...

definitivo

by user

on
Category: Documents
2

views

Report

Comments

Transcript

definitivo
MANAGERIAL ENTRENCHMENT AND
CORPORATE SOCIAL PERFORMANCE †
JORDI SURROCA
Universidad Carlos III de Madrid
Department of Business Administration
Calle Madrid, 126
Getafe (Madrid), Spain, 28903
Phone: (34) 91-624-8640
Fax: (34) 91-624-9607
e-mail: [email protected]
JOSEP A. TRIBÓ *
Universidad Carlos III de Madrid
Department of Business Administration
Calle Madrid, 126
Getafe (Madrid), Spain, 28903
Phone: (34) 91-624-9321
Fax: (34) 91-624-9607
e-mail: [email protected]
* Corresponding author
†
The authors wish to thank Sustainable Investment Research International Company, and Analistas
Internacionales en Sostenibilidad (AISTM) for their helpful comments and access to the SiRi Pro TM
database. We would also like to thank Diego Prior, Stathopoulos Konstantinos and one of the referees as
well as participants at Journal of Business Finance and Accounting Capital Markets Conference (Chapel
Hill, 2007), the Academy of Management Annual Meeting (Philadelphia, 2007), the Symposium of
Corporate Governance & Shareholder Activism (Milan, 2007), the Journal of Banking and Finance 30 th
Anniversary Conference (Beijing, 2006), the Strategic Management Conference (Vienna, 2006), the Foro
de Finanzas (Castellon, 2006) and the seminar participants at Universitat Autònoma of Barcelona for their
useful comments on earlier drafts of this paper. We also acknowledge the financial support of the
Ministerio de Ciencia y Tecnologia (grant #SEC2003-03797), Ministerio de Educación y Ciencia (grant #
SEJ2004-07877-C02-02 and grant # SEJ2006-09401), and the Comunidad de Madrid (Grant #s0505/tic/000230). The usual disclaimers apply.
MANAGERIAL ENTRENCHMENT AND
CORPORATE SOCIAL PERFORMANCE
Abstract
We examine empirically the relationships amongst managerial entrenchment practices,
social performance, and financial performance. We hypothesize that entrenched
managers may collude with non-shareholder stakeholders in order to reinforce their
entrenchment strategy; this is particularly so in firms that have efficient internal control
mechanisms. Moreover, we prove that the combination of entrenchment strategies and
the implementation of socially responsible actions have particularly negative effects on
financial performance. We demonstrate this theoretical contention by using different
proxies for entrenchment like the existence of anti-takeover initiatives, classes of stock
with different voting rights, managerial ownership, managers’ tenure, and smoothing of
earnings. Our empirical investigation uses a database comprising 358 companies, from
22 different countries, for the period 2002-2005.
JEL Classification: G30, M14, M41.
Keywords:
Corporate
Governance,
Stakeholder
Performance, Earnings Management.
2
Activism,
Corporate
Social
1/ INTRODUCTION
Agency theory has contributed to reinforce the long held view that agency
problems are at the core of conflicts that appear within organizations. The literature
distinguishes between two main agency problems: the conflict between large and small
shareholders, which generates minority expropriation issues (Shleifer and Vishny,
1986); and the conflict between managers and shareholders that drives managers to
pursue their own private benefits at the expense of shareholder interests (Jensen and
Meckling, 1976). The latter problem is exacerbated when managers set on
entrenchment: they attempt to neutralize the disciplinary mechanism of the capital
market so as to maintain corporate control (Jensen and Ruback, 1983).
Managers, who place great value on control while holding only a small equity
stake, work to ensure their own job security even though they are no longer competent
or qualified to run the firm (Shleifer and Vishny, 1989). Walsh and Seward (1990)
discuss classes of managerial entrenchment practices. Dual-class recapitalization,
poison pills, supermajority amendments, anti-takeover amendments, and golden
parachutes are examples of such practices. Others (Stulz, 1988; Morck et al., 1988; De
Miguel et al., 2004) emphasize managerial ownership above certain levels as a takeover
deterrence mechanism that promotes managerial entrenchment. The use of these
mechanisms decreases managerial turnover, which is taken as another proxy of
managerial entrenchment (Denis et al., 1997 and Dahya et al., 1998). Finally, managers
can resort to income smoothing and other earnings management practices as a way of
improving their job security (Fudenberg and Tirole, 1995; Yeo et al., 2002).
Although anti-takeover defenses may decrease the efficiency of external control
mechanisms, they do not have the same effect on internal control mechanisms. Hence,
in a context of well developed internal control mechanisms, a manager set on
entrenchment has all the incentives to seek the connivance of non-shareholder
stakeholders against the actions of shareholders. Along this line, we extend the work of
Cespa and Cestone (2007) and hypothesize that entrenched managers collude with
employees, communities, customers, and suppliers to protect themselves from internal
disciplining mechanisms, causing a subsequent reduction in shareholders’ wealth. We
rely on three arguments to justify the manager’s commitment to follow stakeholderfriendly behavior, especially in those firms with efficient internal corporate governance
mechanisms. Firstly, stakeholders generally accumulate certain powers to promote or
penalize top executives (DeAngelo and DeAngelo, 1998; Hellwig, 2000; Rowley and
Berman, 2000); they can engage in costly boycotts and media campaigns (Baron, 2001;
3
Feddersen and Gilligan, 2001; John and Klein, 2003), and stakeholder representatives
may be present on corporate boards (Luoma and Goodstein, 1999; Schneper and
Guillén, 2004). Moreover, due to the fact that the manager retains the confidence of
stakeholders, it will be difficult for displeased shareholders to remove him because they
would have to face pressure from the non-shareholder stakeholders. The second
argument which justifies stakeholder concessions for entrenchment purposes is that, by
colluding with stakeholders, the manager reduces a firm’s attractiveness to potential
raiders. For example, generous long-term stakeholder concessions hinder the raider’s
ability to generate profit (Pagano and Volpin, 2005). Finally, we rely on Sundaramurthy
et al., (1997) to justify the positive, moderating role of internal corporate control
mechanisms that use stakeholder satisfaction as an entrenchment mechanism. These
authors show that the efficacy of some entrenchment mechanisms, like anti-takeover
measures, against external corporate governance mechanisms is moderated by internal
corporate governance mechanisms like board configuration. We extend this result and
argue that the moderating role of internal corporate control mechanisms has a similar
impact on stakeholder satisfaction, when used as an entrenchment mechanism.
In order to gain support from stakeholders, entrenched managers engage in a
broad array of practices to develop relationships with corporate stakeholders and
environmental activists; the so-called corporate social performance (CSP), as explained
in Waddock (2004). This study specifically tests the hypothesis that managerial
entrenchment practices are positively related to improvements in CSP which, in turn,
negatively affect firms’ financial performance. Moreover, we expect this relationship
will be stronger in those firms with well-developed internal corporate governance
mechanisms.
We make use of an international database provided by the Sustainable
Investment Research International (SiRi) Company, an international network of social
research organizations that scrutinizes firms with respect to their practices toward
employees, communities, suppliers, customers, environment, and shareholders. These
data include and expand upon those of Kinder, Lyndemberg, Domini, and Company
(KLD). Our final sample comprises 358 firms from 22 nations.
Our study advances the understanding of stakeholder phenomena by providing
evidence of another explanation for CSP: the entrenchment hypothesis (Cespa and
Cestone, 2007). Also, this study extends agency theory from a stakeholder perspective,
by examining the role employees, communities, customers, and suppliers, may play in
exacerbating or ameliorating conflicts of interests between managers and shareholders.
4
We expand the game-theoretical model of Pagano and Volpin (2005) to include not only
workers but other stakeholders (Cespa and Cestone, 2007), and subsequently, we test
empirically its main propositions. Finally, we provide evidence of the entrenchment
motives that explain earnings manipulations like income smoothing (Yeo et al, 2002;
Fudenberg and Tirole, 1995) as well as their connection with CSP. Managers smooth
earnings, as part of an entrenchment strategy, in order to ensure the stability of cashflow streams so that they can satisfy the short-term interests of shareholders. However,
this practice may generate problems in the medium term. Then, in order to eliminate
possible medium-term problems that may put their jobs at risk, managers will try to
connive with non-shareholder stakeholders by satisfying the interests of this group
(improving a firm’s CSP). Hence, regardless of the entrenchment mechanism used, we
claim that entrenchment practices lead to improvements in a firm’s CSP, particularly in
the presence of strong internal corporate governance mechanisms.
The rest of the article is structured as follows. Section 2 summarizes the most
relevant literature and develops the hypotheses. Section 3 is methodological and
describes the sample, variables and empirical models to be tested. The empirical results
obtained are presented in Section 4, while some extensions are addressed in Section 5.
In the final section of the article, we lay out the main conclusions of this research and
discuss the significance of our results.
2/ THEORY AND HYPOTHESES
Traditionally, agency theory has dominated the analysis of corporate
governance. Its main concern is the separation of ownership and control which
generates minority expropriation problems (Shleifer and Vishny, 1986) as well as
problems between managers and firm owners (Jensen and Meckling, 1976). Focusing
on managerial actions that damage shareholder interests, one of the costliest
manifestations of these actions is managerial entrenchment; and this can take a variety
of forms. Among them are: the issue of common stock with limited voting rights that is
exchanged for a certain number of original common shares; the repurchase of large
blocks of shares from potential acquirers without the approval of shareholders; poison
pills; new security issues; specific acquisitions and divestitures; supermajority
amendments; golden parachutes; and earnings smoothing. Additionally, managers may
try to accumulate stakes large enough to eliminate the threat of takeover but without
incurring the internalization of entrenchment costs. Hence entrenchment is expected to
be found in a middle range of managerial ownership (Morck et al, 1988; and De Miguel
5
et al, 2004). This set of mechanisms defines an entrenchment strategy that could
decrease managerial turnover (Denis et al., 1997 and Dahya et al, 1998).
Remarkably, these strategies are used principally by managers to deal with the
pressure from external corporate control mechanisms, as Shleifer and Vishny (1997)
define. The market for corporate control, product market competition or managerial
labor markets are examples of external mechanisms of corporate governance aimed at
reducing the agency problems linked to managerial actions. However, there are other
control mechanisms that are effective in preventing the aforementioned entrenchment
practices, and are defined as internal (Shleifer and Vishny, 1997). Among these internal
mechanisms are: stock-options and other forms of performance-based payment
schemes; control structures such as the presence of institutional blockholders (Shleifer
and Vishny, 1986); the presence of outsiders on the board of directors (Fama and
Jensen, 1983); or the existence of committees for audit, remuneration and nomination
control (Vafeas, 1999; Anderson and Reeb, 2004).
Hence, those managers set on entrenchment while under the close scrutiny of
internal corporate control mechanisms, need to reinforce their entrenchment strategy by
adopting some additional initiatives. We extend the propositions from Cespa and
Cestone (2007) and highlight stakeholder satisfaction as one of these initiatives used
when internal corporate control mechanisms are well developed. Stakeholders wield
significant power within the firm. They can organize boycotts and lobbies to
demonstrate such power (Baron, 2001; Feddersen and Gilligan, 2001; John and Klein,
2003). Further, stakeholders may exercise their influence via the board of directors,
when the board includes representation from labor, creditors, and regulatory agencies
(Schneper and Guillén, 2004). Under such a control structure, managerial decisions are
monitored and influenced by the presence of stakeholder representatives (Luoma and
Goodstein, 1999). Finally, the actions of stakeholders may influence the threat of a
hostile takeover and with that the own CEO replacement.1 Then, when internal
corporate governance mechanisms are well developed, managers tend to reinforce their
entrenchment strategy by canvassing support from stakeholders so as to channel the
efforts of the latter to the entrenched manager’s own advantage. This strategy enjoys the
benefit of diminishing pressure from activist stakeholders while at the same time
1
Schneper and Guillén (2004) show that the frequency of hostile takeovers is inversely related to
stakeholders (non-shareholder) power, and this result may explain why countries labeled as stakeholderoriented like Germany or Japan are characterized by the low occurrence of hostile takeovers.
6
countering the pressure from other internal corporate control mechanisms. To
operationalize such behavior, managers may engage in practices to create and manage
relationships with corporate stakeholders, the so-called CSP.
Our entrenchment argument for improving CSP is explained by the
stakeholders’ power to influence the firm. As such, our story belongs to the descriptive
realm of stakeholder theory (Mitchell et al., 1997). According to this theory, the degree
to which managers assign priority to competing stakeholders claims – the stakeholder
salience – is positively related to the cumulative number of stakeholder’s attributes of
power, legitimacy, and urgency. Then, a manager who wants to implement an
entrenchment strategy will want to be protected against the actions of powerful
stakeholders. In that case, there are two possibilities: collaboration or confrontation.
Jones (1995) and Hill and Jones (1992) supported the confrontation strategy.
Jones (1995), in deriving implications of his instrumental theory, suggested that
decreases of CSP are connected to a managerial entrenchment strategy. In a similar
vein, Hill and Jones (1992) predicted that managers will undertake strategic actions to
reduce stakeholder power – strategies that negatively affect corporate efficiency.
Our claim, as in Cespa and Cestone (2007), is the opposite: we hypothesize that
stakeholders and incumbent managers will be natural allies, particularly when there are
efficient internal corporate governance mechanisms capable of preventing managerial
entrenchment impulses. In that case, collaboration with stakeholders cannot be blocked
by shareholders easily, based on a “suspicious” entrenchment strategy. This stimulates
managers’ incentives to improve CSP with entrenchment intentions. Hellwig (2000)
pointed out that managers set on entrenchment will find allies in stakeholder sectors
such as the political system, labor, the media, the judiciary, and universities.
In addition, the implementation of expensive policies aimed at improving a
firm’s CSP reduces its attractiveness to a raider. Generous long-term contracts with
workers and suppliers, as well as long-term commitments to support environmental or
philanthropic organizations are too heavy a burden to borne by a raider (Pagano and
Volpin, 2005). These concessions, however, are not fully internalized in share prices for
two reasons. First, as stakeholder theory researchers (e.g., Berman et al., 1999) have
already demonstrated, maintaining good relationships with key stakeholders creates an
organizational resource that leads to more effective use of a firm’s resources; this has a
positive impact on financial performance. Therefore, in a context of information
asymmetries, capital markets will be unable to determine if social concessions are a
means of improving financial performance by generating a valuable organizational
7
resource or if they are part of an entrenchment strategy (market inefficiency). And
second, as mentioned before, social concessions related to the implementation of an
entrenchment strategy are triggered mainly in the presence of strong internal corporate
governance mechanisms. The existence of such strong internal corporate governance
mechanisms, together with the market inefficiency assumption, hinders any steep
reductions in share prices.
Therefore, anticipating the impact of generous CSP initiatives in terms of
reductions in stakeholder activism as well as reductions in the pressure from existing
shareholders and potential raiders, we state the following hypothesis:
Hypothesis 1. Managerial entrenchment practices have a positive impact on a
firm's social performance. This effect is more pronounced in firms with efficient
internal corporate governance mechanisms.
It is important to state that internal corporate governance mechanisms, while
playing a positive moderating role by controlling the connection between entrenchment
and CSP, also constrain managerial discretion. This should hinder the implementation
of expensive social responsible activities. That is, we also expect a negative direct effect
of the internal governance mechanisms on a firm’s CSP.
Finally, as mentioned before, agency theory shows that there is another conflict
of interest within the firm that affects small shareholders, which may generate minority
expropriation by large shareholders. Some authors like Barnea and Rubin (2006) have
argued that certain CSP activities may be connected to expropriation. Thus, it may
result in the case where managers collude with large shareholders in order to
expropriate minority shareholders by promoting intensive CSP policies. This collusion
can be used as an entrenchment mechanism as it reinforces the managerial position with
respect to the largest shareholders. Hence, in order to avoid this spurious connection
between entrenchment and CSP we have to control for variables that capture minority
expropriation risks like ownership concentration.
Types of stakeholders: employees
Within the managerial entrenchment strategy, workers constitute one of the
stakeholder groups that receive preferential attention by the manager because they have
the capacity to influence a firm’s behavior and, at the same time, share common
interests with incumbent managers. As a consequence, it is likely that entrenched
managers, particularly those that face the pressure from efficient internal corporate
8
governance mechanisms, will commit themselves to giving employees more
concessions (e.g., generous salaries).
The capacity of employees to influence decision-making, organizational
arrangements and performance outcomes is well documented in previous literature (see
for example, Scheneper and Guillén, 2004). This power is derived from political action
or legal mechanisms that are at their disposal. By political action we mean that workers
may lobby against/in favor of an incumbent CEO by demonstrating, mobilizing
politicians, appealing to the media, and constituting organized pressure groups like trade
unions (Pagano and Volpin, 2005). In addition, the employees’ power to promote or
penalize top executives is amplified when they have institutionalized legal mechanisms
at their disposal. One such mechanism is the presence of stakeholder directors on
corporate boards, or board subcommittees such as the audit, compensation, executive,
and nominating committees (Luoma and Goodstein, 1999). Furthermore, workers can
directly affect the likelihood of CEO replacement through individual share ownership.
Importantly, employees are not only a powerful stakeholder group, but are
natural allies of managers set on entrenchment. There is a vast literature that
demonstrates that hostile takeovers have negative consequences for workers (Aguilera
and Jackson, 2003) and that they, consequently, tend to oppose such hostile takeovers
(Scheneper and Guillén, 2004). In countries with low employment protection, a hostile
takeover may result in job cuts and causes a worsening in the working conditions.
Successful raiders renegotiate the labor contracts that already exist, cutting wages to a
minimum and stepping up monitoring to maintain workers’ effort (Conyon et al., 2001).
In addition to the previous arguments, managers may be interested in colluding
with employees, not only to gain their support, but also to reduce a firm’s attractiveness
to any potential raiders interested in companies with efficient internal corporate
governance mechanisms. In such cases, employment policy is likely to be used to deter
hostile takeovers (Pagano and Volpin, 2005), particularly in a context of inefficient
financial markets that are more likely to be found in non-Anglo-Saxon countries. In
these countries, social concessions to workers are more commonplace and difficult to
reverse. These arguments suggest the following hypothesis:
Hypothesis 2. Managerial entrenchment practices have a positive impact on
employees’ satisfaction. This effect is more pronounced in those firms with
efficient internal corporate governance mechanisms.
9
Performance analysis
The instrumental approach is an important perspective of stakeholder theory
(Donaldson and Preston, 1995). It advocates the formulation and implementation of
activities that satisfy stakeholders because they control key resources and suggests that
stakeholder satisfaction, in turn, will ensure the long-term survival and success of the
firm (Freeman, 1984; Hillman and Keim, 2001). Accordingly, stakeholders that own
resources relevant to the firm’s success will be more willing to offer their resources to
the extent that their different claims and needs are fulfilled (Strong et al., 2001).
Therefore, under this approach we expect that stakeholder satisfaction leads to higher
commitment, greater effort, and, ultimately, to superior performance (Hosmer, 1994;
Stevens et al., 2005). Thus, stakeholder management has strategic value from a “means
to an end” perspective (Berman et al., 1999), which is opposed to the intrinsic value of
the normative approach.
However, consistent with our previous propositions, we argue that when
managers implement entrenchment practices, the negative effect of such practices on
shareholders’ value (Walsh and Seward, 1990; Sundaramurthy, 2000) is reinforced with
improvements in CSP.
Importantly, Sundaramurthy et al. (1997) and Sundaramurthy (2000) suggest
that the strength or weakness of internal monitoring mechanisms, such as the board
structure and the ownership concentration, moderate the relationship between certain
entrenchment practices like anti-takeover provisions and shareholders’ wealth.
Remarkably, we have suggested in previous hypotheses, that a manager trying to
insulate himself from internal monitoring mechanisms may also follow a generous
policy of social concessions. Therefore, these managerial concessions to stakeholders,
especially in the context of existing internal governance mechanisms that are efficient,
should play a moderating role in the connection between the implementation of
entrenchment practices and financial performance. McWilliams and Siegel (2001)
termed these types of concessions as discretionary CSP and pointed out that it is
negatively related to shareholders’ wealth. Therefore, we expect this moderating role to
be negative. Stated formally:
Hypothesis 3: Managerial entrenchment practices when combined with social
concessions, have a negative impact on financial performance. This is
particularly evident when social concessions are triggered in a context where
there are efficient internal governance mechanisms.
10
Finally, we expect that this result also holds when we focus on the specific
dimension of social performance: workers’ satisfaction. This is because social
concessions to workers are particularly costly and, at the same time, they strongly
reinforce the entrenchment position of the manager with respect to shareholders given
the saliency of these stakeholders to achieve the firm’s success. This stakeholder
characteristic is particularly attractive in a context where efficient governance
mechanisms aimed at preventing entrenchment are present. In such situations, social
concessions, mainly to salient stakeholders like workers, will be needed in order to
reinforce any entrenchment strategy. Hence, the negative impact of entrenchment on
performance will be more evident when combined with workers’ satisfaction:
Hypothesis 4: Managerial entrenchment practices, when combined with social
concessions to workers, have a negative impact on financial performance,
especially in a context where there are efficient internal governance
mechanisms.
3. METHODS
3.1. Sample and Variables
We derive our sample by crossing different databases. The SiRi PRO TM
database, compiled by the Sustainable Investment Research International Company
(SiRi) – the world’s largest company specializing in socially responsible investment
analysis. SiRi performs this analysis based on: reporting procedures, policies and
guidelines, management systems, and key data. The necessary information is extracted
from financial accounts, company documentation, international databases, media
reports, interviews with key stakeholders, and ongoing contact with management.2 The
information extracted from each firm is condensed into 199 information items that
cover major stakeholder issues such as community involvement, environmental impact,
customer policies, employment relations, human rights issues, activities in controversial
areas (e.g. alcohol), supplier relations, and corporate governance. We complement these
2
SiRi does not ask companies if they wish to be included in the survey. SiRi provides detailed profiles on
the largest global companies in collaboration with SiRi’s national partners. Each national partner, using
harmonized methodology, scrutinizes social dimensions of the main corporations in its respective home
market. Beginning with the largest companies, year-to-year, national partners expand the sample of
companies analyzed, with the final objective of covering the whole home capital market (visit
www.siricompany.com for more details).
11
data on corporate responsibility with financial data from 2000-2005, extracted from
OSIRIS, a database compiled by Bureau van Dijk (BvD) that provides information on
financial, ownership and earnings for 38,000 companies, including listed, unlisted and
de-listed companies from over 130 countries. Finally, we obtain data from Bloomberg
on the MSCI world index in order to compute one of the variables of performance (the
abnormal returns). The result of merging these databases is a sample composed of 358
industrial firms3, from 22 different countries, included at least once in the 2002-2005
period.4
3.2. Measures
Corporate social performance (CSP). This variable is approached through the
SiRi PRO TM ratings. Five research fields are devoted to measuring the level of a firm’s
responsibilities to its stakeholders: community, customers, employees, environment,
and vendors and contractors. Another section provides an overview of firms’ corporate
governance practices. However, we have excluded this part from our measure of CSP,
because in our study we focus on the degree of satisfaction of the non-shareholders’
stakeholders. For each stakeholder, the database addresses a firm’s attributes in four
areas: levels of transparency and disclosure; the existence of corporate policies and
principles related to the stakeholder; the importance of management procedures; and the
level of controversies with respect to this stakeholder. In each of these areas, there are
information items that result in a Likert-type scale score ranging from 0 (worst) to 100
(best). Importantly, each information item is weighted according to a methodology
developed by SiRi. These weights are sector-specific and changed annually. The
corporate social performance indicator used in this study is the corresponding SiRi
measure defined as the weighted sum of non-shareholders’ stakeholder scores (the
scores of community, customers, employees, environment, and vendors and
contractors), using SiRi’s weights but excluding the items that give information on
transparency and disclosure for each stakeholder. This reduces any endogeneity
problems because the level of disclosure of certain costly practices aimed at satisfying
3
Although our initial sample had 777 firms, we concentrate on the 358 firms that correspond to 448
observations from which we obtain information on all the variables needed to estimate our specifiations.
This does not generate a bias because the mean values for the main variables in our sample are not
significantly different from those of the original sample composed of 777 firms.
4
The distribution of observations among the most significant countries is: US (35%), UK (12.1%),
Switzerland (11.2%), France (9.6%), Denmark (7.1%), Spain (4.9%), Italy (3.6%) and Japan (3%).
12
stakeholders’ interests may have a clear impact on investment and disinvestment
decisions by outside investors and may well trigger managerial entrenchment practices.
Workers’ satisfaction (Workers). We approach this issue using the score for the
level of a firm’s responsibilities to its employees. This score is an aggregation of 37
indicators that cover different aspects of the firm’s involvement in workers’ issues.
Also, in order to avoid endogeneity problems, we exclude the items related to
transparency from the definition of worker satisfaction.
Performance. We measure performance using three different proxies in order to
provide more robust results. First, the return on assets (ROA), measured as the ratio of
earning before interest and tax to total value of assets. Second, Tobin’s q; which is
defined on a log scale as the ratio of the sum of market equity value, plus a firm’s
liabilities, to the total assets. Last, we use abnormal returns computed using the MSCI
world index as a factor. More specifically, these abnormal returns are computed as the
difference of the expected monthly returns derived from a CAPM model using a
window of 5 years from the firm’s equity return in the corresponding year.
Managerial entrenchment. We use the measures for entrenchment provided by
the SiRi PROTM database that approximate the existence of anti-takeover devices,
limitation of shareholders’ voting rights, existence of multiple classes of stock with
different voting rights, managers’ stake and tenure.5 More specifically, Anti-takeover is
a dummy that it is equal to 1 if the firm has implemented any of the following takeover
measures: voting caps, increased voting rights over time, restrictions on board
appointment rights, and poison pills. Shareholders_Rights is measured by a 3-point
Likert scale: the highest value of this item corresponds to the situation in which major
controversies have impact on the rights and treatment of shareholders, for example,
governance arrangements that affect detrimentally the interests of shareholders, insider
trading scandals involving company directors, or major conflicts of interest among
board members. The intermediate value indicates the existence of controversies, both
major and minor, but where the company has taken credible steps to resolve the
problem. Finally, the other cases receive the lowest value. The variable
OneShare_OneVote is a dummy that it is equal to 1 if the company has multiple classes
of stock with different voting rights and 0 otherwise. Gompers et al. (2006)
demonstrates the perverse effect of having dual-class shares on firm value, caused by
the distinction between cash-flow rights and voting rights. Concerning managerial
5
In the robustness section, we introduce another measure: the income smoothing.
13
ownership, the literature shows that this variable can also be used as an entrenchment
device. To do so, we estimate Tobin’s q in terms of managerial ownership, its quadratic
term, and cubic terms (Morck et al., 1988; De Miguel et al., 2004) In the definition of
managerial ownership, we include the holdings of pension funds, bearing in mind that
managers can exercise some influence in the composition of such a stake (Farinha,
2003). As controls we also incorporate size, leverage and investment, as defined below.
The results show that the relationship between Tobin’s q and managerial ownership is
decreasing in the range between 17% and 69.8%.6 Hence we characterize this
entrenchment region with a dummy (Managown_Entrench) that is equal to 1 when
managerial stake is in this region; and 0, otherwise. Finally, we have used a more direct
measure of entrenchment that is also positively related to managers’ holding; this is the
manager’s tenure. More specifically, Manager_Tenure is defined as a dummy that takes
the value of 1 if the directors’ term in office (including managers) is more than three
years and 0 otherwise. We choose this threshold because Fredrickson et al (1988) show
that a disproportionately large number of CEOs have tenure that lasts three years or less.
Then, a CEO with tenure larger than three years is a potential candidate to become a
manager that follows entrenchment practices.
In addition to providing results using each indicator separately, we also
aggregate the scores for these five indicators. The resulting score is labeled as
Entrenchment.7 Also, in the specifications explaining market measures of performance,
we do not include the variable Managown_Entrench (defined in terms of Tobin’s q) in
the definition of Entrenchment, in order to avoid endogeneity problems. Finally, we
characterize large values of entrenchment through a variable DEntrenchment which is
equal to Entrenchment when the value is larger than the mean of the sector for the
corresponding year; and 0, otherwise.
Internal corporate governance mechanism. According to our framework, we
control for the strength of internal corporate governance. We use different variables to
approach
this
issue.
Audit_Committee,
Nomination_Committee,
and
Remuneration_Committee are 3 dummy variables obtained from SiRi; each one
receiving the value 1 if such a committee exists with independent members. Due to the
high correlation among these variables, we define Control_Committee as a 4-point
6
In De Miguel et al. (2004) the range found using Spanish data is between 30% and 70%.
7
For the sake of robustness, this variable has also been computed using the principal components of the
aforementioned indicators and the results found remain the same, qualitatively.
14
Likert scale in which 0 represents the absence of any one of these committees and 3, the
joint presence of independent audit, nomination, and remuneration committees. Nondual_CEO is a SiRi’s dummy variable that it is equal to 1 when the chairman is not the
CEO, and 0 otherwise. Board_Independence is SiRi’s Likert-type variable that takes
three different values contingent on the percentage of independent directors with respect
to the mean value of the sector. The highest value corresponds to the situation in which
a majority of non-executive directors are considered independent; the intermediate value
indicates that 50% or less of non-executive directors are independent; and zero in other
cases.8 Performance_Evaluation is another SiRi’s Likert-type scale variable that can
take three different values; it takes the value of 2 when two conditions are met: there is
a board performance evaluation system, and there are no controversies on executive
payments like payment unrelated to performance, the existence of golden parachutes,
the repricing of options, or excessive pension benefits. This variable has a value of 1
when only one of the previous conditions is met and 0 in other cases.
Another mechanism for internal control is the role played by large shareholders.
We studied this issue by employing measures of ownership concentration as well the
stakes in the hands of government. State_Ownership is the percentage of ownership in
the hands of the State. Ownership_Concentration is the stake of the three largest
blockholders,9 where we have excluded the manager’s stake in order to avoid
confounding effects with such a variable that is used as a proxy of entrenchment.
Finally, in order to study the existence of differential effects of entrenchment on
CSP contingent on the aforementioned corporate governance mechanisms, we define
the following variables that characterize situations with strong internal corporate
governance mechanisms. DControl_commitee as a dummy that it is equal to 1 if the
variable of Control_Committee is larger than the mean of the sector for the
corresponding year; and 0 otherwise. Following the same logic we define DNon-dual,
DBoard_Independence,
DPerformance_Evaluation,
DState_Ownership,
and
DOwnership_Concentration.
8
According to the SiRi questionnaire, a director is not considered independent when assessing business
may result in a conflict of interest. This precludes, among other things, a director that has held an
executive position within the company group; and/or is on the board of a significant customer or supplier
to the company; and/or has had an association with the company of more than 9 years; and/or is related
through blood, marriage or equivalent to other directors or advisers to the company.
9
Alternatively, we use the stake of the largest shareholder and that of the five largest, and the results
remain qualitatively the same.
15
Control variables. We control for financial structure, dividends, size, firms’ age,
performance, investment, growth opportunities, industry, country, and year. To control
for the financial structure, we use the variable Debt that measures the gearing of the
company. This is defined as the ratio of non-current liabilities plus loans, to
shareholders funds. Dividend is the long-term dividend policy. This is characterized by
the sector average pay-out ratio times the firm’s after-tax profits. We have used this
sectoral component in order to reduce potential endogeneity problems given that
stakeholder satisfaction has a clear impact on the proportion of funds transferred to
shareholders (pay-out ratio). Size is the value of fixed assets; and Age is the number of
years of the company’s existence. We include the aforementioned return on assets
(ROA) as a control for financial performance in the specifications of CSP as dependent
variable. Investment is the ratio of fixed assets to total assets. Growth is equal to 1 when
the rate of increase in sales is larger than the value for the corresponding sector and
year, and 0 otherwise. Intangibles is the ratio of intangible assets to total fixed assets. In
some specifications, we classify the countries according to the origin of their legal
codes. We follow La Porta et al. (1998) and we distinguish four types of countries
according to the origin of a country’s legal code (i) British common law, (ii) French
civil law, (iii) German civil law and (iv) Scandinavian civil law.
4.3. Empirical Analysis
We rely on two basic specifications to contrast our hypotheses, one explaining
CSP in terms of entrenchment and corporate governance; the other to explain
performance in terms of entrenchment and CSP.
We estimate our specifications using panel-data techniques;10 clustering the
standard errors by firm (Petersen, 2006) in order to prevent firm effects given the
persistence of the measures of social performance.11 In accordance with Petersen
(2006), we do not cluster the standard errors by year because the variability of CSP in
each year is already high enough (three times larger than the variability by firm).
10
For robustness (available upon request), we have also used cross-section estimations and the main
results coincides with those found using panel-data techniques.
11
To test whether the effects are fixed or random, we used the Hausman test. When this test tests reveals
that there is no correlation between the firm-specific error component and the explanatory variables, we
use random-effects estimations because it is the most efficient alternative (Arellano and Bond, 1991). The
results show that a majority are fixed-effect estimations.
16
In order to explain a firm’s CSP together with worker satisfaction and test
Hypotheses 1 and 2, we consider the following basic specification: 12
CSPit 1  1   2 Entrenchment it   3Control_Committeeit 
 4 Non-dual_CEOit +5Board_Independenceit  6 Performance_Evaluationit 
[1]
7State_Ownershipit +8Ownership_Concentration it   9 Debt it  10 Dividend it 
11Sizeit  12 Ageit  13ROAit  14 Investment it  15Growth it  16 Intangiblesit   it
We conduct further estimations of specification [1] 13 by breaking the variable
Entrenchment into its basic components: Anti-takeover, Shareholders_Rights,
OneShare_OneVote, Managown_Entrench, and Manager_Tenure. Also, to study the
possible moderating effect of the variables of governance, we crossed the
aforementioned variable DEntrenchment times the dummies measuring the strength of
internal governance mechanisms. The resulting multiplicative variables are the
following: DControl_Committee*DEntrenchment, DNon-dual*DEntrenchment,
DBoard_Independence*DEntrenchment, DPerformance_Evaluation* DEntrenchment,
DState_Ownership*DEntrenchment, and DOwnership_Concentration*DEntrenchment.
We focus on significant values of entrenchment because we expect that they appear as a
reaction to the existence of strong internal control mechanisms. In such situations we
have argued that CSP would be used as part of an entrenchment strategy (Hypothesis 1).
It is important to stress that by using fixed-effect estimations, we eliminate the
unobservable heterogeneity that may be potentially correlated with the independent
variables. For example, the intrinsic characteristics of the manager should condition a
firm’s CSP activities and, at the same time, may be connected to the definition of an
entrenchment strategy as well as with the governance characteristics of the firm.
Additionally, as we explain in the theoretical section, we expect that pressure from
12
We rule out specifications that include the dependent variable lagged by one period because the
persistence in CSP may generate endogeneity problems that can only be tackled through GMM
estimations. Unfortunately, the limited number of years available does not permit us to use such a
technique. Nevertheless, in an unreported estimation available upon request, we introduce the dependent
variable lagged by one period and estimate such a specification using clustered fixed-effects. The results
found are consistent with those reported in Table 3 and, remarkably, the coefficient of the CSP variable
lagged by one period is not very significant. This may suggest that the question of persistence does not
bias significantly the results.
13
For estimates using fixed-effects, temporal, sectoral and country dummies are not needed. For random-
effects estimation, we correct by temporal, sector and country effects by detracting from the dependent
variable its mean value for the corresponding year, sector and country.
17
different stakeholders is connected to different internal corporate control mechanisms.
This means that an endogeneity problem in specification [1], which it is not directly
connected to the unobservable heterogeneity, may exist perfectly. We tackle this
problem by advancing the dependent variable by one period.
Concerning Hypothesis 2, we modify slightly the previous specification and
substitute the dependent variable of CSP with that of Workers’ satisfaction (Workers).
In order to test Hypotheses 3 and 4, we focus on a specification that defines
corporate financial performance in terms of entrenchment and CSP. The basic
specification is:
Performanceit 1  1   2 Entrenchment it   3CSPit   4 DEntrenchment_DCSPit 
5Control_Committeeit   6 Non-dual_CEOit + 7 Board_Independenceit 
[2]
8 Perform _ Evaluationit   9State_Ownershipit +10Ownership_Concent it 
11Debt it  12 Dividend it  13Sizeit  14 Ageit  15Investment it 
16Growth it  17 Intangiblesit   it
From this specification, we test whether entrenchment when combined with a
variation in CSP (Hypothesis 3) or in workers’ satisfaction (Hypothesis 4) have further
negative effects on financial performance. This test requires the inclusion of an
interaction term between variations in CSP and entrenchment practices. Moreover,
consistent with what we have mentioned in the theoretical section, we expect the
complementarity between both variables (entrenchment and increases in CSP) to appear
mainly in a context with efficient governance mechanisms where managers set on
entrenchment will reinforce their strategy with the implementation of social responsible
actions. In such situations, we expect intensive managerial entrenchment complemented
with improvements in socially responsible actions. Thus, we focus on the interactive
variable
DEntrenchment_DCSP
that
crosses
the
aforementioned
variable
DEntrenchment, which is equal to Entrenchment when entrenchment is larger than the
mean of the sector for the corresponding year, with variations in CSP (DCSP).
Following the same logic, we define DEntrenchment_DWorkers. Then, the
complementarity hypothesis is supported if the coefficient of such a variable is negative.
4. RESULTS
Table 1 shows information on the distribution of each variable of our study. The
data show that CSP has a median value which is in the middle of its range of definition
18
(49.4)14 while for Workers it is slightly higher (53.7). For Entrenchment, the median
value is 1 in a range between 0 and 4 and the ROA, as a variable of performance, has a
median value of 5%. Concerning the variables for internal control mechanism, their
median
values
are
3,
1,
1,
1
for
Control_Committee;
Non-dual_CEO;
Board_Independence and Performance_Evaluation respectively, which corresponds to
their maximum values of definition. This is in contrast with the remaining variables for
internal control mechanisms, where State_Ownership has a value of 0.5% and the sum
of the stakes of the three largest shareholders (Ownership_Concentration) is 24%.
Finally, in terms of the control variables, the median size of a company in the sample is
7,064 million €; the age is 82 years; firm Debt is 90%, the Investment variable has a
median value of 45.9% and the median proportion of Intangibles is 26.3%.
Then, the inspection of this second quartile (median values), reveals that almost
all variables have non-null values.15 This eliminates concerns about the existence of
truncated distributions around the zero value. Table 1 also shows the evolution of the
variables comparing the first year of the sample (2002) with the last one (2005). It is
remarkable that the variable for CSP increases steadily from a value of 37% in 2002 to a
value of 57.5% in 2005. This reflects the growing interest in socially responsible
activities with time.16 Precisely this fact allows using such activities as a managerial
entrenchment mechanism. Finally, in Table 1, we present a mean test analysis
comparing firms with large values of CSP to their counterparts. We find that among the
companies with large values of CSP, the Entrenchment variable has significantly larger
values than their counterparts (0.9 versus 0.7), while the variables for Performance
show lower ones (5.2 versus 7.2 for ROA). This conforms to our theoretical
contentions.
14
Luxemburg, Finland, and Denmark are the countries with the largest values of CSP. Remarkably
Denmark is also among the countries with the largest values for the entrenchment variable, which goes in
line with our theoretical contention that relates both variables positively.
15
Among the exceptions there are two components of the entrenchment variable (Anti-takeover and
Managown_Entrench), although the aggregate entrenchment variable has a non-null median value.
16
For example, in the report “Green, Social, and Ethical Funds in Europe: 2005 Review” of Avanzi SRI
Research and SiRi Company, we observe a substantial increase in the number of SRI funds since 2003,
and more importantly that the total amount of SRI assets grew by about 99%: from a value of €12 billion,
in 2003, to a value of €24 billion at the end of second quarter of 2005. This gives us an idea of the
increasing importance of social performance, as these ethical funds invest only in firms with high ratings
in CSR.
19
Table 2 displays the correlation matrix. On inspection of the correlation matrix,
we find that the variable of entrenchment is positively correlated (7%) with that of
CSP17. This is particularly evident for anti-takeover initiatives, the deterioration of
voting rights (OneShare_OneVote) as well as managerial tenure. Additionally, CSP is
also negatively correlated to different internal control mechanisms like the non-duality
between the CEO and the chairman of the board (Non-dual_CEO), and the proportion
of independent directors. We interpret these results as evidence that, when entrenchment
is more difficult as a result of internal control pressures, CSP initiatives are also less
likely. However, we show in the following analysis that, if firms are able to define a
significant entrenchment strategy within a framework where internal governance
mechanisms are well developed, this has a positive impact on a firm’s CSP.
----------------------------------------
Insert Tables 1 and 2 about here
----------------------------------------
Table 3 summarizes the regression analysis of specification [1], whereby we test
the effect of managerial entrenchment practices on CSP. Further, in Column 1 we break
the
variable
Entrenchment
into
its
five
components:
Anti-takeover,
Shareholders_Rights, OneShare_OneVote, Managown_Entrench, and Manager_tenure.
We obtain that anti-takeover measures, as well as managerial tenure have a positive
impact on CSP. Also, when managers’ stake moves into the “entrenchment region”
(between 17% and 69.8%), it has a positive impact on CSP. A second result shows that
when entrenchment is more difficult, due to the presence of a significant proportion of
independent members on the board, or the existence of independent control committees,
there is a negative impact on CSP. This reflects the complementarity idea between
entrenchment and CSP that this paper relies on. Moreover, the aggregate measure of
entrenchment has a positive impact on CSP (Column 2). Once we look at the coefficient
of the multiplicative variables (Column 3), we find that entrenchment has a particular
positive impact on CSP in those scenarios where entrenchment is more difficult ex-ante.
Mainly, when the number of control committees with independent members is larger
than the mean for the sector in the corresponding year; when there is a separation
between the CEO and the chairman of the board, as well as when the performance
17
This result is more significant when we look at variations in CSP; they are positively correlated with
those variations of entrenchment (13.6%). The correlations in differences for other variables show
qualitatively the same results and are available upon request.
20
evaluation systems are more developed than the average for the corresponding sector
and year. In such cases, corporate governance is more developed and, if the manager is
set on entrenchment, he complements this strategy with increases in CSP. This
conforms to Hypothesis 1.
---------------------------------------Insert Table 3 about here
----------------------------------------
In addition, we find that a generous dividend policy hinders the implementation
of policies aimed at satisfying non-shareholder stakeholders. This result is in
accordance with the slack resources hypothesis of Waddock and Graves (1997), which
suggests that better financial performance results in a surplus of resources that provides
firms with the financial wherewithal to consider and do something about social issues
Finally, and consistent with the growing interest in CSP issues, older firms (with an
accumulated reputation) show larger values in CSP.
Once we focus on workers’ satisfaction (Table 4), the results are, in essence, the
same as those for the overall score of CSP. This conforms to Hypothesis 2. The highly
significant coefficient of anti-takeover measures in explaining a firm’s CSP is
remarkable. This is in accordance with Pagano and Volpin (2005), where wage
concessions – that improve workers’ satisfaction – are described as anti-takeover
initiatives. The positive sign for the State_Ownership variable can be explained in terms
of the willingness of government-owned firm managers to satisfy workers as part of an
entrenchment strategy, given that public authorities consider workers as potential
constituencies. This will reinforce the position of managers in such firms. A similar
argument can explain the positive moderating role of the State_Ownership variable in
the connection between entrenchment and CSP. Finally, the non-negative sign of the
dividend variable can be explained by the fact that workers may well have shares in the
firm. In fact, the non-significant sign may be the result of compensating the positive
effect due the dividends received with the negative effect as dividends reduces the
available resources for satisfying workers’ interests.
----------------------------------------
Insert Table 4 about here
----------------------------------------
In order to analyze the consequence of entrenchment combined with CSP, we
show in Tables 5A, 5B and 5C, the results of estimating specification [2] using different
21
measures of performance. In Table 5A, we use ROA; in Table 5B, Tobin’s q; and in
Table 5C, abnormal returns derived from a single factor model.18
----------------------------------------
Insert Table 5A about here
----------------------------------------
Once we focus on the estimations for ROA19, we find that entrenchment has a
negative impact on ROA, a result which is consistent with previous literature (e.g.,
Jensen and Ruback, 1983). Also, we find that this effect is more significant when
entrenchment is large and it is defined in combination with increases in CSP (Column
2) and in workers’ satisfaction (Column 4). That is, there is a complementary
relationship between both variables: when there is an increase in social performance
and/or in workers’ satisfaction, the marginal effect of entrenchment, when large, on the
returns on assets is more negative. This conforms to Hypotheses 3 and 4.
It is important to highlight that the results hold for significant values of
entrenchment (larger than the mean for the sector and year). We have argued previously
that this corresponds to situations of large internal control, where we expect that
entrenchment, when it appears, trigger increases in socially responsible policies that
reinforce the power of the manager. This generates a decrease in performance. Among
the control variables, the positive effect of debt on performance is remarkable. This
effect may be explained by the fact that it reduces managerial discretion (Jensen, 1986)
18
Given the endogeneity between market measures of performance and ownership structure (Demsetz
and Villalonga, 2001), we have detracted from the variable of entrenchment the component on managerial
ownership in those estimations that use market measures of performance (Tables 5B and 5C). Note that
one of the components of entrenchment (Managown_Entrench) is defined in terms of a decreasing
relationship between managerial ownership and Tobin’s q.
19
All variables of performance are defined in relative terms by detracting the mean of the sector and year
from each firm’s performance value. This definition gives information on the real performance of each
firm when compared with the average for each sector in the year analyzed. Also, the demeaning of
performance allows us to eliminate sectoral, temporal and country effects from performance, given that,
contrary to specification [1], some specifications of [2] are estimated using random effects as we have
explained previously (when the Hausman tests reveals that firm-specific error component is not correlated
with the independent variables).
22
-------------------------------------------
Insert Table 5B and 5C about here
-------------------------------------------
The results on Tobin’s q and abnormal returns (Tables 5B and 5C respectively)
conform to those using ROA.
Given the previous set of results, we provide support for the thesis regarding the
existence of an entrenchment motive to justify certain increases in CSP, particularly
when managerial control mechanisms are well developed. In this case, entrenchment is
less likely. However, if it does occur, it is so intensive that it also triggers socially
responsible actions. This is a reinforcing mechanism, given that managerial collusion
with non-shareholder stakeholders allows them to channel stakeholders’ power against
shareholders. In this situation there is a clear negative impact on financial performance.
5/ ROBUSTNESS
5.1/ Income Smoothing
In order to investigate the robustness of our results, we maintain that one form of
earnings manipulation, income smoothing, is an additional entrenchment mechanism.
Managers smooth earnings as a natural entrenchment strategy in order to ensure a stable
stream of profits that will satisfy shareholders (Fudenberg and Tirole, 1995). However,
although earnings manipulations improve financial performance in the short-term, they
damage the medium-term interests of shareholders. The manager anticipates this fact
and has all the incentives to trigger entrenchment initiatives. Within this setting, we
characterize a situation where we expect to find, according to our theory, an increase in
CSP. This is a way to test the robustness of our results.
To test this contention, we use two alternative measures of income smoothing
(Leuz et al., 2003). First, we approach this variable through the correlation between
changes
in
accruals
and
changes
in
cash
flow
(Incsmooth1),
where
Accruals   CA  Cash    CL  STD   DEP , with ΔCA is the change in current
assets; ΔCash is the change in cash; ΔCL is the change in current liabilities; ΔSTD is
the change in debt included in current liabilities; and DEP is depreciation and
amortization. The second measure (Incsmooth2) is the ratio of net income before
23
extraordinary items to the standard deviation of cash-flows.20 In Columns 1 and 2 of
Table 6, we use the first measure; while in Columns 3 and 4, we use the second
measure.
----------------------------------------
Insert Tables 6 about here
----------------------------------------
From this table, we show that income smoothing has a positive impact on CSP
as well as on workers’ satisfaction. It is important to stress that this is not due to the
increase in the short-term financial performance due to earnings manipulation, because
we have controlled by a performance variable through the ROA. Our explanation, which
supports our basic theory, is that income smoothing is connected with entrenchment
practices that may further stimulate improvements in CSP.21
5.2 Expropriation of minority shareholders
We conduct an additional analysis to investigate if the changes of CSP and in
workers’ satisfaction are explained in terms of other agency problems like minority
expropriation, instead of entrenchment, as mentioned in the theoretical part. In the
corporate governance literature (e.g., Shleifer and Vishny, 1997), the presence of
blockholders (e.g., families) is contemplated to have an ambiguous effect on firm’s
financial performance. On the positive side, blockholders diminish the entrenchment
possibilities of managers, which impacts positively on performance; but, on the negative
side, large shareholders may expropriate minority shareholders thereby reducing the
market price of shares. One strategy that blockholders may follow to expropriate
minorities is the overinvestment in CSP (Barnea and Rubin, 2006). By implementing
certain social programs, blockholders receive the full benefits associated with CSP, but
only bear a portion of the costs to implement such policies (proportional to their stakes).
20
Unfortunately, we have no information on cash-flow statements and we estimate accruals from balance
sheet changes in working capital (excluding cash). Although this introduces some bias into the
estimations using the first measure (Hribar and Collins, 2002), the coincidence of the results, once we use
the second measure, gives us confidence in our results.
21
We have conducted an alternative analysis (available upon request) where earnings manipulation is
approached through the discretionary accruals obtained as the difference between the accruals and the
expected ones given by the models of Jones (1991), Dechow et al. (1995), and Kothari et al. (2005),
respectively. The results are consistent with those found using the proxies for income smoothing.
24
This association between ownership concentration and CSP found support in some
recent studies (Barnea and Rubin, 2006; Neubaum and Zahra, 2006).
Keeping in mind this idea, it is important to distinguish improvements in CSP
that are connected to the entrenchment practices from those that may be explained in
terms of the implementation of expropriating strategies. We try to tackle this problem
by introducing in each specification variables for ownership concentration. In Table 7,
we extend this analysis and distinguish firms according to their ownership concentration
(Columns 1 and 2). This is proxied by the stake of the three largest blockholders.
Additionally, we distinguish between family firms – when the largest blockholder
belongs to a family – and non-family ones. Authors agree that expropriation is more
likely among the former firms. Finally, we compare firms where workers have
substantial power (e.g. they are unionized) with their counterparts (Columns 5 and 6).
We define firms with worker power as those where workers have board level
responsibilities and/or benefit from profit sharing programs. We expect that the agency
problem between large blockholders and minority ones will be less likely in firms with
worker power, once we take into consideration the workers’ stake of in the firm.
----------------------------------------
Insert Table 7 about here
----------------------------------------
We see from the results that the impact of our proxy of entrenchment on CSP
has only positive effects in those firms where ownership is not concentrated (lower than
the mean of the sector for the corresponding year), in non-family firms, and in firms
with significant worker power. In these firms, we expect that there is no expropriation
to minority shareholders. This allows ensuring that the effect found on CSP is explained
exclusively in terms of entrenchment and not in terms of expropriation. This gives
further support to our results.
5.3. Leverage, degree of financial markets efficiency, and size
As a robustness check, we conduct three additional analyses. First, we split the
sample into those firms whose leverage is larger than the mean of their corresponding
sector and year. The results, reported in the first two columns of Table 8, show that the
effect of entrenchment on CSP is only visible for high-leverage firms. This is consistent
with the idea that debt complements internal corporate governance mechanisms to
control managers and limit their discretion. As such, in accordance to Hypotheses 1, we
25
expect that the impact if entrenchment on CSP should be more significant for highlyleveraged firms.
----------------------------------------
Insert Table 8 about here
----------------------------------------
A second analysis is performed by comparing firms whose country of origin has
an Anglo-Saxon legal code, to those whose countries have a German, or French-origin
legal code. We have argued previously that financial markets are not fully efficient and
do not fully internalize the costs of implementing costly socially responsible actions
linked to entrenchment. This fact, among other things, explains the use of socially
responsible actions as an entrenchment device, even in the presence of external
corporate governance mechanisms. A consequence of this is that the results connecting
entrenchment and CSP should be more evident in countries with less developed
financial markets. As La Porta et al. (1998) show, countries with French and Germanorigin legal codes have less developed financial markets compared with Anglo-Saxon
ones. Consistently, we find that entrenchment only explains our measure for CSP in
countries whose legal codes have French or German origins. Note also, that countries
with less developed external corporate control mechanisms are also those with more
developed internal corporate control mechanisms. In such cases, according to our
Hypothesis 1, a much stronger impact of entrenchment on CSP should be expected.
A final concern is the existence of a possible sample selection bias given that the
firms in our sample are mainly large firms. In order to investigate such an issue in more
detail, we split the sample into large firms (size larger than the mean of the sector for
the corresponding year) and small ones. The results (Columns 5 and 6) show, as
expected, that the effect is slighter more significant for large firms, but it also appears
for small ones. This means that it may be some size effects but our results are still valid
for smaller firms. This excludes problems linked to sample selection bias.
6. CONCLUSIONS
In this paper, we investigate the effect of the implementation of entrenchment
strategies on socially responsible actions. Our basic premise is that the manager may be
controlled by shareholders – externally through the financial markets and internally
through the board of directors – as well as by the activism of different stakeholders.
Within such a context, entrenchment strategies aimed at hindering the actions of
shareholders are ineffective, unless accompanied by other measures tailored to
neutralize stakeholder pressure. In this case, there are two possibilities: confrontation
26
with stakeholders; or collusion – so as to satisfy their interests. We argue that an
entrenched manager will choose the collusion strategy especially when internal
corporate governance mechanisms are well developed. In such a situation, the collusion
with non-shareholder stakeholders will not only tackle the pressure from stakeholders
but will, more importantly, channel the salience of these stakeholders against agents –
shareholders – who intend to replace the manager.
We tested this claim by looking at the connection between entrenchment and the
scores for corporate social performance (CSP) as well as for worker satisfaction. We
find that there is a clear positive impact by the former on the latter variables, especially
in those firms with efficient internal corporate governance mechanisms like the
existence of independent control committees, the separation between the CEO and the
chairman of the board, and the implementation of performance evaluation schemes.
This confirms the main theoretical contention of the paper: a firm’s socially responsible
activities form part of a manager’s entrenchment strategy.
Additionally, we prove the robustness of this contention by using an alternative
proxy of entrenchment that is the implementation of earnings manipulation practices
based on income smoothing (Fudenberg and Tirole, 1995). Our results are fully
consistent with the theoretical claim.
The second step in our analysis consists of looking at the impact on performance
brought on by an entrenchment strategy. We find that the negative impact on financial
performance is more pronounced when an intensive entrenchment strategy is combined
with increases in socially responsible actions. Accordingly, we argue that an intensive
entrenchment policy is the reaction to the existence of stringent internal governance
mechanisms. In such a situation, although entrenchment is less likely, a manager set on
entrenchment combines such an aggressive strategy with the development of socially
responsible activities. This combination has particular negative effects on shareholders’
wealth. This factor is further evidence that entrenched managers heed stakeholder
satisfaction not only as a consequence of stakeholder activism, but also as a way of
reinforcing their entrenchment strategy against shareholders.
Remarkably, our result also holds true when we focus on a stakeholder group,
the workers. We argue that these stakeholders are amongst the most powerful and the
entrenched manager should pay particular attention to looking after their interests.
Finally, we show that the results are more pronounced in countries with less
efficient financial markets (civil-law countries) and more developed internal corporate
control mechanisms. In such countries, managers can take advantage of that inefficiency
27
and over-invest in socially responsible activities in order to complement an
entrenchment strategy, given that market prices do not fully reflect the cost of
implementing these activities. Also, we discard other explanations linked to the
implementation of expropriating policies in order to explain the increases in social
performance. We also prove that increases in CSP linked to entrenchment appear in
firms with low ownership concentration, particularly in non-family firms, and in firms
where workers enjoy significant power. Minority expropriation is less likely in such
firms; being, therefore, the increases in CSP due to entrenchment and not connected to
expropriating issues.
6.1 Implications
This work forms a bridge between the corporate governance literature and
stakeholder theory. According to this latter line of research, the management of
stakeholders is a good way of improving financial results (Jones, 1995), whereas
corporate governance emphasizes the difficulty in reconciling the demands of a wide set
of stakeholders (Jensen, 2001; Tirole, 2001). We show that trying to satisfy different
stakeholders’ interests, independently of their salience, may have bad consequences on
performance when combined with the implementation of entrenchment policies. Then, it
is not a good policy to give the managers any leeway in determining the degree of
satisfaction of non-shareholder stakeholders because managers may choose socially
responsible
activities,
strategically
to
complement
entrenchment
initiatives.
Furthermore, we find that the existence of strong internal corporate governance
mechanisms is not a guarantee that CSP may not be misused. In fact, the linkage
between CSP and entrenchment is stronger for those firms with strong corporate
governance. Then, how does one deal with such a problem?
There are different possibilities. First, in line with Cespa and Cestone (2007), if
a firm’s CSP may be used as an integral part of an entrenchment strategy, then, some
form of governance mechanism that hinders managerial discretion on social issues is
needed. A possibility is to regulate social issues in order to avoid overinvestment in
socially responsible actions. Undoubtedly, mandatory accounting practices to reflect
these issues on the balance sheet may be a first step in this direction. A second way to
prevent entrenchment problems, especially when involving other stakeholders, is to
transfer ownership to this group. For example, if workers also have shares in a firm, it
will not make sense for the manager to implement simultaneously an entrenchment
strategy of confrontation with shareholders and another of collusion with other
28
stakeholders like workers because the interests of the latter will also be aligned with
those of shareholders. Paradoxically, the interests of shareholders are better defended by
transferring part of their powers to other stakeholders.
6.2. Competing Hypotheses and Future Research
Alternative theoretical arguments may compete with our entrenchment
hypothesis. Firstly, we can explain the positive connection between entrenchment and
CSP by looking at normative or self-promotion motives (see Donaldson and Preston,
1995). Managers with high ethical standards would promote CSP while, at the same
time, define entrenchment strategies that are geared towards the wellbeing of the firm
according to themselves. Within such framework, CSP is used normatively, not
strategically, and generates the same set of results that we obtain: a positive connection
with entrenchment and, when complemented with this latter variable, a negative effect
on performance. In order to distinguish this hypothesis from our theoretical contention,
we need to enlarge the windows of years of analysis given that we suspect that sociallyfriendly initiatives based on ethical grounds will be less sensible to the economic cycle
than strategic ones are. The investigation of this issue is left for future research once
new sets of data were available. However, as preliminary evidence, we find that when
we compare the results for the first year of our sample with those of the last year, the
latter are more significant. This conforms to a strategic use of CSP as an entrenchment
mechanism based on the fact that social awareness has been growing over time. Under a
normative view, we would not observe these differences over time.
A second competing argument is that managers seek the social prestige and
legitimacy of being good corporate citizens. Such managers will try to maintain their
status by implementing CSP while, at the same time, entrenching their position in the
firm so as to enjoy the benefits of such status. The outcome of such a combination is a
negative effect on performance. This hypothesis, however, cannot explain the positive
moderating role of internal corporate control mechanisms in connecting entrenchment
and CSP, just like we find in our empirical analysis.
Finally, the positive association between entrenchment and socially responsible
activities may be explained by the view that managers, who resort to such practices, will
be able to develop longer-term relationships with stakeholders like customers, suppliers,
workers, and communities. This view will expand the set of value-creating exchanges
and, therefore, generate better results. In order to ensure the longevity of such valuable
relationships managers may trigger entrenchment strategies. However, this positive
29
view of managerial entrenchment is not supported by the performance analysis
conducted in this study; we find a negative impact of the combination of entrenchment
and CSP on financial results.
Our work can be expanded in several directions. One research avenue consists of
investigating, in greater depth, the connection between ownership structure and the
entrenchment motive behind some socially responsible activities. The type of
blockholder and its social sensitivity is expected to be relevant in the strategic use of
CSP as an entrenchment mechanism. Finally, a further investigation on different
institutional contexts may also be of interest given the significant differences that exist
in top-management orientation across countries. In Anglo-Saxon countries, managers
are more inclined to satisfy shareholders’ interests, while in Continental Europe and
Japan, managers have traditionally been more sensitive to the development of long-term
relationships with employees, banks, and suppliers. The investigation of this aspect as
well as other issues is left for future research.
30
REFERENCES
Aguilera, R.V. and Jackson, G. (2003), “The cross-national diversity of corporate
governance: Dimensions and determinants”, Academy of Management Review,
28(3), pp. 447-465.
Anderson, R.C. and Reeb, D.M. (2004), “Board composition: Balancing family
influence in S&P 500 firms”, Administrative Science Quarterly, 49, pp. 209–237.
Arellano, M., and Bond, S. (1991), “Some tests of specification for panel data: Monte
Carlo evidence and an application to employment equations”, Review of
Economic Studies, 58, pp. 277-297.
Barnea, A. and Rubin, A. (2006), “Corporate Social Responsibility as a Conflict
Between Shareholders”. EFA 2006 Zurich Meetings Available at SSRN:
http://ssrn.com/abstract=686606
Baron, D.P. (2001), “Private politics, corporate social responsibility, and integrated
strategy”, Journal of Economics and Management Strategy, 10, pp. 7-45.
Berman, S., Wicks, A., Kotha, S. and Jones, T. (1999), “Does stakeholder orientation
matter? The relationship between stakeholder management models and firm
financial performance”, Academy of Management Journal, 42(5), pp. 488-506.
Cespa, G. and Cestone, G. (2007), “Corporate Social Responsibility and Managerial
Entrenchment”, Journal of Economics and Management Strategy (forthcoming).
Conyon, M.J., Girma, S., Thompson, S. and Wright, P.W. (2001), “Do hostile mergers
destroy jobs?”, Journal of Economic Behavior and Organization, 45, pp. 427-440.
Dahya, J., Lonie, A. and Power, D. (1998), “Ownership structure, firm performance and
top executive change: an analysis of UK firms”, Journal of Business Finance and
Accounting, 25, pp. 1089-1118.
De Miguel, A., Pindado, J. and de la Torre, C. (2004), “Ownership structure and firm
value: New evidence from Spain”, Strategic Management Journal, 25(12), pp.
1119-1207.
DeAngelo, H. and DeAngelo, L. (1998), “Ancient redwoods and the politics of finance:
The hostile takeover of the Pacific Lumber Company”, Journal of Financial
Economics, 47(1), pp. 3-53.
Dechow, P., Sloan, R. and Sweeney, A. (1995), “Detecting earnings management”, The
Accounting Review, 70, pp. 193-225.
Demsetz, H. and Villalonga, B. (2001), "Ownership structure and corporate
performance" Journal of Corporate Finance 7 (3), pp. 209-233.
31
Denis, D., Denis, D. and Sarin, A. (1997), “Ownership structure and top executive
turnover”, Journal of Financial Economics, 45, pp. 193-221
Donaldson, T.L. and Preston, L.E. (1995), “The stakeholder theory of the corporation:
Concepts, evidence, and implications”, Academy of Management Review, 20(1),
pp. 65-91.
Fama, E. and Jensen, M. (1983), “Separation of ownership and control”, Journal of Law
and Economics, 26(2), pp. 301-25.
Farinha, J. (2003), “Dividend policy, corporate governance and the managerial
entrenchment hypothesis: an empirical analysis”, Journal of Business Finance and
Accounting, 30 (9&10), pp. 1173-1209.
Feddersen, T. and Gilligan, T. (2001), “Saints and markets: Activists and the supply of
credence goods”, Journal of Economics and Management Strategy, 10, pp. 149171.
Fredrickson, J. W., Hambrick, D. C. and Baumrin, S. (1988), “A model of CEO
dismissal”, Academy of Management Review, 13, pp. 255-270.
Freeman, R.E. (1984). Strategic management: A stakeholder approach. (Pitman: Boston,
MA).
Fudenberg, D. and Tirole, J. (1995), “A theory of income and dividend smoothing based
on incumbency rents”, Journal of Political Economy, 103, pp. 75-95. DOI:
10.1086/261976
Gompers, P. A., Ishii, J. and Metrick, A. (2006), “Extreme governance: An analysis of
dual-class firms in the United States”, mimeo.
Hellwig, M. (2000). “On the economics and politics of corporate finance and corporate
control”. In: Vives, X. (Ed.), Corporate Governance: Theoretical and Empirical
Perspectives. Cambridge University Press, Cambridge, England, pp. 95–136.
Hill, C.W. and Jones, T.M. (1992), “Stakeholder-agency theory”, Journal of
Management Studies, 29, pp. 131-154.
Hillman, A.J. and Keim, G.D. (2001), “Shareholder value, stakeholder management,
and social issues: What’s the bottom line?”, Strategic Management Journal,
22(2), pp. 125-139.
Hosmer, L.T. (1994), “Strategic planning as of ethics mattered”, Strategic Management
Journal, 15, pp. 17-34.
Hribar, P. and Collins, D.W. (2002), Errors in estimating accruals: Implications for
empirical research. Journal of Accounting Research, 40, pp. 105-134.
32
Jensen, M. (2001), “Value maximization, stakeholder theory, and the corporate
objective function”, Journal of Applied Corporate Finance, 14, pp. 8-21.
Jensen, M. (1986), “Agency costs of free cash-flow, corporate finance and takeovers”,
American Economic Review, 76, pp. 323-329.
Jensen, M. and Meckling, W. (1976) “Theory of firm: Managerial behavior, agency cost
and capital structure”, Journal of Financial Economics, 3, pp. 305-360.
Jensen, M. and Ruback, R. (1983), “The market for corporate control: The scientific
evidence”, Journal of Financial Economics, 11, pp. 5-50.
John, A. and Klein, J. (2003), “The boycott puzzle: Consumer motivations for purchase
sacrifice”, Management Science, 49(9), pp. 1196-1209.
Jones, J. (1991) “Earnings management during import relief investigations”, Journal of
Accounting Research, 29, pp. 193-228.
Jones, T.M. (1995), “Instrumental stakeholder theory: A synthesis of ethics and
economics”, Academy of Management Review, 20(2), pp. 404-437.
Kothari, S.P., Leone, A.J. and Wasley, C.E. (2005), “Performance Matched
Discretionary Accrual Measures”, Journal of Accounting and Economics, 39, pp.
163-197.
La Porta, R., Lopez-de-Silanes, F., Shleifer, A. and Vishny, R. (1998), “Law and
Finance,” Journal of Political Economy, 106, 1113-1155.
Leuz, C.; Nandab, D. and Wysockic, P. D. (2003), “Earnings management and investor
protection: an international comparison”, Journal of Financial Economics, 69, pp.
505–527.
Luoma, P. and Goodstein, J. (1999), “Stakeholders and corporate boards: institutional
influences on board composition and structure”, Academy of Management
Journal, 42(5), pp. 553-563.
McWilliams, A. and Siegel, D. (2001), “Corporate social responsibility: A theory of the
firm perspective”, Academy of Management Review, 26(1), pp. 117-127.
Mitchell, R.K., Agle, B.R. and Wood, D.J. (1997), “Toward a theory of stakeholder
identification and salience: Defining the principle of who and what really counts”,
Academy of Management Review, 22(4), pp. 853-886.
Morck, R., Shleifer, A. and Vishny, R. (1988), “Managerial ownership and market
valuation”, Journal of Financial Economics, 20, pp. 293-315.
Neubaum, D.O. and Zahra, S.A. (2006), “Institutional ownership and corporate social
performance: The moderating effects of investment horizon, activism, and
coordination”, Journal of Management, 32, pp. 108-131.
33
Pagano, M. and Volpin, P. (2005), “Managers, workers, and corporate control”, The
Journal of Finance, 60, pp. 841-868.
Petersen, M. A. (2006), “Estimating standard errors in finance panel data sets:
Comparing approaches”, Mimeo.
Rowley, T. and Berman, S. (2000), “A brand new brand of corporate social
performance”, Business and Society, 39(4), pp. 397-418.
Schneper, W.D. and Guillén, M.F. (2004), “Stakeholder rights and corporate
governance: A cross-national study of hostile takeovers”, Administrative Science
Quarterly, 49(2), pp. 263-295.
Shleifer, A. and Vishny, R.W. (1997), “A survey of corporate governance”, Journal of
Finance, 52(2), pp. 737-83.
Shleifer, A. and Vishny, R. (1989), “Management entrenchment: The case of managerspecific investments”, Journal of Financial Economics, 25, pp. 123-140.
Shleifer, A. and Vishny, R. (1986), “Large shareholders and corporate control”, Journal
of Political Economy, 94(3), pp. 461-88.
Stevens, J.M., Steensma, H.K., Harrison, D.A. and Cochran, P.L. (2005) “Symbolic or
substantive document? The influence of ethics codes on financial executives'
decisions”, Strategic Management Journal, 26, pp. 181-195.
Strong, K.C., Ringer, R.C., and Taylor, S.A. (2001), “The ruled of stakeholder
satisfaction (timeliness, honesty, empathy”, Journal of Business Ethics, 32(3), pp.
219-230.
Stulz, R. (1988), “Managerial control of voting rights: financing policies and the market
for corporate control”, Journal of Financial Economics, 20, pp. 25-54.
Sundaramurthy,
C.
(2000),
“Antitakeover
provisions
and
shareholder
value
implications: A review and a contingency framework”, Journal of Management,
26(5), pp. 1005-1030.
Sundaramurthy, C., Mahoney, J.M. and Mahoney, J.T. (1997), “Board structure,
antitakeover provisions, and stockholder wealth”, Strategic Management Journal,
18, pp. 231-245.
Tirole, J. (2001), “Corporate governance”, Econometrica 69(1), pp. 1-35.
Vafeas, N. (1999), “The nature of board nominating committees and their role in
corporate governance”, Journal of Business Finance and Accounting, 26(1), pp.
199-225.
Waddock, S.A. (2004), “Parallel universes: Companies, academics, and the progress of
corporate citizenship”, Business and Society Review, 109(1), pp. 5-42.
34
Waddock, S.A. and Graves, S.B. (1997), “The corporate social performance-financial
performance link”, Strategic Management Journal, 18(4), pp. 303-319.
Walsh, J.P. and Seward, J.K. (1990), “On the efficiency of internal and external
corporate control mechanisms”, Academy of Management Review, 15(3), pp. 421458.
Yeo, G. H. H., Tan, P. M.S., Ho, K. W., and Chen, S. (2002), “Corporate ownership
structure and the informativeness of earnings”, Journal of Business Finance &
Accounting, 29 (7&8), pp. 1023-1040. doi:10.1111/1468-5957.00460.
35
TABLE 1: Descriptive Statistics
Variable
Obs
Mean
Std.
Min
25%
50%
75%
Max
2002
2005 Large_CSP=0 1 Large_CSP=1 1
CSP
448
48.98
14.93
18.51
37.35
49.39
60.96
79.09
37.02
57.47
37.02
57.47
Workers
448
53.34
15.57
14.12
41.50
53.66
65.83
89.60
44.61
62.25
44.61
62.25
Anti-takeover
448
0.47
0.50
0.00
0.00
0.00
1.00
1.00
0.51
0.61
0.51
0.61
OneShare_OneVote
448
0.75
0.43
0.00
1.00
1.00
1.00
1.00
0.79
0.74
0.79
0.74
Shareholders_Rights
448
0.92
0.27
0.00
1.00
1.00
1.00
1.00
0.94
0.92
0.94
0.92
Managown_Entrench
448
0.02
0.12
0.00
0.00
0.00
0.00
1.00
0.03
0.00
0.03
0.00
Manager_Tenure
448
0.36
0.38
0.00
0.00
0.30
0.50
1.00
0.26
0.31
0.26
0.31
Entrenchment
448
1.33
0.68
0.00
1.00
1.00
2.00
4.00
0.74
0.94
0.74
0.94
ROA
448
5.57
6.34
-32.11
2.82
5.04
8.54
35.61
7.24
5.22
7.24
5.22
Tobin’s_q (log)
448
-0.80
0.40
-2.45
-1.01
-0.76
-0.53
0.10
0.01
-0.01
0.01
-0.01
Abn_Returns
448
0.00
0.01
-0.08
0.00
0.00
0.01
0.08
0.08
-0.02
0.08
-0.02
Control_Committee
448
1.81
1.37
0.00
0.00
3.00
3.00
3.00
1.65
1.32
1.65
1.32
Non-dual_CEO
448
0.62
0.49
0.00
0.00
1.00
1.00
1.00
0.69
0.86
0.69
0.86
Board_Independence
448
0.87
0.32
0.00
1.00
1.00
1.00
1.00
0.82
0.80
0.82
0.80
Performance_Evaluation
448
1.36
0.50
0.00
1.00
1.00
2.00
2.00
1.18
1.23
1.18
1.23
State_Ownership
448
0.33
1.08
0.00
0.00
0.50
5.50
89.65
0.57
0.34
0.57
0.34
Ownership_Concentration 448
28.06
16.18
2.31
22.37
24.01
30.67
93.55
24.46
23.12
24.46
23.12
Debt
448
114.16
123.67
0.00
46.73
90.10
154.47
954.56
89.70
117.17
89.70
117.17
Dividend
448 9.94E+05 1.49E+06 0.00E+00 2.36E+05 5.59E+05 1.22E+06 9.49E+06 3.89E+05 1.01E+06
3.89E+05
1.01E+06
Size
448 1.20E+07 1.50E+07 3.70E+04 2.59E+06 7.06E+06 1.67E+07 1.21E+08 9.24E+06 1.65E+07
9.24E+06
1.65E+07
Age
448
73.01
44.42
2.00
30.50
82.00
104.00
338.00
67.30
70.08
67.30
70.08
Investment
448
0.55
0.39
0.10
0.33
0.46
0.66
4.44
0.64
0.58
0.64
0.58
Growth
448
0.21
0.41
0.00
0.00
0.00
1.00
1.00
0.11
0.02
0.11
0.02
Intangibles
448
0.33
0.25
0.00
0.12
0.26
0.51
0.95
0.37
0.29
0.37
0.29
1 We define Large_CSP as equals to 1, when a firm’s CSP has a value larger than the mean of the sector for the corresponding year.
T-test
0.00
0.00
0.00
0.00
0.09
0.31
0.00
0.00
0.00
0.00
0.62
1.00
0.00
0.00
0.31
0.10
0.14
0.52
0.01
0.00
0.00
0.00
0.10
0.00
TABLE 2: Correlation Matrix
1
1 CSP
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
2 Workers
0.71***
3 ROA
-0.09*** -0.08**
22
23
24
1.00
1.00
4 Tobin's q
0.04
0.06* -0.07**
5 Abnormal returns
0.00
0.05** 0.10*** -0.06**
6 Entrenchment
0.07* 0.11*** -0.10*** -0.07* -0.12*** 1.00
7 Anti-takeover
0.06* 0.12*** -0.06* -0.10*** -0.13*** 0.75***
1.00
8 OneShare_OneVote
0.07
*
9 Shareholders_Rights
-0.03
10 Managown_Entrench
-0.04 -0.06** 0.05*
11 Manager_Tenure
0.11*** 0.15*** 0.13*** 0.10*
12 Control_Committee
-0.04
-0.06
0.04
13 Non-dual_CEO
-0.10*
-0.12*
0.18***
14 Board_Independence
-0.06*
-0.01
0.02
15 Performance_Evaluation
-0.01
-0.04
0.05
0.05
-0.06
0.05
1.00
-0.03 -0.14*** -0.02 0.33***
0.01
0.06*
1.00
-0.01
-0.03
0.01
0.10
***
1.00
-0.01
0.04 0.04*
17 Ownership_Concentration -0.05* -0.04*
18 Dividend
0.32*** 0.30***
19 Growth
-0.15*** -0.14***
*
1.00
***
16 State_Ownership
0.01
-0.05** -0.02
1.00
-0.06* -0.17*** -0.72*** -0.17*** -0.08 0.10*** 1.00
0.14*** 0.13*** -0.53*** -0.70*** -0.03
0.00 0.06*
0.00
1.00
-0.06
0.02
-0.08
0.02
-0.07** 0.08***
-0.03
-0.10*** -0.02
-0.03
0.01
0.37***
-0.16***
0.02
-0.05 0.61
0.00
0.00
-0.02
0.02 -0.01
0.02
-0.03
0.05
0.04 -0.04** -0.22*** 0.14*** 0.10**
0.07** -0.38*** -0.50*** -0.07** 0.07** 0.04 -0.12** 0.65***
0.02
0.05*
0.02
0.01 -0.02
0.00
0.10***
-0.05
0.05
-0.09 0.04
0.04
-0.02 0.06
0.00
0.12
***
0.17
***
**
0.02
-0.05 0.04
-0.02 -0.01
0.05
*
-0.07** -0.01
-0.06
-0.21
***
1.00
0.02 0.17*** 1.00
0.17*** 0.14*** 0.13*** -0.02 -0.01 -0.13*** -0.16*** -0.07
0.05*** 0.06*
***
1.00
0.09
-0.01 -0.10*** 1.00
0.00
-0.04 0.14*** 1.00
0.03
0.02
0.05
-0.01 -0.12
***
0.07* -0.07*
1.00
0.15*** 0.13*** 0.02
1.00
20 Size
0.34*** 0.34*** -0.22*** -0.03
-0.01
0.04
-0.01
0.05
-0.05
-0.10
-0.05
0.04
0.04
0.07* -0.02 0.62*** -0.04
1.00
21 Age
0.07*** 0.07***
-0.01
0.04
0.00
0.08** 0.01 -0.04** 0.00
0.02
0.02
-0.01
0.00
0.02 0.05*** 0.10*** 0.00
0.04
22 Debt
0.14*** 0.18*** -0.33*** 0.29*** -0.01
0.10
0.12***
0.10
0.04 -0.10** -0.16*** 0.01 -0.20*** 0.03
23 Investment
-0.03
-0.01
0.12*
0.05 -0.02
-0.06
0.11* 0.19*** 0.03
0.02
0.04
0.05
0.03
0.08
24 Intangibles
21
1.00
0.04
-0.04 -0.07**
0.00
0.04
-0.27*** -0.19*** -0.03 0.12***
-0.09** 0.14**
0.05
0.07
0.05
***p-value 0.01, ** p-value 0.05, *p-value 0.10.
37
1.00
-0.01 0.14*** 0.01 0.12*** -0.06* 0.46*** 0.07*
1.00
-0.10
-0.02
0.03 -0.21*** 0.03 -0.23*** -0.04 -0.24*** 1.00
-0.04 -0.08** 0.05
0.00
0.04
-0.06 0.09** -0.08** -0.14*** 0.03 -0.05 1.00
TABLE 3: The impact of Entrenchment on Corporate Social Performance (CSP)
Table 3 reports the results of conducting fixed-effect estimation on CSP. The dependent variable is led by one period. The variable
CSP is the score provided by SiRi for non-shareholder stakeholders’ degree of satisfaction once we have excluded the component of
disclosure. Anti-takeover is a dummy that it is equal to 1 if the firm has implemented any anti-takeover measures.
OneShare_OneVote is a dummy that it is equal to 1 if the company has multiple classes of stock with different voting rights.
Shareholders_Rights is a variable that takes three values (0, 0.5, 1) depending on the degree of limitation of shareholders’ voting
rights (see text for details). Managown_Entrench is a dummy that is equal to 1 when managerial stake has a value between 17% and
69.8%. In this region there is a negative relationship between managerial stake and Tobin’s q (see text for details). Manager_Tenure
is defined as a dummy that takes the value of 1 if the directors’ (including managers) term in office is more than three years old.
Then, Entrenchment=Anti-takeover + OneShare_OneVote + Shareholders_Rights + Managown_Entrench + Manager_Tenure.
Control_committee is the sum of the following three variables: Nomination_Committee is a dummy that it is equal to 1 if there is a
nomination committee with independent members and zero otherwise. Remuneration_Committee is a dummy that it is equal to 1 if
there is a remuneration committee with independent members and zero otherwise. Audit_Committee is a dummy that it is equal to 1 if
there is an audit committee with independent members and zero otherwise. Non-dual_CEO is a dummy that it is equal to 1 when the
chairman is not the CEO and zero otherwise. Board_Independence is a variable that takes three different values (0, 0.5 and 1)
contingent on the percentage of independent directors with respect to the mean value for the sector. Performance_Evaluation takes
three different values (0, 1, 2) depending on the degree of development of the performance evaluation system (see text for details).
State_Ownership is the stake in the hands of the state. Ownership_Concentration is the stake held by the three largest blockholders.
The crossed variable DControl_Committee*DEntrenchment is the result of multiplying DEntrenchment (that is equal to
Entrenchment when this variable is larger than the mean for the sector in the corresponding year; and zero otherwise), times a
dummy DControl_Committee that is equal to 1 if Control_Committee is larger than the mean for the sector in the corresponding year
and
zero
otherwise.
Following
the
same
pattern,
we
define
DNon-dual_CEO*DEntrenchment;
DBoard_Independence*DEntrenchment; DPerformance_Evaluation* DEntrenchment; DState_Ownership* DEntrenchment;
DOwnership_Concentration *DEntrenchment. Debt is the ratio of non-current liabilities plus loans to shareholders’ funds. Dividend
is the sector average pay-out ratio times the firm’s after-tax profits. Size is the fixed-asset value; Age is the number of years of the
company’s existence, ROA is the EBITDA to the total assets. Investment is the ratio of fixed assets to total assets. Growth is equal to
1 when the rate of increase in sales is larger than the value for the corresponding sector and year, and 0 otherwise. Intangibles is the
ratio of intangible assets to total fixed assets. All the variables are normalized.
CSP (t+1)
CSP (t+1)
CSP (t+1)
Dependent variable
1.977** (2.23)
Anti-takeover
1.792 (1.38)
OneShare_OneVote
-0.136 (-0.2)
Shareholders_Rights
0.894*** (3.32)
Managown_Entrench
2.168** (2.56)
Manager_Tenure
1.566*** (2.44)
2.026* (1.77)
Entrenchment
-2.304* (-1.75)
-2.596** (-2.04)
-1.927* (-1.64)
Control_Committee
0.148 (0.13)
1.502 (0.99)
3.525 (1.38)
Non-dual_CEO
-4.93** (-2.28)
-4.522* (-1.73)
-6.623*** (-3.49)
Board_Independence
-1.13 (-1.41)
-0.893 (-1.1)
-0.149 (-0.19)
Performance_Evaluation
0.489** (2.51)
0.536*** (2.66)
0.332 (1.4)
State_Ownership
0.088 (0.23)
0.478 (1.31)
0.491 (1.24)
Ownership_Concentration
2.399*** (3.93)
DControl_Committee*DEntrenchment
0.988** (2.21)
DNon-dual_CEO*DEntrenchment
-0.919 (-1.41)
DBoard_Independence*DEntrenchment
0.652* (1.85)
DPerformance_Evaluation*DEntrenchment
0.315 (0.77)
DState_Ownership*DEntrenchment
-0.125 (-0.41)
DOwnership_Concentration*DEntrenchment
0.952 (0.59)
-0.518 (-0.26)
1.801 (1.03)
Debt
-4.54* (-1.8)
-5.229** (-1.92)
-4.631* (-1.69)
Dividend
-5.912 (-0.7)
-4.297 (-0.57)
-1.813 (-0.27)
Size
190.98*** (4.24)
199.356*** (4.25)
169.256*** (3.81)
Age
1.522 (1.59)
1.283 (1.15)
1.849* (1.68)
ROA
-2.012 (-0.45)
-3.315 (-0.74)
-3.926 (-0.84)
Investment
-0.498 (-1.64)
-0.35 (-1.13)
-0.466 (-1.62)
Growth
1.307 (0.44)
1.305 (0.46)
0.523 (0.19)
Intangibles
12.705* (1.96)
11.381* (1.67)
14.588** (2.21)
Intercept
48.49
40.75
49.64
R2
7.93***
6.68***
19.88***
Fitness of the model (F test)
57.87***
47.11***
53.57***
Hausman Test (χ -squared)
Effects
Fixed-effects
Fixed-effects
Fixed-effects
448
448
448
Number of observations
***p-value 0.01, ** p-value 0.05, *p-value 0.10. In parentheses the t-statistic.
TABLE 4: The Impact of Entrenchment on Workers’ Satisfaction
Table 4 reports the results of conducting fixed-effect estimation on workers’ satisfaction. The dependent variable is lead by one period.
The variable Workers is the score provided by SiRi of the degree of workers’ satisfaction once we have excluded the component of
disclosure. Anti-takeover is a dummy that it is equal to 1 if the firm has implemented any anti-takeover measures. OneShare_OneVote is
a dummy that it is equal to 1 if the company has multiple classes of stock with different voting rights. Shareholders_Rights is a variable
that takes three values (0, 0.5, 1) depending on the degree of limitation of shareholders’ voting rights (see text for details).
Managown_Entrench is a dummy that is equal to 1 when managerial stake has a value between 17% and 69.8%. In this region there is a
negative relationship between managerial stake and Tobin’s q (see text for details). Manager_Tenure is defined as a dummy that takes
the value of 1 if the directors’ (including managers) term in office is more than three years old. Then, Entrenchment=Anti-takeover +
OneShare_OneVote + Shareholders_Rights + Managown_Entrench + Manager_Tenure. Control_committee is the sum of the following
three variables: Nomination_Committee is a dummy that it is equal to 1 if there is a nomination committee with independent members
and zero otherwise. Remuneration_Committee is a dummy that it is equal to 1 if there is a remuneration committee with independent
members and zero otherwise. Audit_Committee is a dummy that it is equal to 1 if there is an audit committee with independent members
and zero otherwise. Non-dual_CEO is a dummy that it is equal to 1 when the chairman is not the CEO and zero otherwise.
Board_Independence is a variable that takes three different values (0, 0.5 and 1) contingent on the percentage of independent directors
with respect to the mean value for the sector. Performance_Evaluation takes three different values (0, 1, 2) depending on the degree of
development of the performance evaluation system (see text for details). State_Ownership is the stake in the hands of the state.
Ownership_Concentration is the stake held by the three largest blockholders. The crossed variable DControl_Committee*DEntrenchment
is the result of multiplying DEntrenchment (that is equal to Entrenchment when this variable is larger than the mean for the sector in the
corresponding year; and zero otherwise), times a dummy DControl_Committee that is equal to 1 if Control_Committee is larger than the
mean for the sector in the corresponding year and zero otherwise. Following the same pattern, we define DNondual_CEO*DEntrenchment; DBoard_Independence*DEntrenchment; DPerformance_Evaluation* DEntrenchment; DState_Ownership*
DEntrenchment; DOwnership_Concentration *DEntrenchment. Debt is the ratio of non-current liabilities plus loans to shareholders’
funds. Dividend is the sector average pay-out ratio times the firm’s after-tax profits. Size is the fixed-asset value; Age is the number of
years of the company’s existence, ROA is the EBITDA to the total assets. Investment is the ratio of fixed assets to total assets. Growth is
equal to 1 when the rate of increase in sales is larger than the value for the corresponding sector and year, and 0 otherwise. Intangibles is
the ratio of intangible assets to total fixed assets. All the variables are normalized.
Dependent Variable
Workers (t+1)
Workers (t+1)
Workers (t+1)
Anti-takeover
3.241** (1.98)
OneShare_OneVote
2.515 (1.32)
Shareholders_Rights
-0.394 (-0.36)
Managown_Entrench
1.068** (2.11)
Manager_Tenure
3.11** (2.15)
Entrenchment
2.323** (2.17)
2.695* (1.89)
Control_Committee
-5.593*** (-2.56)
-6.156*** (-2.92)
-4.388** (-2.23)
Non-dual_CEO
2.496 (1.13)
4.375** (2.05)
7.304*** (3.4)
Board_Independence
-1.333 (-0.25)
-0.66 (-0.11)
-6.844** (-2.03)
Performance_Evaluation
-2.298** (-1.94)
-2.001* (-1.71)
-9.417* (-1.8)
State_Ownership
1.82*** (3.47)
1.902*** (3.48)
1.769*** (4.05)
Ownership_Concentration
0.489 (0.76)
0.999*1.73)
1.157** (2.08)
DControl_Committee*DEntrenchment
6.466*** (6.45)
DNon-dual_CEO*DEntrenchment
0.599 (0.86)
DBoard_Independence*DEntrenchment
-0.57 (-0.59)
DPerformance_Evaluation*DEntrenchment
1.252* (1.7)
DState_Ownership*DEntrenchment
1.081* (1.69)
DOwnership_Concentration*DEntrenchment
-0.348 (-0.62)
Debt
0.816 (0.25)
-1.259 (-0.32)
5.124 (1.48)
Dividend
-1.037 (-0.27)
-1.996 (-0.52)
3.776 (0.96)
Size
-14.347 (-0.9)
-11.153 (-0.77)
-12.462 (-1.03)
Age
282.167*** (3.13)
285.711*** (3.4)
233.889*** (2.9)
ROA
0.055 (0.02)
-0.327 (-0.14)
-2.029 (-0.61)
Investment
5.724 (0.77)
3.027 (0.39)
-2.588 (-0.31)
Growth
-0.428 (-0.96)
-0.238 (-0.55)
-0.596 (-1.39)
Intangibles
4.881 (1.5)
4.959 (1.62)
4.652 (1.23)
Intercept
8.229 (0.62)
7.596 (0.61)
23.482** (2.37)
R2
45.77
40.29
51.91
Fitness of the model (F test)
3.15***
5.38***
18.93***
Hausman Test (χ -squared)
46.17***
39.99***
51.36***
Effects
Fixed-effects
Fixed-effects
Fixed-effects
Number of observations
448
448
448
***p-value 0.01, ** p-value 0.05, *p-value 0.10. In parentheses the t-statistic.
39
TABLE 5A. The Role of CSP connecting Entrenchment and Performance (ROA)
Table 5A reports the results of conducting fixed-effect estimations of entrenchment and CSP as well as their interaction on a firm’s financial performance
proxied by ROA. The dependent variable is lead by one period in order to prevent endogeneity problems. ROA is defined as the ration of earnings before
interests and taxes to total assets. Also, we detract from this variable the mean for the corresponding year and sector for evaluating the performance in relative
terms. The variable CSP is the score provided by SiRi for non-shareholder stakeholders’ degree of satisfaction once we have excluded the
component of disclosure. The variable Workers is the score provided by SiRi of the degree of workers’ satisfaction once we have excluded the
component of disclosure. Entrenchment=Anti-takeover + OneShare_OneVote + Shareholders_Rights + Managown_Entrench + Manager_Tenure. Antitakeover is a dummy that it is equal to 1 if the firm has implemented any anti-takeover measures. OneShare_OneVote is a dummy that it is equal
to 1 if the company has multiple classes of stock with different voting rights. Shareholders_Rights is a variable that takes three values (0, 0.5, 1)
depending on the degree of limitation of shareholders’ voting rights (see text for details). Managown_Entrench is a dummy that is equal to 1
when managerial stake has a value between 17% and 69.8%. In this region there is a negative relationship between managerial stake and Tobin’s
q (see text for details). Manager_Tenure is defined as a dummy that takes the value of 1 if the directors’(including managers) term in office is
more than three years old. DEntrenchment_DCSP is the product of significant values of entrenchment (larger than the mean for the sector in the
corresponding year) times variations in CSP. DEntrenchment_DWorkers is the product of significant values of entrenchment times variations in workers’
satisfaction. Control_committee is the sum of the following three variables: Nomination_Committee is a dummy that it is equal to 1 if there is a
nomination committee with independent members and zero otherwise. Remuneration_Committee is a dummy that it is equal to 1 if there is a
remuneration committee with independent members and zero otherwise. Audit_Committee is a dummy that it is equal to 1 if there is an audit
committee with independent members and zero otherwise. Non-dual_CEO is a dummy that it is equal to 1 when the chairman is not the CEO and
zero otherwise. Board_Independence is a variable that takes three different values (0, 0.5 and 1) contingent on the percentage of independent
directors with respect to the mean value for the sector. Performance_Evaluation takes three different values (0, 1, 2) depending on the degree of
development of the performance evaluation system (see text for details). State_Ownership is the stake in the hands of the state.
Ownership_Concentration is the stake held by the three largest blockholders. Debt is the ratio of non-current liabilities plus loans to
shareholders’ funds. Dividend is the sector average pay-out ratio times the firm’s after-tax profits. Size is the fixed-asset value; Age is the number
of years of the company’s existence. Investment is the ratio of fixed assets to total assets. Growth is equal to 1 when the rate of increase in sales
is larger than the value for the corresponding sector and year, and 0 otherwise. Intangibles is the ratio of intangible assets to total fixed assets. All
the variables are normalized.
Dependent Variable
Entrenchment
CSP
Workers
DEntrenchment_DCSP
DEntrenchment_DWorkers
Control_Committee
Non-dual_CEO
Board_Independence
Performance_Evaluation
State_Ownership
Ownership_Concentration
Debt
Dividend
Size
Age
Investment
Growth
Intangibles
Intercept
R2
Fitness of the model (F test)
Hausman Test (χ -squared)
Effects
Number of observations
ROA (t+1)
ROA (t+1)
ROA (t+1)
ROA (t+1)
-0.983* (-1.83)
0.473 (0.32)
-0.935** (-1.9)
1.103 (0.89)
-0.986** (-1.92)
-0.819* (-1.69)
-0.098 (-0.02)
-0.647 (-0.11)
-0.165 (-0.16)
-0.912 (-1.14)
-2.7* (-1.61)
0.556 (1.26)
-0.284 (-1.06)
0.48 (0.99)
3.704** (2.49)
-3.561** (-1.89)
-5.081 (-0.57)
22.897 (0.49)
0.055 (0.02)
0.083 (0.56)
0.135 (0.05)
-3.235 (-0.4)
18.09
1.74**
27.35**
Fixed-effects
448
-0.479* (1.62)
-0.372 (-0.37)
-0.808 (-0.98)
-2.232 (-1.25)
0.458 (1.04)
-0.248 (-0.89)
0.481 (0.99)
3.723** (2.56)
-3.206* (-1.66)
-6.646 (-0.74)
33.695 (0.74)
-0.574 (-0.17)
0.099 (0.66)
0.183 (0.07)
-4.682 (-0.6)
19.32
1.90**
27.47**
Fixed-effects
448
-0.670* (1.70)
0.008 (0.01)
-0.792 (-1.1)
-2.798* (-1.66)
0.608 (1.37)
-0.409 (-1.53)
0.473 (1.14)
2.768** (2.42)
-2.185 (-1.33)
-14.985* (-1.64)
15.701 (0.38)
-1.644 (-0.44)
-0.029 (-0.17)
0.92 (0.41)
-7.026 (-1.17)
22.69
2.40***
34.59***
Fixed-effects
448
0.572 (0.75)
-0.706 (-1.13)
-0.57* (-1.68)
0.357 (0.92)
-0.222* (-1.82)
0.095 (0.34)
0.011 (0.04)
-3.106*** (-2.58)
-3.839 (-0.7)
-23.645 (-0.7)
2.983* (1.75)
0.107 (0.77)
-3.805 (-1.41)
2.442 (0.79)
24.07
2.38***
25.56***
Fixed-effects
448
***p-value 0.01, ** p-value 0.05, *p-value 0.10. In parentheses the t-statistic.
40
TABLE 5B The Role of CSP connecting Entrenchment and Performance (Tobin’s q)
Table 5B reports the results of conducting fixed-effect or random-effect estimations of entrenchment and CSP as well as their interaction on a firm’s financial
performance proxied by Tobin´s q. The dependent variable is lead by one period in order to prevent endogeneity problems. Tobin`s q which is defined as the
ratio of the sum of equity value plus a firm’s liabilities to the total assets taken on a log scale. Also, we detract from this variable the mean for the
corresponding year and sector for evaluating the performance in relative terms. The variable CSP is the score provided by SiRi for non-shareholder
stakeholders’ degree of satisfaction once we have excluded the component of disclosure. The variable Workers is the score provided by SiRi of
the degree of workers’ satisfaction once we have excluded the component of disclosure. Entrenchment=Anti-takeover + OneShare_OneVote +
Shareholders_Rights + Managown_Entrench + Manager_Tenure. Anti-takeover is a dummy that it is equal to 1 if the firm has implemented any antitakeover measures. OneShare_OneVote is a dummy that it is equal to 1 if the company has multiple classes of stock with different voting rights.
Shareholders_Rights is a variable that takes three values (0, 0.5, 1) depending on the degree of limitation of shareholders’ voting rights (see text
for details). Managown_Entrench is a dummy that is equal to 1 when managerial stake has a value between 17% and 69.8%. In this region there
is a negative relationship between managerial stake and Tobin’s q (see text for details). Manager_Tenure is defined as a dummy that takes the
value of 1 if the directors’(including managers) term in office is more than three years old. DEntrenchment_DCSP is the product of significant
values of entrenchment (larger than the mean for the sector in the corresponding year) times variations in CSP. DEntrenchment_DWorkers is the
product of significant values of entrenchment times variations in workers’ satisfaction. Control_committee is the sum of the following three variables:
Nomination_Committee is a dummy that it is equal to 1 if there is a nomination committee with independent members and zero otherwise.
Remuneration_Committee is a dummy that it is equal to 1 if there is a remuneration committee with independent members and zero otherwise.
Audit_Committee is a dummy that it is equal to 1 if there is an audit committee with independent members and zero otherwise. Non-dual_CEO is
a dummy that it is equal to 1 when the chairman is not the CEO and zero otherwise. Board_Independence is a variable that takes three different
values (0, 0.5 and 1) contingent on the percentage of independent directors with respect to the mean value for the sector.
Performance_Evaluation takes three different values (0, 1, 2) depending on the degree of development of the performance evaluation system
(see text for details). State_Ownership is the stake in the hands of the state. Ownership_Concentration is the stake held by the three largest
blockholders. Debt is the ratio of non-current liabilities plus loans to shareholders’ funds. Dividend is the sector average pay-out ratio times the
firm’s after-tax profits. Size is the fixed-asset value; Age is the number of years of the company’s existence. Investment is the ratio of fixed assets
to total assets. Growth is equal to 1 when the rate of increase in sales is larger than the value for the corresponding sector and year, and 0
otherwise. Intangibles is the ratio of intangible assets to total fixed assets. All the variables are normalized.
Dependent Variable
Entrenchment
CSP
Workers
DEntrenchment_DCSP
DEntrenchment_DWorkers
Control_Committee
Non-dual_CEO
Board_Independence
Performance_Evaluation
State_Ownership
Ownership_Concentration
Debt
Dividend
Size
Age
Investment
Growth
Intangibles
Intercept
R2
Fitness of the model (F test)
Fitness of the model (χ -sq.)
Hausman Test (χ -squared)
Effects
Number of observations
Tobin’s q (t+1)
Tobin’s q (t+1)
Tobin’s q (t+1)
Tobin’s q (t+1)
-0.54** (-2.01)
0.068 (0.1)
-0.194* (-1.75)
0.128 (1.11)
-0.529** (-2.01)
-0.198** (-1.80)
1.439 (0.69)
0.708* (1.62)
-0.198** (2.01)
0.844*** (2.29)
0.196 (0.34)
-0.547 (-0.58)
-0.72*** (-2.55)
-0.027 (-0.24)
-0.269*** (-2.23)
0.731** (1.97)
-0.058 (-0.05)
3.291 (1.29)
-34.929*** (-2.4)
1.538 (0.79)
-0.113 (-1.19)
0.076 (0.08)
5.697** (2.17)
28.92
3.95***
25.17***
Fixed-effects
448
0.062 (0.49)
0.555*** (5.49)
0.201* (1.74)
-0.46*** (-3.76)
0.109*** (3.55)
-0.14*** (-2.94)
0.01 (0.27)
-0.169 (-1.06)
0.084 (0.83)
-0.007 (-0.13)
0.294*** (2.63)
-0.002 (-0.05)
0.173* (1.69)
0.175 (1.36)
15.88
83.64***
15.73
Random-effects
448
0.814* (0.03)
0.212 (0.39)
-0.366 (-0.37)
-0.716** (-2.58)
-0.029 (-0.25)
-0.294** (-2.33)
0.703** (1.9)
-0.188 (-0.15)
3.304 (1.31)
-36.836** (-2.44)
1.502 (0.76)
-0.114 (-1.19)
0.088 (0.09)
5.244** (1.99)
29.29
4.14***
24.33**
Fixed-effects
448
***p-value 0.01, ** p-value 0.05, *p-value 0.10. In parentheses the t-statistic or the z-statistic (random-effects).
41
-0.299*** (-2.5)
-0.003 (-0.03)
0.558*** (5.48)
0.185 (1.58)
-0.452*** (-3.74)
0.107*** (3.40)
-0.141*** (-3.00)
0.010 (0.27)
-0.210 (-1.29)
0.085 (0.83)
0.000 (0.01)
0.281** (2.57)
0.000 (-0.01)
0.159 (1.63)
-0.228 (-0.98)
17.52
93.12***
15.57
Random-effects
448
TABLE 5C The Role of CSP connecting Entrenchment and Performance (Abnormal returns)
Table 5C reports the results of conducting random-effect estimations of entrenchment and CSP as well as their interaction on a firm’s financial performance
proxied by the abnormal returns. These are computed through a single factor model, using as factor the MSCI world index in US dollars (see text for details). The
dependent variable is lead by one period in order to prevent endogeneity problems. Also, we detract from this variable the mean for the corresponding year and
sector for evaluating the performance in relative terms. The variable CSP is the score provided by SiRi for non-shareholder stakeholders’ degree of
satisfaction once we have excluded the component of disclosure. The variable Workers is the score provided by SiRi of the degree of workers’
satisfaction once we have excluded the component of disclosure. Entrenchment=Anti-takeover + OneShare_OneVote + Shareholders_Rights +
Managown_Entrench + Manager_Tenure. Anti-takeover is a dummy that it is equal to 1 if the firm has implemented any anti-takeover measures.
OneShare_OneVote is a dummy that it is equal to 1 if the company has multiple classes of stock with different voting rights. Shareholders_Rights is
a variable that takes three values (0, 0.5, 1) depending on the degree of limitation of shareholders’ voting rights. Managown_Entrench is a dummy
that is equal to 1 when managerial stake has a value between 17% and 69.8%. In this region there is a negative relationship between managerial stake
and Tobin’s q (see text for details). Manager_Tenure is defined as a dummy that takes the value of 1 if the directors’(including managers) term in
office is more than three years old. DEntrenchment_DCSP is the product of significant values of entrenchment (larger than the mean for the sector in the
corresponding year) times variations in CSP. DEntrenchment_DWorkers is the product of significant values of entrenchment times variations in workers’
satisfaction. Control_committee is the sum of the following three variables: Nomination_Committee is a dummy that it is equal to 1 if there is a
nomination committee with independent members and zero otherwise. Remuneration_Committee is a dummy that it is equal to 1 if there is a
remuneration committee with independent members and zero otherwise. Audit_Committee is a dummy that it is equal to 1 if there is an audit
committee with independent members and zero otherwise. Non-dual_CEO is a dummy that it is equal to 1 when the chairman is not the CEO and
zero otherwise. Board_Independence is a variable that takes three different values (0, 0.5 and 1) contingent on the percentage of independent
directors with respect to the mean value for the sector. Performance_Evaluation takes three different values (0, 1, 2) depending on the degree of
development of the performance evaluation system. State_Ownership is the stake in the hands of the state. Ownership_Concentration is the stake
held by the three largest blockholders. Debt is the ratio of non-current liabilities plus loans to shareholders’ funds. Dividend is the sector average
pay-out ratio times the firm’s after-tax profits. Size is the fixed-asset value; Age is the number of years of the company’s existence. Investment is the
ratio of fixed assets to total assets. Growth is equal to 1 when the rate of increase in sales is larger than the value for the corresponding sector and
year, and 0 otherwise. Intangibles is the ratio of intangible assets to total fixed assets. All the variables are normalized.
Dependent Variable
Abn. Returns (t+1)
Abn. Returns (t+1)
Abn. Returns (t+1)
Abn. Returns (t+1)
Entrenchment
CSP
Workers
DEntrenchment_DCSP
DEntrenchment_DWorkers
Control_Committee
Non-dual_CEO
Board_Independence
Performance_Evaluation
State_Ownership
Ownership_Concentration
-0.133** (-2.11)
0.155** (2.17)
-0.118** (-1.97)
0.044 (0.54)
-0.128** (-2.04)
-0.113** (-1.80)
0.606** (2.32)
0.398 (1.48)
-0.12* (-1.74)
0.041 (0.68)
-0.08 (-1.28)
0.044 (0.72)
0.013 (0.61)
-0.012 (-0.4)
-0.08 (-1.44)
-0.235** (-2.08)
0.073 (0.62)
-0.028 (-0.99)
-0.092 (-1.15)
0.005 (0.16)
-0.081 (-1.26)
-0.288* (-1.65)
4.77
29.89**
13.27
Random-effects
448
-0.139*** (-2.77)
-0.12* (-1.74)
0.068 (0.9)
-0.093 (-1.47)
0.048 (0.79)
0.016 (0.79)
-0.01 (-0.33)
-0.074 (-1.31)
-0.245** (-2.09)
0.075 (0.63)
-0.029 (-1.03)
-0.105 (-1.3)
0.009 (0.3)
-0.089 (-1.39)
-0.193 (-1.04)
5.12
34.44***
14.96
Random-effects
448
Debt
Dividend
Size
Age
Investment
Growth
Intangibles
Intercept
R2
Fitness of the model (χ -sq.)
Hausman Test (χ -squared)
Effects
Number of observations
-0.132** (-2.59)
-0.111 (-1.6)
0.044 (0.74)
-0.075 (-1.2)
0.041 (0.67)
0.015 (0.75)
-0.011 (-0.37)
-0.08 (-1.41)
-0.219** (-1.88)
0.071 (0.6)
-0.03 (-1.05)
-0.092 (-1.14)
0.005 (0.19)
-0.064 (-1.01)
0.056 (0.77)
4.87
27.05**
13.44
Random-effects
448
-0.097 (-1.33)
0.058 (0.77)
-0.106* (-1.62)
0.054 (0.7)
0.016 (0.67)
-0.012 (-0.41)
-0.068 (-1.19)
-0.212* (-1.74)
0.107 (0.74)
-0.034 (-1.17)
-0.123 (-1.34)
0.009 (0.29)
-0.074 (-1.04)
-0.103*** (-0.2)
9.06
37.69*
18.04
Random-effects
448
***p-value 0.01, ** p-value 0.05, *p-value 0.10. In parentheses the t-statistic or the z-statistic (random-effects).
42
TABLE 6. Income Smoothing as proxy for entrenchment
Table 6 reports the results of conducting fixed-effect estimations for income smoothing-based earnings management, as well as
for different governance mechanisms, on a firm’s CSP (led by one period).. We use two measures for income smoothing.
Incsmooth1 is the correlation between changes in accruals and changes in cash flow (columns 1 and 2). Incsmooth2 is the ratio of
net income before extraordinary items to the standard deviation of cash-flows (columns 3 and 4). Control_committee is the sum
of the following three variables: Nomination_Committee is a dummy that it is equal to 1 if there is a nomination committee with
independent members and zero otherwise. Remuneration_Committee is a dummy that it is equal to 1 if there is a remuneration
committee with independent members and zero otherwise. Audit_Committee is a dummy that it is equal to 1 if there is an audit
committee with independent members and zero otherwise. Non-dual_CEO is a dummy that it is equal to 1 when the chairman is
not the CEO and zero otherwise. Board_Independence is a variable that takes three different values (0, 0.5 and 1) contingent on
the percentage of independent directors with respect to the mean value for the sector. Performance_Evaluation takes three
different values (0, 1, 2) depending on the degree of development of the performance evaluation system (see text for details).
State_Ownership is the stake in the hands of the state. Ownership_Concentration is the stake held by the three largest
blockholders. Debt is the ratio of non-current liabilities plus loans to shareholders’ funds. Dividend is the sector average pay-out
ratio times the firm’s after-tax profits. Size is the fixed-asset value; Age is the number of years of the company’s existence, ROA
is the EBITDA to the total assets. Investment is the ratio of fixed assets to total assets. Growth is equal to 1 when the rate of
increase in sales is larger than the value for the corresponding sector and year, and 0 otherwise. Intangibles is the ratio of
intangible assets to total fixed assets. All the variables are normalized.
Dependent Variable
CSP (t+1)
Workers (t+1)
CSP (t+1)
Workers (t+1)
Incsmooth1
44.547*** (3.19)
49.085* (1.67)
Incsmooth2
0.472*** (2.26)
0.857** (1.89)
Control_Committee
-2.175* (-1.68)
-5.74*** (-2.71)
-1.299 (-0.78)
-3.404 (-1.32)
Non-dual_CEO
0.442 (0.31)
3.131 (1.43)
1.41 (0.95)
3.656* (1.66)
Board_Independence
-3.578 (-1.23)
0.55 (0.08)
-2.674 (-0.72)
2.469 (0.3)
Performance_Evaluation
-1.036 (-1.34)
-2.074* (-1.9)
-1.621 (-1.5)
-3.908*** (-2.78)
State_Ownership
0.515** (1.88)
1.864*** (2.83)
0.533** (2.09)
2.039*** (3.28)
Ownership_Concentration
0.372 (0.95)
0.86 (1.39)
0.407 (0.83)
0.912 (1.4)
Debt
Dividend
Size
Age
ROA
Investment
Growth
Intangibles
Intercept
R2
Fitness of the model (F test)
Hausman Test (χ -squared)
Effects
Number of observations
-1.04 (-0.66)
-7.881*** (-2.75)
-6.244 (-0.96)
174.639*** (3.49)
1.654 (1.4)
-3.937 (-0.91)
-0.219 (-0.72)
0.489 (0.17)
13.731* (1.77)
42.01
5.95***
53.25***
Fixed-effects
448
-1.776 (-0.5)
-5.619 (-1.37)
-13.741 (-0.99)
257.078*** (2.9)
0.28 (0.11)
2.137 (0.25)
-0.076 (-0.18)
3.876 (1.22)
9.837 (0.71)
39.56
4.12***
44.99***
Fixed-effects
448
***p-value 0.01, ** p-value 0.05, *p-value 0.10. In parentheses the t-statistic.
43
1.019 (0.5)
-11.708** (-2.57)
-2.829 (-0.36)
187.173*** (3.29)
2.669** (1.85)
-3.839 (-0.74)
-0.264 (-0.71)
0.57 (0.18)
-8.426 (-0.56)
37.69
12.66***
43.17***
Fixed-effects
448
1.406 (0.41)
-13.171** (-2.04)
-8.932 (-0.58)
274.701*** (2.71)
2.036 (0.7)
2.097 (0.22)
-0.217 (-0.44)
3.812 (1.13)
-22.377 (-0.83)
40.63
5.38***
41.32***
Fixed-effects
448
TABLE 7. The Expropriating Effect
Table 7 reports the results of fixed-effect estimations of entrenchment, as well as different governance mechanisms, on a firm’s CSP (led by one period).
Column 1 (2) compare those firms such that the stake of the three largest blockholders is larger (lower) than the mean of the sector for the corresponding
year –concentration = 1 (0)- Columns 3 and 4 compare family firms (when the largest blockholder belongs to a family ) versus non-family ones. Finally,
columns 5 and 6 compare firms where workers have substantial power with their counterparts. We define firms with workers’ power as those where
workers have board level responsibilities and/or they enjoy cash profit sharing programs. The variable CSP is the score provided by SiRi for nonshareholder stakeholders’ degree of satisfaction once we have excluded the component of disclosure. Anti-takeover is a dummy that it is equal to 1 if the
firm has implemented any anti-takeover measures. OneShare_OneVote is a dummy that it is equal to 1 if the company has multiple classes of stock with
different voting rights. Shareholders_Rights is a variable that takes three values (0, 0.5, 1) depending on the degree of limitation of shareholders’ voting
rights (see text for details). Managown_Entrench is a dummy that is equal to 1 when managerial stake has a value between 17% and 69.8%. In this
region there is a negative relationship between managerial stake and Tobin’s q (see text for details). Manager_Tenure is defined as a dummy that takes
the value of 1 if the directors’ (including managers) term in office is more than three years old. Then, Entrenchment=Anti-takeover +
OneShare_OneVote + Shareholders_Rights + Managown_Entrench + Manager_Tenure. Control_committee is the sum of the following three variables:
Nomination_Committee is a dummy that it is equal to 1 if there is a nomination committee with independent members and zero otherwise.
Remuneration_Committee is a dummy that it is equal to 1 if there is a remuneration committee with independent members and zero otherwise.
Audit_Committee is a dummy that it is equal to 1 if there is an audit committee with independent members and zero otherwise. Non-dual_CEO is a
dummy that it is equal to 1 when the chairman is not the CEO and zero otherwise. Board_Independence is a variable that takes three different values (0,
0.5 and 1) contingent on the percentage of independent directors with respect to the mean value for the sector. Performance_Evaluation takes three
different values (0, 1, 2) depending on the degree of development of the performance evaluation system (see text for details). State_Ownership is the
stake in the hands of the state. Ownership_Concentration is the stake held by the three largest blockholders. Debt is the ratio of non-current liabilities
plus loans to shareholders’ funds. Dividend is the sector average pay-out ratio times the firm’s after-tax profits. Size is the fixed-asset value; Age is the
number of years of the company’s existence, ROA is the EBITDA to the total assets. Investment is the ratio of fixed assets to total assets. Growth is
equal to 1 when the rate of increase in sales is larger than the value for the corresponding sector and year, and 0 otherwise. Intangibles is the ratio of
intangible assets to total fixed assets. All the variables are normalized.
Family firms
Non-family firms
CSP (t+1)
CSP (t+1)
High workers’ Low workers’
power
power
CSP (t+1)
CSP (t+1)
1.691** (1.80)
0.17 (0.13)
2.46*** (2.82)
3.17*** (2.64)
1.32 (1.24)
-5.441*** (-2.94)
-1.903 (-0.86)
-3.71*** (-3.15)
-1.97* (-1.51)
-3.99* (-1.72)
-2.19 (-1.54)
Non-dual_CEO
8.736*** (6.46)
-1.279 (-1.06)
2.65*** (2.84)
0.70 (0.60)
-0.86 (-0.64)
4.81*** (2.26)
Board_Independence
-1.678 (-0.91)
1.559* (1.7)
-5.80*** (-7.89)
-4.98 (-1.03)
-1.87 (-1.31)
-3.04 (-1.10)
Performance_Evaluation
2.046*** (2.9)
0.421 (0.29)
0.92 (0.83)
0.02 (0.03)
0.11 (0.12)
0.06 (0.04)
State_Ownership
1.076*** (4.87)
-0.473 (-0.83)
-0.79 (-0.63)
0.51*** (2.61)
-0.72** (-1.92)
0.89*** (2.35)
Ownership_Concentration
0.836** (2.43)
-1.696 (-0.83)
0.39*** (2.53)
-0.17 (-0.35)
1.21* (1.82)
-0.38 (-0.88)
Debt
-14.876*** (-3.12)
0.914 (0.19)
-1.59 (-0.98)
0.06 (0.03)
3.05*** (2.79)
-4.63*** (-2.79)
Dividend
0.291 (0.05)
-12.101*** (-5.64)
-12.32*** (-2.68)
-2.05 (-0.69)
-0.75 (-0.50)
-6.13 (-0.35)
Size
43.058*** (4.82)
8.65* (1.72)
-96.86*** (-2.86)
-1.84 (-0.28)
-4.19 (-0.45)
-11.32 (-0.34)
Age
188.737*** (2.98)
111.591** (1.89)
15.20*** (5.57)
6.95*** (2.91)
9.46** (1.93)
7.39*** (3.02)
ROA
-29.151 (-1.59)
9.757*** (2.93)
-2.23*** (-3.09)
0.98 (0.50)
1.74 (0.62)
0.59 (0.41)
Investment
-43.168** (-2.15)
0.294 (0.03)
-15.86* (-1.69)
0.90 (0.18)
-1.38 (-0.37)
-7.80 (-0.90)
Growth
0.274 (1.1)
-1.006** (-2.33)
-0.16 (-0.39)
-0.37 (-1.20)
-0.87*** (-2.37)
-0.24 (-0.59)
Intangibles
16.746*** (2.66)
-2.784 (-0.46)
3.39 (1.39)
4.91 (1.47)
5.58 (1.44)
6.25** (2.11)
Intercept
-21.49 (-1.3)
33.58*** (7.46)
-1036.14*** (-5.50)
-466.08*** (-2.66)
-607.14** (-1.78)
-522.46*** (-2.78)
R2
92.45
48.44***
62.71***
47.00
6.76***
35.33***
96..20
227.00 ***
68.47***
35.77
4.47 ***
21.91*
43.05
28.25***
25..90***
64.15
8.86***
36.62***
Fixed-effects
Fixed-effects
Fixed-effects
Fixed-effects
Fixed-effects
Fixed-effects
75
373
200
248
Dependent variable
Large ownership
concentration
CSP (t+1)
Low ownership
concentration
CSP (t+1)
Entrenchment
0.818 (0.53)
Control_Committee
Fitness of the model (F test)
Hausman Test (χ-squared)
Effects
Number of observations
173
275
***p-value 0.01, ** p-value 0.05, *p-value 0.10. In parentheses the t-statistic.
44
TABLE 8: ROBUSTNESS
Table 8 reports the results of conducting fixed-effect estimations on a firm’s CSP lead by one period. Columns 1 (2) focus on those firms such that their
leverage is larger (lower) than the mean for the corresponding sector. In columns 3 and 4 the sample is separated in terms of the legal origin of the country
codes (La Porta et al., 1998). In column 3, we focus on those countries with English-origin legal codes, while in column 4, we focus on those countries
with French, German or Scandinavian-origin legal codes. Finally, columns 5 and 6 compare large firms (size larger than the mean of the sector for the
corresponding year) with small ones. The variable CSP is the score provided by SiRi for non-shareholder stakeholders’ degree of satisfaction once we have
excluded the component of disclosure. Anti-takeover is a dummy that it is equal to 1 if the firm has implemented any anti-takeover measures.
OneShare_OneVote is a dummy that it is equal to 1 if the company has multiple classes of stock with different voting rights. Shareholders_Rights is a
variable that takes three values (0, 0.5, 1) depending on the degree of limitation of shareholders’ voting rights (see text for details). Managown_Entrench is
a dummy that is equal to 1 when managerial stake has a value between 17% and 69.8%. In this region there is a negative relationship between managerial
stake and Tobin’s q (see text for details). Manager_Tenure is defined as a dummy that takes the value of 1 if the directors’ (including managers) term in
office is more than three years old. Then, Entrenchment=Anti-takeover + OneShare_OneVote + Shareholders_Rights + Managown_Entrench +
Manager_Tenure. Control_committee is the sum of the following three variables: Nomination_Committee is a dummy that it is equal to 1 if there is a
nomination committee with independent members and zero otherwise. Remuneration_Committee is a dummy that it is equal to 1 if there is a remuneration
committee with independent members and zero otherwise. Audit_Committee is a dummy that it is equal to 1 if there is an audit committee with
independent members and zero otherwise. Non-dual_CEO is a dummy that it is equal to 1 when the chairman is not the CEO and zero otherwise.
Board_Independence is a variable that takes three different values (0, 0.5 and 1) contingent on the percentage of independent directors with respect to the
mean value for the sector. Performance_Evaluation takes three different values (0, 1, 2) depending on the degree of development of the performance
evaluation system (see text for details). State_Ownership is the stake in the hands of the state. Ownership_Concentration is the stake held by the three
largest blockholders. Debt is the ratio of non-current liabilities plus loans to shareholders’ funds. Dividend is the sector average pay-out ratio times the
firm’s after-tax profits. Size is the fixed-asset value; Age is the number of years of the company’s existence, ROA is the EBITDA to the total assets.
Investment is the ratio of fixed assets to total assets. Growth is equal to 1 when the rate of increase in sales is larger than the value for the corresponding
sector and year, and 0 otherwise. Intangibles is the ratio of intangible assets to total fixed assets. All the variables are normalized.
Leverage=1
Leverage=0
Anglosaxon
Non-Anglosaxon Large size
Small size
Dependent variable
Entrenchment
Control_Committee
Non-dual_CEO
Board_Independence
Performance_Evaluation
State_Ownership
Ownership_Concentration
Debt
Dividend
Size
Age
ROA
Investment
Growth
Intangibles
Intercept
CSP (t+1)
CSP (t+1)
CSP (t+1)
CSP (t+1)
CSP (t+1)
CSP (t+1)
2.45** (2.08)
-1.133 (-0.51)
3.624 (1.24)
-4.968 (-1.38)
-3.446*** (-2.42)
0.384 (0.96)
-0.243 (-0.36)
-4.113* (-1.66)
-22.468*** (-3.25)
-14.564 (-0.64)
150.147* (1.73)
18.804** (2.33)
0.049 (0)
-0.443 (-0.8)
-1.909 (-0.32)
-9.317 (-0.3)
-0.193 (-0.22)
-5.322*** (-5.15)
1.691 (0.88)
-2.533 (-1.5)
0.607 (0.59)
0.123 (0.48)
1.084** (1.80)
17.964*** (6.45)
-3.938 (-1.32)
-3.353 (-0.24)
284.882*** (10.84)
1.146 (0.52)
-8.726*** (-2.51)
-1.02*** (-3.26)
-1.541 (-0.66)
65.996*** (18.71)
-1.119 (-0.93)
-0.097 (-0.05)
-3.701*** (-2.28)
-7.672*** (-2.64)
-2.549* (-1.77)
0.187 (0.31)
-0.496 (-0.69)
0.206 (0.08)
-4.087 (-0.93)
-8.998 (-0.94)
208.215*** (3.43)
1.371 (0.83)
-9.851** (-1.85)
-1.032*** (-2.44)
1.418 (0.34)
86.66*** (6.04)
2.502*** (2.76)
-3.706** (-2.2)
5.972*** (3.12)
-2.404 (-1.3)
0.907 (1.08)
0.659*** (3.39)
0.457 (1.21)
-3.446 (-1.23)
-8.351*** (-2.65)
19.365** (2.22)
191.488*** (3.11)
-0.543 (-0.42)
1.065 (0.15)
-0.201 (-0.55)
0.404 (0.09)
-53.851* (-1.75)
1.526** (2.17)
2.292 (0.8)
-3.628*** (-2.39)
0.843 (0.83)
1.04 (0.99)
0.373 (1.14)
0.035 (0.04)
2.496 (1.43)
-14.408*** (-7.32)
18.772*** (4.83)
-184.445 (-1.53)
25.253*** (3.68)
9.621 (1.45)
-0.188 (-0.46)
4.239 (1.12)
67.34*** (4.02)
1.378* (1.71)
-2.4** (-2.04)
3.252* (1.69)
-5.193* (-1.71)
-0.217 (-0.22)
0.569** (2.17)
0.593 (1.43)
-2.63* (-1.76)
-7.039 (-1.64)
6.516 (1.09)
182.279*** (3.64)
1.645 (1.09)
-1.664 (-0.3)
-0.651** (-1.81)
0.167 (0.05)
16.63** (2.02)
R2
Fitness of the model (F test)
Hausman Test (χ -squared)
Effects
Number of observations
61.69
7.94***
35.70***
Fixed-effects
229
68.21
35.95***
60.08***
Fixed-effects
219
54.80
14.85***
36.33***
Fixed-effects
212
57.96
24.43***
32.92***
Fixed-effects
236
64.59
14.19***
30.12**
Fixed-effects
155
56.81
9.20***
45.93***
Fixed-effects
293
***p-value 0.01, ** p-value 0.05, *p-value 0.10. In parentheses the t-statistic.
45
Fly UP