post-succession performance. Similarly, Stannard (2016) shows that the way in 73 which a CEO is dismissed (i.e. publicly or privately) depends on whether the performance issues are accounting- or stock-based, and that when CEOs are dismissed in response to market concerns, positive outcomes are more immediate but shorter-lived. However, if CEO dismissal is used as a scapegoat for another underlying issue, this may be the catalyst for recovery (Hermalin, 2005). Finally, when it comes to pre-turnover performance, research indicates that the relationship is linear. In fact, Friedman and Singh (1989) found that the market tends to react positively to succession when pre-turnover performance is poor and negatively when pre-turnover performance is good. According to Hambrick & D’Aveni (1988), since decline is a slowly developing process, it may not trigger timely corrective action, suggesting that management is likely to remain unchanged throughout much of the decline. Studies have indicated that there is a “downward spiral” that organizations experience (Bozeman & Slusher, 1979; Forrester, 1971; Staw, Sandelands, & Dutton, 1981), whereby “weakness leads to even greater weakness” (Hambrick & D’Aveni; 1988). Since board diversity has been found to enhance CEO performance-turnover sensitivity as well as the likelihood of improved post-dismissal succession (Wahid, 2014), and since lengthy CEO tenure has an adverse effect on corporate turnaround (Abebe, 2010), there is even more reason to believe that the circumstances surrounding the dismissal may affect the relationship between speed of dismissal and speed of recovery. Since it is proposed above that boards with certain social network characteristics will delay CEO dismissal, it is under these circumstances that we expect recovery to be delayed after dismissal. We, therefore, propose the following: P-Recovery. When socially networked boards respond slowly to organizational underperformance, post-succession recovery will be delayed. P-Recovery. The longer the board’s response time to organizational underperformance, the longer it will take the organization to recover upon CEO replacement. Conclusion The propositions presented in this paper are meant to advance research on board characteristics, CEO dismissal, and organizational recovery, with Social Network Theory as the backdrop. The paper reviews several board characteristics that comprise networked boards and explain in detail why such boards delay CEO dismissals, which in turn delay recovery. It is ultimately proposed that non-independent boards with greater than optimal bandwidth will not act in a timely manner to remove CEOs in the face of organizational decline. Reasons for these delays include their lack of attentiveness to the firm as well as their social connections with the CEO and with other board members. It is then proposed that, when CEO dismissal is delayed due to networked boards, organizational recovery will be delayed as well. This paper makes important contributions to the corporate governance literature. First and foremost, it extends the research agenda on board composition, CEO turnover, and performance, to include the element of time. More specifically, it suggests that certain characteristics of corporate boards are more likely to inhibit the types of governance necessary to remove underperforming CEOs, and this, in turn, will impact the time it takes for organizational performance to recover. It also takes a contingency approach with respect to the relationship between speed of dismissal and speed of recovery, contending that the circumstances surrounding dismissal are likely to dictate the relationship. The paper also deepens the application of Social Network Theory to the study of corporate governance, addressing several elements of social networks that are found in non-independent boards and with “overboarded” directors. All arguments favor independence and optimal bandwidth as the type of governance needed to promptly remove underperforming CEOs and to quickly recover from poor performance. To pursue this line of inquiry empirically, research should address accounting- and market- based performance, both before and after CEO dismissal, given the conflicting effects that these have 74 been shown to have on the decision to dismiss as well as the consequences thereafter. To assess board independence, the historical conceptualization as outsider dominance does not take into account the conflicts of interest that many outside directors face. This information is sometimes disclosed in firm proxy statements but not with enough consistency for empirical analysis. It is therefore suggested that an independence index be computed as a function of outsider dominance, diversity (e.g. gender and nationality ratios among board members), director tenure, reciprocal interlocks to the extent that they can be determined, and whether the director served on the board at the time the CEO was selected. Implications Risks of poor performance due to inadequate governance are far-reaching, and shareholders and D&O (Directors and Officers) liability insurers are particularly vulnerable. Even with new management in place, shareholders might lose faith in the board if important strategic actions were delayed and especially if they negatively impact their investments. The anticipated effects of recovery can also discourage future investment in the company. Vigilant boards, composed of independent directors with optimal bandwidth, would be more likely than dense and embedded boards to replace the CEO when it is warranted. It would, therefore, behoove shareholders to participate in the election of directors, rather than turning that privilege over to the very board that took too long to replace the CEO due to its strong network structure. In the event that shareholders are discontented and file suit, D&O liability insurance can provide personal financial protection for directors but is somewhat of a catch 22 for the providers and the companies that purchase the coverage. Since it indemnifies the board members and reduces the risks associated with making poor decisions, there is less incentive to act promptly and in the best interest of shareholders. However, D&O coverage has become increasingly necessary to attract experienced and knowledgeable directors to the board. Insurers cannot afford to pay massive claims from class-action suits and can, therefore, benefit from insuring companies and directors where litigation is less likely. Although research shows that insurers do not actively monitor the management at companies where they provide coverage, according to Social Network Theory, perhaps they should. References Abebe, M. (2010). Top team composition and corporate turnaround under environmental stability and turbulence. Leadership and Organization Development Journal, 31(3): 196-212. Adams, R., & Ferreira, D. (2007). A Theory of Friendly Boards. Journal of Finance, 62(1): 217-250. Barney, J., & Ouchi, W. (Eds.). (1986). Organizational Economics. Jossey-Bass, San Francisco. Bebchuk, L., Weisbach, M. (2010). The state of corporate governance research. Review of Financial Studies, 23(3): 939-961. Bozeman, B., & Slusher, E. (1979). Scarcity and environmental stress in public organizations: A conjectural essay. Administrative Science Quarterly, 11: 335-356. Burrough, B., & Helyar, J. (Eds.). (1990). Barbarians at the Gate. Harper & Row, New York. Burt, R. (1980). On the functional form of corporate cooptation: Empirical findings linking the intensity of market constraint with the frequency of directorate ties. Social Science Research, 9(2): 146-177. Byrne, D. (1997). An overview (and underview) of research and theory within the attraction paradigm. Journal of Social and Personal Relationships, 14(3): 417-431. Carter, D., Simkins, B., & Simpson, W. (2003). Corporate governance, board diversity, and firm value. Financial Review, 38(1): 33-53. Clements, C., Jessup, R., Neill, J., & Wertheim, P. (1918). The relationship between director tenure and director quality. International Journal of Disclosure and Governance, 15(3): 142-161. 75 Condon, J., & Crano, W. (1988). Inferred evaluation and the relations between attitude similarity and interpersonal attraction. Journal of Personality and Social Psychology, 54(5): 798-797. Daily, C., Johnson, J., & Dalton, D. (1999). On the measurements of board composition: Poor consistency and a serious mismatch of theory and operationalization. Decision Sciences, 30, 86– 106. Daily, C., Dalton, D., & Cannella, A. (2003). Corporate governance: Decades of dialogue and data. Academy of Management Review, 28(3): 371-383. Drucker, P. (1981). The Bored Board. In Toward the Next Economics and Other Essays. Harper & Row, New York. Dyl, E. (1988). Corporate control and management compensation: Evidence on the agency problem. Managerial and Decision Economics, 9(1): 21-25. Ertugrul, M., & Krishnan, K. (2011). Can CEO dismissals be proactive? Journal of Corporate Finance, 17(1): 134-151. Fich, E., & Shivdasani, A. (2006). Are busy boards effective monitors? Journal of Finance, 61: 689-724. Finkelstein, S., Hambrick, D., & Canella, A. (2008). Strategic Leadership: Theory and Research on Executives, Top Management Teams, and Boards. Oxford University Press. Forrester, J. (1971). World Dynamics. Wright-Allen, Cambridge MA. Friedman, S., & Singh, H. (1989). CEO succession and stakeholder reaction: The influence of organizational context and event content. Academy of Management Journal, 32(4): 718-744. Gamson, W., & Scotch, N. (1964). Scapegoating in baseball. American Journal of Sociology, 70(1): 69-72. Giambatista, R., Rowe, W., & Riaz, S. (2005). Nothing succeeds like succession: A critical review of leader succession literature since 1994. Leadership Quarterly, 16(6): 963-991. Gibelman, M., & Gelman, S. (2002). On the departure of a chief executive officer: Scenarios and implications. Administration in Social Work, 26(2): 63-82. Gladstein, D., & Reilly, N. (1985). Group decision-making under threat: The tycoon game. Academy of Management Journal, 28(3): 613-627. Glazer, J. (1994). The United Way Scandal. John Wiley and Sons, New York. Granovetter, M. (1985). Economic Action and Social Structure: The Problem of Embeddedness. American Journal of Sociology, 91(3): 481-510. Haleblian, J., & Rajagopalan, N. (2006). A cognitive model of CEO dismissal: Understanding the influence of board perceptions, attributions and efficacy beliefs. Journal of Management Studies, 43(5): 1009-1026. Hambrick, D., & D'Aveni, R. (1988). Large corporate failures as downward spirals. Administrative Science Quarterly, 33(1): 1-23. Hambrick, D., Misangyi, V., & Park, C. (2015). The quad model for identifying a corporate director's potential for effective monitoring: Toward a new theory of board sufficiency. Academy of Management Review, 40(3): 323-344. Harris, I., & Shimizu, K. (2004). Too busy to serve? An examination of the influence of overboarded directors. Journal of Management Studies, 41(5): 775-798. Hazarika, S., Karpoff, J., & Nahata, R. (2012). Internal corporate governance, CEO turnover, and earnings management. Journal of Financial Economics, 104: 44-69. Hermalin, B. (2005). Trends in corporate governance. Journal of Finance, 60(5): 2351-2384. Hu, J., Kim, J., & Lin, Z. (2015). Does timely loss recognition improve the board's ability to learn from market prices? Evidence from worldwide CEO turnovers. Journal of Interational Accounting Research, 14(1): 1-24. Hwang, S. (1993), August 23. Fired Tambrands CEO was unusually close to consulting firm. Wall Street Journal, p. A1. Jiraporn, P., & Gleason, K. (2007). Capital structure, shareholder rights, and corporate governance. The Journal of Financial Research, 30(1): 21-33. 76 Katz, R. (1982). Project communication and performance: An investigation into the effects of group longevity. Administrative Science Quarterly, 27: 81-104. Kay, M. (1994, June 30). JCC board uncovers financial scandal. Washington Jewish Week, pp. 4, 23. Kim, J.-Y., Kim, J.-Y., & Miner, A. (2006). Organizational learning from extreme performance experience: The impact of success and recovery experience. Organization Science, 20(5): 958- 978. Kim, Y. (2011). Speed of CEO dismissal: An attribution-based model of when boards of directors fire CEOs in response to performance downturn. (Doctoral Dissertation): University of North Carolina Chapel Hill. Libit, W., & Freier, T. (2015a). Is 75 the new 68: Director tenure, mandatory director retirement and related issues. Insights, 29(3): 2-11. Libit, W., & Freier, T. (2015b). Director tenure: The next boardroom battle. The Corporate Board. Lublin, J. (1991, June 6). More chief executives are being forced out by tougher boards. Wall Street Journal, p. A1. Milliken, F., & Martins, L. (1996). Searching for common threads: Understanding the multiple effects of diversity in organizational groups. Academy of Management Review, 21(2): 402-433. Mizruchi, M. (1996). What do interlocks do? Analysis, critique, and assessment of research on interlocking directorates. Annual Review of Sociology, 22: 271-298. Mobbs, S. (2013). CEOs under fire: The effects of competition from inside directors on forced CEO turnover and CEO compensation. Journal of Financial and Quantitative Analysis, 48(3): 669-698. Monks, R., & Minow, N. (1996). Watching the watchers: Corporate governance for the 21st century. Blackwell Business, Cambridge, MA. Montoya, R., & Horton, R. (2004). On the importance of cognitive evaluation as a determinant of interpersonal attraction. Journal of Personality and Social Psychology, 86(5): 696-712. Nystrom, P., & Starbuck, W. (1984). To avoid organizational crisis, unlearn. Organizational Dynamics, 12(4): 53-65. Palmer, D. (1983). Broken ties: Interlocking directorates and intercorporate coordination. Administrative Science Quarterly, 28(1): 40-55. Palmer, D., Friedland, R., & Singh, J. (1986). The ties that bind: Organizational and class bases of stability in a corporate interlock network. American Sociological Review, 51(6): 781-796. Perry, T., & Peyer, U. (2005). Board seat accumulation by executives: A shareholder's perspective. Journal of Finance, 60: 2083-2123. Phan, P., & Lee, S. (1995). Human capital or social networks: What constrains CEO dismissals? Academy of Management Journal, 37-41. Pulliam, S., & Pereira, J. (1995), August 14. Reebok CEO Fireman faces criticism by institutional holders. Wall Street Journal, pp. C1, C2. Robinson, G., & Dechant, K. (1997). Building a business case for diversity. Academy of Management Executive, 11(3): 21-31. Shivdasani, A., & Yermack, D. (1999). CEO involvement in the selection of new board members: An empirical analysis. Journal of Finance, LIV(5): 1829-1853. Sinclair, M. (2000), December. Mismanaged finances: Food for the Poor, Feed the Children come under fire. Nonprofit Times, pp. 1, 10, 12. Smith, M. (1996). Shareholder activism by institutional investors: Evidence from CalPERS. Journal of Finance, 51: 227-252. Stannard, T. (2016). Firing on all cylinders? CEO retention and performance: A Learning approach. Staw, B., Sandelands, L., & Dutton, J. (1981). Three-rigidity effects in organizational behavior: A multi-level analysis. Administrative Science Quarterly, 26: 501-524. 77 Triana, M., Miller, T., & Trzebiaowski, T. (2014). The double-edged nature of board gender diversity: Diversity, firm performance, and the power of women directors as predictors of strategic change. Organizational Science, 25(2): 609-632. Uzzi, B. (1996). The sources and consequences of embeddedness for the economic performance of organizations: The network effect. American Sociological Review, 61(4): 674-698. Vafeas, N. (2003). Length of board tenure and outside director independence. Journal of Business Finance & Accounting, 30(7 & 8): 1043-1064. Van Knippenerg, D., De Dreu, C., & Homan, A. (2004). Work group diversity and group performance: An integrative model and research agenda. Journal of Applied Psychology, 89: 1008-1022. Wahid, A. (2012). Director heterogeneity and its impact on board effectiveness (Doctoral Dissertation). UMI Dissertation Publishing. Wellman, B. (1999). Networks in the global village: Life in contemporary communities. Westview, Boulder, CO. Westphal, J., & Bednar, M. (2005). Pluralistic ignorance i corporate boards ad firm's strategic persistence to low firm performance. Administrative Science Quarterly, 50(2): 262-298. Wiley, C. (1995). The ABC's of business ethics: Definitions, philosophies and implementation. Industrial Management, 37(1): 22-26. Zahra, S., & Pearce, J. (1989). Boards of directors and corporate financial performance: A review and integrative model. Journal of Management, 15(2): 291-334. Zhu, D., & Chen, G. (2015). Narcissism, director selection, and risk-taking spending. Strategic Management Journal, 36: 2075-2098. Zucker, L. (1977). The role of institutionalization in cultural persistence. American Sociological Review, 42: 726-743.