Kazan CG

Acknowledgements
This thesis is the last assignment to finalize the Master of Science in Business Administration
with the specialization track Financial Management at the University of Twente.
I would like to thank my family for their support during my study. I also would like to thank
my supervisors for their help during the process of the thesis.
Emre Kazan
March, 2022
1
Abstract
This study examines the impact of board characteristics on firm performance for German
listed firms. Previous literature presents different findings on the impact of board
characteristics on firm performance. Four important theories, namely the agency theory,
stewardship theory, resource dependence theory and human capital theory are described.
Three main board characteristics are analyzed, namely board size, independence and gender
diversity. Performance is measured based on two accounting-based (ROA and ROE) and a
market-based measure (Tobin’s Q). The relationship between the board characteristics and
firm performance is tested by conducting an OLS regression analysis, using data from a
sample of 89 German firms from the Frankfurt Stock Exchange for the period of 2017 to
2019. The results show a negative significant relationship between board size and firm
performance. There is a negative significant relationship between board independence and
the accounting-based measures, whereas the relationship with Tobin’s Q is significant and
positive. Gender diversity has no significant impact on ROA and ROE, and a significant
negative impact on Tobin’s Q.
2
Table of Contents
Acknowledgements ............................................................................................................................1
Abstract ..............................................................................................................................................2
1. Introduction ................................................................................................................................5
1.1. Outline of the Study.............................................................................................................6
2. Literature Review ........................................................................................................................7
2.1. Theoretical Perspectives ......................................................................................................7
2.1.1. Agency theory..............................................................................................................7
2.1.2. Stewardship Theory .....................................................................................................8
2.1.3. Resource Dependence Theory......................................................................................8
2.1.4. Human Capital Theory .................................................................................................9
2.2. Board of Directors ............................................................................................................. 10
2.3. German Board Structure .................................................................................................... 10
2.3.1. The Management Board ............................................................................................ 11
2.3.2. The Supervisory Board ............................................................................................... 12
2.4. Board Characteristics ......................................................................................................... 13
2.4.1. Board Size .................................................................................................................. 13
2.4.2. Board Independence .................................................................................................. 14
2.4.3. Gender Diversity ........................................................................................................ 15
2.5. Empirical Findings .............................................................................................................. 16
2.5.1. Board Size and Firm Performance .............................................................................. 17
2.5.2. Board Independence and Firm Performance .............................................................. 17
2.5.3. Gender Diversity and Firm Performance..................................................................... 18
2.6. Hypothesis Development................................................................................................... 19
2.6.1. Board Size .................................................................................................................. 19
3. Research Methodology ............................................................................................................. 22
3.1. Methods in Previous Studies .............................................................................................. 22
3.1.1. OLS Regression .......................................................................................................... 22
3.1.2. Fixed Effects Regression ............................................................................................. 22
3.1.3. 2SLS Regression ......................................................................................................... 23
3.2. Method in this Study ......................................................................................................... 23
3.3. Variables ........................................................................................................................... 24
3.3.1. Dependent Variables.................................................................................................. 24
3.3.2. Independent Variables ............................................................................................... 24
3
3.3.3. Control Variables ....................................................................................................... 25
4. Sample and Data ....................................................................................................................... 28
4.1. Industry Classification ........................................................................................................ 29
5. Results ...................................................................................................................................... 30
5.1. Descriptive Statistics .......................................................................................................... 30
5.1.1. Dependent Variables.................................................................................................. 30
5.1.2. Independent Variables ............................................................................................... 30
5.1.3. Control Variables ....................................................................................................... 32
5.2. Correlation Analysis ........................................................................................................... 33
5.3. Multicollinearity Test ......................................................................................................... 36
5.4. Regression Analysis ........................................................................................................... 36
5.4.1. Board Size .................................................................................................................. 37
6. Conclusion ................................................................................................................................ 42
7. Limitations and Recommendations ........................................................................................... 44
References........................................................................................................................................ 45
Appendices ....................................................................................................................................... 50
Appendix A: List of sample firms ................................................................................................... 50
Appendix B: Frequencies ............................................................................................................... 51
Appendix C: Regression Standardized Residual.............................................................................. 52
Appendix D: Regression Plots ........................................................................................................ 54
4
1. Introduction
There are several key factors playing a role in the success of a firm and good corporate
governance is a major one of them. Corporate governance is important because it deals with
the procedures and processes according to which firms are directed and controlled by the
board of directors, the CEO and management (Martinez and Alvarez, 2019). Research shows
that firms with good corporate governance have better performance compared to firms with
poor corporate governance (Makhlouf, Laili, Basah and Ramli, 2017; Kao, Hodgkinson and
Jaafar, 2019). Being such an important element for firm success, corporate governance and
its mechanisms have been the focus point of many studies. An important mechanism that
ensures good corporate governance is the board of directors of a firm.
The board of directors is assigned to take on crucial tasks concerning the firm. They are
responsible for guiding and authorizing the firm’s strategic decisions, including mergers and
acquisitions, capital structures, and hiring or firing executives (Terjesen, Couto and
Francisco, 2016). In turn, these issues have an impact on the firm’s financial performance. In
addition, Martinez and Alvarez (2019) describe that the board of directors is a highly
effective corporate governance mechanism because it fulfills two important functions,
namely the supervision of executive management while representing the shareholders, and
providing business resources. It is their responsibility to monitor firm performance and the
behavior of executive managers.
The separation between management and owners is the most interesting and popular issue
to pay attention to. The possible conflicts of interest that may arise from this separation are
referred to as the agency problem (Kao et al., 2019; Martinez and Alvarez, 2019). According
to the agency theory, the board of directors is an essential element and acts as a control
mechanism to ensure that the problems resulting from the principal and agent relationship
are controlled (Martinez and Alvarez, 2019). For the shareholders and investors of a firm the
board of directors are a sort of insurance, because their task is to align the interests of
managers with those shareholders who are not managing the firm directly (Martin and
Herrero, 2018).
Interest in board of directors has been increasing in recent years due to corporate fraud,
scandals in multinationals and negligence of shareholders (Martinez and Alvarez, 2019;
Terjesen et al., 2016). Previous literature present different findings on the topic, which is
understandable since governance can differ from country to country. Therefore it is even
more interesting to do research on this topic with a specific focus country. In this study the
focus will be on three major board characteristics: the size, independence and gender
diversity. Discussions on the ideal board size and the impact of independence have been
very important for firms and research over the years and the ideas have been mixed. Gender
diversity has especially been a focus point in recent years in literature and society. Unlike
most other studies, this paper will thus include three different board characteristics and do
research to their impact on firm performance. Different from other studies, in this research
there will be accounting-based and market-based performance measures. Including multiple
performance measures will make the results of this study more valuable, and form a more
clear idea in practice for firms.
5
The focus in this study will be on German listed firms. Germany is the largest economy in
Europe and the results of this research can present an idea whether the board of directors
has an impact on the performance of German listed firms. Different from other countries,
Germany has a two-tier board structure which will make it interesting to analyze. A two-tier
board structure is divided in a management board and a supervisory board. Another factor
that separates Germany from other countries is codetermination on the supervisory board,
which makes relations on German boards even more complex and interesting to study.
Many previous studies focus on countries with a one-tier board, or studies on Germany tend
to choose either the management or supervisory board to do research on. This study will
analyze both the management and supervisory board of German firms and present new and
valuable findings on the topic from an important European country with a different structure
and law. Another important factor separating this study from most German studies is the in
2016 introduced gender quota that is applicable for German supervisory boards. Results in
this study are valuable in this sense since the gender quota is quite recent and many
previous studies do not present the effects of it on firm performance. Findings in this study
can act as an example for future research in the field of board of directors for firms adapting
a two-tier structure, making this study very interesting. This leads to the research question
of this study:
What is the impact of the board of directors’ characteristics on the firm performance of firms
listed in Germany?
The main objective of this study is to find an answer to the research question, thereby
contributing to the literature written on this topic. In addition to contributing to literature,
the findings can be valuable in practice as well.
1.1. Outline of the Study

The remainder of this study consists of six chapters where subsections are also included.
Chapter 2 is the literature review where four main theories are presented in the theoretical
perspectives. Next to this, Chapter 2 describes the board of directors, presents the German
board structure with the separation between the management and supervisory board. The
three main board characteristics of this study and their implementation in German boards
are explained in this chapter as well. Chapter 2 concludes with empirical findings from
previous studies and hypotheses development based on the theories and empirical findings.
Chapter 3 describes the research methodology. First, several models from comparable
studies are explained, then the model for this study is formed. This chapter ends with
defining the dependent, independent and control variables and their measurements.
Chapter 4 presents the sample and data collection process of this study, and it describes the
industry classification of the sample firms. Chapter 5 presents the results of the research.
First the descriptive statistics are describes, followed by the correlation analysis and
multicollinearity test. The chapters end with an analysis on the regression results. Finally,
Chapter 6 gives the conclusion of the study and in Chapter 7 the limitations and
recommendations are described.
6
2. Literature Review
This section of the paper presents the literature review and is divided in six subsections. First
the theoretical perspective is presented, which includes four important theories on the
topic. Then, there is a description on the board of directors and the German board structure
is explained. The chapter continues with the three board characteristics and their
implementation in German boards. Subsection 5 presents the empirical findings from
previous studies. This chapter ends with a hypotheses development, where the theoretical
perspectives and empirical findings are referred to in forming hypotheses.
2.1. Theoretical Perspectives

This chapter will further describe four important theories, namely the agency theory,
stewardship theory, resource dependence theory and human capital theory. These theories
are commonly used throughout literature on the topic of board of characteristics. They help
in understanding the principal and agent relationship in corporate governance. In addition,
these theories describe how individuals can be valuable and beneficial for firms.
2.1.1. Agency theory

The most important and most used theory in studies regarding corporate governance is the
agency theory. The core of the agency theory is the separation between ownership and
control and explaining the fundamentals of the relationship between the two parties in the
firm (Martinez and Alvarez, 2019). The two parties are commonly referred to as the principal
and the agent, and the agency theory can help understand the potential problems that arise
between these two parties. The principal are the owners and shareholders of a firm and the
agent stands for the managers who are hired by the owners. The problems between these
parties arise when the principal employs the agent to create value for the firm (Bosse and
Phillips, 2016). Hereby, the main goal of the principal is to maximize firm value and
performance, and therefore this task is delegated to the agent (Martinez and Alvarez, 2019).
By delegating the task to the agent, the principal demands the agent to work in the benefit
of the owners (Panda and Leepsa, 2017).
However, in reality the managers are more interested in their own compensation and
interests. The opposite goals and interests of the principal and the agent define the agency
problem, which eventually results in agency costs for the firm (Panda and Leepsa, 2017).
Besides the divergent interests, Bosse and Phillips (2016) add another factor that cause
problems, namely that the agent has better information than the principal often referred to
as information asymmetry. These conditions create the possibility that the agent will not act
in the best interests of the principal, resulting in poor firm performance. By addressing the
problems between the principal and the agent using the agency theory, it is possible to find
solutions and reduce these problems. This theory can be helpful in implementing
governance mechanisms to control, supervise and monitor managers (Panda and Leepsa,
2017). With a proper governance system the agency conflict can be minimized and
eventually result in reduced agency costs.
Hereby, the board of directors plays an important role because the board is considered as an
institution to mitigate agency problems between the owners and the managers (Kao et al.,
2019). According to agency theory, the board of directors is an essential element of the
7
control mechanism to ensure that problems resulting from the principal and agent
relationship are controlled (Martinez and Alvarez, 2019). The board is responsible to present
governance that maximizes the firm’s outcomes by aligning owner and manager interests
and minimizing agency costs (Bosse and Phillips, 2016). The board of directors acts as an
internal governance mechanism by speaking with managers and monitoring them on behalf
of the owners (Bosse and Phillips, 2016; Kao et al., 2019). The main objective is to reduce the
agency problem between owners and managers and improve firm performance.
2.1.2. Stewardship Theory

Stewardship theory is commonly used as an alternative and complementary approach when
focusing on the principal and agent relationship. This theory emphasizes on cooperation and
collaboration, and provides a non-economic premise for explaining principal and agent
relationships (Keay, 2017). The theory describes that the managers act as stewards for the
company they work for. Contrary to the agency theory, stewardship theory assumes that
managers do not behave selfish and act on the best interest of the principal (Schillemans and
Bjurstrom, 2020). Keay (2017) describes that the agent who acts a steward will not be
concerned about their own economic interests, but will act in the best interests of their
company, this approach will eventually lead to a better firm performance and in turn the
personal interests of the manager are also fulfilled. Thus, stewardship theory points out the
idea of service for others and not self-interest where non-financial factors such as intrinsic
satisfaction from achievement, respect, reputation and trust are key factors for the
managers (Duru et al., 2016; Keay, 2017).
Alignment of principal and agent interests is also important in the stewardship theory, but
this theory assumes that interests are already aligned. The managers already work towards
the objectives of their principal, and they are viewed as loyal to the company (Keay, 2017).
The composition, structure and characteristics of the board of directors is used as an
important indicator of stewardship (Schillemans and Bjorstrom, 2020). Contrary to the roles
of the board of directors with the agency theory, the primary function of the board in the
stewardship theory is to lend support, give advice and share experience with managers
(Glinkowska and Kaczmarek, 2015). Thus, the board does not focus on aligning principal and
agent interests but rather on facilitative and empowering structures within the firm
(Donaldson and Davis, 1991). This theory is very helpful in analyzing the board of directors,
their expected functions and how they handle their relationships with managers. A common
example for this theory on boards is that it assumes that insider dominated boards will boost
firm performance (Ramdani and Witteloostuijn, 2010).
2.1.3. Resource Dependence Theory

The key perspective in resource dependence theory is considering the role of external
resources needed by the firm and how these affect firm behavior (Terjesen et al., 2016). This
theory explains the organizational behavior in terms of those critical external resources a
firm must have in order to function and create value. The resource dependence theory
assumes that dependence on critical and important resources influences the actions of
organizations and that organizational decisions and actions can be explained depending on
the particular dependence situation (Nienhuser, 2008). Therefore, the core of this theory is
thus the flow of resources between firms (Johnson, 1995). In order to form an organization
8
providing the critical external resources, an effective corporate system is needed, resulting
in a better firm performance.
The role of the board of directors is very important in this aspect of the resource
dependence theory. According to Martinez and Alvarez (2019) the board of directors help
firms improve their performance by reducing their dependence on the external
environment. Resource dependency theory describes that boards have to be considered as
an asset to the firm since they are providers of resources that are not otherwise available
(Pugliese, Minichilli and Zattoni, 2014). The board is also able to adjust their behavior to the
needs of the firm, hereby securing access to valuable resources and supporting the
managers in making decisions about future directions aiming to increase firm performance
(Pugliese et al., 2014). This makes the board of directors an essential link between the firm
and the required external resources to maximize performance (Martinez and Alvarez, 2019).
Resource dependence theory is also important in the sense that it affects the composition,
size and different characteristics of the board of directors. It suggests for example that a
larger board can provide access to a wider range of resources which influences firm
performance positively (Bennouri et al., 2018). Terjesen et al. (2016) present another
example by explaining that independent directors have access to valuable knowledge and
expertise on the firm and therefore are able to expand their firms’ boundaries with links to
external resources.
2.1.4. Human Capital Theory

Human capital theory focuses on the individual and the qualities he or she has that makes
them unique and valuable such as knowledge and personal experience. This theory can be
considered as a support for the resource dependence theory because it focuses on what
individuals can bring and add to organizations. Human capital is typically developed through
investments in education, training and various experiences (Kor, 2008). Hereby, education
and experience are considered as the main characteristics of human capital, and the
knowledge it brings creates a competitive advantage for the individual (Dimov and
Shepherd, 2005). Having these individuals representing the firm makes is a key factor for its
performance and can make the difference with competing firms in the market (Dimov and
Shepherd, 2005; Gillies, 2015). Human capital theory makes a distinction between skills and
knowledge gained from team-level, firm-level and industry-level experiences (Kor, 2008).
According to Valenti and Horner (2020) extensive experiences in those specific levels impact
decision making and increases the likelihood of positive outcomes.
Human capital theory also states that individuals with more qualities are able to achieve
higher performance for their firms (Dimov and Shepherd, 2005). Each board member brings
a unique set of human capital. Therefore, it is important for firms to form their boards with
individuals who can provide critical resources through their knowledge, expertise and
personal experiences (Terjesen, Singh and Vinnicombe, 2008). It is possible to gain a
competitive advantage by having these individuals representing the board of the firm, here
also lies the link between human capital theory and firm performance (Dimov and Shepherd,
2005). However, it is also important to keep in mind that qualitative aspects of human
capital are more important than quantitative aspects, meaning that firms should not solely
9
focus on attracting as many board members but focus on quality (Dimov and Shepherd,
2005).
2.2. Board of Directors

The board of directors has been the focus point of many studies regarding corporate
governance since many years. Throughout these studies, the board of directors has been
described as a crucial governance mechanism for firms to implement in order to realize firm
success. Studies on the board of directors conclude that the board is the most important and
effective internal governance mechanism (Martinez and Alvarez, 2019; Kao et al., 2019;
Naciti, 2019). Generally, the board of directors of a firm is the body that determines policies
for management and has the ability to make decisions on major company issues. The board
of directors is considered necessary, as the separation of corporate control from ownership
potentially gives managers space to pursue their own interests, at the expense of the
owners of the firm (Bezemer, Peij, de Kruijs and Maassen, 2014). With this, the main
responsibility of the board of directors is protecting the interests of the firm’s stakeholders
through directing operations and supporting the decision-making process (Naciti, 2019).
Together with this responsibility, the board of directors are given many authorities within
the firm such as hiring and firing executive directors, set CEO compensation, and be involved
in financial statements (Bezemer et al., 2014).
Next to the responsibility towards the owners of the firm, according to Martinez and Alvarez
(2019) the board of directors has two main functions in companies. These are the
supervision of executive management and providing resources to the company. In the
supervisory role, the board use their resources to monitor firm performance and the
behavior of managers. By conducting the supervisory role in a good way, the board is even
able to reduce agency problems between the owners and managers of a firm, which
eventually can improve firm performance (Kao et al., 2019). The importance of the
supervisory role of the board can also be seen by international scandals involving major
organizations where the boards fail or struggle to supervise management in a proper
manner (Bezemer et al., 2014). A strong board of directors has shown to reduce agency
problems and encourage managers to operate properly and in the best interest of the
owners (Naciti, 2019).
The main functions and tasks given to the board of directors make it an interesting topic for
previous studies researching the impact of board characteristics on firm performance. Many
previous studies describe that several board characteristics have to be analyzed in order to
be able to present results whether these have an impact on firm performance or not.
Hereby, research is done on the characteristics such as the size of the board, independence
of the board, diversity on the board etc., more on this in the next subsections (Naciti, 2019;
El-Faitouri, 2014). How the board of directors and the different characteristics are formed
also depends on the structure of the board. Since the focus in this study is on German
boards, next section describes their structure further in detail.
2.3. German Board Structure

Studies on corporate governance and the board of directors define two different board
structures, namely the one-tier and the two-tier structures. In one-tier board structures the
10
board consists of both executive and non-executive directors, where the decision-
management and decision-control are integrated in one organizational body (Bezemer et al.,
2014; Bozhinov, Joecks and Scharfenkamp, 2021). This structure is commonly used in
countries as the US and the UK. Contrary to the one-tier structure are thus the two-tier
boards, which is also the structure used by German firms. Since this thesis focuses on
German listed firms, the structure which will be described further in detail will be the two-
tier structure used in Germany.
According to the German Stock Corporation Act, it is mandatory for German firms to have
the two-tier board structure (Jungmann, 2007). The main idea of two-tier boards is that
there is a separation between decision-management and decision-control. In two-tier board
structures there is a formal separation between the two roles of the boards. The executive
directors who are responsible for the daily operations of the firm are part of the
management board, and the non-executive directors who are responsible for the
supervision of executive directors are part of the supervisory board (Bezemer et al., 2014).
There is a clear divide between the management board and the supervisory board since
simultaneous membership on both boards is not permitted (Jungmann, 2007).
2.3.1. The Management Board

The management board thus exists out of executive directors who are being hired, or fired
when found necessary, by the members of the supervisory board. Within this two-tier
structure, the executive directors in the management board are mainly responsible for the
daily operations of the firm and managing and directing the business of the firm (du Plessis,
2004; Fauver and Fuerst, 2006; Block and Gerstner, 2016). They decide about the objectives
of the company and implement the necessary measures to achieve those objectives.
Some specific tasks of the management board are described by Block and Gerstner (2016) as
follows: providing the strategic direction of the company through careful planning of
operations, managing the workforce, controlling the strategic planning of the company and
maintaining the books of account. Next to managing the daily operations of the company,
the management board also sets long term goals and guidelines. Hereby the main objective
is running the business in a way that allows further development and increase financial
performance (Jungmann, 2007).
An important aspect to successfully conduct these tasks is the information sharing between
the members of the management board, this helps in creating an efficient decision-making
process within the management board (Jungmann, 2007). The executive directors on the
management board obtain all information that is necessary for their tasks directly from the
contact with employees and junior management. This creates an additional task for the
management board, which is to minimize the information asymmetry with the supervisory
board. Since the management board is able to obtain all information from within the
company it is their responsibility to share this with the supervisory board. All important
information concerning firm strategy, future projects, business opportunities etc. lies in the
hands of the management board (Jungmann, 2007). The management board must not
withhold any crucial information from members of the supervisory board.
11
The management board is able to minimize the information asymmetry with the supervisory
board by periodically reporting to them on specific matters (Block and Gerstner, 2016). The
reports of the management board must contain all information needed to evaluate their
work by including information concerning strategy and future directions (Jungmann, 2007).
The specific matters that are important to report on are described by du Plessis (2004) as
follows:
• The intended policy of the corporation and fundamental questions regarding the
planning of the undertaking, in particular regarding finance and investments
• The profitability of the firm and the return on its own capital
• The performance of businesses and in particular their turnover and the financial state
of the firm
• Transactions that are of vital importance for the corporation’s profitability
It can be stated that it is important to report all issues regarding planning, business
development and risk management. The matters the management board reports on also
have specific periods of time in which the reports must be given to the supervisory board,
making the flow of information sharing within the company efficient and effective (du
Plessis, 2004).
2.3.2. The Supervisory Board

The supervisory board within the two-tier board structure in Germany exists of non-
executive directors. Members of the supervisory board usually are appointed during the
annual meeting of the company. They can represent several different parties such as
shareholders, employees, labor unions etc., and are given responsibilities and authorities to
protect the interests of those parties. In Germany, there is an important aspect in forming
the supervisory board, referred to as codetermination. With codetermination laws,
employees are given the right to choose or vote to appoint members to the supervisory
board (Block and Gerstner, 2016). The number of employee representatives on the
supervisory board depends on the total number of employees a firm has. For German firms
who have between 500 and 2000 employees, it is required to have one-third employee
representatives on the supervisory board. For companies with more than 2000 employees, it
is required that half of the supervisory board members are formed by employee
representatives (Bozhinov et al., 2021). The supervisory board members are thus formed
and chosen by different parties and are being directly elected by the shareholders or
employees.
The main responsibilities of the supervisory board, as the name refers to is to supervise,
monitor and evaluate the management board´s decisions and their performance. The
supervisory board is given the authority to form the management board by appointing or
removing executives (Fauver and Fuerst, 2006). Appointing executives by the supervisory
board to the management board is usually done through establishing nominating
committees. These committees identify potential candidates for executive positions on the
management board and propose them to the supervisory board who are thus given the
authority to make the final decision (Bozhinov et al., 2021). This makes the members of the
supervisory board the final decision makers regarding the question of whom will be running
the daily operations of the company.
12
The supervisory board is able to evaluate and review the management board by inspecting
the books, reviewing the annual report and overseeing the work of an external auditor
(Block and Gerstner, 2016). In addition, Fauver and Fuerst (2006) describe some specific
tasks within the supervisory board, namely reviewing the executive’s performance, selecting
a chief executive officer, setting executives’ salaries and long-term strategic planning. It is
not possible for the supervisory board to directly interfere with the management of the firm,
but they can address some actions that can only be performed with their permission. Other
ways of influencing management could for example be setting incentives through
remuneration or provide regular advise on strategic decisions (Block and Gerstner, 2016).
Reviewing the responsibilities and tasks of the supervisory board it can be stated that the
focus of the supervisory board is shifting towards advising, controlling and supervising the
management board in order to achieve better firm performance.
A key factor to accomplish abovementioned tasks and responsibilities is the communication

with the management board. As mentioned in the previous subsection, there is a situation of
information asymmetry between the supervisory and management board because the
supervisory board is not permitted to collect information from employees or junior
management on their own (Jungmann, 2007). A possible problem arising from this is the lack
of insider knowledge by the supervisory board since employees do not have the obligation
to report directly to the supervisory board (Fauver and Fuerst, 2006; Block and Gerstner,
2016). This makes the communication and information sharing with the management board
even more important because the members of the supervisory board receive all necessary
information on the firm from the executives on the management board (Fauver and Fuerst,
2006; Jungmann, 2007).
The supervisory board usually has several options to reduce the information asymmetry.
They, for example can choose to hire former members of the management board that will
report to them, but in this case there is a limited number of former members allowed.
Another option is to inspect all reports and documents from the management board in
person or define which data to collect or when exactly the management should deliver their
reports. Yet another option is simply request the members of the management board to join
the board meetings set by the supervisory board to inform them on important subjects
(Block and Gerstner, 2016).
2.4. Board Characteristics

2.4.1. Board Size
Board size is a major characteristic of the board of directors, and determining the optimal
size is an important task for companies (Graf and Stiglbauer, 2009). Board size refers to the
total number of members on the board of directors, who are either involved in the daily
operations of the firm or in the supervisory role. From a corporate governance perspective,
it has been an ongoing question what the ideal board size is for a firm, and since its
importance, the relationship between board size and firm performance has been a focus
point of many studies (Fiss, 2006; Bermig and Frick, 2010; Darmadi, 2011). Previous research
present mixed findings of companies with large and small boards and the reasoning for the
choice of a larger or smaller board also varies (van Ees, Postma and Sterken, 2003; Hidayat
13
and Utama, 2016). This makes it even more difficult for researchers and companies to come
to a conclusion on whether larger or smaller boards are better for firm performance.
The covered theories regarding corporate governance can present different insights or
approaches to board size, depending on the interpretation. For instance, agency theory
mainly describes that smaller boards should be considered over larger boards, since the
increase in board members causes an increase in the agency costs and efficiency of the
board declines which eventually results in poor firm performance (Li, Lu, Mittoo and Zhang,
2015). The smaller number of board members also decreases information asymmetry and
reduces conflicts between members contrary to larger boards which can experience greater
levels of conflict (O’Connell and Cramer, 2010). On the contrary, resource dependence
theory insists that the more members on a board the more available resources for the
company. According to this theory, larger board size can increase firm performance through
sharing of knowledge and experience in the decision-making process (Latif, Kamardin, Mohd
and Adam, 2013). Stewardship theory also supports this view, since the focus of this theory
is that every manager or employer is a good steward of the company meaning their interests
are aligned with those of the owners. So it will not matter how many member there are on
the board, they all will act with loyalty and professionalism towards the company. Having
these mixed ideas from different theories makes the decision on a specific board size even
more difficult for companies.
Adding to this, in Germany as mentioned there is the separation between a management

board and a supervisory board, meaning that German firms have to decide on the number of
members for both boards. For the supervisory board, there are some legal rules and
regulations that determine the size and influence the formation of the board (Graf and
Stiglbauer, 2009). In Germany it is required that the supervisory board has a minimum of
three members and the maximum is up to 21 members, the total number eventually
depends on firm size and number of employees (Boneberg, 2010; Block and Gerstner, 2016).
It is common for German firms with fewer than 2000 employees to have 6 supervisory board
members, 12 members for firms with between 2000 and 10000 employees, and over 16 up
to 21 members for firms with more than 10000 employees (Fiss, 2006). Adding a rule to the
formation is that when one-third of the supervisory board members is elected by the
employees through codetermination, it is required that the total number on the supervisory
board is dividable by three (Boneberg, 2010).
In the case of the management board there is no maximum or minimum amount of

members allowed or required according to law or agreements. The size of the management
board usually depends on the decision of the owners and supervisory board who are in
charge when forming the management board. Thus, it depends on supervisory directors’ and
firm’s characteristics (Bermig and Frick, 2010).
2.4.2. Board Independence

The independence of a board usually refers to the non-executive directors on the board,
who are mainly formed by outsiders of the company. According to Jungmann (2007) a
director is considered independent if there exists no relationship or circumstance which
could affect the director’s judgement. These directors can be considered as a control
mechanism because they are more objective than managers and can provide new
14
viewpoints focused on firm performance (Martinez and Alvarez, 2019). Independent
directors can be better in achieving the supervising task and prevent misuse of company
resources and hereby improve performance (Makhlouf et al., 2017). The presence of
independent board members is also important in aligning the interests of the owners and
managers, which is the core function of the board of directors. There has always been some
kind of fear or mistrust from the owners towards the managers, that they would chase their
own interests rather than that of their firm. Owners believe that independent directors are
able to align their interests with the firm, making the appointment of independent directors
on the board a crucial aspect of this internal governance mechanism.
Because of their independence and objectiveness, independent directors are considered to

be better at assessing firm’s management and are more likely to make better decisions that
are beneficial for firm performance (Kao et al., 2019). Independent directors make this
possible because control is exerted through someone who is independent from the daily
operations of the firm. In the two-tier structure these directors are members of the
supervisory board. In the German two-tier board structure there is thus the clear separation
between the executive directors on the management board and the non-executives on the
supervisory board. However, within the supervisory board of German firms it is not possible
to refer to every non-executive member as independent. The reason for this is the
codetermination law that can apply for German firms, where employee representatives
obtain a place on the supervisory board. These directors are usually chosen from within the
company, and thus cannot be considered as outsiders or independent from the firm (Block
and Gerstner, 2016). As previously mentioned, the number of these members on the
supervisory board depend on the total number of employees of a firm. Therefore,
independence in this study refers to the non-executive directors on the supervisory board
that are no employee representatives.
The characteristic of board independence can also be analyzed using the theoretical
perspectives. According to the agency theory, independent directors will promote better
firm performance by effective monitoring of executives and making sure that interests
between management and shareholders are aligned, resulting in reduced agency costs
(Ramdani and Witteloostuijn, 2010; Kao et al., 2019). A supporting view is given by the
resource dependence theory which claims that independent directors can increase value
creating activities with their knowledge outside the firm and improve performance through
strategic decision-making (Martinez and Alvarez, 2019).
2.4.3. Gender Diversity

Gender diversity on the board of directors has especially been a relevant topic in recent
years and the focus point of many studies. The argument whether more female directors
should be involved in firm strategies got the attention in public debates, academic research
and corporate strategy (Marinova, Plantenga and Remery, 2010). For many years the debate
around gender diversity on boards was considered only as a social issue, but nowadays it is
more perceived as an aspect of increasing value for corporate governance and firm
strategies (Marinova et al., 2010). In general it is believed that the presence of female
directors on boards bring a fresh viewpoint and new perspectives, which could be beneficial
for firm performance (Bennouri et al., 2018; Terjesen et al., 2016).
15
Having more gender diverse boards could be an advantage for firms because women may be
better in understanding particular market conditions than men, this may bring more
creativity and quality to board decision-making (Shukeri, Shin and Shaari, 2012). Increasing
the number of female directors on the board makes it possible to take decisions considering
a wider range of alternatives (Marinova et al., 2010). According to Martinez and Alvarez
(2019) women are able to have a better impact on decision-making when there is more than
one woman present on the board, since this makes them feel stronger and encourages them
to have a stronger voice within the board. In addition to this, having a higher presence of
women on boards also benefits the public image of the firm, which eventually can result in
better firm performance (Marinova et al., 2010).
Theories concerning corporate governance also present ideas on gender diversity on the
board of directors, which are usually positive towards increasing the number of female
directors. Agency theory for instance states that female directors on boards contribute a
wider range of perspectives in decision-making and they can act as a mechanism of control
for board activities (Martinez and Alvarez, 2019). This is mainly because female directors are
more likely to ask questions, debate issues and hold their firm to higher ethical standards
compared to men (Terjesen et al., 2016). These characteristics make the presence of female
directors especially on the supervisory board more important. Having more gender diverse
boards is from the resource dependence and human capital theory perspectives also
beneficial to firm performance. Female directors contribute new skills and valuable
resources, therefore they have to be considered as an essential link between the firm and
external resources, and compared to male directors they have more diverse networks
(Terjesen et al., 2016; Martinez and Alvarez, 2019).
In accordance with the public debate on gender diversity, the topic got the necessary
attention in Germany. The country introduced an act in 2016 which requires that at least
30% of the supervisory board members of publicly listed firms has to be formed by female
directors (Bozhinov, Koch and Schank, 2017). Since the introduction of this gender quota,
the effects have showed, namely the presence of women on supervisory boards increased
from 3% to 28% in 2018 for the largest companies (Weidler, 2020). Similar to the
characteristic of board size, there is no specific law that determines the number of female
directors on the management board. Again, this depends solely on the characteristics of the
firm and the supervisory board. Whether this attention on the topic and the necessary
changes in the formation of the board has an impact on firm performance is a question that
has to be answered.
2.5. Empirical Findings

This section will include a look on empirical findings on the board characteristics and firm
performance. It is valuable to have a look at different studies since findings can be mixed
throughout different researches. Since this study focuses on German listed firms that use a
two-tier board structure, findings will focus on Germany and other countries that adapt a
two-tier structure. This will be helpful in forming the hypotheses for the different board
characteristics by analyzing results from countries with a similar corporate system.
16
2.5.1. Board Size and Firm Performance
An important study regarding the characteristic of board size is conducted by Bermig and
Frick (2010) who focused on supervisory board size in German listed firms and firm
performance. For their sample of 294 firms in the period of 1998-2007 they analyzed
supervisory board size with both market-based and operating performance measures.
Contrary to their expectations, they find a significantly positive relation between board size
and Tobin’s Q, and a non-significant result for their operating performance measures.
Another study conducted by Martinez and Alvarez (2019) focused on a wider scale by
conducting an international study with firms from 34 countries, including Germany, in their
research. They believe that these results will give a great mix of the different corporate
governance systems and provide a general view on board size and its influence on firm
performance. In their analysis they conclude that board size has a positive influence on firm
performance, meaning that larger boards are beneficial for firms. Martinez and Alvarez
(2019) believe this is explainable with resource dependence and human capital theory since
having more board members results in greater knowledge, expertise and more available
resources.
Contrary to the positive relation between board size and firm performance, there are also
studies that find negative relations. An example is the study conducted by Graf and
Stiglbauer (2009), who focus on the size of the German management boards. Analyzing data
from over 100 German firms and their management boards, they conclude that board size
has a negative relation with firm performance. The main reason they present is that
increasing the size of the board lowers efficiency and reduces decision quality because there
are more ideas and perspectives, which makes it take longer to reach consensus. Supporting
these findings is the study of Hidayat and Utama (2016) who focus on Indonesia, which is
another country with the two-tier structure. They conclude that firm performance decreases
as board size increases. Van Ees et al. (2003) focus specifically on the supervisory boards of
Dutch listed firms and observe that size has a negative impact on firm performance. They
believe that the main reason for this is the increasing number of outsiders on the
supervisory board that do not obtain the necessary information on the firm. Yet another
country with the two-tier system is Taiwan where Kao et al. (2019) present findings of a
negative relation between board size and firm performance. Contrary to their hypothesis,
firm performance decreases as board size increases, mainly due to the inefficiency of
decision making and the lack of consensus within the larger boards.
2.5.2. Board Independence and Firm Performance

In their research, Martinez and Alvarez (2019) conclude that a higher presence of
independent directors on the board has a positive impact on firm performance. The
monitoring and supervising abilities of independent directors is in their opinion the main
reason for this positive impact. In line with these findings and the importance of monitoring,
Kao et al. (2019) show that firm performance increases significantly after an increase in the
proportion of independent directors. Another country specific study of Abidin and Jusoff
(2009) also shows that as the percentage of independent non-executive directors on the
board increase, the performance of Malaysian firms increases as well. Next to the
monitoring and supervising benefits of independent directors, Abidin and Jusoff (2009) point
out that these directors also increase the total resources of the firm which eventually results
in greater performance.
17
The Indonesian study conducted by Hidayat and Utama (2016) also finds a significant
positive correlation in the proportion of independent directors to the performance measures
of Tobin’s Q, return on equity and return on assets. This study focuses on the supervisory
role and interest alignment tasks of the independent directors as well. An interesting finding
is that independent directors who served the company less than nine years show a higher
positive significance. According to the study, this indicated that directors cannot be
considered as independent when they served the company for longer than nine years or
were once employees of the company (Hidayat and Utama, 2016). A different approach is
taken by Gani and Jermias (2006) who studied the international manufacturing industry by
categorizing firms according to their strategy, and interesting findings have been presented.
Firms pursuing a strategy of cost efficiency show a positive relation between board
independence and performance. Again, this is a result of the benefits of the monitoring and
supervising tasks of independent directors.
On the contrary, Gani and Jermias (2006) conclude that firms with the main objective of
innovation show a negative relation with board independence and performance. These firms
require the strategic management roles of their directors rather than a monitoring or
supervising role which independent directors mostly provide. This shows that all firms can
have different relations between board independence and performance depending on their
strategies. Ammari, Kadria and Ellouze (2014) investigated a sample of French firms and
present a negative impact of board independence on return on assets and Tobin’s Q.
2.5.3. Gender Diversity and Firm Performance

Joecks, Pull and Vetter (2012) conducted a valuable research to the gender diversity on the
supervisory boards of German firms. They find that gender diversity on the board enhances
firm performance when at least 10 percent is formed by female directors, and boards with
more than 30 percent female representation even perform better than male dominant
boards. They also conclude that when female representation is lower than 10 percent it
might even be associated with reduced firm performance. The international study conducted
by Martinez and Alvarez (2019) find that a higher presence of female directors on the board
increases firm value and performance. They explain this positive relation with the
perspectives of the agency and resource dependence theory, mentioning that female
directors can provide different resources and skills compared to their male colleagues. This is
supported by Green and Homroy (2017) who studied the largest European firms and
conclude that greater female representation on the board, especially when it is integrated in
the governance mechanism of the firm, results in greater firm profitability.
A country specific study conducted by Smith, Smith and Verner (2005) did research on a
sample of 2500 Danish firms presenting findings of a positive relation between gender
diversity and firm performance. Smith et al. (2005) point out a very important factor,
mentioning this positive relation exists when the female directors on the board are elected
by the staff without any ties to the owners of the firm. With another country specific study,
Lückerath-Rovers (2013) examined the link between female directors and return on equity in
the Netherlands. The findings in the study show that companies with female directors on the
board have greater firm performance compared to companies without female directors.
Lückerath-Rovers (2013) mentions that involving female directors on the board is beneficial
18
because companies then make use of the whole talent pool for competent directors. This is
also motivating for other female employees at the company, because they see and believe
that they also can reach the top.
Darmadi (2011) presents other findings in the research to gender diversity and firm
performance in Indonesia. The researcher concludes that a higher proportion of female
board members is associated with lower firm performance. Similar negative findings are
presented by Soare, Detilleux and Deschacht (2021), who focused on Belgium firms in their
research. Both researches describe several factors that could cause these negative impacts.
Firstly, a higher proportion of female directors could lead to over monitoring which can have
an impact on reaching consensus and slow down the decision-making process (Darmadi,
2011). Secondly, female directors tend to focus more on the long term performance of their
firms rather than short term results, therefore it is important to consider the effects of
female directors on long term firm performance. Lastly, women are often still the minority
on the board of directors, and in order to reap the benefits of gender diversity a change in
company culture and mentality is required which will take some time (Soare et al., 2021).
2.6. Hypothesis Development

In this section the hypotheses on the board characteristics for this study will be formed. This
will be done based on the described theories and the empirical findings.
2.6.1. Board Size

When studying the theories it is possible to say that they present different perspectives on
board size and its impact on performance. This makes it not easy to predict whether smaller
or larger boards will be beneficial for firms because all theories are valuable and should be
considered when forming hypotheses. For instance, according to the agency theory, larger
boards increase agency costs and the problem of free riding among the many board
members arises which eventually will negatively affect firm performance (Kao et al., 2019).
In addition, larger boards can reduce efficiency and performance since agreeing on decision
becomes more difficult. On the other hand, resource dependence theory present another
view on board size, considering that larger boards bring more professionals with great
resources thanks to their backgrounds (Makhlouf et al., 2017). Therefore, it is possible that
more members are able to contribute their knowledge which can improve firm performance
(Martinez and Alvarez, 2019).
Previous findings present mixed findings as well, some researches are in favor of the agency
theory while other studies find support for resource dependence or human capital theory.
However, it is possible to say that previous studies are mostly in favor of the agency theory,
presenting results that smaller boards have a positive impact on firm performance. The
primary reasons that smaller boards perform better compared to larger boards are that
decision making, communication, interaction and coordination becomes more efficient (Graf
and Stiglbaur, 2009; Makhlouf et al., 2017; Kao et al., 2019). Regarding the resource
dependence and human capital theory, even if the board is smaller it is expected from the
present board members to be capable individuals with resources beneficial for the firm.
Therefore, the disadvantages of the agency theory for larger boards outweigh the resource
dependence and human capital perspectives in this case. Considering the theories and
findings from previous studies Hypothesis 1A is formed as follows:
19
Hypothesis 1A: smaller board size has a positive impact on firm performance
Based on this main hypothesis for board size, it will also be interesting to do research on the
management and supervisory boards separately, therefore the next hypotheses have been
formed:
Hypothesis 1B: smaller management board size has a positive impact on firm performance
Hypothesis 1C: smaller supervisory board size has a positive impact on firm performance

Independence is a difficult board characteristic to make a prediction on whether it will be
positively or negatively linked with firm performance. There are different opinions on
independence, some are in favor of increasing the number of independent directors and
some have opposing ideas. In theory, independent directors are mostly considered for the
supervisory task within the board because they are not directly linked with management.
Owners tend to have the idea that there is a need for independent directors that have their
best interests in mind, which is in line with the agency theory (Makhlouf et al., 2017). Since
these directors are outsiders of the company they can add valuable resources that other
directors do not have access to (Martinez and Alvarez, 2019). On the contrary, it is discussed
that independent directors are not able to add value in terms of resources because they are
outsiders and possibly do not own all information, expertise or knowledge on the daily
operations and needs of the firm (Ammari et al., 2014).
Evidence from previous research mainly supports the positive impact of board independence
on firm performance. The foremost reasons are in line with the theories, suggesting that
independent directors have a positive impact on firm performance because of their effective
monitoring and supervising activities. Their access to valuable extra resources is also a
positive factor. This leads to the next hypothesis on board independence:
Hypothesis 2: greater proportion of independent directors on boards has a positive impact on

firm performance

In general all theories have positive perspectives on gender diversity on boards and describe
that it can have a positive impact on firm performance. The agency theory’s focus is mainly
aligning the interests between managers and owners by delegating the monitoring and
supervising tasks to board members. According to the agency theory, female directors can
be highly valuable in this task since they can act as a control mechanism for the board
because they hold other standards compared to men (Martinez and Alvarez, 2019; Terjesen
et al., 2016). Also from the perspective of the resource dependence theory a greater
proportion of female directors on boards has benefits for the firm. Female board members
contribute valuable resources firms need with their diverse networks which their male
colleagues cannot provide (Martinez and Alvarez, 2019). Supporting this view is the human
capital theory, since this theory states that every individual should be considered as valuable
to a firm since they all have different expertise, knowledge and backgrounds (Terjesen et al.,
2016).
20
As presented in the previous section, many studies have findings in favor of gender diversity
on boards (Joecks et al., 2012; Green and Homroy, 2017; Martinez and Alvarez, 2019). These
studies describe the advantages of including more women on the board of directors and
how this is impacting firm performance positively. These factors are all in line with the
theories and focus on the increased resources, viewpoints and new perspectives with
involving more female board members. With these results, female directors send a positive
signal to the public and they should be considered as valuable members of the board.
Considering the theoretical perspectives and previous findings, Hypothesis 3A is formed as
follows:
Hypothesis 3A: greater proportion of female directors on boards has a positive impact on
firm performance
As previously mentioned, Germany inducted an act in 2016 with a gender quota which states
that at least thirty percent of the supervisory board of publicly listed firms should be formed
by female directors. The management board on the other hand does not have a specific
quota. It will be interesting to do research on whether this quota has its impact on
performance in reality. Therefore, the next hypotheses have been formed:
Hypothesis 3B: greater proportion of female directors on the management board has a
positive impact on firm performance
Hypothesis 3C: greater proportion of female directors on the supervisory board has a positive
impact on firm performance
21
3. Research Methodology
The research method is a very important factor in a study, since it is the core of presenting
results, form conclusions on the hypotheses and helps in answering the research question. In
studies with multiple dependent, independent and control variables the most used method
is a regression model. Similar studies doing research on the board characteristics and firm
performance mainly use ordinary least square regression (OLS), fixed effects regression (FE),
or two-stage least squares regression (2SLS). Some researchers even use two or more of the
mentioned methods in their studies in order to present the most precise findings. First,
these different regression methods from previous studies will be described, and then the
method for this study will be chosen and the model will be presented. This section also
presents the dependent, independent and control variables and describes their
measurements.
3.1. Methods in Previous Studies

3.1.1. OLS Regression
The most popular method that is used in studies doing research on board characteristics and
firm performance is the OLS regression method (Smith et al., 2005; Graf and Stiglbauer,
2009; Darmadi, 2011; Joecks et al., 2012; Lückerath-Rovers, 2013; Duru et al., 2016; Green
and Homroy, 2017; Bennouri et al., 2018; Martinez and Alvarez, 2019; Soare et al., 2021).
OLS regression estimates the relationship between one or more independent variables and a
dependent variable, more on the variables of this study in the next sections. The estimated
equation is calculated by determining the equation that minimizes the sum of the squared
distances between the sample´s data points and the values predicted by the equation
(Farahani, Rahiminezhad, Same and Immannezhad, 2010; Poston Jr., 2021). It is very useful
as a multivariate model with multiple independent variables, a dependent variable and the
possibility to add control variables. According to Graf and Stiglbauer (2009) OLS is a
traditional method which delivers efficient, unbiased and consistent estimates. The
statistical method of OLS regression requires some assumptions that should be met before
conducting the regression analysis. These assumptions are linearity, random sampling of
observations, no multicollinearity, homoscedasticity and no autocorrelation (Farahani et al.,
2010; Albert Resources, 2020). These assumption can be checked prior to the regression.
3.1.2. Fixed Effects Regression

Another popular method used in prior studies is the fixed effects regression (Smith et al.,
2005; Bermig and Frick, 2010; Hidayat and Utama, 2016; Green and Homroy, 2017;
Makhlouf et al., 2017; Bennouri et al., 2018; Kao et al., 2019). FE is used to explore the
relationship between independent and dependent variables within an entity, where each
entity could influence the dependent variables (Torres-Reyna, 2007). FE regression is mostly
used in studies with panel data which means that observations are collected over a specific
time series for a specific sample (Best and Wolf, 2015). This regression method is used to
reduce selection bias in the estimation of causal effects in data with a requirement that
several assumptions are met (Best and Wolf, 2015; Mummolo and Peterson, 2018).
According to Kao et al. (2019) the FE model assumes that the individual heterogeneity is
associated with independent variables, different to the random effects model. An important
test in determining whether FE regression is appropriate for the study is the Hausman test
(Smith et al., 2005; Hidayat and Utama, 2016; Makhlouf et al., 2017).
22
3.1.3. 2SLS Regression
The 2SLS regression analysis is statistical method that is used in the analysis of structural
equations. It can be seen as the extension of the OLS regression and is mainly used as a
complementary method in studies (Kao et al., 2019; Marinova et al., 2019). Standard linear
regressions assume that errors in the dependent variable are not correlated with the
independent variables. However, this is not always the case and the 2SLS regression can
then be used. Therefore, the 2SLS regression is mainly used when the dependent variable’s
error terms are correlated with the independent variables (Statistics Solutions, 2021).
Because the computed values are based on variables that are not correlated with the errors,
it is argued that the results of the 2SLS method are optimal.
Also similar to the OLS method there are some assumptions that should be met if the 2SLS
regression will be used. First the model has to be correctly identified, the error variance of
all variables should be equal, error terms have to be normally distributed, outliers need to
be removed and observations should be independent (Statistics Solutions, 2021). The 2SLS
regression can also be used in order to control for endogeneity issues between the variables
(Kao et al., 2019).
3.2. Method in this Study

Similar to previous studies, this study includes multiple dependent, independent and control
variables. Therefore a multivariate regression model will be the most appropriate method to
test the hypotheses of this study. As mentioned in the previous section, the most used and
popular model is the OLS regression. This method will also be used in this study because
comparable studies researching different board characteristics and its impact on firm
performance use this method as well. OLS regression presents results that are easy to
understand and analyze. The hypotheses in this study are formed in order to test the impact
of the total board or management and supervisory board separately. Therefore, depending
on the hypothesis tested, either observations on total boards or management and
supervisory boards will be used and included in the model. The model that will be used in
this study is as follows:
PERFit = β1BSIZEit + β2BINDEPit + β3FEMDIRit + β4SIZEit + β5LEVit + β6AGEit + β7YEARit + β8INDit +

εit
Table 1: Model Codes Descriptions

Code Description
PERFit Performance of firm i in period t measured in
ROA, ROE and Tobin’s Q
BSIZEit Board size of firm i in period t
BINDEPit Board independence of firm i in period t
FEMDIRit Gender diversity (female directors) of firm i in
period t
SIZEit Size of firm i in period t
LEVit Leverage of firm i in period t
AGEit Age of firm i in period t
YEARit Year dummy
23
INDit Industry dummy
εit Standard error
3.3. Variables
3.3.1. Dependent Variables
The aim of this study is to analyze the impact of different board characteristics on firm
performance, which means that firm performance is the dependent variable in this study.
There are different measures of performance used through literature, mainly divided in
accounting-based and market-based measures. Performance in this study will be measured
by using three measures, two accounting-based and one market-based. The first accounting-
based measure that will be used in this study is return on assets, from now on referred to as
ROA. This is a very popular performance measure and is used in many studies regarding this
topic (Graf and Stiglbauer, 2009; Bermig and Frick, 2010; Kao et al., 2019). ROA is calculated
as the ratio of net income divided by total assets (Darmadi, 2011; Green and Homroy, 2017;
Makhlouf et al., 2017). The second accounting-based measure is another popular one used
in many studies namely the return on equity, referred to as ROE (Graf and Stiglbauer 2009;
Joekcs et al., 2012; Bennouri et al., 2018). The calculation for ROE is the ratio of net income
divided by the average shareholders’ equity (Lückerath-Rovers, 2013; Kao et al., 2019). The
third and final performance indicator is a market-based measure which is used almost as the
only market-based measure in existing literature namely Tobin’s Q (Bermig and Frick, 2010;
Marinova et al., 2010; Darmadi, 2011; Hidayat and Utama, 2016; Kao et al., 2019). Tobin’s Q
is calculated as market capitalization divided by total assets (Bermig and Frick, 2010;
Marinova et al., 2010; Kao et al., 2019).
3.3.2. Independent Variables

The independent variables in this study are the board characteristics since their impact on
firm performance will be analyzed. The first independent variable is board size of the
German firms in the sample. In a previous German study conducted by Bermig and Frick
(2010) board size is measured as the total members on the supervisory board, and the study
by Graf and Stiglbauer (2009) measured this as the total members on the management
board. This study will combine both the management and supervisory board since the aim is
to analyze the impact of the total board size first and then the management and supervisory
board separately. Therefore, board size is the sum of the total members from the
management and the supervisory board, and as it speaks for itself the size of the
management board and supervisory board are solely the members included in those specific
boards.
The second independent variable is board independence. Martinez and Alvarez (2019)
calculate board independence as the ratio between the total number of independent
members on boards and the total number of board members. Kao et al. (2019) use the same
approach, only difference is that in their study a director is considered independent when all
criteria on independence as stated in the regulations of the country are met. Slightly
different approach is taken by Gani and Jermias (2006) who use the ratio of external
members to total members on the board of directors. In this study it should be considered
that German supervisory boards are different compared to other countries with a two-tier
24
board structure because of the codetermination laws. As previously described, depending on
the number of employees, German supervisory boards include employee representatives
from within the firm. Block and Gerstner (2016) describe that these representatives cannot
be considered as truly independent because of their ties with the company, therefore
independence in this study refers to the members on the supervisory board who are no
employee representatives. Board independence is thus measured as the ratio of supervisory
board members who are no employee representatives to the total number of supervisory
board members.
The third and final independent variable is gender diversity on the board of directors, which
can be measured differently in previous studies. Marinova et al. (2010) use two approaches
to measure gender diversity, the percentage of women on the board (management plus
supervisory board) and the method of a dummy variable indicating whether there is at least
one female director on the board or not. A similar dummy approach is used by Joecks et al.
(2012) in their study to gender diversity on German boards, creating four dummy variables
reflecting different group types ranging from no woman on boards to at least 40 percent
women on boards. It should be mentioned that this study was prior to the gender quota of
30% female representation introduced in Germany in 2016. This particular measurement of
gender diversity is not applicable for this study because every firm is required to have female
representation on boards. The more recent studies focus on the method of proportion or
percentage women on boards, using a ratio of the total number of female directors to the
total number of board members (Lückerath-Rovers, 2013; Green and Homroy, 2017;
Martinez and Alvarez, 2019). Therefore, gender diversity in this study is calculated as the
ratio of female directors on the board (management and/or supervisory board) to the total
number of board members (management plus supervisory board).
3.3.3. Control Variables

When analyzing whether the independent variables have an impact on performance it is
important to consider other factors that can influence performance. These other factors are
the control variables that must be included in the model. There are several control variables
that are often being used in many studies regarding this topic. The first control variable is
the size of the firm, because larger firms can easier generate funds and gain access to funds
from external sources, which can impact performance positively (Kao et al., 2019). Firm size
is measured as the natural logarithm of total assets (Bermig and Frick, 2010; Lückerath-
Rovers, 2013; Hidayat and Utama, 2016; Martinez and Alvarez, 2019). The second control
variable that is commonly used is leverage and is measured as the ratio of total debt to total
assets (Bermig and Frick, 2010; Kao et al., 2019; Martinez and Alvarez, 2019). A firm with a
high leverage ratio can negatively influence firm performance because they have more debt
and less ability to repay debt, therefore it is important to include it as a control variable
(Gani and Jermias, 2006; Kao et al., 2019; Martinez and Alvarez, 2019). The third control
variable is the age of the firm, since it can play a role in the activities that influence
performance. For instance, younger firms are new to the market, have less experience, and
are building up a market position which could affect their earnings (Smith et al., 2005). Older
firms on the other hand do have the experience, market position, and are better to obtain
the necessary resources. Firm age is measured as the number of years the firm exists in the
observation year (Marinova et al., 2010; Hidayat and Utama, 2016). The last two control
25
variables are year and industry dummies, which represent the observation year and firm
industry (Martinez and Alvarez, 2019)
Table 2: Variable Definitions

Variable Code Definition Sources
Dependent Variables
Return on assets ROA Ratio of net income divided Darmadi (2011);
by total assets Green and Homroy
(2017); Makhlouf et
al. (2017)
Return on equity ROE Ratio of net income divided Lückerath-Rovers
by the average shareholders’ (2013); Kao et al.
equity (2019)
Tobin’s Q TOBQ Market capitalization divided Bermig and Frick
by total assets (2010); Marinova et
al. (2010); Kao et al.
(2019)
Independent Variables
Board size BSIZE Sum of total members from Kao et al. (2019);
the management and Martinez and Alvarez
supervisory board (2019)
Management board size MBSIZE Number of directors on the Graf and Stiglbauer
management board (2009)
Supervisory board size SBSIZE Number of directors on the Bermig and Frick
supervisory board (2010)
Board independence BIND Ratio of supervisory board Block and Gerstner
members who are no (2016); Kao et al.
employee representatives to (2019)
the total number of
supervisory board members
Gender diversity FEMDIR Ratio of female directors on Lückerath-Rovers
the board to the total (2013); Green and
number of board members Homroy (2017);
Martinez and Alvarez
(2019)
Management board gender MBFEMDIR Ratio of female directors on Lückerath-Rovers
diversity the management board to (2013); Green and
the total members on the Homroy (2017);
management board Martinez and Alvarez
(2019)
Supervisory board gender SBFEMDIR Ratio of female directors on Lückerath-Rovers
diversity the supervisory board to the (2013); Green and
total members on the Homroy (2017);
supervisory board Martinez and Alvarez
(2019)
Control Variables
26
Firm size SIZE Natural logarithm of total Bermig and Frick
assets (2010); Lückerath-
Rovers (2013);
Hidayat and Utama
(2016); Martinez and
Alvarez (2019)
Leverage LEV Ratio of total debt to total Bermig and Frick
assets (2010); Kao et al.
(2019); Martinez and
Alvarez (2019)
Firm age AGE Number of years the firm Marinova et al.
exists in the observation year (2010); Hidayat and
Utama (2016)
Year YEAR Year dummy Martinez and Alvarez
(2019)
Industry IND Industry dummy Martinez and Alvarez
(2019)
27
4. Sample and Data
This chapter presents the sample of firms that is used in this study, and describes the
process of collecting the data from those firms. It also describes the industry classification of
the sample firms.
The sample in this research has been formed by making a selection of German public firms
from the Frankfurt Stock Exchange. The Frankfurt Stock Exchange is the largest in Germany,
and has been used in several previous studies (Bermig and Frick, 2010; Steger, 2017;
Bozhinov et al., 2021). In order to make the selection of firms in this study, the Orbis
database has been used. This database includes German firms from the four main indices of
the Frankfurt Stock Exchange, namely DAX, MDAX, TecDAX and the German Entrepreneurial
Index (GEX). The initial sample consisted of 104 German firms from the Frankfurt Stock
Exchange that were in the indices above. From this initial sample, 38 companies where listed
in DAX, 44 in MDAX, 9 in TecDAX and 13 companies in GEX. Data on these firms has been
collected for a period of three years, from 2017 to 2019.
Two methods have been used in this study to collect the necessary data from the firms. The
Orbis database has been used and the annual reports of all firms have been collected for the
period of three years. Orbis provides information on many aspects of firms and is in this
study used to collect data on several variables. Firstly, Orbis has been used to collect data on
the dependent variables – the performance indicators ROA, ROE and Tobin’s Q. ROA and
ROE were calculated by using the net income and Tobin’s Q by dividing the market
capitalization by the total assets.
Information on the control variables of firm size, leverage ratio and age of the firm are also
available in Orbis. Firm size is measured in total assets of the firm, which in Orbis was given
in US dollars. Since this study focuses on German firms, the total assets were converted to
Euro currency based on currency rates of the 31 st of December of each sample year. For
leverage, in addition to total assets data on total debt is collected from Orbis as well in order
to calculate the ratio. The founding years of the firms are given in Orbis, this has been used
to calculate the age for each sample year.
Additionally, in order to collect data on the independent variables of board characteristics,

the annual reports of the selected firms are manually collected for the years 2017 to 2019. In
each annual report a search has been made regarding the board characteristics. Every
annual report was unique in the manner of disclosing information on their management and
supervisory boards. Some firms used a very clear manner of presenting the members of the
management and supervisory board by using pictures of the members next to their names.
Other firms only presented the names of the board members without their pictures. The
presence (if applicable) and number of employee representatives on supervisory boards
were also given. For the independent variables board size and board independence this
method of collecting data was very efficient. For the firms who did not present pictures of
their board members in their annual reports, the board members’ names have been
searched on the internet to determine the gender. This approach was needed to collect the
necessary data for the gender diversity variable.
28
During the data collection process several firms have been removed from the initial sample
due to two reasons. In total 15 firms were removed from the sample, 11 companies due to
missing information on the variables and 4 companies which were founded after 2017,
causing a lack of data for the research period. Therefore, the final sample consists of 89
firms with data for three years.
4.1. Industry Classification

Industry dummies will be used as control variables in this study, therefore the firms in this
sample are classified in the industries they operate in. The industry classification of the
German firms in this sample is based on the NACE Rev. 2 codes. Initially there were 11
industries as presented in Table 3. Manufacturing is by far the largest industry with 45 firms
which is 50.56% of the sample. The information and communication industry is the second
largest with 12 firms, and the remaining firms are distributed over the other industries.
Hereby the representation differs, some have just 1 firm others have 3 or 5 firms. For the
analysis it is best to have industries with at least 10 firms each, therefore the firms are
reclassified based on their industries. Combining industries resulted in 4 industries as
presented in Table 3 that will be used to form the dummy variables in the analysis.
Table 3: Industry classifications

Industry Number of firms Reclassified industries
Manufacturing 45 45
Information and 12 12
communication
Mining and quarrying 1
Electricity, gas, steam and air 3
conditioning
Wholesale and retail trade 5
Transportation and storage 5 14
Financial and insurance 4
activities
Professional, scientific and 6
technical activities
Other service activities 1
Real estate activities 6
Arts, entertainment and 1 18
recreation
Total 89
29
5. Results
This section of the paper presents the results of the research. First the descriptive statistics
will be given. Second, the correlation analysis will be presented followed by the
multicollinearity test. Finally, the results of the regression analysis will be presented.
5.1. Descriptive Statistics

5.1.1. Dependent Variables
Looking at Table 4 for the accounting-based performance indicators, the average ROA in this
sample is 3.18% with a median of 4.95%, and for ROE the mean is 9.62% with a median of
11.91%. This means that both accounting-based measures are skewed to the left. Graf and
Stiglbauer (2009) have higher mean values for ROA and ROE in their study, with a mean ROA
of 6.84% and a mean ROE of 15.48%. Not all firms in the sample show a positive ROA or ROE,
the minimum ROA is -128.08% which is far from the mean value. This is also the case with
ROE, which presents a minimum of -196.67%. The best performing firms on ROA and ROE
have values of 26.03% and 58.51% respectively. Despite having firms with negative ROA and
ROE, having positive mean and median values indicates that the majority of German firms in
the sample have positive financial performances. The market-based measure Tobin’s Q has a
mean of 1.57 with a median of 0.93, indicating that this variable is skewed to the right. This
can be explained by the maximum for Tobin’s Q, which is 11.66 for a particular firm. The firm
with the minimum value has a Tobin’s Q of 0.09. Weidler (2018) used the same calculation
for Tobin’s Q, and has some differences in the findings for the sample period of 2012-2018
with a minimum Tobin’s Q of 0.01, a maximum value of 6.66 and a mean of 0.89.
5.1.2. Independent Variables

Board size
For the total board size, the members of the management and supervisory boards are added
together. German firms in this study have an average board size of 15.46 with a median of
15. Total board size is quite diverse in this sample, with the smallest board consisting of 5
members whereas the largest board has 33 members. Analyzing the values of the
management and supervisory board will further explain how the total board size is
composed. For the management board in this sample the minimum is 2 members, and firms
with the largest management board have 10 members. On average, German firms in the
sample have 4.38 management board members. Graf and Stiglbauer (2009) and Bozhinov et
al. (2021) have similar findings in their study to German management boards, they find a
maximum size of 11 and 12 respectively with means of 4.43 and 4.55. Supervisory boards are
usually larger than the management boards, which is also the case in this study. The
supervisory boards have a mean of 11.08 members with a median of 12. The firm with the
smallest supervisory board has 3 members, and the largest supervisory board consists of 23
members. Bozhinov et al. (2021) have comparable findings for supervisory board size in their
study of 95 German firms for the period of 2009-2016, showing a minimum of 6 and
maximum of 21 members with a mean size of 13.48.
Board independence
For board independence it is important to have a look at the number of employee
representatives on the supervisory board and what this means for the total composition of
the supervisory board. Codetermination is not applicable for every firm in the sample,
30
therefore some firms have no employee representatives on the supervisory board at all. This
was the case in 36% of the observations, as can be seen in Appendix B. On average the
supervisory boards have 4.33 employee representatives, and the remaining members are
considered as the independent members, which has a mean of 6.75 per supervisory board.
The majority of the observations, with a percentage of 40.4 have 6 independent supervisory
board members. The maximum representation of employees on a board is 12, and for the
independent members the maximum amount is 13. As mentioned, not every board has
employee representatives, therefore all supervisory board members on those specific cases
are considered as independent members, 3 members is hereby the minimum.
Looking at the mean values of employee representatives and independent supervisory board
members, it is possible to mention that on average the independent members on the
supervisory board outweigh the employee representatives. The board independence ratio of
0.70 confirms this, because this is computed by the ratio of independent members to the
total supervisory board. This means that in this study’s sample, the supervisory boards
consist of 70% independent members and 30% of employee representatives. This difference
is thus explained by the firms where codetermination is not applicable and have no
employee representatives on their supervisory boards. The minimum independence ratio is
0.48, meaning that supervisory boards are composed by half employee representatives and
half independent members. This is mostly applicable for larger firms with over 2000
employees, for which codetermination laws require that half of the supervisory board is
formed by employee representatives. As expected, the maximum board independence ratio
is 1, caused by firms without employee representation.
Gender diversity
German firms in the sample have an average female representation of 3.58 members on
their boards. Not all boards have women directors, and the firm with the highest
representation has 10 female members on its board. The mean for gender diversity ratio is
0.20, thus for the average board only one fifth consists of female members. The most gender
diverse board in the sample has a ratio of 0.43. Female representation on the management
and supervisory board differ quite a bit. On average there are 0.39 female members on the
management board, whereas the mean for the supervisory board is higher with a value of
3.19. Weidler (2018) has comparable findings for German supervisory boards for the period
of 2012-2018 with a mean female representation of 3.49. Bozhinov et al. (2021) on the other
hand present findings that show slightly lower female representation on the supervisory
board with a mean of 2.36. The same study also shows lower female representation on the
management board with a mean of 0.21.
For both the management and supervisory board there are firms without female
representatives. Especially with the management board this percentage is quite high,
namely 67.4% of the observations have no female members on their management boards at
all. This value is rather lower for the supervisory board, with a percentage of 12.4%. The
maximum representation of women on the management board is 2, which counts for 6.4%
of the observations. The mean gender diversity ratio for the management board is 0.07, with
a maximum ratio of 0.40. Supervisory boards have higher female representation with a
mean gender diversity ratio of 0.26. This means that the gender quota of 30% female
members that is required for the supervisory boards is not met, but the difference in this
31
sample is just 4%. The maximum female members on a supervisory board is 9, with a gender
diversity ratio of 0.60. Weidler (2018) found a gender diversity ratio of 0.23. The higher value
in this study could be a result of the gender quota which was introduced in 2016.
5.1.3. Control Variables

Next to the independent and dependent variables, this research also included several
control variables. Looking at the total assets of the firms, it is possible to say that the sample
exists of smaller firms and very large firms. This is because the minimum and maximum
differ in a great amount. The smallest firm in the sample has total assets of €3.5 million, and
the largest firm has an impressive total assets of €489 billion. The average total assets
presents a value of €25.6 billion (median = €4.6 billion). The second control variable is the
leverage ratio, which has a mean of 0.56 (median = 0.57). Another control variable is the age
of the firm, this variable is also quite interesting. On average firms had an age of 58.19 years
(median = 30), the oldest firms was 173 years and the youngest was 2 years old, meaning it
was founded in 2015. These variables show the diversity of firms in the sample, making it
more interesting to analyze.
Table 4: Descriptive Statistics

Variables N Mean Median Std. Minimum Maximum
Deviation
Dependent variables
ROA 267 3.18 4.95 12.68 -128.08 26.03
ROE 267 9.62 11.91 19.76 -196.67 58.51
Tobin’s Q 267 1.57 0.93 1.74 0.09 11.66
Independent variables
MB size 267 4.38 4 1.95 2 10
SB size 267 11.08 12 5.60 3 23
Total board size 267 15.46 15 6.86 5 33
SB employee 267 4.33 6 3.85 0 12
representatives
SB independent 267 6.75 6 2.29 3 13
members
Board independence 267 0.70 0.52 0.23 0.48 1
ratio
MB female members 267 0.39 0 0.61 0 2
SB female members 267 3.19 3 2.27 0 9
Total female members 267 3.58 3 2.60 0 10
MB gender diversity 267 0.07 0 0.10 0 0.40
ratio
SB gender diversity 267 0.26 0.3 0.12 0 0.60
ratio
Total gender diversity 267 0.20 0.22 0.01 0 0.43
ratio
32
Control variables
Total assets 267 25,646 4,630 6,429 3.511 489,466
Leverage ratio 267 0.56 0.57 0.18 0.09 0.96
Firm age 267 58.19 30 50.94 2 173
5.2. Correlation Analysis

The correlation analysis is the step prior to the regression analysis, whereby the correlation
coefficients indicate the strength of the relationship between two variables. A correlation
coefficient greater than zero means that the relationship between the variables is positive,
and a coefficient less than zero indicates a negative relationship (Gogtay and Thatte, 2017).
Hereby especially attention has to be paid to the independent and control variables that will
be put in the same model for the regression analysis. Highly correlated independent and
control variables in the same regression model could be problematic for the analysis.
Therefore a multicollinearity test will be conducted in the next section as well, paying close
attention to strong significant correlations. For this study Pearson´s correlations has been
used. The correlations of the variables are presented in Table 5. As shown in the table, the
accounting-based dependent variables ROA and ROE have a high positive correlation at the
0.01 significance level. On the other hand, the third dependent variable – the market-based
Tobin’s Q – shows no significant correlation with both ROA and ROE.
ROA has no significant correlation with the total board size and the management board size.
On the contrary, ROA shows a positive significant correlation with the supervisory board size
at the 0.05 level, with a Pearson correlation of 0.128. This indicates that an increase in the
number of supervisory board members leads to a greater ROA. Similarly, ROE also has a non-
significant correlation with management board size. However, both the total board size as
the supervisory board size show significant positive correlations, with values of 0.135 and
0.156. Thus, in the case of ROE both increases in members of the total board and supervisory
board have a positive influence. The third dependent variable Tobin’s Q has significant
negative correlations with total board size, management board size and supervisory board
size as well. All correlations are significant at the 0.01 level, indicating that larger boards
cause a lower Tobin’s Q value. Total board size has high positive correlations with the
management and supervisory board, this is as expected since the total board size is formed
by the management and supervisory boards. However, this strong correlation will not be
problematic since these variables will not be put in the same model for the regressions. The
management board size and supervisory board size also show positive significant
correlations at the 0.01 level with a coefficient of 0.549. These will be put in the same
model, therefore it is important to test for multicollinearity in the next section. In addition,
total board size, management board size and supervisory board size all show significant
positive correlations with firm size. This indicates that larger firms have more members on
their boards. The same can be stated for the leverage ratio and the age of firms, since these
correlations are all significant and positive for the board sizes as well.
Furthermore, both ROA and ROE show significant negative correlations with board
independence ratio at the 0.01 level, with Pearson correlations of -0.211 and -0.222
respectively. This means that an increase in the independence ratio negatively influences
33
ROA and ROE, and vice versa. On the other hand, Tobin’s Q has a significant positive
correlation with board independence ratio at the 0.01 level with a value of 0.354.
Interestingly, board independence ratio has negative significant correlations with the control
variables. All correlations are significant at the 0.01 level, firm size has a Pearson correlation
of -0.478. The Pearson correlations of leverage ratio and firm age are -0.226 and -0.399
respectively. This means that as firm size increases, the board independence ratio decreases
and vice versa. It is also the case for leverage ratio and firm age.
Additionally, gender diversity ratio has positive significant correlations with ROA and ROE at
the 0.01 level. This shows that greater gender diversity on boards has a positive impact on
both accounting-based performance indicators. Tobin’s Q has a significant correlation with
gender diversity ratio as well, though this relationship is negative. Contrary to the total
gender diversity ratio, the management board gender diversity does not have significant
correlations with ROA and ROE. Tobin’s Q again shows a negative significant correlation with
the management board diversity. Indicating that an increase in the management board
diversity causes a decrease in the Tobin’s Q value and the other way around. Gender
diversity ratio has positive significant correlations with management board diversity ratio
and supervisory board gender diversity, this is as expected since gender diversity ratio is
formed by management board diversity and supervisory board diversity. Management board
diversity ratio and supervisory board diversity ratio show a positive correlation at the 0.05
level, and is yet another relationship that needs to be tested for multicollinearity in the next
section. Furthermore, both gender diversity ratio and supervisory board gender diversity
ratio have positive significant correlations with all three control variables at the 0.01 level.
This indicates that larger firms, older firms and firms with higher leverage ratios tend to have
more gender diverse board. On the other hand, management board diversity ratio has a
positive correlation with firm size at the 0.01 level, and with leverage ratio at the 0.05 level.
It does not have a significant correlation with firm age.
Lastly, even if the dependent variables do not have an effect on multicollinearity it is

interesting to further analyze the correlations with the control variables. ROA and ROE have
similar relationships with the control variables, both show positive significant correlations
with firm size and firm age at the 0.01 level. This means that as firm size and firm age
increases it positively affects the performance of the firm. Where these performance
indicators differ is the correlation with leverage ratio. ROA shows no significance, whereas
ROE is significant and has a Pearson correlation of 0.134 at the 0.05 level. On the contrary,
Tobin’s Q has negative significant correlations with all three control variables, for firm size
and leverage ratio at the 0.01 level and for firm age at the 0.05 level.
34
Table 5: Correlations of Variables
1 2 3 4 5 6 7 8 9 10 11 12 13
(1) ROA 1
(2) ROE 0.872** 1
(3) Tobin´s Q 0.069 0.042 1
(4) Total board 0.100 0.135* -0.410** 1
size
(5) MB size -0.013 0.028 -0.247** 0.732** 1
(6) SB size 0.128* 0.156* -0.417** 0.971** 0.549** 1
(7) Board -0.211** -0.222** 0.354** -0.753** -0.388** -0.788** 1
independence
ratio
(8) Gender 0.159** 0.183** -0.356** 0.619** 0.366** 0.632** -0.598** 1
diversity ratio
(9) MB gender 0.044 0.078 -0.217** 0.408** 0.500** 0.326** -0.245** 0.429** 1
diversity
(10) SB gender 0.162** 0.165** -0.285** 0.503** 0.311** 0.508** -0.488** 0.927** 0.148* 1
diversity
(11) Firm size 0.258** 0.246** -0.351** 0.684** 0.630** 0.619** -0.478** 0.538** 0.404** 0.485** 1
(12) Leverage -0.020 0.134* -0.441** 0.377** 0.196** 0.394** -0.226** 0.246** 0.139* 0.195** 0.278** 1
ratio
(13) Firm age 0.191** 0.179** -0.150* 0.284** 0.254** 0.260** -0.399** 0.207** 0.108 0.206** 0.277** 0.057 1
Notes: ** correlation is significant at the 0.01 level (2-tailed); * correlation is significant at the 0.05 level (2-tailed)
35
5.3. Multicollinearity Test
Multicollinearity can occur when two or more independent or control variables in the same
regression model are significantly correlated with each other, whereby high multicollinearity
will cause a problem for the results of the regression analysis (Daoud, 2017). Therefore, the
significant correlations between the independent and control variables presented in the
previous section indicate a potential multicollinearity problem when put in the same model.
This means a multicollinearity test has to be conducted prior to the regression analysis. In
the case of a strong significant correlation a consequence for multicollinearity is that the
variance inflation factor (VIF) is inflated (Daoud, 2017). Variance inflation factor values of 5
or higher are considered as highly correlated and show signs of multicollinearity that is
problematic for the regression models (Daoud, 2017).
Table 6 presents the collinearity test for the models that are used in this study, and give the
variance inflation factors (VIF). For all variables it can be seen that the values are between 1
and 2. This means that there will not be a multicollinearity problem for all equations, since
the VIF values are below 5.
Table 6: Multicollinearity Test

Model 1 Model 2 Model 3 Model 4 Model 5
Total board size 2.068
MB size 1.796
SB size 1.931
Board independence 1.463

ratio
Gender diversity ratio 1.436
MB gender diversity ratio 1.201
SB gender diversity ratio 1.330
Firm size 1.909 2.050 1.368 1.516 1.651
Leverage ratio 1.171 1.193 1.101 1.100 1.092
Firm age 1.107 1.108 1.208 1.090 1.092
Notes: the variance inflation factors (VIF) are given.
5.4. Regression Analysis

This section presents the results of the regression analysis which will be helpful in answering
the hypotheses of this study. The OLS regression is conducted for all three dependent
variables, the accounting-based ROA and ROE, and the market-based Tobin’s Q. Table 7
shows the results for ROA for all five models, Table 8 presents the regression results for ROE
and Table 9 for Tobin’s Q. These results will be analyzed for the independent and control
36
variables, whereby the results for the independent variables will either confirm or reject the
hypotheses.
5.4.1. Board Size

The independent variable of board size was categorized in total board size, management
board size and supervisory board size. Total board size represents Models 1 in the regression
analyses, and management board size and supervisory board size are put in Model 2 for all
performance indicators. Hypotheses 1A, 1B and 1C were formed for the variable of board
size. These hypotheses stated that smaller boards will have a positive impact on firm
performance, expecting a negative relationship between the dependent and independent
variables. This means that it is anticipated that an increase in board size will decrease firm
performance.
Looking at Model 1 in Table 7, total board size shows a negative significant relationship at
the 0.05 significance level, with an unstandardized coefficient of -0.263 and a t-statistic of
1.675. This indicates that an increase in total board size decreases the accounting-based
performance indicator ROA. Total board size has a negative significant relationship with the
second accounting-based measure ROE as well. Hereby the unstandardized coefficient is -
0.343 with a t-statistic of -1.406 and a significance at the 0.1 level. Therefore it can be stated
that smaller boards have a positive impact on both ROA and ROE. Furthermore, total board
size shows a negative significant relationship with Tobin´s Q at the 0.01 level with an
unstandardized coefficient of -0.051 and a t-statistic of -2.256. The same can be stated for
the market-based measure, that smaller boards increase the value of Tobin’s Q which is
favorable for firms. Total board size has negative significant relationships with all
performance measures, whereby the significance level differs. This means that Hypothesis
1A is confirmed for ROA, ROE and Tobin’s Q as well.
Similar to total board size, management board size shows a negative significant relationship
with ROA as well, but with a stronger significance at the 0.01 level. The unstandardized
coefficient is -2.252 with a t-statistic of -4.441. This indicates that increasing the members on
the management board results in a lower ROA and that smaller management boards are
more favorable. Thus, smaller management boards in this sample have a positive impact on
ROA. The same conclusion can be drawn for the impact of management board size on ROE.
Hereby the relationship is negative and significant at the 0.05 level, with an unstandardized
coefficient of -2.682 and t-statistic of -3.355. Hypothesis 1B is confirmed for both
accounting-based measures, meaning that smaller management boards have a positive
impact on ROA and ROE. On the contrary, management board size has no significance with
Tobin’s Q meaning that it is not possible to confirm the hypothesis for the market-based
performance measure.
Supervisory board size has different findings for ROA and ROE. The results in Model 2 for the
supervisory board size show no significant relationship with both accounting-based
measures. Therefore, Hypothesis 1C is rejected for ROA and ROE, meaning that the
supervisory board size for firms in this sample do not have an impact on accounting-based
performance. On the other hand, Tobin’s Q does have a negative significant relationship
with supervisory board size at the 0.01 level with a coefficient of -0.065 and t-statistic of -
37
2.842. Smaller supervisory boards in this sample have a positive impact on the market-based
measure, thus Hypothesis 1C is confirmed for Tobin’s Q.

Hypothesis 2 states that a greater proportion of independent directors on boards has a
positive impact on firm performance. Model 3 in the regression tables represents the
regression analysis of the independent variable board independence ratio. First, looking at
Table 7 it can be concluded that board independence ratio has a negative significant relation
with ROA at the 0.1 level. In addition, the relationship with ROE is negatively significant as
well at the 0.05 level, with a coefficient of -10.611 and a t-statistic of -1.742. Thus an
increase in the number of independent board members will negatively influence both ROA
and ROE. The opposite was expected, therefore Hypothesis 2 is rejected for both accounting-
based performance indicators. On the contrary, the market-based measure shows a
significant positive relationship with board independence ratio at the 0.01 significance level.
Hereby the unstandardized coefficient is 1.595 with a t-statistic of 3.327. For Tobin’s Q it can
be stated that increasing the independent board members results in better firm
performance. Hypothesis 2 is confirmed for the market-based measure.

The independent variable gender diversity was categorized in gender diversity ratio,
management board gender diversity ratio and supervisory board gender diversity ratio.
Gender diversity ratio represents Model 4 in the regression analyses, and management
board gender diversity and supervisory board gender diversity ratio are put in Model 5 for all
performance indicators. Hypotheses 3A, 3B and 3C were formed for the variable of gender
diversity. These hypotheses stated that greater proportion of female members on the
boards will have a positive impact on firm performance, expecting a positive relationship
between the dependent and independent variables. This means that it is anticipated that an
increase in gender diversity ratio will increase firm performance.
The regression results show that gender diversity has a non-significant relationship with both
accounting-based performance measures. This is the case for management board gender
diversity and supervisory board gender diversity as well. This means that Hypotheses 3A, 3B
and 3C are rejected for ROA and ROE. On the other hand, gender diversity ratio shows a
negative significant relationship at the 0.01 level with Tobin’s Q. Indicating that an increase
in female board members decreases the Tobin’s Q value. Next to this, management board
gender diversity and supervisory board gender diversity have negative significant
relationships with Tobin’s Q as well, with significance levels of 0.1 and 0.05 respectively.
Contrary to the expectation of the hypothesis, the results show that an increase in gender
diversity ratio on the boards decreases Tobin´s Q and has a negative effect on performance.
Therefore, Hypotheses 3A, 3B and 3C are rejected for the market-based performance
measure as well.
Table 7: OLS Regression - ROA

Constant -33.885*** -38.630*** -16.616* -24.884** -27.352**
38
(-4.073) (-4.736) (-1.746) (-3.269) (-3.460)
Total board -0.263**
size
(-1.675)
MB size -2.252***
(-4.441)
SB size 0.136
(0.753)
Board -6.417*
independence
(-1.634)
ratio
Gender 7.249
diversity ratio
(0.795)
MB gender -7.141
diversity ratio
(-0.889)
SB gender 5.756
diversity ratio
(0.822)
Firm size 1.879*** 2.299*** 1.263** 1.303** 1.433**
(4.203) (5.157) (3.267) (3.128) (3.286)
Leverage -4.865 -6.539 -8.277* -7.519* -7.233
ratio
(-1.036) (-1.429) (-1.817) (-1.654) (-1.597)
Firm age 0.044*** 0.046*** 0.034** 0.041** 0.040**
(2.713) (2.917) (2.027) (2.495) (2.467)
Year dummy Yes Yes Yes Yes Yes
Industry Yes Yes Yes Yes Yes
dummy
Adjusted R2 0.089 0.142 0.088 0.081 0.081
Observations 267 267 267 267 267
Notes: 1. Unstandardized coefficients are given. 2. t-statistics are reported in parentheses. 3. ***
indicates significance at the 0.01 level, ** indicates significance at the 0.05 level and * indicates
significance at the 0.1 level. 4. Firm size is expressed in the natural logarithm of total assets.
Table 8: OLS Regression - ROE

Constant -49.602*** -55.183*** -23.505 -35.810*** -38.150**
(-3.834) (-4.291) (-1.592) (-3.036) (-3.105)
Total board -0.343*
size
(-1.406)
39
MB size -2.682**
(-3.355)
SB size 0.126
(0.442)
Board -10.611**
independence
(-1.742)
ratio
Gender 17.074
diversity ratio
(1.209)
MB gender -2.188
diversity ratio
(-0.175)
SB gender 10.250
diversity ratio
(0.941)
Firm size 2.453*** 2.947*** 1.583** 1.542** 1.681**
(3.529) (4.193) (2.639) (2.388) (2.480)
Leverage 11.170 9.202 6.315 7.287 7.884
ratio
(1.530) (1.275) (0.893) (1.034) (1.120)
Firm age 0.066** 0.069** 0.051* 0.061** 0.061**
(2.621) (2.754) (1.946) (2.427) (2.404)
dummy
Adjusted R2 0.092 0.121 0.096 0.090 0.085
Observations 267 267 267 267 267
Table 9: OLS Regression - Tobin’s Q

Constant 6.392*** 6.556*** 5.229*** 6.852*** 6.785***
(6.242) (6.341) (4.498) (7.378) (6.981)
Total board -0.051***
size
(-2.656)
MB size 0.017
(0.269)
SB size -0.065***
40
(-2.842)
Board 1.595***
independence
(3.327)
ratio
Gender -3.412***
diversity ratio
(-3.068)
MB gender -1.581*
diversity ratio
(-1.602)
SB gender -1.787**
diversity ratio
(-2.075)
Firm size -0.100* -0.115** -0.136*** -0.112** -0.112**
(-1.826) (-2.034) (-2.873) (-2.198) (-2.082)
Leverage -3.122*** -3.065*** -3.275*** -3.374*** -3.476***
ratio
(-5.402) (-5.283) (-5.885) (-6.082) (-6.244)
Firm age -0.002 -0.002 -0.001 -0.003 -0.003
(-1.095) (-1.129) (-0.513) (-1.304) (-1.289)
dummy
Adjusted R2 0.266 0.267 0.277 0.273 0.262
Observations 267 267 267 267 267
41
6. Conclusion
This study examined the impact of board characteristics on firm performance for German
listed firms. The relationship between the board characteristics and firm performance is
tested by conducting an OLS regression analysis, using data from a sample of 89 German
firms from the Frankfurt Stock Exchange. The main objective of this study was to find an
answer to the research question and thereby contributing to the literature written on this
topic. The research question of this study was formulated as follows:
What is the impact of the board of directors’ characteristics on the firm performance of firms
listed in Germany?
The German board structure is quite unique with the separation of the management and
supervisory board and codetermination laws that can be applicable to firms. This unique
structure made the analysis of the results even more interesting. In order to understand the
potential impact of board characteristics, several theories and previous empirical studies
have been researched. Theories that are important in the analysis of board characteristics
and firm performance are agency theory, stewardship theory, resource dependence theory
and human capital theory. These theories have been used throughout literature focusing on
the topic of corporate governance and firm performance. Hypotheses have been formed
based on these theories and previous empirical findings.
The OLS regression model has been formed to test the hypotheses and eventually provide an
answer to the research question. The model included the dependent, independent and
control variables. As dependent variables, two accounting-based measures (ROA and ROE)
and one market-based measure (Tobin’s Q) have been used. The independent variables were
the board characteristics – board size, board independence and gender diversity. In addition,
several control variables were included, these were firm size, leverage ratio, firm age, year
and industry dummies.
The first board characteristic to be tested was the board size and its impact on firm
performance. Board size in this study was classified in total board size, management board
size and supervisory board size. According to agency theory, larger boards increase agency
costs and the problem of free riding among the many board members arises which
eventually has a negative effect on firm performance (Kao et al., 2019). Therefore,
Hypothesis 1 stated that smaller boards have a positive impact on firm performance. This
was tested for total board size, management board size and supervisory board size. Evidence
in this study shows that there is a negative significant relationship between total board size
and all three performance measures meaning that Hypothesis 1A is confirmed. In addition,
Hypothesis 1B regarding the management board size is confirmed for both accounting-based
performance measures. On the other hand, management board size has no significant
relation with Tobin’s Q, thus Hypothesis 1B is rejected for the market-based measure.
Furthermore, supervisory board size has no significance with ROA and ROE, thus Hypothesis
1C is rejected for the accounting-based measures. However, supervisory board size does
have a negative significance with Tobin’s Q, which means that Hypothesis 1C is confirmed for
the market-based measure.
42
The second board characteristic that was tested was board independence, which was
defined as the ratio of supervisory board members who are no employee representatives to
the total number of supervisory board members. Empirical literature provided mixed
findings, however the main idea was that increasing the number of independent members
positively affects firm performance because of their effective monitoring and supervising
activities. Therefore, Hypothesis 2 stated that a greater proportion of independent directors
on boards has a positive impact on firm performance. Opposed to the expected, ROA and
ROE have a negative significant relationship meaning that Hypothesis 2 is rejected. However,
for Tobin´s Q the relationship is significant and positive, thus Hypothesis 2 is confirmed for
the market-based measure.
The third board characteristic was gender diversity, defined as the ratio of female board
members to the total board, management board and supervisory board separately. Gender
diversity and the board of directors have especially been relevant topics in recent years and
received a lot of attention in public debates, academic research and corporate strategy
(Marinova et al., 2010). All theories presented positive perspectives on gender diversity on
boards and described that it can have a positive impact on firm performance. Hypothesis 3
stated that a greater proportion of female directors has positive impact on firm
performance. This study finds no evidence that gender diversity has an impact on
accounting-based performance since there is no significant relationship, causing a rejection
of Hypotheses 3A, 3B and 3C for ROA and ROE. In addition, there is a significant negative
relation between gender diversity and Tobin’s Q. Thus, contrary to the expected, increased
gender diversity has a negative impact on market-based performance causing a rejection of
Hypotheses 3A, 3B and 3C.
Concluding, smaller boards in this study sample are more favorable for firm performance.
Greater board independence has a negative impact on ROA and ROE, and is favorable for
Tobin’s Q. Gender diversity has no impact on accounting-based performance and a
significant negative impact on Tobin’s Q.
43
7. Limitations and Recommendations
This study provides relevant results on the impact of board characteristics on firm
performance in Germany. However, there are some limitations that should be considered
and recommendations for future research on the topic. First of all, the firms in this sample
were chosen from the largest stock exchange and its indices, therefore the results should
not be generalized for all German firms. In addition, the focus of this study was on German
firms which have a unique board structure with the separation of management and
supervisory board and codetermination law. This means that it is not possible to consider
the results as a worldwide effect regarding the specific board characteristics and firm
performance. For future studies on the impact of board characteristics on firm performance
in Germany it is recommended to include a wider variety of firms. Furthermore, for future
studies on board characteristics of two-tier board structures it is recommended to make a
comparison of the same characteristics on different countries. This makes it possible to
analyze whether the results are country or board structure specific.
Secondly, the research period of this study was from 2017 to 2019 in order to measure the
impact of the in 2016 introduced gender quota for German supervisory boards. As the
results show, the impact is not as expected resulting in rejecting Hypothesis 3. A reason for
this could be that there is only one year between the introduction of the gender quota and
the collection of data on gender diversity. Therefore it is possible that the potential positive
effects are not established yet in this sample and research period. A recommendation would
be to widen the range of the research period for this specific characteristic and compare the
results of prior gender quota and after gender quota. This method could help in analyzing
the effects of the gender quota, and whether there is a change in the impact, either
positively or negatively.
Thirdly, in the OLS regression model choices have been made for defining and measuring
specific dependent, independent and control variables. Previous studies on the topic played
an important role in this process and method. A recommendation for future studies would
be to consider different methods to control for the results of firm performance with other
variables. Another point is that board independence in this study was defined as the ratio of
no employee representatives on the supervisory board to the total supervisory board.
Future studies on two-tier board structures could consider another definition for board
independence.
44
References
Abidin, Z. Z., & Jusoff, K. (2009). Board Structure and Corporate Performance in Malaysia.
International Journal of Economics and Finance, 1(1), 150–164.
Albert Resources. (2020). Key Assumptions of OLS: Econometrics Review.

https://www.albert.io/blog/key-assumptions-of-ols-econometrics-review/.
Ammari, A., Kadria, M., & Ellouze, A. (2014). Board Structure and Firm Performance:
Evidence from French Firms Listed in SBF 120. International Journal of Economics and
Financial Issues, 4(3), 580–590.
Bennouri, M., Chtioui, T., Nagati, H., & Nekhili, M. (2018). Female board directorship and
firm performance: What really matters? Journal of Banking and Finance, 88, 267–291.
Bermig, A., & Frick, B. (2010). Board Size, Board Composition, and Firm Performance:
Empirical Evidence from Germany. SSRN Electronic Journal, 1–42.
Best, H., & Wolf, C. (2015). Regression Analysis and Causal Inference. SAGE, 1–32.
Bezemer, P.-J., Peij, S., de Kruijs, L., & Maassen, G. (2014). How two-tier boards can be more
effective. Corporate Governance, 14(1), 15–31.
Block, D., & Gerstner, A.-M. (2016). One-Tier vs. Two-Tier Board Structure: A Comparison
Between the United States and Germany. Comparative Corporate Governance and
Financial Regulation.
Boneberg, F. (2010). The economic consequences of one-third co-determination in German

supervisory boards: First evidence for the service sector from a new source of
enterprise data. EconStor, 1–31.
Bosse, D. A., & Phillips, R. A. (2016). Agency Theory and Bounded Self-Interest. Academy of
Management Review, 41(2), 276–297.
Bozhinov, V., Joecks, J., & Scharfenkamp, K. (2021). Gender spillovers from supervisory
boards to management boards. Managerial and Decision Economics, 42(2), 1–15.
Bozhinov, V., Koch, C., & Schank, T. (n.d.). Has the Push for Equal Gender Representation
Changed the Role of Women on German Supervisory Boards? Schmalenbach Business
Review, 71(3), 385–411.
Daoud, J. I. (2017). Multicollinearity and Regression Analysis. Journal of Physics, 1–7.
Darmadi, S. (2011). Board compensation, corporate governance, and firm performance in

Indonesia. SSRN Electronic Journal, 1–45.
Darmadi, S. (2011). Board diversity and firm performance: the Indonesian evidence.
Corporate Ownership and Control Journal, 8, 1–39.
45
Dimov, D. P., & Shepherd, D. A. (2005). Human capital theory and venture capital firms:
exploring ‘‘home runs’’ and ‘‘strike outs.’’ Journal of Business Venturing, 20, 1–21.
Donaldson, L., & Davis, J. H. (1991). Stewardship Theory or Agency Theory: CEO Governance
and Shareholder Returns. Australian Journal of Management, 16(1), 49–65.
du Plessis, J. J. (2004). The German Two-Tier Board and the German Corporate Governance
Code. European Business Law Review, 1139–1164.
Duru, A., Iyengar, J., & Zampelli, E. (2016). The dynamic relationship between CEO duality
and firm performance: The moderating role of board independence. Journal of
Business Research, 1–9.
Ees, H. van, Postma, T. J. B. M., & Sterken, E. (2003). Board Characteristics and Firm
Performance in the Netherlands. Eastern Economic Journal, 29(1), 41–58.
El-Faitouri, R. (2014). Board of Directors and Tobin’s Q: Evidence from U.K.Firms. Journal of
Finance and Accounting, 2(4), 82–99.
Farahani, H. A., Rahiminezhad, A., Same, L., & Immannezhad, K. (2010). A Comparison of
Partial Least Squares (PLS) and Ordinary Least Squares (OLS) regressions in predicting
of couples mental health based on their communicational patterns. Procedia Social
and Behavioral Sciences, 5, 1459–1463.
Fauver, L., & Fuerst, M. E. (2006). Does Good Corporate Governance Include Employee
Representation? Evidence from German Corporate Boards. Journal of Financial
Economics, 82(3), 673–710.
Fiss, P. C. (2006). Social Influence Effects and Managerial Compensation Evidence From
Germany. Strategic Management Journal, 27, 1013–1031.
Gani, L., & Jermias, J. (2006). Investigating the effect of board independence on performance
across different strategies. The International Journal of Accounting, 41, 295–314.
Gillies, D. (2015). Human Capital Theory in Education. Encyclopedia of Educational

Philosophy and Theory, 1–5.
Glinkowska, B., & Kaczmarek, B. (2015). Classical and modern concepts of corporate
governance (Stewardship Theory and Agency Theory). Management, 19(2), 84–92.
Gogtay, N. J., & Thatte, U. M. (2017). Principles of Correlation Analysis. Journal of the
Association of Physicians of India, 65, 78–81.
Graf, A., & Stiglbauer, M. (2009). Board Size and Firm Operating Performance: Evidence from
Germany. Corporate Board, 5(1), 37–47.
Green, C. P., & Homroy, S. (2017). Female Directors, Board Committees and Firm
Performance. SSRN Electronic Journal, 1–52.
46
Hidayat, A. A., & Utama, S. (2016). Board Characteristics and Firm Performance: Evidence
from Indonesia. International Research Journal of Business Studies, 8(3), 137–154.
Joecks, J., Pull, K., & Vetter, K. (2012). Gender Diversity in the Boardroom and Firm Perfor-
mance: What Exactly Constitutes a “Critical Mass”? Journal of Business Ethics, 1–23.
Johnson, B. L. (1995). Resource Dependence Theory:A Political Economy Model of

Organizations, 1–21.
Jungmann, C. (2007). The Effectiveness of Corporate Governance in One-Tier and Two-Tier

Board Systems - Evidence from the UK and Germany. European Company and Financial
Law Review, 3(4), 426–474.
Kao, M., Hodgkinson, L., & Jaafar, A. (2019). Ownership structure, board of directors and
firm performance: Evidence from Taiwan. Corporate Governance, 19(1), 189–216.
Keay, A. (2017). Stewardship Theory: Is Board Accountability Necessary? International

Journal of Law and Management, 59(6), 1292–1314.
Kor, Y. Y. (2008). Experience-Based Human Capital and Social Capital of Outside Directors.
Journal of Management, 1–26.
Latif, R. A., Kamardin, H., Mohd, K. N. T., & Adam, N. C. (2013). Multiple Directorships, Board
Characteristics and Firm Performance in Malaysia. Management, 3(2), 105–111.
Li, K., Lu, L., Mittoo, U. R., & Zhang, Z. (2015). Board independence, ownership concentration
and corporate performance—Chinese evidence. International Review of Financial
Analysis, 41, 162–175.
Lückerath-Rovers, M. (2013). Women on boards and firm performance. Journal of

Management and Governance, 17, 491–509.
Makhlouf, M., Laili, N., Basah, M., & Ramli, N. (2017). Board of Directors' Effectiveness and
Firm Performance: Evidence from Jordan. Research Journal of Finance and Accounting,
8(18), 23–34.
Marinova, J., Plantenga, J., & Remery, C. (2010). Gender Diversity and Firm Performance:
Evidence from Dutch and Danish Boardrooms. The International Journal of Human
Resource Management, 1–28.
Martin, C., & Herrero, B. (2018). Board of directors: Composition and effects on the
performance of the firm. Economic Research, 31(1), 1015–1041.
Martinez, M., & Alvarez, I. (2019). Do board characteristics drive firm performance? An
international perspective. Review of Managerial Science, 1–47.
Mummolo, J., & Peterson, E. (2018). Improving the Interpretation of Fixed E

ects Regression Results. Semantic Scholar, 1–15.
47
Naciti, V. (2019). Corporate governance and board of directors: The effect of a board
composition on firm sustainability performance. Journal of Cleaner Production, 237, 1–
8.
Nienhüser, W. (2008). Resource dependence theory: How well does it explain behavior of
organizations? Management Revue, 19(1), 9–32.
O'Connell, V., & Cramer, N. (2010). The relationship between firm performance and board
characteristics in Ireland. European Management Journal, 28, 387–399.
Panda, B., & Leepsa, N. M. (2017). Agency theory: Review of Theory and Evidence on
Problems and Perspectives. Indian Journal of Corporate Governance, 10(1), 74–95.
Poston Jr, D. L. (2021). Ordinary Least Squares Regression. International Encyclopedia of the
Social Sciences. https://www.encyclopedia.com/social-sciences/applied-and-social-
sciences-magazines/ordinary-least-squares-regression.
Pugliese, A., Minichilli, A., & Zattoni, A. (2014). 0 Integrating Agency and Resource
Dependence Theory: Firm Profitability, Industry Regulation, and Board Tasks
Performance. Journal of Business Research, 67(6), 1189–1200.
Ramdani, D., & Witteloostuijn, A. (2010). The Impact of Board Independence and CEO
Duality on Firm Performance: A Quantile Regression Analysis for Indonesia, Malaysia,
South Korea and Thailand. British Journal of Management, 21, 607–626.
Schillemans, T., & Bjurstrøm, K. H. (2020). Trust and verification: balancing agency and
stewardship theory in the governance of agencies. International Public Management
Journal, 23(5), 650–676.
Shukeri, S. N., Shin, O. W., & Shaari, M. S. (2012). Does Board of Director’s Characteristics
Affect Firm Performance? Evidence from Malaysian Public Listed Companies.
International Business Research, 5(9), 120–127.
Smith, N., Smith, V., & Verner, M. (2005). Do Women in Top Management Affect Firm
Performance? A Panel Study of 2500 Danish Firms. International Journal of Productivity
and Performance Management, 55(7), 1–37.
Soare, T., Detilleux, C., & Deschacht, N. (2021). The impact of the gender composition of
company boards on firm performance. International Journal of Productivity and
Performance Management, 1–20.
Statistical Solutions. (2021). Two-Stage Least Squares (2SLS) Regression Analysis.

https://www.statisticssolutions.com/free-resources/directory-of-statistical-
analyses/two-stage-least-squares-2sls-regression-analysis/.
Steger, T. (2007). Internationality of Supervisory Boards and Firm Performance: Evidence

From large listed German Companies. SCMS Journal of Indian Management, 5–19.
48
Terjesen, S., Couto, E., & Francisco, P. (2016). Does the presence of independent and female
directors impact firm performance? A multi-country study of board diversity. Journal of
Management and Governance, 20, 447–483.
Terjesen, S., Singh, V., & Vinnicombe, S. (2008). Do women still lack the ‘right’ kind of human
capital for directorships on the FTSE 100 corporate boards? ResearchGate, 152–164.
Torres-Reyna, O. (2007). Panel Data Analysis Fixed and Random Effects using Stata. Princeton
University, 1–40.
Valenti, A., & Horner, S. (2020). THE HUMAN CAPITAL OF BOARDS OF DIRECTORS AND
INNOVATION: AN EMPIRICAL EXAMINATION OF THE PHARMACEUTICAL INDUSTRY.
International Journal of Innovation Management, 24(6), 1–32.
Weidler, H. L. (2020). Does the gender quota make sense? Women on supervisory boards
and firm performance. Barcelona School of Management, 1–39.
49
Appendices
Appendix A: List of sample firms
1&1 AG Deutsche Post AG Jenoptik AG Serviceware SE

Adidas AG Deutsche Telekom AG Jungheinrich AG Siemens AG
Aixtron SE Deutsche Wohnen SE K+S Aktiengesellschaft Siltronic AG
Alstria Office Reit-AG Duerr Kion Group AG SMA Solar Technology

Aktiengesellschaft AG
Aurubis AG E.ON SE Knorr-Bremse AG Software AG
Aves One AG Eckert & Ziegler Lanxess AG Stroeer SE & Co. KGAA
Strahlen- und
Medizintechnik AG
Basf SE Elumeo SE LEG Immobielen SE Symrise AG

Bayer AG Evonik Industries AG Merck KGAA TAG Immobilien AG
Bayerische Motoren Evotec SE Morphosys AG Telefonica
Werke AG Deutschland Holding
AG
Bechtle AG Fraport AG MTU Aero Engines AG ThyssenKrupp AG
Beirsdorf AG Freenet AG Nagarro SE Uniper SE

Brenntag SE Fresenius Medical Nemetschek SE United Internet AG
Care AG & Co.KGAA
Cancom SE Fresenius SE & Co. Nordex SE VA-Q-TEC AG
KGAA
Capsensixx AG Fuchs Petrolub SE Pfeiffer Vacuum Varta AG
Technology AG
Carl Zeiss Meditec AG GEA Group AG Porsche Automobil Volkswagen AG

Holding SE
Continental AG Gerresheimer AG Prosiebensat.1 Media Voltabox AG

SE
Covestro AG Heidelbergcement AG Puma SE Vonovia SE
Creditshelf AG Hella Gmbh & Co. Rational AG Wacker Chemie AG

KGAA
CTS Eventim AG & Co. Hellofresh SE Rheinmetall AG Zalando SE
KGAA
Daimler AG Henkel AG & Co. RWE Zooplus AG
Aktiengesellschaft
Delivery Hero SE Hugo Boss AG SAP SE
Dermapharm Holding Hypoport AG Sartorius AG
SE
Deutsche Lufthansa Infineon Technologies Scout24 AG
AG AG
50
Appendix B: Frequencies
Supervisory Board Employee Representatives
Frequency Percent Cumulative Percent
0 96 36 36
1 4 1.5 37.5
2 17 6.4 43.8
3 6 2.2 46.1
6 66 24.7 70.8
7 2 0.7 71.5
8 32 12 83.5
9 10 3.7 87.3
10 30 11.2 98.5
11 3 1.1 99.6
12 1 0.4 100
Total 267 100
Independent Supervisory Board Members

3 21 7.9 7.9
4 21 7.9 15.7
5 17 6.4 22.1
6 108 40.4 62.5
7 9 3.4 65.9
8 35 13.1 79
9 9 3.4 82.4
10 26 9.7 92.1
11 15 5.6 97.8
12 5 1.9 99.6
13 1 0.4 100
Total 267 100
Management Board Female Directors

0 180 67.4 67.4
1 70 26.2 93.6
2 17 6.4 100
Total 267 100
Supervisory Board Female Directors

0 33 12.4 12.4
1 49 18.4 30.7
2 37 13.9 44.6
3 18 6.7 51.3
4 57 21.3 72.7
5 21 7.9 80.5
6 32 12 92.5
7 11 4.1 96.6
51
8 7 2.6 99.3
9 2 7 100
Total 267 100
Appendix C: Regression Standardized Residual

Regression standardized residual histograms of the models with a significance at the 0.01
level.
ROA Model 2
Tobin’s Q Models 1 - 4
52
53
Appendix D: Regression Plots
Regression plots of significant regression results at the 0.01 level.
54
55

Kazan CG

Uploaded by

Copyright:

Available Formats

Kazan CG

Uploaded by

Document Information

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Kazan CG

Uploaded by

Copyright:

Available Formats

Acknowledgements

1.1. Outline of the Study

2.1. Theoretical Perspectives

2.1.1. Agency theory

2.1.2. Stewardship Theory

2.1.3. Resource Dependence Theory

2.1.4. Human Capital Theory

2.2. Board of Directors

2.3. German Board Structure

2.3.1. The Management Board

2.3.2. The Supervisory Board

A key factor to accomplish abovementioned tasks and responsibilities is the communication

2.4. Board Characteristics

Adding to this, in Germany as mentioned there is the separation between a management

In the case of the management board there is no maximum or minimum amount of

2.4.2. Board Independence

Because of their independence and objectiveness, independent directors are considered to

2.4.3. Gender Diversity

2.5. Empirical Findings

2.5.2. Board Independence and Firm Performance

2.5.3. Gender Diversity and Firm Performance

2.6. Hypothesis Development

2.6.1. Board Size

2.6.2. Board Independence

Hypothesis 2: greater proportion of independent directors on boards has a positive impact on

2.6.3. Gender Diversity

3.1. Methods in Previous Studies

3.1.2. Fixed Effects Regression

3.2. Method in this Study

PERFit = β1BSIZEit + β2BINDEPit + β3FEMDIRit + β4SIZEit + β5LEVit + β6AGEit + β7YEARit + β8INDit +

Table 1: Model Codes Descriptions

3.3.2. Independent Variables

3.3.3. Control Variables

Table 2: Variable Definitions

Additionally, in order to collect data on the independent variables of board characteristics,

4.1. Industry Classification

Table 3: Industry classifications

5.1. Descriptive Statistics

5.1.2. Independent Variables

5.1.3. Control Variables

Table 4: Descriptive Statistics

5.2. Correlation Analysis

Lastly, even if the dependent variables do not have an effect on multicollinearity it is

Table 6: Multicollinearity Test

Total board size 2.068

Board independence 1.463

Gender diversity ratio 1.436

MB gender diversity ratio 1.201

SB gender diversity ratio 1.330

Firm size 1.909 2.050 1.368 1.516 1.651

Leverage ratio 1.171 1.193 1.101 1.100 1.092

Firm age 1.107 1.108 1.208 1.090 1.092

Notes: the variance inflation factors (VIF) are given.

5.4. Regression Analysis

5.4.1. Board Size

5.4.2. Board Independence

5.4.3. Gender Diversity

Table 7: OLS Regression - ROA

Table 8: OLS Regression - ROE

Table 9: OLS Regression - Tobin’s Q