Master thesis for International Business & Management
The Influence of Supervisory Board Characteristics on
Accounting as well as Market based Firm Performance: Evidence
from German Stock Corporations
Author:
Mohssen Sadegh Nezhadian
S1940821
Supervisor:
Dr. W. G. Biemans
Referent:
Prof. Dr. H. van Ees
University of Groningen
Faculty of Economics and Business
Postbus 800, 9700 AV Groningen
The Netherlands
TABLE OF CONTENTS
LIST OF TABLES.………...3
ATTESTATION OF AUTHORSHIP………….………...4
ABSTRACT……….………….………...5
1 INTRODUCTION.………..…..6
2 BACKGROUND INFORMATION TO CORPORATE GOVERNANCE IN GERMANY…..9
3 THEORETICAL BACKGROUND………..……….11
3.1 Importance of corporate governance……….………..12
3.2 Roles and characteristics of corporate boards and firm performance…...………...12
3.3 Theoretical perspectives…………...……….………..16
3.3.1 Agency theory………....16
3.3.2 Resource dependence theory……….………...17
4 LITERATURE REVIEW AND HYPOTHESES DEVELOPMENT………….………18
4.1 Board size……….………19 4.2 Board meetings………20 4.3 Board busyness………22 4.4 Board composition……….…..23 4.5 Board qualification…..………24 4.6 Board remuneration……….24
5 RESEARCH DESIGN AND MEASUREMENT MODEL…………..………25
5.1 Data sample………..26
5.2 Data sources……….26
5.3 Variable construction………...27
5.3.1 Dependent variable: Firm performance………..27
5.3.2 Independent variables: Supervisory board characteristics………..28
5.3.3 Control variables: Firm-level variables………....………..29
5.4 Method………...………..31
5.5 Regression model validation…...………...………..31
6 EMPIRICAL RESULTS AND DISCUSSION………..34
6.1 Descriptive statistics……….34
6.2 Presentation and discussion of the OLS estimation results……….………...……….36
7 CONCLUSION...42
8 LIMITATIONS AND POTENTIALS FOR FUTURE RESEARCH……….45
REFERENCES………..48
LIST OF TABLES
Table I: Number and proportion of firms by industry classification………..26
Table II: Variables and operationalization………..30
Table III: Pearson correlations for variables in the regression model………33
Table IV: Descriptive statistics for variables in the regression model………...35
Table V: OLS Regression results: Dependent variable Return on Assets (ROA)………..37
ATTESTATION OF AUTHORSHIP
I hereby declare that this submission is my own work and that, to the best of my knowledge and belief, it contains no material previously published or written by another person, nor material which to a substantial extent has been submitted for the award of any other degree or diploma of a university or other institution of higher learning.
Mohssen Sadegh Nezhadian
Abstract
In the recent past, there has been a huge interest on the roles and responsibilities of boards in the economic success of firms, especially because of diverse corporate scandals. This study analyzes the relation between key supervisory board characteristics and firm performance particularly using two theoretical perspectives: agency theory and resource dependence theory. Based on theoretical perspectives and the literature, I develop a set of hypotheses to analyze the relationship between supervisory board characteristics and market as well as accounting based firm performance. I test my hypotheses on a sample of 65 non-financial firms listed on the German stock exchange over a four year period from 2004 to 2007. The findings of the study show that supervisory board characteristics such as supervisory board size, number of supervisory board meetings and educational qualification of supervisory board members in terms of PhD degree were not related to firm performance, whereas directors with multiple directorships seemed to have a significant negative impact on firm performance. This suggests that “busyness” destroyed value in terms of networks and better resource accessibility. Bank representatives on the supervisory board and average board member remuneration were positively related with firm performance. Given the findings the two theories, agency and resource dependence theory, were only partly supported by the empirical results of this thesis project. Moreover, the results indicate that no single corporate governance theory explains the interrelation between board characteristics and firm performance and that the results of empirical studies within the Anglo-Saxon one-tier regime cannot be directly assumed to hold within the German two-tier regime.
1. Introduction
The field of corporate governance has extended fast during the recent past, in particular since the corporate failures and scandals in the 2000s. It has decoyed the interest of many academics, researchers, policy makers, firms and the public. Researchers have used various definitions to describe corporate governance. According to Shleifer and Vishny (1997):
“Corporate governance deals with the ways in which suppliers of finance to corporations assure themselves of getting a return on their investment.”
In general, corporate governance refers to the set of mechanisms that impact the decisions made by managers when there is a separation of ownership and control (Larcker et al., 2005). The need for monitoring arises out of information asymmetry and conflict of interests in the event of separation of ownership and control. One internal monitoring mechanism is the board of directors.
The significance of this thesis project is clear from the striking growth in the literature on corporate governance, in particular corporate boards, across business studies such as, economics, management studies, financial management and accounting literatures. Moreover, there has been a rise in research, especially quantitative empirical research, in the field of corporate governance after several notable accounting scandals (e.g., Enron, Ahold, Nicor and WorldCom). The necessity for good corporate governance is indicated by diverse standards established at a national and international level (e.g., the Combined Code in the UK, the Sarbanes-Oxley Act 2002 in the US, and the Organization for Economic Development Code). In addition, in the field of corporate governance there is an ongoing debate whether “good corporate governance” promotes firm performance. Multiple projects examined different questions in this setting, but primary in developed capital markets such as the United States of America, United Kingdom or Japan (e.g., Gompers et al., 2003; Dittmar and Mahrt-Smith, 2007).
In this context prior literature amongst others examines the issue, whether certain board characteristics impact firm performance. Overall, the results of prior research are ambiguous (Dalton et al., 2003; Hermalin and Weisbach, 2010).
The vast majority of these empirical studies focus on one-tier board systems where the board together supervises and controls the corporation (Bremert et al., 2009). Nevertheless, it is interesting to examine whether the findings from one-tier regimes, like in the US or UK, persist in the two-tier board system, prevalent in Continental European countries.1 The German two-tier board system clearly separates management (management board) and supervision (supervisory board) of a stock corporation by transmitting each task to a different board. It is very suitable to examine the relationship between supervisory board characteristics and firm performance since the feature of the German system is the two-tier board system which separates the management board (Vorstand) and the supervisory board (Aufsichtsrat).2
The economic success of a firm is influenced by the management board. The actions of the management board are on the other hand influenced by the supervisory board which controls and advises the management board. Germany is besides the Netherlands and Denmark a major two-tier country in Europe and research in this setting can help to increase the understanding of the interrelations within the two-tier governance regime.
The efficiency of the supervisory board is addressed in the German context particularly using agency theory and resource dependence theory. According to agency theory, in the event of separation of management and ownership, the manager might seek to act in his own interest which is not necessarily in the interests of the shareholders and are different from those interests that are needed to enhance the shareholder returns. In a business world with environmental uncertainty the level of information asymmetry between both parties impacts the monitoring performance of the board. The agency perspective is very acknowledged in explaining the duties of boards in decreasing the agency costs (for instance, monitoring costs) (Fama & Jensen, 1983).
Further, the resource dependence perspective views agents as a resource. They can provide business networks and affect the environment in advantage to their corporation (Carpenter & Westphal, 2001).
1 This system is also found in the Netherlands, Austria, Finland and Denmark
The research question of this thesis is formulated as:
How do supervisory board characteristics affect the firm performance of non-financial firms listed on the German Stock Exchange?
Given the theoretical perspectives relating to boards’ impact on firm performance this study expects to detect measurable effects of supervisory board characteristics on firm performance that are in line with the agency and resource dependence perspective.
This study is motivated by several aspects. For senior managers of corporations, it is very essential to be aware of the influence of supervisory board characteristics on corporate performance because it can influence the confidence of the shareholders. Furthermore, for shareholders, it is essential to learn how corporate governance indicators, here board characteristics, impact firm-specific risk, as they might get an enhanced understanding of the risk-return relation.
For politicians and regulatory institutions in Germany, it is essential to create a regulatory infrastructure which is competitive in order to attract capital. The evidence on the relation between board characteristics and firm performance can help politicians exercise important policies and evaluate the efficiency of these actions (Sami et. al., 2009).
The thesis includes an analysis of various characteristics of the German supervisory board, including aspects of board structure, board composition, board busyness and qualification and board remuneration in correlation with firm performance.
The thesis is structured as follows: the next section provides background information on corporate governance in Germany, followed by an overview of two theories of corporate governance. The fourth section reviews the literature relating to board characteristics and its association or relationship to firm performance. The fifth section describes the research design and measurement model. Findings and analysis are presented in the sixth section, while the conclusion is presented in the seventh section. The thesis is closed by discussing the limitations and the possibilities for future research.
2. Background information to corporate governance in Germany
Institutional division between management and supervision
The federal republic of Germany is a suitable and interesting country for this kind of governance studies, because law amongst others explicitly prescribes a two-tier board system (management board; (Vorstand) and supervisory board; (Aufsichtsrat). The management board includes the executive directors who control the business, and the supervisory board who has a monitoring, supervision and consulting role. Members of the supervisory board have no direct business-related executive power themselves. Thus, the two-tier board structure in Germany provides a clear division between the firm’s management and monitors. Accordingly, this leads to an increased institutional transparency (Bremert et al., 2009).
Role and responsibility of boards
A straight forward consequence of the division between the management board and supervisory board is that their tasks differ. Members of the management board, including the Chief Executive Officer (CEO) manage the firm and, hence, without any exception are not allowed to hold a seat on the supervisory board of the firm in which they control and steer the business and vice versa. As a result, this regulation leads to an institutional separation between the firm’s CEO (Vorstandsvorsitzender) and the chairman (Aufscihtsratvorsitzender) of the supervisory board. Even though, both the management team as well as the supervisory board together contribute to improving the performance of the firm the supervisory board is not authorized to engage itself in the management of the corporation. Nevertheless, it still can declare regulations requiring its permission on a set of essential management decisions, hence, having control over the executive directors in the firm in an indirect way. In addition, the supervisory board appoints and dismisses executive directors (including the CEO) and is engaged in compensation negotiations. However, over the recent past the comprehension of the roles and duties of the supervisory board has changed. Once, having been focused on solely supervising and monitoring the management, the supervisory board is also considered as a consultant for the management team who supports the managers in adequately fulfilling their roles and responsibilities (Schmidt, 2003; Bremert et al, 2009).
The principle of codetermination and supervisory board size
the supervisory board. The concept is interesting for this study, as the supervisory board size and its composition are connected to the principle of codetermination. According to law (Aktiengesetz AktG.) the following requirements are compulsory: For firms with 2.000-10.000 employees it is compulsory to have at least 12 directors serving on the supervisory board, firms with 10.000-20.000 employees are obliged to have at least 16 supervisory board members, and firms where the number of employees exceeds 20.000 are obliged to have at least 20 directors serving on the supervisory board. Thereby, half of the supervisory board members need to be either union functionary or employee of the company.
Compared to other governance regimes in Europe the governance regime in Germany allows a comparably high involvement of workforce in the supervision of the management board. The concept of codetermination is a consequence of the major influences German trade unions have had and nowadays still have on firms. Having employee representatives on the supervisory board helps to effectively address workforce issues like compensation, labor disputes or workforce conditions (Bremert et al., 2009).
Compensation of the supervisory board
The compensation of the supervisory board is set by the general meeting (Hauptversammlung) that typically takes place once a year. Compared to the executive compensation, it is not allowed for non-executive directors to receive stock options as remuneration component. In general, supervisory board members simply receive a fixed compensation. In the event of variable compensation, this is often connected to the payment of the firm’s dividend. Compared to the executive directors supervisory board members receive relatively small compensation. However, with the increasing prominence of corporate governance in Germany, the compensation of the supervisory board was raised remarkably in the last few years. Since 2004 German stock corporations have to disclose information on the level of annual compensation of the executive directors. The disclosure of supervisory board members’ remuneration is not obligatory. Nevertheless, many firms also disclose the remuneration of the supervisory board (Bremert et al., 2009).
The role of banks
regulatory changes banks’ stockholdings in non-financial firms has decreased (Dittmann et al., 2008, Bremert et al., 2009).
Furthermore, it is quite common in Germany that bankers hold seats on the supervisory board and even take the role of the chairmen (Edwards and Nibler, 2000).
In general, the corporate governance regime in Germany can be viewed as stakeholder-oriented compared to the United States or the United Kingdom (Franks and Mayer, 1994). Further, at least three large groups have power and are influential stakeholders. These are employee and/or union representatives, and banks.
3. Theoretical background
This chapter reviews the literature and provides theoretical background relating to corporate governance, in particular board characteristics and firm performance. The theoretical background section has been organized in the following sections: The first section will discuss the importance of corporate governance. The second section will describe the roles and characteristics of corporate boards in particular the supervisory board and its interrelation with firm performance. Finally, the third section will review theoretical perspectives relating to boards’ impact on firm performance.
3.1 Importance of corporate governance
The field of corporate governance deals with a set of incentive and control mechanisms, which addresses conflict of interests between stakeholders of a firm. According to Larcker et al. (2005) corporate governance deals with internal and external mechanism that have an influence the decisions made by executive directors when there is a separation of ownership and control. Managers can utilize the presence of asymmetric information to their advantage and act in their own interests. The remarkable and striking growth in the literature on corporate governance across business studies such as, financial management, management studies, economics and accounting indicates the importance of corporate governance.3
Corporate governance regimes are different among various legal and institutional frameworks. However, there are similarities concerning the internal and external governance mechanisms that are employed. Amongst others they include, corporate boards, ownership structure, risk oversight and remuneration schemes. These aspects of governance mechanisms
to a greater or lesser extent influence fundamental company decision. Accordingly, these decisions in turn affect the economic effect of a firm (Bremert et al., 2009).
Therefore, strong corporate governance standards are the primary element upon which the confidence of shareholders is based.
3.2 Roles and characteristics of corporate boards and firm performance
According to Goodstein, Gautam, and Boeker (1994) corporate boards have three major roles. The first role can be seen as an institutional role, where boards provide a connection between the firm and the external environment in order to secure resources (Williamson, 1996). Second, boards have a monitoring role and appoint or dismiss ineffective executive directors in the board (Barnhart et al., 1994). Third, boards are also engaged in strategic decision making (Fama and Jensen, 1983).4 This thesis particularly focuses on the first and second role.
In the light of these aspects the activity of the board can influence firm performance in two ways, directly and indirectly. First, board activity directly generates beneficial business activities, since strategic decisions conducted in boards are linked to essential matters of the firms’ strategy. Second, consulting and monitoring by non-executive directors have an impact on the decision making of the executives both inside/outside the board. This can have an indirect influence on different dimensions of business activity. In the context of Germany the direct and indirect ways of transmissions can be linked to different corporate boards: Exercising control over the business and the establishment of tactical reasoning are linked to executive managers. Supervision and staffing of managers are main responsibilities of the supervisory board, which has to approve and ratify important decisions such as mergers and acquisitions (Bremert et al., 2009).
In the German context every supervisory board has specific duties and responsibilities but in practice not all of them conduct them similarly. However, it is rational to argue that diverse approaches of cooperation with managers should affect the management’s actions in a different way, and, hence, differently impact firm performance.
Given the strict institutional separation between management and supervision in the German two-tier corporate governance regime one could divide the responsibilities of the supervisory board as follows: First, it appoints and dismisses executive directors and thereby highly exerts influence on a firm’s management philosophy. Second, the supervisory board negotiates with executive directors about their remuneration package and further the non-executive directors
are also responsible for assessing the strategic decisions of the management team. Third, the supervisory board is authorized to constrain the executive director’s scope of action by the right to choose on a set of firm-level decisions in advance. Fourth, the supervisory board constrains the opportunities for managerial freedom by supervising the management’s decision making.
Based on the aforementioned aspects, the supervisory board is considered as an institution that adjusts the firm’s risk-return-relation on behalf of shareholders. In order to do so, they co-determine or limit the risk of investment projects (Bremert et al., 2009).
3.3 Theoretical perspectives
It is widely shared among scholars that no theory of corporate governance can completely explain the general relationship between corporate boards and corporate performance. The interrelation between corporate boards and corporate performance is much more complex. It cannot only be explained by a single corporate governance theory (Jackling, B. and Johl, 2009). This section reviews two major theoretical perspectives of boards and governance mechanisms that are considered relevant for this study namely agency theory and resource dependence theory.
3.3.1 Agency theory
supervisory board, and, hence, agency conflicts can be controlled and minimized, and higher firm performance can be reached (Bathula, 2008).
These agency problems can be eliminated or resolved with well configured contracts that specify the corresponding rights that belong to principals and agents (Jensen and Meckling, 1976). However, the difficulty of developing appropriate contracts, can lead to enhanced managerial discretion which again leads to the same problem (Bathula, 2008).
Given the aforementioned theoretical discussion researchers have suggested internal governance mechanisms, for instance, the board of directors (in this context: the supervisory board) in order to mitigate agency problems. Many academics have utilized the perspective of agency theory to elaborate on the role of the board of directors in influencing corporate performance (Cadbury, 1992; Hampel, 1998; King, 2002).
3.3.2 Resource dependence theory
Resource dependence theory is used to explore the correlation between firms and the elementary and critical resources that are required for superior corporate performance. Moreover, this theory presents a theoretical basis for the role of the board as an essential and critical resource to the corporation (Hillman et al., 2000). In general, these resources amongst others comprise capabilities, assets and knowledge, in order to enhance performance (Daft, 2006). In the light of this understanding the structure of governance and the constitution of the board are considered as a elementary resource that increases the value of the corporation. According to Pfeffer and Salancik (1978) a central reasoning of this theory is that firms try to exercise control over the external environment by gaining resources that are important for the economic success of the firm. Consequently, boards are viewed as an intermediate between the corporation and resources that are necessary in order to maximize performance.
Resource dependence theory also analyzes how the board can help the corporation in getting access to capital (Mizruchi and Stearns, 1988). In order to facilitate the access to financial resources a corporation with a large amount of bank debt might appoint a bank representative to the board (Thompson and McEwen, 1958). Likewise, Mizruchi and Stearns (1988) found that corporations with financial distress are likely to appoint bank representatives to the boards in order to facilitate the access to capital. To sum up, with respect to resource dependence theory the board can not only easily access other resources, but it can rather impact the external environment in its advantage, and in doing so promote corporate performance (Bathula, 2009).
The next chapter adopts the aforementioned insights and other commonly used perspectives to identify key board attributes and their impact on corporate performance.
4. Literature review and hypotheses development
This empirical investigation aims to analyze the impact of supervisory board characteristics on firm performance of large, listed German stock corporations (Aktiengesellschaften).
The vast majority of these empirical studies focus on one-tier board systems prevalent in the US or UK. To my knowledge studies with regard to the interrelation between key supervisory board featrues and corporate performance in the specific and unique setting of the German two-tier board system is limited (e.g., Elston and Chirinko, 2003; Bremert et. al, 2009). For this reason, in the empirical model I particularly refer to the investigations conducted within the US or UK).5 In this chapter I will present the related literature and demonstrate major differences to my model.
The literature review within the framework of two corporate governance theories, form the base for the hypotheses of the thesis. The choice of the explanatory variables is primary based on the set of previous literature and on features of the German governance regime. In the light of the German context I excluded two commonly used variables. First, the aspect of supervisory board independence in terms of proportion of outside directors is negligible in Germany because the supervisory board is a separate instance which is composed by outsiders. Second, in the context of Germany the issue of CEO duality is negligible, since law clearly prohibits the phenomenon of CEO-duality. That is why this thesis project does not consider aspects of supervisory board independence and CEO duality.
Board characteristics are expected to impact firm performance are: board structure in terms of supervisory board size, board activity in terms of supervisory board meetings, board busyness in terms of multiple directorships, board composition in terms of bank representatives on the board and aspects of board qualification in terms of PhDs candidates on the board and supervisory board remuneration. The hypotheses in this study particularly stem from agency theory and resource dependence theory.
The agency perspective is very acknowledged in explaining the duties of boards in decreasing the agency costs (for instance, monitoring costs) (Fama & Jensen, 1983).
Further, the resource dependence perspective views agents as a resource. They can provide business networks and affect the environment in advantage to their corporation (Carpenter & Westphal, 2001).
4.1 Board size
Board structure plays a relevant role in either facilitating or inhibiting the board in fulfilling their work properly.
more non-executive directors on the supervisory board for monitoring and supervision, resource delivery and further ensure the representation of shareholders in the corporation. Consequently, in the light of resource dependence theory a supervisory board with many valuable networks or links to the environment (external environment) can improve and facilitate firm’s access to various elementary resources thus promoting and strengthening corporate governance and corporate performance. Given resource dependencies and resource scarcity, firms establish large boards in order to win directors with different characteristics from multiple backgrounds (Pearce and Zahra, 1992; Bathula, 2008).
In brief, larger boards abet corporations by making sure intense oversight and monitoring of executive directors, providing relevant resources and representing of the shareholders in the corporation.
Consequently, I propose the following:
H1: There is a positive association between supervisory board size and firm performance.
4.2 Board meetings
According to agency theory, the major role of the supervisory board is monitoring the executive directors in order to decrease agency costs and avoid expropriation by the management. The monitoring task of the supervisory board can be enhanced when they meet on a regular basis show high diligence (Carcello, et al., 2002). Zahra and Pearce (1989) argue that diligent meetings are important for good performance of the board. According to Vafeas (1999), it the raise of the number of board meetings is an much more easy way to ensure a more effective governance than to increase or decrease other board characteristics, for instance, ownership structure of directors. In general, the frequency of meetings is considered as a central resource that promotes board diligence. Further, they are good for the board, because they give more time for non-executive directors to monitor management. Raising board meeting frequency, versus changing different board features, is less cost intense and is a better way for corporations to protect shareholder value and promote corporate performance. This leads to the second hypothesis:
4.3 Board busyness
The number of other mandates that non-executive directors hold at other firm boards is also popular under the term “busyness hypothesis” (Ferris, Jagannathan, and Pritchard, 2003). Several studies have demonstrated that board members with multiple mandates positively influence corporate performance (Brown and Maloney, 1999; Miwa and Ramseyer, 2000). Supervisory board members with multiple directorships can add value to the firm when they have valuable links to the external environment and can generate value by accessing critical resources, and suppliers to the firm (Jackling, B. and Johl, 2009). For this reason the busyness aspect has been related with the theory of resource dependencies, because there seems to be at least a theoretical reasoning that directors with a high degree of involvement with the environment open the gate to relevant resources that promote organizational performance (Jackling, B. and Johl, 2009).
Another theoretical reasoning is that a large number of directorships can lead to over-commitment and, hence, inhibit the directors’ ability to supervise and monitor the management adequately and accordingly negatively affect corporate performance (Fich & Shivdasani, 2004). However, it is questionable what a busy director by definition is. Ferris et al. (2003), demonstrated that slightly 6 per cent of board directors held at least three board seats, given the general idea that three mandates indicate a director who is busy.
Thus, with regard to the nexus between multiple directorship and firm performance the theoretical argument is that a supervisory board member with more other mandates has a high level of engagement with the external environment and facilitates the access to various elementary resources that are needed to reach a higher economic success for the firm. This line of argumentation is corresponds with the resource dependence perspective. Therefore I propose the following hypothesis 3:
H3: There is a positive relation between multiple appointments of directors and firm performance.
4.4 Board composition
serves as an equity owner. Bankers are better monitors because they have financial expertise and have better access to relevant information. This is mainly because of their specific role as blockholders and equity lenders; this is clearly demonstrated in the research by Gorton and Schmid (2000) and Edwards and Nibler (2000). However, the results are mixed at best. In their study Dittmann, et al. (2009) found an adverse relationship between bank representatives on the board and firm performance. Consequently, they do not share the opinion that bankers better monitor executive directors. One plausible reason for the different findings is the fact that the equity ownership by banks decreases significantly in recent years due to a taxation law in 2002.
Nevertheless, given that bankers are better monitors due to the information availability or information advantage, I anticipate that they decrease the level of information asymmetry between the supervisory board and the management board and consequently positively influence corporate performance. The latter argument stems from the agency perspective. Accordingly, bank representatives should be able to better monitor executive directors and thereby mitigate agency costs.
As discussed earlier in this study the resource dependency role of the board members also elaborates on the resource provision aspect, here financial resources, and how the board can support the corporation with its access to capital (Pfeffer, 1972; Mizruchi & Stearns, 1988). In the light of the previously mentioned arguments I propose the following fifth hypothesis:
H4: The presence of bank representative on the supervisory board is positively related with firm performance.
4.5 Board qualification
experienced. According to resource dependence theory, PhD qualified directors are strategic resources who enable linkages to relevant resources, hence, promoting firm performance (Ingley and van der Walt, 2001). However, studies about educational qualifications of board members and corporate performance are limited (Bilimoria & Piderit, 1994a; Yermack, 2006). To my knowledge no empirical investigation has studied the interrelation between supervisory board qualification and firm performance in the context of Germany.
Given the previous discussion, I argue that corporations that have more supervisory board members with PhD degree will positively impact firm performance. Accordingly I propose the following hypothesis:
H5: The proportion of directors with PhD degree on the supervisory board is positively related with firm performance.
4.6 Board remuneration
The performance of the supervisory board can be shaped by certain stimuli, for instance, monetary or non-monetary in terms of reputation. This argument is supported by Yermack (2004) who finds evidence that both aspects generate incentives for non-executive board member activity. Cordeiro et al. (2000) found a significant positive association between non-executive director compensation and EPS (earnings per share) growth and further, Lee et al. (2008) examined that non-executive director compensation give incentives that significantly impact corporate performance.
However, the results of previous work are mixed at best. In their study van Ees et al. (2003) expect that director remuneration has an influence on the monitoring task and performance and this in turn would affect the economic success of a firm. However, in their study of Dutch listed firms they did not reveal a significant relationship between board remuneration and firm performance.
H6: Higher average compensation of supervisory board members has a positive effect on firm performance.
5. Research design and measurement model
The thesis examines the nexus between supervisory board features and firm performance. In this chapter I present the measurement model. Further this chapter discusses the sampling, the data sources, the operationalisation of the dependent, independent and control variables, the method used in this research and the model validity.
5.1 Data sample
The total sample comprises German non-financial stock corporations that were publicly listed in one of the two major indices of the German stock exchange (Deutsche Boerse) DAX and MDAX during the period 2004-2007. The initial sample consists of 80 companies. In line with prior research, I eliminated banks from the sample because of their different financial structure (Durnev and Kim, 2003). Furthermore, from the total sample I eliminated all stock corporations that follow the regulation of Kommanditgesellschaft auf Aktien (KGaA), because this is a combination of two different organizational forms, namely a partnership and a stock corporation. This provides 260 firm-year observations from 65 firms in the final dataset. The descriptive statistics are based on this sample of 65 firms.
Table I
Number and proportion of firms by industry classification6
5.2 Data sources
The data for this thesis project stem from sources of secondary data. The majority of data on supervisory board characteristics is taken from annual reports. These annual reports are available on the company websites. For this reason I downloaded the company annual reports for the years 2004 until 2007. From the AMADEUS data set of European firms I obtain the variable that measures firm performance ROA. Further, I utilize the Datastream database at the University of Groningen in order to access financial data for the determination of the second variable that measures firm performance viz. market-to-book value of equity and the control variables firm size and leverage. In case of differences in the data I also proved the validity of the data by comparing them with the data available on the company websites. Finally AMADEUS was used for the determination of the firm-level control variable firm age. The industry classification in this study is based on the German stock exchange classification. The recipient should take into account that my data sources are heterogeneous, as there is no single database with all the data that are relevant for this study. Multiple data sources can lead to discrepancies and bias the outcome of the study.
6The categorizations are in accordance with the German stock exchange classification
Industry Number Percentage
Consumer services 8 12,31
Basic materials 10 15,38
Consumer goods 11 16,92
Industrial manufacturing 18 27,69
Information technology 5 7,69
Pharma & healthcare 6 9,23
Telecommunication 3 4,62
Energy & utilities 4 6,15
5.3 Variable construction
The method of multiple linear regression, precisely ordinary least squares (OLS) is used to test the set of hypotheses. This section introduces the operationalisation and measurement of the dependent variables, independent variables and the control variables.
5.3.1 Dependent variable: Firm performance
The quantitative empirical investigation of the impact of supervisory board characteristics on accounting as well as market based firm performance requires a thoughtful practical use of a proper performance indicator. Nevertheless, the issue about how firm performance can be defined is a debate among researchers. Measures that were used are: Return on investment (ROI), Tobin’s Q, return on assets (ROA), return on equity (ROE), stock returns, earnings per share (EPS), return on invested capital (ROIC) and net profit margin (Bauer, Guenster and Otten, 2004).
However, in general these different performance indicators that were used in prior research can be differentiated into: market based- and accounting based measures (van Ees et al., 2003). It is noteworthy to mention that the reliability of these performance indicators was criticized by many scholars. Kapopoulos and Lazaretou (2007) argue that market based measures are influenced by shareholders psychological perception and thereby influenced by future events. On the other hand accounting based measures are influenced by certain country based accounting standards.
5.3.2 Independent variables: Supervisory board characteristics
The independent variables include supervisory board size, number of supervisory board meetings in the fiscal year, multiple directorships held by directors, bank representatives on the supervisory board, board members with PhD degree and supervisory board remuneration. In general the supervisory board attributes considered in this study can be categorized into the following categories: board structure, board activity (also board diligence), board busyness, board composition, expertise and remuneration. Thus, I consciously take into account the aspects brought up by prior research. However, at this point it is noteworthy to mention that the method of variable construction through single dimensions can be criticized (Larcker et al, 2005).7
Supervisory board size. The supervisory board size is measured by the number of the directors on the supervisory board at the end of the financial year, as can be counted from the annual reports.
Supervisory board meetings. This is measured simply by the number of meetings in a respective fiscal year. Note this measure only gives information about the absolute number of meetings of the board and not the meetings of sub-committees (e.g. audit committee, compensation committee, nomination committee).
Multiple directorships. The variable that is used to proxy for busyness is the average number of appointments of directors or mandates held by non-executive directors on the supervisory board of other stock corporations.
Bank representatives. This is a binary variable which is 1 if at least one bank representative is on the supervisory board of the firm and 0 otherwise.
Supervisory board members with PhD. The indication for this variable is the proportion (percentage) of supervisory board members with PhD degree present on the supervisory board Supervisory board remuneration. The disclosure of supervisory board member remuneration is not obligatory in Germany. Nevertheless, almost every firm discloses the supervisory board members’ remuneration, accordingly this data is to a large extent available through the company annual reports. The variable for remuneration is the average supervisory board member compensation.
7
5.3.3 Control variables: Firm-level variables
The consideration of control variables in a multiple linear regression can positively impact the explanatory power of the supervisory board variables and eliminate or to a certain level reduce the likelihood of biased results caused due to multicollinearity (correlation among the independent variables).
Based on prior literature (e.g. Bai et al. 2004; Joh, 2003; Cho, 1998), I want to include three firm-level control variables that are used in corporate performance studies: These variables are firm size, firm age and leverage.
Firm size because it can be assumed that larger firms (in terms of total assets) enjoy economies of scale (Singh and Davidson, 2003), accordingly, the corresponding agency costs may be higher in larger firms (Jensen and Meckling, 1976). Firm age, as young firms are assumed to generate smaller earnings than firms who have more experience in the market. New firms are in a position where they have to build their market position. In general they face a higher cost structure than older firms (Lipczinsky & Wilson, 2001). On the contrary it is noteworthy to mention that older firms in terms of date of incorporation may be at the end of the product life circle which can lead to less earnings and profits.
Leverage is considered as a general impact of capital structure on firm performance. Firm size.). In order to proxy for firm size I use the natural logarithm of total assets.
Firm age. Firm age is measured by the number of years the firm is on the market from the time the firm established.
Table 2 provides an overview of the variables used for this empirical investigation and their operationalisation.
Table II
Variables and operationalisation
Variables Operationalisation of the variable
Return on Assets (ROA) Operative Income / Assets
Market-to-book value of equity (MB) Market capitalization / Book value per share Supervisory board size (SBSIZE) Number of supervisory board members
Supervisory board meetings (MEET) Total number of board meetings in the fiscal year Multiple directorships (MDSHIP) Average number of other mandates of supervisory
board members
Bank representatives (BANK) Binary variable; 1 if there are one or more bank representatives on the supervisory board, 0 otherwise
Members with PhDs (PHDs) Number of members with PhD present on the supervisory board
Average supervisory board compensation (ASBC)
Natural logarithm of average compensation per supervisory board member
Firm size (FSIZE) Natural logarithm of total assets Firm age (FAGE) Number of years since incorporation Leverage (LEV) Non-current liabilities / total assets
5.4 Method
Based on the theoretical considerations and previous literature I set up the empirical model. The model is developed in order to quantitatively measure the influence of key supervisory board characteristics on corporate performance.
Multiple regression analysis and its variations are major tools in some areas of management disciplines (Thomas, 2006). The method utilized in previous literature analyzing the influence of corporate governance attributes on firm performance applies a linear regression of the following form:
Applying equation (1) to the research objective I obtain:
Firm performance it = α + ΣβSupervisory board characteristics it + ΣγControls it + εit (2)
After inserting the variables from table II into equation (2) I obtain:
ROA / MB it = α + β1SBSIZE it + β2MEET it + β3MDSHIP it + β4BANK it + (3) β5PHDs it + β6ASBC it + γ1FSIZE it + γ2FAGE it + γ3LEV it εt
Here, α is the intercept which is unknown (i.e. constant), β1,2…n designates the unknown parameter for the specific supervisory board variable, γ1,2…n designatess the unknown parameter for the specific firm-level control variables (firm age, firm size, leverage) and εt is a standard error term. For the designation of the dependent and independent variables please see table II.
5.5 Regression model validation
In the methodology of multiple linear regression analysis, precisely, the method of ordinary least squares (OLS), a series of statistical prerequisites of this methodology need to be fulfilled in order to properly perform the method. There are many assumptions that need to be fulfilled. In econometrics, model reliability is certainly one of the most relevant issues in the model design. Certainly, this step is one that is often neglected. An empirical model that neglects the property of the data cannot deliver precise findings to the scientific questions (Thomas, 2006). However, in this thesis I focus on some major assumptions that are considered as important in literature.
Multicollinearity
if a correlation coefficient is 0.7 or higher, irrespective of the sign of the correlation, it is seen to be a very high correlation and accordingly leads to elimination of one variable. Table III presents the correlation matrix for the variables in the regression model.8 However, as can be seen from the table no correlation coefficient is too high to be a problem of multicollinearity. Nevertheless, at this point it is noteworthy to mention that the Pearson correlation coefficient between the two dependent variables (ROA and MB) is considerably high. This is not surprising as both variables measure firm performance.
Table III
Pearson correlations for variables in the regression model
Note: Sample size (n) = 260 firm year observations during the years 2004 - 2007
Heteroscedasticity
This assumption of multiple linear regression deals with homoscedasticity (i.e. constant variances). If the analysis reveals heteroscedasticity, this soft prerequisite is not fulfilled. In this case we face deviation in the variances. Heteroscedasticity does not necessarily lead to biased regression coefficients. In order to test for heteroscedasticity the Breusch-Pagan-Godfrey test is conducted. Therefore, the independent variables are regressed on the squared OLS-residuals. In order to assure homoscedasticity, the product of the associated coefficient of determination (R²) and the number of observation (n) has to be smaller than the critical value of the chi-distribution with the corresponding degree of freedom (R²n < critical value). Consequently, the null hypothesis of homoscedasticity is rejected if the value of the test statistic is larger than the critical value.
8 The correlation coefficients can give expectations about the possible outcome of the hypotheses’ testing.
VARIABLES ROA MB SBSIZE MEET MDSHIP BANK PHDs ASBC FSIZE FAGE LEV
ROA 1 MB 0,798 1 SBSIZE 0,184 0,221* 1 MEET 0,125* 0,143 0,056 1 MDSHIP -0,146** -0,103* 0,061 0,008 1 BANK 0,077 0,128 0,050 0,070 0,114* 1 PHDs 0,054 0,036 0,228 0,001 -0,001 0,034 1 ASBC 0,102* 0,217* 0,127* 0,198* 0,063 0,162* -0,124 1 FSIZE -0,026 -0,139 0,298 -0,035 -0,115 -0,178 0,130* -0,306* 1 FAGE 0,035 0,011 0,045 -0,014 0,039 -0,020 0,046 -0,037 -0,129 1 LEV -0,300** -0,321** 0,132* 0,010 0,023 0,092 0,037 -0,066 0,064 0,039 1
The outcome of the Breusch-Pagan-Godfrey approach confirms the null hypothesis that there is no heteroscedasticity.9 Please see Appendix B for a detailed explanation of the Breusch-Pagan-Godfrey procedure (Schira, 2003).
Outliers
Another important statistical feature that is important in the OLS-methodology is the existence of potential outliers in the sample. The issue of outliers is important in this study, as they can bias the findings of the empirical analysis. In order to identify outliers in the data sample different statistical methods can be used. For instance, the existence of potential outliers can be revealed by exploring the Mahalanobis distances (Tabachnick & Fidell, 2006). This distance is a more dimensional approach of a z-score. A potential outlier is considered as an outlier if the probability connected with its distance is below 0.001. However, in this data sample one outlier was detected, that is the firm Thyssenkrupp AG. The fact that this outlier can have an influence on the analysis leads to the exclusion of this firm from the sample firms.10
6. Empirical results and discussion
This chapter presents the empirical results of the study. First, the descriptive statistics are described. Second, the presentation and discussion of the findings from the quantitative analysis follow.
6.1 Descriptive statistics
Table IV provides descriptive statistics that quantitatively summarizes the data set for both the firm performance variables and the supervisory board variables employed in this study. For each of the variables the mean value, standard deviation value, median value, minimum value and maximum value is presented. The sample size comprises 260 firm year observations. Please note that I present data for the variables that are not converted. In order to meet statistical assumptions of the multiple linear regressions, some of the variables need to be converted. For this reason I take the natural logarithm of the data (For instance, natural logarithm of total assets).
9 The null hypothesis that there is homoscedasticity is approved; (R²n < critical value), (12.32<15.09; p=0.01)
10
Table IV
Descriptive statistics for variables in the regression model
The dependent variables: Return on assets (ROA) and Market-to-book value of equity (MB). The independent variables: Supervisory board size (SBSIZE), Number of supervisory board meetings (MEET), Multiple directorships (MDSHIP), Bank representatives (BANK), Members with PhD (PHDs) and Average supervisory board compensation (ASBC). The firm-level control variables are firm size (FSIZE), firm age (FAGE) and leverage (LEV).
Note: Sample size (n) = 260 firm year observations during the years 2004 - 2007
With respect to financial performance, on average the sample firms seem to be stable as can be seen by the ROA (ROA mean 7.42 per cent).
Further, table IV points out that supervisory board size ranges from 3 to 23 members. Moreover, the table shows that an average board entails approximatively 14 members. They meet between 4 and 9 times in one year, with a mean and standard deviation of 5.33 and 1.46 respectively.
The number of appointments a director holds on a supervisory board of a firm in the sample ranges between 0 and 6 with a mean (and median) of about 3.56 (3.11).
Notably, 66 per cent of the stock corporations have more than one bank representative on their supervisory board. This is not consistent with their blockholder position in German stock corporations (Dittmann et al. 2008).
Paying attention to board remuneration the table shows that directors on average earn 69.130 Euro. Another remarkable observation in the remuneration variable is that remuneration per member ranges from 3.180 to 228.650 Euro.
Performance variables N Mean Std.dev Median Minimum Maximum
ROA (%) 260 7,42 9,73 6,48 -39,54 55,33
MB (ratio) 260 2,01 1,68 1,46 0,62 2,23
Supervisory board variables N Mean Std.dev Median Minimum Maximum
SBSIZE (#) 260 14,32 5,48 16 3 23 MEET (#) 260 5,33 1,46 5 4 9 MDSHIP (#) 260 3,56 1,03 3,11 0 6 BANK (1 or 0) 260 0,66 0,48 1 0 1 PHDs (%) 260 36,03 9,28 38,33 14,66 58,86 ASBC (T €) 260 69,13 57,9 55,41 3,18 228,65
Control variables N Mean Std.dev Median Minimum Maximum
FSIZE (Log of Total assets) 260 12,78 2,24 13,75 8,44 15,92
FAGE (Years) 260 81,26 60,44 80,5 2 251
The mean of total debt and also the median of total debt in the sample seem to be very high (46.98 and 48.35 per cent of total assets respectively) .They vary between 5.41 and 87.56 per cent of total assets.
As can be seen from the descriptive statistics table the results also show the stock corporations are, because the mean is 81 years and varies between 2 and 251 years.
6.2 Presentation and discussion of the OLS estimation results
Table V and VI present the results of the OLS estimation, describing the impact of supervisory board characteristics on firm performance. Further, table V and VI shows the findings for the analysis with industry dummies and without industry dummies (Model 1 and Model 2). The high F-statistic with a significance of less than 5 percent in both model specifications for ROA and MB indicates that any changes in the supervisory board variables give information to the explanation of the changes in accounting as well as market based firm performance.
Table V
OLS Regression results: Dependent variable Return on Assets (ROA) This table shows the results of the OLS regressions of ROA on supervisory board variables and firm-level control variables. In model 1 the analysis does not control for industry effects, whereas in model 2 the analysis controls for industry effects.
ROAt = α + β1SBSIZEt + β2MEETt + β3MDSHIPt + β4BANKt + β 5PHDst + β6ASBCt +
γ1FSIZEt + γ2FAGEt + γ3LEVt + (γ4INDt) + εt
*, **, *** = statistically significant at less than the .10, .05, and .01 significance level
Model 1 Model 2 Coefficient Coefficient
Supervisory board variables
Table VI
OLS Regression results: Dependent variable Market-to-book value of equity (MB) This table shows the results of the OLS regressions of MB on supervisory board variables and firm-level control variables. In model 1 the analysis does not control for industry effects, whereas in model 2 the analysis controls for industry effects.
MBt = α + β1SBSIZEt + β2MEETt + β3MDSHIPt + β4BANKt + β 5PHDst + β6ASBCt +
γ1FSIZEt + γ2FAGEt + γ3LEVt + (γ4INDt) + εt
*, **, *** = statistically significant at less than the .10, .05, and .01 significance level
Designation:
MB = Market to book value of equity
SBSIZE = Number of supervisory board members
MEET = Absolute number of supervisory board meetings in the fiscal year MDSHIP = Average number directorships of supervisory board members
BANK = Binary variable; 1 if more than one bank representative is on the board PHDs = Number of members with PhD present on the supervisory board ASBC = Natural log average compensation per supervisory board member FSIZE = Natural log of total assets
FAGE = Number of years since incorporation
LEV = Non-current liabilities divided by total assets
Model 1 Model 2 Coefficient Coefficient
Supervisory board variables
Board size
The variable tested in hypothesis 1 is supervisory board size. With respect to this variable my analysis yields that supervisory board size does not have an influence on both accounting as well as market based firm performance, because the regression coefficient turns out to be insignificant. Further, even after including industry dummies the findings remain insignificant. One possible reason for this finding can be the legal fact, that the size of the German supervisory board is predefined by law, as it depends on the number of employees in the stock corporation (please see chapter 2). In contrast to the one-tier governance system prevalent in the US and UK, board size is regulated in the German two-tier regime. As stated earlier in this thesis project stock corporations that achieve a predefined number of employees are obliged to accept a minimum number of directors on the board. Given the fact that the data set shows that the minimum number of directors on average is not essentially exceeded, I argue that board size is not a strategic tool for optimization issues and therefore is a strict straightforward rule that has to be followed. Consequently, the results do not provide support for hypothesis 1 and further do not give evidence for the resource dependence perspective. Moreover, this result does not confirm the findings from Yermack (1996) and Eisenberg, et al. (1998) who revealed a significant negative association between board size and firm performance in the US. For the Netherlands van Ees et al. (2003) found a negative impact of the size of the supervisory board on firm performance. However, the comparison of findings between other countries and Germany concerning board size should be made with caution, in particular because of the differences in the legal framework (e.g. minimum number of directors on the board is prescribed by law).
Board meetings
Bank representatives
In all specifications the corresponding regression coefficient shows a positive sign but only in one case it is significant at the 5 percent level (ROA with industry dummies) and at the 10 per cent level (MB with industries included). This result is not strong, however, this finding still supports hypothesis 4. Accordingly, both the agency and resource dependence perspective are in line with the result. Bankers are better monitors because they reduce the level of information asymmetry between the supervisory board and the management. The findings provide evidence for the fact that bankers have indeed a better access to information. In their role as blockholders they have a high incentive to monitor management. Thereby, they reduce agency costs due to the lower degree of information asymmetry between the management and the supervisory board. Furthermore, the result of the empirical analysis is in line with the findings of Gorton and Schmid (2000).
Members with PhD
Hypothesis 5 predicted that there is a positive association between the proportion of directors with PhD degree and corporate performance. This hypothesis is rejected by the findings, as the corresponding regression coefficient is not significant. All estimators in the models are not significant. It appears that PhD qualified members (skills of research and analysis), do not benefit the board and, hence, do not positively affect firm performance. Given the argument by Yermack (2006) financial and accounting based expertise are important and influence firm performance. He revealed that financial qualifications of board members are in advantage to firms in the United States. Given this fact, it is useful to find out what skills and competencies are important in the German context. In other words, it is necessary to find out what skills are needed in order to increase board performance.
Average supervisory board compensation
compensation like it is common in the United States. Given the findings, it seems that compensation fulfills its role as an incentive mechanism.
Moreover, I disagree with the argument that remuneration of directors is a reaction of firm performance, or put differently is dependent on firm performance. This would mean that more profitable corporations would pay their directors more. I disagree because the data show that compensation is at a relatively small level. Accordingly, not only successful companies can afford a higher payment, as the sample also comprises non-profitable firms.
Given the findings of the empirical investigation share the opinion that supervisory board compensation can be applied as an incentive tool in order to allow for better board performance to positively affect the overall corporate performance. Consequently, the results support hypothesis 6.
7. Conclusion
In the recent past, there has been a huge interest on the roles and responsibilities of boards in the economic success of firms, especially because of diverse corporate scandals.
To understand the role of boards, I relied upon the agency theory and resource dependence theory. This thesis intends to fill a gap in the literature on the interrelation between supervisory board characteristics and firm performance by empirically examining this issue in the light of German stock corporations. I test my hypotheses on a sample of 65 non-financial firms listed on the German stock exchange over a four year period from 2004 to 2007. I conducted the research within the German one-tier board setting, in order to answer the following research question:
How do supervisory board characteristics affect the firm performance of non-financial firms listed on the German Stock Exchange?
To answer the research question an empirical research on supervisory board characteristics and its impact on firm performance was conducted.
The data for this study was taken from publicly listed non-financial firms on the German stock exchange during the years 2004 – 2007.
Contrary to the second hypothesis statement the number of board meetings is unrelated to my performance measures. This finding may suggest that there is a non-linear association between number of meetings and corporate performance. The relation might be more complex than a linear pattern.
Furthermore, a negative association was found between multiple directorships and firm performance.
Moreover, although being not very strong, the findings show that bank representatives on the supervisory board exert a positive influence on accounting as well as market performance. Accordingly, both the agency and resource dependence perspective are in line with the finding.
A remarkable finding of this study is that the PhD degree of board members, with skills of research and analysis, does not benefit the board and moreover does not affect firm performance at all.
As hypothesized, the average supervisory board director compensation is consistently interrelated with firm performance. This supports the general idea that remuneration can serve as an incentive tool and motivate board members to work diligently.
In summary, I found that supervisory board characteristics (multiple directorships, bank representatives on the board, board remuneration) show an association with firm performance. Given the findings the two theories, agency and resource dependence theory, were only partly supported by the empirical results of this thesis project. Moreover, the results indicate that no single corporate governance theory explains the interrelation between board characteristics and firm performance and that the results of empirical studies within the Anglo-Saxon one-tier regime cannot be directly assumed to hold within the German two-one-tier regime.
This study contributes to academic knowledge in several ways:
order to adequately answer this question a research has to consider both specific business aspects of a firm and the legal environment the firm is operating.
8. Limitations and potentials for future research
As with other empirical studies in this field the findings presented in this thesis are subject to some limitations that the recipient should take into account. This chapter presents the limitations of the research and gives recommendations for future research.
First, the data belongs to German stock corporations. The regardless transfer of these findings to other countries is not possible which limits the external validity of this thesis. Therefore, these findings cannot be generalized to other firms in other countries.
Second, only a limited number of variables are taken into account in this study, whereas there are plenty of other dimensions of board. This further limits the generalizability of the findings. Consequently, future empirical study should take more or other variables into account, as the database for German firms is increasing.
Third, I acknowledge that my sample is quite small. A deeper analysis, with more firms in the sample would allow for more generalizability. Accordingly, this aspect can be considered by future research which should settle on more firms.
Fifth, this study does not reveal a significant positive relationship between PhD degree of supervisory board members and corporate performance. Following the work of Yermack (2006) it is necessary to identify specific skills that would benefit the work of the board. Yermack (2006) suggested considering accounting and financial skills, as they would benefit the board.
Sixth, the study has examined the impact of supervisory board variables on firm performance, as measured by return on assets and market-to-book value of equity. It is interesting whether the findings would remain robust if the variable Tobin’s Q is applied. So, it may be useful to re-analyze this study with other commonly used firm performance measures.
Further, as with all quantitative multiple regression analyzes, endogeneity can be a caveat. The method of instrumental variables (IV) can be applied in order to account for endogeneity. Here, it would be difficult to select the variable that serves as the instrument. Another limitation with respect to the method is that I pool the data as if there is no variance over time. Consequently, I applied cross-section analysis while in fact I have a panel of four years. Thus, the neglect of time variation is a limitation.
can reveal specific contingency conditions the supervisory board characteristics and corporate performance measures may be dependent on.
