UNITED STATES
SECURITIES AND EXCHANGE COMMISSION
Washington, D.C. 20549
SCHEDULE 14A
(Rule 14a-101)
Proxy Statement Pursuant to Section 14(a) of the Securities
Exchange Act of 1934 (Amendment No. )
|
Filed by the Registrant [X]
|
|
Filed by a Party other than the Registrant [ ]
|
|
Check the appropriate box:
|
|
[ ] |
Preliminary Proxy Statement
|
|
[ ]
|
Confidential, for Use of the Commission Only (as permitted by Rule 14a-6(e)(2))
|
|
[ ] |
Definitive Proxy Statement
|
|
[X]
|
Definitive Additional Materials
|
|
[ ]
|
Soliciting Material Pursuant to §240.14a-12
|
|
Honeywell International Inc.
|
|
(Name of Registrant as Specified In Its Charter)
|
|
|
|
|
|
(Name of Person(s) Filing Proxy Statement, if other than the Registrant)
|
Payment of Filing Fee (Check the appropriate box):
| [X] |
No fee required. |
| |
|
| [ ] |
Fee computed on
table below per Exchange Act Rules 14a-6(i)(1) and 0-11. |
| |
|
|
| |
1) |
Title of each class of securities to which transaction applies: |
| |
|
|
| |
2) |
Aggregate number of securities to which transaction applies: |
| |
|
|
| |
3) |
Per unit price or other underlying
value of transaction computed pursuant to Exchange Act Rule 0-11 (set forth
the amount on which the filing fee is calculated and state how it was determined): |
| |
|
|
| |
4) |
Proposed maximum aggregate value of transaction: |
| |
|
|
| |
5) |
Total fee paid: |
| |
|
| |
|
| [ ] |
Fee previously paid with preliminary
materials. |
| |
|
| [ ] |
Check box if any part of the fee is
offset as provided by Exchange Act Rule 0-11(a)(2) and identify the filing
for which the offsetting fee was paid previously. Identify the previous
filing by registration statement number, or the Form or Schedule and the
date of its filing. |
| |
|
| |
1) |
Amount Previously Paid: |
| |
|
|
| |
2) |
Form, Schedule or Registration Statement No. |
| |
|
| |
|
|
| |
3) |
Filing Party: |
| |
|
|
| |
4) |
Date Filed: |
| |
|
Academy of Management Journal
APPRENTICE, DEPARTURE, AND DEMOTION: AN
EXAMINATION OF THE THREE TYPES OF CEO-BOARD CHAIR
SEPARATION
|
|
|
|
|
|
Journal: |
|
|
Academy of
Management Journal |
|
|
|
|
|
|
Manuscript ID: |
|
|
AMJ-2011-0121.R4 |
|
|
|
|
|
|
Manuscript Type: |
|
|
Revision |
|
|
|
|
|
Keyword: |
|
|
Board of directors
< Upper Echelons/Corporate Governance < Business Policy and Strategy
< Topic Areas, Upper echelons/corporate governance (General) < Upper
Echelons/Corporate Governance < Business Policy and Strategy < Topic
Areas, Agency theory < Theoretical Perspectives, Organization and
management theory (General) < Organization and Management Theory <
Topic Areas |
|
|
|
|
|
|
Abstract: |
|
|
Past research has
consistently shown that separation of the CEO and Board Chair roles has no
systematic effect on firm performance. In this study, we introduce the
different forms of separation: apprentice, departure, and demotion. Using
S&P 1500 and Fortune 1000 firms, we find that separation of the two
leadership roles positively impacts future firm performance when current
performance is poor, but negatively impacts future firm performance when
current performance is high. We find that this effect is most dramatic for
demotion separations. Finally, we test theory about the permanency of
CEO-Board Chair separation. Our results show that the different types of
CEO-Board Chair separation have very distinct consequences. |
|
|
|
|
|
|
Academy of Management Journal
|
|
1
Apprentice, Departure, and Demotion: An Examination of
the
Three Types of CEO-Board Chair Separation
RYAN KRAUSE
Department of Management and Entrepreneurship
Kelley School of Business
Indiana University
1309 E. Tenth St.
Bloomington, IN 47405
rakrause@indiana.edu
MATTHEW SEMADENI
Department of Management and Entrepreneurship
Kelley School of Business
Indiana University
1309 E. Tenth St.
Bloomington, IN 47405
semadeni@indiana.edu
Keywords:
Boards of
Directors, Upper Echelons/Corporate Governance, Agency Theory, Organization and
Management Theory
Acknowledgements:
We wish to thank Dan Dalton,
Associate Editor Yan (Anthea) Zhang, and three anonymous AMJ reviewers for
their helpful comments on earlier drafts of this paper.
|
|
|
|
|
|
Academy
of Management Journal
|
|
2
APPRENTICE, DEPARTURE, AND DEMOTION: AN
EXAMINATION OF THE
THREE TYPES OF CEO-BOARD CHAIR SEPARATION
ABSTRACT
Past
research has consistently shown that separation of the CEO and Board Chair
roles has no systematic effect on firm performance. In this study, we introduce
the different forms of separation: apprentice, departure, and demotion. Using
S&P 1500 and Fortune 1000 firms, we find that separation of the two
leadership roles positively impacts future firm performance when current
performance is poor, but negatively impacts future firm performance when
current performance is high. We find that this effect is most dramatic for
demotion separations. Finally, we test theory about the permanency of CEO-Board
Chair separation. Our results show that the different types of CEO-Board Chair
separation have very distinct consequences.
|
|
|
|
|
|
Academy of Management Journal
|
|
3
INTRODUCTION
CEO
duality—the assignment of the CEO and Board Chair roles to one individual—has
fascinated scholars and governance experts for decades (Boyd, 1995; Daily &
Dalton, 1993; Lorsch & MacIver, 1989). Finkelstein, Hambrick, and Cannella
(2009: 230) described duality as “one of the most contentious issues in public
debates about the role of boards of directors”, and with good reason.
Traditionally, scholars have applied one of two opposing theoretical
perspectives in studying duality. Those applying agency theory have
hypothesized that firms with combined CEO and Board Chair roles would
underperform relative to firms with separate roles (Daily & Dalton, 1994;
Rechner & Dalton, 1991). Those viewing the phenomenon from the “unity of
command” perspective (Fayol, 1949) argued the opposite (Finkelstein &
D’Aveni, 1994). Overall, empirical inquiry has consistently failed to uncover
any significant, systematic relationship between CEO duality and firm
performance, and thus failed to support either theory (Dalton, Daily,
Ellstrand, & Johnson, 1998; Dalton, Hitt, Certo, & Dalton, 2007). In
spite of this, governance experts and practitioners consistently advocate the
separate board leadership structure as best practice (MacAvoy & Millstein,
2004; Monks & Minow, 2008).
Building
on prior work that incorporated a contingency perspective (Boyd, 1995;
Finkelstein & D’Aveni, 1994), we argue that separating a firm’s leadership
roles will impact firm performance differently depending on the performance at
the time of the separation. Integrating agency theory and the unity of command
perspective, we hypothesize that when the unity of command inherent in the
combined leadership structure is producing strong performance, independent
oversight will dampen future performance. Alternatively, if the combined
structure is generating weak firm performance, imposing independent oversight
on the CEO will improve future performance. Analysis of S&P 1500 and Fortune
1000 firms from 2002 to 2006 strongly supports this reasoning.
|
|
|
|
|
|
Academy of Management Journal
|
|
4
While
the insight that firms should only separate the CEO and Board Chair roles under
conditions of poor performance is crucial, how firms choose to execute the
separation is just as important, if not more so. It is surprising, then, that
no research to date has explored the different types of separation. We
contribute to the CEO duality literature by identifying the three possible
types of CEO-Board Chair separation and developing theory related to them.
The
first type of separation has been touched on briefly in the literature, but not
as part of a comprehensive framework. This is the “apprentice” separation, in
which a sitting CEO/Board Chair relinquishes the chief executive title but
remains Chair, effectively a violation of agency theorists’ assumption of
independence (Coles & Hesterly, 2000; Daily & Dalton, 1997b; Quigley
& Hambrick, In Press). In the second type, “departure” separations, firms
separate the roles by replacing the CEO/Chair with two different individuals.
In such a scenario, the firm experiences not only a CEO succession event, but
also a significant governance change. By contrast, removal of the Board Chair
title from a sitting CEO, what we term a “demotion”, involves only a governance
change. While all three types of separation reduce the unity of command
inherent in the combined structure, demotion separations with their imposition
of truly independent oversight, generate the greatest potential for a change in
the firm’s strategic direction. As such, we hypothesize that when the status
quo merits change, demotion separations will yield the greatest improvement in
performance, and when the status quo merits no change, demotion separations
will yield the greatest reduction in firm performance. Our results strongly
support this reasoning as well.
Finally,
our study of the different types of CEO-Board Chair separation would be
incomplete without a discussion of permanency, especially since we are treating
board leadership structure as a strategic concern. Based on Vancil’s (1987)
“passing the baton” concept, we hypothesize that apprentice separations will be
the most likely to result in a recombination of the
|
|
|
|
|
|
Academy of Management Journal
|
|
5
CEO
and Board Chair roles. Adapting this logic to cases in which the former CEO is not serving as Board Chair, we then argue
that departure separations will be more likely to result in a recombination if
post-separation firm performance is high than if it is low. Our data supported
both these hypotheses.
The
present study contributes to the literature on CEO duality, agency theory, and
organization theory by suggesting for the first time that in studying CEO-Board
Chair separation, the relevant question is not whether the roles should be
separated, but rather when and how the firm should choose to separate
them. We introduce three distinct types of CEO-Board Chair separation, which
have yet to be treated independently, and develop theory related to their
consequences. For each type, we propose and test the theory that the choice to
separate the CEO and Board Chair positions impacts future firm performance
differently depending on the presence or absence of a performance problem prior
to the separation. If practitioners are to follow the conventional wisdom,
which is to separate the two roles, then they will have to choose from one of
the three types we identify. The theory and results we present in this paper
suggest that demotion separations, involving the strongest change of direction,
yield the most dramatic performance consequences—positive or negative.
Conversely, apprentice separations, arguably the most likely to promote the
status quo, are therefore the most likely to result in a recombination of the
leadership roles.
With
this paper, we join many in the scholarly community in insisting that board
leadership structure impacts firms beyond the minimal effect empirically
demonstrated so far (Brickley, Coles, & Jarrell, 1997; Dalton & Dalton,
2009a, 2010). The evidence we present corroborates this claim, with the caveat
that it is the change in structure and the type of change that ultimately
affect performance.
|
|
|
|
|
|
Academy of Management Journal
|
|
THEORY
AND HYPOTHESES
6
In
the CEO duality literature, those advocating for a separate leadership
structure have traditionally relied on agency theory to justify their arguments
(Dalton et al., 1998; Fama & Jensen, 1983a, 1983b). Building on the initial
work of Berle and Means (1932), agency theorists have examined the problems
that arise when the control and management of organizations are separated from
their ownership. In an agency relationship, the principals (shareholders) hire
an agent (CEO or top management team) to perform a service on the principals’
behalf, “which involves delegating some decision making authority to the agent”
(Jensen & Meckling, 1976: 308). According to agency theory, an independent
board of directors, one untainted by conflicts of interest, is essential to
ensuring that managers promote shareholder welfare, and do not extract
excessive rents from the firm (Jensen, 1993; Mizruchi, 1983).
It
should be clear, then, why CEO duality has garnered so much interest from
scholars taking an agency theory perspective. Positioning the highest-ranking
manager at the head of the body charged with monitoring his or her behavior
violates the principle of independent governance. Early in the development of
agency theory, researchers expressed concern regarding the practice of
combining the CEO and Board Chair roles (Fama & Jensen, 1983a, 1983b;
Mizruchi, 1983). According to Jensen (1993: 866), ‘Without the direction of an
independent leader, it is much more difficult for the board to perform its
critical function’. Brickley, Coles, and Jarrell (1997: 190) quaintly summed up
the agency-based view of duality as the “CEO grading his own homework”.
Not
all scholars of CEO duality adopt this viewpoint, however. Advocates of duality
have drawn on the “unity of command” principle to justify their conclusion that
the benefits of duality outweigh the costs (Dalton et al., 2007). Unity of
command has a long history of support in organization theory. Fayol (1949: 25)
wrote that “A body with two heads is in the social as in the animal sphere a
monster, and has difficulty in surviving”. He argued that unity of command
|
|
|
|
|
|
Academy of Management Journal
|
|
7
was
essential to achieving unity of action, coordination, and focus. He wrote that
“Should [unity of command] be violated, authority is undermined, discipline is
in jeopardy, order disturbed and stability threatened” (Fayol, 1949: 24).
Gulick and Urwick (1937: 93) echoed Fayol’s concerns about separated
leadership, arguing that “The rigid adherence to the principle of unity of
command may have its absurdities; these are, however, unimportant in comparison
with the certainty of confusion, inefficiency, and irresponsibility which arise
from the violation of the principle”. Modern proponents of the unity of command
principle maintain that members of an organization should be responsible to
only one individual, and that the organization should have one voice in engaging
with its environment, both internal and external (Lorsch & Zelleke, 2005).
The latter is especially pertinent during a time of crisis (Dalton &
Dalton, 2009a).
Finkelstein
and D’Aveni (1994) directly tested the contrasting predictions of the unity of
command and agency perspectives. They argued that both frameworks have
validity, but that one takes precedence over the other depending on the
surrounding circumstances. They showed that more vigilant boards, assumed to be
more focused on shareholder value, opted for the combined CEO/Board Chair when
other measures of CEO power were relatively low, and chose a separate
leadership structure when CEO power was relatively high. The authors
interpreted their findings to suggest that vigilant boards weigh the benefits
of duality (i.e. unity of command) against its costs (i.e. lack of
independence) when determining the proper board leadership structure.
Given
the controversy over CEO duality, both among academics and practitioners
(Finkelstein et al., 2009; Lorsch & Zelleke, 2005; Wilson, 2008), the
logical question to ask is what relationship duality has with firm performance.
After two decades of research into the CEO duality-firm performance
relationship (Boyd, 1995; Brickley et al., 1997; Coles, McWilliams, & Sen,
2001; Donaldson & Davis, 1991; Rechner & Dalton, 1989, 1991), neither
|
|
|
|
|
|
Academy
of Management Journal
|
|
8
of
the dominant perspectives has been confirmed. A stream of research, reviews,
and meta-analyses have shown, fairly conclusively, that CEO duality has no
substantive, systematic relationship with firm performance (Baliga, Moyer,
& Rao, 1996; Dahya, Garcia, & van Bommel, 2009; Dalton et al., 1998;
Dalton et al., 2007).
As
Dalton et al. (2007: 13)
described it, the lack of an evident relationship between board leadership
structure and firm performance exhibits a “level of consistency…unusual in any
literature”. It might seem that in the face of such convincing evidence, the
chapter of governance research devoted to CEO duality and firm performance
should be brought to a close. We do not believe, however, that the existing
literature fully encapsulates the phenomenon of CEO duality. With this study we
intend to illustrate the importance of performance at the time of CEO-Board
Chair separations. We examine, not simply the choice to separate the two roles,
but rather when and how those roles are separated, and what
implications these two contextual elements have for firm performance.
Context Matters
The Question of When. Most
empirical examinations of CEO duality’s relationship with performance have
focused on duality as a long-term phenomenon, rather than as a strategic choice
on the part of boards of directors (Iyengar & Zampelli, 2009; Rechner &
Dalton, 1991). In the past, this approach made sense because the vast majority
of boards maintained a combined structure, and any firms that adopted a
separate leadership structure typically did so temporarily as part of the CEO
succession process (Brickley et al., 1997; Vancil, 1987). This is no longer the
case. Over the past decade, many boards have permanently separated their
leadership roles, significantly driving down the percentage of firms with
combined structures year after year (Spencer Stuart, 2010). While a select few
studies have examined changes in board leadership structure (Baliga et al.,
1996), these studies have found no strong relationship between such
|
|
|
|
|
|
Academy of Management Journal
|
|
9
changes
and firm performance. Despite the lack of corroborating evidence, separation
remains a staple of corporate governance recommendations in practice (Monks
& Minow, 2008).
We
contend that the reason for the persistent non-findings is that CEO-Board Chair
separation is only organizationally beneficial under certain circumstances, and
is harmful under others. Agency theory rests on the assumption that conflicts
of interest between managers and shareholders will lead managers to act in ways
that are destructive to shareholder value (Jensen & Meckling, 1976).
Conversely, the unity of command argument is predicated on the assumption that
concentration of power at the top is necessary if CEOs are to lead their
organizations effectively (Fayol, 1949).
We
argue that the appropriateness of a corporate governance alteration like
CEO-Board Chair separation depends on the performance of the firm at the time
of separation. Ultimately, the question of whether a board should transition
from a combined structure to a separate structure boils down to whether the
unity of command inherent in the combined structure is producing superior
performance, as traditional organization theory suggests it should (Gulick
& Urwick, 1937; Perrow, 1986).
If
the firm’s present performance indicates that unity of command is benefiting
the organization, what then would be the result of a separation in this
context? Many reasons exist to suggest that the result will be a strong
reversal of fortune for the firm. Finkelstein and D’aveni (1994: 1083) note
that the combined structure, through clear lines of authority and
responsibility, “helps to avoid confusion among top managers as to who is the
boss, facilitating effective decision-making”. They write that “strong and unambiguous
CEO leadership is an integral part of the success of an organization”.
Conversely, the separate leadership structure “creates multiple authority
relationships that promote role conflict among top managers (Galbraith, 1977)”.
|
|
|
|
|
|
Academy of Management Journal
|
|
10
In
addition, Finkelstein and D’aveni (1994: 1083) cite Miller and Friesen in
noting that strong and unified leadership is important for firm success and
that “‘the diffusion of power makes it difficult to take any decisive actions’
(1977: 268)”. Introducing a separate Board Chair, whose objectives and
strategic priorities may differ from those of the sitting CEO, can lead to
conflict and confusion over who has authority over the firm’s direction. Should
a disagreement between the CEO and the Board Chair arise, agency theory holds
that the independent judgment of the Chair should prevail. There is always the
possibility, however, that the Chair is misguided, and a standoff between the
two parties can bring the organization to a halt.
In
the end, these issues boil down to the old truism, “If it ain’t broke, don’t
fix it.” Superior performance under the leadership of a dual CEO/Board Chair
suggests that the unity of command associated with that structure is working,
or at least not causing a noticeable problem. Boards that separate the
positions under such conditions impose additional independent oversight where
none is needed. Also, regardless of whether the sitting CEO remains in that
role or not, separation implies a change in direction. Boards implementing this
change under such circumstances risk reversing strategies and systems that are
currently producing superior returns. Such boards introduce all the pitfalls of
separation previously mentioned without gaining the benefits. Separation in
this case is a solution in search of a problem, with dire consequences
resulting from its implementation.
What
happens, though, if a firm’s governance system is “broke”? As Simon (1946)
famously observed, the unity of command principle is a “proverb of
administration” and applies perfectly right up until the point where it no
longer applies at all. If a firm’s performance is suffering when it chooses to
separate the CEO and Board Chair roles, we predict that the separation will
lead to improved performance in the future. The basic agency logic applies here
because the poor performance indicates that the unity of command associated
with the combined
|
|
|
|
|
|
Academy of Management Journal
|
|
11
leadership
structure is failing to produce the superior firm performance its advocates
promise. Imposing independent oversight would force a change in direction,
which would be warranted given the firm’s lackluster performance.
The
choice to introduce an independent voice with structural power over the CEO
(Finkelstein, 1992) into board discussions can lead to more critical reviews of
the CEO’s strategies, and ideally more shareholder-focused strategic action
(Lorsch & MacIver, 1989). Fundamentally, the separate leadership structure
has always been advocated as a solution to a problem. In contrast to previous
research, we do not assume that such a problem either must exist or must not
exist. We merely suggest that when a problem does exist, the independent
oversight that results from separation constitutes a productive solution. When
a problem does not exist, and the unity of command inherent in the combined
structure is producing superior shareholder returns, imposition of independent
oversight through separation will only hurt firm performance going forward.
|
|
|
|
|
Hypothesis 1: When a firm is performing well, separation of the CEO
and Board Chair roles decreases firm performance, but when a firm is
performing poorly, separation increases firm performance.
|
The Question of How. Perhaps even more important than the
question of when boards should separate their leadership roles is the question
of how they should do this. Researchers have devoted scant attention to the
problem of how the CEO and Board Chair positions are, or should be, separated.
If a firm currently combines the two roles as the majority of U.S. firms do
(Spencer Stuart, 2010), and the firm’s board elects to separate them as many
governance experts recommend (MacAvoy & Millstein, 2004; Monks & Minow,
2008; Wilson, 2008), there is a limited number of ways such a separation could
be carried out, each with different implications for the firm.
|
|
|
|
|
|
Academy of Management Journal
|
|
12
The
board faces three alternatives if it decides to separate the CEO and Board
Chair roles. The first option, in which the former CEO/Chair remains Board
Chair and a different person is appointed CEO, creates what Dalton and Dalton
(2009a: 79; 2009b: 32) call the “apprentice” CEO. This situation often arises
as a transitional component of a relay succession, whereby the former CEO
remains Board Chair for a short period of time with every intention of
relinquishing that position when the new CEO is deemed fully ready to take the
helm of the firm (Brickley et al., 1997; Harris & Helfat, 1998; Vancil,
1987). In other cases, however, the structural change is permanent, and the
former CEO remains the leader of the board indefinitely, as is often the case
when firm founders are replaced as CEO, but remain Board Chair (Daily &
Dalton, 1997b).
Another
option boards have for separating the two roles is to retain the firm’s current
CEO and appoint a new Chair to oversee him or her. Unlike the apprentice
scenario, this second type of separation cannot be used as a management
transition or succession mechanism, as the CEO is not changing, only the Chair
is. This kind of separation is a pure governance play, and we have termed it a
“demotion” separation, because the CEO necessarily loses a position of
authority he or she previously enjoyed. With that said, the demotion need not
always be foisted on an unwilling CEO. The CEO might request for the board to
appoint a separate Chair in order to bolster the perceived legitimacy of the
firm’s governance practices (Scott, 2001), as Whole Foods founder and CEO John
Mackey did when he relinquished his long-held title of Board Chair (Mackey,
2009). Of course, given the reduction in the CEO’s power due to losing the
Chair position (Daily & Johnson, 1997; Finkelstein, 1992), it is also
highly likely that firms will select this form of separation against the wishes
of the CEO, as Bank of America shareholders did when they voted to relieve CEO
Kenneth Lewis of his dual role as Chairman of the Board (Fitzpatrick &
Eckblad, 2009).
|
|
|
|
|
|
Academy of Management Journal
|
|
13
Finally,
a board might separate the roles of CEO and Chairman as part of a CEO
succession event, but one that looks distinctly unlike an apprentice
separation. After a dual CEO departs—for whatever reason—the board could
replace him or her with two separate individuals, one as Board Chair and the
other as CEO. Such a separation might ease the transition to a separate
structure from a dual structure, since the sitting CEO would not have to
continue to serve in a capacity reduced from that to which he or she had grown
accustomed. Like the apprentice scenario, this type of separation can be used
as a gradual succession mechanism, whereby incoming CEOs are not entrusted—or burdened,
depending on one’s point of view—with dual responsibilities until they have
proven themselves capable. We label this type of separation a “departure”
separation.
Thus,
if a board of directors has determined that the time is right to separate its
leadership roles, it must choose from one of the three forms of separation
discussed above. While we expect the contingent relationship proposed in
Hypothesis 1 to hold for all three types of separation, we also expect demotion
separations to exhibit more dramatic effects on future performance than the
other two types. The reason for this is simply that, while all separation types
violate the unity of command principle, demotions, through the imposition of
independent oversight on a sitting CEO, suggest the greatest potential for a
strategic reversal. Thus, when a separation is warranted, the type of
separation that addresses the problem most directly will engender the most
benefit, and when a separation is not warranted, it will cause the greatest
harm.
Many
have noted that apprentice separations are consistent with the letter, but not
the spirit, of the agency theory recommendation of a separate leadership
structure (Dalton & Dalton, 2009a, 2009b). Over two decades ago, Lorsch and
MacIver (1989: 185-6) warned that “it’s not in the corporation’s best interest
to have its retired CEO become chairman. He or she might
|
|
|
|
|
|
Academy of Management Journal
|
|
14
inhibit a
successor from making necessary changes, because of personal objections, or
because the new CEO was too sensitive to a predecessor’s feelings.” Quigley and
Hambrick (In Press) referred to a CEO succession where the former CEO remains
the Board Chair as a “partial succession”, and found that such arrangements
significantly limit the strategic change that new CEOs can effect. The board
that elects to create an apprentice of their new CEO attains the illusion of
independence without erecting any real barrier between the management and
control of the firm.
The
new apprentice CEO is likely to “stay the course,” as “the presence of the
former CEO as chairperson may inhibit the ability of incumbent CEOs to initiate
changes for fear of offending their predecessors” (Daily & Dalton, 1997a:
129). If firm performance has been relatively good, separating the CEO and
Board Chair roles in an apprentice fashion would likely have a less negative
impact on future performance. This is because apprentice separations are not a
strong governance change, as Daily and Dalton have noted (1997b; 1997a).
Conversely, if performance were poor, an apprentice separation would provide a
less effective solution to the problem.
We
can say the same for departure separations, but for different reasons. If the
sitting CEO/Board Chair departs from the firm, replaced by two separate
individuals, a significant change has, indeed, occurred. The continuity of
leadership inherent in an apprentice separation is not present with the
departure separation. As a result, we would not expect the perpetuation of past
strategies—whether good or bad—that we would in the event of an apprentice
separation. There is, however, a fundamental difference between departure
separations and demotion separations that makes demotion separations more
effective (or in the context of strong firm performance, more destructive). In
essence, the departure separation should produce little
|
|
|
|
|
|
Academy of Management Journal
|
|
15
change in firm
performance above and beyond what would accompany a non-separation-creating
succession (Shen & Cannella, 2002; Weisbach, 1995).
Consider
a scenario in which a poorly performing CEO/Board Chair resigns or is
dismissed. If the successor had been heir apparent (Cannella & Shen, 2001),
he or she may have been complicit in the poorly performing strategies of the
previous CEO, but this connection is by no means clear. Appointment of a
separate Chair might simply create an unnecessary division of command. If the
board chooses an outside replacement for the CEO, the same principle applies
even more strongly. The replacement CEO certainly had no involvement in the
firm’s lackluster performance. Outside CEOs are typically viewed as agents of
change (Kesner & Sebora, 1994), which is why poorly performing firms have
been shown to select outside CEO successors over insiders (Boeker &
Goodstein, 1993). Whether an insider or an outsider, if the new CEO would
leverage his or her unity of command effectively without independent oversight,
hindering him or her with a separate Board Chair only adds to the list of
impediments facing outside CEOs’ turnaround efforts (Friedman & Saul, 1991;
Zhang & Rajagopalan, 2004). If not, the governance change might help the
firm or it might not, depending on the Board Chair, but there is no reason to believe
past performance will moderate this effect much more than it would for a normal
succession.
Demotion
separations do not suffer from the same issues as the other two forms of
separation. By keeping the sitting CEO in the managerial role but removing the
title of Board Chair, the board sends a strong message that things need to
change, and that the newly independent Chair will oversee that transformation.
Whereas a departure separation portends as much strategic change as any CEO
succession, the demotion separation suggests the greatest potential for
strategic—and thus, performance—reversal.
Demotion essentially amounts to a
|
|
|
|
|
|
Academy of Management Journal
|
|
16
change of
direction without a change of CEO. Apprentice separations are a change of CEO
without a change in direction, and departure separations change both CEO and
direction.
For
this reason, demotions are also the most disastrous when implemented in times
of good firm performance. When performance is good, apprentice separations will
likely change little, and thus produce little decline in performance. Departure
separations will produce change, but this change could be positive, negative,
or neutral. Either way, it should produce little change over what a normal
succession would. Demotion separations, however, put the sitting CEO in a
difficult position. If performance was strong prior to the separation, the
sitting CEO will have been demoted for no demonstrable reason, and the unity of
command that had been producing superior shareholder returns will no longer
exist. In addition, imposing such a structure on a high-performing CEO can only
decrease the CEO’s commitment to the firm (Lange, Boivie, & Westphal,
2011). Research has shown that CEOs often resent having to answer to a
superior, all the more so if such a structure is seen as undeserved (Dalton
& Dalton, 2009b; Roberts, 2002). For these reasons, keeping the CEO in the
management position following a separation is the most dangerous if a
performance problem did not exist to warrant the separation.
Therefore,
while we predict that all CEO-Board Chair separations will change the firm’s
performance direction, we also predict that demotion separations, through the
imposition of independent oversight on a sitting CEO, will have the strongest
effect.
|
|
|
|
|
Hypothesis 2: The interactive effect of separation of the CEO and
Board Chair roles with current firm performance on future firm performance is
stronger when the separation is a demotion than when it is (H2a) an
apprentice separation or (H2b) a departure separation.
|
After the Separation: Coming Full Circle
|
|
|
|
|
|
Academy of Management Journal
|
|
17
Finally,
because we are viewing board leadership structure as a strategic choice, with context-dependent
consequences, we must consider the likelihood that the firm comes full circle
and recombines the leadership roles following a separation (Davidson, Nemec,
& Worrell, 2001; Davidson, Worrell, & Nemec, 1998). We must determine
which separations are the most likely to be temporary, and which others are
likely to last. Below, we extend our discussion of the three types of CEO-Board
Chair separation and discuss the likelihood that each type of separation
results in a recombination.
As
mentioned above, apprentice separations are often employed as part of the CEO
succession process, referred to as “passing the baton” (Vancil, 1987). In such
instances, the new CEO remains in a probationary period while the board
evaluates his or her performance. “If the new CEO passes this test” Brickley,
Coles, and Jarrell (1997: 194) write, “then typically the new CEO earns the
additional title of chairman, and the old chairman resigns from the board.”
Interestingly, even if the CEO fails the test, an apprentice separation is
still likely to precede role recombination, but in such an instance, the Board
Chair would return to the CEO role (Dalton & Dalton, 2009b). Brickley,
Coles, and Jarrell (1997: 195) cite Vancil (1987) in noting that “the transition
period…is deliberately structured to allow the board to readily oust the new
CEO, should he or she “drop the baton”. Dalton and Dalton (2009b) refer to this
as the “boomerang CEO” scenario, and such recombinations have occurred in
recent years at such prominent firms as Dell and Seagate Technology. In
contrast to departure and demotion separations, apprentice separations’ value
is in continuity of leadership, not in independence. Thus, regardless of
whether an apprentice separation is followed by high or low performance, it is
likely to result in role recombination, much more so than is either of the
other two types of CEO-Board Chair separation.
|
|
|
|
|
|
Academy of Management Journal
|
|
18
Demotion
separations will be equally uniform in their recombination consequences. As we
discussed above, demotion separations are a pure governance play and by
definition not part of a CEO succession event. The “probation period” logic
behind Vancil’s (1987) concept of passing the baton does not apply to demotion
separations because the CEO is not new to the job. The lack of such a logic
makes the demotion separation far less likely than the apprentice separation to
lead to recombination. Of course, the board could eventually hire a new CEO and
award him or her both leadership roles, in the same way that any board with a
separate leadership structure could. The absence of a probation period logic,
however, makes demotion separations systematically less likely than apprentice
separations to precede recombination.
The
recombination consequences of departure separations are similar to those of the
apprentice separation, but with one important difference. In an apprentice
separation, the board is sacrificing unity of command for continuity of leadership. In a demotion
separation, the board is sacrificing unity of command for independent
oversight. Departure separations, on the other hand, suggest the imposition of
independent oversight, but on an unknown entity (i.e. the new CEO). Following
the departure separation, the new CEO enters a probation period similar to that
which follows an apprentice separation.
There
is one fundamental difference between the apprentice and departure separations
in terms of their recombination prospects. With an apprentice separation, the
subsequent probation period simply constitutes a transition from unity of
command under one leader to unity of command under a new leader—or back to the
old leader if circumstances warrant. Following a departure separation, unity of
command under the old CEO is not really an option, and so the probation period
in this case constitutes a trial and assessment of the new CEO. In this time,
the board must determine whether it wants to transition to unity of command
under, or to permanent
|
|
|
|
|
|
Academy of Management Journal
|
|
19
independent
oversight of, the new CEO. This determination will hinge on the performance of
the firm following the departure separation.
As
with the apprentice separation, positive firm performance following a departure
separation will indicate to the board that the new CEO has “earned” the Board
Chair role, and thus unity of command under his or her leadership would benefit
the firm. However, if performance suffers following the departure separation,
the board is unlikely to oust the CEO and grant the new Board Chair the
additional title of CEO. In contrast to the Board Chair after an apprentice
separation, the Board Chair after a departure separation does not have the
firm-specific knowledge of an
ex-CEO (Lorsch & Zelleke, 2005), and would thus make a less obvious choice
to take the new CEO’s place (Brickley et al., 1997). Whereas bringing back the
former CEO suggests the return of better times in the firm’s history
(Fahlenbrach, Minton, & Pan, 2011), appointing the independent chair as CEO
brings no such fanfare. Therefore, in the event of poor firm performance
following a departure separation, we expect that boards would determine that
unity of command under the new CEO’s leadership is unwarranted. Instead, they
would choose to make independent oversight of the new CEO permanent.
The
theory developed above suggests a distinct permanency profile for each type of
CEO-Board Chair separation. Apprentice separations are clearly the least permanent.
Because these separations typically constitute a passing of the baton, they
will be far more likely than either departure or demotion separations to lead
to a recombination. Demotion separations will be less likely, given that they
are a pure governance play. Overall, departure separations will be less likely
than apprentice separations to precede recombination; however, the likelihood of recombination
following a departure separation will grow as post-separation firm performance
increases. Consistent with these three permanency profiles, we offer the
following two hypotheses.
Academy
of Management Journal
20
|
|
|
|
|
Hypothesis 3: The likelihood of a firm combining its leadership roles
is greater following an apprentice separation than following (H3a) a
departure separation or (H3b) a demotion separation.
|
|
|
|
|
|
Hypothesis 4: The likelihood of a departure separation leading to a
subsequent combination of the leadership roles increases as post-separation
firm performance increases.
|
METHODS
Our
sample for this study consists of all the firms in the Corporate Library
database, which includes the firms in the S&P 1500 and Fortune 1000
indices, between 2002 and 2006. We chose these years because they constitute a
period of substantial increase in the number of large firms separating their
CEO and Board Chair roles (Spencer Stuart, 2010). In addition, we believe it to
be good practice for studies in corporate governance to exclude years preceding
the passage of the Sarbanes-Oxley Act of 2002. The law fundamentally changed so
many aspects of corporate governance that examination of years prior to the
law’s passage limits a study’s practical relevance. All governance data was
retrieved from The Corporate Library’s Companies, Directors, and CEOs
databases. Performance data was retrieved from Compustat, CRSP, and IBES.
Dependent Variables
Performance. We
measured firm performance in two ways. We calculated Stock Return as the holding period return on the company’s
stock from beginning to end of a calendar year. We also measured performance in
terms of the mean composite analyst rating of the company’s stock in a given
year. Analyst ratings can take one of five categorical values, ranging from
“Strong Buy” to “Strong Sell”. Data on these ratings, obtained from IBES, were
coded from 1 for “Strong Buy” to 5 for “Strong Sell”. We reverse coded this
variable, Analyst Rating, so as to
make the results more intuitive.
Academy
of Management Journal
21
CEO-Board Chair Combination.
We measured the combination of the CEO and Board Chair roles as a dichotomous
variable, with a 1 signifying that the firm switched from a separate leadership
structure in the previous year to a combined leadership structure in the
present year, and a 0 signifying no such change.
Independent Variables
CEO-Board Chair Separation.
We measured the separation of the CEO and Board Chair roles as the dichotomous
variable Separation, with a 1
signifying that the firm switched from a combined leadership structure in the
previous year to a separate leadership structure in the present year, and a 0
signifying no such change. These separations were then divided into the three
types. If the sitting CEO in the previous year remained CEO in the year of
separation, the variable Demotion
took a value of 1, otherwise the variable was coded as 0. If the sitting CEO in
the previous year remained Board Chair in the year of separation, the variable Apprentice took a value of 1, and if the
sitting CEO in the previous year was neither CEO nor Board Chair in the year of
separation, the variable Departure
took a value of 1.
Controls
We
included a number of control variables in our analyses. Board Independence was measured as the
percentage of directors on the board who were classified as independent on the
firm’s proxy statement in a given year. This is an important variable to
include as it has been shown to correlate with board leadership structure
(Finkelstein & D’Aveni, 1994). We used the natural log of annual revenues
to measure Firm Size, consistent
with previous research (Walters, Kroll, & Wright, 2010). Industry has been
shown to impact firm performance (Rumelt, 1991), so to control for
industry-level performance effects, we calculated the median performance values
at the 4-digit SIC code level for both performance measures. The resulting
variables are Industry Stock Return
and Industry Analyst Rating. In
order to measure only the effects of
Academy
of Management Journal
22
separation on
future performance, independent of the often-related event of a CEO succession
(Harris & Helfat, 1998), we included a dichotomous variable labeled CEO Change, which took a value of 1 if the
firm changed CEOs between the previous year and the current year.
We
also included a number of CEO characteristic variables: CEO Age, CEO Tenure, and CEO Ownership, which is the percentage of
company stock owned by the CEO in the prior year. These factors have been shown
to correlate with succession and changes in governance (Wiersema & Zhang,
2011). We used year dummy variables to control for potential contemporaneous
correlation (Certo & Semadeni, 2006).
ANALYSIS AND RESULTS
For
our analysis of performance consequences of separation, we excluded any firms
that maintained a separate board leadership structure throughout the period of
analysis. This is consistent with Goertz’s (2006) Possibility Principle, which
states that comparisons between subjects that experienced a particular outcome
and subjects that did not experience the outcome can only be made if the
subjects that did not experience the outcome of interest could possibly have.
Since firms with consistently separate CEO and Board Chair roles could not
possibly have separated the positions again, we cannot use them to determine
the effects of separation. Thus we limited our sample to firms that had a combined
leadership structure and changed it, and firms that had a combined leadership
structure but did not change it. In the three-year window, 1,095 firms
potentially could have experienced a separation, and thus constituted our
sample. We also excluded any separation firms that re-combined their leadership
roles the following year. We did this so as to isolate only those separations
that influenced at least a full year of firm performance.1 From this
sample, 309 did separate their CEO and Board Chair positions, 202 of which were
apprentice separations, 78 were departure separations, and 29 were demotion
|
|
|
|
|
|
|
1 As a
robustness check, we also conducted our analyses with these separation events
added back in, and the results did not materially change.
|
Academy
of Management Journal
23
separations.
While demotion separations were certainly the smallest category, they still
constituted almost 10% of all separations in the period of study.
For
our analysis of the recombination of the leadership roles, we drew a different
sample from our population, this time consisting of all firms with separate
leadership structures, as these are the only firms that could potentially
combine their leadership positions (Goertz, 2006). This selection method produced
843 such firms, inclusive of all the separation firms detailed above, plus all
those that did recombine their leadership roles the following year. In total,
then, this sample included 272 apprentice separations, 105 departure
separations, and 34 demotion separations in addition to the baseline firms,
those who entered the sample with a separate structure.
Descriptive
statistics and pair-wise correlations for variables included in our analyses of
the performance effects of separation appear in Table 1. Our five-year panel
enabled us to examine separation events from 2003 to 2005. Data on board
leadership structure from 2002 was required to determine a separation in 2003,
and data from 2006 was required to determine whether separations in 2005 were
immediately followed by a recombination.
---------------------------------------
Insert Table 1 About Here
---------------------------------------
We
tested Hypothesis 1 using a fixed-effects model. This model is appropriate
because it controls for all firm-specific effects, and only analyzes changes
occurring within firms. This is important because Hypothesis 1 specifically
refers to improvements in an individual firm’s performance, rather than the
cross-sectional superiority of one firm’s performance over another’s. Highly
significant Hausman test results confirmed that fixed-effects regression, and
not random-effects regression, was the appropriate method of analysis (Certo
& Semadeni, 2006). In addition to the control variables listed above, we
also included an interaction of CEO
Academy
of Management Journal
24
Change with
Firm Performance so that any interaction effects of separation can be solely
attributed to the interaction between separation and firm performance, and not
to CEO change.2 The results of our fixed-effects regression testing
Hypothesis 1 are shown in Table 2. Both measures of future firm performance are
included in the table. In the Stock Return models, all performance metrics are
stock returns. In the Analyst Rating models, all performance metrics are
analyst ratings. Due to concerns over multicollinearity, we standardized the
analyst ratings, and improvement in variance inflation factors (VIFs) reflected
the value of this change. The results were the same regardless of whether we
used standardized or unstandardized values.
---------------------------------------
Insert Table 2 About Here
---------------------------------------
As
Table 2 shows, our results strongly confirm Hypothesis 1 for both future stock
return (ß = -0.32, p < 0.01) and future analyst ratings (ß = -0.08, p <
0.05). The interaction for stock return is shown in Figure 1 and the
interaction for analyst ratings is shown in Figure 2. Consistent with prior
research, separation of the CEO and Board Chair roles has no main effect on
future performance. In both sets of analyses, the VIFs for firm size, board
independence, and CEO age were elevated. We removed these variables from the
models, and the results were identical, while all VIFs fell below 3. As such,
we are confident that multicollinearity is not a problem in these models.
--------------------------------------------------
Insert Figures 1 and 2 About Here
--------------------------------------------------
We
fitted the same fixed-effects model for Hypotheses 2a and 2b as for Hypothesis
1, but for Hypotheses 2a and 2b, we divided the separation variable into its
three forms: apprentice, departure, and demotion. Once again, the Hausman test
confirmed the appropriateness of this
|
|
|
|
|
|
|
2 As a
robustness check, we also analyzed the models with only a main effect of CEO
change included, and with CEO change excluded entirely. The results were
highly consistent with the results we report in this manuscript.
|
Academy
of Management Journal
25
method of
analysis. The results from this model are shown in Table 3. For both stock
return and analyst ratings, demotion separations exhibit the same significant
interaction with firm performance that separations in general do [(ß = -1.41, p
< 0.001) and (ß = -0.22, p < 0.001), respectively]. Hypotheses 2a and 2b
do not just predict a significant relationship, however; they predict that the
relationship will be stronger for demotions than for either of the other two
types of separation. Testing these hypotheses requires the use of Wald F tests
to determine the degree of significant difference between regression
coefficients.
---------------------------------------
Insert Table 3 About Here
---------------------------------------
The
F scores associated with these tests are shown in Table 4, and the interactions
for stock return and analyst ratings are depicted in Figure 3 and Figure 4,
respectively. Both hypotheses were strongly supported in the analysis of future
stock return, in that demotion separations exhibited a much stronger negative
interaction with performance than did either apprentice (F = 11.28, p <
0.001) or departure (F = 9.90, p < 0.01) separations. In the analysis of
future analyst ratings, the hypothesized difference between the effect of
demotion and apprentice separations was strongly supported (F = 15.45, p <
0.001) and the difference between the interaction effects of demotion and
departure separations was marginally significant (F = 3.69, p < 0.1). Given
the results of the fixed-effects regression and the Wald tests, we believe that
the data generally support both Hypothesis 2a and Hypothesis 2b.
---------------------------------------
Insert Table 4 About Here
---------------------------------------
-------------------------------------------------
Insert Figures 3 and 4 About Here
-------------------------------------------------
Descriptive
statistics and pair-wise correlations for variables included in our analyses of
separation permanency appear in Table 5. Because we were no longer excluding
firms that
Academy
of Management Journal
26
recombined
their leadership roles in the year after separation, this sample includes
observations from 2002 to 2006.
We
tested Hypotheses 3 and 4 using time-fixed effects logistic regression with
standard errors clustered by firm to adjust for firm-level effects.3
The results of these analyses are shown in Table 6.4 We modeled the
likelihood of a firm combining its leadership roles for all firm-years in which
a combination could have occurred. All performance variables are measured in
the focal year. To examine the difference between the effects of apprentice,
departure, and demotion separations, we performed Wald x2 tests on the difference
between the coefficients. Apprentice separations clearly exhibit a strong
positive impact on the likelihood that a firm will combine its leadership roles
relative to firms that did not experience a separation (ß = 1.33, p <
0.001), but also relative to departure separations (x2 = 7.49, p < 0.01) and to demotion
separations (x2 =
4.60, p < 0.05). Neither demotion nor departure separations exhibited a
significant main effect on the likelihood of combination. These results provide
strong support for Hypotheses 3a and 3b.
---------------------------------------
Insert Table 5 About Here
---------------------------------------
Hypothesis
4 is tested in Model 3 of Table 6. The interactions of departure separation
with post-separation firm performance is significantly positive for both
performance measures: stock return (ß = 2.19, p < 0.05) and analyst ratings
(ß = 0.81, p < 0.05), indicating that while departure separations do not
generally lead to recombination, higher post-separation performance increases
the likelihood that it will occur. This finding supports Hypothesis 4. Because
research has shown interactions in non-linear models to sometimes be deceptive
(Ai & Norton, 2003;
|
|
|
|
|
|
|
3 Firm
fixed-effects regression is not feasible in this context because the lack of
variance in the dependent variable would exclude from the analysis any firm
that never changed its leadership structure. Hypotheses 3 and 4 refer to
differences between firms, rather than within firms, and thus this method is
appropriate.
|
|
4 We excluded
firm size from these models because its inclusion introduced substantial
multicollinearity.
|
Academy
of Management Journal
27
Wiersema &
Bowen, 2009), we also conducted the analyses at above-mean and below-mean
levels of firm performance (Shaver, 2007). At below-mean levels of stock
return, departure separations had no significant impact on the likelihood of
role combination. At above-mean levels of stock return, however, the effect was
positive and significant (ß = 1.08, p < 0.05). The same holds for analyst
ratings. At below-mean levels, departure separations had no significant impact
on the likelihood of role combination. At above-mean levels of analyst ratings,
the effect was positive and significant (ß = 1.16, p < 0.01). We believe this
additional analysis adds considerable validity to the support for Hypothesis 4.
In post-hoc analyses, we also added interactions of apprentice and demotion
separations with future performance into the model, and found no effect,
consistent with our theory.
---------------------------------------
Insert Table 6 About Here
---------------------------------------
Finally,
we modeled CEO-Board Chair separation on past firm performance to make sure
that endogeneity is not a problem in determining the effects of separation on
future firm performance. Our results showed no effect. In other words,
consistent with Iyengar and Zampelli (2009), we find no evidence that boards
select their leadership structure based on firm performance. In order to
further guard against endogeneity, we also tested Hypotheses 1 and 2 using
instrumental variables regression with board independence and CEO age as
instruments, and this analysis supported all three hypotheses across both
measures of firm performance. Therefore, we are confident that we have
established causal priority in our findings.
DISCUSSION
These
results support our theory that CEO-Board Chair separation’s impact on a firm
depends heavily on how and when it is implemented. The data strongly supported
Hypothesis 1, indicating that separations during times of poor firm performance
lead to higher future
Academy
of Management Journal
28
performance,
in terms of both stock return and analyst ratings. However, separations during
times of healthy performance lead to lower performance in the future. Thus, in
addition to the tradeoff between unity of command and independent oversight
that firms face when choosing a board leadership structure, firms choosing to
separate the roles of CEO and Board Chair must also be cognizant of when they choose to do it.
However,
as Figure 1 and Figure 2 show, the size of this effect is relatively small. Our
results suggest that this is because how
a firm separates the CEO and Board Chair roles matters just as much as, if not
more than, when the separation occurs. Ours is the first study to distinguish
between the three types of CEO-Board Chair separation events—apprentice,
departure, and demotion—and analyze their distinct consequences. We find that
distinguishing between these types of separation makes a world of difference. Across
both measures of performance, demotion separations exhibited significantly
stronger effects on future performance than either apprentice separations or
departure separations. This finding provided compelling support for Hypothesis
2. At low levels of both stock return and analyst ratings, firms that separated
their CEO and Board Chair roles in a demotion fashion experienced significantly
better future stock return and analyst ratings, respectively, than did firms
that separated the roles in either an apprentice or a departure separation.
We
have visually depicted the interactions involved in these hypotheses in Figures
3 and 4. Note that we include all forms of separation in these figures, even if
the relationship proved not to be significant. We do this to demonstrate just
how important context is when choosing a board leadership structure. In Figure
3, for instance, demoting a CEO in a year when the firm is returning about 30
percent to shareholders (1 standard deviation above a return of 0) results in a
decrease in shareholder value of about 42 percent the next year, relative to no
separation. The opposite effect holds, however, if the firm lost 30 percent in
the year of separation. By contrast,
Academy
of Management Journal
29
departure
separations produced no significant return above or below what regression to
the mean would generate. Apprentice separations produce a return of close to 8
percent following a 30 percent stock loss, paltry in comparison to the
demotion’s 42 percent.
Figure
4 shows a similar pattern. A firm with a Strong Sell rating would experience a
full rating improvement following a Demotion separation. A firm with a Strong
Buy rating would experience about three fifths of a rating decline following a
Demotion separation. Clearly, both the stock return and analyst rating measures
indicate that demotion separations are by far the most potent, but that boards
should only demote their CEO in times of poor firm performance—the poorer the
better.
That
boards must understand the performance context at the time of separation raises
the question of why firms do not choose their leadership structures based on
firm performance. Consistent with recent research (Iyengar & Zampelli,
2009), post-hoc analyses indicated that even though firms should only separate
the CEO and Board Chair roles in times of poor performance, performance does
not significantly affect the likelihood of boards choosing separation. To
better understand this dynamic, we examined the percentage of high- and
low-performing firms that separated their leadership roles. Across our panel,
10.6 percent of firms whose stocks outperformed their industry separated the
CEO and Board Chair positions. Of firms performing below their industry
average, the figure was almost identical: 10.7 percent. This disconnect between
the context in which separation occurs and the context in which it should occur
constitutes a major opportunity for practitioners to gain valuable insight from
the results of our study.
Our
research also produced interesting insights about the permanency of CEO-Board
Chair separations, allowing us to contribute to the CEO succession literature
as well as the CEO duality literature. Hypothesis 3 was supported, with
evidence showing that apprentice
|
|
|
|
|
|
Academy of Management Journal
|
|
30
separations
are far-and-away the most likely to result in a recombination of the leadership
roles the following year. This is because apprentice separations are a typical
form of succession mechanism known as “passing the baton” (Vancil, 1987). While
past scholars have theorized that the Board Chair position is often awarded as
the final step of a relay succession (Brickley et al., 1997), we contribute new
empirical verification of this assertion. In addition, scholars of CEO turnover
have noted the importance of outgoing CEO duality in the relay succession
process (Zhang & Rajagopalan, 2004). We hope that in future studies of
relay successions, researchers will recognize the distinction between the types
of CEO-Board Chair separation outlined and developed in this paper.
Departure
separations, while less likely to lead to recombination than apprentice
separations, do become more likely as post-separation performance increases.
They are, in a sense, a test of the firm’s new CEO, who cannot benefit from the
wisdom of the previous CEO as an apprentice can. This phenomenon has received
virtually no attention in the CEO succession literature, but the potential for
contribution is extensive. For instance, Zhang and Rajagopalan (2004) report a
positive effect of outgoing CEO duality on the likelihood of a relay
succession. While this is an interesting finding, the authors did not treat it
as material to their theory.5 There are, however, many questions one
could ask in this context that relate to departure separations. What influences
the likelihood that an outgoing CEO will retain the Chair role or leave it?
Does this influence the appointment of an heir apparent to the CEO role? If the
CEO leaves the firm entirely, what determines whether the new CEO will serve as
Chair or a departure separation will occur? Are heirs apparent more or less
likely to be designated prior to a departure separation? These are all
compelling questions that we hope scholars will address going forward.
|
|
|
|
|
|
|
5 The variable
was included as a control in their study.
|
|
|
|
|
|
|
Academy of Management Journal
|
|
31
Demotion
separations, the most potent of the three in terms of performance, is the least
likely to result in a recombination. The results of Hypotheses 2 through 4 seem
to suggest that demotion separations are the most unambiguous governance change
out of the three, with the most dramatic impact on performance and the lowest
likelihood of reversal. Notably, they are also the rarest, although demotions
are becoming more common with each passing year. We hope that governance
scholars will develop this concept further, as it taps into the underlying
independence logic behind CEO-Board Chair separation more consistently than
does either of the other two types of separation.
We
believe our study constitutes a substantial contribution to the corporate
governance literature as well as a powerful new lens through which boards of
directors can examine their choice of leadership structure. The findings we
present raise several questions boards can ask themselves: Is performance
suffering under our CEO’s command? Is it bad enough to justify a change in
board leadership structure? How should we go about doing this? Our post-hoc
analyses suggested that firms are not considering performance when choosing to
separate their leadership roles. As the evidence has shown, failing to ask the
above questions and take performance and separation type into account prior to
implementing a separation can have dire consequences.
Finally,
in opening the discussion about the different forms of CEO-Board Chair
separation, we hope to rekindle the CEO duality conversation. While we believe
our study has pushed understanding of this issue forward, ours are preliminary
efforts, and we have painted the phenomenon of strategic leadership with the
broad brush so often applied in such circumstances. Specifically, while we
theorize about the nature of the CEO and Board Chair’s working
relationship—with an apprentice separation yielding minimal independent
oversight and a demotion separation yielding the opposite—our measures are
still just proxies. For decades,
|
|
|
|
|
|
Academy of Management Journal
|
|
32
researchers have
taken a bird’s eye view of board leadership and inferred certain facts about
the roles of the CEO and Board Chair from what they see. With this paper, we
continue in that tradition. We hope, however, that scholars can build on the
insights we have provided to investigate deeper into boardroom processes and
better identify exactly how CEOs and Board Chairs interact when those roles are
separate.
With
better information on the roles that CEOs and Board Chairs adopt within their
organizations, scholarship on board leadership structure could expand beyond
the traditional paradigms of agency and unity of command, to incorporate
insights from other theories regarding the combination or separation of tasks:
specialization (Taylor, 1911), synergy (Chatterjee, 1986; Lubatkin, 1987), as
well as the resource- and knowledge-based views of the firm (Barney, 1991;
Grant, 1996). We encourage researchers to dig deeper into apprentice,
departure, and demotion separations to discover exactly how Board Chairs interact
with CEOs, or how boards conceptualize their firms’ two leadership roles,
following such transformational events. We provide this study as the first step
down what we anticipate will be a fruitful path of scholarship.
LIMITATIONS
Naturally,
our study is subject to some limitations. The first concerns the scope of our
study. We believe we have extensively addressed the phenomenon of CEO-Board
Chair separation: its types, its performance implications, and its likelihood
of reversal. However, we have not addressed the other side of the coin, which
is CEO-Board Chair combination. Given the importance that scholars and
practitioners put on separation (Monks & Minow, 2008), as well as the
absolute dearth of theory regarding the three possible types of separation, we
limited our study to developing said theory. We hope, however, that future
research will develop similar theory regarding combination. This theory will
need to be distinct, as combination has its own
|
|
|
|
|
|
Academy of Management Journal
|
|
33
typology: a
sitting CEO can become Chair, a sitting Chair can become CEO, or a successor
can be appointed to both roles. We have broached this subject by looking at
re-combinations, but this is only a small slice of the larger potential theoretical
contribution.
In
addition, our study assumes that the board of directors makes the choice to
separate its leadership roles. While this choice always technically falls on
the board, they may sometimes be “forced” into such a choice following a CEO
resignation. While we control for CEO turnover in our models, we are not able
to account fully for the circumstances surrounding each succession and each
separation. This is a limitation of our study, as a separation initiated by the
board might look quite different from a separation initiated by a departing
CEO. Nevertheless, the ultimate decision about whether to separate or not still
resides with the board, and as such, we do not see this as a threat to the
validity of our findings.
Finally,
we limit our study to examining the consequences of CEO-Board Chair separation,
excluding any development or tests of theory related to the antecedents. We
anticipate that a host of factors not only determine the choice to separate the
board leadership roles, but also determine which type of separation the board
will choose. While performance does not determine the board’s choice of
leadership structure (Iyengar & Zampelli, 2009), perhaps firms select their
type of separation differently depending on performance at the time. There are
likely other variables that play a role as well, such as CEO age, CEO tenure,
or board independence. We hope that future research will identify how these and
other predictors impact each type of separation.
CONCLUSION
Board
leadership structure is an issue that every publicly traded firm must address.
While extant research on CEO duality has predominantly treated the phenomenon
as a steady state, recommendations that boards separate their leadership
positions require a structural
|
|
|
|
|
|
Academy of Management Journal
|
|
34
change. Our paper is the first
to recognize that this change must occur in a specific context, and the first
to theorize about how this context will impact the separation’s effectiveness.
We are the first to generate theory about the three possible forms of
separation: apprentice, departure, and demotion. We anticipate that future
research on CEO-Board Chair separation will benefit from this distinction.
Our
research shows that the question of choosing to separate the CEO and Board
Chair positions is not one of whether,
but rather one of when and how. If firms are to separate their board
leadership roles, they should do so only in times of weak firm performance, and
then only in a demotion fashion. We hope that this paper will open a new avenue
of research in the much-traversed terrain of CEO duality. Through this avenue,
perhaps scholars can finally begin to discuss CEO-Board Chair separation, not
as a panacea or a curse, but rather as a tool to be employed selectively and
with care.
|
|
|
|
|
|
Academy of Management Journal
|
|
35
REFERENCES
Ai, C. & Norton, E. C. 2003. Interaction terms in logit and probit
models. Economics
Letters, 80(1): 123-129.
Baliga, B. R., Moyer, R. C., & Rao, R. S. 1996. CEO duality and
firm performance: What’s the fuss? Strategic Management Journal, 17(1): 41-53.
Barney, J. B. 1991. Firm resources and sustained competitive advantage.
Journal of Management,
17(1): 99-120.
Berle, A. A. & Means, G. C. 1932. The modern corporation and private property.
New York,: Macmillan.
Boeker, W. & Goodstein, J. 1993. Performance and successor choice:
The moderating effects of governance and ownership. Academy of Management Journal,
36(1): 172-186.
Boyd, B. K. 1995. CEO duality and firm performance: A contingency
model. Strategic
Management Journal, 16(4): 301-312.
Brickley, J. A., Coles, J. L., & Jarrell, G. 1997. Leadership
structure: Separating the CEO and chairman of the board. Journal of Corporate Finance,
3(3): 189-220.
Cannella, A. A. & Shen, W. 2001. So close and yet so far: Promotion
versus exit for CEO heirs apparent. Academy of Management Journal, 44(2):
252-270.
Certo, S. T. & Semadeni, M. 2006. Strategy research and panel data:
Evidence and implications. Journal
of Management, 32(3): 449-471.
Chatterjee, S. 1986. Types of synergy and economic value: The impact of
acquisitions on merging and rival firms. Strategic Management Journal, 7(2): 119-139.
Coles, J. W. & Hesterly, W. S. 2000. Independence of the chairman
and board composition: Firm choices and shareholder value. Journal of Management,
26(2): 195-214.
Coles, J. W., McWilliams, V. B., & Sen, N. 2001. An examination of
the relationship of governance mechanisms to performance. Journal of Management,
27(1): 23-50.
Dahya, J., Garcia, L. G., & van Bommel, J. 2009. One man two hats:
What’s all the commotion! The
Financial Review, 44: 179-212.
Daily, C. M. & Dalton, D. R. 1993. Board of directors leadership
and structure: Control and performance implications. Entrepreneurship Theory &
Practice, 17(3): 65-81.
Daily, C. M. & Dalton, D. R. 1994. Bankruptcy and corporate
governance: The impact of board composition and structure. Academy of Management Journal,
37(6): 1603-1617.
Daily, C. M. & Dalton, D. R. 1997a. Separate, but not independent:
Board leadership structure in large corporations. Corporate Governance: An International Review,
5(3): 126-136.
Daily, C. M. & Dalton, D. R. 1997b. CEO and board chair roles held
jointly or separately: Much ado about nothing? Academy of Management Executive, 11(3):
11-20.
Daily, C. M. & Johnson, J. L. 1997. Sources of CEO power and firm
financial performance: A longitudinal assessment. Journal of Management, 23(2): 97-117.
Dalton, D. R., Daily, C. M., Ellstrand, A. E., & Johnson, J. L.
1998. Meta-analytic reviews of board composition, leadership structure, and
financial performance. Strategic
Management Journal, 19(3): 269-290.
Dalton, D. R., Hitt, M. A., Certo, S. T., & Dalton, C. M. 2007. The
fundamental agency problem and its mitigation: Independence, equity, and the
market for corporate control. The Academy of Management Annals, 1: 1-64.
|
|
|
|
|
|
Academy of Management Journal
|
|
36
Dalton, D. R. & Dalton, C. M. 2009a. The joint CEO/board
chairperson leadership issue in sharp relief. In J. A. Conger (Ed.), Boardroom realities: Building
leaders across your board. San Francisco, CA: Jossey-Bass
Publishers.
Dalton, D. R. & Dalton, C. M. 2009b. Let’s get real on role
separation: Is splitting the CEO and chairperson positions leading edge...or
over the edge? Directors
& Boards, 33(5): 30-32.
Dalton, D. R. & Dalton, C. M. 2010. At the intersection of ethics
and finance: A collision of theories and collapsing applications. In J. R.
Boatright (Ed.), Finance
ethics: Critical issues in financial theory and practice.
Hoboken, NJ: John Wiley & Sons, Inc.
Davidson, W. N., III, Worrell, D. L., & Nemec, C. 1998. CEO
duality, succession-planning and agency theory: Research agenda. Strategic Management Journal,
19(9): 905-908.
Davidson, W. N., III, Nemec, C., & Worrell, D. L. 2001. Succession
planning vs. agency theory: A test of Harris and Helfat’s interpretation of
plurality announcement market returns. Strategic Management Journal, 22(2): 179-184.
Donaldson, L. & Davis, J. H. 1991. Stewardship theory or agency
theory: CEO governance and shareholder returns. Australian Journal of Management, 16:
49-64.
Fahlenbrach, R., Minton, B. A., & Pan, C. H. 2011. Former CEO
directors: Lingering CEOs or valuable resources? Review of Financial Studies, 24(10):
3486-3518.
Fama, E. F. & Jensen, M. C. 1983a. Agency problems and residual
claims. Journal of Law
and Economics, 26: 327-349.
Fama, E. F. & Jensen, M. C. 1983b. Separation of ownership and
control. Journal of Law
and Economics, 26: 301-325.
Fayol, H. 1949. General
and industrial management. London: Pitman.
Finkelstein, S. 1992. Power in top management teams: Dimensions,
measurement, and validation. Academy of Management Journal, 35(3): 505-538.
Finkelstein, S. & D’Aveni, R. A. 1994. CEO duality as a
double-edged sword: How boards of directors balance entrenchment avoidance and
unity of command. Academy
of Management Journal, 37(5): 1079-1108.
Finkelstein, S., Hambrick, D. C., & Cannella, A. A. 2009. Strategic leadership : Theory and
research on executives, top management teams, and boards. New York:
Oxford University Press.
Fitzpatrick, D. & Eckblad, M. 2009. Lewis voted out as BofA
chairman. Wall Street
Journal. New York, NY: Dow Jones & Co. 30 April. A1.
Friedman, S. D. & Saul, K. 1991. A leader’s wake: Organization
member reactions to CEO succession. Journal of Management, 17(3): 619-642.
Galbraith, J. R. 1977. Organization design. Reading, Mass.: Addison-Wesley Pub.
Co.
Goertz, G. 2006. Social
science concepts : A user’s guide. Princeton: Princeton
University Press.
Grant, R. M. 1996. Toward a knowledge-based theory of the firm. Strategic Management Journal,
17(Winter): 109-122.
Gulick, L. & Urwick, L. 1937. Papers in the science of
administration. In J. S. Ott & J. Shafritz (Eds.), Classics in organizational theory:
87-102. Chicago, IL: The Dorsey Press.
Harris, D. & Helfat, C. E. 1998. CEO duality, succession,
capabilities and agency theory: Commentary and research agenda. Strategic Management Journal,
19(9): 901-904.
Iyengar, R. J. & Zampelli, E. M. 2009. Self-selection, endogeneity,
and the relationship between CEO duality and firm performance. Strategic Management Journal,
30(10): 1092-1112.
Jensen, M. C. & Meckling, W. 1976. Theory of the firm: Managerial
behavior, agency costs, and ownership structure. Journal of Financial Economics, 3:
305-360.
|
|
|
|
|
|
Academy of Management Journal
|
|
37
Jensen, M. C. 1993. The modern industrial revolution, exit, and the
failure of internal control systems. The Journal of Finance, 48(3): 831-880.
Kesner, I. F. & Sebora, T. C. 1994. Executive succession: Past,
present & future. Journal
of Management, 20(2): 327-372.
Lange, D., Boivie, S., & Westphal, J. D. 2011. Predicting
organizational identification at the CEO level. Proceedings of the Seventieth Annual Meeting of the
Academy of Management: 1-6.
Lorsch, J. W. & MacIver, E. 1989. Pawns or potentates : The reality of america’s
corporate boards. Boston, Mass.: Harvard Business School Press.
Lorsch, J. W. & Zelleke, A. 2005. Should the CEO be the chairman? Sloan Management Review,
46(2): 71-74.
Lubatkin, M. 1987. Merger strategies and stockholder value. Strategic Management Journal,
8(1): 39-53.
MacAvoy, P. W. & Millstein, I. M. 2004. The recurrent crisis in corporate
governance. Stanford, Calif.: Stanford Business Books.
Mackey, J. 2009. Latest 8k filing, The CEO’s Blog, Vol. 2010. Austin, TX: Whole
Foods Market.
Miller, D. & Friesen, P. H. 1977. Strategy-making in context: Ten
empirical archetypes. Journal
of Management Studies, 14: 253-280.
Mizruchi, M. S. 1983. Who controls whom - An examination of the
relation between management and boards of directors in large american
corporations. Academy of
Management Review, 8(3): 426-435.
Monks, R. A. G. & Minow, N. 2008. Corporate governance (4th ed.).
Chichester, England ; Hoboken, NJ: John Wiley & Sons.
Perrow, C. 1986. Complex
organizations : A critical essay (3rd ed.). New York: Random
House.
Quigley, T. J. & Hambrick, D. C. In Press. When the former CEO
stays on as board chair: Effects on successor discretion, strategic change, and
performance. Strategic
Management Journal.
Rechner, P. L. & Dalton, D. R. 1989. The impact of CEO as board
chairperson on corporate performance: Evidence vs. Rhetoric. Academy of Management Executive,
3(2): 141-143.
Rechner, P. L. & Dalton, D. R. 1991. CEO duality and organizational
performance: A longitudinal analysis. Strategic Management Journal, 12(2): 155-160.
Roberts, J. 2002. Building the complementary board. The work of the Plc
chairman. Long Range
Planning, 35: 493-520.
Rumelt, R. P. 1991. How much does industry matter? Strategic Management Journal,
12(3): 167-185.
Scott, W. R. 2001. Institutions
and organizations (2nd ed.). Thousand Oaks, Calif.: Sage
Publications.
Shaver, J. M. 2007. Interpreting empirical results in strategy and
management research. Research
Methodology in Strategy and Management, 4: 273-293.
Shen, W. & Cannella, A. A. 2002. Revisiting the performance
consequences of CEO succession: The impacts of successor type, postsuccession
senior executive turnover, and departing CEO tenure. Academy of Management Journal,
45(4): 717-733.
Simon, H. 1946. The proverbs of administration. Public Administration Review,
6(1): 53-67.
Spencer Stuart. 2010. Spencer stuart board index. Chicago, IL: Spencer
Stuart.
|
|
|
|
|
|
Academy of Management Journal
|
|
38
Taylor, F. W. 1911. The principles of scientific management. New York,
London,: Harper & Brothers.
Vancil, R. F. 1987. Passing the baton : Managing the process of ceo succession.
Boston, Mass.: Harvard Business School Press.
Walters, B. A., Kroll, M., & Wright, P. 2010. The impact of TMT
board member control and environment on post-IPO performance. Academy of Management Journal,
53(3): 572-595.
Weisbach, M. S. 1995. CEO turnover and the firms investment decisions. Journal of Financial Economics,
37(2): 159-188.
Wiersema, M. & Zhang, Y. 2011. CEO dismissal: The role of
investment analysts. Strategic
Management Journal, 32(11): 1161-1182.
Wiersema, M. F. & Bowen, H. P. 2009. The use of limited dependent
variable techniques in strategy research: Issues and methods. Strategic Management Journal,
30(6): 679-692.
Wilson, G. 2008. How to rein in the imperial CEO. The Wall Street Journal.
New York: Dow Jones & Co. July 9, 2008. A15.
Zhang, Y. & Rajagopalan, N. 2004. When the known devil is better
than an unknown god: An empirical study of the antecedents and consequences of
relay CEO successions. Academy
of Management Journal, 47(4): 483-500.
Academy
of Management Journal
39
TABLE 1
Means, Standard Deviations, and Correlations
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
Variable
|
Mean
|
S.D.
|
1
|
2
|
3
|
4
|
5
|
6
|
7
|
8
|
9
|
10
|
11
|
12
|
13
|
14
|
15
|
16
|
17
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1
|
|
Future Stock Performance
|
0.14
|
0.27
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2
|
|
Current Stock Performance
|
0.22
|
0.33
|
0.04
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
3
|
|
Future Analyst Rating (Unstandardized)
|
3.54
|
0.47
|
0.12
|
0.23
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
4
|
|
Current Analyst Rating
|
-0.12
|
0.97
|
-0.01
|
0.06
|
0.73
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
5
|
|
Future Industry Average Stock Return
|
0.15
|
0.18
|
0.63
|
0.08
|
0.09
|
0.00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
6
|
|
Current Industry Average Stock Return
|
0.22
|
0.24
|
0.05
|
0.69
|
0.12
|
0.01
|
0.09
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
7
|
|
Future Industry Average Analyst Rating
|
-0.07
|
0.95
|
0.09
|
0.15
|
0.59
|
0.42
|
0.13
|
0.17
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
8
|
|
Current Industry Average Analyst Rating
|
-0.18
|
0.97
|
0.01
|
0.00
|
0.44
|
0.59
|
0.01
|
-0.01
|
0.64
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
9
|
|
Board Independence
|
0.70
|
0.16
|
0.02
|
-0.01
|
-0.01
|
-0.02
|
0.00
|
-0.03
|
-0.03
|
-0.01
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
10
|
|
CEO Age
|
56.11
|
7.18
|
-0.04
|
-0.05
|
-0.04
|
-0.01
|
-0.02
|
-0.05
|
-0.06
|
-0.03
|
0.04
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
11
|
|
CEO Tenure
|
9.22
|
8.30
|
-0.03
|
-0.04
|
0.00
|
0.02
|
-0.03
|
-0.02
|
0.02
|
0.02
|
0.03
|
0.45
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
12
|
|
CEO Ownership
|
0.01
|
0.03
|
0.00
|
0.02
|
0.02
|
0.01
|
-0.01
|
0.01
|
0.01
|
0.01
|
-0.12
|
0.04
|
0.01
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
13
|
|
Firm Size
|
7.53
|
1.53
|
0.02
|
-0.04
|
0.00
|
0.04
|
0.07
|
-0.02
|
-0.04
|
0.00
|
0.14
|
0.02
|
-0.17
|
-0.05
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
14
|
|
CEO Change
|
0.12
|
0.33
|
-0.02
|
-0.02
|
-0.03
|
-0.03
|
-0.03
|
-0.04
|
0.00
|
-0.02
|
0.00
|
-0.19
|
-0.30
|
-0.01
|
0.02
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
15
|
|
Separation
|
0.08
|
0.27
|
-0.04
|
-0.02
|
-0.02
|
-0.04
|
-0.04
|
-0.05
|
0.02
|
0.00
|
-0.03
|
-0.23
|
-0.28
|
0.00
|
-0.05
|
0.67
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
16
|
|
Apprentice
|
0.05
|
0.22
|
-0.02
|
-0.03
|
0.01
|
-0.02
|
-0.01
|
-0.04
|
0.03
|
0.00
|
-0.02
|
-0.19
|
-0.23
|
0.02
|
-0.02
|
0.59
|
0.16
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
17
|
|
Departure
|
0.02
|
0.15
|
-0.01
|
0.00
|
-0.04
|
-0.02
|
-0.05
|
-0.02
|
-0.01
|
0.01
|
-0.02
|
-0.12
|
-0.15
|
-0.02
|
-0.04
|
0.37
|
-0.01
|
-0.04
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
18
|
|
Demotion
|
0.01
|
0.10
|
-0.05
|
0.02
|
-0.02
|
-0.03
|
-0.03
|
0.00
|
-0.01
|
-0.02
|
-0.02
|
-0.04
|
-0.02
|
-0.01
|
-0.03
|
-0.03
|
-0.02
|
-0.02
|
-0.01
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
N = 2,477
All
correlations greater than 0.04 are significant at the 0.05 level.
Academy
of Management Journal
40
TABLE 2
Effects of Separation on Performance
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
Stock
Return
|
|
Analyst
Ratings
|
|
|
|
|
|
|
|
|
|
|
|
Model
1
|
|
Model
2
|
|
Model
3
|
|
Model
1
|
|
Model
2
|
|
Model
3
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
Constant
|
|
|
1.69
|
***
|
|
1.76
|
***
|
|
1.70
|
***
|
|
4.12
|
***
|
|
4.08
|
***
|
|
4.00
|
***
|
|
|
|
|
(0.32
|
)
|
|
(0.32
|
)
|
|
(0.32
|
)
|
|
(0.37
|
)
|
|
(0.37
|
)
|
|
(0.37
|
)
|
|
Industry Performance
|
|
|
0.29
|
***
|
|
0.29
|
***
|
|
0.29
|
***
|
|
0.00
|
|
|
0.00
|
|
|
0.00
|
|
|
|
|
|
(0.04
|
)
|
|
(0.04
|
)
|
|
(0.04
|
)
|
|
(0.01
|
)
|
|
(0.01
|
)
|
|
(0.01
|
)
|
|
Future Industry
Performance
|
|
|
0.94
|
***
|
|
0.94
|
***
|
|
0.94
|
***
|
|
0.23
|
***
|
|
0.23
|
***
|
|
0.23
|
***
|
|
|
|
|
(0.04
|
)
|
|
(0.04
|
)
|
|
(0.04
|
)
|
|
(0.01
|
)
|
|
(0.01
|
)
|
|
(0.01
|
)
|
|
Board Independence
|
|
|
0.08
|
|
|
0.08
|
|
|
0.08
|
|
|
0.18
|
*
|
|
0.17
|
*
|
|
0.17
|
*
|
|
|
|
|
(0.07
|
)
|
|
(0.07
|
)
|
|
(0.07
|
)
|
|
(0.08
|
)
|
|
(0.08
|
)
|
|
(0.08
|
)
|
|
CEO Ownership
|
|
|
0.13
|
|
|
0.12
|
|
|
0.10
|
|
|
0.62
|
|
|
0.63
|
|
|
0.60
|
|
|
|
|
|
(0.44
|
)
|
|
(0.44
|
)
|
|
(0.44
|
)
|
|
(0.50
|
)
|
|
(0.50
|
)
|
|
(0.50
|
)
|
|
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
|
)
|
|
CEO Tenure
|
|
|
-0.00
|
|
|
-0.00
|
|
|
-0.00
|
|
|
0.00
|
|
|
0.00
|
|
|
0.00
|
|
|
|
|
|
(0.00
|
)
|
|
(0.00
|
)
|
|
(0.00
|
)
|
|
(0.00
|
)
|
|
(0.00
|
)
|
|
(0.00
|
)
|
|
Firm Size
|
|
|
-0.22
|
***
|
|
-0.22
|
***
|
|
-0.21
|
***
|
|
-0.05
|
|
|
-0.05
|
|
|
-0.04
|
|
|
|
|
|
(0.04
|
)
|
|
(0.04
|
)
|
|
(0.04
|
)
|
|
(0.05
|
)
|
|
(0.05
|
)
|
|
(0.05
|
)
|
|
Firm Performance
|
|
|
-0.27
|
***
|
|
-0.27
|
***
|
|
-0.27
|
***
|
|
0.05
|
***
|
|
0.05
|
***
|
|
0.05
|
***
|
|
|
|
|
(0.02
|
)
|
|
(0.02
|
)
|
|
(0.02
|
)
|
|
(0.01
|
)
|
|
(0.01
|
)
|
|
(0.01
|
)
|
|
CEO Change
|
|
|
0.02
|
|
|
0.04
|
|
|
0.01
|
|
|
-0.04
|
|
|
-0.05
|
*
|
|
-0.04
|
|
|
|
|
|
(0.02
|
)
|
|
(0.03
|
)
|
|
(0.03
|
)
|
|
(0.02
|
)
|
|
(0.03
|
)
|
|
(0.03
|
)
|
|
CEO Change*Firm
Performance
|
|
|
-0.14
|
*
|
|
-0.16
|
**
|
|
-0.04
|
|
|
-0.04
|
|
|
-0.04
|
|
|
-0.01
|
|
|
|
|
|
(0.06
|
)
|
|
(0.06
|
)
|
|
(0.07
|
)
|
|
(0.02
|
)
|
|
(0.02
|
)
|
|
(0.02
|
)
|
|
Separation
|
|
|
|
|
|
-0.07
|
|
|
-0.02
|
|
|
|
|
|
0.04
|
|
|
0.02
|
|
|
|
|
|
|
|
|
(0.03
|
)
|
|
(0.04
|
)
|
|
|
|
|
(0.04
|
)
|
|
(0.04
|
)
|
|
Separation*Firm
Performance
|
|
|
|
|
|
|
|
|
-0.32
|
**
|
|
|
|
|
|
|
|
-0.08*
|
|
|
|
|
|
|
|
|
|
|
|
(0.10
|
)
|
|
|
|
|
|
|
|
(0.03
|
)
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
Observations
|
|
|
2403
|
|
|
2403
|
|
|
2403
|
|
|
2355
|
|
|
2355
|
|
|
2355
|
|
|
Number of Groups
|
|
|
1053
|
|
|
1053
|
|
|
1053
|
|
|
1033
|
|
|
1033
|
|
|
1033
|
|
|
R Squared
|
|
|
0.45
|
|
|
0.45
|
|
|
0.46
|
|
|
0.28
|
|
|
0.28
|
|
|
0.28
|
|
|
Log Likelihood
|
|
|
1249.88
|
|
|
1253.07
|
|
|
1262.17
|
|
|
928.17
|
|
|
929.00
|
|
|
934.69
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
Year dummy coefficients
included in analysis but omitted from table for parsimony
Standard errors in parentheses
* p < 0.05
** p < 0.01
*** p < 0.001
Academy
of Management Journal
41
TABLE 3
Effects of Apprentice, Departure, and
Demotion Separations on Performance
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
Stock
Return
|
|
Analyst
Ratings
|
|
|
|
|
|
|
|
|
|
|
|
Model
1
|
|
Model
2
|
|
Model
3
|
|
Model
1
|
|
Model
2
|
|
Model
3
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
Constant
|
|
|
1.69
|
***
|
|
1.72
|
***
|
|
1.62
|
***
|
|
4.12
|
***
|
|
4.17
|
***
|
|
4.09
|
***
|
|
|
|
|
(0.32
|
)
|
|
(0.32
|
)
|
|
(0.32
|
)
|
|
(0.37
|
)
|
|
(0.37
|
)
|
|
(0.37
|
)
|
|
Industry Performance
|
|
|
0.29
|
***
|
|
0.29
|
***
|
|
0.28
|
***
|
|
0.00
|
|
|
0.00
|
|
|
0.00
|
|
|
|
|
|
(0.04
|
)
|
|
(0.04
|
)
|
|
(0.04
|
)
|
|
(0.01
|
)
|
|
(0.01
|
)
|
|
(0.01
|
)
|
|
Future Industry
Performance
|
|
|
0.94
|
***
|
|
0.94
|
***
|
|
0.94
|
***
|
|
0.23
|
***
|
|
0.23
|
***
|
|
0.23
|
***
|
|
|
|
|
(0.04
|
)
|
|
(0.04
|
)
|
|
(0.04
|
)
|
|
(0.01
|
)
|
|
(0.01
|
)
|
|
(0.01
|
)
|
|
Board Independence
|
|
|
0.08
|
|
|
0.08
|
|
|
0.07
|
|
|
0.18
|
*
|
|
0.17
|
*
|
|
0.15
|
|
|
|
|
|
(0.07
|
)
|
|
(0.07
|
)
|
|
(0.07
|
)
|
|
(0.08
|
)
|
|
(0.08
|
)
|
|
(0.08
|
)
|
|
CEO Ownership
|
|
|
0.13
|
|
|
0.12
|
|
|
0.13
|
|
|
0.62
|
|
|
0.63
|
|
|
0.61
|
|
|
|
|
|
(0.44
|
)
|
|
(0.44
|
)
|
|
(0.44
|
)
|
|
(0.50
|
)
|
|
(0.50
|
)
|
|
(0.50
|
)
|
|
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
|
)
|
|
CEO Tenure
|
|
|
-0.00
|
|
|
-0.00
|
|
|
-0.00
|
|
|
0.00
|
|
|
0.00
|
|
|
0.00
|
|
|
|
|
|
(0.00
|
)
|
|
(0.00
|
)
|
|
(0.00
|
)
|
|
(0.00
|
)
|
|
(0.00
|
)
|
|
(0.00
|
)
|
|
Firm Size
|
|
|
-0.22
|
***
|
|
-0.21
|
***
|
|
-0.20
|
***
|
|
-0.05
|
|
|
-0.06
|
|
|
-0.05
|
|
|
|
|
|
(0.04
|
)
|
|
(0.04
|
)
|
|
(0.04
|
)
|
|
(0.05
|
)
|
|
(0.05
|
)
|
|
(0.05
|
)
|
|
Firm Performance
|
|
|
-0.27
|
***
|
|
-0.27
|
***
|
|
-0.26
|
***
|
|
0.05
|
***
|
|
0.05
|
***
|
|
0.05
|
***
|
|
|
|
|
(0.02
|
)
|
|
(0.02
|
)
|
|
(0.02
|
)
|
|
(0.01
|
)
|
|
(0.01
|
)
|
|
(0.01
|
)
|
|
CEO Change
|
|
|
0.02
|
|
|
0.04
|
|
|
0.02
|
|
|
-0.04
|
|
|
-0.04
|
|
|
-0.04
|
|
|
|
|
|
(0.02
|
)
|
|
(0.03
|
)
|
|
(0.03
|
)
|
|
(0.02
|
)
|
|
(0.03
|
)
|
|
(0.03
|
)
|
|
CEO Change*Firm
Performance
|
|
|
-0.14
|
*
|
|
-0.16
|
**
|
|
-0.11
|
|
|
-0.04
|
|
|
-0.04
|
|
|
-0.03
|
|
|
|
|
|
(0.06
|
)
|
|
(0.06
|
)
|
|
(0.07
|
)
|
|
(0.02
|
)
|
|
(0.02
|
)
|
|
(0.02
|
)
|
|
Apprentice Separation
|
|
|
|
|
|
-0.07
|
|
|
-0.03
|
|
|
|
|
|
0.01
|
|
|
0.00
|
|
|
|
|
|
|
|
|
(0.04
|
)
|
|
(0.05
|
)
|
|
|
|
|
(0.05
|
)
|
|
(0.05
|
)
|
|
Departure Separation
|
|
|
|
|
|
-0.01
|
|
|
0.01
|
|
|
|
|
|
-0.02
|
|
|
-0.04
|
|
|
|
|
|
|
|
|
(0.05
|
)
|
|
(0.06
|
)
|
|
|
|
|
(0.06
|
)
|
|
(0.06
|
)
|
|
Demotion Separation
|
|
|
|
|
|
-0.14
|
|
|
0.08
|
|
|
|
|
|
0.18
|
*
|
|
0.18
|
*
|
|
|
|
|
|
|
|
(0.07
|
)
|
|
(0.09
|
)
|
|
|
|
|
(0.08
|
)
|
|
(0.08
|
)
|
|
Apprentice Separation*Firm Performance
|
|
|
|
|
|
|
|
|
-0.26
|
*
|
|
|
|
|
|
|
|
0.00
|
|
|
|
|
|
|
|
|
|
|
|
(0.13
|
)
|
|
|
|
|
|
|
|
(0.04
|
)
|
|
Departure Separation*Firm Performance
|
|
|
|
|
|
|
|
|
-0.03
|
|
|
|
|
|
|
|
|
-0.06
|
|
|
|
|
|
|
|
|
|
|
|
(0.15
|
)
|
|
|
|
|
|
|
|
(0.06
|
)
|
|
Demotion Separation* Firm Performance
|
|
|
|
|
|
|
|
|
-1.41
|
***
|
|
|
|
|
|
|
|
-0.22
|
***
|
|
|
|
|
|
|
|
|
|
|
(0.31
|
)
|
|
|
|
|
|
|
|
(0.05
|
)
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
Observations
|
|
|
2403
|
|
|
2403
|
|
|
2403
|
|
|
2355
|
|
|
2355
|
|
|
2355
|
|
|
Number of Groups
|
|
|
1053
|
|
|
1053
|
|
|
1053
|
|
|
1033
|
|
|
1033
|
|
|
1033
|
|
|
R Squared
|
|
|
0.45
|
|
|
0.45
|
|
|
0.46
|
|
|
0.28
|
|
|
0.28
|
|
|
0.29
|
|
|
Log Likelihood
|
|
|
1249.88
|
|
|
1255.32
|
|
|
1278.26
|
|
|
928.17
|
|
|
932.55
|
|
|
948.07
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
Year dummy coefficients
included in analysis but omitted from table for parsimony
Standard errors in
parentheses
|
|
* p
< 0.05
|
|
** p
< 0.01
|
|
*** p
< 0.001
|
Academy
of Management Journal
42
TABLE 4
Wald Tests of Coefficient Differences
|
|
|
|
|
|
|
|
|
|
|
|
Wald F Test: Demotion Coefficient is More Negative Than…
|
|
|
|
|
|
|
|
|
|
Apprentice Coefficient
|
|
Departure Coefficient
|
|
|
|
|
|
|
|
|
|
Dependent
Variable: Future Stock Return
|
|
11.28
|
***
|
|
9.90
|
**
|
|
|
|
|
|
|
|
|
|
|
|
Dependent
Variable: Future Analyst Rating
|
|
15.45
|
***
|
|
3.69
|
†
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
†
p < 0.1
|
|
|
|
|
|
|
|
|
*
p < 0.05
|
|
|
|
|
|
|
|
|
** p < 0.01
|
|
|
|
|
|
|
|
|
*** p < 0.001
|
|
|
|
|
|
|
|
TABLE 5
Means, Standard Deviations, and Correlations
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
Variable
|
|
Mean
|
S.D.
|
1
|
2
|
3
|
4
|
5
|
6
|
7
|
8
|
9
|
10
|
11
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1
|
|
Combination
|
|
|
0.13
|
|
0.34
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2
|
|
Stock Performance
|
|
|
0.20
|
|
0.33
|
|
0.03
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
3
|
|
Analyst Rating
|
|
|
3.54
|
|
0.50
|
|
0.01
|
|
0.05
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
4
|
|
Industry Stock Performance
|
|
|
0.21
|
|
0.23
|
|
0.06
|
|
0.70
|
|
0.02
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
5
|
|
Industry Analyst Rating
|
|
|
3.55
|
|
0.30
|
|
-0.03
|
|
0.03
|
|
0.60
|
|
0.05
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
6
|
|
CEO Ownership
|
|
|
0.00
|
|
0.03
|
|
-0.05
|
|
0.00
|
|
0.05
|
|
-0.01
|
|
0.03
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
7
|
|
CEO Age
|
|
|
51.88
|
|
6.45
|
|
0.00
|
|
0.01
|
|
-0.03
|
|
0.01
|
|
0.00
|
|
-0.03
|
|
|
|
|
|
|
|
|
|
|
|
|
8
|
|
CEO Tenure
|
|
|
4.26
|
|
4.77
|
|
-0.06
|
|
0.02
|
|
0.03
|
|
0.04
|
|
0.01
|
|
0.00
|
|
0.01
|
|
|
|
|
|
|
|
|
|
|
9
|
|
Board Independence
|
|
|
0.65
|
|
0.16
|
|
0.03
|
|
-0.03
|
|
0.00
|
|
-0.04
|
|
-0.02
|
|
-0.10
|
|
0.02
|
|
0.01
|
|
|
|
|
|
|
|
|
10
|
|
Apprentice
|
|
|
0.10
|
|
0.30
|
|
0.13
|
|
-0.08
|
|
0.01
|
|
-0.09
|
|
0.03
|
|
0.03
|
|
-0.07
|
|
0.00
|
|
-0.01
|
|
|
|
|
|
|
11
|
|
Demotion
|
|
|
0.01
|
|
0.12
|
|
-0.01
|
|
-0.07
|
|
-0.01
|
|
-0.06
|
|
-0.01
|
|
-0.02
|
|
0.02
|
|
0.00
|
|
0.04
|
|
-0.03
|
|
|
|
|
12
|
|
Departure
|
|
|
0.04
|
|
0.19
|
|
0.02
|
|
-0.03
|
|
-0.01
|
|
-0.08
|
|
0.01
|
|
-0.03
|
|
-0.03
|
|
0.00
|
|
0.10
|
|
-0.05
|
|
-0.02
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
N = 1,844
All correlations greater
than 0.04 are significant at the 0.05 level
Academy
of Management Journal
43
TABLE 6
Logit Model of CEO-Board Chair Combination
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
Model 1
|
|
Model 2
|
|
Model 3
|
|
|
|
|
|
|
|
|
|
|
|
Constant
|
|
|
-1.74
|
*
|
|
-2.03
|
**
|
|
-2.05
|
**
|
|
|
|
|
(0.73
|
)
|
|
(0.75
|
)
|
|
(0.75
|
)
|
|
Stock Return
|
|
|
-0.28
|
|
|
-0.24
|
|
|
-0.40
|
|
|
|
|
|
(0.28
|
)
|
|
(0.29
|
)
|
|
(0.30
|
)
|
|
Analyst
Rating
|
|
|
0.09
|
|
|
0.09
|
|
|
0.05
|
|
|
|
|
|
(0.09
|
)
|
|
(0.09
|
)
|
|
(0.09
|
)
|
|
Industry
Return
|
|
|
0.56
|
|
|
0.55
|
|
|
0.65
|
|
|
|
|
|
(0.50
|
)
|
|
(0.51
|
)
|
|
(0.51
|
)
|
|
Industry
Analyst Rating
|
|
|
-0.09
|
|
|
-0.10
|
|
|
-0.09
|
|
|
|
|
|
(0.09
|
)
|
|
(0.09
|
)
|
|
(0.09
|
)
|
|
CEO
Ownership
|
|
|
-88.67
|
***
|
|
-89.90
|
***
|
|
-88.56
|
***
|
|
|
|
|
(19.98
|
)
|
|
(20.15
|
)
|
|
(19.96
|
)
|
|
CEO Age
|
|
|
0.00
|
|
|
0.00
|
|
|
0.00
|
|
|
|
|
|
(0.01
|
)
|
|
(0.01
|
)
|
|
(0.01
|
)
|
|
CEO Tenure
|
|
|
-0.07
|
**
|
|
-0.03
|
|
|
-0.03
|
|
|
|
|
|
(0.02
|
)
|
|
(0.02
|
)
|
|
(0.02
|
)
|
|
Board
Independence
|
|
|
0.67
|
|
|
0.70
|
|
|
0.77
|
|
|
|
|
|
(0.48
|
)
|
|
(0.48
|
)
|
|
(0.48
|
)
|
|
Apprentice
Separation
|
|
|
|
|
|
1.33
|
***
|
|
1.33
|
***
|
|
|
|
|
|
|
|
(0.21
|
)
|
|
(0.21
|
)
|
|
Demotion
Separation
|
|
|
|
|
|
-0.30
|
|
|
-0.33
|
|
|
|
|
|
|
|
|
(0.74
|
)
|
|
(0.75
|
)
|
|
Departure
Separation
|
|
|
|
|
|
0.38
|
|
|
-0.24
|
|
|
|
|
|
|
|
|
(0.35
|
)
|
|
(0.56
|
)
|
|
Departure
Separation X Stock Return
|
|
|
|
|
|
|
|
|
2.19
|
*
|
|
|
|
|
|
|
|
|
|
|
(1.08
|
)
|
|
Departure
Separation X Analyst Rating
|
|
|
|
|
|
|
|
|
0.81
|
*
|
|
|
|
|
|
|
|
|
|
|
(0.40
|
)
|
|
|
|
|
|
|
|
|
|
|
|
|
|
Observations
|
|
|
1844
|
|
|
1844
|
|
|
1844
|
|
|
Number of
Firms
|
|
|
829
|
|
|
829
|
|
|
829
|
|
|
Log
Likelihood
|
|
|
-675.68
|
|
|
-656.82
|
|
|
-652.27
|
|
|
Chi Squared
|
|
|
52.44
|
|
|
88.69
|
|
|
96.33
|
|
|
|
|
|
|
|
|
|
|
|
|
|
Year dummy coefficients
included in analysis but omitted from table for parsimony
Standard errors in
parentheses
† p
< 0.1
* p < 0.05
** p < 0.01
*** p < 0.001
Academy
of Management Journal
44
FIGURE 1
The Effect of Separation on Stock Return
FIGURE 2
The Effect of Separation on Analyst Ratings
Academy
of Management Journal
45
FIGURE 3
The Effect of Apprentice, Departure, and
Demotion Separations on Stock Returna,b
aAll relationships are shown relative
to the baseline effect of current stock performance on future stock
performance.
bWe acknowledge that the interaction for
Departure is not significant and we add it only for comparison.
Academy
of Management Journal
46
FIGURE 4
The Effect of Apprentice, Departure, and
Demotion Separations on Analyst Ratingsa,b
aAll relationships are shown relative
to the baseline effect of current analyst ratings on future analyst ratings.
bWe acknowledge that the interactions for
Departure and Apprentice are not significant and we add them only for
comparison.
Academy
of Management Journal
47
Ryan Krause (rakrause@indiana.edu) is a doctoral candidate specializing in strategic
management and organization theory at the Kelley School of Business at Indiana
University. His research interests include board leadership and composition,
executive and director succession, and stakeholder management.
Matthew
Semadeni (semadeni@indiana.edu) is an associate
professor of strategic management in the Kelley School of Business at Indiana
University. He received his Ph.D. from Texas A&M University. His research
interests include top management teams, competitive strategy, and
innovation/knowledge creation.