THE EFFECTS OF BOARD STRUCTURE ON CSR PERFORMANCE: EVIDENCE FROM THE EUROPEAN BANKING INDUSTRY

THE EFFECTS OF BOARD STRUCTURE ON CSR

PERFORMANCE: EVIDENCE FROM THE EUROPEAN

BANKING INDUSTRY

Mark Vermaning University of Groningen Faculty of Economics and Business

Supervisor Boris van Oostveen June 2017

ABSTRACT

By using a dataset covering a 12-year time-period, this study extends research on corporate governance and corporate social responsibility into the European banking sector, and accounts for time specific influences in examining the effects of board independence, board size and board gender diversity on CSR performance in 16 European countries. The results of this study suggest a positive relationship between the percentage of independent board members and the CSR performance, but show no evidence of a relationship between board size and CSR performance. Finally, in contradiction to previous research, this study finds no significant relationship between gender diversity and CSR performance after the inclusion of time-effects. Additionally, this study shows that time-effects can have an important effect on the results of the study and should be incorporated in future CSR research and the results suggest that there appear to be no generalization issues between the different banking industries.

Keywords: Corporate governance, corporate social responsibility, board of directors, board

independence, board size, gender diversity

JEL classification: G30, G34, J16, M14, Q55

I

NTRODUCTION

Research on the US banking industry and the banking industries of several developing countries suggests that that various board structure characteristics can have a significant positive effect on the CSR performance of the banks. However, it is unclear if these results can be transferred to the European banking industry. By using a dataset covering a 12-year time-period, this study extends research on corporate governance and corporate social responsibility into the European banking sector, and accounts for time specific influences in examining the effects of board independence, board size and board gender diversity on CSR performance in 16 European countries.

In the past 15 years, the attention of firms for Corporate Social Responsibility (CSR) has increased significantly. The attention stems from firms realizing that they can have a big influence on society, from a demand in the market for sustainable products and production, and from customers, employees, investors and other stakeholders who have encouraged firms to make considerable investments into CSR (McWilliams and Siegel, 2000; Pondeville, Swaen, and De Rongé, 2013). Firms now have to create value while thinking about the consequences it has on the environment and society (Van Marrewijk, 2003). Scientific research investigating the corporate social responsibility of firms increased as well.

that corporate governance at banks was ineffective at preventing the practices that lead to large economic problems in society. Therefore, more empirical research is needed on the role of corporate governance at banks, and research that combines corporate governance and CSR could lead to useful insights (Grove, Patelli, and Victoravich, 2011). One of the most important corporate governance mechanisms in companies is the board of directors. The board has a lot of influence on the CSR strategy and the CSR policies of the firm. Consequently, it is useful to understand the effects that the board can have on the level of corporate social responsibility of the bank. There are a few studies that looked into this relationship, but they mainly investigated the effects of the board of directors on the CSR of banks in developing countries such as Bangladesh, Pakistan and Turkey, as well as the banking market of the United States (Khan, 2010; Sharif and Rashid, 2014; Jizi, Salama, Dixon, and Stratling, 2014; Kiliç, Kuzey, and Uyar, 2015). These, studies find that the composition of the board can have an influence on the level of CSR.

Halme, 2009). Because of the mentioned differences, research on the influence of corporate governance on corporate social responsibility needs to be extended into the European banking sector.

The results of this study suggest a positive relationship between the percentage of independent board members and the CSR performance, although the effect seems to mainly stem from a strong influence of board independence on the economic CSR pillar. Furthermore, the results show no evidence of a relationship between board size and CSR performance. Finally, in contradiction to previous research, this study finds no significant relationship between board gender diversity and CSR performance after the inclusion of time-effects.

This study has three main contributions. Firstly, this study adds to the corporate social responsibility and corporate governance literature by providing a deeper understanding of the relationship between board structure and CSR performance. Secondly, the results of the study show that there does not appear to be any generalization issues between the different banking industries. And thirdly, by controlling for effects that influence the CSR performance of European banks at given moments in time but are consistent across firms, this study shows that time-effects can have an important effect on the results of the study and should be incorporated in future research.

The next section discusses the main theories and concepts that provide the theoretical perspective for the rest of this study, as well as a review of previous literature leading to three new hypotheses. The third section discusses the methodology and the different models that are used. The fourth section describes the data set and the mea sures in the model. The fifth section presents the results of the data analysis. And finally, the sixth section provides a conclusion to the study, as well as the limitations and suggestions for further research.

L

ITERATURE

R

EVIEW

the interests of the owners of the firm and its management aligned. Legitimacy theory and stakeholder theory explain the rationale behind the CSR activities of firms and link CSR to corporate governance and the board, by showing that CSR can be important for shareholder wealth. Previous research on CSR and corporate governance is discussed. Additionally, resource dependence theory is introduced to support the hypotheses development at the end of this chapter.

Agency theory and corporate governance

The board is one of the most important corporate governance structures in a firm, and through the monitoring and disciplining function it aligns the actions of the managers with the interests of the shareholders (Jensen and Meckling, 1976; Eisenhardt, 1989). Agency Theory discusses the relationship between the principles, often the shareholders, and the agents, which are the company’s executives and managers (Jensen and Meckling, 1976). The principles hire the agents to manage the firm on their behalves and the agents are expected to act in the best interest of the principles (Eisenhardt, 1989). However, because of a misalignment of interests and risk preferences this can lead to problems (Eisenhardt, 1989; Johnson and Greening, 1999). Agency theory therefore provides the basis for corporate governance and the installment of a board which is concerned with fixing these problems (Cadbury, 1992; Adams, Hermalin, and Weisbach, 2010; Jizi, Salama, Dixon, and Stratling, 2014). Managers who fail to meet the expectations of the shareholders can be punished or eventually be fired by the board. From the perspective of agency theory, shareholders want to maximize their wealth by increasing the performance and value of the firm, and therefore the managers and the board are expected to do the same.

have investigated the influence of board size on the performance of the bank. Adam and Mehran (2012) studied the performance of banks from a 34-year time-period and found a positive effect of board size on performance. Belkhir (2009) finds the same result. Additionally, Pathan (2009) finds that having a smaller board increases risk -taking in the firm. Staikouras et al. (2007) have the only study that find a negative effect between board size and the performance of the banks.

Stakeholder theory, legitimacy theory, and corporate social responsibility

Stakeholder theory and legitimacy theory help to explain the concept of corporate social responsibility. Stakeholder theory states that to be able to be successful and survive in the long-run, firms need the support of their stakeholders (Donaldson and Preston, 1995). A stakeholder is a person or entity that has an interest or a concern in the firm, and stakeholders can include investors, customers, managers and other employees, business partners, the government, and society in general (Donaldson and Preston, 1995). Legitimacy theory is an extension of stakeholder theory and is often used to explain the social reporting practices of firms (O’Dwyer, 2002; Sharif and Rashid, 2014). Legitimacy theory describes a social contract between a firm and society, and states that the firm should target its activities at meeting the requirements of this contract (Cormier and Gordon, 2001). In practice, this means that firms should generally follow the expectations of society and respond to social pressures (Kiliç, Kuzey, and Uyar, 2015). Furthermore, legitimacy theory describes is successful in meeting the expectations of society, it will legitimize the company and create a trustworthy image (Cormier and Gordon, 2001). Together, legitimacy theory and stakeholder theory describe the relationship that a bank has with its stakeholders. Corporate social responsibility is the overarching term for managing this relationship and it therefore builds on stakeholder theory and legitimacy theory.

To motivate the measurement of CSR performance in this study, and since corporate social responsibility is such a broad term, a definition supported by theory is necessary (Van Marrewijk, 2003). The concept of the triple bottom line provides a theoretical basis and describes corporate social responsibility by dividing it into three parts: economic, social and environmental responsibility (Dyllick and Hockerts, 2002). It leads to the following definition of CSR: “Meeting the social and environmental needs of a firms’ stakeholders without

2003). In this study, CSR performance will follow this definition and is based on economic performance, social performance and environmental performance.

Stakeholder theory and legitimacy theory not only explain the concept of corporate social responsibility, they also provide a link between CSR and corporate governance. Evidence shows that focusing on corporate social responsibility does not necessarily compromise profitability, but on the contrary, can enhance the performance, longevity and value of the firm (Giannarakis and Theotokas, 2011; Jizi, Salama, Dixon, and Stratling, 2014). After the financial crisis of 2008, pressure on the banks to improve their corporate social responsibility activities increased significantly. Matching the perspective of legitimacy theory, banks now have to take into account a long-term view and acknowledge and act on their obligations to society to regain the lost trust and survive as a company (Giannarakis and Theotokas, 2011). CSR activities of banks are not only necessary to survive, but they can also create financial and strategic advantages (Salama, Anderson, and Toms, 2011). CSR activities can decrease the firm risk and increase the banks’ profits and share price (Salama, Anderson, and Toms, 2011; Jizi, Salama, Dixon, and Stratling, 2014). Furthermore, the CSR activities of banks can increase brand loyalty, customer satisfaction and employee commitment (Matten; 2006; Mackenzie, 2007). This means that corporate social responsibility activities can have a positive effect on shareholder wealth, and a successful board is therefore expected to stimulate these activities and increase CSR performance.

Corporate governance and corporate social responsibility research

So far, several papers have investigated the influence of board structure on the corporate social responsibility of firms. Most studies focus on US firms. Various studies investigate the effects of independent directors on the CSR performance, as well as the effects of having female board members, and find significant positive effects (Zahra, Oviatt, and Minyard, 1993; Ibrahim, Howard, and Angelidis, 2003; Bear, Rahman, and Post, 2010; Jo and Harjoto, 2011; Zhang, Zhu, and Ding, 2013). Post, Rahman, and Rubow (2011) find similar results, but also add a significant influence of board member age.

banking industry often use CSR disclosure as a double for CSR performance where a higher CSR disclosure relates to a better CSR performance. Jizi, Salama, Dixon, and Stratling (2014) investigate the effects of corporate governance attributes on the level of CSR disclosure of US banks and find that board independence and board size are positively related to CSR disclosure. Additionally they find that, in contrary to the expectations, CEO duality is positively related to CSR disclosure. Kiliç, Kuzey, and Uyar (2015) look at the Turkish banking market and find positive effects of board size, board composition and board diversity. Khan (2010) studies the private commercial banks of Bangladesh and finds no impact of gender diversity in the board of directors on the CSR activities of the banks, but outside directors and foreign nationalities have a significant influence. Finally, Sharif and Rashid (2014) study the CSR performance of commercial banks in Pakistan and find a positive influence of having outside directors on the board.

Based on previous literature a significant influence of board structure on CSR performance is expected, however, evidence on this relationship does not yet exist for European banks. This paper will extend research on corporate governance, the board of directors, and CSR into the European banking market and focus on three board characteristics: board independence, board size and gender diversity. The hypotheses for this study are developed below.

Hypotheses development

From an agency theory perspective, increasing the number of independent directors on the board improves the ability of the board to monitor and control management, and will therefore have a positive effect on CSR performance. Independent directors allow for more objective monitoring because they are less closely involved with the daily operations of the firm and are less reliant on the goodwill of the CEO. Additionally, their remuneration is not linked to the performance of the firm, which makes them less focused on short-term financial results (Cheng and Courtenay, 2006; Jizi, Salama, Dixon, and Stratling, 2014).

H1: The percentage of independent directors on the board is positively related to the CSR performance of European banks.

From an agency perspective, a larger board is expected to be better suited at handling the high workload that is present at banks, and to be more effective at its monitoring and control function. Therefore, a larger board is expected to positively influence CSR performance. Having a smaller board does have benefits. A smaller board is more effective at coordination and communication, and brings a higher level of commitment and individual accountability for the board members (Yermack, 1996; Jizi, Salama, Dixon, and Stratling, 2014). However, smaller boards can become ineffective when the complexity of the firm is high (Pathan, 2009). A complex firm has a high workload, and a larger board will be able to divide the workload over more people and is therefore more effective at its tasks (De Andres and Vallelado, 2008; Belkhir, 2009; Grove, Patelli, and Victoravich, 2011). Since banks are seen as complex firms, a larger board is likely to have a positive effect on the CSR performance (Jizi, Salama, Dixon, and Stratling, 2014; Kiliç, Kuzey, and Uyar, 2015).

As an addition to the aforementioned theories, resource dependence theory helps as a fourth theory to support the second and third hypothesis. Resource dependence theory views a firm as an open system which is dependent on contingencies in the external environment (Hillman et al., 2009). The environment controls resources that are important to the firm, and the firm has to try to obtain the resources that it needs (Pfeffer and Salancik, 2003). Resource dependence theory argues that the task of the board is to minimize resource dependence, provide access to channels of information, provide access to specific resources, and improve legitimacy (Pfeffer and Salancik, 2003; Hillman, Withers, and Collins, 2009).

From a resource dependence perspective, a larger board provides a more knowledge and access to resources, and will be better at its tasks. Therefore board size is expected to positively influence CSR performance. A larger board means more people, which relates to a higher range of expertise that is available to the firm (Jizi, Salama, Dixon, and Stratling, 2014). Additionally, Carpenter and Sanders (1998) find that a larger board is related to a higher level of internalization on the board, leading to a larger range of critical resources. It leads to the following hypothesis:

Boards with a higher percentage of females on the board are better at monitoring and controlling the management of the bank, and therefore a positive influence on CSR performance is expected. Banks are considered to be complex organizations and in a complex environment a greater variety of perspectives and talents is beneficial for problem solving tasks (Anderson, Reeb, Upadhyay, and Zhao, 2011). Having more females on the board will provide this variety, since their perspectives and talents differ from those of men (Anderson, Reeb, Upadhyay, and Zhao, 2011). Additionally, Adams and Ferreira (2009) find that females have less attendance problems. Banks with a higher percentage of females are therefore expected to be better at the monitoring and control tasks.

Furthermore, having more females on the board can be seen as a response to the wishes of stakeholders, which is positive for the CSR performance of the firm. Hillman, Shropshire, and Cannella (2007) argue that, in recent years, firms often receive pressure to increase the number of female board members. Increasing the percentage of females on the board will improve the firm’s reputation and increase its credibility. From a stakeholder and legitimacy perspective, this improvement in reputation and credibility means an increase in CSR performance, because the wishes of the company’s stakeholders are incorporated in the firm’s actions.

Finally, because of the different perspectives and solutions that females bring to the board, the board will be better in completing its tasks, which will increase CSR performance. Hillman et al. (2007) argue that female board members are able to provide more and different solutions. Moreover, female board members are more likely to have expert knowledge in different areas than business, and can therefore bring in new perspectives (Hillman, Cannella, and Harris, 2002). Carter, Simkins, and Simpson (2003) also state that gender diversity can increase the collective knowledge that is present in the board. Additionally, Stearns and Mizruchi (1993) argue that females are able to provide the firm with more and alternative resource channels. From a resource dependence perspective, this means that the different perspectives, solutions and resource channels will enable the board to perform better and increase CSR performance. Consequently, the third hypothesis is:

M

ETHODOLOGY

The aim of this study is to investigate the influence of board independence, board size, and gender diversity on the CSR performance of European banks, while controlling for several other variables. Since the 2008 financial crisis, researchers have realized that corporate social responsibility is very important for banks. Research has shown that corporate governance attributes can influence the CSR activities of banks, but research on this relationship does not yet exist for the European banking industry. The following equation specifies the initial regression model:

Performancei,t = β0 + β1BIi,t + β2BSi,t + β3GDi,t + β4Controlsi,t +

εi,t

In this model (1), β0 represents the constant and β1, β2, β3, and β4 are the regression

coefficients for BI, BS, GD, and Controls respectively. Performance is a proxy for CSR performance, BI is a proxy for board independence, BS is a proxy for board size, GD is a proxy for gender diversity on the board, and Controls is a proxy for the control variables. Furthermore,

εi,t

The data analysis will consist of three parts. Firstly, this study will use a similar type of analysis as performed in most previous research, which is a pooled ordinary least squares regression (OLS). This forms a benchmark for the other estimated models and gives a basic indication of the effects of the corporate governance attributes on the CSR performance of European banks. Additionally, it allows for a direct comparison with the results from the previous studies that used this approach. Table A.II in the appendix presents an overview of the various methodological approaches used in previous studies.

pooled OLS model should be rejected in favor of the fixed-effects model. The Breush-Pagan Lagrange Multiplier test for random effects is used to compare the random-effects model and the pooled OLS model. Finally, if both the fixed-effects and the random-effects model are superior to the pooled OLS model, a Hausman test will compare the fixed-effects model and the random effects model. The Hausman test looks at the correlation between unobservable heterogeneity and the explanatory variables, and if the null hypothesis – the difference in coefficients is not systematic – is rejected, firm fixed-effects will be used. If the null hypothesis is not rejected, the random-effects model will be used.

The fixed-effects model is often used in financial models, however, recent research indicates that using the random effects model can have several advantages and is often useful in analyzing panel data. Bell and Jones (2015), for example, argue that the random-effects model provides a greater flexibility and generalizability in creating a model and in analyzing the results. Bell, Fairbrother, and Jones (2016) even argue that the random-effects model can sometimes be the preferred method over the fixed effects model. Furthermore, the random-effects model allows for time-invariant variables to be included in the model, which could explicitly show influences that in the fixed-effects model are only included in the constant. Bell and Jones (2015), add that the flexibility of the random-effects model provides the opportunity to extend the random-effects model in various ways. Therefore, a third variation of the model is included in this study.

The third part extends the methodology as used by Kiliç, Kuzey, and Uyar (2015) by adding time-effects to the firm fixed-effects model or the random-effects model. Most previous studies use data from a 3-4 year time-period, however, because this study uses data from a 12 year time-period, the data is well suited to test for characteristics that affect the CSR performance of European banks on specific moments in time. It prevents problems caused by dependence in the residuals from effects that influence the CSR performance of all firms at a given point in time. To test whether there are time-effects present in the data a Wald test is executed. If the null hypothesis, the time-effects are equal to 0, is rejected, a time-dummy variable will be included in the model, which will adjust the estimates accordingly.

Wooldridge test indicates if there is an autocorrelation issue. If this is the case, clustered standard errors will be used to account for both heteroscedasticity and autocorrelation. As a final test, a Jarque-Bera test is conducted to test if the variables follow a normal distribution. When variables are non-normal, the data can possibly be improved by using the natural logarithm of the variable’s data.

D

ATA

The dataset consists of 63 listed banks operating in 16 different European countries covering the years 2004 to 2015. It leads to a total of about 750 observations. Since most similar CSR research works with about 200 to 300 observations, this is considered to be quite a large sample. The firm characteristics are collected from Thomson Reuters’ Worldscope database, which contains detailed financial information on firms from all over the world. The CSR rating and the board characteristics are collected using Thomson Reuters’ ESG Asset4 database, which contains mainly environmental, social and governance information of a large number of organizations. The definitions of the variables that are used in this study are summarized in Table I.

CSR rating

Table I

Definitions of the firm and board characteristics for 63 listed European banks in the period 2004 to 2015

The ESG Asset4 CSR rating originally consists of scores on four different pillars: economic performance, environmental performance, social performance and corporate governance performance. Since this study wants to investigate the effects of corporate governance on the CSR performance, the corporate governance pillar will not be used. It leads to a CSR rating consisting of: economic performance, scored on client loyalty, performance and shareholder loyalty; environmental performance, scored on resource reduction, emission reduction and product innovation; and social performance, scored on employment quality, health and safety, training and development, human rights, community responsibility, and product responsibility. Economic rating are based on a total of 24 key performance indicators, environmental ratings are based on a total of 70 key performance indicators, and social ratings are based on 88 key performance indicators. The three main pillars are fused on an equally-weighted basis to form the final CSR rating.

Definitions Source

Firm characteristics

Leverage Total debt over total assets Worldscope

Firm size The distance of each bank's log total assets from the sample

mean, scaled by the log total assets' standard deviation Worldscope Firm risk Systematic risk indicated by the firm’s beta Worldscope CSR performance (ln) The natural logarithm of the firm's CSR performance based on

the CSR rating ESG Asset4

Board Characteristics

Board independence The number of total outside directors over the number of total

board members ESG Asset4

Board size The total number of board members ESG Asset4

Gender diversity The number of female board members over the number of total

board members ESG Asset4

Audit committee independence The number of independent audit committee members over the

number of total audit committee members ESG Asset4 Meeting attendance The average number of board members present at a board

meeting over the total number of board members ESG Asset4 Board meetings (ln) The natural logarithm of the total number of board meetings ESG Asset4 CEO separation The CEO of the firm is not a board member ESG Asset4 Board expertise

The number of board members with an industry specific or strong financial background over the number of total board members

Independent Variables

The three main independent variables are: board independence, measured by the number of outside directors over the number of total board members; board size, measured by the total number of board members; and gender diversity, measured by the number of female board members over the number of total board members. The data for these variables is retrieved from the ESG Asset4 database.

Control Variables

Several control variables are used to test for other influences. Table A.II, presented in the appendix, shows the various control variables that are used in previous studies. This study will control for audit committee independence, measured by the number of independent committee members over the total number of members on the audit committee; CEO separation, rated as one if the CEO is not a board member and rated as zero if the CEO is a board member; board meetings, measured by the number of yearly meetings of the board; board expertise, measured as the number of board members with an industry specific or strong financial background over the total number of board members; firm risk, measured by the firm’s beta, which is assumed to be constant over time; leverage, measured by total debt over total assets; firm size, measured by the distance of each bank’s log total assets from the sample mean and scaled by the log total assets’ standard deviation; and meeting attendance, measure by the average number of board members that is present at the board meetings divided by the total number of board members.

therefore lead to better monitoring and decision-making. Based on previous research, a positive relationship between firm risk and CSR performance is expected (Jo and Harjoto, 2011). The sixth control variable, leverage, is expected to have a positive influence on the CSR performance, because a positive ratio indicates that a firm is in a good position to negotiate with the liabilities and has the resources to perform well on corporate social responsibility (Sharif and Rashid, 2014). Additionally, a positive relationship between firm size and CSR performance because larger firms have more resources to perform well on CSR and are also able to have a bigger impact. Additionally, larger firms are often under greater pressure from society to have a good CSR performance (Khan, 2010). Finally, a positive influence of meeting attendance is expected, because attendance at the board meetings allows for better communication between board members, and keeps the board members better informed, which will increase the board’s effectiveness.

Descriptive Statistics

Table II shows the descriptive statistics of the firm-level and board-level variables used in this study. Table II indicates that most variables have 756 observations, however, there are some missing observations for the CSR rating, board size, board meetings and CEO separation. On average, the board of European consist for 42.4 percent of outside directors. The average debt to assets ratio is 0.324. The average board size of European banks is 14 with a minimum of 4 board members and a maximum of 35 board member, and an average of 35.5 percent of the banks has no CEO on the board. Furthermore, on average the boards consist for 13.6 percent out of female board directors, which seems to be quite low.

diversity worldwide and new regulations aimed at increasing the gender diversity within firms.

Table III displays the pairwise correlation coefficients between the variables used in this study. Based on the correlation matrix there are no signs of multicollinearity. The Variance Inflation Factors (VIF) confirm this with no VIF-value higher than 2 (higher than 4 is considered a problem).

Table II

Descriptive statistics for firm and board characteristics for 63 listed European banks in the period 2004 to 2015

Mean Stdev Min Max Obs

Firm-level characteristics CSR rating (ln) 3.967 0.713 1.562 4.584 730 Firm size 0.012 1.004 -2.451 2.029 756 Leverage 0.324 0.181 0.000 0.945 756 Firm risk 1.272 0.670 -0.030 4.430 756 Board-level characteristics Board independence 0.424 0.277 0.000 1.000 756 Board size 14.421 5.074 4.000 35.000 748 Gender diversity 0.136 0.126 0.000 0.615 756 Ceo separation 0.355 0.479 0.000 1.000 723 Board expertise 0.395 0.295 0.000 1.000 756 Meeting attendance 0.789 0.336 0.000 1.000 756 Board meetings (ln) 2.372 0.454 1.099 3.761 694

Table III

Pairwise Correlation Matrix and VIF values

R

ESULTS

The results section consists of three parts. In the first part the results from the OLS regression are displayed in order to determine get a basic understanding of the relationship between CSR performance and board independence, board size, and gender diversity. Additionally, it allows for a comparison with most previous studies on this relationship. In the second part, the first regression is extended by using a random effects model, and the results are displayed and discussed. The random effects model will correct the estimates for unobserved heterogeneity. As a follow-up, the third part extends the random effects model by correcting the estimates for time-effects. The final results are presented and discussed.

Pooled OLS regression

Table IV shows the results of the pooled OLS regression, where the CSR performance is regressed on the board independence, board size, board gender diversity and the control variables. Column 1 presents the results with the CSR rating as the independent variable. additionally, the separate results for the CSR rating components are displayed. Column 2, 3 and 4 present the results with the independent variable being the economic rating, the social rating and the environmental rating, respectively. This structure is consistent throughout this paper. The OLS regression uses cluster robust standard errors to correct for heteroscedasticity and autocorrelation1. Column 1 of Table IV shows that board independence has a strong positive effect on the CSR performance at a 5% significance level (β = 0.551). Column 2, 3 and 4 show that this effect is caused through an improvement in the economic performance and the environmental performance at a 5% significance level (β = 0.648; β = 0.648). No significant effect of board independence on the social performance seems to be present. Furthermore, Table IV shows that gender diversity on the board has a strong positive influence on the CSR performance of the banks through all corporate social responsibility

1

Table IV

CSR performance and the influence of board independence, board size, and board gender diversity in European banks, 2004-2015

This table shows the results of the OLS regression without including the random-effects and time-effects, using a sample consisting of data from 63 listed European banks, originating from 16 countries, and covering the period 2004 to 2015. The data regarding the firm characteristics are retrieved from the Thomson Reuters Worldscope database and the CSR ratings and the board characteristics are retrieved from the Thomson Reuters ESG Asset4 database. Columns 1 shows results from the CSR rating regressed on board independence, board size, and board gender diversity, and where CEO separation, audit committee independence, board meetings, firm size, leverage, meeting attendance, and firm risk are controlled for. The CSR rating consists of three equally-weighted ratings: an economic rating, a social rating, and an environmental rating. Columns 2, 3 and 4 show the separate regression results of the respective ratings. Board independence equals the number of outside directors over the number of total board members, board size equals the total number of board members, gender diversity equals the number of female board members over the number of total board members, CEO separation equals one if the CEO is not a board member and equals zero if the CEO is a board member, board expertise equals the number of board members with an industry specific or strong financial background over the total number of board members, audit committee independence equals, the number of independent committee members over the total number of m embers on the audit committee, board meetings equals the number of yearly meetings of the board, firm size equals the distance of each bank’s log total assets from the sample mean and scaled by the log total assets’ standard deviation, leverage equals total debt over total assets, meeting attendance equals the average percentage of board members that is present at the board meetings, and firm risk equals systematic risk as indicated by the firm’s beta. For the CSR ratings and the board meetings the natural logarithm is used. *, **, *** represents significance at the 10%, 5%, and 1% levels, respectively. Robust standard errors are presented in parentheses under the coefficients.

Dependent variable

CSR rating (ln) Econ. rating (ln) Social rating (ln) Env. rating (ln)

(1) (2) (3) (4) Board independence 0.551** (0.268) 0.648** (0.315) 0.557 (0.341) 0.484** (0.238) Board size 0.001 (0.012) 0.006 (0.012) 0.002 (0.016) -0.008 (0.017) Gender diversity 0.901*** (0.321) 0.574** (0.443) 1.184*** (0.425) 1.111*** (0.294) CEO separation -0.075 (0.098) -0.014 (0.108) -0.109 (0.132) -0.120 (0.107) Board expertise 0.223* (0.127) 0.123*** (0.160) 0.294* (0.156) 0.239 (0.151)

Audit committee indep. -0.106

pillars, and is significant at the 1% level (β = 0.901). The results on the effects of board independence and gender diversity on CSR performance are consistent with OLS regression results of prior research (Bear, Rahman, and Post, 2010; Jizi, Salama, Dixon, and Stratling, 2014; Sharif and Rashid, 2014; Kiliç, Kuzey, and Uyar, 2015). Table IV shows no evidence of a relationship between board size and CSR performance, which contrasts the findings of Jizi, Salama, Dixon, and Stratling (2014). Furthermore, firm size and firm risk both appear to have a positive influence on the CSR performance, significant at the 1% level (β = 0.315; β = 0.289). Column 2 shows that board expertise only (positively) affects the economic performance, significant at the 1% level (β = 0.123).

Random-effects model

Since omitted variables and heterogeneity could possibly influence the results of the OLS regression, a fixed-effects model or a random-effects model can be used to account for these biases. In this study, the random effects model is the most suitable and accounts for differences across the banks that are uncorrelated with the independent variables2. Table V shows the results of the regression where random effects are included. Column 1 suggests a positive relationship between board independence and CSR performance, significant at the 5% level (β = 0.174). However, where Columns 2, 3 and 4 first indicated that the CSR performance was influenced through the economic performance and the environmental performance, they now indicate an effect through the economic performance and social performance (β = 0.378; β = 0.045). Furthermore, Table V again indicates a positive relationship between board gender diversity and CSR performance through all three CSR pillars, significant at the 1% level (β = 0.481). Similar to the first regression, there appears to be no significant relationship between board size and CSR performance. Kiliç, Kuzey, and Uyar (2015) investigated the effects on US banks and is the only similar study that also applied the fixed-effects or random effects extension. In their study they also find significant positive effects of board independence and gender diversity, and no significant effects of board size. Additionally, Table V indicates positive influences of both firm size and firm risk

2

Table V

CSR performance and the influence of board independence, board size, and board gender diversity in European banks, 2004-2015

This table shows the results of the OLS regression with the inclusion of random-effects, but time-effects are excluded. The sample consists of data from 63 listed European banks, originating from 16 countries, and covering the period 2004 to 2015. The data regarding the firm characteristics are retrieved from the Thomson Reuters Worldscope database and the CSR ratings and the board characteristics are retrieved from the Thomson Reuters ESG Asset4 database. Columns 1 shows results from the CSR rating regressed on board independence, board size, and board gender diversity, and where CEO separation, audit committee independence, board meetings, firm size, leverage, meeting attendance, and firm risk are controlled for. The CSR rating consists of three equally-weighted ratings: an economic rating, a social rating, and an environmental rating. Columns 2, 3 and 4 show the separate regression results of the respective ratings. Board independence equals the number of outside directors over the number of total board members, board size equals the total number of board members, gender diversity equals the number of female board members over the number of total board members, CEO separation equals one if the CEO is not a board member and equals zero if the CEO is a board member, board expertise equals the number of board members with an industry specific or strong financial background over the total number of board members, audit committee independence equals, the number of independent committee members over the total number of m embers on the audit committee, board meetings equals the number of yearly meetings of the board, firm size equals the distance of each bank’s log total assets from the sample mean and scaled by the log total assets’ standard deviation, leverage equals total debt over total assets, meeting attendance equals the average percentage of board members that is present at the board meetings, and firm risk equals systematic risk as indicated by the firm’s beta. For the CSR ratings and the board meetings the natural logarithm is used. *, **, *** represents significance at the 10%, 5%, and 1% levels, respectively. Robust standard errors are presented in parentheses under the coefficients.

Dependent variable

CSR rating (ln) Econ. rating (ln) Social rating (ln) Env. rating (ln)

(1) (2) (3) (4) Board independence 0.174** (0.081) 0.378** (0.110) 0.045*** (0.128) 0.125 (0.098) Board size -0.001 (0.008) 0.011 (0.012) -0.003 (0.007) -0.011 (0.009) Gender diversity 0.481*** (0.183) 0.297*** (0.299) 0.587*** (0.220) 0.672*** (0.206) CEO separation -0.050 (0.066) -0.112 (0.077) -0.097 (0.116) 0.007 (0.070) Board expertise 0.105 (0.067) 0.096 (0.100) 0.114 (0.080) 0.134 (0.082)

Audit committee indep. -0.099

on CSR performance, significant at the 1% level (β = 0.324; β = 0.272). The influence of board expertise on the economic performance of the banks has disappeared in this second regression.

Random-effects model with time-effects

In this third part, the random effects model is extended by correcting the estimates for time-effects3. Since the sample consists of data covering a 12-year period, it is well suited to test for characteristics that affect the CSR performance of European banks on specific moments in time. A time-dummy variable is added to the model and the results are presented in Table VI. Column 1 shows a positive relationship between board independence and CSR performance, significant at the 1% level (β = 0.232). However, the effect on the CSR rating appears to mainly stem from the strong effect of board independence on the economic performance, which is significant at the 1% level (β = 0.232). In contrast to the first two regressions, no significant effect of board independence on social and environmental performance is found. Although, for the environmental performance it is on the edge of the 10% significance level, which might influence the significance of the effect on the total CSR performance. Interestingly, with the addition of time-effects to the regression model, there is no longer evidence of an effect of board gender diversity on the CSR performance of the banks. The results suggest that that there are effects that influence the percentage of females on the board over time for all firms, which distorted the estimations of the effect gender diversity on CSR performance in the first two regressions, and it becomes apparent that time-effects must be accounted for. The results also question the findings of earlier studies that find significant positive results of gender diversity on the CSR performance, but do not include time-effects in the regression (Bear, Rahman, and Post, 2010; Post, Rahman, and Rubow; Kiliç, Kuzey, and Uyar, 2015). Similar to the first two regressions, no significant relationship between board size and CSR performance is found. Column 1 now does show a positive relationship between board expertise and CSR performance, which is significant at the 5% level (β = 0.125). Additionally, positive relationships between firm size and CSR performance, and firm risk and CSR performance are found, significant at the 1% level (β = 0.302; β = 0.247).

3

Table VI

CSR performance and the influence of board independence, board size, and board gender diversity in European banks, 2004-2015

This table shows the results of the OLS regression with the inclusion of random-effects and time-effects. The sample consists of data from 63 listed European banks, originating from 16 countries, and covering the period 2004 to 2015. The data regarding the firm characteristics are retrieved from the Thomson Reuters Worldscope database and the CSR ratings and the board characteristics are retrieved from the Thomson Reuters ESG Asset4 database. Columns 1 shows results from the CSR rating regressed on board independence, board size, and board gender diversity, and where CEO separation, audit committee independence, board meetings, firm size, leverage, meeting attendance, and firm risk are controlled for. The CSR rating consists of three equally-weighted ratings: an economic rating, a social rating, and an environmental rating. Columns 2, 3 and 4 show the separate regression results of the respective ratings. Board independence equals the number of outside directors over the number of total board members, board size equals the total number of board members, gender diversity equals the number of female board members over the number of total board members, CEO separation equals one if the CEO is not a board member and equals zero if the CEO is a board member, board expertise equals the number of board members with an industry specific or strong financial background over the total number of board members, audit committee independence equals, the number of independent committee members over the total number of members on the audit committee, board meetings equals the number of yearly meetings of the board, firm size equals the distance of each bank’s log total assets from the sample mean and scaled by the log total assets’ standard deviation, leverage equals total debt over total assets, meeting attendance equals the average percentage of board members that is present at the board meetings, and firm risk equals systematic risk as indicated by the firm’s beta. For the CSR ratings and the board meetings the natural logarithm is used. *, **, *** represents significance at the 10%, 5%, and 1% levels, respectively. Robust standard errors are presented in parentheses under the coefficients.

Dependent variable

CSR rating (ln) Econ. rating (ln) Social rating (ln) Env. rating (ln)

(1) (2) (3) (4) Board independence 0.232*** (0.088) 0.445*** (0.105) 0.147 (0.146) 0.149 (0.099) Board size 0.004 (0.008) 0.020* (0.012) 0.001* (0.007) -0.008 (0.009) Gender diversity 0.195 (0.219) -0.036 (0.363) 1.392 (0.264) 0.345 (0.255) CEO separation -0.048 (0.064) -0.120 (0.078) -0.108 (0.111) 0.024 (0.068) Board expertise 0.125** (0.061) 0.093** (0.109) 0.234*** (0.086) 0.100 (0.070)

Audit committee indep. -0.103

C

ONCLUSIONS

By using a dataset covering a 12-year time-period, this study extends research on corporate governance and corporate social responsibility into the European banking sector, and accounts for time specific influences in examining the effects of board independence, board size and board gender diversity on CSR performance in 16 European countries. Additionally, panel data methodology is applied to account for unobserved heterogeneity.

The first part of the research generates estimates using a pooled OLS regression to give a basic indication of the effects of board independence, board size, and board gender diversity on the CSR performance of European banks, and it allows for a direct comparison with the results from previous studies. The results also form a benchmark for the other estimated models in this study. The first regression shows positive relationships between board independence and CSR performance, and between board gender diversity and CSR performance. These results are consistent with findings of prior research (Bear, Rahman, and Post, 2010; Jizi, Salama, Dixon, and Stratling, 2014; Sharif and Rashid, 2014; Kiliç, Kuzey, and Uyar, 2015). The first regression does not show evidence of a relationship between board size and CSR performance. This contrasts the findings of Jizi, Salama, Dixon, and Stratling (2014), who do find a positive relationship between board size and CSR performance.

The second part of this study further investigates the mentioned relationships by testing for random effects, which accounts for omitted variables and heterogeneity bias. The regression results indicate a positive relationship between board independence and CSR performance, through the economic and social CSR pillars. Furthermore, the results show evidence of a positive relationship between board gender diversity and CSR performance, through all CSR pillars, suggesting that having a higher percentage of females on the board increases economic performance, social performance, and environmental performance. Again, no evidence is found of a relationship between board size and CSR performance. The results of this second regression are similar to the findings of Kiliç, Kuzey, and Uyar (2015), who applied a similar methodology.

of time-effects in the model changes the outcome of the regression significantly. The results still indicate a positive relationship between board independence and CSR performance, however, the effect on the CSR performance now appears to mainly stem from a strong effect through the economic pillar. Additionally, and probably most notable, the regression results no longer show evidence of a relationship between gender diversity and CSR performance. This means that there are effects that influence the CSR performance of all firms at a given point in time, which distorted the estimations of the effects of gender diversity on CSR performance in the first two regressions. An explanation for the time-effects could be the increase in regulation and pressure on firms to improve their CSR performance in recent years, as initiated by the governments of many European countries. Matten and Moon (2008), for example, mention increased efforts of the governments in the UK, Germany, France, and Italy to shape the CSR debate, and describe a similar mentality throughout the rest of Europe. The results, however, question the findings of earlier studies that show a significant positive relationship between gender diversity and CSR performance, and which do not account for time-effects in the regression (Bear, Rahman, and Post, 2010; Post, Rahman, and Rubow; Kiliç, Kuzey, and Uyar, 2015). More research on the effects of board gender diversity on CSR performance, and the possible influence of time-effects, is necessary to get a better understanding on this relationship. Finally, all three regressions show no evidence of a significant relationship between board size and CSR performance. This contradicts the results of previous research, which did indicate a positive influence of board size on CSR performance (Rao, Tilt, and Lester, 2012; Jizi, Salama, Dixon, and Stratling, 2014). However, Kiliç, Kuzey, and Uyar (2015) also reported an insignificant effect of board size. Larger boards are expected to be more effective at dealing with a high workload in a complex environment, such as a bank (Grove, Patelli, and Victoravich, 2011). However, it could be that the more efficient communication and coordination in a smaller board is more important for European banks. The effects of board size on CSR performance should be investigated further in future research.

this study. Future research should try to address these problems. From an empirical perspective, this study provides a deeper understanding of the relationship between board structure and CSR performance. This study used only European data, since it was uncertain if the results from other banking industries could be transferred the European market. However, the first two regressions of this study show similar results as previous research, and therefore generalization does not appear to be a problem. Future research can investigate the effects of board structure on CSR performance using worldwide data to get a large sample, and to shed more light on influences of board size and board gender diversity on CSR performance specifically. Additionally, this study shows that time-effects can have an important effect on the results of the study and should be incorporated in future research.

R

EFERENCES

Adams, R., Ferreira, D., 2009. Women in the boardroom and their impact on governance and performance. Journal of financial economics, 94(2), 291-309.

Adams, R., Hermalin, B., Weisbach, M., 2010. The role of boards of directors in corporate governance: A conceptual framework and survey. Journal of Economic Literature, 48(1), 58-107.

Anderson, R., Reeb, M., Upadhyay, A., Zhao, W., 2011. The economics of director heterogeneity. Financial Management, 40(1), 5-38.

Bear, S., Rahman, N., Post, C., 2010. The impact of board diversity and gender composition on corporate social responsibility and firm reputation. Journal of Business Ethics, 97(2), 207-221.

Belkhir, M., 2009. Board of directors' size and performance in the banking industry. International Journal of Managerial Finance, 5(2), 201-221.

Bell, A., Jones, K., 2015. Explaining fixed effects: Random effects modeling of time-series cross-sectional and panel data. Political Science Research and Methods, 3(01), 133-153.

Bell, A., Fairbrother, M., Jones, K., 2016. Fixed and random effects: making an informed choice. Bonn, I., 2004. Board structure and firm performance: Evidence from Australia. Journal of Management & Organization, 10(01), 14-24.

Carter, D., Simkins, B., Simpson, W., 2003. Corporate governance, board diversity, and firm value. Financial review, 38(1), 33-53.

Cormier, D., Gordon, I., 2001. An examination of social and environmental reporting strategies. Accounting, Auditing & Accountability Journal, 14(5), 587-617.

Cheng, E., Courtenay, S., 2006. Board composition, regulatory regime and voluntary disclosure. The international journal of accounting, 41(3), 262-289.

Cadbury, A., 1992. Report of the committee on the financial aspects of corporate governance (Vol. 1). Gee.

De Andres, P., Vallelado, E., 2008. Corporate governance in banking: The role of the board of directors. Journal of banking & finance, 32(12), 2570-2580.

Dobers, P., Halme, M., 2009. Corporate social responsibility and developing countries. Corporate social responsibility and environmental Management, 16(5), 237-249.

Donaldson, T., Preston, L., 1995. The stakeholder theory of the corporation: Concepts, evidence, and implications. Academy of management Review, 20(1), 65-91.

Dyllick, T., Hockerts, K., 2002. Beyond the business case for corporate sustainability. Business strategy and the environment, 11(2), 130-141.

Eisenhardt, K., 1989. Agency theory: An assessment and review. Academy of management review, 14(1), 57-74.

Giannarakis, G., Theotokas, I., 2011. The effect of financial crisis in corporate social responsibility performance. International Journal of Marketing Studies, 3(1), 2-10.

Graves, S., Waddock, S., 1994. Institutional owners and corporate social performance. Academy of Management journal, 37(4), 1034-1046.

Grove, H., Patelli, L., Victoravich, L., Xu, P., 2011. Corporate governance and performance in the wake of the financial crisis: Evidence from US commercial banks. Corporate Governance: An International Review, 19(5), 418-436.

Hillman, A., Cannella, A., Harris, I., 2002. Women and racial minorities in the boardroom: How do directors differ?. Journal of Management, 28(6), 747-763.

Hillman, A., Withers, M., Collins, B., 2009. Resource dependence theory: A review. Journal of management.

Ibrahim, N., Howard, D., Angelidis, J., 2003. Board members in the service industry: An empirical examination of the relationship between corporate social responsibility orientation and directorial type. Journal of Business Ethics, 47(4), 393-401.

Jackling, B., Johl, S., 2009. Board structure and firm performance: Evidence from India's top companies. Corporate Governance: An International Review, 17(4), 492-509.

Jensen, M., Meckling, W., 1976. Theory of the firm: Managerial behavior, agency costs and ownership structure. Journal of financial economics, 3(4), 305-360.

Jizi, M., Salama, A., Dixon, R., Stratling, R., 2014. Corporate governance and corporate social responsibility disclosure: Evidence from the US banking sector. Journal of Business Ethics, 125(4), 601-615.

Jo, H., Harjoto, M., 2011. Corporate governance and firm value: The impact of corporate social responsibility. Journal of business ethics, 103(3), 351-383.

Johnson, R., Greening, D., 1999. The effects of corporate governance and institutional ownership types on corporate social performance. Academy of management journal, 42(5), 564-576.

Jungmann, C., 2006. 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.

Khan, H., 2010. The effect of corporate governance elements on corporate social responsibility (CSR) reporting: Empirical evidence from private commercial banks of Bangladesh. International Journal of Law and Management, 52(2), 82-109.

Kolk, A., Pinkse, J., 2010. The integration of corporate governance in corporate social responsibility disclosures. Corporate Social Responsibility and Environmental Management, 17(1), 15-26.

Kiliç, M., Kuzey, C., Uyar, A., 2015. The impact of ownership and board structure on Corporate Social Responsibility (CSR) reporting in the Turkish banking industry. Corporate Governance, 15(3), 357-374.

La Porta, R., Lopez-de-Silanes, F., Shleifer, A., Vishny, R., 2000. Investor protection and corporate governance. Journal of financial economics, 58(1), 3-27.

Li, S., Fetscherin, M., Alon, I., Lattemann, C., Yeh, K., 2010. Corporate social responsibility in emerging markets. Management International Review, 50(5), 635-654.

Matten, D., 2006. Why do companies engage in corporate social responsibility? Background, reasons and basic concepts. The ICCA handbook on corporate social responsibility, 3-46.

Matten, D., Moon, J., 2008. ―Implicit‖ and ―explicit‖ CSR: A conceptual framework for a comparative understanding of corporate social responsibility. Academy of management Review, 33(2), 404-424.

McWilliams, A., Siegel, D., 2000. Corporate social responsibility and financial performance: correlation or misspecification?. Strategic Management Journal, 21(5), 603-609.

O’Dwyer, B., 2002. Managerial perceptions of corporate social disclosure: An Irish story. Accounting, Auditing & Accountability Journal, 15(3), 406-436.

Pathan, S., 2009. Strong boards, CEO power and bank risk-taking. Journal of Banking & Finance, 33(7), 1340-1350.

Pfeffer, J., Salancik, G., 2003. The external control of organizations: A resource dependence perspective. Stanford University Press.

Pondeville, S., Swaen, V., De Rongé, Y., 2013. Environmental management control systems: The role of contextual and strategic factors. Management accounting research, 24(4), 317-332.

Post, C., Rahman, N., Rubow, E., 2011. Green governance: Boards of directors’ composition and environmental corporate social responsibility. Business & Society, 50(1), 189-223.

Rao, K., Tilt, C., Lester, L., 2012. Corporate governance and environmental reporting: an Australian study. Corporate Governance: The international journal of business in society, 12(2), 143-163.

Rosenstein, S., Wyatt, J., 1990. Outside directors, board independence, and shareholder wealth. Journal of financial economics, 26(2), 175-191.

Sanders, W., Carpenter, M., 1998. Internationalization and firm governance: The roles of CEO compensation, top team composition, and board structure. Academy of Management journal, 41(2), 158-178.

Shleifer, A., Vishny, R., 1997. A survey of corporate governance. The journal of finance, 52(2), 737-783.

Said, R., Hj Zainuddin, Y., Haron, H., 2009. The relationship between corporate social responsibility disclosure and corporate governance characteristics in Malaysian public listed companies. Social Responsibility Journal, 5(2), 212-226.

Sharif, M., Rashid, K., 2014. Corporate governance and corporate social responsibility (CSR)

reporting: an empirical evidence from commercial banks (CB) of Pakistan. Quality & Quantity, 48(5), 2501-2521.

Staikouras, P., Staikouras, C., Agoraki, M., 2007. The effect of board size and composition on European bank performance. European Journal of Law and Economics, 23(1), 1-27.

Stearns, L., Mizruchi, M., 1993. Board composition and corporate financing: The impact of financial institution representation on borrowing. Academy of management Journal, 36(3), 603-618.

Vaisey, S., Miles, A., 2017. What you can—and can’t—do with three-wave panel data. Sociological Methods & Research, 46(1), 44-67.

Van Marrewijk, M., 2003. Concepts and definitions of CSR and corporate sustainability: Between agency and communion. Journal of business ethics, 44(2), 95-105.

Yermack, D., 1996. Higher market valuation of companies with a small board of directors. Journal of financial economics, 40(2), 185-211.

Zahra, S., Oviatt, B., Minyard, K., 1993. Effects of corporate ownership and board structure on corporate social responsibility and financial performance. In Academy of Management Proceedings (Vol. 1993, No. 1, pp. 336-340). Academy of Management.

A

PPENDIX

Table A.I

Descriptive statistics for firm and board characteristics for 63 listed European banks in the period 2004 to 2015, decomposed by year

Mean Stdev Min Max Obs

Table A.I Continued

Mean Stdev Min Max Obs

Table A.I Continued

Mean Stdev Min Max Obs

Table A.I Continued

Mean Stdev Min Max Obs

Table A.I Continued

Mean Stdev Min Max Obs

CEO Separation 2004 0.415 0.497 0.000 1.000 53 2005 0.414 0.497 0.000 1.000 58 2006 0.362 0.485 0.000 1.000 58 2007 0.386 0.491 0.000 1.000 57 2008 0.403 0.495 0.000 1.000 62 2009 0.381 0.490 0.000 1.000 63 2010 0.323 0.471 0.000 1.000 62 2011 0.323 0.471 0.000 1.000 62 2012 0.306 0.465 0.000 1.000 62 2013 0.290 0.458 0.000 1.000 62 2014 0.323 0.471 0.000 1.000 62 2015 0.355 0.482 0.000 1.000 62 Total 0.355 0.479 0.000 1.000 723

Audit Committee Indep.

Table A.I Continued

Mean Stdev Min Max Obs

Table A.II

Overview of the variables and methodology used in previous studies

Kolk and Pinkse (2010) Li, Fetscherin, Alon, Lattemann and Yeh (2010) Kiliç, Kuzey and Uyar (2015) Jizi, Salama, Dixon and Stratling (2014) Sharif and Rashid (2014) Khan (2010) Bear, Rahman and Post (2010) Post, Rahman and Rubow (2011)

Percentage of outside directors x x x x x x

Board size x x Percentage of females x x x x CEO duality/separation x x x Audit committee x Board meetings x Board expertise x x

Board member age x

Board member nationality x x

Firm size x x x x x Leverage x x x Profitability x x x x Firm risk x Ownership structure x Industry x Country x x OLS regression x x x x x x Random effects x x

Firm/country fixed effects x x

Time fixed effects

Tobit regression x