Board diversity and Sustainability

(1)

Board diversity and Sustainability

Does diversity in boards has an influence on the reported sustainability policies of firms?

Empirical evidence from Dutch listed firms.

Master Thesis

By

Marijn De Groot

s1874454

June 2014

University of Groningen

Faculty of Economics and Business

Msc. International Business and Management

Mr. Dr. Kees van Veen

Mr. Q. Dong, M.

Prinsenstraat 9a

9711 CL Groningen

(2)

ABSTRACT

Manuscript type: Empirical

Research question: Does diversity in boards has an influence on the reported sustainability policies

of Dutch listed firms?

Research findings: Hypotheses are developed in order to test if diversity in boards (mean age,

gender diversity and nationality diversity) has an influence on the sustainability policies and on the different dimensions within sustainability of 57 Dutch listed firms in 2013. In particular, sustainability consists of three dimensions: economic, environmental and social dimension. This study takes into account both the supervisory and executive board of Dutch listed firms.

Contrary to expectations, there is (almost) no influence found of diversity in boards on the reported sustainability policies of Dutch listed firms. In particular, there were no results found of the influence of mean age, gender diversity and nationality diversity on the transparency of sustainability. Also, there were no results found for mean age, gender diversity and nationality diversity on the extent to which sustainability policies of firms are explicit available in external reporting and the extent to which firms report about economic (taxation), environmental (biodiversity) and social (human rights) issues concerning sustainability. None of the developed hypotheses are supported.

When diversity in boards in its entirety is tested on sustainability, there is a slight positive influence found for mean age on the extent to which firms report about social issues concerning sustainability (human rights). However, this is only the case because mean age, gender diversity, nationality diversity and the control variables (firm size, firm industry and board size) are all taken into account.

Theoretical implications: This study is the first that takes into account several characteristics of

board members (age, gender and nationality) in order to test the influence of diversity in boards on the reported sustainability policies of firms. Also, this study extends previous research of the influence of gender diversity on sustainability policies by focusing in particular on Dutch listed firms.

Policy implications: This study shows that diversity in boards has (almost) no influence on the

sustainability policies of Dutch listed firms. Overall, the results implicate for practice that Dutch listed firms are not able to use the mean age of board members, gender diversity or nationality diversity in the board to influence the reporting on sustainability policies.

Keywords: Diversity in boards, mean age, gender diversity, nationality diversity, sustainability

(3)

3

LIST OF ABBREVIATIONS

AEX – Amsterdam Exchange Index AMX – Amsterdam Midkap Index AScX – Amsterdam Small Capp Index BEE – Black economic empowerment B2B – Business-to-business

B2C – Business-to-consumer

CSP – Corporate Social Performance CSR – Corporate Social Responsibility CSR – Corporate Social Reporting SR – Sustainability Reporting TMT – Top Management Team TMTs – Top Management Teams UE – Upper Echelon

(4)

1. INTRODUCTION

Throughout recent years, firms look different at social and environmental issues compared to the past and pay more attention to the integration and implementation of these issues (Epstein, 2008). In particular, the Corporate Social Responsibility (CSR) and the Corporate Social Performance (CSP) of a firm are important (Wong et al., 2011). Wong et al. (2011) argue that Corporate Social Performance refers to “a business organization’s configuration of principles of social responsibility, processes of social responsiveness, and policies, programs, and observable outcomes as they relate to the firm’s societal relationships”. Some of the world largest firms have significant commitments to social and environmental issues (Epstein, 2008). In particular, sustainability is becoming more valuable in the decision-making processes of an organization (Peloza et al., 2012). Epstein (2008) defined sustainability as “economic development that meets the needs of the present generation without compromising the ability of future generations to meet their own needs”. Sustainability can be seen as a ‘CSR synonym’ (Strand, 2013).

The corporate reputation of a firm is one of the most valuable assets of a firm. Recently, sustainability plays an important role because it has become a component of this reputation. Stakeholders play also an important role for the reputation of a firm and can give the firm an extra benefit (Peloza et al., 2012). Peloza at al. (2012) did research with respect to three key stakeholders (investment professionals, purchasing managers and graduating students) perceptions and plot those against third-party rankings of actual firm performance on sustainability metrics. Results show that sustainability is important to these stakeholders; they use sustainability in their decision-making process. Also, they argue that many other stakeholders (such as employees and customers) report that sustainability plays an important role in their decision-making process (Peloza et al., 2012). Peloza et al. (2012) conclude in their paper that it is important to integrate sustainability in the culture of an organization and into its business practices; this is done by integrating sustainability into existing business operations. Research by Nidumolu et al. (2009) indicates that sustainability is now a key driver of innovation and only when a firm makes sustainability a goal, it is able to accomplish a competitive advantage.

(6)

operations is very important (Epstein and Roy, 2001). Epstein and Roy (2001) argue that, “The alignment of strategy, structure, and management systems are essential for firms to both coordinate activities and motivate employees toward implementing a sustainability strategy”. Top Management Teams are able to make a change very quickly when focusing on a particular problem. Sustainability should work together with the general strategies and frameworks of a firm (Nidumolu et al., 2009). In particular, the goals of the boards are important for building a strong sustainable strategy (Epstein, 2008). Epstein (2008) argues that leadership, engagement, alignment, diversity, evaluation and responsibility are important core principles that can help boards in formulating strategies in general and improve sustainability in particular. On the basis of research, Epstein (2008) concludes that sustainability strategy occurs when using a top-down approach. Boards have the highest authority in a firm, the ‘foundation of power’. Naturally, the Top Management Teams exercise this power (Sundaramurthy and Lewis, 2003). Top Management should be committed to the sustainability strategy in order to ensure that the implementation of the strategy is most effective (Epstein, 2008). The Upper Echelon (UE) theory argues that there is a relationship between board characteristics and the strategic outcomes of a firm. In particular, policies are important strategic outcomes (Carpenter et al., 2004). On the basis of the Upper Echelon theory and research by Epstein (2008) there could be a possible relationship between diversity in boards and sustainability policies of firms. This study will focus in particular on boards instead of Top Management Teams.

(7)

7

2. EARLIER FINDINGS AND PROBLEM FORMULATION

2.1 Research question

Research by Jhunjhunwala and Mishra (2012) shows that there are several views (political, social and business views) calling formore demographically spread boards, in particular there is more demand for diverse boards. Barkema and Shvyrkov (2007) argue that demographic diversity refers to different preferences, beliefs, experiences, skills and information networks. Board diversity ensures more variation of perspectives that increases the creativity and innovative solutions to problems. Diverse ideas arise when there is a heterogeneous composition and a diversity of backgrounds. A heterogeneous composition means a mix between insiders and outsiders. In fact, both insiders and outsiders may contribute with contrasting perspectives. This could lead to productive conflicts and these conflicts can lead to more creative debates of board members. Heterogeneous backgrounds indicatefor example different educational, functional or occupational backgrounds (Sundaramurthy and Lewis, 2003). Recent study by Lawler and Mohrman (2013) indicate that the role of boards for sustainability is very important because they are responsible for the long-term performance and survival of their firms. In particular, Lawler and Mohrman (2013) show that board leadership is critical for making sustainability a core strategic and operational focus. Sustainability is a difficult issue and a problem is that sustainability programs come and go depending on the interests and believes of corporate leaders at a particular point in time (Lawler and Mohrman, 2013). Lawler and Mohrman (2013) argue, “Boards must provide direction, holding management responsible for sustainable performance in order for sustainable performance to be integrated into the firm”. However, boards have rarely primary responsibility for sustainability strategy (Lawler and Mohrman, 2013).

(8)

Therefore, this study will include all three dimensions of sustainability with respect to the research of the possible relationship between diversity in boards and sustainability.

A lot research is done with respect to women on boards (gender diversity on boards) and sustainability. In general, research shows that women on boards have a positive influence on sustainability (Galbreath, 2011). However, there is no research done with respect to gender diversity and sustainability of Dutch listed firms. Dutch listed firms usually consist of two-tier boards, distinct by the supervisory and executive board (Van Veen and Elbertsen, 2008). This research aims to extend the research on gender diversity and sustainability with respect to the two-tier boards of Dutch listed firms. Also, there is no research done with respect to other forms of diversity in boards and sustainability policies of firms, for example age diversity and nationality diversity. In fact, this research aims to fulfill these gaps and take into account more forms of board diversity in order to achieve an objective measurement of the possible relationship between board diversity and sustainability policies of Dutch listed firms.

Hafsi and Turgut (2013) argue that there is a difference between diversity of boards and diversity in boards. In fact, diversity of boards is about dissimilarities in board attributes, for example board independence, leadership structure and director ownership. Diversity in boards is about dissimilarities in directors’ attributes. In particular, this means the diversity within a board, for example age and gender of the board members (Hafsi and Turgut, 2013). This research will focus on diversity in boards, because this study aims to discover if specific board characteristics (age, gender and nationality diversity in the board) has an influence on sustainability policies of firms. Jhunjhunwala and Mishra (2012) define board diversity as “the heterogeneous composition of the board in terms of gender, age, race, education, experience, nationality, lifestyle, culture, religion and many other facets that make each of us unique as individuals”. This definition of board diversity is consistent with this research.

(9)

9 strategies (Epstein, 2008). Recent literature shows a research gap with respect to the diversity in boards and sustainability policies of firms. As noted above, there is no research done with respect to the influence of multiple characteristics of board members on sustainability policies of firms. This leads to the following main research question:

Does diversity in boards has an influence on the reported sustainability policies of Dutch listed firms?

(10)

3. THEORETICAL CONSIDERATIONS

3.1 Historical background of board characteristics and board functions

Recent literature investigates sustainability in relation to firms and how to incorporate sustainability in their daily activities, directly and indirectly (Parisi, 2012). Corporate Social Responsibility (CSR) or sustainability consists of three important dimensions: an economic, environmental and social one (Torugsa et al., 2013). Bardy and Massaro (2012) argue that sustainability is about integrating the three dimensions and these dimensions are interrelated. In particular, the interconnectedness between these three dimensions is very important (Gao and Bansal, 2013). Wolff (2011) studied three French firms in the building industry with respect to integrating sustainable development. In particular, Wolff (2011) argues, “It is found that the firms showed that they were capable of integrating sustainable development principles into their way of management; they also developed their rules of governance –in particular with regard to the constitution and the operating rules of their board of directors”. It turns out that Top Management plays an important role for an effective sustainability strategy (Nidumolu et al., 2009).

Carter et al. (2010) argue that there are at least four very important functions of boards. These four important functions are: “monitoring and controlling managers, providing information and counsel to managers, monitoring compliance with applicable laws and regulations, and linking the corporation to the external environment”. There are several theoretical perspectives with respect to these board functions, but also with respect to board diversity. In particular, the ‘resource dependence theory’ provides arguments with respect to board diversity. This theory argues that diversity is useful for gaining access to different external environments because different board members can have different external contacts. Also, board diversity can provide/improve valuable information that can help boards to make better decisions. With respect to board diversity, the ‘human capital theory’ adds some important information to the resource dependence theory. This theory argues that unique human capital has an impact on the performance of an organization. However, this effect could be either positive or negative (Carter et al., 2010).

(11)

11 3.2 Hypotheses

Sustainability is a difficult concept, because every individual operationalizes sustainability in a different manner and there is no general definition of sustainability. In particular, for each individual, sustainability can mean something different (Daizy et al., 2013). In this study, chosen is to operationalize sustainability to the extent to which sustainability policies are explicit available in external reporting. Sustainability reporting is used as tool to measure to what extent firms consider sustainability as an important issue in order to discover the visible sustainability policies of firms. Recently, sustainability reporting is becoming more important. To firms, the aim of sustainability reporting is to reveal their own business practices and to compare them with those of the environment (Christoﬁ et al., 2012). Also, firms report externally about sustainability to communicate what they do with respect to CSR. This is done with the aim of showing that the firm cares about people and environment (Breitbarth et al., 2010). Sustainability reporting developed fast over de past years (Christoﬁ et al., 2012). Daizy et al. (2013) argue, “Sustainability reporting is gaining momentum globally as an important communication tool for firms to disclose their sustainability plans and performance and enhance stakeholder confidence”. Sustainability reporting increased because of the need to increase corporate transparency, but also because of the accountability concerning social and environmental issues (Breitbarth et al., 2010). The term ‘corporate sustainability reporting (CSR) or sustainability reporting (SR)’ is equal to the term ‘corporate social responsibility reporting’ and covers the three pillars (economic, environmental and social dimensions) of sustainability (Daizy et al., 2013). External sustainability reporting is related to the transparency of a firm. In particular, it is assumed that when firms report externally about sustainability and thus are transparent with respect to this area, they consider sustainability as an important issue. This study will focus on the extent to which sustainability policies are explicit available in external reporting.

(12)

Sustainability is seen as an important factor for the corporate reputation of a firm (Peloza et al., 2012). This implies when firms focus on sustainability, they will also be transparent with respect to their sustainability policies because this will contribute to a positive corporate reputation.

H1: Diversity in boards has a positive influence on the extent to which firms are transparent with respect to sustainability.

In this study, diversity in boards will be viewed as heterogeneous in terms of age, gender and nationality characteristics of board members. The first board member characteristic that will be discussed is age diversity. Age diversity refers to different age ranges of board members. Most boards need age diversity because different age groups foster different perspectives (Kang et al., 2007). In particular, research by Jhunjhunwala and Mishra (2012) indicates that younger people are in general more flexible and understand new technologies and concepts more easily. Also, younger people take more risk compared to older people. In turn, the firm could also benefit from senior members because these senior board members have a wider experience and stronger networks that could be used (Jhunjhunwala and Mishra, 2012).

Ntim and Soobaroyen (2013) did research with respect to black economic empowerment (BEE) disclosures by South African listed firms and the influence of ownership and board characteristics. Black economic empowerment is a form of Corporate Social Responsibility. Results supported the positive relationship between age diversity and the extent of black economic empowerment disclosures (Ntim and Soobaroyen, 2013). This research has shown that age diversity has a positive influence on a particular form of Corporate Social Responsibility, namely the black economic empowerment disclosures. On the basis of this research, it is expected that age diversity in the board has a positive influence on Corporate Social Responsibility (also called sustainability).

Due to data limitations, mean age is used instead of age diversity. Mean age is a characteristic of age diversity because it gives an indication of the distribution of age in the board. It could be possible that the mix of young and old people in the board will create an optimal and different combination of perspectives in order to positively influence sustainability policies of firms. For example, an optimal combination of the boldness of young people and the wide experience of old people. It is expected that mean age has a positive influence on sustainability policies of firms.

H2: Mean age in the board has a positive influence on the extent to which sustainability policies of firms are explicit available in external reporting.

(13)

13

H2b: Mean age in the board has a positive influence on the environmental dimension within sustainability policies of firms.

H2c: Mean age in the board has a positive influence on the social dimension within sustainability policies.

There has been some research done on gender diversity in boards and the linkage with sustainability. A reason to pay attention to gender diversity with respect to the composition of boards is that “gender diversity has been advocated as means of improving organizational value and performance by inculcating boards with new insights, new information, and new perspectives”. In fact, on board level these ‘new insights, new information and new perspectives’ are important issues with respect to meeting sustainability challenges (Galbreath, 2011). Results of the study by Galbreath (2011) shows that women on boards have in general a positive influence on sustainability. However, there are doubts whether women address all the dimensions of sustainability, for example due to sex-based biases or stereotyping. In particular, this study finds that women on boards have a positive significant effect on economic growth and social responsiveness. An explanation for this is that women on boards are more capable to improve stakeholder relationships and ethical conduct that result in a positive effect on economic growth and social responsiveness (Galbreath, 2011). Stakeholders play an important role and organizations want to satisfy the demands of their stakeholders (Garvare and Johansson, 2010). In general, stakeholders argue that they find sustainability important and use this in their decision-making process (Peloza et al., 2012). However, Galbreath (2011) argues that women on boards have a non-significant effect on environmental quality. Galbreath (2011) argues that environmental issues are associated with “hard science, technology and technical or engineering processes” and men (on the board) have in general more degrees and backgrounds in technical disciplines and are thus better able to solve environmental issues instead of female board members. On the basis of this previous research, the following hypotheses are developed. It is important to note that it is expected that gender diversity will have a negative impact on the environmental dimension within sustainability.

H3: Gender diversity in the board has a positive influence on the extent to which sustainability policies of firms are explicit available in external reporting.

H3a: Gender diversity in the board has a positive influence on the economic dimension within sustainability policies of firms.

(14)

H3c: Gender diversity in the board has a positive influence on the social dimension within sustainability policies of firms.

Finally, nationality diversity will be discussed. Nationality diversity refers to board members who come from different countries. Nationality diversity in boards could be very useful because firms need board members from different countries in order to understand how countries operate and what their business environment looks like. Also, board members with different backgrounds are needed to understand and to communicate with all the people they work with on a global scale (Jhunjhunwala and Mishra, 2012). Jhunjhunwala and Mishra (2012) argue that these board members have “different lifestyles, culture and upbringing backgrounds that will bring new perspectives and solutions to the table”. Research shows that Dutch firms have in general a high nationality diversity in their boards (Van Veen and Marsman, 2008). In this study, it is expected that the included Dutch listed firms also have a high nationality diversity in their boards.

As noted above, Ntim and Soobaroyen (2013) did research with respect to a particular form of CSR (black economic empowerment disclosures by South African listed firms) and the influence of ownership and board characteristics. The research indicated that age diversity has a positive influence on black economic empowerment disclosures, but they also proved that nationality diversity has a positive effect on this particular form of CSR (Ntim and Soobaroyen, 2013). On the basis of this evidence it is expected that nationality diversity has a positive influence on CSR or sustainability. Also, nationality diversity leads to different perspectives that foster strategic innovation (Barkema and Shvyrkov, 2007). Sustainability is now a key driver of innovation (Nidumolu et al., 2009) and therefore it is also expected that nationality diversity will influence sustainability in a positive way.

H4: Nationality diversity in the board has a positive influence on the extent to which sustainability policies of firms are explicit available in external reporting.

H4a: Nationality diversity in the board has a positive influence on the economic dimension within sustainability policies of firms.

H4b: Nationality diversity in the board has a positive influence on the economic dimension within sustainability policies of firms.

(15)

15

4. DATA AND METHODOLOGY

4.1 Data collection

In order to test the assumptions described in the hypotheses, there are two types of datasets needed. The first dataset contains data with respect to the composition of boards of the included Dutch listed firms. The board data that is needed should provide information about board diversity in the year 2013. In particular, information about mean age, gender diversity and nationality diversity of Dutch listed firms. This data is obtained by using the ‘Dutch Interlock Database’ of the University of Groningen. The database contains secondary data of inter alia various information about boards of 109 Dutch listed firms for the period 2006-2014. Every year, several master students from the University of Groningen collected this data on the bases of the published annual reports1 by the relevant Dutch listed firms. The valuable board information of the annual reports of the firms is entered manually into the database.

The second required dataset contains data on the firm level. Information is needed about the transparency of Dutch listed firms with respect to sustainability. This information is obtained by using data of the Dutch report: ‘sustainability of listed firms in gear’2. This report contains information about sustainable development at 64 Dutch listed firms in 2013. The Dutch association of investors for sustainable development (VBDO3) collected this data by attending shareholders’ meetings and asking several questions about the sustainability policy conducted by the relevant firm. Also, VBDO used published annual reports4 of the firms. In this way, VBDO collected data with respect to transparency of reporting on sustainability of Dutch listed firms. VBDO did research on several sustainability reporting themes, for example biodiversity and human rights. Overall, they found an increase in sustainability reporting by Dutch listed firms. In particular, Dutch firms report more extensively on corporate social responsibility, both in annual reports as well as in presentations to shareholders (Remmers et al., 2013).

In order to test the hypotheses and answer the main research question, both board data and firm-level sustainability data should be combined. In particular, descriptive and statistical analyses are needed in order to answer the research question.

1_{The annual reports are publicly available on the website of the relevant firm.}

(16)

4.2 Sample size and selection

This study focus in particular on the analysis of Dutch listed firms. The sample is composed of 57 Dutch listed firms5. The 57 Dutch listed firms are chosen based on the available firms in both the board and firm-level sustainability datasets of 2013. The Dutch Interlock Database (board data) consists of 109 Dutch listed firms and the report of sustainability of firms in gear (firm-level sustainability data) consists of 64 Dutch listed firms. Combining both datasets, 57 Dutch listed firms can be found in both the board and firm-level sustainability datasets.

Most of these firms are listed AEX, AMX or AScX. Amsterdam Exchange Index (AEX) indicates the most important Dutch listed firms; in fact these are the largest firms. Amsterdam Midcap Exchange (AMX) indicates medium sized firms and Amsterdam Small Cap Exchange (AScX) indicates the smallest listed firms.

The selected Dutch listed firms are distinct from each other in several ways; for example, they differ in firm size and the sectors in which they operate. These different firms are selected in order to create an objective overall picture with respect to transparency of sustainability of Dutch listed firms.

4.3 Variables and measurement

There are several variables needed in order to test if the assumptions made in this research can be empirically supported. The variables can be derived from the formulated hypotheses. In particular, there are independent variables and dependent variables. Also, control variables are used. This section will focus on these three variables. In particular, focus is on the definitions of the variables and on how they are measured. The independent variables can be measured using board data of the ‘Dutch Interlock Database’ of 2013. The dependent variables can be measured using the firm-level sustainability data of the Dutch report ‘sustainability of listed firms in gear’ of 2013 (Remmers et al., 2013). The control variables can also be measured by using data of the report ‘sustainability of listed firms in gear’. Besides specific information about sustainability, this report also contains some general information with respect to the relevant firms. Part of this information is suitable for the use as control variable. Also, the control variables can be measured by using the ‘Dutch interlock database’.

4.3.1 Independent variables

The main independent variable is board diversity. This variable is measured on the basis of several diversity factors: mean age, gender diversity and nationality diversity. The majority of Dutch firms consist of a two-tier board, distinct by the supervisory board and executive board (Van Veen and

(17)

17 Elbertsen, 2008). Measuring the number of board members will be on the basis of the total absolute number of board members of both the supervisory and executive board. Mean age will be measured on the basis of the mean age of all the board members. Gender diversity will be measured on the basis of the absolute and relative number of women on the board. Nationality diversity will be measured on the basis of the absolute and relative number of foreign board members on the board.

4.3.2 Dependent variables

The first dependent variable that can be indicated is transparency of sustainability. Transparency of sustainability can be defined as the extent to which the firm provides information regarding sustainability (Navarro Galera et al., 2014). The overall information firms provide regarding sustainability is used to measure the transparency of sustainability in an appropriate way. Data of the Dutch report ‘sustainability of listed firms in gear 2013’ contains a list of the most transparent firms, which is a ‘transparency benchmark’6. In particular, transparency scores are assigned to Dutch listed firms with regard to the extent to which they are transparent with respect to sustainability in 20127. A high score on this indicator means that the relevant firm provides much information about sustainability and therefore gets a high transparency score (Remmers et al., 2013).

The second dependent variable is sustainability policies. As noted earlier, sustainability is very difficult to measure because every individual operationalizes sustainability in a different manner (Daizy et al., 2013). In this study, it is chosen to operataionalize sustainability (policies) as: ‘the extent to which sustainability policies of firms are explicit available in external reporting’. Sustainability consists of three dimensions: economic, environmental and social dimension. All three dimensions should be taking into account in order to measure sustainability in an objective and appropriate way (Singh et al., 2012). The extent to which sustainability policies of firms are explicit available in external reporting is measured by the/a (total) sustainability score, which takes into account all three dimensions within sustainability. It is important to note that there is a distinction between the measurement of ‘transparency of sustainability’ (first dependent variable) and ‘sustainability policies’ (second dependent variable). Transparency of sustainability will be measured by the overall information firms provide regarding sustainability. Sustainability polices will be measured by the explicit reporting of the three dimensions within sustainability. In particular, this is the explicit reporting about economic, environmental and social issues concerning sustainability.

The main hypotheses about board diversity (and other board variables) on sustainability policies are also divided in sub hypotheses to examine the effect of board diversity on the different dimensions

6

(18)

within sustainability. In particular, in these sub hypotheses the different dimensions within sustainability are the dependent variables.

The first dimension within sustainability is the economic dimension, this is seen as the extent to which firms report about economic issues concerning sustainability. Throughout recent years, there is more discussion about tax evasion with relation to CSR/sustainability. In particular, tax evasion theme is a major political issue and most of the Dutch government parties are against tax evasion. The economic dimension within sustainability can be measured using tax scores. In particular, the extent to which the involved Dutch listed firms find taxation a relevant theme for sustainability. The measurement is based on four factors: “1) The firm sees taxes as CSR themes, 2) The firm mentions the tax payments by country, 3) The firm has no shelf firms in tax havens and 4) The firm provides access to tax strategy”. Scores are assigned to these individual factors. Herewith, a maximum total score of 10 points can be achieved (Remmers et al., 2013). VBDO made the specific choice to use these tax scores in order to measure the economic dimension within sustainability. This is a limited operationalization and there is a better method needed. In this study, due to data limitations, the tax scores by VBDO are used to measure the economic dimension within sustainability.

The second dimension within sustainability is the environmental dimension, this is seen as the extent to which firms report about environmental issues concerning sustainability. The environmental dimension can be measured using biodiversity scores. In particular, with this theme VBDO asks the included firms questions with respect to “the extent to which conservation and sustainable use of biodiversity and ecosystems is included in the strategy and operations, so that it leads to a lasting and significant reduction of the negative impact on biodiversity and ecosystems, or contributing to the restoration of biodiversity and ecosystems (no net loss)” (Remmers et al., 2013). VBDO assigns several scores with respect to the theme biodiversity. These scores are based on five factors: “1) Biodiversity identified as activity in the annual report or sustainability report, 2) Reporting on impact on biodiversity of the product or process, 3) Cooperation with civil society organizations, 4) No net loss projects and 5) Ecol. profit net loss research”. With achieving each of these individual factors, 2 points can be earned. Herewith, a total maximum of 10 point can be earned with respect to the biodiversity theme (Remmers et al., 2013).

(19)

19 diligence (Remmers et al., 2013). Remmers et al. (2013) defined due diligence as “the process in which firms identify the actual and potential negative impact of their actions, prevent and reduce and explain how they deal with the risks identified”. Both factors are assessed with an individual maximum score of 5 points. A high total score (maximum of 10 point) indicates a good human rights policy and due diligence. In particular, this means that the firm pays an adequate amount of attention with respect to human rights (Remmers et al., 2013).

Summarized, sustainability policies can be measured on the basis of sustainability scores. In particular, taxation, biodiversity and human rights scores measure the three dimensions within sustainability8. The use of these variables will be appropriate because previous researchers, which did research with respect to the three dimensions within sustainability, used similar variables (Collado and Fensterseifer, 2011).

4.3.3 Control variables

In order to increase the accuracy of the predictions in this study, the choice is made to include three control variables. In particular, these variables proved to be useful in previous studies with respect to board and/or CSR related studies (Hafsi and Turgut, 2013).

The first control variable is firm size. Galbreath (2011) argues that there is no convincing (empirical or theoretical) evidence with respect to the use of one particular measure of firm size. In this study, chosen is to measure firm size by the absolute number of employees working in the firm. An other often chosen control variable with respect to CSR is firm industry (Boulouta, 2013). Chosen is to use firm industry as a second control variable. In particular, this variable includes 14 sectors the 57 Dutch firms are active in: industry, financial, retail, transport, service, construction, electronics, real estate, food, chemistry, ICT, pharmacy, logistics and media sector. In order to measure firm industry in an appropriate way, a classification is made. Firm industry will be transformed into a dichotomous variable, with 1 for business-to-business sector (B2B) and 0 for business-to-consumer sector (B2C). B2B is defined as trade that takes place between two firms in the supply chain, except the relationship of a firm with the last link in the supply chain (the end consumer) (Jones, 2010; Slack et al., 2010). B2C is defined as trade that takes place between a firm and the firm’s network of individual customers (Jones, 2010). In particular, this is trade between the firm and the end customer (the last link in the supply chain) (Slack et al., 2010). B2B sector consists of the sectors: industry, construction, electronics, real estate, food, chemistry, pharmacy and logistics. B2C sector consists of the sectors: financial, retail, transport, service, ICT and media. This classification is based on previous research and on the specific activities of the firms in their sectors.

(20)

Finally, board size will be used as a control variable. Board size is the total number of members in the board (total of executive and supervisory board).

4.4 Data analysis methods

This section will discuss the statistical analysis in order to test if the hypotheses should be supported or rejected. Also, this section will mention the software that is used to test the hypotheses. The main goal of these statistical analyses is to answer the research question.

4.4.1 Statistical analysis

There are several statistical analyses needed in order to draw some conclusions with regard to the main research question. In particular, statistical tests must be carried out in order to review if the hypotheses can be supported or have to be rejected.

First, in each hypothesis must be examined if there is a relationship between two variables. In particular, to test if there is a relation between the variables without drawing conclusions with respect to causality. In order to test this possible relationship, the correlation test is an appropriate test to carry out. In fact, there are several different types of correlation tests (Huizingh, 2010). In this study, the variables of hypothesis 1 to H4c meet the requirements of the Pearson correlation and therefore will be tested with the use of the Pearson correlation to discover if there is a relationship between the variables. Also, the control variables firm size and board size meet the requirements of the Pearson correlation and will also be tested with this type of correlation. Firm industry is transformed into a dummy variable in order to meet the requirements of the Pearson correlation (and also for the regression analyses). In particular, firm industry is a dichotomous variable, with 1 for business-to-business sector (B2B) and with 0 for business-to-consumer sector (B2C).

(21)

21 First, for each hypothesis a set of multiple regression analyses will be performed of the control variables on sustainability in order to create a base model for the analyses. In particular, the absolute number of employees (ANE), firm industry (FI) and board size (BS) on the different sustainability variables (Y). Y is the dependent variable; in hypothesis 1 the dependent variable Y is transparency of

sustainability. In hypotheses 2, 3 and 4 the dependent variable Y is sustainability policies. In the sub

hypotheses 2a to 2c, 3a to 3c and 4a to 4c the dependent variable Y are the (economic, environmental and social) dimensions within sustainability policies. The following regression model is used for this set of analyses:

Y = β0 + β1 ANE + β2 FI + β3 BS + e

Second, the hypotheses are tested in order to test if the developed hypotheses are supported or should be rejected. In order to test hypothesis 1, multiple regression analysis will be performed of diversity in boards and control variables on transparency of sustainability. In particular, multiple regression analysis will be performed of mean age (MA), absolute number of women (ANW), relative number of women (RNW), absolute number of foreign members on the board (ANF), relative number of foreign members on the board (RNF), absolute number of employees (ANE), firm industry (FI) and board size (BS) on transparency of sustainability (TOS). The following regression model is used for hypothesis 1:

TOS = β0 + β1 MA + β2 ANW + β3 RNW + β4 ANF + β5 RNF + β6 ANE + β7 FI + β8 BS + e In order to test hypotheses 2 and sub hypotheses, multiple regression analysis will be performed of mean age and control variables on sustainability. In particular, multiple regression analyses will be performed of mean age (MA), absolute number of employees (ANE), firm industry (FI) and board size (BS) on the different sustainability variables (Y). In hypothesis 2 the dependent variable Y is

sustainability policies (SP). In hypotheses 2a to 2c the dependent variable Y are the different dimensions within sustainability. In particular, in H2a the dependent variable Y is the economic dimension (ECD), H2b the dependent variable Y is environmental dimension (EVD) and in H2c the

dependent variable Y is the social dimension (SD) within sustainability. The following regression model is used for hypothesis 2 and sub hypotheses:

Y = β0 + β1 MA + β2 ANE + β3 FI + β4 BS + e

(22)

environmental dimension (EVD) and in H3c the dependent variable Y is the social dimension (SD) within sustainability. The following regression model is used for hypothesis 3 and sub hypotheses:

Y = β0 + β1 ANW + β2 RNW + β3 ANE + β4 FI + β5 BS + e

In order to test hypotheses 4 and sub hypotheses, multiple regression analyses will be performed of nationality diversity and control variables on sustainability. In particular, multiple regression analyses will be performed of absolute number of foreign members on the board (ANF), relative number of foreign members on the board (RNF), absolute number of employees (ANE), firm industry (FI) and board size (BS) on the different sustainability variables (Y). In hypothesis 4 the dependent variable Y is sustainability policies (SP). In hypotheses 4a to 4c the dependent variable Y are the different

dimensions within sustainability. In particular, in H4a the dependent variable Y is the economic dimension (ECD), H4b the dependent variable Y is environmental dimension (EVD) and in H4c the

dependent variable Y is the social dimension (SD) within sustainability. The following regression model is used for hypothesis 4 and sub hypotheses:

Y = β0 + β1 ANF + β2 RNF + β3 ANE + β4 FI + β5 BS + e

Finally, multiple regression analyses will be performed of mean age (MA), absolute number of women (ANW), relative number of women (RNW), absolute number of foreign members on the board (ANF), relative number of foreign members on the board (RNF), absolute number of employees (ANE), firm industry (FI) and board size (BS) on the different sustainability variables (Y). Y is the dependent variable; in hypothesis 1 the dependent variable Y is transparency of sustainability

(TOS). In hypotheses 2, 3 and 4 the dependent variable Y is sustainability policies (SP). In the sub

hypotheses 2a to 2c, 3a to 3c and 4a to 4c the dependent variable Y are the (economic, environmental and social) dimensions within sustainability policies.

The following regression model is used for this analysis:

Y = β0 + β1 MA + β2 ANW + β3 RNW + β4 ANF + β5 RNF + β6 ANE + β7 FI + β8 BS + e. In each of these regression models, β0 is a constant, the intercept. The e is the error term; this term indicates the difference between the true value Y and the value of Y predicted by the model (Huizingh, 2010). For the correlation test and the multiple regression analysis, a significance level of 0.05 (p < 0.05) will be adopted.

4.4.2 Software

In order to perform the statistical analysis, software is needed. In this study, SPSS version 20 is used in order to test all the hypotheses9.

(23)

23

5. RESULTS

5.1 Descriptive results

Before testing the hypotheses, it is useful to describe the variables in order to get an overview and an insight of the involved variables. In particular, this section will focus on the descriptive results of the variables.

To begin with, it is interesting to look at the board data (independent variables) of the involved 57 Dutch firms. The 57 Dutch listed firms consist of a total of 568 board members with an average of 9,96 members on the board (total of supervisory and executive board). In particular, this means there are approximately 10 people on the board.

Board diversity is measured by different age, gender and nationality. The mean age of the board members ranges from 48,00 to 67,60 years. Statistics show that Dutch board members have a mean age of 58,57 years. Looking into the statistics of gender, results show that there are 73 female board members of the 568 total number of board members (only 12,85% on the board is women). Dutch listed firms have a mean of 1,28 women on their boards. Statistics also show that these 57 Dutch listed firms have a maximum of 4 women on their boards. Summarized, the results show that there are still relatively few women on Dutch boards compared to the number of men. The last board variable is nationality; the number/percentage of foreigners in Dutch boards. Results show that there are 190 foreigners on the boards (33,45%) of the 57 Dutch listed firms, with a mean of 3,33 foreign members on the board. Statistics also show that the majority of board members in Dutch listed firms are from Dutch nationality. Table 1 summarizes the most important descriptive results of the board data.

Table 1 Descriptive result of board data of 57 Dutch listed firms.

It is also interesting to look at the sustainability data (dependent variable) of the involved 57 Dutch listed firms. Different sustainability scores are used to measure the extent to which sustainability policies of Dutch firms are explicit available in external reporting. These descriptive results can be useful in order to draw some conclusions with respect to the extent to which sustainability policies of firms are explicit available in external reporting.

Board size Age of the

board members Women on the board (gender) Foreigners on the board (nationality)

Total number 568 members 73 members 190 members

Mean number in the board 9,96 members 58,57 years 1,28 members 3,33 members

(24)

Transparency scores are used in order to discover the overall extent firms are transparent with respect to sustainability. Statistics show that the transparency score ranges from 20 to 199 points. The mean transparency score is 119,60. However, results indicate that there are some values missing. In particular, there is no transparency score given to 2 firms and thus the sample consist of 55 firms instead of 57 Dutch listed firms.

On each individual dimension within sustainability, the firms could earn a maximum of 10 points. The first dimension within sustainability is the economic dimension measured by tax scores. The mean tax score achieved is 2,33 points. Results show that the majority of firms scored 0 or 3 points on taxation. Also, with respect to taxation are 2 values missing and the sample for the economic dimension within sustainability consist thus of 55 Dutch listed firms. The second dimension within sustainability is the environmental dimension measured by biodiversity. Overall, the firms scored well on the environmental dimension within sustainability with a mean score of 4,90 points. However, there are a lot of values missing. In particular, 26 values are missing which indicates that the sample only consists of 31 firms instead of 57 Dutch listed firms. This smaller sample could potentially have an impact on the reliability of the biodiversity scores. The last dimension within sustainability is the social dimension measured by human rights score. Results show a mean score of 4,05 points on human rights. In the sample for human rights are no values missing.

The economic, environmental and social dimensions encompass the (total) sustainability score. In particular, the extent to which sustainability policies of firms are explicit available in external reporting are measured by using the sustainability score. On each individual dimension within sustainability the firm could earn 10 points and thus the firms could earn a maximum of 30 points on total sustainability. However, as noted earlier there are some values missing on these individual dimensions within sustainability. Therefore, the sample for the sustainability score consist only of the firms that have received scores for all three individual dimensions. In particular, there are 26 values missing and thus the sample consists of 31 Dutch listed firms with respect to the total sustainability score. These 31 firms achieved a mean sustainability score of 12,29 points of the total maximum of 30 points that they could have scored. Results show that none of these firms scored a sustainability score higher than 20 points.

(25)

25 Table 2 Descriptive results of sustainability data of Dutch listed firms.

The control variables are also part of the statistical analysis and therefore descriptive results are useful. Statistics show the mean absolute number of employees working in the firm is 38.247,16 employees. In sum, there are 2.180.088 employees working in the 57 included Dutch listed firms. In particular, the absolute number of employees working in Dutch listed firms’ measures firm size. The control variable firm industry is measured by the sectors the included Dutch firms are active in. In particular, these 57 firms are active in 14 sectors. The majority of firms (12,3% of the firms) are active in the industry sector. Also, the financial, service and construction sector are popular sectors. However, the pharmacy sector is the least popular. In particular, only one firm is active in this sector. The sectors are divided in two groups (in order to measure firm industry in an appropriate way): B2B sector and B2C sector. B2B sector consists of 7 sectors: industry, construction, electronics, real estate, food, chemistry, pharmacy and logistics. B2C sector consists of 6 sectors: financial, retail, transport, service, ICT and media. This classification is based on previous research and on the activities of the firms in their sectors. Descriptive results show there are 33 firms (57,9%) included in the B2B sector and 24 firms (42,1%) included in the B2C sector of the total 57 firms in the sample for firm industry. The most important descriptive results of firm industry are summarized in table 3. The control variable board size was already discussed above because it was useful to describe the other board variables.

Number of firms active in this sector Percentage of firms active in this sector

Industry (B2B) 7 12,3% Financial (B2C) 6 10,5% Retail (B2C) 4 7% Transport (B2C) 2 3,5% Service (B2C) 6 10,5% Construction (B2B) 6 10,5% Electronics (B2B) 5 8,8% Real estate (B2B) 4 7,0% Food (B2B) 5 8,8% Chemistry (B2B) 3 5,3% ICT (B2C) 3 5,3% Pharmacy (B2B) 1 1,8% Logistics (B2B) 2 3,5% Media (B2C) 3 5,3% Total 57 firms 100%

Table 3 Descriptive results of the control variable firm industry.

Transparency Taxation Biodiversity Human rights Sustainability

(total)

Sample size 55 firms 55 firms 31 firms 57 firms 31 firms

(26)

5.2 Testing hypotheses

After describing the relevant variables, this section will focus on testing the hypotheses using Pearson correlation and multiple linear regression analysis. As noted earlier, the correlations will be performed first in order to discover if there is a relationship between the variables.

5.2.1 Pearson correlation

In this section, the variables in each hypothesis will be discussed and tested on whether there is a correlation between the variables. Table 3 summarizes all the performed Pearson correlations. Hypothesis 1 consists of the variables: diversity in boards and transparency of sustainability. Diversity in boards is measured by the variables: mean age of board members, absolute and relative number of women on the board (gender diversity) and absolute and relative number of foreign members on the board (nationality diversity). Transparency score is used to measure the (dependent) variable transparency of sustainability. Results show that the mean age of board members does not correlate significantly with transparency score (r = 0.191, p = 0.081). However, the Pearson correlation shows a significant effect between the correlation of the absolute number of women on the board and transparency score (r = 0.454, p < 0.01). In particular, this means when the absolute number of women on the board is higher, the transparency score is higher (thus, firms are more transparent with respect to sustainability) and vice versa. Also, the relative number of women on the board correlates significantly with the transparency score (r = 0.319, p < 0.01). This means when the number of women is higher compared to the total number of board members, the transparency score is higher (thus, firms are more transparent with respect to sustainability) and vice versa. The last variables that are components of board diversity are the absolute and relative number of foreign members on the board. Contrary to expectations, results show both absolute numbers of foreign board members (r = 0.199, p = 0.073) and relative number of foreign board members (r = 0.115, p = 0.202) does not significantly correlate with transparency score.

Summarized, there is only a correlation found between the absolute number of women and the relative number of women on the board (gender diversity) and the extent to which firms are transparent with respect to sustainability (transparency of sustainability).

(27)

27 board members and the economic dimension within sustainability (taxation), hypothesis 2b consists of mean age and the environmental dimension within sustainability (biodiversity) and hypothesis 2c consists of mean age and the social dimension within sustainability (human rights). Pearson correlation results show no significant correlation between the mean age of board members and taxation (r = -0.128, p = 0.420) and biodiversity (r = 0.047, p = 0.401). However, results show a significant correlation between mean age and human rights (r = 0.254, p < 0.05). This indicates that higher mean age leads to more reporting about social issues concerning sustainability and vice versa. Summarized, mean age has no relationship to the extent to which sustainability policies of firms are explicit available in external reporting and the economic and environmental dimension within sustainability. However, mean age has a relationship with the social dimension within sustainability (human rights).

Hypothesis 3 consists of the variables: gender diversity and sustainability policies. The absolute number of women and the relative number of women on the board measure gender diversity. As noted earlier, sustainability score, tax score, biodiversity score and human rights score, measures sustainability and the different dimensions. Contrary to expectations, the Pearson correlation with respect to hypothesis 3 shows no significant result for both the absolute number of women on the board and sustainability score (r = 0.132, p = 0.239) as well as the relative number of women on the board and sustainability score (r = 0.139, p = 0.228). Also, hypothesis 3a shows no significant result; there is no significant correlation between the absolute number of women (r = -0.145, p = 0.146) and the relative number women on the board and tax score (r = -0.005, p = 0.485). However, hypothesis 3b and 3c shows significant correlation. In particular, there is a relationship found between the absolute number of women (r = 0.339, p < 0.05) and relative number of women on the board (r = 0.319, p < 0.05) and biodiversity score. Also, there is a relationship found and between the absolute number of women (r= 0.354, p< 0.01) and relative number of women on the board and human rights score (r = 0.229, p < 0.05), although these relationships are weak. In particular, this means when gender diversity is higher there is more reporting about environmental and social issues concerning sustainability and vice versa. Summarized, there is a significant correlation found between gender diversity and the environmental (biodiversity) and social dimension within sustainability (human rights).

(28)

significant result. In particular, there is no significant correlation between the absolute number (r = -0.172, p = 0.105) and relative number of foreign members on the board and tax score (r = -0.136, p = 0.160). Nationality diversity shows a significant relation with the environmental dimension within sustainability. Nationality diversity and social dimension within sustainability are partial correlated. There is a signification correlation between the absolute number (r = 0.348, p < 0.05) and relative number of foreign board members and biodiversity score (r = 0.433, p < 0.01) and also there is a relationship found between the absolute number of foreign board members and human rights score (r = 0.226, p < 0.05). This result indicates that when nationality diversity is higher there is more reporting about environmental issues concerning sustainability and vice versa. Also, the results indicate when the number of foreign board members is higher this leads to more reporting about social issues concerning sustainability and vice versa. However, there is no significant result found for the relative number of foreign board members and human rights score (r = 0.202, p = 0.066). Summarized, with respect to nationality diversity and sustainability there is only a significant correlation found between nationality diversity and the environmental dimension within sustainability (biodiversity) and between the number of foreign board members and the social dimension within sustainability (human rights).

In addition to the variables, which are mentioned in the hypotheses, there are also Pearson correlations performed with respect to the control variables.

The first control variable is firm size. The absolute number of employees working in the firm measures firm size. Results show a significant correlation between absolute number of employees and biodiversity score (r = 0.407, p < 0.05) and between the number of employees and human rights score (r = 0.228, p < 0.05). This indicates that bigger firms report externally more about environmental and social issues concerning sustainability and vice versa. However, results show no significant correlation between the absolute number of employees and transparency score (r = 0.177, p = 0.098), sustainability score (r = 0.290, p = 0.057) and tax score (r = -0.179, p = 0.096). The second control variable is firm industry. Firm industry is measured by the 14 sectors the firms are active in. Firm industry is transformed into a dichotomous variable, with 1 for B2B and 0 for B2C. Results show only a significant correlation between firm industry and sustainability score (r = 0.393, p < 0.05). This result indicates that firm industry has an influence on the extent to which sustainability policies are explicit available in external reporting and vice versa. There are no significant results for a relation between firm industry and transparency of sustainability (r = 0.006, p = 0.482) and the economic (r = 0.150, p = 0.137), environmental (r = 0.251, p = 0.086) and social dimension within sustainability (r = 0.135, p = 0.158).

(29)

29 correlates with transparency score (r = 0.518, p < 0.01) and with tax score (r = -0.316, p < 0.01) and human rights score (r = 0.363, p < 0.01). In particular, this result indicates that the board size has an influence on the extent to which firms are transparent with respect to sustainability and the extent to which they report about economic and social issues concerning sustainability and vice versa. There was no significant relation found between board size and sustainability score (r = -0.111, p = 0.275) and biodiversity score (r = 0.062, p = 0.370). In particular, this indicates there is no relationship found between board size and the extent to which sustainability policies are explicit available in external reporting and the extent to which firms report about environmental issues concerning sustainability and vice versa.

Summarized, results show how firm size correlates to the extent to which firms report about environmental and social issues concerning sustainability. With respect to the control variable firm industry, there is a relationship found between firm industry and the extent to which sustainability policies are explicit available in external reporting. Results show that board size correlates with the extent to which firms are transparent with respect to sustainability. Also, board size correlates with the extent to which firms report about economic and social issues concerning sustainability.

Finally, it is also interesting to examine the possible correlations between the sustainability variables (the dependent variables).

It is expected that transparency of sustainability is associated with the extent to which sustainability policies are explicit available in external reporting and with the extent to which firms report about economic, environmental and social issues concerning sustainability. It is expected that when firms are transparent with respect to sustainability, they will also be transparent with respect to sustainability policies and the different dimensions within sustainability. However, the results show that transparency of sustainability only significantly correlates with tax score (r = -0.290, p < 0.05) and human rights score (r = 0.611, p < 0.01). There are no significant results found for transparency of sustainability and sustainability score (r = 0.138, p = 0.230) and biodiversity score (r = 0.130, p = 0.244). The results indicate that transparency of sustainability has an influence on the extent to which firms report about economic and environmental issues concerning sustainability and vice versa.

(30)

(31)

Variables (1) (2) (3) (4) (5) (6) (7) (8) (9) (10) (11) (12) (13)

(1) Mean age Correlation

Sig. (1-tailed) N 1 57 -.063 .321 57 -.100 .229 57 .042 .377 57 .162 .114 57 .029 .415 57 .187 .082 57 .020 .441 57 .191 .081 55 .196 .145 31 -.028 .420 55 .047 .401 31 .254* .029 57 (2) Women absolute number (gender diversity) Correlation Sig. (1-tailed) N -.063 .321 57 1 57 .886** .000 57 .521** .000 57 .436** .000 57 .411** .001 57 .023 .431 57 .596** .000 57 .454** .000 55 .132 .239 31 -.145 .146 55 .339* .031 31 .354** .003 57 (3) Women relative number (gender diversity) Correlation Sig. (1-tailed) N -.100 .229 57 .886** .000 57 1 57 .243* .034 57 .217 .052 57 .261* .025 57 -.017 .449 57 .261* .025 57 .319** .009 55 .139 .228 31 -.005 .458 55 .319* .040 31 .229* .043 57 (4) Foreign absolute number (nationality diversity) Correlation Sig. (1-tailed) N .042 .377 57 .521** .000 57 .243* .034 57 1 57 .924** .000 57 .401** .001 57 .216 .053 57 .764** .000 57 .199 .073 55 .156 .201 31 -.172 .105 55 .348* .028 31 .266* .023 57 (5) Foreign relative number (nationality diversity) Correlation Sig. (1-tailed) N .162 .114 57 .436** .000 57 .217 .052 57 .924** .000 57 1 57 .324** .007 57 .190 .078 57 .562** .000 57 .115 .202 55 .194 .148 31 -.136 .160 55 .433** .007 31 .202 .066 57 (6) Employee number (firm size) Correlation Sig. (1-tailed) N .029 .415 57 .411** .001 57 .261* .025 57 .401** .001 57 .324** .007 57 1 57 -.110 .208 57 421** .001 57 .177 .098 55 .290 .057 31 -.179 .096 55 .407* .012 31 .228* .044 57

(7) Firm Industry Correlation Sig. (1-tailed) N .187 .082 57 .023 .431 57 -.017 .449 57 .216 .053 57 .190 .078 57 -.110 .208 57 1 57 .108 .212 57 .006 .482 55 .383* .017 31 .150 .137 55 .251 .086 31 .135 .158 57

(8) Board size Correlation

Sig. (1-tailed) N .020 .441 57 .596** .000 57 .261* .025 57 .764** .000 57 .562** .000 57 .421** .001 57 .108 .212 57 1 57 .518** .000 55 -.111 .275 31 -.361** .003 55 .062 .370 31 .363** .003 57 (9) Transparency score Correlation Sig. (1-tailed) N .191 .081 55 .454** .000 55 .319** .009 55 .199 .073 55 .115 .202 55 .177 .098 55 .006 .482 55 .518** .000 55 1 55 .138 .230 31 -.290* .017 54 .130 .244 31 .611** .000 55 (10) Sustainability score Correlation Sig. (1-tailed) N .196 .145 31 .132 .239 31 .139 .228 31 .156 .201 31 .194 .148 31 .290 .057 31 .383* .017 31 -.111 .275 31 .138 .230 31 1 31 .592** .000 31 .650** .000 31 .667** .000 31 (11) Tax score (economic dimension) Correlation Sig. (1-tailed) N -.028 .420 55 -.145 .146 55 -.005 .458 55 -.172 .105 55 -.136 .160 55 -.179 .096 55 .150 .137 55 -.361** .003 55 -.290* .017 54 .592** .000 31 1 55 .124 .253 31 -.141 .151 55 (12) Biodiversity score (environmental dimension) Correlation Sig. (1-tailed) N .047 .401 31 .339* .031 31 .319* .040 31 .348* .028 31 .433** .007 31 .407* .012 31 .251 .086 31 .062 .370 31 .130 .244 31 .650** .000 31 .124 .253 31 1 31 .263 .076 31 (13) Human rights score (social dimension) Correlation Sig. (1-tailed) N .254* .029 57 .354** .003 57 .229* .043 57 .266* .023 57 .202 .066 57 .228* .044 57 .135 .158 57 363** .003 57 .611** .000 55 .667** .000 31 -.141 .151 55 .263 .076 31 1 57

Table 3 Correlation matrix.

Board diversity and Sustainability