Journal of Cleaner Production 135 (2016) 760e770

Contents lists available at ScienceDirect

Journal of Cleaner Production

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

Corporate social responsibility as an entrenchment strategy, with a

focus on the implications of family ownership

zaro Rodrguez-Ariza b, Isabel-Mara Garca-Sa

nchez a
Jennifer Martnez-Ferrero a, *, La
a
b

Multidisciplinary Institute for Enterprise (IME), University of Salamanca, Campus Miguel de Unamuno, FES, 37007, Salamanca, Spain

micas y Empresariales, Departamento de economa nanciera y contabilidad, Universidad de Granada, Spain
Facultad de Ciencias Econo

a r t i c l e i n f o

a b s t r a c t

Article history:
Received 17 May 2015
Received in revised form
30 April 2016
Accepted 21 June 2016
Available online 28 June 2016

Taking as its basis the classical agency conict between owners and managers, this article investigates
issues of managerial discretion, entrenchment and corporate social responsibility (CSR) in family rms.
Using an international sample, its purpose is to examine the promotion of CSR as a strategic shield
against the costs of managerial discretion and to determine whether this use of CSR is moderated by
family ownership. The results obtained support the argument that CSR may provide managers who
manipulate earnings, as a discretionary practice, with the opportunity to entrench themselves. This
would be an outcome of the decrease in activism and surveillance by stakeholders whose social and
environmental demands are satised by the exercise of CSR. Thus, by satisfying stakeholders' demands,
managers can use a socially responsible strategy as a mechanism for self-defence. Moreover, our results
show that CSR is moderated by the ownership structure of family rms. Family owners serve as active
monitors of managers, thus alleviating the classical agency problem and decreasing both the risks
associated with managerial discretion and the use of CSR as entrenchment.
2016 Elsevier Ltd. All rights reserved.

Keywords:
Corporate social responsibility
Entrenchment strategy
Managerial discretion
Family rms

1. Introduction
From the perspective of the agency conict between owners and
managers (Shleifer and Vishny, 1989), this paper examines the
possible existence of an entrenchment mechanism based on actions of corporate social responsibility (hereinafter CSR); CSR has
been dened as the corporate integrated responsibilities encompassing the economic, legal, ethical and discretionary expectations
that the society has of organizations (Carroll, 1979).
The rationale for this study is the separation between property
and control that is the basis for the agency theory; according to this
theory, a shareholder (the principal) delegates the management of
the rm to a manager (the agent). The latter acts for the former.
However, because of their conict of interests between the principal and the agent and the differences in their access to information, and because it is difcult for the principal to check on the
manager's activities (Jensen and Meckling, 1976), the central
assumption of this paper is that the self-interest of the agent results
in opportunistic and/or discretionary behaviour. This behaviour

* Corresponding author.
E-mail addresses: jenny_marfe@usal.es (J. Martnez-Ferrero), lazaro@ugr.es

nchez).
(L. Rodrguez-Ariza), lajefa@usal.es (I.-M. Garca-Sa
http://dx.doi.org/10.1016/j.jclepro.2016.06.133
0959-6526/ 2016 Elsevier Ltd. All rights reserved.

arises from an excessive autonomy in decision-making processes

that gives managers the opportunity to pursue their own selfinterest (Eisenhardt, 1989) rather than the corporate benet.
By exploring this discretionary behaviour and the agency cost
for CSR, we hypothesise that managers could over-invest in social
and environmental concessions as a self-defence strategy (Rowley

n, 2004). Their aim is to
and Berman, 2000; Schneper and Guille
ensure their job security, to strengthen their position, to avoid
stakeholders' reactions, and, overall, to pre-empt the costs of
managerial discretion. Thus, as a hedging strategy to avoid stakeholders' negative reactions, for example through costly boycotts
and lobbying, media campaigns, or greater activism and scrutiny

, 2008), managers
(Pagano and Volpin, 2005; Surroca and Tribo
could satisfy stakeholders' demands by following CSR practices
and, in this way, expropriating shareholder wealth (Cespa and

, 2008).
Cestone, 2007; Prior et al., 2008; Surroca and Tribo
In addition to addressing the possibility that CSR practices may
mask practices of managerial discretion and facilitate entrenchment (the expropriation of wealth from shareholders), this study
makes a contribution with its focus on family ownership as a
possible control mechanism that underlies the relationship
mentioned above (La Porta et al., 1998). It has been suggested that
the presence of blockholders may constitute a mechanism that

J. Martnez-Ferrero et al. / Journal of Cleaner Production 135 (2016) 760e770

inhibits this type of entrenchment practice (La Porta et al., 1998;

, 2008). In this respect, family rms are, in genSurroca and Tribo
eral, characterised by a high concentration of ownership in the
family, who are often majority shareholders and can perform
various functions in the management of the rm (Haalien and
Huse, 2005). Thus, although family members can act as shareholders and as decision-makers, this paper only examines family
ownership as a control mechanism, focusing on the special agency
problem that appears in family-owned rms, where family members are majority shareholders (Chen et al., 2008). In this situation,
the classical agency problem between managers and shareholders
is diminished because the shareholders have more information and
can monitor the managers more closely than in a non-family
business (Chau and Gray, 2010; Chen et al., 2008; Chrisman et al.,
2004). Therefore, as a secondary research goal, we examine the
moderating effect of family ownership.
In summary, this paper addresses the following questions: (1)
Does the agency conict between owners and managers promote
managerial discretion? (2) Are CSR-oriented policies used as a tool
for managerial self-defence? (3) Compared to non-family rms,
does family ownership minimise agency problems and thus the use
of CSR as an entrenchment mechanism?
The remainder of this paper is structured as follows. In Section 2,
we describe the theoretical background that supports our research
hypotheses. Section 3 describes the research model, data and
sample. Finally, Sections 4 and 5 present the results obtained and
the conclusions drawn, respectively.
2. Research hypotheses
2.1. Theoretical background: Agency theory
Most large companies are owned by a multitude of shareholders
and investors. Such companies are characterised by a clear separation between property and control which, in turn, leads to conicting interests, to different levels of risk aversion between
shareholders and managers, and, moreover, to different abilities to
access information, which limits the verication of the managers'
activities (Jensen and Meckling, 1976; Jones, 1995). According to
agency theory, this separation can lead to conicts of interest: (i)
between shareholders and company managers (the Type 1 agency
problem), as a result of which managers' decisions may not always
coincide with shareholders' objectives (Jensen and Meckling, 1976);
and (ii) between minority and majority shareholders (the Type 2
agency problem), which can generate expropriation issues as a
result of information asymmetries between the two groups
(Shleifer and Vishny, 1989).
The fundamental concern of this study is to investigate managerial discretion behaviour, which is an aspect of the potential
agency conict between owners and managers (Type 1) under the
basic assumption of the self-interest of managers and owners.
But, rst, in order to put this concern into context, the question
that may be formulated is the following: do the managers' interests
run parallel to the owners' interests? Considering the degree of
freedom and discretion permitted to managers in order to
encourage corporate decisions (Langfred, 2004), that is, their autonomy in the decision-making process, Shimizu (2012) distinguishes between two opposing types of managerial behaviour with
respect to shareholders' interests: (1) from a positive point of view,
managers need to be encouraged in autonomous behaviour so that
they pursue corporate entrepreneurship as a mechanism for
exploring new strategic decisions outside the scope of the current
strategy; and (2) from a negative point of view, as an agency
problem, excessive autonomy gives managers the opportunity to
pursue their self-interest (Eisenhardt, 1989). A high degree of

761

autonomy can lead managers to act opportunistically, being more

concerned with their own interests than in the corporate strategy
(Shimizu, 2012).
According to the second perspective proposed by Shimizu
(2012), on which this paper is focused, when shareholders' and
managers' interests are not well aligned, managers may take
advantage of their position and their wide managerial autonomy
and support actions that promote their own benet. This implication, derived from the separation of the attributes of ownership and
control, was initially reected in the research of Berle and Means
(1932), who termed it managerial discretion. According to these
authors, agency conict is the basis for managerial discretion,
which is seen as the opportunity for managers in the decisionmaking process to serve their own objectives rather than the objectives of their principals by, for example, using investors' funds to
obtain private benets (such as goods for their personal use or
excessive remuneration) or making investment decisions that are
prejudicial to the shareholders' interests (Healy and Palepu, 2001).
Moreover, in this agency context, it is necessary to recognise
that there are some mechanisms by which managers can retain
control and protect their jobs, personal status and prestige, and
lessen the monitoring by the shareholders (Florackis and Ozkan,
2009) while continuing their discretionary behaviour. That is,
there are mechanisms that allow managers to preserve their private benets in a way that runs counter to the maximisation of the
owners' wealth (Shleifer and Vishny, 1989).
From this type of agency problem e the conict of interests that
arises from the separation of the attributes of ownership and
control e and in a context in which there is information asymmetry
between managers and owners, one area that merits attention is
that of the power of the shareholders to control the managers; the
most important instrument here is the vote. More concretely,
attention should be given to the role played a specic interest
group, namely the family shareholders in a family rm. Following
Chen et al. (2008), we dene a family rm as one in which the
family founders remain in senior managerial positions, are present
on the board or are able to act as blockholders. Nonetheless, as
regards the amount of information available to the different parties,
family rms are not homogeneous. Majority shareholders tend to
have access to more information than minority shareholders (Ali
et al., 2007; Chen et al., 2008; Landry et al., 2013). Moreover,
since in a family rm the majority shareholders are actively
involved in both controlling and monitoring managers, there is less
information asymmetry between them and the managers than in
non-family rms (Chen et al., 2008). Accordingly, the classical
agency problem between managers and shareholders is smaller,
which will have an impact on the ability of managers to use their
discretion in decision-making processes; family shareholders have
more information and can control managers more effectively than
the shareholders in a non-family business (Chen et al., 2008;
Chrisman et al., 2004).1
In summary, the conict of interests between a principal and an
agent (a Type 1 agency conict) is of fundamental interest in this

1
Note that an exceptional agency problem arises in a family business when there
are family and non-family investors (Chau and Gray, 2010), i.e. majority and minority shareholders. Family members have access to more information than outsiders, since they participate actively in most business activities, and so Type 2
agency problems can occur. Such a conict between the majority and the minority
shareholders' interests could lead to one or the other of the two perspectives
proposed by Cho et al. (2013): the adverse selection effect and the information
efciency effect. According to the former, family members may act opportunistically and aggravate information differences; according to the latter, more informed
investors can disseminate information to other investors, thus reducing information asymmetries.

762

J. Martnez-Ferrero et al. / Journal of Cleaner Production 135 (2016) 760e770

study. In the following sections, we examine the context, manifestations and implications of this conict, and consider whether it
might be moderated by the smaller information asymmetries in
family-owned rms.
2.2. CSR as a strategic shield against the costs of managerial
discretion
One of the most costly manifestations of the conict between
managers and shareholders (Type 1 agency conict) is the result of
managerial discretion. In 1932, Berle and Means dened this
agency cost as any behaviour arising from the conict of interests
between owners and managers as a result of which the latter can
serve their own pockets better by proting at the expense of the
company than by making prots for it (p. 125). Thus, from an
agency perspective, where the principal and agent have different
interests, maximising shareholder value is not the main priority of
managers when they are allocating investment funds; managerial
decisions grow from managers' basic desire to obtain benets for
themselves (Berle and Means, 1932). These benets are power,
status, salary, security, power and prestige (Barnard, 1962). In many
cases, managers exercise this discretion by recommending investments that do not maximise shareholder value but instead
increase their own remuneration (Williamson, 1963). Alternatively,
managers may seek to maximise sales (Baumol, 1959) in order to
increase the funds available for their discretionary use. Activities
like these that are carried out with the aim of beneting managers
are subject to the constraint that satisfactory levels of prot must
be achieved and thus performance pressures circumvented, and
they are associated with commitments to disburse large amounts of
cash (Jensen, 1986) and to reduce the possibility of the rm
becoming the object of a takeover (Stulz, 1988).
Preliminary evidence has highlighted the detrimental consequences of the conict of interests that leads to managerial
discretion behaviour. Among other negative impacts, the aim of
managers to pursue their own rather than the shareholders' objectives can decrease corporate value and prejudice the rm's
reputation and corporate image (Roychowdhury, 2006), can provoke a loss of support and trust, and can increase activism and
surveillance by interest groups and regulatory authorities (Zahra
et al., 2005). Moreover, when rm owners identify that managers
have personal objectives that do not serve the goal of maximising
corporate benet, they respond to this exercise of managerial
discretion by penalising these top executives and, thus, increasing
the rm's turnover risk (Rowley and Berman, 2000) and their

n, 2004) and provoking
possible replacement (Schneper and Guille
costly boycotts, lobbying and media campaigns (Prior et al., 2008;

, 2008).
Surroca and Tribo
In view of these possible consequences, and seeking to reduce
the risk of being sanctioned for the opportunism of satisfying their
own interests, managers may choose to entrench (Shleifer and
Vishny, 1989), which should be understood here as expropriating
wealth from the shareholders (Florackis and Ozkan (2009). With
this understanding, managerial entrenchment remains as another
agency cost arising from the conict of interest between principal
and agent and from the earlier exercise of managerial discretion.
Through entrenchment, managers who have given higher priority
to meeting their own goals rather than the corporate goals preserve
their position and avoiding close scrutiny by activist stakeholders
(Cespa and Cestone, 2007).
In the context of this conict of interests, Pawlina and Renneboog (2005) argue that managers prefer overinvestment as a
mechanism for entrenchment. Among the possible available strategies for overinvestment, and as a hedging strategy against disciplinary initiatives, managers may increase the nancial resources

allocated to social and environmental concessions, presenting this

, 2008).
as an exercise of CSR (Prior et al., 2008; Surroca and Tribo
The premise of the overinvestment in CSR actions is consistent with
the implicit argument of instrumental theory (Jones, 1995), that
certain types of socially responsible performance can be promoted
as a means of building trusting and cooperative relationships with
stakeholders who control key resources that will ensure the longterm survival of the rm (Freeman, 1984). Furthermore, the
premise of the possible use of CSR as a self-defence strategy is
consistent with the contention that CSR action usually contributes
to the preservation of the manager's private benets of control
(Cheng et al., 2013): entrenchment is also an agency cost.
In other words, according to agency theory, the principal's
objective is to maximize prots, but the managers have non-prot
objectives. This conict of interests, and the autonomy of managers
in the decision-making process, allows managers to pursue their
own benet rather than the owners' objectives. Proting from information asymmetries between the principal and the agent, they
can obtain personal benets that they try to mask. To put it more
precisely, in order to mask their discretionary behaviour and as a
result of their agency, managers decide to increase the nancial
resources allocated to CSR actions. In doing this they have a dual
goal: (i) to achieve their non-prot aims (for example, to preserve
their jobs, prestige and reputation and to avoid boycotts and
lobbying), and (ii) to avoid the shareholders' identication of their
discretionary actions that run counter to the goal of prot
maximisation.
In accordance with this argument, some authors have adopted
an agency cost perspective on CSR practices, taking a negative view
of the managerial motivations for pursuing CSR. Friedman (1970)
was the rst to suggest concern in this respect, anticipating later
studies that conrmed that managers do indeed make opportunistic use of CSR to advance their careers, to protect their positions
or to obtain personal gain (Kim et al., 2012; Prior et al., 2008). Thus
managers serve their own interests. In this respect, too,
Hemingway and Maclagan (2004) hypothesised that, whereas
there is a conict of interests and information asymmetry between
the principal and the agent, the application of CSR-oriented policies
could divert stakeholders' attention and distract stakeholders from
activism and scrutiny into activities resulting from managerial
discretion. Similarly, Fabrizi et al. (2014) presented evidence of
monetary managerial incentives that go beyond the impact of CSR
actions on prot maximisation, and showed that in order to reinforce their legitimacy in the eyes of the stakeholders, entrenched
CEOs are more likely to invest in CSR. These authors conrmed the

(2008), who reported that
earlier evidence of Surroca and Tribo
entrenched managers satisfy stakeholders using CSR actions as an
entrenchment strategy, and of Gargouri et al. (2010), who showed
that managers opt for CSR practices with the aim of decreasing the
risk of dismissal that results from aggressive discretionary accruals
(a proxy for managerial discretion).
Overall, when considering agency costs, the use of CSR as an
entrenchment strategy can enable managers to reinforce their job
security (Shleifer and Vishny, 1989), and, at the same time, build up
their personal reputations in the community and enhance their
personal status (Fabrizi et al., 2014). Accordingly, managers may
adopt CSR-oriented policies and in this way satisfy their stakeholders' expectations, neutralising possible negative reactions to
perceived discretionary accruals that are the result of the agency

,
conict of interest (Cespa and Cestone, 2007; Surroca and Tribo
2008).
In view of these considerations, and from an agency cost
perspective, we hypothesise that the promotion of CSR practices
provides managers with a powerful tool for complying with
stakeholders' demands (which are certainly different from their

J. Martnez-Ferrero et al. / Journal of Cleaner Production 135 (2016) 760e770

own), and that they thereby avoid negative reactions to their

discretionary behaviour. In other words, as an implication of
agency, CSR is used as an entrenchment mechanism. The following
hypothesis is proposed:
Hypothesis 1: From an agency cost perspective and for their
own benet, managers satisfy stakeholders' demands via CSR actions in order to avoid the costs of managerial discretion.
2.3. The moderating effect of family ownership
In an agency scenario where there is a conict of interest and
information asymmetry, shareholders cannot determine whether
social and environmental concessions are being promoted by
managers as part of an ethical and sustainable business policy or as
a strategy to entrench themselves, serving their own interests. In
this paper, we hypothesise, on the one hand, that entrenched
managers might satisfy stakeholders' demands e their social and
environmental concerns e via CSR in order to reduce or eliminate
any negative impact of their discretionary behaviour. In other
words, to obtain personal benets from their managerial positions,
managers adopt discretionary decisions that they try to mask
through CSR actions e they aim to entrench and, in this way, to
retain their jobs and personal status. But moreover, we hypothesise
on the other hand that there are some mechanisms by which these
agency manifestations could be controlled. For example, internal
mechanisms might detect and/or prevent such an entrenchment
strategy. Such mechanisms could include the presence of block
, 2008), a condition
holders (La Porta et al., 1998; Surroca and Tribo
that is evidently fullled in family rms (Chen et al., 2008). In the
context of family rms, according to Chen et al. (2008), majority
family shareholders can be actively involved both in the daily activities of the rm and in management responsibilities, exerting a
dual pressure on ownership and management. It is necessary to
point out that this paper focuses on examining the family ownership aspect, without assessing the effect of the appointment of
family members to the management or to the board (Jaggi et al.,
2009).
Under this premise, and bearing in mind the level of Type 1
agency conict that may arise in family rms, we follow Shleifer
and Vishny (1989) in arguing that family shareholders can use
their stronger monitoring and supervisory powers to decrease the
risk of entrenchment via CSR. Therefore, although managers have
self-serving interests, the power of the shareholders to control
managers minimizes the risk of managerial discretion and thus of
entrenchment. Their greater control and monitoring of managerial
decisions prevents managers from using their decision-making
discretion to pursue their own interests at the expense of the
shareholder wealth. Although family rms operate in different
ways and constitute a heterogeneous group (Cennamo et al., 2012),
different studies agree that such rms present certain characteristics that decrease the propensity of managers to act in their own
interests and to expropriate wealth from shareholders; see, for
example, Ali et al. (2007), who show that a decrease in Type 1
agency conicts between managers and shareholders is associated
with a lower level of discretionary accruals (our proxy for managerial discretion). Similarly, Jaggi et al. (2009) nd that families can
limit the ability of managers to manipulate earnings, can monitor
managerial behaviour and then can reduce the managerial opportunities to engage in earning management practices.
Therefore, it is to be expected that some aspects of family
ownership will decrease the probability of there being managerial
discretion and of entrenchment, because there will be fewer
manifestations of the agency problem. First, family shareholders
have access to more information than outsiders, which reduces
information asymmetries between managers and majority owners

763

(Chen et al., 2008). Then, in rms characterised by a high level of

ownership in the hands of a family or families, shareholders have
more access to the information that allows them to check and supervise managerial decisions. In other words, there is less of an
agency conict that could lead to managerial entrenchment
through CSR actions. Second, family owners are better placed to
monitor and supervise managers, and so agency problems are
alleviated (Landry et al., 2013), and the risks associated with
managerial discretion are decreased or averted (Khan et al., 2013;
Wang, 2006). Moreover, this effect is reinforced by the majority
owners' greater knowledge about the rm's activities (Anderson
and Reeb, 2003). Third, family rms tend to focus on long-term
investment in order to ensure the family's succession and the
transmission of their legacy (Singal, 2014). There is support for this
from Jiraporn and DaDalt (2009), who demonstrate that family
rms are long-term investors and prevent managers from engaging
in unproductive activities such as manipulating earnings. In
consequence, the different levels of risk aversion arising from
agency theory lead to family owners seeking to prevent shortsighted managerial practices that could compromise the rm's
reputation and even its survival (Anderson and Reeb, 2003). Family
ownership therefore allows shareholders to preserve their wealth
and prevents managers from satisfying their personal interests and
then masking their discretionary behaviour via entrenchment tools
such as CSR. Family ownership limits agency costs.
As well as being less liable to Type 1 agency conict and better
protected against the risks of managerial discretion and of CSR as
an entrenchment strategy, family rms tend to promote CSR as an

niz and Cabrera, 2005) rather than as an
ethical option (De
entrenchment strategy. Although the evidence reported in prior
studies is mixed, most authors agree that family owners assign a
high priority to following CSR practices to comply with stakeholders' expectations (Cruz et al., 2014), and are more attentive to
social issues and their stakeholders than are non-family owners
(Van Gils et al., 2014). Many researchers have explained this by
referring to the tendency of family owners to take a longer-term
view of their investments (Berrone et al., 2012), to take greater
account of questions of image, reputation, identication and so on
(Dyer and Whetten, 2006) and, ultimately, to seek to preserve their
own socio-emotional wealth (Cennamo et al., 2012). These are the
affective endowments referred to by Berrone et al. (2012).
Furthermore, a family business is traditionally expected to be
proactive in developing connections with stakeholders, acting as a
good steward in the community in which it is operating, and caring

mezfor employees' welfare and the working environment (Go
Meja et al., 2007; Bammens et al., 2014).
Overall, as Florackis and Ozkan (2009) argue, the existence of
controlling shareholders may have an impact on the level of
managerial entrenchment as a manifestation of the agency problem. That is, it is expected that family shareholders will control and
monitor managerial actions more closely, decreasing information
asymmetries, and minimising the opportunity for managers to
serve their personal objectives. They thus decrease the risks associated with managerial discretion. Moreover, it would be expected
that these rms would make greater social and environmental
concessions, doing so in accordance with their socio-emotional
wealth, the non-economic aspects that provide utility to family

mez-Meja et al., 2007). Accordingly, we expect that
members (Go
family ownership would moderate managers' use of CSR as a selfdefence strategy, because of the lower agency costs (managerial
discretion and entrenchment) of these rms. The following hypothesis is proposed:
Hypothesis 2: From an agency cost perspective and because
there is less of a conict of interests between majority shareholders and managers, family ownership moderates the use of CSR

764

J. Martnez-Ferrero et al. / Journal of Cleaner Production 135 (2016) 760e770

as an entrenchment strategy against the costs of managerial

discretion.
3. Method
3.1. Population and sample
The data source was formed using two databases: (i) Thomson
One Analytic, for the accounting and nancial information provided
in consolidated nancial statements; (ii) the Ethical Investment
Research Service (EIRIS), for data on corporate social responsibility.
Following La Porta et al. (2002), nancial rms were excluded from
our sample due to the different characteristics of their equity, and
because they are not comparable to non-nancial rms. We also
excluded companies whose listed company status varied, i.e. the
sample included only companies that remained listed throughout
the sample period (2002e2010). The nal sample was composed of
1,275 international non-nancial companies listed for the period
2002e2010. This sample was unbalanced, because full data were
not available for all companies and for all years. The data consisted
of 9,594 observations obtained from 20 countries (United States,
United Kingdom, Canada, Australia, Germany, Netherlands, New
Zealand, Austria, Denmark, Finland, Sweden, Switzerland, France,
Italy, Spain, Belgium, Japan, Singapore and Korea, plus the Chinese
special administrative region of Hong Kong). The sample was also
divided by industry-sector groups, in accordance with the Compustat economic activity code, under the following headings:
business materials, consumer discretionary, consumer staples,
health care, industrial, information technology, and nally, utilities.
3.2. Measures
3.2.1. Dependent variable: CSR practices
CSR practices were measured using a multidimensional
construct addressing all the actions carried out by the rms in
question, especially in social and environmental contexts (Carroll,
1979). In the present case, CSR information was compiled from
the EIRIS database, widely used in academic research (see Louche
et al., 2012; Dam and Scholtens, 2012; Fabrizi et al., 2014,
Martnez-Ferrero et al., 2014; Cuadrado-Ballesteros et al., 2015,
among others). Through the information that companies disclose
online and through questionnaires and surveys sent to companies,
EIRIS address different areas including environmental issues, human rights, and relationships with stakeholders (see Fig. 1),2,
assigning text-grades to specic attributes in each of these three
areas. This procedure might involve a subjective assessment of
relevant corporate practices, but the topics addressed and the
questions posed are designed in such a way as to enable a
reasonable assessment of the activities evaluated. Moreover, EIRIS
combines a broad range of environmental and socio-labour data
points to assess how companies respond to the various sustainability challenges they face and to identify corporate leadership in
tackling environmental and social challenges through policies,

2
The rst of the three CSR areas concerns items such as the company's environmental management system and policy, its impact on the environment, and
whether it has published reports on this question. In addition, general note is taken
of the scope of the company's strategies, policies, systems and reporting in the eld
of human rights. Other factors taken into account include the company's management systems, the quantitative information provided, the general level of
commitment to stakeholders, its policies and practices in support of equal opportunities and diversity, the health systems and safety-at-work procedures that are
implemented, the support given to employee training and development, relationships with customers and suppliers, and the level of commitment to the community and to social projects. Fig. 1 shows the composition of the CSR index in detail.

systems, reporting
and documented improvements in
performance.
To obtain the socially responsible commitment of companies,
we make use of an aggregate CSR measure that takes into consideration a range of important issues across companies according to
the 20 issues shown in Fig. 1: environmental, human right and
stakeholders' aspects. Similar to Martnez-Ferrero et al. (2014), and
Fabrizi et al. (2014), we transform the EIRIS text-grade rating for
each measure into a number-grade rating. According to the scoring
criteria of EIRIS (inadequate, weak, moderate, good and exceptional), we assign ve values:
3,
1, 0, 1 and 3. Overall, companies
are considered socially responsible in a specic aspect when the
score is above the threshold of 0 and are otherwise not considered
socially responsible. Because of CSR is determined from the nonweighted sum of these 20 items, it is in the range
60 to 60. Similar
to previous studies (Martnez-Ferrero et al., 2014), we use this
scoring criterion for two reasons: rstly, we think that negative
values represent non-socially responsible behaviours in a better
way; secondly, we do not use the value 2 (and
2) with the aim of
strongly discriminating among socially and non-socially responsible behaviours.
3.2.2. Independent variable: Managerial discretion
Among others, a passive attitude to accounting alternatives can
be used as proxy of managerial discretion (Prior et al., 2008; Jaggi
et al., 2009; Gargouri et al., 2010; Kim et al., 2012) because of
managers may exercise their discretionality by preferring accounting procedures that are most appropriate for their own purposes, in what has been termed earnings management (EM).
Discretionary accounting can have a major impact, if managers
engage in EM practices and report accounting results that do not
correspond to those actually achieved (Kim et al., 2012). Thus,
managerial discretion can be a proxy for EM, which can be
considered a discretionary practice (Garca Osma et al., 2005). Most
studies have sought to estimate the incidence of such practices by

, 2008).
examining accruals (Prior et al., 2008; Surroca and Tribo
With respect to EM, the discretionary component of accruals
adjustment could be used as a measure of accounting manipulation. Since not all accruals are discretionary, the discretionary
component must be separated from the non-discretionary one in
order to determine the presence and extent of EM. Total accruals
adjustment (TAA) has two components, discretionary accruals
(DAA) and non-discretionary accruals (NDAA) (Garca Osma et al.,
2005). TAA are dened as follows (Jones, 1991):

TAAit DCAit
DCASHit 
DCLit
DRLTPit 
DAit

(A)
where DCAit is the change in the current assets; DCASHit reects
the change in the cash held and short-term nancial investments;
DCLit is the change in current liabilities; DRLTPit is the change in
reclassied long-term obligations; and DAit is the depreciation and
amortisation.
On the basis of Equation (A), accruals are calculated using an
explanatory model. The difference between actual and expected
accrual adjustments represents the discretionary or unexplained
component of accrual adjustments, and acts as a measurement of
EM in the reporting of results. The discretionary accruals adjustment (DAA) represents the abnormal accruals, and these are used
as a measure of EM. DAA is calculated by applying the most
commonly used model for this purpose, namely the modied
version of the Jones model proposed by Dechow et al. (1995).
Firstly, based on the standard Jones model (1991), the following
model is estimated for each industry-year-quarter using OLS
estimation:

J. Martnez-Ferrero et al. / Journal of Cleaner Production 135 (2016) 760e770

765

Environmental index
Environmental policies and commitment
Environmental management systems
Environmental reporting
Level of environmental impact improvement
Human rights index
Extent of policies addressing human rights issues
Extent of systems addressing human rights issues
Extent of reporting addressing human rights issues
Stakeholders index
Policies toward stakeholders overall
Management systems for stakeholders overall
Quantitative reporting for stakeholders overall
Level of engagement with stakeholders overall
Policies on equal opportunities and diversity issues
Systems and practices to support equal opportunities and diversity issues
Health and safety systems
Systems and practices to advance job creation and security
Systems to manage employee relations
Systems to support employee training and development
Policies on maintaining good relations with customers - suppliers
Systems to maintain good relations with customers - suppliers
Level of commitment with community or charitable work
Fig. 1. CSR practices.

TAAit
1
a1;t
Ai;t
1
Ai;t
1

!
a2;t

DSalesit
Ai;t
1

!
a3;t

PPEit
Ai;t
1

!
t
(B)

where A represents the total assets; DSales is the change in sales;

and PPE represents property, plant and equipment.
Dechow et al. (1995), in their modied version of the Jones
model, assumed that not all sales are necessarily non-discretionary,
and included in their model the variations in sales minus accounts
receivable and minus the property, plant and equipment item,
which is used to measure the depreciation costs contained in the
discretionary adjustments. Thus, using the estimated coefcients
from model B, we compute the DAA as a proxy of managerial
discretion (termed MangDiscre):

"
b 1;t
MangDiscre TAAit
a
b 3;t
a

PPEit
Ai;t
1

1
Ai;t
1
!#

!
b 2;t
a

DSales
A*Rit

Ai;t
1

(C)
where A*R represents accounts receivable.
3.2.3. Moderator variable
There is no universal denition of family rm, and numerous
approaches to this concept have been proposed. Nevertheless, most
authors focus on aspects such as management, ownership or succession to dene a company as a family business (see Chen et al.,
2008; Chrisman et al., 2004). Some take the presence of family
members on the board as the most signicant yardstick (Anderson

and Reeb, 2003); others, such as Block and Wagner (2013), dene
family rms as companies in which at least two members of the
founding family are active in the rm as owners; and Chau and Gray
(2010) use the percentage of common shares held by the founding
family or their relatives as the measure of family ownership.
Family ownership is most commonly represented/measured by
the use of percentages of voting rights: the thresholds suggested
range from 5% (Berrone et al., 2012; Chen et al., 2008), to 10% (Chen
and Jaggi, 2001; Mok et al., 1992), and even 25% (Chau and Leung,
2006). Among the ample range of possibilities, in our study the
explanatory variable of ownership concentration is taken as
Family, a dummy variable (Kashmiri and Mahajan, 2010; Landry
et al., 2013) that takes the value 1 if the largest shareholder is a
family member with more than 10% of the votes, and 0 otherwise
(Mok et al., 1992; Dahya et al., 2008; Cuadrado-Ballesteros et al.,
2015).

3.2.4. Control variables

To avoid effects of bias on the results obtained, several control
variables were included taking into account their effect on CSR.
Company size (Size) is measured by the logarithm of the total
assets and is commonly used as a determinant variable of eco
,
nomic, social and environmental practices (Surroca and Tribo
2008). Another variable that has been widely used in previous
studies is the level of rm leverage (Debt), which represents the
company's debt or risk of non-compliance (Prior et al., 2008).
Risk represents the level of systematic risk and is measured by
the beta of the market model. In a CSR analysis, and according to
Margolis and Walsh (2003), it is necessary to consider the effect of
the industry sector (Industry) in which the company operates,
because every area of economic activity presents different characteristics. In order to represent the sector in which the company

766

J. Martnez-Ferrero et al. / Journal of Cleaner Production 135 (2016) 760e770

operates, a dummy variable was created, in accordance with the

Compustat economic sector code. Finally, R&D_Intensity is
measured as the ratio of R&D expenditure to total revenue. Some
studies, such as McWilliams and Siegel (2001), hold that CSR is also
dependent on R&D costs.
With respect to the control variables and their effect on CSR
practices, several studies have empirically conrmed the positive
relationship between size and social practices, concluding that
larger companies are able to pay more attention to their stakeholders' interests (Prior et al., 2008). In this respect, too, Kim et al.
(2012), among others, have shown that more socially responsible
companies have more opportunities for growth, achieve better
economic performance and present lower levels of debt than
companies that are less socially engaged. Furthermore, Prior et al.
(2008) report that nancial resources are strongly linked to CSR
practices. Similarly, Kim et al. (2012) concluded that the greater
the resources available to a company, the lower its level of social
practices. Spagnolo (2005) found that companies may use CSR as a
means of decreasing volatility, through agreements with their
stakeholders; furthermore, as shown by McWilliams and Siegel
(2001), CSR is also dependent on R&D costs since this type of
investment could encourage CSR-related processes and
innovation.
3.3. Model and analysis technique
To test our research hypotheses, simultaneous equations were
calculated for the panel data, applying the estimator proposed by
Arellano and Bond (1991). The use of panel data enables company
performance to be assessed over time, by analysing observations
from several consecutive years for the same company. The models
in our sample were analysed using the generalised method of
moments (GMM). Unlike within-groups or generalised least
squares estimators, this method accounts for endogeneity and
controls unobservable heterogeneity.
This paper proposes, rst, that CSR may be used as an
entrenchment strategy in order to avoid the costs of managerial
discretion and, second, that family ownership of a rm could
moderate this use of CSR. To test these propositions, the following
relationship is proposed, which is split into two dependency
models in accordance with our hypotheses:

managerial discretion as explanatory variables. The results obtained for Model 2 allow us to examine whether family ownership
moderates the use of CSR as an entrenchment strategy against the
costs of managerial discretion:

CSR a1 MangDiscreit a2 Familyit a3 MangDiscre  Familyit

a4 Sizeit a5 Debtit a6 Riskit a7 Industryit
a8 R&DIntensityit hi mit
(2)
where i ranges from 1 to 1,275; t takes the values of the years from
2002 to 2010; and a represent the estimating parameters; i
represents the unobservable heterogeneity; and mi represents the
error term.
4. Empirical results and discussion
Table 1 describes the distribution of the sample and of the
family rms, by country and by sector. Approximately 13% (1,261
observations out of 9,594) of the entities were family rms, but
these were not homogeneously distributed among the 20 countries. The countries with the highest percentages of family rms
were South Korea, France, Italy and Spain; the last three of these are
western European countries where there is a strong concentration
of ownership and voting rights in the hands of families. The
countries with the lowest presence of family rms were Singapore
and Japan. The representation of family rms was similar across all
the sectors analysed. Nonetheless, the sector with the highest
presence of family rms was consumer staples (19.4% relative frequency) and that with the smallest presence was utilities (5%
relative frequency).
Table 2 summarises the Pearson correlation coefcients of the
main variables used in this study. The correlation matrix for these
coefcients indicates the relative positions of the events with
respect to the two variables, via a numerical expression indicating
the degree of relationship between the two variables and the extent
to which they relate. The coefcients between the independent
variables are not very strong, and so there are no multicollinearity
problems that might confound the estimation. Table 2 shows that
there is a positive association between managerial discretion and
CSR, an association that is analysed below. This result can be

CSR fMangDiscre; Family; MangDiscre  Family; Control Variables

For the rst of our study aims (to examine whether CSR is used
as a strategic shield against the costs of managerial discretion),
Model 1 regresses CSR practices (CSR), while controlling for the
managerial discretion variable and the control variables. The results
obtained for Model 1 enable us to examine whether managers
satisfy stakeholders' demands via CSR actions in order to avoid the
costs of managerial discretion (MangDiscre).

CSRit 1 MangDiscreit 2 Sizeit 3 Debtit 4 Riskit

5 Industryit 6R&DIntensityit hi mit

(1)

With respect to the second study aim (to examine whether

family ownership of the rm moderates the above relationship),
Model 2 regresses CSR, including managerial discretion and the
family ownership indicator variables and their interaction with

interpreted as evidence that when managers adopt discretionary

practices, they may also employ an entrenchment strategy based on
CSR. Moreover, in relation to the moderating factor of the family
rms, we obtain evidence from two standpoints. On the one hand,
family ownership has an impact on economic issues such as CSR,
demonstrating a greater ethical and social commitment by family
rms than non-family rms. But on the other hand, and as discussed below, there is a negative association between family
ownership and managerial discretion, which appears to arise from
family members' greater control and monitoring of managerial
decisions in these rms. These results can be interpreted as
meaning that when a rm is owned by (majority) family shareholders, CSR practices are less likely to be promoted as a mechanism to satisfy stakeholders and in this way to avoid the costs of
managerial discretion.
Table 3 shows the results of the analyses of Models 1 and 2.

J. Martnez-Ferrero et al. / Journal of Cleaner Production 135 (2016) 760e770

Table 1
Descriptive statistics: family rms by country and by sector.

Table 3
CSR entrenchment based on managerial discretion in family rms.

Non family rm

Family rm

Absolute

Relative (%)

Absolute

Relative (%)

292
59
79
339
89
70
208
281
335
127
1822
108
85
258
0
135
133
120
1417
2376

94.5%
95.2%
83.2%
79.2%
90.8%
88.6%
57.8%
77.0%
83.8%
706%
96.1%
91.5%
91.4%
97.7%
0%
69.2%
89.9%
70.2%
92.0%
85.6%

17
3
16
89
9
9
152
84
65
53
73
10
8
6
16
60
15
51
124
401

5.5%
4.8%
16.8%
20.8%
9.2%
11.4%
42.2%
23.0%
16.2%
29.4%
3.9%
8.5%
8.6%
2.3%
100%
30.8%
10.1%
29.8%
8.0%
14.4%

Business materials
Consumer Discretionary
Consumer Staples
Health Care
Industrial Field
Information
Utilities
Others

1,583
1,759
2,154
1,096
909
525
114
193

88.8%
80.6%
88.7%
89.4%
87.6%
87.2%
95.0%
89.8%

200
423
274
130
129
77
6
22

11.2%
19.4%
11.3%
10.6%
12.4%
12.8%
5%
10.2%

Total

8333

86.86%

1261

13.14%

Country
Australia
Austria
Belgium
Canada
Denmark
Finland
France
Germany
Hong Kong
Italy
Japan
Netherlands
New Zealand
Singapore
South Korea
Spain
Sweden
Switzerland
UK
USA

767

Dependent variable: CSR

Industry

Empirically, the results of Model 1 indicate that the effect of

managerial discretion on the adoption of CSR-oriented policies is
positive and signicant at 99% (coeff. 0.0001; p-value<0.001). In
other words, the higher the level of managerial discretion in a
company, the stronger its commitment to CSR. These initial results
support the agency perspective of CSR actions, as proposed by
Friedman (1970), according to which CSR is used as a hedging
strategy against possible negative reactions to practices of managerial discretion. From a theoretical perspective, applying agency
theory, managerial discretion and entrenchment are manifestations of the agency conict in the sense that when the objectives of
the principal and those of the agent are not the same, managers can
benet from their managerial position to promote actions that
serve their own interests. Furthermore, they will adopt strategies
that are necessary to mask this behaviour, such as a CSR commitment. Our evidence conrms the prior ndings of Hemingway and

(2008),
Maclagan (2004), Prior et al. (2008) and Surroca and Tribo
among others, who suggested that CSR actions e as agency costs e
are employed as a mechanism for managers' private benet (Fabrizi

Variable

Model 1

Model 2

Coef.

Std. err.

Coef.

Std. err.

MangDiscre
Family
MangDiscre*Family
Size
Debt
Risk
Industry
R&Dintensity
Year

0.0001***
e
e

0.4799***

0.00632**

0.01381***
7.7942***
0.0264***
Controlled

2.13E-06
e
e
0.0392
0.0031
0.0008
2.4545
0.0007

0.001***
3.299***

0.003*

0.469***

0.006***

0.014*
7.262***
0.026***
Controlled

2.03E-06
0.703
0.002
0.039
0.003
0.001
2.430
0.001

Z
m1
m2
Hansen

6756.11
0.08

0.23
118.42

5023.44
0.08

0.23
138.44

Notes:
i) Heteroskedasticity consistent asymptotic standard error in parentheses.
ii) p < 0.1; *p < 0.05; **p < 0.01; ***p < 0.001.
iii) z is a Wald test of the joint signicance of the reported coefcients, asymptotically distributed as c2 under the null hypothesis of no relationship, degrees of
freedom and signicance in parentheses.
iv) mi (m1 and m2) is a serial correlation test of order i using residuals in rst differences, asymptotically distributed as N(0,1) under the null hypothesis of no serial
correlation.
v) Hansen is a test of over-identifying restrictions, asymptotically distributed as c 2
under the null hypothesis of non-correlation between the instruments and the error
term; degrees of freedom and signicance in parentheses.

et al., 2014; Kim et al., 2012) as they seek to enhance their job security, to strengthen their position and to avoid negative reactions
by stakeholders to the practice of managerial discretion.
Table 3 also reects the moderating effect of family ownership
on the use of CSR as an entrenchment strategy (see Model 2). The
results obtained reveal the positive and signicant impact of
managerial discretion on CSR variable (coeff. 0.01 and pvalue<0.001), supporting the hypothesis that managers promote
CSR practices as a powerful tool of entrenchment in order to serve
their personal aims. Here, we again explore CSR from an agency
cost perspective. Furthermore, the results for the Family variable
highlight the positive and signicant impact of family ownership
on the promotion of socially responsible practices (coeff. 3.299 and
p-value<0.001). We conrm the stronger ethical orientation of
family owners, and that they favour non-economic and nonnancial priorities (such as CSR), which means that they aim to
satisfy stakeholders' demands for CSR. In this respect, too, we
support the conclusions of Berrone et al. (2012), Cennamo et al.
(2012) and Van Gils et al. (2014), among others, who observe a
proactive family orientation toward social and environmental
practices. Thus CSR is a mechanism for preserving their image,
reputation, socio-emotional wealth and support from the community. Finally, the interaction between family ownership and the
indicator of managerial discretion (MangDiscre*Family) is

Table 2
Bivariate correlations.

1.
2.
3.
4.
5.
6.
7.
8.

CSR
MangDiscre
Family
Size
Debt
Risk
Industry
R&Dintensity

1
0.307
0.036
0.333
0.016

0.033

0.143

0.014

1

0.015

0.380

0.031
0.011
0.049
0.002

1
0.101

0.018

0.010

0.025
0.002

1
0.048

0.007

0.139

0.047

1
0.005
0.004

0.001

1

0.002
0.001

1
0.036

768

J. Martnez-Ferrero et al. / Journal of Cleaner Production 135 (2016) 760e770

revealed to have a negative and signicant effect on CSR (coeff.

-0.003, and p-value<0.05). This result supports the argument that
family ownership constitutes an effective internal mechanism that
inhibits managerial discretion and thus renders the use of CSR as a
self-defence strategy irrelevant. In other words, family ownership
limits managerial discretion and CSR as entrenchment, or family
rms are characterised by lower agency conicts. In the same way
as Shleifer and Vishny (1989), La Porta et al. (1998), Chen et al.

(2008), we nd that in family rms
(2008) and Surroca and Tribo
the classical agency conict between owners and managers is less
likely to arise, because family shareholders exert greater monitoring and a higher degree of supervisory control over managerial
decisions. In other words, the smaller agency conict in family
rms decreases the likelihood that managers will promote CSR by
way of entrenchment. In conclusion, and following Landry et al.
(2013) and Khan et al. (2013), we conrm that the higher the degree of family ownership, the lower the probability of a Type 1
agency conict occurring, and thus, the lower the risk of managerial
discretion risk behaviour (a practice that might compromise the
rm's survival and the succession process) as a manifestation of the
agency conict (Singal, 2014).
The effect of the control variables remains unchanged in the two
proposed models; corporate size and debt and the degree of market
risk are all associated with a lower level of CSR practices. Larger,
more highly indebted companies, and those that operate in capital
markets with higher risk, undertake fewer CSR practices. Following

(2008), we nd that CSR
Prior et al. (2008) and Surroca and Tribo
practices depend on company size and debt, and on the level of
systematic risk. Moreover, industry sector and R&D intensity are
positively associated with social, economic and environmental actions. In agreement with Margolis and Walsh (2003), we nd that
the effect of the industry in which the company operates and the
intensity of R&D (McWilliams and Siegel, 2001) both modify the
level of CSR-related activity.
Several additional analyses are performed to determine the

Table 4
Robust analysis. CSR entrenchment based on managerial discretion in family rms.
Dependent variable: CSR
Variable

Robust 1 (without
control variables)

Robust 2 (family_2)

Coef.

Std. err.

Coef.

Std. err.

MangDiscre
Family_2
MangDiscre*Family_2
Size
Debt
Risk
Industry
R&Dintensity
Year

0.001**
3.071***

0.003
e
e
e
e
e
Controlled

0.000
0.752
0.002
e
e
e
e
e

0.002***
1.001***

0.003***

0.496***

0.006***

0.014*
8.112***
0.026***
Controlled

0.001
0.028
0.006
0.038
0.003
0.001
2.361
0.007

Z
m1
m2
Hansen

897.31

0.18

1.15
56.94

5244.54
0.05

0.23
135.20

Notes:
i) Heteroskedasticity consistent asymptotic standard error in parentheses.
ii) p < 0.1; *p < 0.05; **p < 0.01; ***p < 0.001.
iii) z is a Wald test of the joint signicance of the reported coefcients, asymptotically distributed as c2 under the null hypothesis of no relationship, degrees of
freedom and signicance in parentheses.
iv) mi (m1 and m2) is a serial correlation test of order i using residuals in rst differences, asymptotically distributed as N(0,1) under the null hypothesis of no serial
correlation.
v) Hansen is a test of over-identifying restrictions, asymptotically distributed as c 2
under the null hypothesis of non-correlation between the instruments and the error
term; degrees of freedom and signicance in parentheses.

moderating effect of family rms on managerial discretion and

entrenchment, in order to conrm the robustness of our results.
First, we investigate the robustness of our results for Model 2
without taking into account the control variables (see Table 4),
and we nd that the inuence of the independent variables and
their interaction continues to be signicant. This nding supports
our prior evidence of the moderating impact of family ownership
on the possible connection between managers and stakeholders via
CSR practices, which we suggest are undertaken in order to avoid
the costs arising from managerial discretion.
Table 4 also reects the robustness of our results when we apply
an alternative proxy for family rms, with an increased percentage
of family ownership required for a company to be considered a
family rm. For this purpose, we create a new variable (Family_2),
which takes the value 1 if the largest shareholder is a family
member with more than 25% of the votes, and 0 otherwise. This
new proxy of family ownership conrms the relationships
described previously.
In short, the two robust analyses corroborate our original
premise that entrenchment based on CSR is used as a defence
strategy and that family ownership of the rm moderates this
relationship.
5. Conclusions
The central idea of this paper arises from the classical agency
conict between owners and managers. This conict led us to
examine whether CSR practices constitute an entrenchment strategy aimed at avoiding the costs of managerial discretion, and
whether family ownership may have a moderating role; that is, we
wished to examine CSR from an agency cost perspective. Our
empirical ndings corroborate the proposition that managers who
manipulate earnings adopt CSR practices as a means of dissuading
shareholders and stakeholders from reacting against this managerial discretion. Moreover, family ownership constitutes an
effective control mechanism for decreasing the risk of managerial
discretion and hence the use of CSR as an entrenchment strategy. In
other words, within an agency cost perspective, managers satisfy
stakeholders' demands via CSR actions in order to serve their own
interests and to avoid the costs of managerial discretion; the implications of agency conict are limited by family ownership,
through which majority shareholders can retain control and supervision of managerial actions.
Various important contributions are made in this paper: rst,
the results obtained show that CSR is indeed employed as a strategy
of entrenchment and self-defence, which is different from most of
the prior literature in this eld because of our alternative approach
to CSR practices (Freeman, 1984). We expand the focus of the prior
literature regarding the possible use of CSR as an entrenchment
strategy in the following ways. While Cespa and Cestone (2007)
referred, theoretically, to the possible discretionary use of CSR
from an agency perspective, we construct several dependency
models for panel data, applying the GMM estimator of Arellano and
Bond (1991). We also extend the evidence of Pagano and Volpin
(2005), who had focused exclusively on employees as stakeholders. In contrast, the present study examines a broader
construct of CSR. In addition, while Scholtens and Kang (2013) and
Cheng et al. (2013) focused on a single context (Asian rms, and US
rms, respectively), we examine an international sample of 20
countries, which provides results that can be generalised more
easily.
Nevertheless, the main contribution of this paper is its analysis
of the moderating role of family rms, which contributes to the
growing literature about them. With respect to the inclusion of
family ownership as a moderating variable, we extend the earlier

J. Martnez-Ferrero et al. / Journal of Cleaner Production 135 (2016) 760e770

(2008) and
evidence of Prior et al. (2008), Surroca and Tribo
Martnez-Ferrero et al. (2014), who only examined the possible use
of CSR as a means of entrenchment, without examining the impact
of ownership structure. With respect to previous studies of family
rms, we expand the research of Ali et al. (2007), who examined
the discretionary accruals of US rms but did not analyse their
relationship to CSR; we also extend the study of Chen et al. (2008),
who examined the impact of different levels of agency conict on
family rms, but focused on Type 2 agency conict. We contribute
to knowledge in this respect by examining Type 1 agency conict
between family owners and managers.
5.1. Theoretical and practical implications
The results of this study have a number of implications for
theory and practice. First, the rationale for this paper is found in
agency theory. From the perspective of this theory, managerial
discretion and entrenchment are viewed as opportunities for
managers to serve their own interests rather than the owners'
objectives. Our results corroborate the potential existence of Type 1
agency conict between owners and managers, because managers
may implement autonomous actions aimed at obtaining private
benets (Berle and Means, 1932; Shimizu, 2012). In rms characterised by a clear separation between ownership and control,
managerial discretion and CSR as entrenchment are the result of a
conict between the interests of managers (who have discretion)
and those of owners. Second, our ndings support the theory of
entrenchment (Shleifer and Vishny, 1989), according to which
managers may overinvest in CSR practices as a mechanism for
expropriating wealth from the shareholders (Cespa and Cestone,

,
2007; Khan et al., 2013; Prior et al., 2008; Surroca and Tribo
2008). However, the implementation of CSR as a self-defence
mechanism is moderated by family ownership. In this respect,
very little previous research has been reported, and we propose
theoretical arguments based on the description of family owners as
an internal control mechanism. From the agency theory setting,
although it is known that managers' objectives are different from
those of shareholders, large family shareholders that control a rm
may develop various strategies to prevent managers from using
their discretion in decision-making processes, thus constraining
managerial discretion and the use of CSR as an entrenchment
mechanism.
The main practical implication of this study concerns the conclusions that should be drawn by shareholders and other stakeholders in companies that are directly affected by the use of CSR as
an entrenchment strategy. Our evidence provides useful information to investors who wish to evaluate the use of CSR as an
entrenchment strategy, especially for family-owned rms. Our
ndings will be of interest to investors and public authorities who
are seeking to assess the negative cost of managerial discretion;
and to policy makers and regulators, who could make use of them
to improve market transparency by introducing new requirements
to constrain managerial discretion in decision-making processes,
and thus restrict the possible use of CSR as a hedging strategy.
5.2. Limitations and areas for future research
The results of this study should be interpreted carefully, since
this research is subject to certain limitations. First, CSR is measured
as the unweighted sum of three different measurements (in the
areas of the environment, human rights and stakeholders), based
on numerical scales. Although we believe this measure to be reliable and accurate (following previous studies), we are cautious
about the possible bias included in it because it may not capture the
true underlying practices. Moreover, we have only examined CSR as

769

a global construct. Thus, we cannot conclude whether the use of

CSR as an entrenchment tool varies according to the area of CSR
being analysed. Because the CSR indicator is an aggregate measure,
it does not tell us whether bad performance in one area is
compensated by good performance in another. Future research
could try to overcome this limitation with additional measures.
Furthermore, family ownership and family management should be
considered in greater detail to provide a better characterisation of
the evidence discussed. A limitation related to the family rm
measure is that this paper is focuses only on the moderating role of
family ownership, without examining the appointment of family
members to the board as managers (Jaggi et al., 2009). Our results
could be checked with an additional analysis of the family control
of the rm; a precise measure of family management, such as the
percentage of family members in senior management positions,
would be a valuable addition to the analytical tools employed.
Moreover, the analysis conducted in this paper should be extended
to the context of Type 2 agency conicts (between majority and
minority shareholders) and expanded to include informational
differences (Chen et al., 2008). Finally, this analysis, conducted in an
international context, is focused on countries with different systems of corporate governance and different legislative and legal
frameworks. Future research could examine the relationships of
our analysis in the context of greater stakeholder or shareholder
protection and different legal systems, cultural values, etc (La Porta
et al., 2002).
Acknowledgements
The authors wish to acknowledge the nancial support from the
Ministry of Science and Innovation for the research project
ECO2013-43838-P and from the Multidisciplinary Institute of Enterprise (MIE) of the University of Salamanca.
References
Ali, A., Chen, T.Y., & Radhakrishnan, S., 2007. Corporate disclosures by family rms. J.
Account. Econ. 44 (1), 238e286.
Anderson, R.C., Reeb, D.V., 2003. Founding-family ownership, corporate diversication, and rm leverage. J. Law Econ. 46, 653e684.
Arellano, M., Bond, S., 1991. Some tests of specication for panel data: monte Carlo
evidence and an application to employment equations. Rev. Econ. Stud. 58 (2),
277e297.
Bammens, Y., Notelaers, G., Van Gils, A., 2014. Implications of family business
employment for employees' innovative work involvement. Fam. Bus. Rev. 28
(2), 123e144.
Barnard, C., 1962. Functions of the Executive (Cambridge).
Baumol, W.J., 1959. Business Behaviour. Value and Growth (New York).
Berle, A.A., Means, G.G.C., 1932. The Modern Corporation and Private Property.
Transaction Publishers.

mez-Meja, L.R., 2012. Socioemotional wealth in family
Berrone, P., Cruz, C., Go
rms: theoretical dimensions, assessment approaches, and agenda for future
research. Fam. Bus. Rev. 25, 258e279.
Block, J.H., Wagner, M., 2013. The effect of family ownership on different dimensions of corporate social responsibility: evidence from large US rms. Bus.
Strategy Environ. 23 (7), 475e492.
Carroll, A.B., 1979. A three-dimensional conceptual model of corporate performance. Acad. Manag. Rev. 4 (4), 497e505.

mez-Meja, L.R., 2012. Socioemotional wealth
Cennamo, C., Berrone, P., Cruz, C., Go
and proactive stakeholder engagement: why family-controlled rms care more
about their stakeholders. Entrepreneursh. Theory Pract. 36 (6), 1153e1173.
Cespa, G., Cestone, G., 2007. Corporate social responsibility and managerial
entrenchment. J. Econ. Manag. Strategy 16 (3), 741e771.
Chau, G., Gray, S.J., 2010. Family ownership, board independence and voluntary
disclosure: evidence from Hong Kong. J. Int. Account. Auditing Tax. 19 (1),
93e109.
Chau, G., Leung, P., 2006. The impact of board composition and family ownership on
audit committee formation: evidence from Hong Kong. J. Int. Account. Auditing
Tax. 15 (1), 1e15.
Chen, C.J., Jaggi, B., 2001. Association between independent non-executive directors,
family control and nancial disclosures in Hong Kong. J. Account. Public Policy
19 (4), 285e310.
Chen, S., Chen, X., Cheng, Q., 2008. Do family rms provide more or less voluntary
disclosure? J. Account. Res. 46 (3), 499e536.

770

J. Martnez-Ferrero et al. / Journal of Cleaner Production 135 (2016) 760e770

Cheng, I.H., Hong, H., Shue, K., 2013. Do Managers Do Good with Other People's
Money? National Bureau of Economic Research (No. w19432).
Cho, S.Y., Lee, C., Pfeiffer, R.J., 2013. Corporate social responsibility performance and
information asymmetry. J. Account. Public Policy 32, 71e83.
Chrisman, J.J., Chua, J.H., Litz, R.A., 2004. Comparing the agency costs of family and
non-family rms: conceptual issues and exploratory evidence. Entrepreneursh.
Theory Pract. 28 (4), 335e354.

s-Galdeano, L., Berrone, P., 2014. Are family rms
Cruz, C., Larraza-Kintana, M., Garce
really more socially responsible? Entrepreneursh. Theory Pract. 38 (6),
1295e1316.
Cuadrado-Ballesteros, B., Rodrguez-Ariza, L., & Garca-S

anchez, I.M., 2015. The role
of independent directors at family rms in relation to corporate social responsibility disclosures. Int. Bus. Rev. 24 (5), 890e901.
Dahya, J., Dimitrov, O., & McConnell, J.J., 2008. Dominant shareholders, corporate
boards, and corporate value: a cross-country analysis. J. Financ. Econ. 87 (1),
73e100.
Dam, L., Scholtens, B., 2012. Ownership concentration and CSR policy of european
multinational enterprises. J. Bus. Ethics 1e10.
Dechow, P.M., Sloan, R.G., Sweeney, A.P., 1995. Detecting earnings management.
Account. Rev. 70 (2), 193e225.

niz, M.C., Cabrera, M.K., 2005. Corporate social responsibility and family business
De
in Spain. J. Bus. Ethics 56 (1), 27e41.
Dyer, W.G., Whetten, D.A., 2006. Family rms and social responsibility: preliminary
evidence from the S&P 500. Entrepreneursh. Theory Pract. 30 (6), 785e802.
Eisenhardt, K.M., 1989. Agency theory: an assessment and review. Acad. Manag.
Rev. 14 (1), 57e74.
Fabrizi, M., Mallin, C., Michelon, G., 2014. The role of CEO's personal incentives in
driving Corporate Social Responsibility. J. Bus. Ethics 124 (2), 311e326.
Florackis, C., Ozkan, A., 2009. The impact of managerial entrenchment on agency
costs: an empirical investigation using Uk panel data. Eur. Financ. Manag. 15 (3),
497e528.
Freeman, R.E., 1984. Strategic Management. A Stakeholders Approach. Pitman,
London.
Friedman, M., 1970. A theoretical framework for monetary analysis. J. Political Econ.
193e238.
Garca Osma, B., Gill de Albornoz Noguer, B., Gisbert Clemente, A., 2005. La inves
n sobre earnings management. Rev. Espan
~ ola Financ. Contab. 34 (127),
tigacio
1001e1034.
Gargouri, R.M., Shabou, R., Francoeur, C., 2010. The relationship between corporate
social performance and earnings management. Can. J. Adm. Sci./Rev. Can. Des
Sci. De l'Adm. 27 (4), 320e334.

mez-Meja, L.R., Hynes, K.T., Nun
~ ez-Nickel, M., Moyano-Fuentes, H., 2007. SociGo
oemotional wealth and business risk in family-controlled rms: evidence from
Spanish olive oil mills. Adm. Sci. Q. 52, 106e137.
Haalien, L., Huse, M., 2005. Board of Directors in Norwegian Family Businesses:
Results from the Value Creating Board Survey. Research Report No 7-2005.
Norwegian School of Management BI, Oslo.
Healy, P.M., Palepu, K.G., 2001. Information asymmetry, corporate disclosure, and
the capital markets: a review of the empirical disclosure literature. J. Account.
Econ. 31 (1), 405e440.
Hemingway, C.A., Maclagan, P.W., 2004. Managers' personal values as drivers of
corporate social responsibility. J. Bus. Ethics 50 (1), 33e44.
Jaggi, B., Leung, S., Gul, F., 2009. Family control, board independence and earnings
management: evidence based on Hong Kong rms. J. Account. Public Policy 28
(4), 281e300.
Jensen, M.C., Meckling, W.H., 1976. Theory of the rm: managerial behavior, agency
costs and ownership structure. J. Financial Econ. 3 (4), 305e360.
Jensen, M.C., 1986. Agency costs of free cash ow, corporate nance, and takeovers.
Am. Econ. Rev. 76 (2), 323e329.
Jiraporn, P., DaDalt, P.J., 2009. Does founding family control affect earnings management? Appl. Econ. Lett. 16 (2), 113e119.
Jones, J.J., 1991. Earnings management during import relief investigations.
J. Account. Res. 29 (2), 193e228.
Jones, T.M., 1995. Instrumental stakeholder theory: a synthesis of ethics and economics. Acad. Manag. Rev. 20 (2), 404e437.
Kashmiri, S., Mahajan, V., 2010. What's in a name? an analysis of the strategic
behavior of family rms. Int. J. Res. Mark. 27 (3), 271e280.
Khan, I., Chand, P.V., Patel, A., 2013. The impact of ownership structure on voluntary
corporate disclosure in annual reports: evidence from Fiji. Account. Tax. 5 (1),
47e58.
Kim, Y., Park, M.S., Wier, B., 2012. Is earnings quality associated with corporate
social responsibility? Account. Rev. 87 (3), 761e796.
La Porta, R.L., Lopez-de-Silane, F., Shleifer, A., Vishny, R., 1998. Agency problems and
Dividend Policies Around the World. National Bureau of Economic Research
(No. w6594).

La Porta, R., Lopez-de-Silanes, F., Shleifer, A., 2002. Government ownership of banks.
J. Finance 57 (1), 265e301.
Landry, S., Deslandes, M., Fortin, A., 2013. Tax aggressiveness, corporate social responsibility, and ownership structure. J. Account. Eth. Public Policy 14 (3),
611e645.
Langfred, C.W., 2004. Too much of a good thing? negative effects of high trust and
individual autonomy in self-managing teams. Acad. Manag. J. 47 (3), 385e399.
Louche, C., Arenas, D., van Cranenburgh, K.C., 2012. From preaching to investing:
attitudes of religious organisations towards responsible investment. J. Bus.
Ethics 110 (3), 301e320.
Margolis, J.D., Walsh, J.P., 2003. Misery loves companies: rethinking social initiatives
by business. Adm. Sci. Q. 48 (2), 268e305.
Martnez-Ferrero, J., Banerjee, S., Garca-S

anchez, I.M., 2014. Corporate social responsibility as a strategic shield against costs of earnings management practices. J. Bus. Ethics 1e20.
McWilliams, A., Siegel, D.S., 2001. Corporate social responsibility: a theory of the
rm perspective. Acad. Manag. Rev. 26 (1), 117e127.
Mok, H.M., Lam, K., Cheung, I., 1992. Family control and return covariation in Hong
Kong's common stocks. J. Bus. Finance Account. 19 (2), 277e293.
Pagano, M., Volpin, P.F., 2005. Managers, workers, and corporate control. J. Finance
60 (2), 841e868.

, J.A., 2008. Are socially responsible managers really
Prior, D., Surroca, J., Tribo
ethical? Exploring the relationship between earnings management and
corporate social responsibility. Corp. Gov. An Int. Rev. 16 (3), 160e177.
Rowley, T., Berman, S., 2000. A brand new brand of corporate social performance.
Bus. Soc. 39 (4), 397e418.
Roychowdhury, S., 2006. Earnings management through real activities manipulation. J. Account. Econ. 42 (3), 335e370.

n, M.F., 2004. Stakeholder rights and corporate governance: a
Schneper, W.D., Guille
cross-national study of hostile takeovers. Adm. Sci. Q. 49 (2), 263e295.
Scholtens, B., Kang, F.C., 2013. Corporate social responsibility and earnings management: evidence from Asian economies. Corp. Social Responsib. Environ.
Manag. 20 (2), 95e112.
Shimizu, K., 2012. Risks of corporate entrepreneurship: autonomy and agency issues. Organ. Sci. 23 (1), 194e206.
Shleifer, A., Vishny, R.W., 1989. Management entrenchment: the case of managerspecic investments. J. nancial Econ. 25 (1), 123e139.
Singal, M., 2014. Corporate social responsibility in the hospitality and tourism industry: do family control and nancial condition matter? Int. J. Hosp. Manag. 36
(1), 81e89.
Spagnolo, G., 2005. Managerial incentives and collusive behavior. Eur. Econ. 49 (6),
1501e1523.
Stulz, R., 1988. Managerial control of voting rights: nancing policies and the
market for corporate control. J. Financ. Econ. 20, 25e54.

, J.A., 2008. Managerial entrenchment and corporate social perSurroca, J., Tribo
formance. J. Bus. Financ. Account. 35 (5e6), 748e789.
Van Gils, A., Dibrell, C., Neubaum, D.O., Craig, J.B., 2014. Social issues in the family
enterprise. Fam. Bus. Rev. 27 (3), 193e205.
Wang, D., 2006. Founding family ownership and earnings quality. J. Account. Res. 44
(3), 619e656.
Williamson, O.E., 1963. Managerial discretion and business behavior. Am. Econ. Rev.
53 (5), 1032e1057.
Zahra, S.A., Priem, R.L., Rasheed, A.A., 2005. The antecedents and consequences of
top management fraud. J. Manag. 31 (6), 803e828.

Jennifer Martnez-Ferrero is Assistant Professor at the University of Salamanca, where

she earned her PhD in business. Her research interests cover both private and public
sectors, in particular the relationship between earnings management and corporate
social responsibility and their effects on nancial performance, cost of capital and
corporate reputation.

zaro Rodrguez-Ariza is Professor at University of Granada, where he earned his

La
PhD in business. In addition, he is director of one of the departments of family business
in Spain. His research has focused on both the public and private sectors, covering both
aspects of accounting and business organization.

nchez is Professor of Accounting at University of Salamanca,

Isabel Mara Garca-Sa
where she earned her PhD in business. Her main lines of research are focused on
corporate social responsibility, e-government and business ethics. In addition, she is
focused on public sector researches.