an analysis in Europe.
Master Thesis
MSc Business Administration, International Management Track.
University of Amsterdam/Amsterdam Business School
Name: Frits Thissen
Student no.: 10283218
Submission date: 29-06-2015 (final version) Supervisor: Mr. Robert Kleinknecht MSc
2 STATEMENT OF ORIGINALITY
This document is written by Frits Thissen, who declares to take full responsibility for the
contents of this document.
I declare that the text and the work presented in this document is original and that no sources
other than those mentioned in the text and its references have been used in creating it.
The Faculty of Economics and Business is responsible solely for the supervision of
3 ABSTRACT
Time horizon is an important element in strategic decision making. An excessive focus on
short-term results may result in short-termism, which has several negative performance
implications. Executive time horizon is influenced by many factors. This thesis examines how
employee participation in corporate decision making, by means of works councils and
employee board level representation, influences the time horizon of executives at the largest
firms from 17 European countries. Content analysis of quarterly results conference calls by
executives and regression analyses revealed a limited influence of employee participation on
executive time horizons. These findings are in line with the mixed results from earlier studies
and stress the need for further research. Practitioners are advised to cooperate with works
councils or employee representatives and consider a longer time horizon, as this may prove
beneficial in the long run.
Keywords: co-determination, works councils, employee board level representation, executive time horizons, short-termism.
4 TABLE OF CONTENTS: Statement of originality 2 Abstract 3 1. Introduction 6 2. Literature review 8
2.1 Managerial time horizons and economic short-termism 8
2.2 The agency problem 11
2.3 Shareholder and stakeholder model of corporate governance 12
2.4 Stakeholder influence on management 16
2.5 Aligning shareholder and stakeholder interests: the case of 17 employees
2.6 Employee participation 18
2.6.1 Works councils 20
2.6.2 Employee board level representation 21
3. Conceptual framework 23
4. Methodology 26
4.1 Content analysis and description of material 26
4.2 Procedure 27 4.3 Operationalization of variables 29 4.4 Control variables 32 5. Results 36 5.1 Preliminary analyses 37 5.2 Correlations 38 5.3 Hypothesis testing 42 5.4 Robustness of measures 45
5.4.1 Analysis using Q&A part of conference calls 45
5.4.2 Analysis using R&D/Sales ratio 47
5.4.3 Analysis using alternative calculation of time horizon 50
6. Discussion 52
6.1 Discussion of results 52
6.2 Theoretical discussion 53
5 6.4 Limitations and suggestions for future research 58
6.5 Conclusion 61
7. References 63
6 1. Introduction
Employee representation in corporate decision making is deemed of great importance in many
European countries. In the academic literature, this process is usually referred to as
co-determination. Co-determination can take several forms, with works councils and employee
board level representation (EBLR) being the most common ones. Benefits of works councils
include the enhancement of information exchange, employee consultation and conflict
settlement (Freeman & Lazear, 1995), whereas EBLR leads to more effective decision
making because of reduced information asymmetry and greater diversity among decision
makers (Kleinknecht, 2014). However, particularly from a shareholder perspective, not all
effects of employee participation can be named as positive. Scholars say EBLR can also slow
down decision making, because different backgrounds can inhibit effective communication
(Kleinknecht, 2014). In addition, works councils can increase bargaining power of labor and
can provide opportunities for ‘rent seeking’ (Freeman and Lazear, 1995).
Since employee representation in decision making, especially in the form of EBLR,
influences the strategic planning process, it can be expected that it influences the temporal
orientation of corporate decision makers as well, since this is an important part of the strategic
planning process. As Gao (2010) states, managers are concerned with the short- or long-term
firm value when making decisions. An excessive focus on short-term firm value can lead to
economic short-termism, a phenomenon that is defined by Laverty (1996: 826) as: “decisions
and outcomes that pursue a course of action that is best for the short term, but suboptimal over the long run”. Several explanations for economic short-termism exist, like flawed
management practice, managerial opportunism, stock market myopia, fluid or impatient
capital and information asymmetry (Laverty, 1996). However, academic research has not yet
examined the influence that co-determination has on the time horizon of corporate decision
7 for the best interests of firm employees, especially when it comes to issues like job security
and employment conditions. Since these are long-term focused issues (employees will want to
know their job is secure for an extended period of time), co-determination could contribute to
corporate decision making that is long-term focused. Conversely, the absence of
co-determination might contribute to short-term focused decision making, i.e. economic or
corporate short-termism. The relation between employee representation and executive time
horizon will be the focus of this thesis, using the following research question, which will act
as a guideline throughout this thesis:
RQ1: “Does employee participation, by means of co-determination, influence executive time horizons?”
The results from this study have several implications. First, the bodies of academic
literature regarding co-determination and managerial time horizons are brought together.
Another contribution from this study stems from the fact that research into executive time
horizons is relatively scarce and performed in diverse ways. The main reason for this is the
difficulty in measuring executive time horizons. In order to quantify an abstract concept like
this, most studies (e.g. Souder & Bromiley, 2012; Souder & Shaver, 2010) have used
variables like R&D or capital expenditures as an indicator of long-term focused investments.
By using another methodology, in this case content analysis of conference call transcripts,
executive time horizons are examined from another viewpoint which could shed more light on
the factors influencing the phenomenon.
The following section will provide an overview of the academic literature regarding
the phenomena of managerial time horizons and economic short-termism. These concepts will
be discussed in detail, followed by a discussion of various forms of employee representation.
8 hypotheses regarding employee representation and their influence on executive time horizons.
Later, the used method will be explained extensively, followed by an overview of the results
from the statistical analysis performed. The thesis will end with a discussion of the findings
and a conclusion.
2. Literature review
2.1 Managerial time horizons and economic short-termism
The concept of time varies greatly among individuals and cultures and is one of the
fundamental and consequential reference points within strategic decision making
(Mosakowski & Early, 2000). Basic to the debate is the problem of intertemporal choice,
which is a characteristic of decisions in which the timing of costs and benefits are spread out
over time (Laverty, 1996; Lowenstein & Thaler, 1989). This also implies that “the course of
action that is best in the short term is not the same course of action that is best over the long run” (Laverty, 1996: 828). In other words, managers need to balance short-term results and
the long-term wellbeing of the firm. This balancing inevitably involves a tradeoff, which
often leads to choices that favor the short-term at the expense of the long term and is harmful
for the performance of the firm (Marginson & McAulay, 2008). Managerial time horizon, the
dependent variable in this thesis, is defined as the time into the future that managers typically
focus on when making decisions (DesJardine & Bansal, 2014). When a manager or executive
is disproportionately concerned with the long term, he or she has a long time horizon.
Conversely, when a manager or executive focuses disproportionately on the short term, he or
she has a short time horizon . An excessive managerial focus on the short term can lead to
short-termism (DesJardine & Bansal, 2014).
The phenomenon of (economic) short-termism, defined as “decisions and outcomes
9 (Laverty, 1996: 826), has several antecedents and consequences. An interesting psychological
explanation for the existence of short-termism is offered by DesJardine and Bansal (2014),
who ground their study in prospect theory. According to prospect theory, individuals evaluate
the outcomes of their decisions based on a certain benchmark or reference point (Kahneman
& Tversky, 1979). When the outcome of a decision exceeds the set reference point,
individuals experience a feeling of gain, whereas a feeling of loss is experienced when the
outcome of a decision does not meet the reference point (DesJardine & Bansal, 2014). As
individuals want to avoid this feeling of loss more strongly than they want to experience the
feeling of gain, they make suboptimal decisions, especially when the mentioned reference
point is based within a too narrow time frame. Another psychological explanation could be
that people, in general, favor short-term rewards over long-term rewards (Souder & Bromiley,
2012).
In order to further explain the existence of the phenomenon of short-termism, Laverty
(1996) has conducted an extensive literature review and offered a number of explanations.
First, several forms of flawed management practice contribute to short time horizons. For
instance, measuring performance over too brief of a period, before the long-term (adverse)
consequences of short-term decisions become apparent. Furthermore, managerial or executive
decisions might be optimal for the manager or executive as a person, but might not be for the
organization as a whole. For instance, managers or executives tend to make investments or
associate with projects that offer a relatively quick payback, which enhances their reputation.
Their enhanced reputation might lead to the manager or executive exiting the firm while the
short-term success is visible, but before the actual outcome of the investment or project
becomes apparent. Conversely, managers might be afraid of being dismissed before the
success of their long-term investment or project is realized (Narayanan, 1985). The focus on personal gain can also be explained by managers’ or executives’ compensation being based
10 on bonuses tied to short-term performance (Souder & Shaver, 2010). Business schools have
been assigned some blame as well, as they emphasize current balance sheet results as an
indicator of managerial effectiveness (Naranayan, 1985).
Last, stock market pressures contribute to managerial short-termism, as short-term
performance is needed to attract capital and the stock market tends to undervalue investments
that only pay off in the long run (Laverty, 1996). The latter observation can be associated with
a strong focus on shareholder value, as posed by the shareholder model of corporate
governance. This focus may lead to managers being occupied with maximizing firm value by
making investments with a short payoff horizon, thus maximizing short-term shareholder
value, whereas the firm might actually be better off with higher yielding, long-term
investments.
Short-termism, which is regarded as a result of short executive time horizons, is
associated with several performance-related outcomes. For instance, Marginson and McAulay
(2008) argue that “a preference for short-term performance leads to unintended consequences
for the long-term value adding activities of the firm” (Marginson & McAulay, 2008: 273). In
a broader context, Hayes and Abernathy (1980) found a shift among American managers
towards a view of short-term results instead of long-term innovation, and posed that this did
severely threaten the competitiveness and innovativeness of American firms. Although
research has suggested that managers and executives are forced into favoring short-term
performance over long-term company wellbeing, research has also found opposing results.
Marginson and McAulay (2008) for instance, have reported about earlier studies that found
significant positive stock market returns associated with the announcement of large R&D
11 The debate about the causes and consequences of short-termism lead to a central issue
in this thesis: should managers and executives adhere to shareholders and try to maximize
shareholder value, i.e. focus on results within a short time frame? Or should they try to
balance the interests of all stakeholders of the firm, including employees represented by
works councils or board level representatives, even if there is no short term reward? Two
models of corporate governance address this question, which will be elaborated on later in
this thesis.
2.2 The agency problem
An agency relationship is a contract under which one or more persons (the principals) engage
another person (the agent) to perform some service on their behalf, that involves delegating
some decision making authority to the agent (Jensen & Meckling, 1976). As Shleifer and
Vishny (1997) describe, the essence of the agency problem is in the separation of ownership
and control of the firm. Managers need financiers to generate capital for productive use,
whereas financiers need the managers’ human capital to generate returns on these funds. In
this case, managers act as so-called agents, in order to generate returns for their principals, the
financiers. But how can financiers be sure that their funds will be used productively and
efficiently, and will not be wasted on useless projects?
Agency theorists have argued that corporate managers are not disciplined by the
market mechanism and that they could use their control over corporate resources for their
personal gain, or at least to pursue objectives that were not those of shareholders (Lazonick & O’Sullivan, 2000). According to Shleifer and Vishny (1997), the agency problem is ideally
settled by drawing up a contract that specifies what the manager does with the funds. But, as
the future is unforeseeable, it is not feasible to develop a complete contract that covers every
12 decisions in unforeseen circumstances. It is possible for financiers to retain these rights, but it
is not uncommon that financiers do not possess the qualifications to make these decisions,
which leads to managers usually being in possession of these rights (Shleifer & Vishny,
1997). This is a reflection of the agency problem, as shareholders may fear that managers will
make improper use of their funds and as a result, may want short-term returns (Jackson &
Petraki, 2011). To align managerial decisions with shareholders’ long-term interests, several
things can be done. First, experienced managers should be appointed to make major decisions,
as managerial experience was found to be inversely related to incentives for short term results
(Naranayan, 1985). Furthermore, following the successful example in Japan, managers should
be given long-term contracts. Last, managers’ task divisions should not be changed too
frequently, as this will lead to a shorter managerial planning horizon and in turn, will increase
the propensity for short-term results (Naranayan, 1985).
Although widely recognized and used in academic literature, the agency theory has
faced criticism as well. For instance, Gospel & Pendleton (2003) agree with Aguilera and
Jackson (2003) in their statements that the principle-agent model in the agency theory
abstracts investor-management relationships from their institutional context, interests among
both investors and managers are not engaged with and other actors like labor are excluded.
Especially this influence of labor on the investor-management relationship is relevant for this
thesis. Works councils and employee board level representation are two ways in which labor
can influence the investor-management relationship, as will be explained later.
2.3 Shareholder and stakeholder model of corporate governance
As explained in the section above, the agency problem poses a large challenge upon
managers. Do they adhere to the wishes of the owners of their company? Or should they
13 Managers may very well choose to actively balance the interests of all stakeholders (e.g.
employees, customers, trade unions etc.) that are involved in the firm. In the corporate
governance literature, two well-known models regarding this choice exist; the Anglo-Saxon,
market-based model, as seen in the United States and the United Kingdom, and the
collaborative model, seen in continental Europe and Japan (Aguilera & Jackson, 2003). The
former is often referred to as the shareholder value oriented model of corporate governance.
The theoretical rationale behind this model holds that “when the corporate enterprise
maximizes shareholder value, everyone – workers, consumers, suppliers and distributors – will be better off” (Lazonick & O’Sullivan, 2000: 27). Thus, when shareholder value is
maximized, the performance of the whole economy is enhanced, not just the interests of shareholders (Lazonick & O’Sullivan, 2000). The shareholder value model as we know it
today, is characterized by dispersed firm ownership with markets for corporate control, legal
regulation and contractual incentives as the main governance mechanisms.
The shareholder value model can also be associated with the ‘market-outsider’ system
of corporate ownership, in which firms are generally owned by private or institutional
portfolio investors with little active interest in the day to day operations of the firm (Almond,
Edwards & Clark, 2003). Firms are viewed as a “bundle of assets that are deployed in order to
maximize relatively short-term earnings” (Fligstein & Brantley, 1992: 287). Furthermore,
market-outsider systems have large and active equity markets, big firms are publicly listed
and firms are able to raise significant amounts of capital from either debt or equity markets
(Gospel & Pendleton, 2003). In the shareholder value model and market outsider systems, the
focus tends to be on the maximization of short-term returns to shareholders, which has caused
a sharp decline in the timescale of corporate investments. Rather than retaining profits for
14 because of a liquid capital market in business systems like the UK and the US, there is little
attachment to the long-term fortune of the business (Almond, Edwards & Clark, 2003).
In the collaborative model, referred to as the stakeholder-oriented model, firms are
owned by larger blockholders, such as banks and families, that retain greater control and
operate in a context that is less market-oriented and has weaker managerial incentives
(Aguilera & Jackson, 2003). Stakeholders are “all individuals or groups who can substantially
affect, or be affected by, the welfare of the firm” (Freeman, 1984: 48). This includes not only
the financial claimholders, but also employees, customers, communities and government
officials (Jensen, 2001). A distinction between primary and secondary stakeholders can be
made. Primary stakeholders are groups whose support is necessary for the firm to exist, such
as financiers or employees. Secondary stakeholders do not have a formal claim on the
company, but do fall under the moral duties of the firm, like environmental or employee
satisfaction issues (Garcia-Castro, Ariño & Canela, 2011; Parmar, Freeman, Harrison, Wicks,
Purnell & De Colle, 2010).
The stakeholder model is associated with the insider-model of corporate ownership. In
this system, owners of large firms, which often consist of banks, have a long-term relationship
with the firm and links to day-to-day management are close (Almond, Edwards & Clark,
2003). The closer integration of financiers into the firm and the consideration of multiple
stakeholders (apart from just shareholders) leads to a long-term perspective by firms,
especially since profits are retained or reinvested in the firm. This, in turn, leads to longer
payout horizons (Gospel & Pendleton, 2003). Where the shareholder model has shareholder
value maximization as the ultimate criterion when making decisions, the stakeholder-oriented
model poses that managers should make decisions that take the interests of all the
stakeholders in the firm, into account. Maintaining the quality of the relationships between a
15 As Parmar et al. (2010) state, stakeholder theory has often been qualified as
non-economic, as it incorporates a social or moral dimension into the functions of the firm rather
than solely focus on running the firm effectively. However, to explain why stakeholder
management should be associated with better (financial) performance, several motivations
exist. For instance, increased trust among management and stakeholders can lower transaction
costs (Williamson, 1975) and an excellent reputation flowing from effective stakeholder
management is attractive in the marketplace to potential business partners, employees and
customers (Parmar et al, 2010; Puncheva, 2008). Furthermore, Harrison, Bosse and Phillips
(2010) have argued that stakeholder management can serve as a source of competitive
advantage, as the firm is presented with a larger number of better business opportunities and
stakeholders are more likely to reveal valuable information that can lead to enhanced
efficiency and innovation. As Parmar et al. (2010) note, a very strong economic justification
for the stakeholder model was offered by Choi and Wang (2009), who found that good
stakeholder relations not only helped firms achieve superior economic performance over a
long period of time, but also helped distressed firms to improve their performance in the short
run. Garcia-Castro et al. (2011) found some evidence that stakeholder management negatively
affected firm performance in the short run, but had a positive long-run impact.
Hill and Jones (1992) have applied the agency theory to the relationship between management and various stakeholders. The main difference compared to the ‘original’ agency
theory is that Hill and Jones’ stakeholder-agency model encompasses the implicit and explicit
contractual relationships between management and all stakeholders of the firm, instead of
focusing solely on the relationship between management and shareholders. In the
stakeholder-agency model, the manager has a unique position, as he or she is the only actor that engages
16 actor with direct control over decision making in the firm. Thus, they can be seen as the agent
of all stakeholders (Hill & Jones, 1992).
2.4 Stakeholder influence on management
Stakeholder management is a two-way process, meaning that firms can affect stakeholders,
but stakeholders can also affect firms. Effective stakeholder management is a complex task.
Over the years, the influence of (external) stakeholders on a firm’s strategy has increased
dramatically (Parmar et al., 2010). Stakeholders possess leverage over the firm because firms
are dependent on them for resources (Frooman, 1999). They can exert control over firm
management in two ways: (1) determining whether the firm gets the desired resources, and (2)
determining whether the firm gets to use these resources in the way it wants. The first strategy
is generally chosen when the power balance is in favor of the stakeholder or when the
stakeholder has a low level of dependency on the firm, whereas the second strategy is
preferred when power is more evenly divided between the two or when the stakeholder is
highly dependent on the firm (Frooman, 1999).
In order to deal with stakeholder influence effectively, management can analyze these
stakeholders and their interests. Mitchell, Agle and Wood (1997) have identified three
attributes that determine how much attention management will give to a stakeholder. These
attributes are: (1) power to influence the firm, (2) legitimacy of the stakeholder’s relationship with the firm and (3) urgency of the stakeholder’s claim on the firm. Stakeholders can possess
one, a combination of two, or all three of these attributes. “Latent stakeholders” are
stakeholders that possess only one of the three attributes. The salience level of these
stakeholders is low and management may choose to do nothing about these stakeholders as
time, energy and other resources to manage relationships are limited. “Expectant
17 They are called expectant, because they are believed to be “expecting something” from
management. This, in turn, leads to an increase in management’s responsiveness and the level
of engagement. The most salient stakeholders are the ones that possess all three, meaning they
have the power to influence the firm, their relationship with the firm is legitimate and the stakeholder’s claim on the firm is urgent. These stakeholders are called “definitive
stakeholders” and are the ones that will receive the most attention from management
(Mitchell et al., 1997).
2.5 Aligning shareholder and stakeholder interests: the case of employees
Following academic literature, there is no universal answer whether it is best to focus
specifically on shareholder value, or to balance the needs of all stakeholders in the firm.
Especially in the short-run, decisions require making tradeoffs among the competing interests
of different stakeholders. One of the most common scenarios is the divergence between
financial performance and the welfare of certain stakeholder groups (Garcia-Castro et al.,
2011). However, especially in the long-run, the interests of shareholders and other
stakeholders may very well be aligned.
Consider the case of employees. Satisfying employee claims for higher wages or better
employment conditions may spark productivity and thus, provide management with higher
resources (Hill & Jones, 1992). Edmans (2011) found that employee satisfaction was
positively correlated with shareholder returns, which illustrates that adhering to the needs of a
stakeholder, in this case the employees of the firm, can also enhance shareholder value. In
addition, Sheridan (1992) investigated the costs of losing employees among accounting firms
and reported that firms suffered a cost of about $9.000 when a new employee had to replace a
two-year employee. When a new employee had to replace a three-year employee, firms
18 long-term employees are more likely to know all aspects of their job and accordingly, are
more productive. Furthermore, these findings stress the importance of employee satisfaction
and retention. In order to ensure that the needs of employees are addressed at higher levels in
the company, works councils or EBLR may be useful tools. The following section will
explain the importance of employee participation and elaborate on the phenomena of works
councils and EBLR in further detail.
2.6 Employee participation
As Almond, Edwards and Clarke (2003) note, the ability of employees to affect managerial
decisions, either as individuals or by collective means, is central to the discussion of varieties
of capitalism and business systems. In general, shareholder oriented, or market
outsider-model firms, tend to discourage institutions of employee voice. Managerial decisions like
restructuring tend to be on terms set by shareholders, rather than on those of employees.
Unions or works councils are considered to pose barriers to actions that benefit the firm itself
within a relatively short time period, such as downsizing (Gospel & Pendleton, 2003).
Furthermore, employee protection is weaker in market-outsider systems like the UK and the
US. Managers in these systems may be pressured to adhere to shareholder interests and
consequently, break implicit contracts with employees regarding job security or long-term
career progression. When in distress, large firms in outsider systems tend to cut labor costs in
order to maintain profitability and consequently, maintain access to capital markets, credit
ratings and resist takeovers. In contrast, firms in market insider systems tend to absorb
financial distress by reductions in returns to shareholders, rather than cuts in labor (Gospel &
Pembleton, 2003).
The difference in employee protection between market in- and outsider systems is
19 employee protection and high stock market development. The UK and the US, both
economies with high stock market capitalizations were found to have relatively weak systems
of employee protection, whereas Germany and France had lower stock market capitalizations
and strong systems of employee protection (Hall & Soskice, 2001). However, Garcia-Castro
et al. (2011) report of research by Jacoby (2005), which has suggested that firms in Japan and
Germany, two countries that are traditionally considered stakeholder-oriented, are growingly
adopting more aggressive shareholder-first policies at the cost of some of the traditional
long-term relationships that they maintained with employees and other stakeholders.
Employee participation can take several forms. In academic research regarding the
subject of employee participation in higher level decision making, many terms are used
interchangeably to describe concepts that are similar in overall objectives, but not in their
respective processes. Works councils and EBLR are mostly viewed as part of a broader
research focus on a concept called determination (Frege, 2002). Literature on
co-determination mainly focuses on Germany, as until the 1970’s, co-co-determination was largely
restricted to this country (Gold, 2011). According to FitzRoy and Kraft (2005: 233), “the legal
framework of co-determination includes the right to elect plant level works councils with
far-reaching powers in most personnel-related decisions and since 1976, a requirement that half
of the supervisory board of firms with over 2000 employees, consist of labor representatives”.
In this thesis, the approach by Frege (2002) is adopted, using co-determination as a term to
describe institutionalized participation of workers at both the firm (EBLR) and the workplace
level (works councils). The concept of co-determination is built on the notion that suppliers of
capital and suppliers of labor run the firm cooperatively (Gorton & Schmid, 2004). In the
following section, the concepts of works councils and EBLR will be explained in further
20 2.6.1 Works councils
Frege (2002: 223) defines works councils as: “institutionalized bodies of collective worker
participation at the workplace level, with specific informatory, consultative and
codetermination rights in personnel, social and economic affairs”. Works councils are found
in many European economies and consist of elected bodies of employees, that have rights to
information, consultation and in some cases co-determination of employment conditions. The
decision whether to establish a council tends to rest on the potential benefits that either the
firm or the workers see in doing so. If both workers and managers believe the benefits of a
council exceed the costs of establishing and maintaining it, there is an incentive to make the
effort. Although there is no formal connection to trade unions, most works councils are
dominated by union members (FitzRoy & Kraft, 1993). The function of works councils is to
foster cooperation between labor forces and management, with the ultimate goal of
‘increasing the size of the enterprise pie’ (Freeman & Lazear, 1995). As works councils are
not allowed to call a strike, this includes peaceful and co-operative relations with management
(Frege, 2002). However, at least in the case of German codetermination laws, works councils
do have to be informed in advance about upcoming plans that the firm may have regarding
layoffs, plant closure, internal transfers or changes in work practice. Furthermore,
consultation with management about these issues is required and in the case of layoffs,
management needs to negotiate a ‘social plan’ for the employees affected (FitzRoy & Kraft,
1993).
Benefits of works councils include the enhancement of information exchange,
employee consultation and conflict settlement. From the firm’s perspective however, works
councils tend to slow down decision making, provide opportunities for rent seeking and may
increase the bargaining power of labor, which creates obstacles to restructuring and
21 Results of research into the economic effects of works councils are mixed. For
instance, FitzRoy and Kraft (1987) found work councils associated with a significant
reduction in productivity, whereas Addison, Schnabel and Wagner (2001) found a positive
association between works council presence and productivity. Hübler and Jirjahn (2003)
reported similar findings, but found that the positive effect of works councils was significant,
only where a plant was covered by a collective labor agreement. These mixed findings can be
interpreted as another sign of the need for further clarification of the effects of works
councils.
2.6.2 Employee board level representation (EBLR)
Another concept that serves to enhance employee participation and stems from the
co-determination literature, is employee board level representation (EBLR). Although quite
similar in their objectives, works councils and EBLR do differ in process and content. The
main difference is that in the case of EBLR, employee representatives have seats in the
executive or supervisory boards of a firm, thus directly having an influence in the strategic
planning and higher level decision making process. EBLR allows for direct contact between
finance and labor, resulting in greater awareness of labor’s interests and the value of human
capital among financiers (Gospel & Pendleton, 2003). In Germany and Japan, the direct
involvement of finance in corporate governance has resulted in a pattern of strong internal
employment, characterized by more in-house training, longer job tenures and lower employee
turnover (Gospel & Pendleton, 2003). Furthermore, the presence of employee representatives
in the boardroom helps protect employee interest against potentially opportunistic behavior of
shareholders (Gorton & Schmid, 2004).
When looking at the differences between works councils and EBLR, is becomes
22 employee representatives on the board can be directly involved in the strategic
decision-making process. Works councils can thus be seen as institutions of management coordination
with core employees, whereas EBLR has a tendency towards intervention in strategic and
economic decisions of firms (Kleinknecht, 2014).
Benefits of EBLR tend to be quite similar to those of works councils. As FitzRoy and
Kraft (2005) note, EBLR can enable the use of employee information, which will lead to
more co-operative solutions and the reduction or solving of conflict between capital suppliers
(i.e. owners) and workers. On the other hand, decisions will take longer when a consensus is
needed. Such an environment can reduce the innovativeness of an organization. Thus, EBLR
can lead to a tendency to maintain the status quo in order to avoid conflict (FitzRoy & Kraft,
2005).
As was the case with works councils, empirical research into the economic effects of
EBLR has yielded inconclusive results. As FitzRoy and Kraft (1993: 366) note, “there have
been few attempts to quantify economic effects, and they all suffer from inadequate data and methodology.” In the same article, these authors report a negative influence on productivity
and profitability, whereas in a review of the literature, Gold (2011: 43) poses that EBLR “may
not do much economic good, but it does no harm either”. Gurdon and Rai (1990) found a
negative influence on productivity, but on the other hand, found that assets were used more
effectively. Addison & Schnabel (2009) report that EBLR has a minimal impact on company
performance and moreover, surveys have indicated that employee board level representatives
exercise rather little influence on the board. This seems to stem from a dilemma that
employee representatives face: influence on the board was found to be inversely related to
member alienation. In other words, the greater their influence on the board, the less influence
employee representatives have on their members, as they may be suspected of ‘selling out’ to
23 research by Benelli, Loderer and Lys (1987). They found that less risky investments were
undertaken by companies with larger degrees of co-determination. This finding was, however,
not significant.
3. Conceptual framework
Although the exact nature is still unknown, the review of previous research seems to indicate
that the concept of co-determination has the potential to influence executive decision making.
As the time horizon forms an important dimension of each managerial decision, it can be
expected that the time horizons of managerial decisions can thus be influenced by
co-determination as well. As can be derived from the literature review, managers of firms with
works councils or EBLR, in this study regarded as stakeholder-oriented firms, need to balance
the interests of many actors involved in the activities of their firm. This means that all
decisions are subject to discussion from multiple viewpoints and striving for a solution that is
most optimal for the largest part of the stakeholders.
Works councils are important actors in these discussions. Since they are not allowed to
call strikes to enforce their desired outcomes, works councils strive to maintain peaceful and
co-operative relations with managers within the firm. Furthermore, in Germany, works
councils need to be informed in advance about any upcoming plans that the firm may have
that involve layoffs, plant closure, internal transfers or changes in work practice. When these
issues arise, consultation between management and councils is required and in the case of
layoffs, management is required to develop a plan for the workers involved. Looking at the
role works councils play in the decision making process within the firm, it seems that works
councils are primarily concerned with job stability for employees, which is a long-term issue.
24 Furthermore, the required consultation with works councils in the case of important
decisions regarding layoffs or other employee-related concerns, involve a time consuming
process as well. This would indicate another factor that places its influence on the time
horizon of managers within the firm. Both methods in which stakeholders can influence
management, as posed by Frooman (1999) apply to works councils, as works councils are
capable of withholding resources from management and can also determine the way in which
management may use these resources. Since works councils do not possess a lot of power
over the firm, the latter strategy is more likely to be chosen. As works councils have power to
influence firm strategy and their claims may very well be legitimate, works councils can be expected to at least be treated as “expectant” stakeholders, as mentioned in the model by
Mitchell et al. (1997). If their claims are urgent as well, works councils could even be
considered “definitive” stakeholders and thus, will receive more attention from management.
For instance, in case of layoffs during periods of financial distress, claims by works councils
can be considered urgent.
Last, works councils are associated with increased bargaining power of labor, which
also stretches the decision making process. Managers of firms with works councils, can
therefore be expected to have longer time horizons (i.e. a stronger focus on long-term results)
than managers of firms without works councils, as the latter do not need to consult councils in
the case of important employee-related decisions. In line with this logic, the following
hypothesis can be derived:
H1: Executives of firms with works councils have longer time horizons, than executives of firms without works councils
It can be assumed that, as was the case with works councils in Hypothesis 1,
25 them with job stability for a long period. Employee board level representation is assumed to
influence the decision making process during the strategic planning phase, whereas works
councils mostly influence the implementation phase. Therefore, the influence of EBLR on
executive time horizons, may be stronger or more direct. One of the most important features
of EBLR is the facilitation of direct interaction between the financiers and the labor forces of
the company. During these meetings, usually board meetings, employee representatives can
bring employee concerns to the table and the interests of employees can be defended. The
desire for consensus may, however, slow down the decision making process. Furthermore, the
presence of employee representatives in the boardroom helps protect employee interest
against potentially opportunistic behavior of shareholders (Gorton & Schmid, 2004), which is
generally short-term oriented.
Similar to works councils, as stakeholders, employee board level representatives have
power and legitimacy to influence the firm and employee board level representatives can both
withhold resources and determine the way in which management can use these resources. The
power of EBLR might be stronger than that of works councils, as the firm is influenced at a
higher level. Still, the power of EBLR might not be large enough for the first strategy by
Frooman (1999) to be the most likely one chosen. Whether the representatives’ claims are
urgent depends on the issue at hand. The presence of employee representatives and their
expected focus on stable employment conditions that are beneficial for employees is
pressuring managers to take a long-term perspective when making decisions and since
employee board level representatives in most cases possess at least two of the attributes from
the model by Mitchell et al. (1997), it is likely that management pays attention to their claims.
In sum, managers at firms that have EBLR are expected to have longer time horizons
26 employee representatives at board level and the decision making process is expected to be
quicker. Following this, the following hypothesis can be derived:
H2: Executives of firms with EBLR have longer time horizons, than executives of firms without EBLR
4. Methodology
The empirical research carried out in this thesis consisted of an exploratory, quantitative
archival study, combined with a content analysis methodology. Executive time horizon is a
phenomenon that is difficult to quantify. It cannot simply be retrieved from a database and it
is difficult to accurately transform it into a number. A quantitative content analysis of
conference calls by executives can be used as a proper methodology to study this
phenomenon. The sample used in this study was based on a database of conference calls from
executives at the largest companies in 17 European countries in 2006. Conference calls from a
total of 303 companies were analyzed. As the current research was studying a topic at a
particular time, the research design can be considered cross-sectional.
4.1 Content analysis and description of material
Content analysis of the conference calls provided the necessary data to quantify the dependent
variable, executive time horizons, and provided further insight into the temporal orientation of
the executives at companies included in the database. Content analysis is a well-suited
research method for this purpose, as it provides a replicable methodology that enables access
to deep individual or collective structures such as values, intentions, attitudes, and cognitions
and as such, is applicable to a broad range of organizational phenomena like managerial
short-termism. Furthermore, content analysis allows rendering the rich meaning associated with
organizational documents combined with quantitative analysis (Duriau, Reger & Pfarrer,
27 the given data and the replicability, as it “implies fixed and observer-independent categories
and procedures that must be codified without reference to the analyst and the material being analyzed” (Krippendorff, 1989: 407).
The use of conference calls in a content analysis to determine executive time horizon,
has several reasons. As DesJardine and Bansal (2014) note, conference calls are unscripted,
which helps to deflect impression management. Furthermore, conference call transcripts are
archived and publicly available for publicly traded firms. Last, discussing the short- and
long-term strategy is a central component of conference calls (DesJardine & Bansal, 2014).
The conference call transcripts were retrieved from the Eikon database by Thomson Reuters
Financial, which is a database that comprises content from over 400 stock exchanges
worldwide, including transcripts of quarterly financial results conference calls by executives, usually the firm’s Chief Executive Officer (CEO) and/or Chief Financial Officer (CFO).
Access to the database was secured by the thesis project supervisor. Each conference call (and
corresponding transcript) consisted of two parts. First, a presentation-style addressing of all
conference call participants by firm executives, during which the financial results from the
preceeding quarter were discussed. Afterwards, in the second part, analysts, journalists and
shareholders were given the opportunity to ask questions regarding the results. Ideally, all
four quarterly conference calls from the year 2006 were present in the database. However, this
did not always prove to be the case. In some cases, firms only gave yearly or half-yearly
financial results conference calls.
4.2 Procedure
The content analysis was performed using word counting software, in this case CATScanner.
The codes and included keywords were comprised from the mentioned study by DesJardine
28 reflecting a short-term focus and one with words reflecting a long-term focus. Both sets of
keywords can be found in the appendix of this thesis. Furthermore, the content analysis was
performed in collaboration with fellow students that were participating in the same thesis
project. This way, a larger sample could be established, thus improving the reliability of the
research. The specific method used, was a replication of the content analysis performed by
DesJardine and Bansal (2014), who studied the influence that financial analysts had on the
temporal horizons of CEOs.
In order to ensure that all steps were performed in a way that maximized reliability, all
participating students met several times prior to the analysis, in collaboration with the project
supervisor, Mr. Robert Kleinknecht. During these meetings, guidelines for the analysis were
set and potential pitfalls were discussed. These guidelines included that both the presentation
and the Q&A part of the conference calls were to be included in the transcripts, so a maximal
amount of information could be analyzed. Questions asked by financial analysts or journalists
were deleted from the transcripts, since these questions were not the words of firm executives
and thus, were not a reflection of firm strategy. As this was quite an extensive and time
consuming task, the collaborating students took shifts in which each student spent four hours
deleting the questions from the transcripts.
Another important guideline was the decision that there would not be made any
difference between the CEO or the CFO talking during the conference calls. The first reason
for this, was that in many cases, the CEO did not address the public, but left this task to the
CFO, the head of Investor Relations, or another board member. Furthermore, as both are CEO
and CFO are executives, it was agreed on that statements made by either of them, reflected
firm strategy and time horizon. Last, after discussion, the list of keywords from the
DesJardine and Bansal (2014) article was retained integrally. There was some discussion about whether to delete the word ‘quarter’ from the keyword list, since most conference calls
29 were quarterly financial results presentations, in which the word ‘quarter’ was mentioned
many times, without necessarily being part of statements of strategic intent. However, since
DesJardine and Bansal (2014) validated their keyword list extensively, the students agreed to
retain this word in the keyword list used. Last, in some cases, at least one quarterly transcript
was missing in the database. The participating students agreed that as much content as
possible needed to be included and chose to include all available material from 2006 for each
company. This did, however, result in transcripts that varied in size. On average, the
transcripts had 21.278 words (SD = 11.267) ranging from a minimum of 1.159 to a maximum
of 56.779.
For each company with transcripts available, two files were created. One file covered
all the content from the presentation part of each conference call that the company had given.
The second file covered all content from the Q&A part of each conference call that the
company had given. Using the CATScanner, both files for each company were analyzed,
which provided information on the number of words, number of characters, number of
short-term related keywords and number of long-short-term related keywords in the transcript. These
numbers could be used to construct a measure for executive time horizon, as will be explained
in further detail in the next section.
4.3 Operationalization of variables
The measures for the independent variables, works councils and EBLR, were based on
secondary data provided by the supervisor of the project. This database was used by the
supervisor in his own (previous) research and consisted of company data from the largest
publicly traded companies in 17 European countries in 2006. The database included financial
30 companies involved. Most importantly, it specifically provided information about the
existence of either works councils or EBLR at the companies included in the database.
The existence of works councils at the firms covered in the database, was measured using a dummy variable, in which a “0” indicated that no works councils were present at the
given firm, and a “1” indicated that works councils were present. Out of the 303 companies
included in the sample, 136 had works councils installed and 167 had not. The sample of
companies from this study included German works councils, but also works councils from
The Netherlands, Belgium, Ireland, Portugal, Greece, Denmark, Sweden, Norway, Finland,
Austria, Italy, Spain, France and Switzerland. Table 1.1, on page 31, contains an overview of
the number of firms per country and the number of firms that had works councils and/or
EBLR installed.
The second independent variable, EBLR, was measured in a similar way. A dummy
variable indicated whether employee board level representatives were present at the given firm, using a “0” to indicate that the firm had no employee board level representatives, and a
“1” to indicate that employee board level representatives were present at the firm. In total,
one-third (101 out of 303) firms had employee board level representatives, with an average of
5 employee representatives in these firms. The firms with employee board level
representatives came from Germany, Luxembourg, Greece, Denmark, Sweden, Norway,
Finland, Austria, France and Switzerland. For an overview of the distribution, see Table 1.1
31 The measure for the dependent variable, firms’ time horizon was established using the
data from the content analysis. Again, the same methodology as in DesJardine and Bansal
(2014) was applied, which means that time horizon was calculated using the following
formula:
Time Horizon = # of Long time horizon words
(# of Short time horizon words + # of Long time horizon words)
This formula was chosen because its application results in a ratio of short-term to all
time-related keywords, rather than displaying the number of short-term time-related keywords as a
percentage of the total number of words or characters in each transcripts. It ignores the
amount of words or characters in the transcript and focuses on actual short- and long-term Table 1.1: Distribution of firms in the database
Country Number of firms Firms with works councils Firms with EBLR
1. The Netherlands 25 13 0 2. Germany 49 27 43 3. Belgium 8 6 0 4. Luxembourg 2 1 1 5. Ireland 9 1 0 6. Portugal 9 1 0 7. Greece 8 1 2 8. Denmark 13 3 11 9. Sweden 25 18 20 10. Norway 16 5 8 11. Finland 20 12 2 12. Austria 8 4 6 13. Italy 28 6 0 14. Spain 17 2 0 15. France 51 23 6 16. Turkey 4 0 0 17. Switzerland 11 8 2 Total 303 131 101
32 related keywords. Since the amount of words and characters varied among all transcripts, the
given formula was deemed the most suited for the present study.
In the content analysis, the presentation and Q&A part of each transcript were
analyzed separately. The application of the formula resulted in the construction of three
versions of the dependent variable. The first version (called THPres) was based on the results
from the analysis of the presentation part of the conference calls only. The second version
(called THQA) was based on the results of the analysis of the Q&A part of the conference
call. Last, the scores of both the presentation part, and the Q&A part were added up, resulting in a ‘total’ amount of short- and long-term related keywords for each company, to which the
formula was applied. This way, the third version of the dependent variable (called THTOT) was constructed. This last, ‘total’ version was used in the statistical analysis, as it was based
on the maximum amount of content available. Executive time horizon was thus measured
using a continuous variable named time horizon ratio (THTOT). A higher ratio indicated a
longer time horizon, thus a stronger focus on long term results. Conversely, a low ratio
indicated a shorter time horizon, thus a stronger focus on short term results.
4.4 Control variables
In order to isolate the effect of works councils and/or EBLR on executive time horizons from
other influences, a number of control variables was used in the statistical analysis. The first
control variable used in this study was the national culture of the countries of origin of the
firms in the database. Managerial time horizons may be affected by national culture-related
factors, especially those related to time. Hofstede (2003) is well known for his research on
cultural differences between countries. In his research, carried out at IBM, he initially
distinguished four dimensions on which national cultures can be compared among countries:
33 masculinity versus femininity. Later a fifth dimension, long term orientation (LTO), was
added. Each national culture that was researched by Hofstede, received a score between 1 and
100 on each of his dimensions. As the LTO dimension of national culture was most relevant
for this research, this dimension was used as the benchmark on which the different countries
were grouped for use as a control variable in the statistical analysis.
In order to assess the long term orientation of the national cultures of the countries in which the firms in this study are active, the scores of each of these countries on Hofstede’s
long term orientation dimension were assessed and compared. The used scores were retrieved
from the World Value Survey results found in Hofstede, Hofstede and Minkov (2010), as this
work contained a larger and more up-to-date sample of scores for the long-term orientation
dimension, and covered all countries included in the sample. It needs to be noted that the scores on Hofstede’s dimensions are not concrete numbers to describe reality, but are
generalized and are ought to be relative (Hofstede, Hofstede & Minkov, 2010). In the case of
long-term orientation, the lower the score on this dimension, the more short term oriented the
national culture was.
The scores of the 17 European countries included in this study ranged from 24
(Ireland) to 83 (Germany). The overview of scores lead to the grouping of the countries into
three groups: (1) short-term oriented national culture (NatCulSTO), which included all countries with scores lower than 40 , (2) ‘medium-term’ oriented national culture, which
included all countries with scores 40-65 (NatCulMTO) and (3) long-term oriented culture,
which included all countries with scores above 65 (NatCulLTO). The largest group, ‘medium term’ orientation national culture, represented cultures that were neither short- nor long term
focused and was used as a reference category. The distribution of countries among the
different groups, including their relative scores on Hofstede’s long term orientation dimension
34 Another way that was examined to classify countries is by using the research of La
Porta, Lopez-de-Silanes, Shleifer and Vishny (1998), who grouped countries based on either
English, French, German or Scandinavian origin and corresponding corporate governance
practices. This method was adopted initially, but after extensive testing and given the fact that
the English origin group from the sample of the current research merely consisted of 9 Irish
firms, it was decided not to use this method in the statistical analysis, as it would reduce the
validity of the research.
The second control variable that was used, was the type of industry in which the firm
was active. Different industry characteristics can result in different managerial time horizons.
The financial industry, for instance, is notorious because of its focus on short-term results.
Other industries might have completely different time horizons. Industries like real estate or
the energy sector can be expected to be focused on the long term. Time horizons might also
vary within industries. For instance, two years may be a long time horizon in the technology
sector, whereas in the mining industry, this is considered a short time horizon (DesJardine &
Bansal, 2014). A total of 23 industries was present in the database, represented by a dummy Table 1.2: Categorization of National Culture variable
Culture Group Countries included N
Short-term oriented National Culture Ireland (24), Portugal (28), 67
(scores < 40) Denmark (35), Norway (35),
Finland (38)
Medium-term oriented National Culture Greece (45), Turkey (46), 143
(scores 40-65) Spain (48), Sweden (53),
Austria (60), Italy (61), France (63), Luxembourg (64)
Long-term oriented National Culture The Netherlands (67), 93 (scores >65) Switzerland (74), Belgium (82),
35 variable for each industry. In order to simplify the analysis, these industries were grouped
based on their NACE-code. A NACE (Nomenclature of Economic Activities) code is used to
group organizations based on their economic activities. Each code is divided into four digits.
The first two digits identify the division in which the firm is active (e.g. agriculture, mining,
etc.), the third digit represents the group within this division and the fourth digit stands for the
specific class of activity. The grouping of industries was based on the first two digits of each industry’s NACE code, retrieved from an online overview established by the European
Commission (2010). This resulted in the following industry groups, that were later recoded
into dummy variables. The ‘Other industries’ category served as the reference group. Table
1.3 below provides an overview of the categorization of the industry variable:
The third control variable was firm size. It was expected that as firms were in different
stages of their development, time horizons of their managers would differ accordingly. For
instance, a smaller firm, aiming at growth might have a stronger interest in short-term results
in order to grow, whereas a larger, established firm might be looking for ways to retain its Table 1.3: Categorization of Industry variable
Industry Group Industries included N
1. Construction, Mining & Manufacturing Construction, Mining, Petroleum, 60
(CMM) Lumber, Glass, Steel
2. Fast Moving Consumer Goods (FMCG) Food, Apparel, Wholesale, Retail 35
3. Financial Banks, Insurance, Holdings 54
4. Media Radio/Telephone, Advertising 40
5. Engineering Electrics, Electronics, Engines, 57
Motor vehicles
6. Other industries Chemicals, Transport, Real Estate, 57 Other industries
36 leading position for a longer period of time. Firm size was measured by the firms’ number of
employees in 2006, as was recorded in the database. The measure consisted of a continuous
variable, in which the number given represented the actual number of employees that the firm
had. The companies in the sample had an average of 44.472 employees (SD = 72.133).
The last control variable was firm performance. As business moves in cycles, it is not
uncommon for managers to alter their time horizons when business is bad. During times of
low or negative growth (i.e. losses), short-term results are needed in order to get the company
back on track and to win back the trust of all those involved (customers, shareholders, other
stakeholders, etc.). Conversely, in times of strong growth, managers may be interested in
ways to sustain this growth for a long period, thereby stretching their time horizons to a
longer timeframe when making decisions. Several variables, capable of indicating firm
performance, were present in the database, with the most common ones being 2006 Earnings
before Interest, Tax, Depreciation & Amortization (EBITDA), Return on Equity (ROE),
Return on Assets (ROA). Return on Assets (ROA) was expected to be the best indication of
firm performance and was used in the analyses, following the example of DesJardine and
Bansal (2014). This variable was used as a continuous variable, in which the number given
represented the actual ROA percentage achieved by the firm in 2006. The firms in the
database had an average Return on Assets of 6.69% (SD = 18.20).
5. Results
Hypotheses were tested using multiple regression analysis in SPSS. Regression analysis is a
useful tool in which a linear model is used to predict values of an outcome variable, based on
one or more predictor variables (Field, 2013). In this case, the outcome variable to be
predicted was executive time horizon and the predictor variables were the presence of works
37 and industry. In anticipation of the regression analyses, a few preliminary analyses needed to
be performed.
5.1 Preliminary analyses.
The first preliminary step was to check for errors in the data. This was done by frequency
checks of the relevant variables in SPSS. No large errors or outliers were found. Second,
descriptive statistics, skewness, kurtosis and normality were checked for the dependent,
independent and control variables. Descriptive statistics can be found in the correlation matrix
at the end of this section. A closer look at the distribution of the time horizon variable
(THTOT), revealed an asymmetrical distribution, in the form of positive skew (1.114). This
indicates that the frequent scores are clustered at the lower end of the distribution and the tail
points towards the higher end of the distribution (Field, 2013). Also, the distribution was
leptokurtic (1.446). In order to normalize the distribution, a logarithmic (Log10)
transformation of the time horizon variable was done. Taking the logarithm of a variable is a
good way to reduce positive skew, as it flattens the right side of the distribution (Field, 2013).
However, as it is impossible to get a logarithmic value from zero or negative numbers, a
constant needed to be added. Since the time horizon variable was measured by a ratio in
which all the observations had values between 0 and 1, a constant of 1 (Xi + 1) was added to
the transformation. The logarithmic transformation corrected the time horizon variable’s skew
to a statistic of 0.916, which, after consultation with the supervisor, was deemed acceptable
for use in further analyses.
As most of the independent variables were dummy variables, no comments regarding
skewness, kurtosis and normality can be made. The other predictors that were not dummy
variables, firm size and firm performance did show distributions that did not meet the
normality requirement. However, since they were not the dependent variables and after
38 5.2 Correlations
After normalizing the dependent variable, correlations for all variables in the model were
checked. Assessing the correlations between the variables in the model provides an insight
into the way variables are related and can provide a general idea of what the regression model
might be going to look like. Correlations, along with descriptive statistics for all variables
included in the first regression model can be found in the matrix on pages 40 and 41. A few
remarks regarding the implications need to be made. First, the control variables ‘national culture’ and ‘industry type’ were measured using dummy variables. Therefore, they were
mutually exclusive and negative correlations between them were logical and inevitable.
Second, the mean and standard deviation of the time horizon variable in the table were based
on the logarithmic values of this variable. The statistics of the original variable were as
follows: M = 0,16, SD = 0,07.
Looking at the correlation matrix, some significant correlations between national
culture or industries and the existence of works councils and/or EBLR could be distinguished.
Both works councils and EBLR correlated positively with long-term oriented national
cultures. This is not surprising given when keeping in mind that Germany, the country from
which works councils and EBLR have emerged, was included in this group and short-term
oriented economies tend to dislike employee participation. Furthermore, works councils
correlated positively with firms in the construction, mining & manufacturing industry and
negatively with firms from the financial and media industries. Works councils also correlated
strongly with firm size, indicating that works councils tend to be more present at larger firms.
A significant correlation between works councils and EBLR existed, indicating a tendency for
39 Like works councils, EBLR correlated positively with long-term oriented national
culture. In contrast to works councils, no negative correlation with short-term oriented
cultures existed. Regarding industries, EBLR correlated with firms from the engineering
industry and with firm size, implying that EBLR is mainly present at firms from the industry
mentioned before and at larger firms.
Executive time horizon seemed to significantly correlate negatively with short-term
oriented national culture, indicating that managerial time horizons are shorter in countries
with a short-term oriented culture. Furthermore, a positive correlation between executive time
horizon and firm size was found, indicating that executives at larger firms tend to have
relatively long time horizons. Most importantly though, no significant correlation was found
between executive time horizon and the existence of works councils at the firm. The
correlation between executive time horizon and the presence of EBLR was significant.
However, contrary to what might be expected based on the literature, this correlation was
negative, implying that executives at firms with EBLR present actually had shorter time
