European Management Journal 35 (2017) 336e350

Contents lists available at ScienceDirect

European Management Journal

journal homepage: www.elsevier.com/locate/emj

The cost of CEO duality: Evidence from French leadership

compensation
!raldine Broye a, *, Abel François b, Yves Moulin c
Ge
a
EM Strasbourg Business School, Strasbourg University (LaRGE), France
b
Lille 1 University, Sciences and Technologies (LEM), France
c
Lorraine University (CEREFIGE), France

a r t i c l e i n f o a b s t r a c t

Article history: This paper aims to provide a detailed analysis of the relationship between board leadership structures
Received 2 February 2016 and executive compensation. According to agency theory, the combined position of CEO and Chairperson
Received in revised form of the Board (COB) entails greater compensation for the CEO in order to reduce conflicts of interest. In the
6 December 2016
literature, combined board structure is generally considered to generate additional costs for companies.
Accepted 23 January 2017
Available online 31 January 2017
However, the choice of two separate structures implies the payment of incentive compensation for the
COB in addition to that defined for the CEO. This paper investigates the financial cost of duality when
compensation packages are set for both leaders. Our results suggest that although combined board
Keywords:
Corporate governance
structure is associated with higher incentive compensation for the CEO, the overall compensation cost to
Executive compensation the company is no higher when the chairperson's compensation is considered.
Duality © 2017 Elsevier Ltd. All rights reserved.
Chairperson of the board
France

1. Introduction compensation that can be accorded to CEO-Chairpersons.

Separating the positions of Chairperson of the Board (COB) and
Chief Executive Officer (CEO) is seen as an essential key to ensuring
good corporate governance. Agency theory argues that combining
the functions of decision and control reduces the board's ability to
effectively monitor the decisions and actions of the CEO, thus giving
the latter more latitude to satisfy his own interests rather than
those of shareholders (Fama & Jensen, 1983; Jensen, 1993). Ac-
cording to managerial power theory, CEOs holding the chairperson
title may use their increased power for rent extraction, thus
obtaining higher compensation at the expense of shareholders
(Bebchuk & Fried, 2004). Alternatively, optimal contracting theory
argues that the incentive compensation of a CEO holding the po-
sition of chairperson is likely to increase in order to mitigate the
reduced effectiveness of direct monitoring (Jensen & Murphy, 1990;
Murphy, 1999). Overall, although a combined chair position may
have benefits such as improving the decision-making process and
the ability to exchange information (Adams & Ferreira, 2007;
Finkelstein & D'Aveni, 1994), this leadership structure may entail
greater bonding costs for the firms, mainly due to the higher
* Corresponding author.
E-mail address: geraldine.broye@unistra.fr (G. Broye).
The aim of this paper is to shed new light on the relationship
between board structure and executive compensation by consid-
ering the role of the chairperson, and specifically his/her own
compensation contract. Separate leadership requires a compensa-
tion contract for the chairperson, in addition to that defined for the
CEO. Specifically, the separation of ownership and control in firms
can generate conflicts of interest between the chairperson and the
shareholders, which can result in incentive compensation packages
being set for the chairperson. It is therefore crucial to consider
compensation paid to the chairperson when analyzing the impact
of the choice of board structure on executive compensation. Ac-
cording to Brickley, Coles, and Jarrell (1997), the decision to sepa-
rate the titles should be analyzed in terms of costs and benefits. In
this light, additional compensation awarded to the chairperson in a
separate structure is likely to result in additional costs for firms. Our
study argues that although the cost of CEO compensation may be
lower in firms with separate titles, the cumulative cost of leader-
ship in these firms is equivalent to the cost associated with CEO
compensation in dual firms, if not higher. The main objective of this
paper is to analyze how far the choice of board structure impacts
the total compensation of leadership, whether single or joint.
This study uses a sample of French public firms from 2005 to
2011 to address two main empirical questions. Firstly, how does

http://dx.doi.org/10.1016/j.emj.2017.01.007
0263-2373/© 2017 Elsevier Ltd. All rights reserved.
 G. Broye et al. / European Management Journal 35 (2017) 336e350
duality affect French CEO compensation? In line with previous
empirical literature, we investigate whether the choice of a com-
bined structure, and the consequent lack of effective monitoring by
a chairperson, entails higher CEO compensation levels. We partic-
ularly seek to pinpoint how the choice of board structure affects the
design of CEO compensation packages and how this choice shapes
the relationship between CEO compensation and firm performance.
As the CEOs of firms with separate titles might be more effectively
monitored by the board, we examine whether incentive pay con-
tracts are used as complements or alternatively as substitutes.
Secondly, how does duality affect French leadership compensation?
We consider the total compensation costs for CEO and COB lead-
ership, and investigate whether a separate board structure is still an
additional cost to the company when COB compensation is
included. By addressing this issue, we identify the determinants of
COB compensation.
This study makes several new contributions to the literature.
First, this paper sheds new light on the relationship between
duality and executive compensation and on the predictions of
optimal contracting and managerial power theories. Previous US-
based empirical studies show diverging results on this issue and
do not provide a definitive conclusion about the validity of these
two theories (e.g. Capezio, Shields, & O'Donnell, 2011; Cordeiro &
Veliyath, 2003; Cyert, Kang, & Kumar, 2002; Dey, Engel, & Liu,
2011). The recent meta-analysis from Van Essen, Otten, and
Carberry (2015) suggests that managerial power does have a sig-
nificant influence over the pay-setting process, but that optimal
contracting arrangements may also exist. The present paper as-
sesses the validity of these theories within the French context.
Second, this paper is a valuable addition to recent cross-national
governance research. Several studies have highlighted the “under-
contextualized” nature of agency theory, and the necessity to
explain the diversity of corporate governance arrangements across
different institutional contexts (Aguilera, Filatotchev, Gospel, &
Jackson, 2008; Aguilera & Jackson, 2003; Desender, Aguilera,
Crespi, & Garcia-Cestona, 2013; Schiehll & Martins, 2016; Van
Essen, Engelen, & Carney, 2013). These studies suggest that the
effectiveness of governance prescriptions may be contingent on a
variety of legal and institutional factors. Van Essen et al. (2013)
specifically document that the positive effects of CEO duality
depend on individual country factors such as the rule of law and
investor rights (La Porta, Lopez-de-Silanes, Shleifer, & Vishny,
2000). Besides, the governance mechanisms interact at both the
firm-level and the country-level, and corporate governance can be
viewed as the result of interrelated country- and firm-level factors
(Aguilera et al., 2008; Schiehll & Martins, 2016).
The present study builds on these arguments to examine com-
plementarities and substitutions among firm-level governance
mechanisms such as board composition and incentive compensa-
tion, within the French context. The majority of literature on CEO
compensation and board structure to date has been conducted in
the Anglo-American context. This environment is characterized by
dispersed ownership where board composition and contractual
incentives are key governance mechanisms (Aguilera & Jackson,
2003; Fernandes, Ferreira, Matos, & Murphy, 2013). This study
provides the opportunity to examine an environment where listed
firms typically have large shareholders and a high proportion of
family-based ownership (Belot, Ginglinger, Slovina and Sushka,
2014). Since large shareholders are likely to exercise more direct
monitoring of managers, a concentrated ownership environment
should entail less necessity for incentive compensation and a
reduced monitoring role of the board. CEO compensation in France
is indeed significantly lower and less equity-based than in the US.
Nevertheless, compensation levels in France have increased
considerably in recent years, as well as the relative proportion of
337
incentive compensation (Goyer & Jung, 2011). Additionally, as large
shareholders have both the power and incentives to discipline
CEOs, they may encourage performance-based compensation.
While ownership concentration is likely to limit agency conflicts
between managers and shareholders, it is also a potential source of
conflicts between large and small shareholders, and blockholders
can use their discretion to extract private benefit (Schleifer and
Vishny, 1997). In a context of weak minority shareholder protec-
tion, as observed in the French context (La Porta et al., 2000), the
board is likely to play an important monitoring role to protect
minority shareholders. In this case the choice of board structure
remains a key issue.
The French institutional setting provides an interesting frame-
work for the study of duality, as companies can freely choose to
unify or split CEO and COB positions within a unitary board (as in
the US model) or a two-tier board (as in the German model). This
leadership structure model provides a stimulating context in which
the relationship between CEO duality and compensation can be
examined.
Lastly, this study provides greater insight into the ongoing
debate surrounding the costs and benefits of implementing good
corporate governance mechanisms in different countries (Aguilera
et al., 2008; Doidge, Karolyi, & Stulz, 2007), and specifically the
costs and benefits of separating roles (Brickley et al., 1997). The
paper adds to the literature by providing a close examination of
how CEO duality impacts leadership compensation within a novel
framework that addresses COB compensation. This contribution
seems significant given the scarcity of literature on the role - and
more particularly the compensation - of the COB (Roberts, 2002).
Nevertheless, the leadership provided by the joint CEO e COB po-
sitions is a key issue for companies that have chosen a separate
board structure. As the leader of the board, the chairperson plays a
pivotal role in the company (Karabadse & Karabadse, 2007;
Roberts, 2002; Waelchli & Zeller, 2013). A COB can make an effec-
tive and positive contribution to the strategic direction and
corporate success of companies. He/she has a central role in
determining the effectiveness of a board and its contribution to the
quality of executive decision making. The COB receives compen-
sation to hold this key position. Our study provides insights into the
financial cost of separating titles when COB compensation is taken
into account.
2. Board structure and leadership compensation: The French
context
2.1. Board structure in France
French companies have the opportunity to choose freely be-
tween three board structures. The first type, introduced by French
law in 1966, is a two-tier board structure. This two-tier system
consists of a supervisory board of non-executive directors and a
separate management board of executive managers. The CEO runs
the management board and the COB chairs the supervisory board;
their respective roles and responsibilities are clearly separated. The
management board conducts the day-to-day management and
corporate activities of the firm, while the supervisory board su-
pervises the management board and the monitoring of managerial
activities. The chairperson sets the agenda for board meetings,
leads discussions among non-executive directors and coordinates
their activities. He/she plays a key role in ensuring that managerial
decisions are in line with the overall strategic guidelines of the
company, and in encouraging effective corporate governance
practices.
The second type of structure is a traditional single-tier board
structure, where the board of directors is typically chaired by one
338 G. Broye et al. / European Management Journal 35 (2017) 336e350
person holding the titles of both COB and CEO. Within this struc-
ture, the CEO has a combined operational and supervisory role,
with full power to act in the firm's name. While chairing the board,
the CEO is expected to take on the role and responsibilities of a non-
executive chairperson.
A third type of structure was introduced by French law in 2001,
and allows the separation of the positions of COB and CEO within a
unitary board of directors. In this case the CEO sits on the board but
this “dissociated board” is chaired by a separate, non-executive
COB. This structure makes it possible to separate the functions of
decision and control within a single board in the same way as in the
US context, and the non-executive COB is expected to play an
identical role to that of the chairperson in a two-tier structure.
This analysis defines dual firms as firms in which CEO and COB
positions are combined. Conversely, the term ‘separate structure’ is
used for any case where these positions are split, within unitary or
two-tier board structures. The term “leadership” refers to the CEO
in combined structures and the CEO and the COB in separate
structures.
Changing between a single-tier and two-tier board requires the
involvement of both the board and shareholders (Belot, Ginglinger,
Slovin, & Sushka, 2014). The board has the initial responsibility of
proposing a change in board structure. It submits a charter
amendment proposal to an Extraordinary General Meeting of
shareholders. Conversely, a unitary board can freely decide to
separate or unify CEO and COB positions without the approval of
the shareholders. Since the enactment of the 2001 act, “dissociated
boards” have multiplied in France, enhancing a trend to separate
the positions of chairperson and CEO. Nevertheless, combined
structures remain the most common type, and are used by 57% of
French listed companies (AMF, 2015). This percentage is very close
to that found recently in the United States by Yang and Zhao (2014),
who provide evidence of a significant trend in recent years towards
separating the positions of CEO and COB. They report that over 80%
of large U.S. companies chose to combine the positions in the early
1990s, and that this figure fell to 54% in 2010. The choice of struc-
ture in France is different to that observed in other European
countries, where a majority of companies separate CEO and COB
positions. This is particularly true in the United Kingdom, Germany,
Austria, Denmark and Finland. French companies are amongst
those with the highest proportion of combined structure boards in
Europe (Heidrick & Struggles, 2011 ).
2.2. CEO and COB compensation in French firms
Executive compensation in French firms usually consists of a
fixed salary, bonuses and equity-based compensation. According to
the Expert Corporate Governance Service survey (ECGS, 2012), 37% of
total CEO compensation is equity based, whilst bonuses make up
35% of the sum. France is ranked fifth in Europe in terms of average
compensation, and total executive compensation is significantly
lower than that of counterparts in the UK, Germany or Switzerland.
In the UK, equity-based compensation alone represents 66% of total
compensation.
In France, the board is responsible for setting CEO compensation
packages1. The board defines salary and bonuses. Although the
initial approval of the shareholders is a prerequisite to the issuance
of any stock option plan or performance shares plan, the board sets
the conditions for the stock option allocation plan (i.e. the
1
France recently introduced the principle of advisory say on pay votes by
shareholders. While say on pay is not a legal requirement, it has been included in
2013 in the French code of governance AFEP-MEDEF as a soft law (AFEP/MEDEF,
2013).
beneficiaries, number of shares allocated and exercise price). In
dual firms, the CEO is awarded a global compensation package
without explicitly separating the compensation received for his/her
position as chairperson. The French Code of Governance (Code
AFEP/MEDEF, 2013) provides several guidelines on compensation
policy, and underlines the necessity to publish clear and trans-
parent rules in annual reports with regard to any performance
criteria used to define executive bonuses and the granting of op-
tions. However, the relative lack of academic research on CEO
compensation in France means that the relationship between pay
and firm performance remains unclear (Broye & Moulin, 2010).
The decision to have a separate board structure entails the
provision of a compensation contract for a non-executive chair-
person, in addition to that defined for the CEO. Like for CEOs, it is
the role of the board to set the level and the structure of COB
compensation (Article L.225-47 of the Commercial Code). Part of
the compensation is paid in the form of meeting fees. The annual
general meeting of shareholders determines an annual amount for
meeting fees, to be shared amongst all directors. The board is
responsible for deciding how to allocate this sum, and how much of
it will be awarded to the COB. Apart from these fees, chairpersons
are likely to receive compensation from other sources. The board of
directors can decide to grant chairpersons stock options, and award
them bonuses for special assignments or duties. Although these
payments should normally be enacted by regulated agreement,
they are, in practice, rarely under shareholder control. The market
authority therefore considers the determination of the COB's
compensation to be too opaque, and specifies that the chairperson
should not receive a bonus or equity-based compensation unless it
is justified by specific circumstances (AMF, 2014, 2015). Overall, the
compensation of non-executive French chairpersons is particularly
high in comparison to that of other European counterparts
(Heidrick & Struggles, 2011).
3. Theoretical framework and hypothesis development
The objective of this paper is to examine the costs entailed by
setting a compensation contract for the chairperson, then relate
this analysis to the costs incurred when a firm separates the posi-
tions of CEO and COB.
3.1. CEO compensation and duality
Two theories address the relationship between the leadership
structure of a firm and the design of CEO compensation contracts,
namely optimal contracting theory and managerial power theory.
Optimal contracting theory suggests that compensation contracts
are set up to solve agency problems and to provide managers with
an incentive to maximize shareholder value (Core, Guay, & Larcker,
2003; Jensen & Murphy, 1990; Murphy, 1999). Under this theory,
when CEO and COB positions are combined, boards should provide
CEOs with efficient incentives to compensate for the reduced
effectiveness of direct monitoring. Consequently, CEOs are ex-
pected to receive a higher proportion of performance-based
compensation, and a compensation package that is closely tied to
firm performance (Dey et al., 2011). This emphasis on contingent
compensation results in increased compensation risk, and risk-
averse CEOs are likely to require higher levels of fixed salary and
total compensation. Accordingly, optimal contracting theory pre-
dicts greater CEO compensation in dual firms, where levels of both
fixed and incentive compensation are higher and the incentive
compensation is more strongly related to performance.
Alternatively, managerial power theory argues that CEOs hold-
ing the title of COB are likely to use their position within the board
to exert pressure on directors and achieve their own objectives
 G. Broye et al. / European Management Journal 35 (2017) 336e350
(Bebchuk & Fried, 2003, 2004; Weisbach, 2007). This theory pre-
dicts that duality will lead to greater structural power through the
authority of the CEO's preeminent formal position (Finkelstein,
1992). The term ‘structural power’ refers to the influence of CEOs
over the board. CEOs who also hold the COB position have increased
power over the director selection process, and can influence the
nomination of friendly board members (Hermalin & Weisbach,
1998; Westphal & Zajac, 1995). Their responsibility for organizing
boards meetings and setting the agendas also gives them the op-
portunity to control the information provided to the board, hence
weakening the board's ability to monitor (Adams & Ferreira, 2007;
Bebchuk & Fried, 2004; Tuggle, Sirmon, Reutzel, & Bierman, 2010).
As CEOs combining both roles have more power over board
members, they are able to negotiate for compensation arrange-
ments that serve their own interests, including a generous
compensation package (Bebchuk & Fried, 2004; Boyd, 1994;
Grabke-Rundell & Gomez-Mejia, 2002).
In dual firms, powerful CEOs are therefore likely to request high
levels of pay. Conversely, the chairperson and the board in firms
with separate structures have more freedom to exercise a more
effective monitoring role, and can limit the level of executive
compensation. CEO compensation is therefore expected to be
higher in combined structures. Managerial power theory also
suggests that the design of CEO compensation contracts reflects
managerial preference rather than optimal incentive contract (Van
Essen et al., 2015). Risk-averse CEOs are likely to prefer a higher
salary to riskier, performance-based compensation (Mehran, 1995).
Accordingly, powerful CEOs are able to dissociate their compensa-
tion from firm performance, making the relationship between
compensation and performance weak or non-existent.
Van Essen et al. (2015) argue that optimal contracting and
managerial power theories are not competing perspectives, and
should rather be viewed as complementary theories to explain
compensation arrangement. Overall, both theories predict higher
levels of salary and total compensation in dual firms, and this im-
plies an increased cost for firms that wish to combine the two
positions. However, the two theories differ on the expected level of
incentive compensation within the package and the expected
relationship between compensation and performance. Empirical
studies (largely conducted in the Anglo-American context) have
reached diverging conclusions on these alternative theoretical ex-
pectations. Core, Holthausen, and Larcker (1999) and Cyert et al.
(2002) find that combining the positions positively impacts CEO
compensation, and suggest that CEOs use their power and
entrenchment within the board to extract rents at the expense of
shareholders. Other studies find no relationship between duality
and CEO compensation (Brickley et al., 1997; Capezio et al., 2011;
Conyon & Peck, 1998; Cordeiro & Veliyath, 2003). Fernandes et al.
(2013) describe a positive impact of duality on compensation
level for US firms, but a negative effect for non-US firms. Dey et al.
(2011) show that executive pay-performance sensitivity was
significantly lower when the company opted for a separation of
positions, and significantly higher after a switch to a combined
structure. Their findings support the optimal contracting hypoth-
esis. Conversely, Firth, Fung, and Rui (2007) find that the relation-
ship between pay and performance is weaker for Chinese firms that
have chosen a combined structure. Van Essen et al. (2015) provided
a recent meta-analysis of 219 US studies and found that duality is
positively related to the level of total compensation, but has no
significant effect on the performance-pay relationship.
The present study allows analyzing how duality influences CEO
compensation arrangements in a context characterized by
concentrated ownership and weaker managerial incentives. The
presence of controlling shareholders reduces the need to rely on
the board for monitoring management, as large shareholders have
339
both the incentives and abilities to actively monitor CEOs
(Desender et al., 2013). The greater capacity of large shareholders to
exercise direct control on CEOs may also make boards feel less
obliged to complete their monitoring role through more incentive
compensation (Aguilera & Jackson, 2003). Fernandes et al. (2013)
suggest that higher institutional ownership and more indepen-
dent boards in US firms entail higher CEO compensation and
increased use of equity-based pay than in non-US firms. The links
between board structure and compensation in the French context
may therefore be less evident than those expected in the Anglo-
American context.
However, high concentration ownership and weak legal pro-
tection for minority shareholders (La Porta et al., 2000) may result
in differing interests between blockholders and minority share-
holders, leading to agency conflicts between large and small in-
vestors (Shleifer & Vishny, 1997). In this context, the board is
expected to strengthen its monitoring role to prevent the expro-
priation of small investors. Specifically, separate board structures
are expected to play a role in minimizing agency problems between
large and small shareholders (Fraile & Fradejas, 2014; Goyer & Jung,
2011), and directors may be willing to set optimal incentive
compensation for managers to protect shareholder value.
The context of ownership concentration may also have conse-
quences on duality effects under the managerial power perspective.
Desender et al. (2013) argue that duality obstructs the board's
monitoring role when its members represent dispersed share-
holders. Conversely, the presence of controlling shareholders on
the board may overcome the dominance of powerful CEOs holding
the COB position. The representation of active large shareholders
on the board can also encourage the setting of optimal compen-
sation contracts to motivate CEOs. In a context of concentrated
ownership, there may therefore be a decrease in CEO/COB power
over the board. This study tests the following hypotheses about the
impact of duality on the level and design of French CEO
compensation:
Hypothesis 1. CEO total compensation is higher in dual firms than in
firms with separate board structures.
Hypothesis 2a. (optimal contract). Dual firms set a CEO compen-
sation package made up of a higher fixed salary, a higher level of
incentive compensation, and incentive compensation that is more
strongly linked to firm performance.
Hypothesis 2b. (managerial power). Dual firms set a CEO
compensation package with a higher fixed salary, lower incentive
compensation, and incentive compensation that is unrelated to firm
performance.
3.2. Leadership compensation and duality
Several publications suggest that the choice of leadership
structure requires a trade-off between expected costs and benefits
(Brickley et al., 1997; Dey et al., 2011; Linck, Netter, & Yang, 2008).
The main expected benefit of splitting the roles lies in reducing
agency costs and therefore increasing the firm performance
through more effective monitoring (Baliga, Moyer, & Rao, 1996;
Krause & Semadeni, 2013). Adams and Ferreira (2007) claim that
separated structures allow the cleanest separation of the board's
advisory and monitoring roles. Duality may also have strategic
leadership benefits, such as improving the decision-making pro-
cess, creating unity in the command structure of a firm (Finkelstein
& D'Aveni, 1994), and fostering information exchange (Adams &
Ferreira, 2007).
The literature also pinpoints some of the potential costs of
separating the positions, such as the costly and potentially
340 G. Broye et al. / European Management Journal 35 (2017) 336e350
incomplete transfer of critical information between the CEO and
the COB (Brickley et al., 1997), the potential for rivalry between
the CEO and the COB (Anderson & Anthony, 1986), and costs
related to confusion about authority (Finkelstein & D'Aveni,
1994). Brickley et al. (1997) also mention the agency costs
incurred by the separation of positions. Although the appoint-
ment of a chairperson may reduce number of agency conflicts
through increased monitoring of CEO behavior, it creates other
agency costs inherent to the monitoring of the COB. Indeed,
chairpersons are likely to take advantage of their position on the
board to favor decisions that will satisfy their own interests at
the expense of shareholders. Brickley et al. (1997) mention that
curiously, the discussion about separate leadership has
completely ignored the critical issue of the non-executive
chairperson's incentives. However, the literature has high-
lighted the value of incentive compensation granted to directors
in order to reduce conflicts of interest arising from separation of
ownership and control and consequently increase the perfor-
mance of the company (Cordeiro, Veliyath, & Romal, 2007;
Deutsch, 2007; Yermack, 2004). In this line, we expect firms
with separate titles to set a specific compensation contract for
the chairperson who runs the board of directors, leading to
additional costs. We propose to further analyze the cost of
separating these two positions by taking the compensation of the
COB into account.
Section 3.1 documents how the optimal contracting and
managerial power arguments both predict higher total compen-
sation for CEOs in combined board structures. This increase in
compensation is considered to incur additional cost for firms with
combined titles. However, the examination of CEO compensation
without factoring the chairperson compensation cannot provide a
reliable comparison of how much each board structure costs in
terms of compensation. It seems more appropriate to compare the
compensation of CEOs in dual firms with the cumulative
compensation of both the CEO and the COB in firms with separate
positions.
It is possible that paying compensation to the chairperson in-
creases the cost to firms with separate positions, and that the
compensation of a CEO holding the COB title may be equivalent to
the compensation of two leaders who share the same role and
responsibility in a firm with separate structure. When examining
CEO compensation in dual firms and in firms with separate struc-
tures, it is important to bear in mind that we are comparing the
compensation of CEOs playing different roles; a CEO who is also the
chairperson can be expected to have a more complex job than a
counterpart in a firm that separates the two positions, and may
thus require higher compensation for holding both positions (Core
et al., 1999). This alternative argument may also explain higher CEO
compensation levels in dual firms. Examining the aggregate
compensation for the CEO and the COB in firms with separate
structures makes it possible to compare the compensation awarded
for the same job, whether it is carried out by one or two heads. We
propose to test the hypothesis that leadership compensation re-
mains similar whether the leadership is single or joint, and that
overall, the choice to combine these positions does not incur higher
costs in terms of compensation.
Hypothesis 3. The compensation of the CEO combining both posi-
tions in dual firms is equivalent to the compensation of the CEO added
to that of the COB in firms separating these positions.
Addressing leadership compensation permits further analysis
of the optimal contracting and managerial power arguments. If
CEO compensation in dual firms is similar to, or even higher than
the leadership compensation in firms with separate structures,
the result would be consistent with a greater rent extraction or
an optimal incentive contract for the CEO holding the chair-
person title. A more thorough analysis is obtained through the
examination of how duality impacts the leadership compensa-
tion structure. In the case of combined positions, the CEO is likely
to assume full responsibility for the firm’s performance, and
compensation should therefore include incentive compensation
and should be closely linked to performance. Conversely, in the
case of separate positions, the CEO and the COB share this re-
sponsibility. The role of the chairperson is potentially critical to
the performance of the company, and high interaction is neces-
sary between the two leaders to ensure corporate success
(Florou, 2005; Waelchli and Zeller, 2013). This sharing of re-
sponsibility and the implementation of incentive compensation
for the chairperson probably entail a lower incentive for a CEO in
a separate structure than for a CEO who also holds the title of
chairperson. Yet the compensation of leadership and the
compensation of one CEO combining COB title should both be
associated with overall firm performance.
Alternatively, if the optimal contracting view dominates, CEOs in
dual firms should have higher incentives and their compensation
should be more strongly dependent on performance compared to
the compensation packages for two separate leaders in firms with
separate structures. If the managerial power view dominates, CEOs
in dual firms should receive less incentive compensation and their
remuneration should be less related to performance than that of two
separate leaders in firms with separate structures. This study tests
the link between firm performance and compensation of the CEO
and the chairperson team, versus the link between firm perfor-
mance and compensation of a CEO who combines both positions.
Hypothesis 4a. (neutral). The compensation packages for the CEO
and the COB in firms with separate structures are equivalent to the
compensation package of the CEO holding the COB position in dual
firms.
Hypothesis 4b. (optimal contract). The compensation includes more
incentive compensation and has a stronger link to firm performance
for the CEO in dual firms than for both the CEO and the COB in firms
with separate structures.
Hypothesis 4c. (managerial power). The compensation includes less
incentive compensation and is less connected to firm performance for
the CEO in dual firms than for both the CEO and the COB in firms with
separate structures.
3.3. Chairperson compensation
The skills and behavior of the chairperson have a direct impact
on the effectiveness of the board (Leblanc, 2005; Roberts, 2002).
Given that the chairperson can play a significant role in the quality
of the decision-making process and the strategic direction of the
firm, he/she can contribute to its successful financial performance.
However, agency problems are likely to emerge from a divergence
of interest between the chairperson and shareholders (Brickley
et al., 1997). If we assume that the COB is a self-serving agent
who does not necessarily act in the interest of shareholders, then
the question arises of which factors determine his/her compensa-
tion, and more specifically whether this compensation is linked to
performance.
Previous studies have investigated the determinants of board
director compensation (e.g. Boyd, 1996; Cordeiro et al., 2007; Ryan
& Wiggins, 2004). This literature provides evidence that outside
director compensation is one of the principle incentive mecha-
nisms motivating directors to behave in the interest of shareholders
(Yermack, 2004). In the US, Cordeiro et al. (2007) mention that
director compensation has considerably shifted towards a greater
 G. Broye et al. / European Management Journal 35 (2017) 336e350
use of stock-based incentive compensation since the mid-1990s, as
incentive pay is expected to be an effective means of aligning the
interests of shareholders and directors. Empirical evidence shows a
relationship between incentive compensation to directors and firm
performance (Cordeiro et al., 2007; Deutsch, 2007).
In France, outside directors cannot be awarded stock options
or restricted shares. Although this equity-based compensation
can be used as a reward mechanism to remunerate chairpersons
of the board, it is a rare occurrence in French firms. We never-
theless expect the COB compensation contract to be closely tied
to firm performance, thus providing COBs with stronger in-
centives to monitor managerial discretion and contribute to
corporate performance. Based on these arguments, we hypoth-
esize as follows:
Hypothesis 5. In firms with separate structures, the compensation of
the COB is positively related to firm performance.
4. Methods
The cost of separating positions is assessed using a sample of
French companies listed on the SBF 120 (Soci! et!
e des Bourses Fran-
çaises 120 index). This index lists the 120 largest equities listed in
France. In this section the sample is described, then the variables
are presented.
4.1. Sample
To construct the sample, we start with firms that have been
present in the SBF 120 Index for at least one year during the
2005e2011 period. Financial companies are excluded due to the
specificity of their accounts. Observations with missing data are
also excluded, as well as any companies that changed their CEO
and/or chairperson during the year. Information relating to the
compensation of CEOs and COBs was collected manually from
annual reports. Financial and market data were gathered from
Amadeus and Eurofidai databases. The final sample contains 577
observations (corresponding to 113 observed firms).
4.2. Dependent variables
CEO total compensation includes salary, bonus and equity-based
compensation. The latter includes the value of any stock options
and shares granted during the year, which is estimated using the
Black and Scholes formula. We also examine CEO fixed salary and
CEO incentive compensation (i.e. the sum of bonus and equity-based
compensation) as a proxy for performance-based compensation.
CEO incentive compensation is also calculated in relative terms to
capture the proportion of compensation that is contingent on
performance. CEO relative incentive is the ratio of incentive
compensation to total compensation.
Leadership total compensation corresponds to the compensation
of leadership within the firm. This represents the CEO's total
compensation in firms with combined structures, given that the
CEO receives one global compensation for his/her position as CEO
and COB. In firms with separate structures, leadership compensa-
tion is the sum of CEO and COB total compensations. We also
consider Leadership fixed salary, Leadership incentive compensation,
and Leadership relative incentive. In firms with separate structures,
COB total compensation is composed of fixed meeting fees and
incentive compensation, which includes bonuses and equity-based
compensation.
The logarithm form is used to analyze compensation variables in
the regressions, thus reducing the influence of observations on tail
and size distribution.
341
4.3. Independent variables
For the assessment of how duality affects CEO and Leadership
compensation, we used the Dual variable which equals 1 if the two
positions are combined (duality), and zero when the two positions
are separated. In line with agency theory, we predict a close and
positive link between CEO compensation and firm performance
that should help align the interests of shareholders with those of
the CEOs. Firm performance is estimated through two financial
measures commonly used in compensation literature (Tosi, Werner,
Katz, & Gomez-Mejia, 2000), i.e. an accounting measure of per-
formance, Return on assets (ROA), and a market-based performance
measure, Stock return. Both variables are lagged to measure prior
performance. We also use two interactive variables, Dual x ROA and
Dual x Stock return, to study the impact of duality on the link be-
tween pay and performance.
We control for a number of other variables that could impact
CEO and COB compensations. The literature shows that size is a
crucial factor when defining the levels of executive compensation
(Jensen & Murphy, 1990; Tosi et al., 2000). Running a larger firm
involves a higher level of responsibility with more hierarchical
levels to coach, demanding greater experience and skills. A larger
company should also require chairpersons to accomplish more
extensive and complex work. We therefore expect larger firms to
offer higher levels of compensation for CEOs and COBs. Firm size is
measured using Log assets, the logarithm of total assets. As CEOs are
responsible for developing growth opportunities, they should be
rewarded when these opportunities are high (Baber, Janakiraman,
& Kang, 1996). These growth opportunities also represent a
higher risk factor for CEOs, and this should be reflected by a higher
level of compensation. Similarly, the COBs of firms with increased
growth opportunities are expected to deal with many diverse is-
sues that demand a more complicated monitoring and evaluation
of CEO performances, thus adding to the COB's workload. These
growth opportunities are measured through Market to book, i.e. the
ratio comparing the company's market capitalization to the book
value of its equity, and we expect to see a positive association with
CEO and COB compensation level.
As ownership structure plays a key role in compensation prac-
tices (Croci, Gonenc, & Ozkan, 2012), we control for various
ownership proxies. French ownership is characterized by concen-
trated ownership, with a high proportion of family-based firms.
This means that large shareholders, and specifically the large
ownership of CEOs or family, are likely to significantly influence
compensation policies. Institutional ownership, although much less
developed than in Anglo-Saxon businesses, has recently increased
in French listed firms and may also be influential (Goyer & Jung,
2011). Previous literature argues that large shareholders, and
particularly family or institutional shareholders, are likely to play a
more active role in monitoring the management (Jensen &
Meckling, 1976), hence limiting agency conflicts not only between
shareholders and CEOs but also between shareholders and chair-
persons. As these investors have both the power and incentives to
discipline CEOs, they are expected to encourage performance-
based compensation and ensure that pay levels are not excessive
(Croci et al., 2012; Fernandes et al., 2013; Van Essen et al., 2015).
When this active monitoring is also applied to COBs, it should
encourage incentive compensation for the chairperson and also
limit the level of his/her compensation. Large, family and institu-
tional ownerships are therefore expected to be associated with
lower total compensation and higher incentive compensation for
the CEO and COB. Besides, as large CEO ownership is likely to
mitigate agency conflicts, it is expected to limit excessive
compensation levels and reduce the need for incentive compen-
sation packages (Core et al., 1999). We predict a negative
342 G. Broye et al. / European Management Journal 35 (2017) 336e350
association between CEO ownership and total and incentive
compensation for the CEO and COB. The following variables are
used to control for the impact of ownership structure on the
compensation of CEOs and COBs: Largest shareholder, which is the
percentage of shares held by the largest shareholder, Institutional
main shareholder, which is equal to 1 if the main owner is an
institutional investor, CEO main shareholder, which is equal to 1 if
the main owner is the CEO2, and Family firm, which is equal to 1 if
the family ultimately holds more than 20% of the shares (following
Faccio & Lang, 2002).
We also control for additional board characteristics. The disci-
plinary role played by the board is likely to be more effective when
it includes a higher proportion of independent outside directors
(Jensen, 1993). Independent directors are expected to be more
resistant to management pressure and also ensure that CEOs do not
expropriate wealth from shareholders in the form of excess pay. A
higher proportion of independent directors is therefore expected to
lower CEO compensation and favor incentive compensation linked
to performance (Chhaochharia & Grinstein, 2009; Croci et al.,
2012). We expect to note a negative (positive) association with
total (incentive) compensation for the CEO. Inversely, we can
expect independent directors to favor a higher level of pay for the
COB as an incentive to monitor the management more effectively
(Ryan & Wiggins, 2004). A higher proportion of independent di-
rectors is expected to be positively associated with the level of COB
compensation. The independent directors variable is the percentage
of independent directors on the board. We also control for Board
size. Larger boards are likely to be associated with less effective
monitoring, as they entail coordination and communication prob-
lems, as well as difficulties in decision making (Jensen, 1993).
Smaller boards are expected to supervise compensation practices
more efficiently, meaning lower total compensation and higher
incentive compensation.
We control for CEO characteristics. CEO age and CEO tenure
variables reflect accumulated experience, the acquisition of
expertise in this role, and specific knowledge of the business and its
environment, all of which should be compensated by higher pay.
Alternatively, these variables may reflect CEO power, in which case
the CEO is likely to exert more pressure on the board and influence
compensation policy for his own advantage (Hill & Phan, 1991).
Moreover, we can expect a negative relationship between CEO
tenure or age and the level of COB compensation. Indeed, an
entrenched CEO can use his power to limit the effective control of
the board by reducing the incentive compensation of directors
(Ryan & Wiggins, 2004), and thereby, that of the chairperson.
To explain COB compensation, we additionally control for COB
tenure and COB age. It is expected that the longer tenure and age of
individuals indicates greater experience, and thus a higher level of
compensation for the chairperson. Having held a previous position
as CEO of the firm or being the founder of the company also gives
the COB significant experience and a thorough knowledge of the
company. This managerial experience should be associated with a
higher level of compensation (Broye & Moulin, 2014). We use the
dummy variable COB previous CEO, which is equal to 1 if the COB
was previously the CEO of the company, and COB founder, which is
equal to 1 if the COB is the founder. We also control for COB
ownership where a large ownership is likely to limit the need for
incentive compensation. The dummy variable COB main share-
holder, which is equal to 1 if the main owner is the COB, is expected
to be negatively associated with total and incentive compensation
2 4
In France, CEO main shareholders are often the founders of the company. We
alternatively used in our models a dummy variable measuring whether the CEO is
founder, and we obtained similar results.
for the COB. We also control for COB in committee, which is equal to
1 if the COB sits on the compensation committee3. We suggest that
the COB's involvement in this committee is likely to provide him
more power to negotiate a higher compensation.
Finally, year and industry fixed effects are controlled in each
regression. We use Industry dummy variables according to ICB
classification4.
5. Results
5.1. Descriptive statistics
Table 1 presents descriptive statistics for our sample.
For the full sample (577 observations), the mean CEO compen-
sation is V2,295,000 and total compensation varies strongly,
ranging from V109,000 to V25,449,000. Equity-based compensa-
tion is the largest element of this compensation, with an average of
V918,000 (40% of total compensation). While this component of
compensation can reach up to V18,200,000, firms did not grant
equity-based compensation to the CEO in 45% of the observations.
This result indicates that French CEOs receive a much lower fraction
of their compensation in the form of stock options than US CEOs
(Fernandes et al., 2013). Fixed salary and bonus represent an
average compensation of V702,000 and V683,000, respectively.
In firms with separate positions (233 observations), COBs
receive an annual average compensation of V370,000. This
compensation also varies widely, as it ranges from V0 to
V3,587,000, with a median of V196,000. On average, this
compensation is made up of 9% fees, 85% bonus and 6% stock op-
tions and shares. Bonuses are therefore the main component of
total COB compensation, while the awarding of stock options to
chairpersons remains a relatively exceptional event in French
companies. Only 11 of our observations illustrate a firm offering
equity-based compensation to their COB.
Regarding board leadership structures, the two positions had
been split in 40% of cases. The French context is therefore similar to
the US context, in which 46% of US companies adopted separate
board structures (Yang & Zhao, 2014). The average proportion of
independent directors is 46.5%. The largest shareholder holds an
average 33.4% of shares, which illustrates the high ownership
concentration in French firms. Moreover, 44% of our observations
are family firms.
Table 1 also provides firm and CEO characteristics for dual firms
and firms with separate structures. Firm size is revealed to be
significantly higher in dual firms, whereas ROA is significantly
lower. Dual firms also have a significantly smaller proportion of
independent directors on their board, and have larger boards. This
result suggests that board structure and board independence are
complementary governance mechanisms used to enhance board
effectiveness, rather than substitutes designed to constrain agency
conflicts (Faleye, 2007). This is also consistent with the managerial
power argument, in which CEOs holding the chairperson title use
their power to favor internal directors. Our results also show that
dual firm CEOs are older, with longer tenure and more ownership.
This suggests that the greater experience of older and long-tenure
CEOs makes them more likely to hold the position as chairperson. It
may also indicate that CEOs are more entrenched and use their
3
We did not control for the CEO sitting on the compensation committee, as this
situation is very rare in France, and has been illegal since 2008. In our sample, there
are only 15 observations of the CEO sitting on the compensation committee, and
the variable is not significant.
The Industry Benchmark Categorization (ICB) distinguishes ten sectors: “oil and
gas”, “basic materials”, “industrials”, “consumer goods”, “health care”, “consumer
services”, “telecommunications”, “utilities”, “financials” and “technology”.
 G. Broye et al. / European Management Journal 35 (2017) 336e350 343

Table 1
Descriptive statistics.

Full sample Dual firms Firms with separate structures

(N ¼ 577) (N ¼ 344) (N ¼ 233)

Mean Median Std. Dev. Mean Median Std. Dev. Mean Median Std. Dev.

CEO salary 702.0 650.0 386.7 705.5 647.5 406.8 696.9 670.0 355.8
CEO bonus 683.3 560.9 628.1 666.9 524.3 636.1 707.6 615.3 616.6
CEO equity-based compensation 918.1 180.6 1779.4 971.4 171.7 1975.8 839.5 220.9 1442.1

CEO compensation (1000V) 2295.0 1689.3 2420.1 2334.8 1592.9 2676.6 2236.3 1800.0 1985.8

COB fees 34.6 24.5 43.9

COB bonus 312.7 154.8 501.2
COB equity-based compensation 22.9 0 119.5

COB compensation (1000V) 370.2 196.5 541.3

Leadership compensation (1000V) 2444.5 1795.0 2535.9 2606.5 1998.8 2308.8

Dual 0.60 1 0.49

ROA (%) 7.04 6.45 8.1 6.09*** 6.2 7.9 8.43 6.45 8.21
Stock return (%) 19.80 16.03 50.6 20.78 18.41 52.70 18.35 14.83 47.33
Total assets (MV) 16,86 4,65 31,08 19,26** 4,68 36,47 13,37 4,53 20,28
Market to book (%) 2.12 1.81 1.72 2.12 1.79 1.66 2.12 1.83 1.82
Largest shareholder (%) 33.4 28.1 23.4 34.6 28 23.9 31.8 29.3 22.6
Family firm 0.44 0 0.50 0.44 0 0.50 0.45 0 0.50
Instit. main shareholder 0.26 0 0.44 0.26 0 0.44 0.27 0 0.44
CEO main shareholder 0.22 0 0.42 0.31*** 0 0.46 0.1 0 0.3
Independent directors (%) 46.5 45.5 20.3 44.3*** 44.3 20.0 49.7 46.1 20
Board size 11.41 11 3.80 11.66* 11 4.16 11.05 11 3.16
CEO tenure 8.5 6 8 10.6*** 8 8.8 5.5 6 5.6
CEO age 56.1 56 6.6 57.2*** 57 6.4 54.4 56 6.6
COB tenure 4.53 3 3.95
COB age 64.4 65 8.6
COB founder 0.15 0 0.36
COB previous CEO 0.46 0 0.5
COB main shareholder 0.26 0 0.44
COB in committee 0.39 0 0.49

Dual ¼ 1 if CEO and COB positions are combined, 0 otherwise; ROA ¼ lagged return on assets, i.e. ratio of operating profit to total assets; Stock return ¼ lagged stock return;
Total assets ¼ firm total assets; Market to book ¼ Ratio of market equity value to book value; Largest shareholder ¼ Percentage of shares held by the largest shareholder;
Family firm ¼ 1 if a family holds more than 20% of the shares, 0 otherwise; Instit. main shareholder ¼ 1 if an institutional investor is the main shareholder, 0 otherwise; CEO
main shareholder ¼ 1 if the CEO is the main shareholder, 0 otherwise; Independent directors ¼ percentage of outside independent directors; Board size ¼ number of directors
in the board; COB founder ¼ 1 if the COB is the founder of the company, 0 otherwise; COB previous CEO ¼ 1 if the COB was previously the CEO of the firm, 0 otherwise; COB
main shareholder ¼ 1 if the COB is the main shareholder, 0 otherwise; COB in committee ¼ 1 if the COB is in the compensation committee, 0 otherwise.
Means comparisons between dual firms and firms with separate structures (t-test). ***, ** and * mean respectively t-test is significant at 1%, 5% and 10%.

power to favor combined structures and influence the choice of
internal directors. Institutional and largest shareholder ownership
levels are similar for both types of firm.
5.2. The impact of duality on CEO and leadership compensation
The results of the estimations based on OLS regressions using
pooled panel are reported in Table 2. We first analyze the impact of
the leadership structure on CEO compensation (Table 2, Panel A),
then examine its effect on leadership compensation (Table 2, Panel
B). Dummy variables are included in each model to control for year
and industry fixed effects (the coefficients are not tabulated for
convenient reason).
For each regression model, we examined variance inflation
factors (VIF) to check for multicollinearity. Maximum VIF values
were far below 10, the conventional threshold for excessive mul-
ticollinearity. The potential endogeneity problem of the dual vari-
able was addressed by performing a DurbineWueHausman test. In
a first step, a logit model was run to determine the board structure
and obtain a Pearson residual value5. In a second step, this residual
variable was incorporated into the original regression models. For
each model, the coefficients of the residual were not significant at
5%, meaning that endogeneity is not a major concern in this
empirical work6.
Additional robustness checks were conducted. The suppression
of fixed effects related to either the industries or the years did not
significantly affect the estimation outcomes. In the same vein, the
successive exclusions of each firm or each year from the sample led
to very similar results. Lastly, we checked that the distributions of
the residuals did not denote any latent problem. These robustness
tests underline the quality of our empirical results.
In Panel A, regression 1a, total CEO compensation is significantly
higher in combined structures than in separate structures. This
result provides strong support for Hypothesis 1, and is consistent
with the findings of Van Essen et al. (2015). In line with both
optimal contracting and managerial power theories, this result
suggests that the choice of a combined structure entails an addi-
tional cost to firms through the provision of higher CEO compen-
sation. More specifically, we find that the higher compensation

5
We used the following variables to determine the Dual variable: Log assets,
Market to book, Log largest shareholder, Institutional main shareholder, Independent
directors, Board size, Family firm, CEO tenure, CEO age, CEO main shareholder, Year
dummies and Industry dummies. The pseudo-R2 of the estimation is 0.22.
6
The coefficients (standard errors) are respectively 0.04 (0.06), 0.02 (0.06), 0.04
(0.03), 0.04 (0.03), 0.12 (0.19), 0.07 (0.18), 0.11 (0.27) and 0.07 (0.26) for regressions
1a, 1b, 2a, 2b, 3a, 3b, 4a and 4b in Panel A; and 0.04 (0.05), 0.02 (0.05), 0.03 (0.03),
0.03 (0.03), 0.12 (0.21), 0.07 (0.20), 0.12 (0.27) and 0.08 (0.26) for regressions 5a, 5b,
6a, 6b, 7a, 7b, 8a and 8b in Panel B.
344 G. Broye et al. / European Management Journal 35 (2017) 336e350

Table 2
Board leadership structure and leadership compensation. This table reports coefficient estimates using logarithm of CEO compensation in Panel A, and logarithm of leadership
compensation in Panel B as dependent variables. Standard errors (in parentheses) are corrected following the method of White (1980). ***, ** and * indicate significance at the
1%, 5% and 10% level, respectively.

Panel A e CEO compensation (N ¼ 577)

Log Total compensation Log Fixed Log Incentive compensation Log Relative
salary incentive

(1a) (1b) (2a) (2b) (3a) (3b) (4a) (4b)

Dual 0.14** 0.08 0.087*** 0.052 0.086 "0.15 "0.070 "0.25

(0.061) (0.004) (0.033) (0.039) (0.16) (0.24) (0.14) (0.22)
ROA 0.0049* 0.004 "0.002 "0.003 0.023** 0.016 0.019* 0.014
(0.003) (0.005) (0.002) (0.002) (0.010) (0.012) (0.010) (0.010)
Stock return 0.18*** 0.017 "0.015 "0.062 0.57*** 0.12 0.45*** 0.11
(0.053) (0.071) (0.028) (0.055) (0.16) (0.19) (0.14) (0.17)
Dual x ROA 0.002 0.0029 0.014 0.010
(0.006) (0.0029) (0.020) (0.019)
Dual x Stock return 0.25*** 0.073 0.68** 0.52**
(0.094) (0.060) (0.26) (0.23)
Log assets 0.30*** 0.30*** 0.22*** 0.22*** 0.33*** 0.32*** 0.060 0.055
(0.030) (0.030) (0.017) (0.018) (0.094) (0.095) (0.085) (0.085)
Market to book 0.040** 0.042** 0.019 0.020* 0.089* 0.096** 0.056 0.062
(0.018) (0.017) (0.012) (0.012) (0.047) (0.047) (0.041) (0.041)
Log largest shareholder "0.16*** "0.16*** "0.039* "0.040* "0.42*** "0.43*** "0.25*** "0.26***
(0.042) (0.041) (0.022) (0.022) (0.11) (0.11) (0.094) (0.094)
Family firm 0.17** 0.17** 0.060 0.057 0.14 0.13 "0.14 "0.14
(0.079) (0.079) (0.043) (0.043) (0.21) (0.21) (0.18) (0.17)
Instit. main shareholder 0.098 0.097 0.047 0.047 0.47*** 0.47*** 0.45*** 0.45***
(0.065) (0.065) (0.037) (0.037) (0.17) (0.17) (0.14) (0.14)
CEO main shareholder "0.37*** "0.36*** "0.13** "0.13** "0.85** "0.84** "0.62** "0.62**
(0.11) (0.11) (0.061) (0.061) (0.34) (0.34) (0.31) (0.31)
Independent directors 0.74*** 0.74*** 0.49*** 0.49*** 1.20** 1.20** 0.65 0.65
(0.19) (0.19) (0.11) (0.11) (0.53) (0.53) (0.51) (0.51)
Board size 0.051*** 0.050*** 0.011 0.010 0.21*** 0.21*** 0.18*** 0.18***
(0.012) (0.012) (0.007) (0.007) (0.037) (0.037) (0.033) (0.033)
CEO tenure 0.002 0.002 "0.003 "0.003 "0.012 "0.013 "0.016 "0.017
(0.005) (0.005) (0.003) (0.003) (0.016) (0.016) (0.015) (0.015)
CEO age "0.004 "0.003 0.006* 0.006* "0.007 "0.006 0.0001 0.001
(0.005) (0.005) (0.003) (0.003) (0.017) (0.017) (0.016) (0.016)
Industry fixed effects yes yes yes yes yes yes yes yes
Years fixed effects yes yes yes yes yes yes yes yes
Constant 8.69*** 8.74*** 9.18*** 9.23*** 4.43** 4.69** "0.89 "0.70
(0.59) (0.60) (0.31) (0.32) (1.82) (1.82) (1.64) (1.63)

Adj. R2 0.57 0.57 0.63 0.63 0.44 0.44 0.34 0.34

max VIF 2.58 3.43 2.58 3.43 2.58 3.43 2.58 3.43

Panel B e Leadership compensation (N ¼ 577)

Log Total compensation Log Fixed Log Incentive compensation Log Relative
salary incentive

(5a) (5b) (6a) (6b) (7a) (7b) (8a) (8b)

Dual "0.010 "0.083 0.024 "0.008 "0.27* "0.58 *** "0.31** "0.56 ***
(0.059) (0.072) (0.033) (0.038) (0.15) (0.21) (0.13) (0.19)
ROA 0.004 0.003 "0.002 "0.003 0.017* 0.004 0.012 "0.000
(0.003) (0.005) (0.002) (0.002) (0.009) (0.008) (0.009) (0.005)
Stock return 0.18*** 0.0043 "0.020 "0.071 0.56*** 0.14 0.45*** 0.16
(0.052) (0.065) (0.028) (0.054) (0.15) (0.15) (0.13) (0.13)
Dual x ROA 0.003 0.002 0.026 0.023
(0.006) (0.003) (0.018) (0.017)
Dual x Stock return 0.27*** 0.079 0.65*** 0.45**
(0.088) (0.059) (0.23) (0.20)
Log assets 0.30*** 0.30*** 0.21*** 0.21*** 0.34*** 0.33*** 0.049 0.038
(0.029) (0.029) (0.017) (0.017) (0.088) (0.088) (0.079) (0.078)
Market to book 0.040** 0.042** 0.019 0.020* 0.098*** 0.11*** 0.072** 0.080***
(0.017) (0.016) (0.012) (0.012) (0.038) (0.038) (0.029) (0.030)
Log largest shareholder "0.17*** "0.17*** "0.045** "0.045** "0.37*** "0.37*** "0.18** "0.18**
(0.041) (0.040) (0.022) (0.022) (0.10) (0.10) (0.083) (0.082)
Family firm 0.18** 0.18** 0.068 0.066 0.32* 0.30 0.076 0.055
(0.076) (0.076) (0.043) (0.043) (0.19) (0.19) (0.16) (0.16)
Instit. main shareholder 0.069 0.068 0.050 0.051 0.30** 0.30** 0.27** 0.27**
(0.062) (0.062) (0.036) (0.036) (0.15) (0.15) (0.12) (0.12)
CEO main shareholder "0.39*** "0.39*** "0.12** "0.12** "1.34*** "1.35*** "1.22*** "1.23***
(0.11) (0.11) (0.060) (0.060) (0.30) (0.29) (0.27) (0.26)
Independent directors 0.75*** 0.75*** 0.48*** 0.48*** 1.42*** 1.43*** 0.96** 0.97**
(0.19) (0.19) (0.10) (0.10) (0.51) (0.51) (0.48) (0.48)
Board size 0.052*** 0.052*** 0.010 0.010 0.20*** 0.20*** 0.17*** 0.17***
 G. Broye et al. / European Management Journal 35 (2017) 336e350 345

Table 2 (continued )

Panel B e Leadership compensation (N ¼ 577)

Log Total compensation Log Fixed Log Incentive compensation Log Relative
salary incentive

(5a) (5b) (6a) (6b) (7a) (7b) (8a) (8b)

(0.011) (0.011) (0.007) (0.007) (0.035) (0.035) (0.031) (0.031)

CEO tenure 0.003 0.002 "0.003 "0.003 0.007 0.005 0.007 0.006
(0.005) (0.005) (0.003) (0.003) (0.014) (0.014) (0.013) (0.013)
CEO age "0.006 "0.005 0.007** 0.007** "0.022 "0.021 "0.017 "0.017
(0.005) (0.005) (0.003) (0.003) (0.014) (0.014) (0.013) (0.013)
Industry fixed effects yes yes yes yes yes yes yes yes
Years fixed effects yes yes yes yes yes yes yes yes
Constant 8.87*** 8.94*** 9.28*** 9.32*** 5.38*** 5.79*** 0.28 0.64
(0.58) (0.59) (0.31) (0.31) (1.67) (1.65) (1.48) (1.46)

Adj. R2 0.59 0.60 0.63 0.63 0.50 0.50 0.39 0.39

Max VIF 2.58 3.43 2.58 3.43 2.58 3.43 2.58 3.43

provided to CEOs holding the COB title is related to a higher salary
(regression 2a), whereas incentive compensation is similar in both
governance structures (regressions 3a and 4a). Duality does not,
therefore, have any significant effect on the level and the propor-
tion of incentive compensation. This finding is not consistent with
the predictions of both the optimal contracting and the managerial
power expectations. However, the interaction between duality and
performance variables will provide further results.
Overall, and in line with agency theory, CEO compensation is
significantly and positively related to firm performance. Re-
gressions 1a, 3a and 4a show that performance measures play an
important role in the determination of total and incentive executive
compensation. These results are similar to the findings of Core at al.
(1999) and Fernandes et al. (2013), but contrast with the results of
Croci et al. (2012). We observe that executive compensation is more
strongly associated with market performance than with accounting
performance. Regressions 1b, 3b and 4b introduce interactive var-
iables, and show that the sensitivity of CEO compensation to mar-
ket performance is higher in dual firms than in firms with separate
structures. In particular, incentive compensation in dual firms has a
significantly stronger link to stock return than that observed in
firms with separate structures. These results support the optimal
contracting argument and hypothesis 2a, and are consistent with
previous empirical evidence reported by Dey et al. (2011). In
contrast, these results do not support hypothesis 2b and therefore
invalidate the managerial power argument, which predicts that
powerful CEOs holding the COB title might influence the board to
serve their own interests, specifically through compensation that is
not tied to performance. Our findings rather suggest that CEO
compensation in dual firms is more closely related to market per-
formance in order to align the interests of CEOs with those of
shareholders when the monitoring role of the board is reduced. In
line with this argument, the higher pay for CEOs combining posi-
tions might be explained by the setting of a higher fixed salary to
compensate for the risk incurred by the CEO. Overall, while this
study does not find any evidence pointing to higher levels of
incentive compensation in dual firms, its findings clearly support
the optimal contracting expectation of a stronger link between pay
and market performance. The relationship between accounting
performance and executive compensation is not affected by the
type of board structure.
Panel B shows our results when leadership compensation ele-
ments are used as dependent variables. When we include COB
compensation in our estimation of leadership compensation, we
find that the coefficient of the Dual variable is no longer significant
(regression 5a). Thus, when taking into account the cost linked to
the compensation of both CEO and COB in firms with separate
positions, we find that the aggregate compensation of leadership is
equivalent to the compensation of a CEO combining both titles. This
result supports Hypothesis 3. Specifically, our findings indicate that
the fixed salary given to CEOs in dual firms is not significantly
different in comparison to the combined fixed salary of both leaders
in firms with separate structures (regression 6a). These results
challenge the argument that the combined structure is associated
with an additional cost. Overall, the need to set additional
compensation for chairperson in separate structures offsets the
granting of a higher compensation for dual firm CEOs in terms of
cost.
We also find that leadership total and incentive compensation
are significantly associated with market performance (regressions
5a, 7a and 8a). Incentive compensation is more slightly related to
accounting performance. The use of interactive variables in re-
gressions 5b, 6b and 7b makes it possible to assess the impact of
duality on the relationship between leadership compensation and
firm performance. The interactive variable Dual x Stock return is
positive and highly significant in regressions 5b, 7b and 8b. This
finding implies that the sensitivity of CEO total and incentive
compensation to market performance remains greater in dual firms
than in firms with separate structures, even when COB compen-
sation is taken into account. This evidence suggests that dual firms
are willing to set incentive compensation for CEOs that is closely
linked to performance, thus remaining consistent with optimal
contracting theory and supporting Hypothesis 4b. However, this
choice does not involve a higher level of compensation in dual
firms. Overall, the choice to combine CEO and COB positions leads
the board to increase CEO motivation through an incentive
compensation that is closely tied to market performance, but this
does not incur higher costs than those observed in firms with
separate structures. Regressions 7a, 7b, 8a and 8b also show that
CEOs holding the COB position receive a lower level of incentive
compensation in comparison with the sum of incentive compen-
sation packages for the CEO and the COB in separate structures.
In other words, our results do not support Hypothesis 4a and
indeed provide evidence that the compensation packages for the
CEO and the COB in firms with separate structures are different to
the package set for the CEO holding the COB position in dual firms.
We show that the leadership compensation is more strongly linked
to market performance in dual firms, thus supporting Hypothesis
4b and optimal contracting theory and invalidating Hypothesis
4c. However, our results do not support the prediction of
Hypothesis 4b that the compensation package would include more
incentive compensation in dual firms: on the contrary, we find that
346 G. Broye et al. / European Management Journal 35 (2017) 336e350
incentive compensation is higher in firms with separate structures.
This finding can be interpreted as follows: Given the lack of sig-
nificant difference in incentive CEO compensation in combined and
separate structures (regressions 3a, 3b, 4a and 4b), this finding
suggests that when we consider the incentive compensation set for
the COB, the global incentive compensation of the CEO and the COB
is higher than the incentive compensation set for the CEO alone in
combined structure.
In accordance with our expectations and with previous studies,
CEO compensation and leadership compensation are shown to be
higher in larger firms and in firms with strong growth opportu-
nities. Ownership concentration is negatively associated with the
levels of total and incentive compensation, suggesting that more
direct and active monitoring provided by large shareholders
involve limited compensation levels and smaller incentives. Hence,
contrary to our expectations, this increased monitoring acts as a
substitute for incentive compensation, rather than a complement.
These results are in line with the findings of Core et al. (1999) and
Van Essen et al. (2015). Conversely, the Institutional main share-
holder variable has an insignificant effect on total compensation
and a positive and significant effect on incentive compensation.
Institutional shareholders therefore seem to favor a higher level of
performance-based compensation for CEOs, probably because their
large portfolio does not put them in a position to carry out strong,
direct monitoring. This result is consistent with works by Croci et al.
(2012) and Fernandes et al. (2013), who find that institutional
shareholdings are associated with more equity-based compensa-
tion. In contrast to our expectations and to Croci et al. (2012), we
find that family firms are associated with higher level of total
compensation and are not associated with incentive compensation.
A possible interpretation is that families over-compensate the CEOs
in order to buy their loyalty, and allow them to expropriate mi-
nority shareholders (Croci et al., 2012). As expected, and in line with
agency theory expectation, the CEO main shareholder variable is
negatively associated with total, fixed and incentive compensation,
suggesting that CEO large ownership reduces the need for any
compensation contracts that encourage CEOs to act in the interest
of shareholders (Core et al., 1999).
Regarding board composition, a higher proportion of inde-
pendent directors is associated with a higher level of total
compensation. Whilst contrary to our expectations, this result is
similar to those of Fernandes et al. (2013), Croci et al. (2012) and
Van Essen et al. (2015). This finding may be consistent with the
argument that independent directors are willing to increase CEO
compensation through increased reliance on incentive compen-
sation. We observe that the Independent directors variable is
positively and significantly associated with both fixed and
incentive compensation. However, the variable is not signifi-
cantly associated with the ratio of CEO incentive compensation to
total compensation. Independent directors can therefore be
considered to effectively act to increase the level of CEO
performance-based compensation, but this increase is not sig-
nificant in relation to total compensation. Thus, we cannot
conclude that independent boards play a significant role in the
CEO incentive policy. These results are not consistent with Croci
et al. (2012), who describe a positive association between board
independence and the proportion of equity-based compensation
to total compensation in Continental European firms. Like Croci
et al. (2012), this study shows that larger boards are associated
with higher total and incentive compensation. Despite potential
coordination problems, large boards seem to provide more
effective monitoring through more incentive compensation ar-
rangements. Finally, we find that CEO tenure has no significant
impact on CEO compensation, whereas CEO age implies a higher
level of fixed salary.
5.3. Determinants of COB compensation
Table 3 presents the determinants of COB compensation from
the sub-sample of firms with separate positions. The highest VIF
value in these regressions is 3.45, indicating that there is no
problem of multicollinearity.
The analysis of COB compensation enables us to explore our
primary results in more depth. We find that the chairperson's total
and incentive compensation are not associated with firm perfor-
mance. This result is not consistent with Hypothesis 5, and fails to
support the prediction that the chairperson's compensation con-
tracts are designed to motivate COBs to maximize the performance
of the firm. As previously mentioned, very few French chairpersons
are granted stock options or restricted shares. Our finding further
demonstrates that firms do not attempt to align the interests of
COBs and shareholders by using incentives closely linked to
performance.
We also find evidence that the level of total and incentive COB
compensation is positively and significantly related to firm size, but
is not associated with growth opportunities. Regarding the
ownership structure variables, we find that Log largest shareholder
and Institutional main shareholder are negatively associated with
COB total compensation, and that Family firm and Institutional main
shareholder are significantly and negatively related to COB incentive
compensation. Overall, these results suggest that ownership con-
centration, and specifically family and institutional blockholders,
are associated with lower compensation and incentives for the COB.
We previously predicted that these investors would press for more
incentives to be given to the COB in order to ensure the efficient
monitoring of the CEO. On the contrary, these results suggest a
substitution effect, and are consistent with the view that active
monitoring by controlling shareholders reduces the need to offer
the chairperson high levels of compensation. Similarly, we find that
CEO ownership is negatively associated with total and incentive
compensation. These results support the expectation that the COB
will receive less incentive to monitor effectively when there is little
risk of agency conflicts occurring.
This study also finds that the proportion of independent di-
rectors is not significantly associated with the level of total and
incentive COB compensation. This result is not consistent with the
expectation that independent directors will press for higher levels
of pay and incentive compensation for the COB to monitor man-
agement more effectively. Board size has also no incidence on the
level of total and incentive COB compensation. Overall, these re-
sults suggest that the COB compensation is more impacted by the
type of ownership structure than by board composition.
CEO tenure is positively related to COB total and incentive
compensation, whereas CEO age has a negative impact on the
incentive compensation of the COB and a positive impact on his/her
fixed salary. These results do not allow us to clearly conclude that
potential CEO power, as estimated through CEO tenure and CEO
age, systematically results in giving less incentives to the COB to
achieve effective monitoring. The age and tenure of the COB have
no impact on his/her total and incentive compensation, although
COBs with a longer tenure receive higher levels of fixed compen-
sation. However, the chairperson's compensation is significantly
influenced by his/her previous position: COBs who served as the
CEO of the company before chairing the board received signifi-
cantly higher levels of total and incentive compensation. The firm
therefore values the skills acquired through this previous experi-
ence. Inversely, the COB founder variable is weakly and negatively
associated with total compensation, and is not related to COB
incentive compensation. Our results also provide evidence that the
COB main shareholder variable has no impact on COB compensation.
When the COB sits on the compensation committee, he/she
 G. Broye et al. / European Management Journal 35 (2017) 336e350 347

Table 3
COB compensation determinants. This table reports coefficient estimates using logarithm of COB compensation as dependent variables. Standard errors (in parentheses) are
corrected following the method of White (1980). ***, ** and * indicate significance at the 1%, 5% and 10% level, respectively.

COB compensation (N ¼ 233)

Log Total compensation Log Fixed salary Log Incentive compensation Log Relative incentive

ROA 0.016 0.032* "0.014 "0.012

(0.009) (0.018) (0.020) (0.009)
Stock return "0.20 "0.32 0.26 0.18
(0.18) (0.21) (0.37) (0.16)
Log assets 0.39*** 0.047 0.90*** 0.33***
(0.12) (0.15) (0.17) (0.069)
Market to book 0.021 0.012 0.002 "0.009
(0.057) (0.060) (0.094) (0.039)
Log largest shareholder "0.65*** "0.30 "0.35 0.074
(0.24) (0.18) (0.23) (0.089)
Family firm 0.46 1.28*** "0.80** "0.67***
(0.33) (0.30) (0.39) (0.16)
Instit. main shareholder "0.68** "0.74** "1.28*** "0.54***
(0.29) (0.30) (0.43) (0.17)
CEO main shareholder "1.54*** 0.12 "2.13*** "0.75***
(0.40) (0.44) (0.52) (0.23)
Independent directors 0.97 1.34* 1.56 0.17
(0.71) (0.71) (1.05) (0.45)
Board size 0.014 0.014 "0.081 "0.060*
(0.047) (0.062) (0.072) (0.030)
CEO tenure 0.036** "0.028 0.10** 0.042**
(0.017) (0.022) (0.041) (0.019)
CEO age 0.0011 0.077*** "0.071*** "0.030***
(0.016) (0.019) (0.024) (0.011)
COB tenure 0.043 0.11*** "0.029 "0.021
(0.028) (0.032) (0.056) (0.021)
COB age 0.004 "0.006 0.008 "0.001
(0.011) (0.012) (0.018) (0.007)
COB previous CEO 0.93*** "0.45 1.32*** 0.51***
(0.26) (0.31) (0.42) (0.16)
COB main shareholder "0.38 "0.34 "0.019 0.081
(0.26) (0.35) (0.40) (0.17)
COB founder "0.65* 0.16 "0.89 "0.36
(0.38) (0.41) (0.59) (0.23)
COB in committee "0.56** 0.73*** "1.42*** "0.48***
(0.24) (0.27) (0.36) (0.15)
Industry fixed effects yes yes yes yes
Years fixed effects yes yes yes yes
Constant 6.49*** 3.36 1.11 "3.73**
(2.04) (2.48) (3.48) (1.44)

Adj. R2 0.55 0.44 0.60 0.56

Max VIF 3.29 3.29 3.29 3.45

receives less total and incentive compensation but more fixed compensation. Specifically, this paper questioned the validity of the
salary. An interpretation of this result is that the COB receives more predictions of optimal contracting theory versus managerial power
fixed pay for his/her contribution to the committee, but does not theory in the French context (Dey et al., 2011; Van Essen et al.,
use his/her position in the committee to push for more incentive 2015). Our results show that the choice of a combined structure
compensation. is associated with higher CEO compensation. Overall, our findings
tend to support optimal contracting theory, meaning that the CEO
compensation contract is used as a means to reduce agency con-
6. Discussion
flicts. We suggest that board structure is likely to play a key role in
the protection of minority shareholders’ interests and avoid their
This research examines the relationship between board struc-
expropriation through excessive CEO compensation.
ture choice and leadership compensation by considering the
This finding suggests that combined structures entail additional
chairperson's compensation contract. To our knowledge, this is the
costs for firms. However, our study extends previous research by
first study to take the COB's compensation into account when
taking COB compensation into account and focusing on leadership
addressing the implications of board leadership structure in terms
compensation. Our paper concurs with the analysis made by
of potential costs. The paper aims to contribute to the literature
Brickley et al. (1997) or Dey et al. (2011), who recommend inves-
investigating the costs and benefits of duality choice, with a focus
tigating the costs and benefits of alternative leadership structures
on the specific French context. The study of this context sheds light
in greater detail. We argue that it is important to consider COB
on an institutional environment characterized by concentrated
compensation and provide new insights into the financial cost of
ownership and weak protection of investors, and by a “mixed”
separating titles. Our results show that overall, the cost linked to
board structure model allowing some freedom in board structure
leadership compensation is equivalent in firms with a combined or
selection.
a separate structure, because the separation of these two positions
Consistent with the previous literature on duality, we first
entails COB compensation arrangements. Specifically, we suggest
examined the relationship between leadership duality and CEO
348 G. Broye et al. / European Management Journal 35 (2017) 336e350
that incentives for the chairperson are necessary to avoid potential
agency conflicts between shareholders and the chairperson. How-
ever, we have no evidence that this incentive compensation is
related to firm performance. By evaluating the specific cost asso-
ciated with leadership compensation, we conclude that the two
types of structure are equivalent in terms of compensation costs.
Overall, our study highlights that this dimension must be taken into
consideration when studying the trade-offs inherent to board
structure choice.
This paper has also implications for cross-national governance
research (Aguilera et al., 2008; Schiehll & Martins, 2016; Van Essen
et al., 2013). Our study highlights the necessity to consider the
institutional context in which board structure models can be
selected and compensation policy for both CEOs and COBs can be
shaped. Even if the role played by the incentive compensation
could be considered weakened by the high ownership concentra-
tion that is inherent to the French context, this study reveals that
this mechanism is still used in France, with CEO compensation ar-
rangements that are linked to market performance. Our results also
suggest that board structure significantly impacts compensation
policy in France. Our results also evidence the interactions between
compensation incentives, board composition and ownership
structure. In the French context we find some similar comple-
mentary and substitution effects to those previously reported in the
Anglo-Saxon framework.
Our findings have practical implications for regulators and
practitioners who publicly debate the benefits and costs of duality,
and commonly advocate the separation of titles as the best practice
to ensure “good” corporate governance. This study sheds new light
on the outcomes of board structure choice by highlighting an
under-researched aspect associated with the financial cost of
duality for firms. We believe that this research might be useful in
helping firms to understand the different implications of their
board structure choice. It may also help regulators in the European
Union to identify the strengths and weaknesses of alternative
models given the recent movement toward greater freedom of
choice about board structure among member states (Belot et al.,
2014). Specifically, our findings can be of use to practitioners
seeking a better understanding of the interdependences between
board composition and the compensation practices implemented
for both the CEO and the COB.
This analysis also provides regulators and practitioners with
new insights into the determinants of COB pay within the French
business context, where the exact role played by the chairperson
and his/her level of pay is a current source of controversy. The
French market authority recommends the removal of variable and
equity-based compensation for chairpersons (AMF, 2015). This
position may seem surprising in view of literature demonstrating
that the incentive compensation of directors positively impacts
firm performance. The AMF recommendation, however, is primarily
based on an insufficient grasp of the precise role played by French
chairpersons, and a lack of clarity about how this role is linked to
the level of compensation. Companies rarely communicate about
the specific role played by the chairpersons and to what extent
their contribution has an influence on the strategic direction and
performance of the firm. It would therefore be worthwhile to clarify
the respective roles and responsibilities of the CEO and the COB
(Proxinvest, 2013; Saint-Geours, 2009). Overall, this study might
encourage firms to consider how the COB contributes to strategy-
making and firm performance, and to set compensation that is
sufficiently attractive to achieve this.
6.1. Limitations
Our study has several limitations. Comparative governance
research has largely documented that the national legal and insti-
tutional environment affects the degree of complementarity or
substitutability among different firm-level governance mecha-
nisms (Schiehll and Martins, 2016), and their specific costs (Doidge
et al., 2007). Since this study draws on French firms from an
external context with weak investor protection and high ownership
concentration, it may not be possible to generalize its conclusions
to other environmental contexts. We might expect the compensa-
tion cost of duality to be higher in the Anglo-American context,
which is characterized by higher reliance on incentives for the CEO
and by more dispersed ownership.
Another limitation of this study is that it focuses on the
compensation cost of separating titles, but does not address other
types of costs. Aguilera et al. (2008) argue that the implementation
of corporate governance practices may have several types of asso-
ciated costs, such as costs of compliance, opportunity costs, pro-
prietary costs, or reputational costs. Furthermore, while focusing
on the financial cost of separating titles, our study overlooks po-
tential benefits, such as improved corporate performance (Dey
et al., 2011; Krause & Semadeni, 2013).
This paper also has limitations regarding the choice of variables
used to explain compensation elements. Although we controlled
for some characteristics of CEOs and COBs in our models, we did not
include critical personal characteristics associated to the human
and social capital of leaders (Combs & Skill, 2003; Damiani & Ricci,
2014; Hambrick & Mason, 1984; Harris & Helfat, 1997). This issue
requires a thorough analysis in the French context given the specific
educational background of CEOs, the particular path of access to top
executive positions via prestigious “grandes ! ecoles” and civil service
“corps”, and the high levels of social capital of elite CEOs (Goyer &
Jung, 2011). Additionally, this study focused solely on CEO and
COB characteristics, and did not consider the influence of board
members, or top management team specificities such as social-
psychological factors, on the pay setting process (Fredrickson,
Davis-Blake, & Sanders, 2010).
6.2. Future research
This paper focuses on compensation cost, a specific dimension
of the consequences of separating the CEO and COB roles in firms
within the French national context. Further research is needed to
better understand the global effect of duality from a multidimen-
sional perspective in the light of the overall costs and benefits of
splitting the roles.
Most importantly, further research is required to estimate the
overall implications of duality across countries. Cross-country
analysis should provide a more context-dependent understanding
of outcomes associated with the choice of board structure. Previous
literature provides evidence that institutional settings considerably
affect these outcomes. For instance, the board model adopted in a
particular institutional context influences the advisory role of the
board, the board's ability to exchange information and knowledge
resources, and the extent to which boards can contribute to the
strategy-making process (Adams & Ferreira, 2007; Heyden,
Oehmichen, Nichting, & Volberda, 2015). It is now necessary to
investigate how institutional aspects also affect the role played by
the chairperson and his/her contribution to the firm e from both a
monitoring and an advisory perspective. Future research might also
benefit from the use of our framework in different external settings
to assess the influence of national contexts on COB compensation
policy and the cost of duality.
7. Conclusion
Our study contributes to the literature investigating duality
 G. Broye et al. / European Management Journal 35 (2017) 336e350
choice and enriches comparative corporate governance research by
providing a contextualized study on how the choice of duality af-
fects leadership compensation. Our findings show that COB
compensation is likely to influence the cost of duality, and this
dimension should be considered in future research concerning the
advantages and disadvantages of splitting the positions of CEO and
COB.
Acknowledgements
The authors would like to thank the editor, Sabina Siebert, the
associate editor, Mariano Heyden, and the two anonymous re-
viewers for their insightful comments and suggestions.
References
Adams, R., & Ferreira, D. (2007). A theory of friendly boards. Journal of Finance, 62,
217e250.
AFEP/MEDEF. (2013). Code de gouvernement d’entreprise des soci! et!
es cot!
ees.
Aguilera, R. V., Filatotchev, I., Gospel, H., & Jackson, G. (2008). An organizational
approach to comparative corporate governance: Costs, contingencies, and
complementarities. Organization Science, 19(3), 475e492.
Aguilera, R. V., & Jackson, G. (2003). The cross-national diversity of corporate
governance: Dimensions and determinants. Academy of Management Review,
28(3), 447e465.
Anderson, C. A., & Anthony, R. N. (1986). The new corporate directors: Insights for
board members and executives. Wiley.
Autorite! des Marche !s Financiers. (2014). Rapport de l’AMF sur le gouvernement
d’entreprise et la re !mune !ration des dirigeants.
Autorite! des Marche !s Financiers. (2015). Rapport de l’AMF sur le gouvernement
d’entreprise et la re !mune !ration des dirigeants.
Baber, W., Janakiraman, S., & Kang, S.-H. (1996). Investment opportunities and the
structure of executive compensation. Journal of Accounting and Economics, 21,
297e318.
Baliga, R., Moyer, C., & Rao, R. (1996). CEO duality and firm performance: What's the
fuss? Strategic Management Journal, 17, 41e43.
Bebchuk, L., & Fried, J. (2003). Executive compensation as an agency problem.
Journal of Economic Perspectives, 17(3), 71e92.
Bebchuk, L., & Fried, J. (2004). Pay without performance: The unfulfilled promise of
executive compensation. Cambridge: Harvard University Press.
Belot, F., Ginglinger, E., Slovin, M., & Sushka, M. (2014). Freedom of choice between
unitary and two-tier boards: An empirical analysis. Journal of Financial Eco-
nomics, 112, 364e385.
Boyd, B. (1994). Board control and CEO compensation. Strategic Management Jour-
nal, 15(5), 335e344.
Boyd, B. (1996). Determinants of US outside director compensation. Corporate
Governance: An International Review, 4(4), 202e211.
Brickley, J., Coles, J., & Jarrell, G. (1997). Leadership structure: Separating the CEO
and chairman of the board. Journal of Corporate Finance, 3, 189e220.
Broye, G., & Moulin, Y. (2010). Re !mune!ration des dirigeants et gouvernance des
entreprises: Le cas des entreprises françaises cote !es. Finance Contro^le Strat!
egie,
13(1), 67e98.
Broye, G., & Moulin, Y. (2014). La re !mune !ration des pre!sidents non exe !cutifs
de!pend-elle de leur capital humain? Management International, 19(1), 204e218.
Capezio, A., Shields, J., & O'Donnell, M. (2011). Too good to be true: Board structural
independence as a moderator of CEO pay-for-firm-performance. Journal of
Management Studies, 48(3), 487e513.
Chhaochharia, V., & Grinstein, Y. (2009). CEO compensation and board structure.
Journal of Finance, 64(1), 231e261.
Combs, J. G., & Skill, M. S. (2003). Managerialist and human capital explanations for
key executive pay premiums: A contingency perspective. Academy of Manage-
ment Journal, 46(1), 63e73.
Conyon, M., & Peck, S. (1998). Board control, compensation committees, and top
management compensation. Academy of Management Journal, 41(2), 146e157.
Cordeiro, J., & Veliyath, R. (2003). Beyond pay for performance: A panel study of the
determinants of CEO compensation. American Business Review, 21(1), 56e66.
Cordeiro, J., Veliyath, R., & Romal, J. (2007). Moderators of the relationship between
director stock-based compensation and firm performance. Corporate Gover-
nance: An International Review, 16(6), 1384e1393.
Core, J. E., Guay, W. R., & Larcker, D. F. (2003). Executive equity compensation and
incentives: A survey. Economic Policy Review, 9(1), 27e50.
Core, J. E., Holthausen, R. W., & Larcker, D. F. (1999). Corporate governance, chief
executive officer compensation, and firm performance. Journal of Financial
Economics, 51(3), 371e406.
Croci, E., Gonenc, H., & Ozkan, N. (2012). CEO compensation, family control, and
institutional investors in Continental Europe. Journal of Banking & Finance,
36(12), 3318e3335.
Cyert, R. M., Kang, S. H., & Kumar, P. (2002). Corporate governance, takeovers, and
top-management compensation: Theory and evidence. Management Science,
48(4), 453e469.
349
Damiani, M., & Ricci, A. (2014). Managers' education and the choice of different
variable pay schemes: Evidence from Italian firms. European Management
Journal, 32(6), 891e902.
Desender, K. A., Aguilera, R. V., Crespi, R., & Garcia-Cestona, M. (2013). When does
ownership matter? Board characteristics and behavior. Strategic Management
Journal, 34(7), 823e842.
Deutsch, Y. (2007). The influence of outside directors' stock-option compensation
on firms' R&D. Corporate Governance: An International Review, 15(5), 816e827.
Dey, A., Engel, E., & Liu, X. (2011). CEO and board chair roles: To split or not to split?
Journal of Corporate Finance, 17(5), 1595e1618.
Doidge, C., Karolyi, G. A., & Stulz, R. M. (2007). Why do countries matter so much for
corporate governance? Journal of Financial Economics, 86(1), 1e39.
Expert Corporate Governance Service (ECGS). (2012). Directors' pay 2011: An inter-
national comparison.
Faccio, M., & Lang, L. H. (2002). The ultimate ownership of Western European
corporations. Journal of Financial Economics, 65(3), 365e395.
Faleye, O. (2007). Does one hat fit all? The case of corporate leadership. Journal of
Management & Governance, 11, 239e259.
Fama, E., & Jensen, M. (1983). Separation of ownership and control. Journal of Law
and Economics, 26, 301e326.
Fernandes, N., Ferreira, M., Matos, P., & Murphy, K. (2013). Are US CEOs paid more?
New international evidence. Review of Financial Studies, 26(2), 323e367.
Finkelstein, S. (1992). Power in top management teams: Dimensions, measurement,
and validation. Academy of Management Journal, 35(3), 505e538.
Finkelstein, S., & D'Aveni, R. (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), 1079e1108.
Firth, M., Fung, P., & Rui, O. (2007). How ownership and corporate governance in-
fluence chief executive pay in China's listed firms. Journal of Business Research,
60, 776e785.
Florou, A. (2005). Top director shake-up: The link between chairman and CEO
dismissal in the UK. Journal of Business Finance & Accounting, 32(1), 97e128.
Fraile, I. A., & Fradejas, N. A. (2014). Ownership structure and board composition in
a high ownership concentration context. European Management Journal, 32(4),
646e657.
Fredrickson, J. W., Davis-Blake, A., & Sanders, W. M. (2010). Sharing the wealth:
Social comparisons and pay dispersion in the CEO's top team. Strategic Man-
agement Journal, 31(10), 1031e1053.
Goyer, M., & Jung, D. K. (2011). Diversity of institutional investors and foreign
blockholdings in France: The evolution of an institutionally hybrid economy.
Corporate Governance: An International Review, 19(6), 562e584.
Grabke-Rundell, A., & Gomez-Mejia, L. R. (2002). Power as a determinant of exec-
utive compensation. Human Resource Management Review, 12(1), 3e23.
Hambrick, D. C., & Mason, P. A. (1984). Upper echelons: The organization as a
reflection of its top managers. Academy of Management Review, 9(2), 193e206.
Harris, D., & Helfat, C. (1997). Specificity of CEO human capital and compensation.
Strategic Management Journal, 18(11), 895e920.
Heidrick, Struggles (2011). European corporate governance report 2011.
Hermalin, B. E., & Weisbach, M. S. (1998). Endogenously chosen boards of directors
and their monitoring of the CEO. American Economic Review, 88(1), 96e118.
Heyden, M. L., Oehmichen, J., Nichting, S., & Volberda, H. W. (2015). Board back-
ground heterogeneity and exploration-exploitation: The role of the institu-
tionally adopted board model. Global Strategy Journal, 5(2), 154e176.
Hill, C., & Phan, P. (1991). CEO tenure as a determinant of CEO pay. Academy of
Management Journal, 34(3), 707e717.
Jensen, M. (1993). The modern industrial revolution, exit, and the failure of internal
control systems. Journal of Finance, 48(3), 831e880.
Jensen, M., & Meckling, W. (1976). Theory of the firm: Managerial behavior, agency
costs and ownership structure. Journal of Financial Economics, 3(4), 305e360.
Jensen, M., & Murphy, K. (1990). Performance pay and top-management incentives.
Journal of Political Economy, 98, 225e264.
Karabadse, A., & Karabadse, N. (2007). The return of the chairman. Business Strategy
Review, Winter, 60e65.
Krause, R., & Semadeni, M. (2013). Apprentice, departure, and demotion: An ex-
amination of the three types of CEO-board chair separation. Academy of Man-
agement Journal, 56(3), 805e826.
La Porta, R., Lopez-de-Silanes, F., Shleifer, A., & Vishny, R. (2000). Investor protection
and corporate governance. Journal of Financial Economics, 58(1), 3e27.
Leblanc, R. (2005). Assessing board leadership. Corporate Governance: An Interna-
tional Review, 13(5), 654e666.
Linck, J. S., Netter, J. M., & Yang, T. (2008). The determinants of board structure.
Journal of Financial Economics, 87(2), 308e328.
Mehran, H. (1995). Executive compensation structure, ownership, and firm per-
formance. Journal of Financial Economics, 38(2), 163e184.
Murphy, K. J. (1999). Executive compensation. Handbook of Labor Economics, 3,
2485e2563.
Proxinvest. (2013). La r! emun! eration des dirigeants des soci! es du SBF (Vol. 120).
et!
Roberts, J. (2002). Building the complementary board. The work of the Plc
chairman. Long Range Planning, 35, 493e520.
Ryan, H. E., & Wiggins, R. A. (2004). Who is in whose pocket? Director compen-
sation, board independence, and barriers to effective monitoring. Journal of
Financial Economics, 73, 497e524.
Saint-Geours, J. P. (2009). Gouvernement d’entreprise: Le bice !phalisme en question.
Lettre de l’IFA, 19, 16e17.
Schiehll, E., & Martins, H. C. (2016). Cross-national governance research: A
350 G. Broye et al. / European Management Journal 35 (2017) 336e350
systematic review and assessment. Corporate Governance: An International Re-
view, 24(3), 181e199.
Shleifer, A., & Vishny, R. W. (1997). A survey of corporate governance. The Journal of
Finance, 52(2), 737e783.
Tosi, H., Werner, S., Katz, J., & Gomez-Mejia, L. (2000). How much does performance
matter? A meta-analysis of CEO pay studies. Journal of Management, 26(2),
301e339.
Tuggle, C. S., Sirmon, D. G., Reutzel, C. R., & Bierman, L. (2010). Commanding board
of director attention: Investigating how organizational performance and CEO
duality affect board members' attention to monitoring. Strategic Management
Journal, 31(9), 946e968.
Van Essen, M., Engelen, P. J., & Carney, M. (2013). Does “good” corporate governance
help in a crisis? The impact of country- and firm- level governance mechanisms
in the European financial crisis. Corporate Governance: An International Review,
21(3), 201e224.
Van Essen, M., Otten, J., & Carberry, E. J. (2015). Assessing managerial power theory:
A meta-analytic approach to understanding the determinants of CEO
compensation. Journal of Management, 41(1), 164e202.
Waelchli, U., & Zeller, J. (2013). Old captains at the helm: Chairman age and firm
performance. Journal of Banking & Finance, 37, 1612e1628.
Weisbach, M. S. (2007). Optimal executive compensation versus managerial power:
A review of Lucian Bebchuk and Jesse Fried's pay without performance: The
unfulfilled promise of executive compensation. Journal of Economic Literature,
45(2), 419e428.
Westphal, J. D., & Zajac, E. J. (1995). Who shall govern? CEO/board power, de-
mographic similarity, and new director selection. Administrative Science Quar-
terly, 40, 60e83.
White, H. (1980). A heteroskedasticity-consistent covariance matrix estimator and a
direct test for heteroscedasticity. Econometrica, 48(4), 817e838.
Yang, T., & Zhao, S. (2014). CEO duality and firm performance: Evidence from an
exogenous shock to the competitive environment. Journal of Banking & Finance,
49, 534e553.
Yermack, D. (2004). Compensation, retention, and reputation incentives for outside
directors. Journal of Finance, 59, 2281e2308.