Received: 9 December 2019 Revised: 1 April 2021 Accepted: 3 April 2021

DOI: 10.1111/corg.12376

ORIGINAL ARTICLE

Corporate performance and CEO dismissal: The role of social

category faultlines

Alana Vandebeek1 | Wim Voordeckers2 | Jolien Huybrechts3 |

Frank Lambrechts2

1
Faculty of Economics and Business, Work and
Organisation Studies, KU Leuven, Antwerpen, Abstract
Belgium Research Question/Issue: This study investigates the moderating role of social cate-
2
Faculty of Business Economics, RCEF
gory faultlines in the relationship between firm performance and CEO dismissal. We
Research Center for Entrepreneurship and
Family Firms, Hasselt University, Hasselt, also examine how two board contingencies—the presence of board evaluation and
Belgium
the number of board committees—affect how social category faultlines moderate the
3
School of Business and Economics,
Department of Organization & Strategy, performance–CEO dismissal relationship.
Maastricht University, Maastricht, The Research Findings/Insights: Using panel data on Belgian listed firms covering 2006
Netherlands
to 2014, we find that the negative relationship between firm performance and the
Correspondence likelihood of CEO dismissal is significantly weaker when boards experience social cat-
Wim Voordeckers, Hasselt University, Faculty
of Business Economics, RCEF Research Center egory faultlines. Further investigation of board contingencies demonstrates that
for Entrepreneurship and Family Firms, social category faultlines have a stronger influence on the performance–CEO dis-
Martelarenlaan 42, 3500, Hasselt, Belgium.
Email: [email protected] missal relationship when the board does not engage in board evaluation or has
greater numbers of board committees.
Theoretical/Academic Implications: This study builds on the social identity perspec-
tive by providing empirical support for social category faultlines within the context
of boards of directors. Social category faultlines lead to identity-based subgroups
within the board, which reduce directors' identification with the board as a whole,
ultimately impacting the board's capability to dismiss a poorly performing CEO.
Moreover, our study demonstrates how board contingencies can promote and/or
weaken directors' identification with the board, which affects the salience of
identity-based subgroups.
Practitioner/Policy Implications: This study offers insights for practitioners inter-
ested in improving board effectiveness. Our evidence implies that decisions con-
cerning the appointment of directors should be based on criteria that take faultlines
and subgroups into account. Furthermore, our findings highlight the importance of
strengthening the superordinate board identity by implementing initiatives such as
board evaluation.

KEYWORDS
corporate governance, board composition, board diversity, faultlines, CEO dismissal

Corp Govern Int Rev. 2021;1–25. wileyonlinelibrary.com/journal/corg © 2021 John Wiley & Sons Ltd 1
2 VANDEBEEK ET AL.

1 | I N T RO DU CT I O N Vandebeek et al., 2016; Veltrop et al., 2015). To date, only one study
has focused on how faultlines affect directors' capability to dismiss an
The duty of boards of directors to select, compensate, and fire the underperforming CEO (Van Peteghem et al., 2018). This study found
CEO is a fundamental element of corporate governance (Lel & that faultlines negatively impact the firm performance–CEO dismissal
Miller, 2008). The decision to replace a CEO is one of the most impor- relationship from a sample of US-listed firms using a faultlines mea-
tant responsibilities of a board because it can have long-term implica- sure based on a diverse mix of informational (i.e., directly job related)
tions for the firm's investments, operations, and financing (Huson and social-category attributes (i.e., entailing a social identity; Jehn
et al., 2001). Directors have a fiduciary duty to make replacement et al., 2008; Jehn & Rupert, 2008). However, this faultlines approach
decisions in an informed way, in good faith, and with the best inter- suffers from a limitation in that “measuring a single faultline based on
ests of the corporation in mind (Johnson et al., 1996). characteristics that reflect very different aspects of individuals
Extensive research has indicated that the relationship between may hinder interpretation of its effect” (Hutzschenreuter &
corporate performance and CEO dismissal is a measure of the effec- Horstkotte, 2013, p. 719), making precise theorizing very difficult
tiveness of a corporate governance system (Durukan et al., 2012). (Bezrukova et al., 2009). In fact, recent faultline literature makes a dis-
Empirical studies have shown that firm performance is indeed consis- tinction between faultlines based on social category attributes, which
tently negatively related to CEO dismissal (Brunello et al., 2003; entail a certain social identity (e.g., gender and age), and faultlines
Denis & Denis, 1995; Fiordelisi & Ricci, 2014; Furtado & Karan, 1990; based on informational attributes, which are job-related and revolve
Kesner & Dalton, 1994; Lau et al., 2009; Ocasio, 1994). However, the around knowledge (e.g., education and work experience; Bezrukova &
direct relationship between firm performance and CEO dismissal fails Uparna, 2009; Chung et al., 2015; Jehn et al., 2008; Jehn &
to explain why poorly performing CEOs sometimes remain in their Rupert, 2008). Informational faultlines are driven by informational
position, and it has been suggested that the relationship between processes and are most relevant for strategic decisions in which effec-
poor performance and CEO dismissal is moderated by overlooked fac- tive pooling of information and integration of alternative perspectives
tors (He & Fang, 2016). Specifically, most of the board research has is important (e.g., different directors have different pieces of informa-
considered the board's composition as one of the most fundamental tion that need to be integrated for optimal decision-making;
dimensions (Haleblian & Rajagopalan, 2006) and has studied the Bezrukova et al., 2009). On the other hand, social category faultlines
effects of board composition in terms of the type of directorship are explained by social categorization processes (Ashforth &
(i.e., inside, independent, or affiliated director) and also from a demo- Mael, 1989; Hogg & Terry, 2000; Tajfel & Turner, 2004) and are most
graphic perspective that defines board composition in terms of char- relevant in common-goal groups that need to accomplish a specific
acteristics such as age and gender. task (in our case, fire a CEO following the information available [i.-
While the notion of diversity in boards' demographic characteris- e., poor performance]; Bezrukova et al., 2009). We will contribute to
tics has received considerable attention (Erhardt et al., 2003; this literature by theorizing and testing the effect of social category
Siciliano, 1996), there is still little attention paid to the interaction of faultlines on the capability of boards to dismiss a poorly
multiple demographic attributes. In contrast to earlier studies of diver- performing CEO.
sity that examined the heterogeneity or dispersion of one diversity Building on the social identity perspective, we provide a meaning-
attribute at a time, faultline theory suggests that the interaction of ful extension of the work on faultlines in the strategic management lit-
multiple demographic attributes affects group dynamics to a greater erature and find economically and statistically significant evidence
extent than separate attributes (Lau & Murnighan, 1998). Group that performance–CEO dismissal sensitivity is negatively influenced
faultlines are defined as hypothetical dividing lines that split a by social category faultlines. Furthermore, this study explicitly
group into multiple subgroups that are relatively homogeneous based responds to recent calls to focus on moderators that can affect
on a set of diversity attributes (Lau & Murnighan, 1998) and considers faultline dynamics (cf. Van Peteghem et al., 2018; Veltrop
how a bundle of different demographic characteristics may influence et al., 2015). We find that boards that do not engage in board evalua-
group processes and performance outcomes (Thatcher & Patel, 2012). tion and boards with greater numbers of board committees suffer
Because boards of directors are groups of individuals who provide cor- more from the negative impact of social category faultlines. We use a
porate oversight (Forbes & Milliken, 1999; Merchant & Pick, 2010), sample of Belgian listed firms across the time span of 2006–2014,
group faultlines can be expected to play an important role in deter- which provides us with a unique institutional context. Belgium applies
mining their effectiveness. In this study, we will examine the effect of a one-tier Continental European corporate governance system and
group faultlines (based on the attributes of age, gender, and family can be classified as a French-civil law country (Cuervo, 2002; La Porta
membership) on the capability of boards to dismiss a poorly et al., 1998). These national characteristics differ from the Anglo-
performing CEO. Saxon institutional contexts (e.g., the United Kingdom and the United
There are a number of studies that examine group faultlines on States) on which the CEO dismissal literature has typically focused
boards of directors and have reported largely negative organizational (e.g., Dah et al., 2014; Fiordelisi & Ricci, 2014; Van Peteghem
and team consequences (i.e., Basco et al., 2019; Basco & et al., 2018). While the Anglo-Saxon system is based on market con-
Voordeckers, 2015; Crucke & Knockaert, 2016; Kaczmarek trol, the Continental European system is characterized by large-
et al., 2012a, 2012b; Tuggle et al., 2010; Van Peteghem et al., 2018; shareholder control (Cuervo, 2002), and the vast majority of Belgian
VANDEBEEK ET AL. 3

companies are family owned (La Porta et al., 1999). It thus follows that personally responsible for firm-level outcomes, and their actions are
Belgian boards are often partly composed of family relatives linked to performance (Fama, 1980).
(Voordeckers et al., 2007). Therefore, our institutional context pro- However, not all boards may be capable of dismissing a poorly
vides us with an interesting social identity to examine—identification performing CEO. It is argued that board characteristics can affect the
as a family member—as a relevant base for social category faultline likelihood of forced CEO dismissal following poor performance
formation on boards of directors. Furthermore, a comprehensive (Haleblian & Rajagopalan, 2006). For example, Weisbach (1988) found
Belgian Corporate Governance Code was developed during the study a stronger performance–CEO dismissal sensitivity when firms had a
period (in 2009), which is enforced by law for public companies. greater share of independent directors on the board. Independent
Following the development of the Code, the Belgian federal directors have a greater incentive to monitor because being a director
government adopted gender quota legislation for boards of directors of a well-performing company signals their competence to the direc-
on July 28, 2011, imposing a female quota of one third, to tor market. However, even boards that are well constructed and orga-
which boards must adhere within the next 5 years. Therefore, our nized can fall victim to the destructive group patterns that occur
institutional context also provides us with valuable variations when groups of individuals are placed in decision-making roles
among the number of female directors on a board, as we investigate a (Merchant & Pick, 2010). Most boards follow generally agreed-upon
period in which changes have occurred with respect to gender ideals and are influenced by institutional pressures when constructing
representation on corporate boards. their boards, but they do not consider the potential group faultlines
that can exist due to differences in board composition.

2 | T H E O R E T I C A L BA C K G R O U N D A N D
HYPOTHESIS DEVELOPMENT 2.2 | Faultline theory and boards of directors

2.1 | Firm performance and CEO dismissal The concept of faultlines is an emerging topic in the corporate gover-
nance literature, and only a few studies have investigated the relation-
A large body of research has been attempting to determine the drivers ship between faultlines and board effectiveness. For example, Tuggle
of CEO dismissal, with a focus on its relationship with firm perfor- et al. (2010) found that group faultlines influenced the amount of time
mance. This relationship has been explained with reference to agency the board spent on entrepreneurial issues in board meetings. Veltrop
theory (Lausten, 2002), according to which principals can assure et al. (2015) used a slightly different conceptualization of faultlines
themselves that the agent will make optimal decisions only if the (i.e., factional faultlines), which become stronger when demographic
agent is monitored (Jensen & Meckling, 1976). The agency perspec- attributes align within the factional affiliations on the board, and
tive suggests that boards of directors have the responsibility to effi- argued that these negatively affected board performance. Kaczmarek
ciently monitor and control managers and the CEO (Fama, 1980; et al. (2012b) found that board faultlines impact firm performance,
Fama & Jensen, 1983) and that boards should dismiss the CEO when along with certain conditions that either exacerbated or ameliorated
firm performance is poor (Tushman & Rosenkopf, 1996). Therefore, this relationship. Vandebeek et al. (2016) and Crucke and
scholars have assessed whether boards succeed in their role as an Knockaert (2016) discovered a negative relationship between
effective internal control mechanism by investigating the incidence of faultlines and board task performance. Finally, Van Peteghem
CEO dismissal in poorly performing firms (Denis & Denis, 1995). et al. (2018) found that faultlines were associated with lower firm per-
Poor performance is likely to be attributed to the leader of the formance, higher abnormal CEO compensation, and lower CEO
organization (Harrison et al., 1988). A CEO should be replaced to turnover–performance sensitivity using a single faultline measure
determine if a different individual can improve the firm's situation. based on two different kinds of characteristics (i.e., informational and
Directors are expected to see it as the duty of the CEO to guide the social category). However, if different kinds of characteristics are used
firm to success in any environment and to hold the CEO responsible in one single measure, it becomes more difficult to determine which
for the firm's failures (Puffer & Weintrop, 1991; Walsh & characteristics drive the observed effect (Hutzschenreuter &
Seward, 1990). It thus follows that poor performance can act as a trig- Horstkotte, 2013). Moreover, informational and social category attri-
ger (Ocasio, 1994) to drive boards of directors to induce a leadership butes have been associated with different effects. Empirical research
change (Harrison et al., 1988). Moreover, pressured by external mar- on the effects of informational faultlines is equivocal, with studies
kets and increased scrutiny by the financial press, listed firms are reporting positive effects (e.g., Rupert et al., 2016), negative effects
expected to act quickly to replace a CEO in response to poor perfor- (e.g., Bezrukova et al., 2012) or even no effects on organizational out-
mance (Faleye et al., 2011; Farrell & Whidbee, 2002). As boards of comes (e.g., Chung et al., 2015). However, social category faultlines
directors have a fiduciary responsibility to act in the interests of the have consistently been related to a number of negative effects on
firm (Johnson et al., 1996), they should act to address poor perfor- group functioning, such as intra-group conflict (Jehn &
mance in order to avoid shareholder lawsuits and loss of investor con- Bezrukova, 2010; Molleman, 2005), decreased trust and cohesion
fidence (Harrison et al., 1988). In sum, CEOs should be held (Cronin et al., 2011; Polzer et al., 2006), and communication issues
accountable for poor performance because they are assumed to be (Lau & Murnighan, 2005). Therefore, social category faultlines may
4 VANDEBEEK ET AL.

P 2
explain why particular board compositions are less able to dismiss a index (calculated as 1 pi , where pi is the proportion of the board
poorly performing CEO. in the ith category) of 0.5. While the second board shows a strong
In this study, we will examine social category faultlines based on faultline, it would generate Blau's heterogeneity index of 0.5 as well.
the alignment of three attributes: gender, age, and family membership.
Gender and age are attributes that entail a social identity and can
shape perceptions and behaviors through mechanisms of categoriza- 2.3 | Social category faultlines, identification with
tion and stereotyping (cf. Bezrukova et al., 2009; Chung et al., 2015). the board, and CEO dismissal
Stereotyping can make it difficult to think as a collective and to act
productively on a problem. For example, Kunze and Bruch (2010) Board effectiveness has been described as the capability of the board
found a negative effect of age-based faultlines on teams' productive to perform its tasks effectively and to continue working together as a
energy because members of different age-based subgroups are found group (Forbes & Milliken, 1999). The ability to work together plays an
to mainly promote the diverse aims of their own subgroups and to important role, as boards are supposed to be governed via member
have a reduced motivation to contribute to the group as a whole. consensus, and most board votes should be unanimous
Gender and age are also considered important aspects of diversity (Ocasio, 1994). Prior research indicates that the expected behavior of
within boards (Van der Walt & Ingley, 2003). In addition, during our board members follows from the directors' overarching goal of carry-
observed period, a new law was introduced in Belgium that urged ing out their fiduciary duties to the shareholders (Bainbridge, 2002;
firms to install boards of directors whose composition was at least Lan & Heracleous, 2010). The more a director identifies with the
one-third women. Rather than using a voluntary approach to promote board as a whole, the more likely they are to apply themselves to their
women on boards, Belgium uses a regulatory enabling approach. This fiduciary duty and fulfill their resource provision and monitoring roles
provides us with valuable variations over time in terms of directors' (Hillman et al., 2008). However, from a social categorization perspec-
gender and age, as gender quota could end the board's nature as an tive, social category faultlines can “separate” the board as a whole, as
“old boys club,” as more (and often younger) women are being they signal different social identities and induce the members to per-
appointed (Burgess & Tharenou, 2002). ceive each other as being part of a specific social category (Harrison &
In addition, we observe a third important attribute among the Klein, 2007), creating identity-based subgroups. Such subgroup for-
directors of Belgian boards. Belgium is characterized by a high number mation can create division, as group members may feel little personal
of family businesses, and family members act as directors on many of attachment to members of other subgroups, leading to decreased
their boards. Family members are expected to form a highly commit- cohesiveness or trust in the group (Lau & Murnighan, 2005; Polzer
ted in-group because they have often been involved in the firm since et al., 2006) and an “us-versus-them” attitude, which facilitates the
its foundation, consider themselves part of the organization, and value fragmentation of the group (Bezrukova et al., 2009; Li &
membership in it more than nonfamily members do (Deephouse & Hambrick, 2005). Divisions among directors can render the board
Jaskiewicz, 2013; Uhlaner et al., 2007). These members have the same unable to take effective management action (i.e., lead to a deadlock
culture, values, and identity (Sundaramurthy & Kreiner, 2008). There- on the board), which can even prompt directors to neglect stake-
fore, we will treat family membership as an additional social category holders' interests (Donaldson et al., 2020; Kim, 2003). Moreover,
attribute. faultlines forming divisive subgroups can also channel intense sub-
To illustrate faultline formation on boards in our sample, Table 1 group power struggles (Hinds et al., 2014). Such subcategorization
provides an example and a comparison with Blau's diversity index reduces group cohesiveness and causes subgroup resistance (Rink &
(Blau, 1977), which has been the most widely used measure to cap- Jehn, 2010). The formation of subgroups can cause groups to lose
ture variations within a group of people (Harrison & Klein, 2007). As focus on the group as a whole, and subgroups can become polarized
presented in this table, the first board has no faultline because there around particular viewpoints as a result of the pressure exerted within
is no alignment of the three attributes of family membership, gender, the subgroups on conformity and competition between the subgroups
and age. This board composition would generate Blau's heterogeneity (Bezrukova et al., 2007; Thatcher et al., 2003).

T A B L E 1 Two examples of boards'

Family membership Gender Age
social category faultline strength and
Board 1: No faultline (Blau = 0.5) Family member Female 48 comparison to Blau's heterogeneity index
Family member Male 32
Outsider Female 31
Outsider Male 47
Board 2: Strong faultline (Blau = 0.5) Family member Male 48
Family member Male 47
Outsider Female 31
Outsider Female 32
VANDEBEEK ET AL. 5

Polarized opinions can negatively impact the quality of decisions, Identification with the board can engender a strong team identity,
as directors will search and affirm information that reinforces their which in turn is expected to act as “social glue,” or a common uniting
preferred position, disconnecting the board as a whole even more force through which the members remain committed to each other
(Oehmichen et al., 2017). Furthermore, such polarization can prevent and focused on the team's goals (Rink & Jehn, 2010). Such mecha-
subgroup members from sharing task-relevant information across sub- nisms could prevent identity-based subgroups from eliciting negative
groups and taking such information seriously when it comes from group processes (such as polarization) based on stereotyping and out-
members of another subgroup (Jehn & Rupert, 2008; Milliken & group biases.
Martins, 1996). The resulting communication and coordination diffi-
culties could increase the decision-making power of the CEO
(Cheng, 2008; Eisenberg et al., 1998; Ryan & Wiggins, 2004; Van 2.4.1 | Promote identification with the board
Peteghem et al., 2018). When performance is bad, the CEO can take
advantage of social category faultlines in the board, profiting from the Board evaluation is increasingly being used to improve board effec-
division and the inability of board members to form a strong alliance tiveness (Kiel & Nicholson, 2005; Long, 2006; Rasmussen, 2015). The
against him (Ocasio, 1994). Due to various viewpoints and arguments Belgian Corporate Governance Code 2009 recommends that boards
among subgroups in board meetings, it can take longer to reach a con- carry out such evaluations. Board evaluations can highlight potential
sensus on the appropriate course of action; the board can get stuck at issues raised by strong social category faultlines and allow board
a crossroad, resulting in a “wait and see” attitude (Weber & members to work toward a resolution before these problems become
Wiersema, 2017) and a delay of the CEO dismissal decision. Indeed, major crises (Kiel & Nicholson, 2005). Board evaluations can increase
Donaldson et al. (2020) argue that a deadlocked board can lead to an directors' identification as a member of the board by directing their
entrenched CEO, because directors on a diverse board anticipate attention to the bigger picture and discussing organizational goals and
disagreeing in the future and thus have the incentive to block policies. shared objectives (Nederveen Pieterse et al., 2013; van Knippenberg
Indeed, if a particular diverse board composition leads to the exis- et al., 2011). Strengthening the collective identity of the director as a
tence of strong social category faultlines, schisms and problematic member of the board can minimize the effects of subgroup categori-
inter-subgroup relations can arise among directors, resulting in less zation and identification and can motivate the directors to strive to
attention being paid to the superordinate strategic priority that these reach an agreement (Bezrukova et al., 2009; Withers, Corley, &
directors are expected to share (“the bigger picture”; Porck Hillman, 2012).
et al., 2018). Directors may identify more as members of a subgroup Evaluation can improve the effectiveness of relationships
than as part of the board as a whole, thereby compromising their abil- between and among directors, “through the exploration of motivation
ity to work as a unified group. In sum, it will be more difficult for the and shared principles, levels of debate and challenge, and the clarity
board to replace a poorly performing CEO if social category faultlines and transparency of information flows between management, board
are strong, and the probability of CEO dismissal is likely to be less sen- members and committees” (Long, 2006, p. 554). The evaluation pro-
sitive to performance on boards that have social category faultlines cess can clarify expectations, encourage involvement, develop com-
than on those that do not. Accordingly, we propose the following, mitment and a sense of ownership, and ensure that directors
ceteris paribus: understand their legal duties and responsibilities (Kiel &
Nicholson, 2005). Moreover, board reflexivity (i.e., the extent to which
Hypothesis 1. The negative relationship between firm performance group members reflect upon and communicate about objectives, strat-
and the likelihood of CEO dismissal is weaker when boards egies, and processes and adapt them to circumstances; West
experience social category faultlines. et al., 1997) is likely to transpire during evaluations or debriefings.
In sum, evaluating their own performance provides directors with
a greater shared understanding of what they are trying to achieve,
2.4 | Role of board contingencies how they should work together to achieve these goals, and how they
should communicate to achieve the necessary collaboration
Studies have argued that faultline effects depend on contextual condi- (Bouwen & Fry, 1996), and the resulting conversational process can
tions (Meyer et al., 2014). According to the social identity perspective, shift attention from the disruptive identity-based subgroups to the
the presence of social category faultlines may not always elicit inter- board as a whole. Thus, we propose the following:
group bias to the same extent, and some contexts can reduce the
salience of social identities (Chung et al., 2015). Faultline effects have Hypothesis 2. The weakening moderating role of social category
also been found to be context dependent in board research (Crucke & faultlines on the negative relationship between firm perfor-
Knockaert, 2016; Kaczmarek et al., 2012b; Tuggle et al., 2010; mance and CEO dismissal is stronger when boards do not
Vandebeek et al., 2016; Veltrop et al., 2015). Therefore, we will engage in board evaluation (i.e., the board is less capable of dis-
explore the role of two board contingencies—the presence of board missing a poorly performing CEO when social category
evaluation and the number of established board committees—that faultlines are strong and the board does not engage in board
can promote or weaken identification with the board as a whole. evaluation).
6 VANDEBEEK ET AL.

2.4.2 | Weaken identification with the board underlying differences in asset structure, we excluded financial insti-
tutions, real estate firms, and financial holdings. First, the individual-
Boards can choose to either perform their roles through the full board level director information needed to calculate faultline strength was
or to delegate their authority to standing committees that are account- hand-collected from multiple sources, such as financial reports, com-
able to the board. The Belgian Corporate Governance Code 2009 rec- pany websites, press archives, and social media (e.g., LinkedIn). When
ommends that boards set up an audit and a remuneration and a director's demographic information was missing, we contacted the
nomination committee to perform particular board tasks. However, director personally to collect the data.2 Second, corporate governance
some boards have more than these required committees and include and accounting data were collected using financial reports and the
additional ones, such as strategy, corporate governance, financial, sci- financial database of Bureau Van Dijk.
entific, or stock option committees, which have various important tasks The original dataset consisted of information on 106 unique pub-
and roles. Committee members are specifically chosen to perform these licly traded firms covering 2006 to 2014. A total of 814 firm years
tasks and have additional responsibilities, which can give them a sense were obtained, which covered information on 7342 board members.3
of power and create tension between committee members and others These firms account for approximately 71% of the total market capi-
(Spira & Bender, 2004). In Belgium, firms also need to justify why these talization of Euronext Brussels. To estimate the statistical models, we
members were selected to serve on a certain committee. required non-missing data on CEO dismissal and all independent and
When directors are acting in a certain role or identifying with a control variables. We also used a lagged independent variable strat-
particular group, they undergo a process of self-categorization that egy, which required firms to have at least two consecutive years of
steers them to act in a manner appropriate for that role (Hogg & data during the study period. We compared CEOs across years to
Terry, 2000). Due to social category faultlines, perceptions of salient identify CEO turnover. To ensure that we considered only forced
conflicting subgroup identities are formed by board members through CEO turnovers, we hand-collected CEO turnover events. Following
the proceedings of the entire board. Salient subgroup identities can Wiersema and Zhang (2011), we coded CEO turnovers as forced only
be de-emphasized by directing directors' attention to the board as a if (a) the company's announcement indicated that the CEO's departure
whole and their common goals and fiduciary duties. However, we was forced; (b) the news coverage and press releases on the CEO suc-
argue that if boards have greater numbers of board committees, the cession event provided strong evidence that the CEO was removed
collective identification with the board will be weakened. Rather than by the board as a consequence of performance concerns; (c) the CEO
having one superordinate group that is involved in the entire decision- resigned effective immediately after a board meeting with no desig-
making process, all decisions are now being prepared within numerous nated CEO replacement; or (d) the CEO left abruptly with the board
board committees and later presented to the board as a whole to beginning a search for a new CEO (and the CEO did not leave to pur-
allow ratification of the decisions (Schwartz-Ziv & Weisbach, 2013), sue other opportunities). After data screening, in which we deleted
which might reduce a director's feeling of being part of an autono- unusable entries due to missing values required for our covariates, the
mous fiduciary group that makes decisions in the interests of the cor- final sample comprised 95 unique firms with 620 firm-year observa-
poration (Lan & Heracleous, 2010). Therefore, the weaker the tions between 2006 and 2014.
identification with the board as a whole, the less commitment to a Within our sample, CEO dismissal events represent 4.52% of firm
common goal (in our case, dismissing a poorly performing CEO) and years. This relative proportion is in line with prior corporate gover-
the more likely that social category faultlines will have a stronger neg- nance research, such as Van Peteghem et al. (2018) and Bushman
ative impact on the board's capability to dismiss a poorly performing et al. (2010). The remaining observations of the sample were kept as a
CEO. Accordingly, we propose the following: control sample of non-forced dismissal events. Table 2 reports the
descriptive statistics and correlations for our dependent, independent,
Hypothesis 3. The weakening moderating role of social category and control variables. An average firm has an industry-adjusted return
faultlines on the negative relationship between firm perfor- on assets (ROA) of .03, and 57% of the firms are family run.4 Fur-
mance and CEO dismissal is stronger when boards have greater thermore, an average CEO owns 25% of total shares outstanding,5 is
numbers of board committees (i.e., the board is less capable of 53 years old, and has an average tenure of 8 years. An average board
dismissing a poorly performing CEO when social category has nine members, and 42% are independent directors.6 The
faultlines are strong and the board has greater numbers of largest shareholders own an average of 44% of the total shares
board committees). outstanding.

3 | METHOD 3.2 | Measures

3.1 | Sample and data collection 3.2.1 | Dependent variable

This study uses data on Belgian firms1 listed on the Brussels Stock The dependent variable in our econometric models is forced CEO dis-
Exchange (Euronext Brussels) from 2006 to 2014. Because of the missal, which is captured using a binary dummy variable equal to 1 in
 VANDEBEEK ET AL.

TABLE 2 Mean, standard deviations, and correlations

M SD Min Max 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15

1. CEO dismissal 0.04 0.19 0 1 1.00

2. CEO tenure 1.90 0.71 0 3.69 0.12*** 1.00
(log)
3. CEO ownership 0.25 0.29 0 0.97 0.11*** 0.32*** 1.00
4. CEO age 52.90 7.87 34 88 0.00 0.24*** 0.03 1.00
***
5. CEO duality 0.16 0.36 0 1 0.03 0.11 0.05 0.27*** 1.00
6. Family CEO 0.36 0.48 0 1 0.11*** 0.28*** 0.83*** 0.01 0.18*** 1.00
*** *** ***
7. Independent 0.42 0.16 0 1 0.04 0.04 0.15 0.11 0.12 0.12*** 1.00
director ratio
8. Block 0.44 0.23 0 0.97 0.13*** 0.19*** 0.61*** 0.03 0.04 0.46*** 0.23*** 1.00
ownership
9. Family business 0.57 0.50 0 1 0.10** 0.14*** 0.53*** 0.01 0.05 0.58*** 0.16*** 0.56*** 1.00
*** *** ** ***
10. Board size 9.00 3.11 3 24 0.05 0.00 0.15 0.34 0.10 0.20 0.06 0.03 0.05 1.00
11. Board 2.21 0.84 0 5 0.00 0.04 0.08* 0.17*** 0.16*** 0.18*** 0.27*** 0.01 0.04 0.50*** 1.00
committees
12. Board 0.36 0.13 0 1 0.06 0.03 0.01 0.04 0.08** 0.04 0.07* 0.06 0.06 0.14*** 0.14*** 1.00
evaluation
13. Informational 0.40 0.13 0.05 0.86 0.02 0.09** 0.05 0.14*** 0.07* 0.04 0.03 0.09** 0.15*** 0.27*** 0.29*** 0.03 1.00
Faultlines
14. Ind. Adj. ROA 0.03 0.21 2.39 1.78 0.16*** 0.15*** 0.21*** 0.01 0.00 0.16*** 0.02 0.21*** 0.18*** 0.05 0.13*** 0.01 0.04 1.00
** *** *** ** *** *** ***
15. Social 0.58 0.12 0.17 0.89 0.01 0.08 0.24 0.05 0.03 0.27 0.09 0.26 0.25 0.23 0.13*** 0.02 0.04 0.02 1.00
category
Faultlines

*p < .10. **p < .05. ***p < .01; 95 public firms (representing 620 firm-year observations) from 2006 to 2014.
7
8 VANDEBEEK ET AL.

a given year if a company experiences a forced turnover of the CEO addition, following the recommendations of Meyer et al. (2014), we
and 0 otherwise. scale our numeric attribute (i.e., age) by its standard deviation. In this
study, the average social category faultline strength is 0.58.
We measured firm performance using the industry-adjusted ROA,
3.2.2 | Independent variables calculated as the operating income before interest and tax (EBIT)
divided by the book value of total assets, minus the industry average
The measure of social category faultlines is based on a cluster-based ROA. Following Barber and Lyon (1996), industry average ROA was
approach—average silhouette width (ASW) faultline clustering.7 Clus- measured using median ROA for all listed firms given their industry
ter analysis groups team members into clusters (i.e., subgroups) and size.9 We used ROA because this measure is less influenced by
according to their similarity, such that the clusters have maximum macroeconomic factors beyond managers' control and is often used in
internal homogeneity and between-cluster heterogeneity. The ASW the literature (Durukan et al., 2012; Fiordelisi & Ricci, 2014; He &
measure is a measure of the quality of a group's partitioning with ref- Fang, 2016; Huson et al., 2001; Lau et al., 2009; Mobbs, 2013;
erence to the within-subgroup homogeneity, the between-subgroup Visintin et al., 2017). We then multiplied our faultline measure by the
separation, and the optimal number of clusters. The ASW measure industry-adjusted ROA to create our interaction variable.10
goes beyond most faultline measures because it does not require the We will examine the moderating role of two contextual variables
attributes to be categorical and allowed us to work with continuous on the relationship between social category faultlines and the likeli-
measures such as age. hood of CEO dismissal following poor performance. First, we will
The ASW algorithm detects subgroups and calculates faultline examine this relationship for boards that do not engage in board eval-
strength in several steps. First, it determines a certain set of all possi- uation versus boards that do. This was measured with a dummy vari-
ble subgroup partitions, which is based on the team members' attri- able equal to 1 if the annual report contained a description of a board
bute distribution. The cluster analysis configures a number of evaluation being carried out and 0 otherwise. Next, we counted the
subgroups wherein each team member is placed in their subgroup various board committees present on the board, as stated in each
of size one. Next, these subgroups are merged into new, larger sub- company's annual report and compared boards with fewer board com-
groups. It is always the two most similar subgroups that are merged. mittees than the average to boards with more board committees than
In the end, one cluster encompasses the entire team. During this pro- the average.11 These two variables were used to split the sample into
cedure, all the subgroup configurations are stored in the algorithm's subsamples and are also included as control variables.
memory, and the ASW value is computed. The ASW value portrays
how all team members on average fit to their subgroup, and ranges
between 1 and 1. If no homogeneous subgroups exist, ASW is equal 3.2.3 | Control variables
to 0. If ASW equals 1, all existing subgroups are completely homoge-
neous. If ASW equals to 1, the formed subgroups are not adequate Our models also contain control variables for individual, board, and
and members of the same subgroup are more dissimilar to each other firm characteristics as identified by the dismissal literature to mitigate
than to members of different subgroups.8 The algorithm chooses the potential confounding effects. We control for CEO Tenure, measured
subgroup configuration that returns the highest ASW value. Our cal- as the logarithmic transformation of the number of years the CEO has
culations were performed using the asw.cluster package for faultline been found to hold the position, and CEO Ownership, measured as the
calculation in R (Meyer & Glenz, 2013). We operationalized social cat- proportion of shares owned by the CEO, because these variables have
egory faultline strength along with three demographic characteristics: been shown to be important indicators of CEO power, and CEOs can
family membership, gender, and age. First, we categorized directors more easily prevent their own involuntary dismissal (Daily &
into family and non-family members. The presence of both family and Johnson, 1997; Fredrickson et al., 1988). Furthermore, we control for
non-family members on a board is known to create particular group CEO Age because of the strong relationship with CEO turnover
dynamics (Collin & Ahlberg, 2012; Vandebeek et al., 2016; Zattoni (Goyal & Park, 2002; Weisbach, 1988). At the board level, we control
et al., 2015). Boards of directors in non-family firms will benefit from for Independent Director Ratio by measuring the total proportion of
the absence of family directors since they have no allegiance to a cer- independent directors on the board, because this variable can give an
tain family, and these boards will have no accompanying family iden- indication of structural CEO power (Mobbs, 2013; Weisbach, 1988).
tity or dynamics (Cannella et al., 2015; Fredrickson et al., 1988; As concentrated ownership could decrease CEO power (Boeker &
Vandebeek et al., 2016). Second, we used board members' gender and Goodstein, 1993), we include Block Ownership, measured as the own-
the numeric value of their age. These characteristics are often used in ership percentage of the largest shareholder. Next, we include Board
faultline research because they represent certain social categories Size to control for group size (Hutzschenreuter & Horstkotte, 2013),
(i.e., women vs. men and younger vs. older directors) that lead to simi- measured as the total number of board members. Lastly, while Infor-
larity attraction and social identity attribution (Bezrukova et al., 2009; mational Faultlines are more relevant for strategic decisions in which
Crucke & Knockaert, 2016; Thatcher & Patel, 2012; Vandebeek effective pooling of information and integration of alternative per-
et al., 2016). The ASW method allows us to use the numeric value of spectives is important, we follow prior faultline literature
age, rather than categorizing age prior to the faultlines calculation. In (e.g., Bezrukova et al., 2009; Chung et al., 2015) and control for this
VANDEBEEK ET AL. 9

type of faultlines, based on the attributes “type of directorship,” (1) boards that do not engage in board evaluation versus boards that
“board tenure,” “educational level,” and “educational do and (2) boards with fewer versus more board committees to test
12,13
specialization.” H2 and H3. Using a split-sample design is preferable to the alternative
Furthermore, our model also controls for firms run by families by method of a three-way interaction, which will force unobserved varia-
including a dummy that equals 1 if the business is a family firm (Family tions (i.e., the variation in outcomes beyond that explained by the
Business) and 0 otherwise. Family firms can be less likely to exhibit independent variables [i.e., the error term ε]) for both groups to be
CEO dismissal, as they are more focused on long-term goals (Visintin the same, because there is only one error term (Hoetker, 2007). In
et al., 2017). Finally, each of the regression models contains a set of fact, the unobserved variation of two groups may differ, and con-
industry dummies to capture industry variations that might impact the straining the unobserved variation to be identical will affect the main
performance–CEO dismissal relationship. For example, it has been coefficient under consideration. Furthermore, the scale of all coeffi-
shown that CEO dismissal can be greater in highly competitive indus- cients in the model will be influenced. When the model is estimated
tries than in less competitive ones (DeFond & Park, 1999). Year separately for each group, the coefficients and standard errors are
dummies are included to control for macroeconomic shocks. consistent within each group, and we can compare the statistical sig-
nificance of the coefficients across groups (Hoetker, 2007).14
To ease reverse causation and endogenous variable concerns, all
3.3 | Data analysis right-hand side variables were lagged by 1 year in the analysis to
address the potential for a simultaneous determination of the depen-
Before analyzing the data, we assessed whether there was a dent and independent variables.
multicollinearity problem in our sample and computed variance infla-
tion factors (VIFs) for all the variables. The explanatory variables had
low VIF values (ranging between 1.12 and 2.02), which led us to con- 4 | RE SU LT S
clude that there was no multicollinearity problem in our sample (Hair
et al., 2014). There were no outliers in our dataset, as indicated by Table 3 reports the results of the hierarchical regression models used
two measures of influence (DFBETA and Cook's D). to test our hypotheses, considering CEO dismissal as the dependent
In order to test the research hypotheses, we conducted two types variable and including the direct effects of the explanatory and control
of analysis. First, we used a pooled logit model with clustered error variables. Models 1 to 3 are estimated using a pooled logit model with
terms, as indicated by a likelihood-ratio test. The pooled logit model clustered error terms. Models 4 to 7 are estimated using firthlogit
with clustered errors (by firm) takes into account that multiple observa- models to avoid the problem of separation. While the maximum likeli-
tions of the firm are not independent of each other, resulting in robust hood estimation of the logistic model omits a number of observations
errors (Greene, 2007). Next, we noted that a number of CEOs in our because they perfectly predict (forced turnover = 0), the firthlogit
sample could be family CEOs because the Belgian setting is character- model always produces finite, consistent estimates of regression
ized by a high level of family ownership. While we controlled for family parameters. As a result, the number of observations is higher for the
dynamics in our models by including a dummy that equals one if the firthlogit models than for the maximum likelihood models. Marginal
business is a family firm, we also ran our models with a dummy that effects are reported in all models. Table 3 shows a significantly nega-
equals 1 if the CEO is a family CEO and 0 otherwise but found that the tive relationship between firm performance and CEO dismissal (dy/
forced dismissal of family CEOs was a rare event and created problems dx = 0.33, p < 0.05), which suggests that CEOs in underperforming
of separation. In addition, the research shows that CEO duality firms are more likely to be dismissed. These results are consistent with
(i.e., chairman of the board and CEO are the same individual) is an prior studies on CEO dismissal.
important factor that can influence CEO dismissal because it indicates Furthermore, we find that the interaction term of faultline
their power (Finkelstein & D'aveni, 1994). However, including CEO strength and firm performance is positively significant (dy/dx = 0.56,
duality as a control variable also created problems of separation. There- p < 0.05). A positive interaction between faultline strength and firm
fore, we used the “firthlogit” command in Stata to run the Firth's penal- performance suggests a moderating effect of faultlines on the
ized maximum likelihood estimation (PMLE) for rare events in order to performance–CEO dismissal relationship. These results support H1's
include the control variables family CEO and CEO duality. The PMLE proposal that the negative relationship between firm performance and
approach is designed to address small-sample bias in the maximum like- the likelihood of CEO dismissal is weaker when boards experience
lihood estimation of the logistic model and always leads to finite esti- stronger social category faultlines. The control variable “family CEO”
mates of regression parameters, even in the case of rare events and was found to have a significantly negative effect on CEO dismissal,
separation problems (Coveney, 2015; Firth, 1993). Firth's logistic but the effect of social category faultlines on the performance–CEO
regression surpasses exact logistic regression, which is computationally dismissal relationship remained positive and significant.
intensive and can consider only a small number of (preferably dichoto- According to Hoetker (2007, p. 336), it is important to include
mous) covariates (Heinze & Schemper, 2002; Zorn, 2005). graphical presentations when interpreting main effects, as “the signifi-
We tested H1 by introducing the interaction between firm per- cance of the interaction effect cannot be determined just by the sig-
formance and social category faultlines. We then split the sample into nificance of the interaction coefficient.” Therefore, to provide a richer
10 VANDEBEEK ET AL.

TABLE 3 Social category faultlines (SCF) and performance–CEO dismissal sensitivity

Variables Model 1 Model 2 Model 3 Model 4 Model 5 Model 6 Model 7

CEO tenure (log) 0.01 (0.01) 0.01 (0.01) 0.01 (0.01) 0.01 (0.01) 0.00 (0.01) 0.00 (0.01) 0.00 (0.01)
CEO ownership 0.08 (0.06) 0.08 (0.05) 0.08 (0.05) 0.03 (0.06) 0.03 (0.06) 0.02 (0.06) 0.02 (0.06)

CEO age 0.00 (0.00) 0.00 (0.00) 0.00 (0.00) 0.00 (0.00) 0.00 (0.00) 0.00 (0.00) 0.00 (0.00)

CEO duality 0.02 (0.04) 0.04 (0.04) 0.03 (0.05) 0.03 (0.05)

Family CEO 0.15** (0.07) 0.15** (0.07) 0.14* (0.07) 0.14* (0.07)

Board evaluation 0.00 (0.02)

Board committees 0.00 (0.02)

Independent director ratio 0.04 (0.04) 0.03 (0.04) 0.04 (0.04) 0.04 (0.06) 0.03 (0.06) 0.04 (0.06) 0.04 (0.07)

Block ownership 0.04 (0.04) 0.04 (0.04) 0.03 (0.04) 0.04 (0.06) 0.04 (0.06) 0.04 (0.06) 0.04 (0.06)

Family business dummy 0.03 (0.02) 0.02 (0.02) 0.02 (0.02) 0.02 (0.02) 0.02 (0.02) 0.02 (0.02) 0.02 (0.02)

Board size 0.00 (0.00) 0.00 (0.00) 0.00 (0.00) 0.00 (0.00) 0.00 (0.00) 0.00 (0.00) 0.00 (0.00)

Inform. Faultlines 0.06 (0.06) 0.06 (0.06) 0.05 (0.06) 0.05 (0.07) 0.06 (0.07) 0.05 (0.07) 0.05 (0.07)

Ind. Adj. ROA 0.03 (0.03) 0.33** (0.15) 0.04 (0.03) 0.34* (0.18) 0.35* (0.18)

Social category Faultlines (SCF) 0.01 (0.07) 0.04 (0.06) 0.04 (0.09) 0.02 (0.09) 0.02 (0.09)

Ind. Adj. ROA * SCF 0.56** (0.28) 0.56* (0.32) 0.56* (0.33)

Year dummies Yes Yes Yes Yes Yes Yes Yes

Industry dummies Yes Yes Yes Yes Yes Yes Yes

No. of observations 620 620 620 650 650 650 650

No. of firms 95 95 95 99 99 99 99
χ2 51.05*** 53.53*** 57.21*** 7.66*** 7.66*** 6.69*** 6.21**

McFadden's pseudo-R2 0.16 0.17 0.19 0.26 0.28 0.29 0.27

Note: In this table, we examine the relationship between social category faultlines and performance–CEO dismissal sensitivity. We use data on Belgian firms listed on
the Brussels Stock Exchange (Euronext Brussels) from 2006 to 2014. The dependent variable is a dummy variable equal to 1 in a given year if a company experiences
a forced turnover of the CEO and 0 otherwise. Models 1 to 3 are estimated using a pooled logit model with clustered error terms. Models 4 to 7 are estimated using
firthlogit models to avoid the problem of separation. All variables are defined in Section 3.2. All variables are lagged with one period. We control for year and industry
effects in all models. Standard errors are reported in parentheses below coefficient estimates. Marginal effects are reported.
*, **, and *** denote statistical significance at the 10%, 5%, and 1% levels, respectively.

understanding of the variables' effects, we provide a graphical repre-

sentation of the relationship between firm performance and CEO dis-
missal in Figure 1, which shows a significant difference in the slopes
of boards with weak and strong social category faultline strength.
Moreover, using the Stata margins command, we find that a one stan-
dard deviation increase of social category faultline strength (from 0.47
to 0.59) decreases the probability of CEO dismissal at an industry-
adjusted ROA of 0.2 with 19.27%.
To test H2 and H3, we split the sample according to our specified
board contingencies. Table 4 indicates that the moderating role of
social category faultlines on the performance–CEO dismissal relation-
ship is significant for boards that do not engage in board evaluation
(dy/dx = 1.03, p < 0.01) but not for boards that do (dy/dx = 0.79,
p > 0.10). To compare the regression coefficients across the two
groups, we used the Stata command suest to test whether the coeffi-
cients from the two models are equal.15 The test indicates that the dif- F I G U R E 1 The moderating role of social category faultlines on
ference is significant at the 1% level (χ 2 = 6.84, p < 0.01). This result the likelihood of dismissal
 VANDEBEEK ET AL.

TABLE 4 Subsample analysis: Board evaluation

No board evaluation Board evaluation

Variables Model 1 Model 2 Model 3 Model 4 Model 5 Model 6 Model 7 Model 8 Model 9 Model 10 Model 11 Model 12 Model 13 Model 14

CEO tenure (log) 0.01 0.00 0.00 0.00 0.00 0.00 0.00 0.03* 0.06*** 0.07*** 0.02 0.03 0.03 0.03
(0.01) (0.01) (0.01) (0.02) (0.02) (0.02) (0.02) (0.02) (0.02) (0.02) (0.02) (0.03) (0.03) (0.03)
CEO ownership 0.11 0.10 0.10 0.05 0.05 0.03 0.02 0.10 0.14* 0.20** 0.12 0.14 0.13 0.13
(0.07) (0.07) (0.07) (0.09) (0.08) (0.08) (0.09) (0.08) (0.09) (0.10) (0.15) (0.17) (0.17) (0.18)
CEO age 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00
(0.00) (0.00) (0.00) (0.00) (0.00) (0.00) (0.00) (0.00) (0.00) (0.00) (0.00) (0.00) (0.00) (0.00)
CEO duality 0.05 0.06 0.05 0.05 0.08 0.07 0.07 0.07
(0.06) (0.06) (0.06) (0.06) (0.09) (0.09) (0.09) (0.09)
Family CEO 0.21** 0.20** 0.18** 0.18* 0.06 0.06 0.05 0.06
(0.10) (0.09) (0.09) (0.09) (0.09) (0.10) (0.10) (0.10)
Board committees 0.01 0.01
(0.02) (0.03)
Ind. Director ratio 0.02 0.03 0.02 0.03 0.03 0.02 0.00 0.37*** 0.51** 0.40** 0.23 0.22 0.21 0.21
(0.05) (0.05) (0.05) (0.08) (0.08) (0.08) (0.10) (0.13) (0.23) (0.17) (0.15) (0.17) (0.17) (0.18)
Block ownership 0.00 0.01 0.01 0.01 0.03 0.02 0.02 0.19*** 0.37*** 0.42*** 0.17 0.21 0.21 0.20
(0.05) (0.05) (0.05) (0.08) (0.08) (0.08) (0.08) (0.05) (0.14) (0.14) (0.11) (0.15) (0.16) (0.16)
Family business 0.03 0.03 0.02 0.00 0.00 0.01 0.01 0.04 0.04 0.00 0.02 0.01 0.02 0.02
(0.02) (0.02) (0.02) (0.03) (0.03) (0.03) (0.03) (0.03) (0.03) (0.03) (0.06) (0.06) (0.06) (0.06)
Board size 0.00 0.00 0.00 0.01 0.01 0.01 0.01 0.01 0.01** 0.02** 0.01 0.01 0.01 0.01
(0.00) (0.00) (0.00) (0.01) (0.01) (0.01) (0.01) (0.00) (0.01) (0.01) (0.01) (0.01) (0.01) (0.01)
Inform. Faultlines 0.02 0.02 0.01 0.01 0.02 0.00 0.01 0.46*** 0.68** 0.54** 0.37 0.36 0.35 0.34
(0.07) (0.07) (0.08) (0.09) (0.09) (0.09) (0.10) (0.18) (0.28) (0.22) (0.25) (0.31) (0.31) (0.29)
Industry adjusted ROA 0.04 .60*** 0.07* .60** .60** 0.08 .31 (0.34) 0.05 0.00 0.05
(0.03) (0.17) (0.04) (0.25) (0.25) (0.07) (0.13) (0.44) (0.45)
Social category 0.02 0.07 0.06 0.06 0.46 0.50 0.09 0.11 0.10
faultlines (SCF) (0.10) (0.09) (0.13) (0.13) (0.34) (0.33) (0.26) (0.29) (0.28)
Industry adjusted ROA 1.03*** 0.99** 0.99** 0.79 0.12 0.02
* SCF (0.31) (0.45) (0.46) (0.54) (0.79) (0.81)

(Continues)
11
 12 VANDEBEEK ET AL.

Note: In this table, we examine the relationship between social category faultlines and performance–CEO dismissal sensitivity in boards that do engage in evaluation versus boards who do not. We use data on Belgian
provides strong support for H2. In the firthlogit model, the interaction

firms listed on the Brussels Stock Exchange (Euronext Brussels) from 2006 to 2014. The dependent variable is a dummy variable equal to 1 in a given year if a company experiences a forced turnover of the CEO and

variables are defined in Section 3.2. All variables are lagged with one period. We control for year and industry effects in all models. Standard errors are reported in parentheses below coefficient estimates. We show
Model 14

0 otherwise. Models 1 to 3 and 8 to 10 are estimated using a pooled logit model with clustered error terms. Models 4 to 7 and 11 to 14 are estimated using firthlogit models to avoid the problem of separation. All
of social category faultlines and firm performance remained strong

0.43
234
Yes
Yes

70
and significant for boards that did not engage in board evaluation.
Table 5 provides evidence supporting H3, as the coefficient is
Model 13

more significant for boards that have more board committees

0.49
234
Yes
Yes

70
(dy/dx = 3.00, p < 0.001) than for boards that have fewer (dy/
dx = 0.57, p > 0.10). The difference between these two groups is signif-
icant at the 10% level (χ 2 = 3.27, p < 0.10). In the firthlogit model, the
Model 12

0.50
234
Yes
Yes

70

interaction of social category faultlines and firm performance remained

strong and significant for boards that had more board committees.
Board evaluation

To better illustrate the patterns revealed in Tables 4 and 5, we

Model 11

plot the relationship between firm performance, social category

0.53
234
Yes
Yes

70

the Wald chi-square test of Model 3 versus Model 10. Marginal effects are reported. Between-group coefficient difference on SCF * Ind. adj. ROA(χ 2 for SUR): 6.84***.
faultlines, and the probability of forced CEO dismissal by separating
the total sample based on the presence of evaluation and the number
Model 10

of board committees in Figure 2. Figure 2a shows a significant differ-

0.66
119
Yes
Yes

58

ence in the slopes of boards with different social category faultline

strengths within the subsample of boards that do not engage in board
evaluation. Figure 2b shows that slopes indicating the dismissal–
Model 9

0.65
119
Yes
Yes

58

performance relationship are almost parallel in boards, which do

engage in board evaluation. This suggests that the moderating role of
social category faultlines is negligible for the subsample of boards that
Model 8

do engage in board evaluation. This pattern is consistent with the pre-

0.60
119
Yes
Yes

58

diction of H2. Similarly, we can tell that slopes indicating the

dismissal–performance relationship exhibit much smaller differences
Model 7

between weak and strong social category faultline strength for boards
0.30
416
Yes
Yes

95

with fewer board committees (Figure 2c) than for boards with more
committees (Figure 2d). These results support H3.
Model 6

Lastly, to ensure the robustness of our findings, we reran the

0.33
416
Yes
Yes

95

analysis using the average industry-adjusted ROA of two previous

years. Table 6 provides the results of this robustness analysis for our
Model 5

main model, which confirms our findings, and with an even stronger
0.30
416
Yes
Yes

95

significant interaction effect (dy/dx = 0.71, p < 0.001). Tables 7 and 8

*, **, and *** denote statistical significance at the 10%, 5%, and 1% levels, respectively.

contain the robustness analysis for H2 and H3, respectively, and gen-
No board evaluation

Between-group coefficient difference on SCF * Ind. adj. ROA (χ 2 for SUR): 6.84***

erate results that are consistent with our main subsample analysis.
Model 4

0.27
416
Yes
Yes

95

Due to data limitations (i.e., a decrease in data observations due to

missing values), we were not able to run a robustness analysis using
market-based performance measures (such as Tobin's Q) for the sub-
Model 3

sample analysis (H2 and H3). However, we do find similar results for
0.20
401
Yes
Yes

91

our main model (H1). These additional results are available upon
request.
Model 2

0.14
401
Yes
Yes

91

5 | DI SCU SSION
Model 1

0.13
401
Yes
Yes

91

Using unique, hand-collected panel data containing individual informa-

tion on the board members of publicly traded Belgian firms, our study
(Continued)

investigates the impact of social category faultlines on the sensitivity

of CEO dismissal to firm performance. The study provides empirical
No. of observations
Industry dummies

and theoretical contributions to the growing body of faultline research

Year dummies

pseudo-R2
McFadden's

on boards of directors. In particular, we extend the board-specific

No. of firms
TABLE 4

Variables

faultline research (e.g., Van Peteghem et al., 2018) by using Belgian

data and focusing on the specific effect of social category faultlines. In
this way, we also respond to the call to consider the nature (or type)
 VANDEBEEK ET AL.

TABLE 5 Subsample analysis: Board committees

Fewer board committees More board committees

Variables Model 1 Model 2 Model 3 Model 4 Model 5 Model 6 Model 7 Model 8 Model 9 Model 10 Model 11 Model 12 Model 13 Model 14

CEO tenure (log) 0.00 0.00 0.00 0.01 0.00 0.00 0.00 0.01 0.01 0.04 0.00 0.01 0.04 0.04
(0.02) (0.02) (0.01) (0.01) (0.02) (0.02) (0.02) (0.03) (0.04) (0.03) (0.03) (0.03) (0.04) (0.03)
CEO ownership 0.13 0.13 0.12 0.08 0.11 0.12 0.12 0.15 0.17 0.20* 0.04 0.05 0.09 0.12
(0.09) (0.08) (0.09) (0.10) (0.10) (0.10) (0.10) (0.12) (0.13) (0.12) (0.14) (0.15) (0.14) (0.14)
CEO age 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.01 0.01** 0.00 0.00 0.01* 0.01*
(0.00) (0.00) (0.00) (0.00) (0.00) (0.00) (0.00) (0.00) (0.01) (0.00) (0.00) (0.00) (0.00) (0.00)
CEO duality 0.02 0.02 0.01 0.01 0.07 0.11 0.08 0.06
(0.05) (0.05) (0.05) (0.05) (0.13) (0.22) (0.16) (0.14)
Family CEO 0.15** 0.17** 0.17** 0.18** 0.09 0.07 0.07 0.09
(0.08) (0.07) (0.07) (0.07) (0.14) (0.15) (0.15) (0.15)
Board evaluation 0.03 0.04
(0.03) (0.05)
Ind0. Director ratio 0.04 0.04 0.04 0.02 0.01 0.00 0.02 0.06 0.06 0.14 0.03 0.04 0.11 0.10
(0.06) (0.06) (0.07) (0.08) (0.08) (0.08) (0.09) (0.23) (0.25) (0.30) (0.21) (0.21) (0.24) (0.24)
Block ownership 0.10 0.10 0.10 0.12 0.13 0.15 0.16* 0.05 0.08 0.12 0.02 0.07 0.09 0.11
(0.07) (0.06) (0.07) (0.09) (0.09) (0.09) (0.10) (0.13) (0.13) (0.11) (0.12) (0.13) (0.14) (0.14)
Family business 0.00 0.01 0.00 0.01 0.00 0.01 0.00 0.08 0.08 0.07 0.05 0.05 0.04 0.03
(0.03) (0.03) (0.03) (0.03) (0.03) (0.03) (0.03) (0.06) (0.05) (0.06) (0.07) (0.07) (0.07) (0.07)
Board size 0.01 0.00 0.00 0.01 0.01 0.01 0.01 0.00 0.01 0.00 0.00 0.01 0.00 0.00
(0.01) (0.00) (0.00) (0.01) (0.01) (0.01) (0.01) (0.01) (0.01) (0.01) (0.01) (0.01) (0.01) (0.01)
Inform0. faultlines 0.06 0.09 0.08 0.04 0.05 0.04 0.04 0.25 0.23 0.33 0.23 0.18 0.23 0.19
(0.09) (0.10) (0.11) (0.08) (0.08) (0.08) (0.08) (0.23) (0.25) (0.32) (0.30) (0.30) (0.34) (0.34)
Industry adjusted ROA 0.10* 0.42 0.10** 0.58** 0.56** 0.01 1.59*** 0.04 1.62*** 1.57***
(0.06) (0.26) (0.05) (0.27) (0.27) (0.06) (0.40) (0.09) (0.62) (0.58)
Social category faultlines 0.06 0.02 0.14 0.04 0.06 0.16 0.37** 0.20 0.45 0.46
(SCF) (0.10) (0.11) (0.11) (0.12) (0.12) (0.31) (0.18) (0.27) (0.30) (0.30)
Industry adjusted ROA * 0.57 0.81* 0.78* 3.00*** 3.01*** 2.92***
SCF (0.53) (0.44) (0.44) (0.74) (1.12) (1.05)

(Continues)
13
 14 VANDEBEEK ET AL.

of member differences in faultline measures (cf. Bezrukova

Note: In this table, we examine the relationship between social category faultlines and performance–CEO dismissal sensitivity in boards who have fewer board committees than average versus boards that have

avoid the problem of separation. All variables are defined in Section 3.2. All variables are lagged with one period. We control for year and industry effects in all models. Standard errors are reported in parentheses
a forced turnover of the CEO and 0 otherwise. Models 1 to 3 and 8 to 10 are estimated using a pooled logit model with clustered error terms. Models 4 to 7 and 11 to 14 are estimated using firthlogit models to
more. We use data on Belgian firms listed on the Brussels Stock Exchange (Euronext Brussels) from 2006 to 2014. The dependent variable is a dummy variable equal to 1 in a given year if a company experiences
Model 14
et al., 2009; Chung et al., 2015; Hutzschenreuter & Horstkotte, 2013;

below coefficient estimates. We show the Wald chi-square test of Model 3 versus Model 10. Marginal effects are reported. Between-group coefficient difference on SCF * Ind. adj. ROA(χ 2 for SUR): 3.27*.
0.39
198
Yes
Yes

45
Jehn et al., 2008; Jehn & Rupert, 2008). Our study shows that social
category faultlines are a vital dimension in explaining the capability of
Model 13

boards to dismiss a poorly performing CEO. By looking at faultlines at

0.43
198
Yes
Yes

45
the board level,16 we examine the impact of social category faultlines
on board-level decision-making, such as deciding on the dismissal of a
poorly performing CEO. Moreover, by focusing on social category
Model 12

0.36
198
Yes
Yes

faultlines in particular, we were able to build on the social identity per-

45
More board committees

spective in more detail, enabling us to uncover specific mechanisms of

social category faultlines and the influence of contextual factors.
Model 11

Social category faultlines can drive the existence of identity-based

0.36
198
Yes
Yes

45

subgroups, creating a divided board in which directors identify

strongly with their subgroup rather than with the collective board as a
Model 10

whole. Social category faultlines can elicit social categorization pro-

0.37
144
Yes
Yes

42

cesses (i.e., stereotypic perceptions of dissimilar directors and inter-

group biases). These processes can engender disruptive group
dynamics between different subgroups and negatively influence atti-
Model 9

0.25
144
Yes
Yes

42

tudes, collaboration, trust, and interactions between in- and outgroup

members (Hogg et al., 1990; Hogg & Terry, 2000).
With our research into corporate boards, we also add theoretical
Model 8

0.24
144
Yes
Yes

42

insights to some recent advancements in strategic leadership and

faultlines research, which has primarily focused on faultlines within
top management teams (Calabrò et al., 2020; Kavuşan et al., 2020;
Model 7

0.40
452
Yes
Yes

78

Richard et al., 2019) or the CEO-TMT interface (Georgakakis

et al., 2017; Georgakakis et al., 2019; Georgakakis & Buyl, 2020). As
board members act as strategic leaders (Heracleous, 1999), the board
Model 6

0.42
452
Yes
Yes

78

as a whole can function as a strategic leadership interface (which can

be defined as “the interdependent social situations in which the attri-
butes, aspirations, and/or activities of strategic leaders and/or salient
Model 5

0.41
452
Yes
Yes

74

stakeholders come into contact with and influence each other”; Sim-
Fewer board committees

sek et al., 2018, p. 283), in which the presence of social category

*, **, and *** denote statistical significance at the 10%, 5%, and 1% levels, respectively.

faultlines can impact strategic decision-making and shape collective

Model 4

outcomes.
0.35
452
Yes
Yes

74

Between-group coefficient difference on SCF * Ind. adj. ROA (χ 2 for SUR): 3.27*

Furthermore, our study sheds light on the contextual conditions

that can promote or weaken directors' identification with the board.
Model 3

We have investigated the impact of social category faultlines on

0.27
334
Yes
Yes

74

performance–CEO dismissal sensitivity on boards that engage in

board evaluation versus boards that do not. Boards that engage
in board evaluation increase directors' attention to the bigger picture,
Model 2

0.25
334
Yes
Yes

74

reminding board members of their common fiduciary responsibility.

Such feedback is necessary for board members to determine whether
their actions are sufficient to fulfill their board roles (Locke &
Model 1

Latham, 2002). We have found that the negative moderating role of

0.20
334
Yes
Yes

74

social category faultlines on the performance–CEO dismissal relation-

ship is not significant for boards that engage in board evaluation. For
(Continued)

boards that do not engage in board evaluation, however, there is a

McFadden's pseudo-R2

very strong weakening effect of social category faultlines on the nega-

No. of observations
Industry dummies

tive relationship between firm performance and the likelihood of CEO

Year dummies

dismissal. We have also found that the moderating effect of social cat-
No. of firms
TABLE 5

Variables

egory faultlines is stronger for firms with greater numbers of board

committees. The weakening effect of social category faultlines on the
performance–CEO dismissal relationship is greater and more
VANDEBEEK ET AL. 15

F I G U R E 2 Board contingencies and

the moderating role of social category
Faultlines on the likelihood of dismissal

TABLE 6 Robustness analysis for H1

Variable Model 1 Model 3 Model 3 Model 4 Model 5 Model 6 Model 7

CEO tenure (log) 0.01 (0.01) 0.00 (0.01) 0.00 (0.01) 0.01 (0.01) 0.00 (0.01) 0.00 (0.01) 0.00 (0.01)
CEO ownership 0.08 (0.06) 0.08 (0.06) 0.08 (0.05) 0.03 (0.06) 0.03 (0.06) 0.02 (0.06) 0.02 (0.06)
CEO age 0.00 (0.00) 0.00 (0.00) 0.00 (0.00) 0.00 (0.00) 0.00 (0.00) 0.00 (0.00) 0.00 (0.00)
CEO duality 0.02 (0.04) 0.03 (0.04) 0.02 (0.04) 0.03 (0.04)
Family CEO 0.15** (0.07) 0.16** (0.07) 0.14* (0.07) 0.14* (0.07)
Board evaluation 0.00 (0.02)
Board committees 0.00 (0.02)
Independent director ratio 0.04 (0.04) 0.03 (0.04) 0.05 (0.04) 0.04 (0.06) 0.04 (0.06) 0.05 (0.06) 0.05 (0.07)
Block ownership 0.04 (0.04) 0.04 (0.04) 0.03 (0.04) 0.04 (0.06) 0.04 (0.06) 0.04 (0.06) 0.04 (0.06)
Family business dummy 0.03 (0.02) 0.03 (0.02) 0.02 (0.02) 0.02 (0.02) 0.02 (0.02) 0.02 (0.02) 0.02 (0.02)
Board size 0.00 (0.00) 0.00 (0.00) 0.00 (0.00) 0.00 (0.00) 0.00 (0.00) 0.00 (0.00) 0.00 (0.00)
Informational Faultlines 0.06 (0.06) 0.06 (0.06) 0.05 (0.06) 0.05 (0.07) 0.06 (0.07) 0.06 (0.07) 0.06 (0.07)
2-year Ind0. Adj0. ROA 00.3 (0.03) 0.42*** (0.14) 0.05 (0.04) 0.41** (0.20) 0.41** (0.21)
Social category Faultlines (SCF) 0.01 (0.07) 0.06 (0.07) 0.04 (0.09) 0.03 (0.10) 0.03 (0.10)
2-year Ind. Adj. ROA * SCF 0.71*** (0.26) 0.66* (0.36) 0.67* (0.37)
Year dummies Yes Yes Yes Yes Yes Yes Yes
Industry dummies Yes Yes Yes Yes Yes Yes Yes
No. of observations 620 620 620 650 650 650 650
No. of firms 95 95 95 99 99 99 99
χ 2
51.05*** 55.88*** 74.05*** 7.66*** 7.91*** 6.36** 6.59**
McFadden's pseudo-R2 0.16 0.17 0.20 0.24 0.25 0.26 0.27

Note: In this table, we use a different performance measure to show robustness of the relationship between social category faultlines and performance–CEO
dismissal sensitivity. We use data on Belgian firms listed on the Brussels Stock Exchange (Euronext Brussels) from 2006 to 2014. The dependent variable is a
dummy variable equal to 1 in a given year if a company experiences a forced turnover of the CEO and 0 otherwise. The 2-year Ind. Adj. ROA was measured using
the average industry-adjusted firm performance of two previous years. Models 1 to 3 are estimated using a pooled logit model with clustered error terms. Model
4 to 7 are estimated using firthlogit models to avoid the problem of separation. All variables are lagged with one period. We control for year and industry effects
in all models. Standard errors are reported in parentheses below coefficient estimates. Marginal effects are reported.
*, **, and *** denote statistical significance at the 10%, 5%, and 1% levels, respectively.
 16

TABLE 7 Robustness analysis for H2

No board evaluation Board evaluation

Variables Model 1 Model 2 Model 3 Model 4 Model 5 Model 6 Model 7 Model 8 Model 9 Model 10 Model 11 Model 12 Model 13 Model 14
CEO tenure (log) 0.01 0.00 0.00 0.00 0.00 0.01 0.01 0.03* 0.05*** 0.05 0.02 0.03 0.03 0.03
(0.01) (0.01) (0.01) (0.02) (0.02) (0.02) (0.02) (0.02) (0.02) (0.04) (0.02) (0.03) (0.03) (0.03)
CEO ownership 0.11 0.10 0.10 0.05 0.05 0.03 0.02 0.10 0.12 0.15 0.12 0.12 0.13 0.13
(0.07) (0.07) (0.07) (0.09) (0.08) (0.08) (0.09) (0.08) (0.08) (0.27) (0.15) (0.17) (0.17) (0.17)
CEO age 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00
(0.00) (0.00) (0.00) (0.00) (0.00) (0.00) (0.00) (0.00) (0.00) (0.00) (0.00) (0.00) (0.00) (0.00)
CEO duality 0.05 0.05 0.05 0.05 0.08 0.07 0.06 0.06
(0.06) (0.05) (0.06) (0.06) (0.09) (0.09) (0.09) (0.09)
Family CEO 0.21** 0.21** 0.18** 0.18* 0.06 0.05 0.06 0.06
(0.10) (0.09) (0.09) (0.09) (0.09) (0.10) (0.10) (0.10)
Board committees 0.01 0.01
(0.02) (0.03)
Ind. Director ratio 0.02 0.03 0.00 0.03 0.03 0.00 0.02 0.37*** 0.55** 1.00 0.23 0.23 0.19 0.19
(0.05) (0.05) (0.05) (0.08) (0.08) (0.08) (0.10) (0.13) (0.22) (0.71) (0.15) (0.17) (0.15) (0.15)
Block ownership 0.00 0.01 0.01 0.01 0.02 0.02 0.02 0.19*** 0.37*** 0.36 0.17 0.20 0.21 0.20
(0.05) (0.05) (0.05) (0.08) (0.08) (0.08) (0.08) (0.05) (0.13) (0.26) (0.11) (0.14) (0.15) (0.15)
Family business 0.03 0.03 0.02 0.00 0.01 0.00 0.01 0.04 0.05** 0.14 0.02 0.02 0.01 0.02
(0.02) (0.02) (0.02) (0.03) (0.03) (0.03) (0.03) (0.03) (0.03) (0.10) (0.06) (0.06) (0.06) (0.06)
Board size 0.00 0.00 0.00 0.01 0.01 0.01 0.01 0.01 0.01** 1.43 0.37 0.37 0.32 0.30
(0.00) (0.00) (0.00) (0.01) (0.01) (0.01) (0.01) (0.00) 0.74*** (0.99) (0.25) (0.31) (0.26) (0.25)
(0.26)
2-year Ind. Adj. 0.07* 0.68*** 0.10* 0.70** 0.70** 0.05 1.27 0.01 0.15 0.12
ROA (0.04) (0.19) (0.05) (0.31) (0.31) (0.08) (0.88) (0.15) (0.37) (0.39)
Social category 0.02 0.11 0.04 0.09 0.09 0.49 0.62 0.08 0.08 0.07
Faultlines (SCF) (0.10) (0.10) (0.13) (0.14) (0.14) (0.33) (0.70) (0.27) (0.27) (0.27)
2-year Ind. Adj. 1.14*** 1.12** 1.10* 2.45 0.37 0.32
ROA * SCF (0.35) (0.56) (0.57) (1.80) (0.57) (0.59)
VANDEBEEK ET AL.
 VANDEBEEK ET AL. 17

significant. These results allow us to suggest that having greater num-

in parentheses below coefficient estimates. We show the Wald chi-square test of Model 3 versus Model 10. Marginal effects are reported. Between-group coefficient difference on SCF * Ind. adj. ROA
adjusted firm performance of two previous years. Models 1 to 3 and 8 to 10 are estimated using a pooled logit model with clustered error terms. Models 4 to 7 and 11 to 14 are estimated using firthlogit
models to avoid the problem of separation. All variables are defined in Section 3.2. All variables are lagged with one period. We control for year and industry effects in all models. Standard errors are reported
dependent variable is a dummy variable equal to 1 in a given year if a company experiences a forced turnover of the CEO and 0 otherwise. The 2-year Ind. Adj. ROA was measured using the average industry-
Note: In this table, we perform a robustness analysis with a different performance measure. We use data on Belgian firms listed on the Brussels Stock Exchange (Euronext Brussels) from 2006 to 2014. The
Model 14
bers of board committees may impair directors' identification with the

0.43
234
Yes
Yes

70
board as a whole. Collective identification with the board should be
made salient, as it is likely to shape director behaviors (Khurana &
Model 13

Pick, 2004; Withers, Corley, & Hillman, 2012) and can weaken dys-

0.43
234
Yes
Yes

70
functional group dynamics caused by social category faultlines. Our
results therefore add some caution to the usual governance recom-
mendation of installing board committees in the main board as a way
Model 12

0.46
234
Yes
Yes

to improve corporate governance (Spira & Bender, 2004). While previ-

70

ous research has identified benefits to having boards working in com-

Board evaluation

mittees, our study suggests that having too many committees may
Model 11

have unintended consequences. Despite their importance, these two

0.50
234
Yes
Yes

70

board contingencies (i.e., board evaluation and board committees)

have received relatively little attention in the corporate governance
Model 10

literature. While the advantages of a comprehensive board evaluation

0.68
119
Yes
Yes

58

system seem clear, board evaluations have not received much atten-
tion in either theory or practice (Minichilli et al., 2007). Furthermore,
little attention has been paid to the work of board committees and
Model 9

0.64

their impact on board functioning, partly due to the challenges of

119
Yes
Yes

58

studying boards at work (Spira & Bender, 2004). Our study sheds light
on the important role of these two board contingencies in shaping
directors' social identifications.
Model 8

0.60
119
Yes
Yes

58

Our findings also offer practical implications for corporate gover-

nance. Specifically, our findings suggest that the composition of the
board in terms of social category attributes can affect its capability to
Model 7

dismiss a poorly performing CEO. These findings may have overarch-

0.30
416
Yes
Yes

95

ing implications for director-selection criteria. Considering the exis-

tence of faultlines on their boards can help firms develop a more
Model 6

integrative perspective of director selection by taking into account

0.29
416
Yes
Yes

95

both the needs of the firm and the group dynamics on the board
(Withers, Hillman, & Cannella, 2012). Firms often seek directors with
Model 5

*, **, and *** denote statistical significance at the 10%, 5%, and 1% levels, respectively.

a certain profile in function of the firm's needs but should also analyze
0.27
416
Yes
Yes

95

how that profile matches with the profiles of those corporate direc-
Between-group coefficient difference on SCF * Ind. adj. ROA (χ 2 for SUR): 3.43*
No board evaluation

tors already present. Practitioners concerned with board composition

can benefit from an understanding of board processes related to
Model 4

0.24
416
Yes
Yes

95

group faultlines, as it clarifies the complexity of board design

(Forbes & Milliken, 1999). Nomination committees could take the
existence of subgroups into account while establishing relevant
Model 3

criteria in the director nomination and selection process. In addition,

0.20
401
Yes
Yes

91

our findings highlight several practical solutions that boards can imple-
ment to ameliorate the negative effects of faultlines. For example,
Model 2

board evaluations can allow directors to redirect their focus to the

0.15
401
Yes
Yes

91

board as a whole, reminding them of the superordinate goal

(i.e., acting in the best interests of the stakeholders). Awareness can
Model 1

be enhanced by providing practitioners with more knowledge on the

0.13
401
Yes
Yes

91

advantages of doing such evaluations and by giving directions on how

(Continued)

to carry them out. Lastly, while delegating specific tasks from the main
board to board committees has been strongly recommended as a suit-
No. of observations
Industry dummies

(χ 2 for SUR): 3.43*.

able mechanism for improving corporate governance because it can

Year dummies

pseudo-R2
McFadden's
No. of firms

harness the contribution of non-executive directors (Spira &

Variables
TABLE 7

Bender, 2004), firms should be made aware that establishing more

board committees can have a negative side effect due to the presence
of social category faultlines.
 18

TABLE 8 Robustness analysis for H3

Fewer board committees More board committees

Variables Model 1 Model 2 Model 3 Model 4 Model 5 Model 6 Model 7 Model 8 Model 9 Model 10 Model 11 Model 12 Model 13 Model 14
CEO tenure (log) 0.00 0.00 0.00 0.01 0.00 0.00 0.01 0.01 0.01 0.04 0.00 0.01 0.04 0.04
(0.02) (0.02) (0.01) (0.01) (0.02) (0.02) (0.02) (0.03) (0.04) (0.04) (0.03) (0.03) (0.03) (0.03)
CEO ownership 0.13 0.12 0.13 0.08 0.11 0.11 0.11 0.15 0.16 0.20* 0.04 0.05 0.08 0.10
(0.09) (0.08) (0.08) (0.10) (0.10) (0.10) (0.11) (0.12) (0.13) (0.11) (0.14) (0.15) (0.13) (0.14)
CEO age 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.01 0.01** 0.00 0.00 0.01* 0.01*
(0.00) (0.00) (0.00) (0.00) (0.00) (0.00) (0.00) (0.00) (0.01) (0.00) (0.00) (0.00) (0.00) (0.00)
CEO duality 0.02 0.01 0.01 0.00 0.07 0.11 0.03 0.02
(0.05) (0.05) (0.05) (0.05) (0.13) (0.22) (0.14) (0.12)
Family CEO 0.15** 0.17** 0.17** 0.17** 0.09 0.07 0.09 0.11
(0.08) (0.07) (0.07) (0.07) (0.14) (0.15) (0.14) (0.14)
Board evaluation 0.03 0.04
(0.03) (0.05)
Ind. Director ratio 0.04 0.04 0.04 0.02 0.00 0.00 0.02 0.06 0.06 0.08 0.03 0.04 0.06 0.05
(0.06) (0.07) (0.06) (0.08) (0.08) (0.09) (0.09) (0.23) (0.25) (0.26) (0.21) (0.21) (0.23) (0.23)
Block ownership 0.10 0.10 0.10 0.12 0.13 0.14 0.16 0.05 0.08 0.15 0.02 0.07 0.10 0.10
(0.07) (0.07) (0.06) (0.09) (0.09) (0.10) (0.10) (0.13) (0.13) (0.12) (0.12) (0.13) (0.14) (0.14)
Family business 0.00 0.01 0.01 0.01 0.00 0.00 0.00 0.08 0.08 0.07 0.05 0.05 0.03 0.02
(0.03) (0.03) (0.03) (0.03) (0.03) (0.03) (0.03) (0.06) (0.05) (0.06) (0.07) (0.07) (0.07) (0.07)
Board size 0.01 0.00 0.00 0.01 0.01 0.01 0.01 0.00 0.01 0.00 0.00 0.01 0.00 0.00
(0.01) (0.00) (0.00) (0.01) (0.01) (0.01) (0.01) (0.01) (0.01) (0.01) (0.01) (0.01) (0.01) (0.01)
Inform. Faultlines 0.06 0.09 0.07 0.04 0.05 0.04 0.04 0.25 0.22 0.30 0.23 0.19 0.18 0.14
(0.09) (0.11) (0.12) (0.08) (0.08) (0.09) (0.09) (0.23) (0.24) (0.27) (0.30) (0.30) (0.32) (0.32)
2-year Ind. Adj. ROA 0.08 0.50** 0.09 0.46 0.48 0.01 2.05*** 0.06 1.72*** 1.70***
(0.05) (0.21) (0.05) (0.30) (0.29) (0.11) (0.62) (0.12) (0.66) (0.64)
Social category 0.07 0.05 0.14 0.04 0.05 0.17 0.41** 0.21 0.44 0.44
Faultlines (SCF) (0.10) (0.12) (0.11) (0.13) (0.13) (0.28) (0.20) (0.27) (0.29) (0.28)
2-year Ind. Adj. ROA 0.71** 0.63 0.65 3.89*** 3.21** 3.19***
* SCF (0.36) (0.51) (0.50) (1.21) (1.30) (1.24)
VANDEBEEK ET AL.
 VANDEBEEK ET AL. 19

Our study also has some limitations, which suggest an agenda for

dependent variable is a dummy variable equal to 1 in a given year if a company experiences a forced turnover of the CEO and 0 otherwise. The 2-year Ind. Adj. ROA was measured using the average industry-

to avoid the problem of separation. All variables are defined in Section 3.2. All variables are lagged with one period. We control for year and industry effects in all models. Standard errors are reported
Note: In this table, we perform a robustness analysis with a different performance measure. We use data on Belgian firms listed on the Brussels Stock Exchange (Euronext Brussels) from 2006 to 2014. The

adjusted firm performance of two previous years. Models 1 to 3 and 8 to 10 are estimated using a pooled logit model with clustered error terms. Models 4 to 7 and 11 to 14 are estimated using firthlogit models
Model 14
future research. First, while focusing on a sample of listed firms in

0.38
198
Yes
Yes

45
Belgium from 2006 to 2014 provides us with a unique institutional
context that differs from the Anglo-Saxon institutional contexts on
which the CEO dismissal literature has typically focused, our findings
Model 13

0.37
198
may be most useful for countries that apply the same board system
Yes
Yes

45
Belgium uses (i.e., a one-tier Continental European system). While
our results on the relationship between faultlines and the likelihood
Model 12

of CEO dismissal following poor performance are consistent with

0.32
198
Yes
Yes

45
More board committees

the findings using US data, the large-shareholder control tradition

in Belgium allows for additional faultline formation related to family
ownership. The market-control, or Anglo-Saxon, system is charac-
Model 11

0.31
198
Yes
Yes

terized by diffuse ownership except for institutional investors

45

(Cuervo, 2002). Therefore, a comparative study of this research topic

that examines board systems with distinct ownership structures could
Model 10

advance our understanding of the role of faultlines on boards of direc-

0.38
144
Yes
Yes

42

tors. In addition, our study faced some data limitations regarding the
in parentheses below coefficient estimates. We show the Wald chi-square test of Model 3 versus Model 10. Marginal effects are reported.

use of market-based performance measures. While our current perfor-

mance measures have enough merit to test our hypotheses, future
Model 9

0.25
144
Yes
Yes

42

comparative studies could use market-based performance measures

to expand the results of the current study (e.g., by using a larger
database).
Model 8

0.24
144
Yes
Yes

42

Second, while we use well-studied measures of faultlines to study

the presence and extent of these faultlines, we do not capture the full
complex web of the social relationships among board members and
Model 7

0.38
452
Yes
Yes

the CEO. Further research should be done in the boardroom to exam-

78

ine how faultlines unfold on the board and if events unfold the way
theory suggests. Our tests provide initial evidence of these relation-
Model 6

0.37
452

ships and should stimulate future inquiry that considers different

Yes
Yes

78

research methods and even more nuanced or sophisticated measures

of these relationships. Micro-level studies on the internal dynamics of
Model 5

*, **, and *** denote statistical significance at the 10%, 5%, and 1% levels, respectively.

subgroup processes would constitute a valuable complement to our

0.36
452
Yes
Yes

74
Fewer board committees

study and enable us to better understand the deeper level psychologi-

Between-group coefficient difference on SCF * Ind. adj. ROA (χ 2 for SUR): 3.42*

cal factors underlying social category faultlines and identity-based

Model 4

subgroup formation.
0.32
452
Yes
Yes

74

Third, while we account for the effects of identity-based sub-

groups (formed by social category faultlines) and knowledge-based
subgroups (formed by informational faultlines), we do not account for
Model 3

0.25

resource-based subgroups (formed by variety faultlines). While

334
Yes
Yes

74

identity-based subgroups will mostly affect the board's role as an

effective internal control mechanism, resource-based subgroups might
Model 2

play a more important role for predicting resource consuming firm

0.23
334
Yes
Yes

74

outcomes such as innovation (Li & Atuahene-Gima, 2001). According

to Carton and Cummings (2012), a team may break into subgroups
Model 1

when one set of members seeks access to materials and supplies that
0.20
334
Yes
Yes

74

another set of members possesses. Such resource-based subgroups

(Continued)

are formed when key resources are constrained, and some members
McFadden's pseudo-

have more access to them than others. Future research could investi-
No. of observations
Industry dummies

gate the effect of resource-based faultlines on resources-consuming

Year dummies

No. of firms

activities. Another interesting research avenue is to investigate poten-

Variables
TABLE 8

tial scapegoating behavior of powerful CEOs. It could be the case that

R2

the (powerful) CEO takes advantage of the us-versus-them tensions

that exist in the board and scapegoats performance responsibility to
20 VANDEBEEK ET AL.

lower level senior managers (e.g., chief operating officer [COO] or ACKNOWLEDG MENTS
chief financial officer [CFO] dismissal after poor performance rather We would like to thank Prof. Till Talaulicar, co-editor-in-chief of Cor-
than CEO dismissal). Therefore, future research could take COO or porate Governance: An International Review, the assigned editor,
CFO dismissal into account. Prof. Eugene Kang, and the two anonymous reviewers for their
We hypothesized and empirically found that the negative rela- insightful and constructive comments and suggestions during the
tionship between firm performance and the likelihood of CEO dis- review process.
missal is weaker when boards experience social category faultlines.
Surprisingly, the relationship between firm performance and the likeli- DATA AVAILABILITY STAT EMEN T
hood of CEO dismissal can even become positive under strong social Data available on request due to privacy/ethical restrictions.
category faultlines. Although this finding has to be considered with
some caution as we only have a very limited number of CEO turnover OR CID
cases (only seven) when performance is strong, we believe this kind of Alana Vandebeek https://orcid.org/0000-0002-0550-9485
CEO turnover event should be explored in greater detail in future Wim Voordeckers https://orcid.org/0000-0001-7041-3933
research. We speculate that the status quo and undecidedness that Jolien Huybrechts https://orcid.org/0000-0001-6520-7641
characterizes boards with strong faultlines is not only reflected in the Frank Lambrechts https://orcid.org/0000-0003-0970-0372
execution of their monitoring role (e.g., failure to dismiss the CEO
when performance is poor) but also in the execution of their strategic NOTES
role (e.g., failure to take the necessary strategic change decisions 1
Belgian companies operate under a one-tier board system (i.e., both
when performance is good). Indeed, sustaining superior firm perfor- executive and non-executive members form one board), which is com-
mon in both Continental European (e.g., France, Italy, and Spain) and
mance becomes challenging for firms in today's increasingly turbulent
Anglo-Saxon systems (e.g., the United States, the United Kingdom,
environments (Luciano et al., 2020) and requires timely strategic Canada, and Australia).
changes. As strategic leadership requires the coordinated efforts of 2
Of the data, 8% was collected by contacting directors via e-mail, tele-
boards and top management teams, it is presumable that successful, phone, or personal letter.
high-performing CEOs (who aim to explore new opportunities and ini- 3
The total number of board members represents the number of board
tiate necessary strategic changes) will come into conflict with an members' data points; for each board, a faultline strength was
undecided board, creating a strong unworkable polarization between calculated.
4
CEO and board, which may ultimately lead to the dismissal of the A family firm was classified as such if at least 20% of the shares were
family owned and the board included members of the family. This defi-
CEO. We invite future studies to further scrutinize this proposed
nition is in line with commonly used definitions of public family firms
explanation. (Chakrabarty, 2009; La Porta et al., 1999; Villalonga & Amit, 2006).
Furthermore, while we consider two board contingencies that 5
The average CEO ownership can be explained by the Belgian institu-
influence the moderating role of social category faultlines on tional context. Due to the large number of family businesses in our
performance–CEO dismissal sensitivity, there may also be other fac- sample, there are a number of CEOs with a high level of ownership.
tors that can moderate the relationship. For example, measuring the However, there is also a large proportion of CEOs with a very small
ownership interest.
strength of a director's identification with the organization may be an
6
The size of boards in our sample is comparable to US sample studies
interesting avenue (Hillman et al., 2008; Melkumov et al., 2015). The
such as Van Peteghem et al. (2018), but the independent director ratio
more strongly directors identify with the organization, the greater is lower (42% vs. 76.5%) in US boards. This may be explained by the
the likelihood that they will act in the interests of the organization, high portion of family firms in Belgian data samples.
7
which may have an ameliorating effect on the disruptive effect of We followed the recommendations made by Meyer et al. (2014) for
social category faultlines. On the other hand, directors could also choosing the most appropriate faultline measure given our research
setting.
identify strongly with being a CEO. Most directors of publicly traded
8
While a negative ASW is a theoretical possibility, our board composi-
firms are current or former CEOs (Fich, 2005). As these directors may
tions did not return any negative ASW values.
strongly identify with being a CEO, they may exhibit empathy with 9
We matched firms based on their size measured as the number of
the CEO if a tough decision has to be made, such as CEO dismissal
employees. In Belgium, a business is considered small if it employs
(Hillman et al., 2008). A higher ratio of such directors may also fewer than 50 employees, medium sized if it employs fewer than
increase the possibility of an additional identity-based subgroup, 250 employees, and large otherwise.
10
based on their shared strength of identification as a CEO. This could The components of the interaction term are not mean centered. While
strengthen the effect of social category faultlines on performance– mean centering could facilitate interpretation in some contexts, it will
not change the substantive effects (Franzese & Kam, 2009).
CEO dismissal sensitivity.
11
Using the median generated similar results.
We hope our understanding of the antecedents of CEO dis-
12
Directorship type is divided into three categories: “executive director,”
missal may be enriched through further study of this domain by
“independent director,” and “affiliated director” (i.e., nonexecutive and
creating awareness of the influence of group faultlines on the
non-independent). To measure educational background, we categorize
capability of board members to dismiss a CEO following poor firm directors into four educational areas—sciences, economics, laws, and
performance. business—representing the discipline in which they received their
VANDEBEEK ET AL. 21

highest degree. We classify the education of directors who received Bezrukova, K., & Uparna, J. (2009). Group splits and culture shifts: A new
their degree in another area as “other.” We also classify each director map of the creativity terrain. In M. A. Neale, B. Mannix, & J. Goncalo
into one of four educational levels: high school, professional bachelor's (Eds.), Research on managing groups and teams (Vol. 12) (pp. 161–191).
degree, academic master's degree, and PhD. Tenure is measured as the Stamford, CT: JAI Press.
number of years that the director has served on the board. Blau, P. M. (1977). Inequality and heterogeneity: A primitive theory of social
13
We also performed additional analyses to check whether informational structure (Vol. 7). New York: Free Press.
faultlines moderate the main effect. As expected, we did not find any Boeker, W., & Goodstein, J. (1993). Performance and successor choice:
significant effect of informational faultlines in our specific research set- The moderating effects of governance and ownership. Academy of
ting because the performance information is readily available and visible Management Journal, 36(1), 172–186. https://doi.org/10.5465/
for all board members—which is the case for financial performance 256517
because this is a key quarterly financial metric provided to the board by Bouwen, R., & Fry, R. (1996). Facilitating group development: Interven-
the financial department. These results are available upon request. tions for a relational and contextual construction. In M. A. West (Ed.),
14
This test could be done only for the pooled logit models. Handbook of work group psychology (pp. 531–552). London, UK: John
15
Wiley & Sons Ltd.
The suest command cannot be used when comparing coefficients of
Brunello, G., Graziano, C., & Parigi, B. M. (2003). CEO turnover in insider-
firth logit models, because suest requires that the “predict command”
dominated boards: The Italian case. Journal of Banking & Finance, 27(6),
allows the score option (which is not the case for firthlogit models).
1027–1051. https://doi.org/10.1016/S0378-4266(02)00244-3
16
For our specific research setting, it is more relevant to investigate the Burgess, Z., & Tharenou, P. (2002). Women board directors: Characteris-
functioning of the board as a whole compared to investigating potential tics of the few. Journal of Business Ethics, 37(1), 39–49. https://doi.
faultlines on the nominating committee level. We expect that the nomi- org/10.1023/A:1014726001155
nating committee will not have a great influence on its own on the CEO Bushman, R., Dai, Z., & Wang, X. (2010). Risk and CEO turnover. Journal of
dismissal decision following poor performance. In line with corporate Financial Economics, 96(3), 381–398. https://doi.org/10.1016/j.jfineco.
governance codes around the world (e.g., the United Kingdom and the 2010.03.001
United States), the 2009 Belgian Corporate Governance Code states Calabrò, A., Santulli, R., Torchia, M., & Gallucci, C. (2020). Entrepreneurial
that the nomination committee is mainly responsible for establishing orientation and family firm performance: The moderating role of TMT
procedures for the appointment of directors and seeking candidates for identity-based and knowledge-based faultlines. Entrepreneurship The-
directorships, which are vacant and to submit them to the board ory and Practice, Forthcoming. https://doi.org/10.1177/
for approval. Therefore, it is not their main goal to make the CEO dis- 1042258720973997
missal decision following poor performance. This decision falls to the Cannella, A. A. Jr., Jones, C. D., & Withers, M. C. (2015). Family- versus
board as a whole (Schwartz-Ziv & Weisbach, 2013). lone-founder-controlled public corporations: Social identity theory and
board of directors. Academy of Management Journal, 58(2), 436–459.
https://doi.org/10.5465/amj.2012.0045
RE FE R ENC E S Carton, A. M., & Cummings, J. N. (2012). A theory of subgroups in work
Ashforth, B. E., & Mael, F. (1989). Social identity theory and the organiza- teams. Academy of Management Review, 37(3), 441–470. https://doi.
tion. Academy of Management Review, 14(1), 20–39. https://doi.org/ org/10.5465/amr.2009.0322
10.5465/amr.1989.4278999 Chakrabarty, S. (2009). The influence of national culture and institutional
Bainbridge, S. M. (2002). Director primacy: The means and ends of corpo- voids on family ownership of large firms: A country level empirical
rate governance. Northwestern University Law Review, 97(2), 547–606. study. Journal of International Management, 15(1), 32–45. https://doi.
https://doi.org/10.2139/ssrn.300860 org/10.1016/j.intman.2008.06.002
Barber, B. M., & Lyon, J. D. (1996). Detecting abnormal operating perfor- Cheng, S. (2008). Board size and the variability of corporate performance.
mance: The empirical power and specification of test statistics. Journal Journal of Financial Economics, 87(1), 157–176. https://doi.org/10.
of Financial Economics, 41(3), 359–399. https://doi.org/10.1016/ 1016/j.jfineco.2006.10.006
0304-405X(96)84701-5 Chung, Y., Liao, H., Jackson, S. E., Subramony, M., Colakoglu, S., & Jiang, Y.
Basco, R., Campopiano, G., Calabrò, A., & Kraus, S. (2019). They are not all (2015). Cracking but not breaking: Joint effects of faultline strength
the same! Investigating the effect of executive versus non- and diversity climate on loyal behavior. Academy of Management Jour-
executive family board members on firm performance. Journal of Small nal, 58(5), 1495–1515. https://doi.org/10.5465/amj.2011.0829
Business Management, 57(2), 637–357. https://doi.org/10.1111/jsbm. Collin, S. Y., & Ahlberg, J. (2012). Blood in the boardroom: Family relation-
12535 ships influencing the functions of the board. Journal of Family Business
Basco, R., & Voordeckers, W. (2015). The relationship between the board Strategy, 3(4), 207–219. https://doi.org/10.1016/j.jfbs.2012.11.001
of directors and firm performance in private family firms: A test of the Coveney, J. (2015). FIRTHLOGIT: Stata module to calculate bias reduction
demographic versus behavioral approach. Journal of Management & in logistic regression: Boston College Department of Economics.
Organization, 21(4), 1–25. https://doi.org/10.1017/jmo.2015.23 Cronin, M. A., Bezrukova, K., Weingart, L. R., & Tinsley, C. H. (2011). Sub-
Bezrukova, K., Jehn, K., Zanutto, E., & Thatcher, S. (2009). Do workgroup groups within a team: The role of cognitive and affective integration.
faultlines help or hurt? A moderated model of faultlines, team identifi- Journal of Organizational Behavior, 32(6), 831–849. https://doi.org/10.
cation, and group performance. Organization Science, 20(1), 35–50. 1002/job.707
https://doi.org/10.1287/orsc.1080.0379 Crucke, S., & Knockaert, M. (2016). When stakeholder representation
Bezrukova, K., Thatcher, S. M., & Jehn, K. A. (2007). Group heterogeneity leads to faultlines. A study of board service performance in social
and faultlines: Comparing alignment and dispersion theories of group enterprises. Journal of Management Studies, 53(5), 768–793. https://
composition. In K. J. Behfar & L. L. Thompson (Eds.), Conflict in organi- doi.org/10.1111/joms.12197
zational groups: New directions in theory and practice (pp. 57–92). Cuervo, A. (2002). Corporate governance mechanisms: A plea for less code
Evanston, IL: The Northwestern University Press. of good governance and more market control. Corporate Governance:
Bezrukova, K., Thatcher, S. M. B., Jehn, K. A., & Spell, C. S. (2012). The An International Review, 10(2), 84–93. https://doi.org/10.1111/1467-
effects of alignments: Examining group faultlines, organizational cul- 8683.00272
tures, and performance. Journal of Applied Psychology, 97(1), 77–92. Dah, M. A., Frye, M. B., & Hurst, M. (2014). Board changes and CEO turn-
https://doi.org/10.1037/a0023684 over: The unanticipated effects of the Sarbanes–Oxley Act. Journal of
22 VANDEBEEK ET AL.

Banking & Finance, 41, 97–108. https://doi.org/10.1016/j.jbankfin. evidence. Financial Management, 19(2), 60–75. https://www.jstor.org/
2014.01.006 stable/3665635
Daily, C. M., & Johnson, J. L. (1997). Sources of CEO power and firm Georgakakis, D., & Buyl, T. (2020). Guardians of the previous regime: Post-
financial performance: A longitudinal assessment. Journal of Manage- CEO succession factional subgroups and firm performance. Long Range
ment, 23(2), 97–117. https://doi.org/10.1016/S0149-2063(97) Planning, 53(3), 101971. https://doi.org/10.1016/j.lrp.2020.101971
90039-8 Georgakakis, D., Greve, P., & Ruigrok, W. (2017). Top management team
Deephouse, D. L., & Jaskiewicz, P. (2013). Do family firms have better rep- faultlines and firm performance: Examining the CEO-TMT interface.
utations than non-family firms? An integration of socioemotional The Leadership Quarterly, 28(6), 741–758. https://doi.org/10.1016/j.
wealth and social identity theories. Journal of Management Studies, 50 leaqua.2017.03.004
(3), 337–360. https://doi.org/10.1111/joms.12015 Georgakakis, D., Heyden, M. L., Oehmichen, J. D., & Ekanayake, U. I.
DeFond, M. L., & Park, C. W. (1999). The effect of competition on CEO (2019). Four decades of CEO–TMT interface research: A review
turnover. Journal of Accounting and Economics, 27(1), 35–56. https:// inspired by role theory. The Leadership Quarterly, Forthcoming,
doi.org/10.1016/S0165-4101(98)00044-5 101354. https://doi.org/10.1016/j.leaqua.2019.101354
Denis, D. J., & Denis, D. K. (1995). Performance changes following top Goyal, V. K., & Park, C. W. (2002). Board leadership structure and CEO
management dismissals. The Journal of Finance, 50(4), 1029–1057. turnover. Journal of Corporate Finance, 8(1), 49–66. https://doi.org/10.
https://doi.org/10.1111/j.1540-6261.1995.tb04049.x 1016/S0929-1199(01)00028-1
Donaldson, J. R., Malenko, N., & Piacentino, G. (2020). Deadlock on the Greene, W. H. (2007). Econometric analysis (6th ed.). Upper Saddle River,
board. The Review of Financial Studies, 33(10), 4445–4488. https://doi. NJ: Pearson Prentice Hall.
org/10.1093/rfs/hhaa006 Hair, J. F. J., Black, W. C., Babin, B. J., & Anderson, R. E. (2014). Multivariate
Durukan, B., Ozkan, S., & Dalkilic, F. (2012). CEO turnover and corporate data analysis (Vol. 7). Essex, UK: Prentice-Hall.
performance relationship in pre-and post-IFRS period: Evidence Haleblian, J., & Rajagopalan, N. (2006). A cognitive model of CEO dis-
from Turkey. Journal of Business Economics and Management, 13(3), missal: Understanding the influence of board perceptions, attributions
421–442. https://doi.org/10.3846/16111699.2011.620145 and efficacy beliefs. Journal of Management Studies, 43(5), 1009–1026.
Eisenberg, T., Sundgren, S., & Wells, M. T. (1998). Larger board size and https://doi.org/10.1111/j.1467-6486.2006.00627.x
decreasing firm value in small firms. Journal of Financial Economics, Harrison, D. A., & Klein, K. J. (2007). What's the difference? Diversity con-
48(1), 35–54. https://doi.org/10.1016/S0304-405X(98)00003-8 structs as separation, variety, or disparity in organizations. Academy of
Erhardt, N. L., Werbel, J. D., & Shrader, C. B. (2003). Board of director Management Review, 32(4), 1199–1228. https://doi.org/10.5465/amr.
diversity and firm financial performance. Corporate Governance: An 2007.26586096
International Review, 11(2), 102–111. https://doi.org/10.1111/1467- Harrison, J. R., Torres, D. L., & Kukalis, S. (1988). The changing of the
8683.00011 guard: Turnover and structural change in the top-management posi-
Faleye, O., Hoitash, R., & Hoitash, U. (2011). The costs of intense board tions. Administrative Science Quarterly, 33(2), 211–232. https://doi.
monitoring. Journal of Financial Economics, 101(1), 160–181. https:// org/10.2307/2393056
doi.org/10.1016/j.jfineco.2011.02.010 He, L., & Fang, J. (2016). CEO overpayment and dismissal: The role of attri-
Fama, E. F. (1980). Agency problems and the theory of the firm. Journal of bution and attention. Corporate Governance: An International Review,
Political Economy, 88(2), 288–307. https://doi.org/10.1086/260866 24(1), 24–41. https://doi.org/10.1111/corg.12129
Fama, E. F., & Jensen, M. C. (1983). Separation of ownership and control. Heinze, G., & Schemper, M. (2002). A solution to the problem of separa-
Journal of Law and Economics, 26(2), 301–326. https://doi.org/10. tion in logistic regression. Statistics in Medicine, 21(16), 2409–2419.
1086/467037 https://doi.org/10.1002/sim.1047
Farrell, K. A., & Whidbee, D. A. (2002). Monitoring by the financial press Heracleous, L. (1999). The board of directors as leaders of the organisa-
and forced CEO turnover. Journal of Banking & Finance, 26(12), 2249– tion. Corporate Governance: An International Review, 7(3), 256–265.
2276. https://doi.org/10.1016/S0378-4266(01)00183-2 https://doi.org/10.1111/1467-8683.00157
Fich, E. M. (2005). Are some outside directors better than others? Evi- Hillman, A. J., Nicholson, G., & Shropshire, C. (2008). Directors' multiple
dence from director appointments by Fortune 1000 firms. The Journal identities, identification, and board monitoring and resource provision.
of Business, 78(5), 1943–1972. https://doi.org/10.1086/431448 Organization Science, 19(3), 441–456. https://doi.org/10.1287/orsc.
Finkelstein, S., & D'aveni, R. A. (1994). CEO duality as a double-edged 1080.0355
sword: How boards of directors balance entrenchment avoidance Hinds, P. J., Neeley, T. B., & Cramton, C. D. (2014). Language as a lightning
and unity of command. Academy of Management Journal, 37(5), rod: Power contests, emotion regulation, and subgroup dynamics in
1079–1108. https://doi.org/10.5465/256667 global teams. Journal of International Business Studies, 45(5), 536–561.
Fiordelisi, F., & Ricci, O. (2014). Corporate culture and CEO turnover. Jour- https://doi.org/10.1057/jibs.2013.62
nal of Corporate Finance, 28, 66–82. https://doi.org/10.1016/j. Hoetker, G. (2007). The use of logit and probit models in strategic manage-
jcorpfin.2013.11.009 ment research: Critical issues. Strategic Management Journal, 28(4),
Firth, D. (1993). Bias reduction of maximum likelihood estimates. Bio- 331–343. https://doi.org/10.1002/smj.582
metrika, 80(1), 27–38. https://doi.org/10.2307/2336755 Hogg, M. A., & Terry, D. J. (2000). Social identity and self-categorization
Forbes, D. P., & Milliken, F. J. (1999). Cognition and corporate governance: processes in organizational contexts. Academy of Management Review,
Understanding boards of directors as strategic decision-making 25(1), 121–140. https://doi.org/10.5465/amr.2000.2791606
groups. Academy of Management Review, 24(3), 489–505. https://doi. Hogg, M. A., Turner, J. C., & Davidson, B. (1990). Polarized norms and
org/10.5465/amr.1999.2202133 social frames of reference: A test of the self-categorization theory of
Franzese, R. J., & Kam, C. (2009). Modeling and interpreting interactive group polarization. Basic and Applied Social Psychology, 11(1), 77–100.
hypotheses in regression analysis. Michigan: University of Michigan https://doi.org/10.1207/s15324834basp1101_6
Press. Huson, M. R., Parrino, R., & Starks, L. T. (2001). Internal monitoring mecha-
Fredrickson, J. W., Hambrick, D. C., & Baumrin, S. (1988). A model of CEO nisms and CEO turnover: A long-term perspective. The Journal of
dismissal. Academy of Management Review, 13(2), 255–270. https:// Finance, 56(6), 2265–2297. https://doi.org/10.1111/0022-1082.
doi.org/10.5465/amr.1988.4306882 00405
Furtado, E. P., & Karan, V. (1990). Causes, consequences, and shareholder Hutzschenreuter, T., & Horstkotte, J. (2013). Performance effects of top
wealth effects of management turnover: A review of the empirical management team demographic faultlines in the process of product
VANDEBEEK ET AL. 23

diversification. Strategic Management Journal, 34(6), 704–726. https:// Management Review, 23(2), 325–340. https://doi.org/10.5465/amr.
doi.org/10.1002/smj.2035 1998.533229
Jehn, K. A., & Bezrukova, K. (2010). The faultline activation process Lau, D. C., & Murnighan, J. K. (2005). Interactions within groups and sub-
and the effects of activated faultlines on coalition formation, groups: The effects of demographic faultlines. Academy of Manage-
conflict, and group outcomes. Organizational Behavior and Human Deci- ment Journal, 48(4), 645–659. https://doi.org/10.5465/amj.2005.
sion Processes, 112(1), 24–42. https://doi.org/10.1016/j.obhdp.2009. 17843943
11.008 Lau, J., Sinnadurai, P., & Wright, S. (2009). Corporate governance and chief
Jehn, K. A., Bezrukova, K., & Thatcher, S. (2008). Conflict, diversity, and executive officer dismissal following poor performance: Australian evi-
faultlines in workgroups. In C. K. W. D. Dreu & M. J. Gelfand (Eds.), dence. Accounting and Finance, 49(1), 161–182. https://doi.org/10.
The organizational frontiers series. The psychology of conflict and conflict 1111/j.1467-629X.2008.00278.x
management in organizations (pp. 179–210). Taylor & Francis Lausten, M. (2002). CEO turnover, firm performance and corporate gover-
Group/Lawrence Erlbaum Associates nance: Empirical evidence on Danish firms. International Journal of
Jehn, K. A., & Rupert, J. (2008). Group faultlines and team learning: How Industrial Organization, 20(3), 391–414. https://doi.org/10.1016/
to benefit from different perspectives. In M. London & V. Sessa (Eds.), S0167-7187(00)00083-7
Work group learning: Understanding, improving & assessing how groups Lel, U., & Miller, D. P. (2008). International cross-listing, firm performance,
learn in organizations (pp. 121–149). Mahwah, NJ: Lawrence Erlbaum and top management turnover: A test of the bonding hypothesis. The
Asssociates. Journal of Finance, 63(4), 1897–1937. https://doi.org/10.1111/j.1540-
Jensen, M. C., & Meckling, W. H. (1976). Theory of the firm: Managerial 6261.2008.01377.x
behavior, agency costs and ownership structure. Journal of Financial Li, H., & Atuahene-Gima, K. (2001). Product innovation strategy and the
Economics, 3(4), 305–360. https://doi.org/10.1016/0304-405X(76) performance of new technology ventures in China. Academy of Man-
90026-X agement Journal, 44(6), 1123–1134. https://doi.org/10.5465/
Johnson, J. L., Ellstrand, A. E., & Daily, C. M. (1996). Boards of directors: A 3069392
review and research agenda. Journal of Management, 22(3), 409–438. Li, J., & Hambrick, D. C. (2005). Factional groups: A new vantage on demo-
https://doi.org/10.1016/S0149-2063(96)90031-8 graphic faultlines, conflict, and disintegration in work teams. Academy
Kaczmarek, S., Kimino, S., & Pye, A. (2012a). Antecedents of board compo- of Management Journal, 48(5), 794–813. https://doi.org/10.5465/amj.
sition: The role of nomination committees. Corporate Governance: An 2005.18803923
International Review, 20(5), 474–489. https://doi.org/10.1111/j.1467- Locke, E. A., & Latham, G. P. (2002). Building a practically useful theory of
8683.2012.00913.x goal setting and task motivation: A 35-year odyssey. American Psychol-
Kaczmarek, S., Kimino, S., & Pye, A. (2012b). Board task-related faultlines ogist, 57(9), 705–717. https://doi.org/10.1037/0003-066X.57.9.705
and firm performance: A decade of evidence. Corporate Governance: Long, T. (2006). This year's model: Influences on board and director evalu-
An International Review, 20(3), 337–351. https://doi.org/10.1111/j. ation. Corporate Governance: An International Review, 14(6), 547–557.
1467-8683.2011.00895.x https://doi.org/10.1111/j.1467-8683.2006.00529.x
Kavuşan, K., Ateş, N. Y., & Nadolska, A. (2020). Acquisition target Luciano, M. M., Nahrgang, J. D., & Shropshire, C. (2020). Strategic leader-
selection and technological relatedness: The moderating role of ship systems: Viewing top management teams and boards of directors
Top Management Team demographic faultlines. Strategic Organization, from a multiteam systems perspective. Academy of Management
Forthcoming, 147612702091932. https://doi.org/10.1177/147612 Review, 45(3), 675–701. https://doi.org/10.5465/amr.2017.0485
7020919329 Melkumov, D., Breit, E., & Khoreva, V. (2015). Directors' social identifica-
Kesner, I. F., & Dalton, D. R. (1994). Top management turnover and CEO tions and board tasks: Evidence from Finland. Corporate Governance:
succession: An investigation of the effects of turnover on perfor- An International Review, 23(1), 42–59. https://doi.org/10.1111/corg.
mance. Journal of Management Studies, 31(5), 701–713. https://doi. 12088
org/10.1111/j.1467-6486.1994.tb00635.x Merchant, K. A., & Pick, K. (2010). Blind spots, biases, and other pathologies
Khurana, R., & Pick, K. (2004). The social nature of boards. Brooklyn Law in the boardroom. New York, NY: Business Expert Press. https://doi.
Review, 70(4), 1259–1285. https://brooklynworks.brooklaw.edu/blr/ org/10.4128/9781606490716
vol70/iss4/6 Meyer, B., & Glenz, A. (2013). Team faultline measures: A computational
Kiel, G. C., & Nicholson, G. J. (2005). Evaluating boards and directors. Cor- comparison and a new approach to multiple subgroups. Organizational
porate Governance: An International Review, 13(5), 613–631. https:// Research Methods, 16(3), 393–424. https://doi.org/10.1177/
doi.org/10.1111/j.1467-8683.2005.00455.x 2F1094428113484970
Kim, S. M. (2003). The provisional director remedy for corporate deadlock: Meyer, B., Glenz, A., Antino, M., Rico, R., & Gonz alez-Roma, V. (2014).
A proposed model statute. Washington and Lee Law Review, 60, Faultlines and subgroups: A meta-review and measurement guide.
111–185. https://scholarlycommons.law.wlu.edu/wlulr/vol60/iss1/4 Small Group Research, 45(6), 633–670. https://doi.org/10.1177/
Kunze, F., & Bruch, H. (2010). Age-based faultlines and perceived produc- 2F1046496414552195
tive energy: The moderation of transformational leadership. Small Milliken, F. J., & Martins, L. L. (1996). Searching for common threads:
Group Research, 41(5), 593–620. https://doi.org/10.1177/ Understanding the multiple effects of diversity in organizational
2F1046496410366307 groups. Academy of Management Review, 21(2), 402–433. https://doi.
La Porta, R., Lopez-de-Silanes, F., & Shleifer, A. (1999). Corporate owner- org/10.5465/amr.1996.9605060217
ship around the world. The Journal of Finance, 54(2), 471–517. https:// Minichilli, A., Gabrielsson, J., & Huse, M. (2007). Board evaluations: Making
doi.org/10.1111/0022-1082.00115 a fit between the purpose and the system. Corporate Governance: An
La Porta, R., Lopez-de-Silanes, F., Shleifer, A., & Vishny, R. W. (1998). Law International Review, 15(4), 609–622. https://doi.org/10.1111/j.1467-
and finance. Journal of Political Economy, 106(6), 1113–1155. https:// 8683.2007.00591.x
doi.org/10.1086/250042 Mobbs, S. (2013). CEOs under fire: The effects of competition from inside
Lan, L. L., & Heracleous, L. (2010). Rethinking agency theory: The view directors on forced CEO turnover and CEO compensation. Journal of
from law. Academy of Management Review, 35(2), 294–314. https:// Financial and Quantitative Analysis, 48(3), 669–698. https://doi.org/10.
doi.org/10.5465/amr.35.2.zok294 1017/S0022109013000318
Lau, D. C., & Murnighan, J. K. (1998). Demographic diversity and faultlines: Molleman, E. (2005). Diversity in demographic characteristics, abilities and
The compositional dynamics of organizational groups. Academy of personality traits: Do faultlines affect team functioning? Group
24 VANDEBEEK ET AL.

Decision and Negotiation, 14(3), 173–193. https://doi.org/10.1007/ Thatcher, S., Jehn, K., & Zanutto, E. (2003). Cracks in diversity research:
s10726-005-6490-7 The effects of diversity faultlines on conflict and performance. Group
Nederveen Pieterse, A., Van Knippenberg, D., & Van Dierendonck, D. Decision and Negotiation, 12(3), 217–241. https://doi.org/10.1023/A:
(2013). Cultural diversity and team performance: The role of team 1023325406946
member goal orientation. Academy of Management Journal, 56(3), Thatcher, S., & Patel, P. (2012). Group faultlines: A review, integration, and
782–804. https://doi.org/10.5465/amj.2010.0992 guide to future research. Journal of Management, 38(4), 969–1009.
Ocasio, W. (1994). Political dynamics and the circulation of power: CEO https://doi.org/10.1177/2F0149206311426187
succession in U.S. industrial corporations, 1960–1990. Administrative Tuggle, C. S., Schnatterly, K., & Johnson, R. A. (2010). Attention patterns in
Science Quarterly, 39(2), 285–312. https://doi.org/10.2307/2393237 the boardroom: How board composition and processes affect discus-
Oehmichen, J., Heyden, M. L., Georgakakis, D., & Volberda, H. W. (2017). sion of entrepreneurial issues. Academy of Management Journal, 53(3),
Boards of directors and organizational ambidexterity in knowledge- 550–571. https://doi.org/10.5465/amj.2010.51468687
intensive firms. The International Journal of Human Resource Manage- Tushman, M. L., & Rosenkopf, L. (1996). Executive succession, strategic
ment, 28(2), 283–306. https://doi.org/10.1080/09585192.2016. reorientation and performance growth: A longitudinal study in the US
1244904 cement industry. Management Science, 42(7), 939–953. https://doi.
Polzer, J. T., Crisp, C. B., Jarvenpaa, S. L., & Kim, J. W. (2006). Extending org/10.1287/mnsc.42.7.939
the faultline model to geographically dispersed teams: How colocated Uhlaner, L., Wright, M., & Huse, M. (2007). Private firms and corporate
subgroups can impair group functioning. Academy of Management Jour- governance: An integrated economic and management perspective.
nal, 49(4), 679–692. https://doi.org/10.5465/amj.2006.22083024 Small Business Economics, 29(3), 225–241. https://doi.org/10.1007/
Porck, J. P., van Knippenberg, D., Tarakci, M., Ateş, N. Y., Groenen, P. J., & s11187-006-9032-z
de Haas, M. (2018). Do group and organizational identification help or Van der Walt, N., & Ingley, C. (2003). Board dynamics and the influence of
hurt intergroup strategic consensus? Journal of Management, 46(2), professional background, gender and ethnic diversity of directors. Cor-
234–260. https://doi.org/10.1177/0149206318788434 porate Governance: An International Review, 11(3), 218–234. https://
Puffer, S. M., & Weintrop, J. B. (1991). Corporate performance and CEO doi.org/10.1111/1467-8683.00320
turnover: The role of performance expectations. Administrative Science van Knippenberg, D., Dawson, J. F., West, M. A., & Homan, A. C. (2011).
Quarterly, 36(1), 1–19. https://doi.org/10.2307/2393427 Diversity faultlines, shared objectives, and top management team per-
Rasmussen, J. (2015). Do board evaluations measure board effectiveness? formance. Human Relations, 64(3), 307–336. https://doi.org/10.1177/
International Studies of Management and Organization, 45(1), 80–98. 0018726710378384
https://doi.org/10.1080/00208825.2015.1005999 Van Peteghem, M., Bruynseels, L., & Gaeremynck, A. (2018). Beyond diver-
Richard, O. C., Wu, J., Markoczy, L. A., & Chung, Y. (2019). Top manage- sity: A tale of faultlines and frictions in the board of directors. The
ment team demographic-faultline strength and strategic change: What Accounting Review, 93(2), 339–367. https://doi.org/10.2308/accr-
role does environmental dynamism play? Strategic Management Jour- 51818
nal, 40(6), 987–1009. https://doi.org/10.1002/smj.3009 Vandebeek, A., Voordeckers, W., Lambrechts, F., & Huybrechts, J. (2016).
Rink, F. A., & Jehn, K. A. (2010). How identity processes affect faultline Board role performance and faultlines in family firms: The moderating
perceptions and the functioning of diverse teams. In R. J. Crisp (Ed.), role of formal board evaluation. Journal of Family Business Strategy,
Psychology of social and cultural diversity (pp. 281–296). Chichester, 7(4), 249–259. https://doi.org/10.1016/j.jfbs.2016.10.002
UK: John Wiley & Sons. https://doi.org/10.1002/9781444325447. Veltrop, D. B., Hermes, N., Postma, T. J. B. M., & Haan, J. (2015). A tale of
ch12 two factions: Why and when factional demographic faultlines hurt
Rupert, J., Blomme, R. J., Dragt, M. J., & Jehn, K. (2016). Being different, board performance. Corporate Governance: An International Review,
but close: How and when faultlines enhance team learning. European 23(2), 145–160. https://doi.org/10.1111/corg.12098
Management Review, 13(4), 275–290. https://doi.org/10.1111/emre. Villalonga, B., & Amit, R. (2006). How do family ownership, control and
12083 management affect firm value? Journal of Financial Economics, 80(2),
Ryan, H. E. Jr., & Wiggins, R. A. III (2004). Who is in whose pocket? Direc- 385–417. https://doi.org/10.1016/j.jfineco.2004.12.005
tor compensation, board independence, and barriers to effective moni- Visintin, F., Pittino, D., & Minichilli, A. (2017). Financial performance and
toring. Journal of Financial Economics, 73(3), 497–524. https://doi.org/ non-family CEO turnover in private family firms under different
10.1016/j.jfineco.2003.11.002 conditions of ownership and governance. Corporate Governance: An
Schwartz-Ziv, M., & Weisbach, M. S. (2013). What do boards really do? International Review, 25(5), 312–337. https://doi.org/10.1111/corg.
Evidence from minutes of board meetings. Journal of Financial Econom- 12201
ics, 108(2), 349–366. https://doi.org/10.1016/j.jfineco.2012.04.011 Voordeckers, W., Van Gils, A., & Van den Heuvel, J. (2007). Board compo-
Siciliano, J. I. (1996). The relationship of board member diversity to organi- sition in small and medium-sized family firms. Journal of Small Business
zational performance. Journal of Business Ethics, 15(12), 1313–1320. Management, 45(1), 137–156. https://doi.org/10.1111/j.1540-627X.
https://doi.org/10.1007/BF00411816 2007.00204.x
Simsek, Z., Heavey, C., & Fox, B. C. (2018). Interfaces of strategic leaders: Walsh, J. P., & Seward, J. K. (1990). On the efficiency of internal and exter-
A conceptual framework, review, and research agenda. Journal of nal corporate control mechanisms. Academy of Management Review,
Management, 44(1), 280–324. https://doi.org/10.1177/014920631 15(3), 421–458. https://doi.org/10.5465/amr.1990.4308826
7739108 Weber, L., & Wiersema, M. (2017). Dismissing a tarnished CEO? Psycho-
Spira, L. F., & Bender, R. (2004). Compare and contrast: Perspectives on logical mechanisms and unconscious biases in the board's evaluation.
board committees. Corporate Governance: An International Review, 12 California Management Review, 59(3), 22–41. https://doi.org/10.1177/
(4), 489–499. https://doi.org/10.1111/j.1467-8683.2004.00389.x 2F0008125617712257
Sundaramurthy, C., & Kreiner, G. E. (2008). Governing by managing iden- Weisbach, M. S. (1988). Outside directors and CEO turnover. Journal of
tity boundaries: The case of family businesses. Entrepreneurship Theory Financial Economics, 20, 431–460. https://doi.org/10.1016/0304-
and Practice, 32(3), 415–436. https://doi.org/10.1111/j.1540-6520. 405X(88)90053-0
2008.00234.x West, M. A., Garrod, S., & Carletta, J. (1997). Group decision-making and
Tajfel, H., & Turner, J. C. (2004). The social identity theory of intergroup effectiveness: Unexplored boundaries. In C. L. Cooper & S. E. Jackson
behavior. In S. Worchel & W. G. Austin (Eds.), The psychology of inter- (Eds.), Creating tomorrow's organizations (pp. 293–317). Chichester,
group relations (pp. 7–24). Chicago, IL: Nelson-Hall. UK: John Wiley & Sons, Ltd.
VANDEBEEK ET AL. 25

Wiersema, M. F., & Zhang, Y. (2011). CEO dismissal: The role of invest-
ment analysts. Strategic Management Journal, 32(11), 1161–1182. Corporate Governance: an International Review and Family Business
https://doi.org/10.1002/smj.932 Review. He is also author of several family business textbooks.
Withers, M. C., Corley, K. G., & Hillman, A. J. (2012). Stay or leave: Direc-
tor identities and voluntary exit from the board during organizational Jolien Huybrechts (PhD in Business Economics, Hasselt Univer-
crisis. Organization Science, 23(3), 835–850. https://doi.org/10.1287/ sity) is Associate Professor at Maastricht University, School of
orsc.1110.0660 Business and Economics, Department of Organization, Strategy &
Withers, M. C., Hillman, A. J., & Cannella, A. A. Jr. (2012).
Entrepreneurship. Her work focuses on family firms, nonfamily
A multidisciplinary review of the director selection literature. Journal
of Management, 38(1), 243–277. https://doi.org/10.1177/2F0149206 CEOs, corporate governance, entrepreneurship, and risk-taking.
311428671 Jolien has published in journals such as Family Business Review,
Zattoni, A., Gnan, L., & Huse, M. (2015). Does family involvement influ- Entrepreneurship Theory and Practice, Journal of Small Business
ence firm performance? Exploring the mediating effects of board pro-
Management, Academy of Management Learning & Education, Jour-
cesses and tasks. Journal of Management, 41(4), 1214–1243. https://
doi.org/10.1177/2F0149206312463936 nal of Family Business Strategy, Entrepreneurship and Regional
Zorn, C. (2005). A solution to separation in binary response models. Politi- Development, and European Management Review. She is a member
cal Analysis, 13(2), 157–170. https://doi.org/10.1093/pan/mpi009 of the Editorial Review Board of Journal of Family Business Strat-
egy and Editorial Board member of Entrepreneurship Research
Journal.
AUTHOR BIOGRAPHI ES
Frank Lambrechts is a professor and the associate dean of the
Alana Vandebeek is a Postdoctoral Research Fellow at KU Leuven Faculty of Business Economics at Hasselt University (Belgium). He
(Belgium), Faculty of Economics and Business, the department of holds a PhD in Organizational Psychology and a PhD in Business
Work and Organisation Studies. Her research focuses on corpo- Economics. At the Research Center for Entrepreneurship & Family
rate governance with a strong emphasis on group dynamics and Firms, his research focuses on collaboration dynamics, systemic
diversity on boards. She holds a PhD in Business Economics from learning, strength-based change, and family businesses. Frank has
Hasselt University (Belgium). Alana has participated in several published in journals such as Academy of Management Learning &
international conferences and has published in Journal of Family Education, Journal of Business Ethics, Family Business Review, Jour-
Business Strategy. nal of Family Business Strategy, and Journal of Management Inquiry.
He also serves as an associate editor of the Journal of Family Busi-
Wim Voordeckers is a Professor of Entrepreneurial Finance and
ness Strategy.
Family Firm Governance at the Research Center of Entrepreneur-
ship and Family Firms (RCEF) at Hasselt University in Belgium. His
primary research interests include financing decisions, leadership,
corporate governance, and board and TMT behavior in relation-
How to cite this article: Vandebeek A, Voordeckers W,
ship with strategic change in family firms. He has published arti-
Huybrechts J, Lambrechts F. Corporate performance and CEO
cles in several academic journals including Journal of Management
dismissal: The role of social category faultlines. Corp Govern Int
Studies, Journal of Banking & Finance, Family Business Review, Jour-
Rev. 2021;1–25. https://doi.org/10.1111/corg.12376
nal of Business Ethics, and Small Business Economics. He serves on
the editorial review board of Entrepreneurship Theory & Practice,