Gender Diversity in Dutch Boardrooms: Impact on the Monitoring and Advising Duties

(1)

[1]

Date: June 29, 2015

Name: H.J.M. Bonestroo

First name: Jelle

Student ID: S2597845

First supervisor: Mr. Ph.D. M.Sc. S. Mukherjee Second supervisor: Mrs. Ph.D. M.Sc. Y. Karaibrahimoglu

Faculty: Economics and Business

Subject: Master Thesis Accountancy

Code: EBM869A20

a

Gender Diversity in Dutch Boardrooms:

Impact on the Monitoring and

(2)

[2]

Gender Diversity in Dutch Boardrooms:

Impact on the Monitoring and

Advising Duties

Date: June 29, 2015

Name: H.J.M. Bonestroo

First name: Jelle

Student ID: S2597845

Address: Horseler Pol 4

Postcode: 3863XD Nijkerk

Mail: jbonestroo@chello.nl

Phone number: 00316-29496269

First supervisor: Ph.D. M.Sc. S. Mukherjee

Second supervisor: Mrs. Ph.D. M.Sc. Y. Karaibrahimoglu

Faculty: Economics and Business

Subject: Master Thesis Accountancy

Code: EBM869A20

Word Count excl. tables: 13.531

Abstract: Regulators pleaded for more female directors in the Netherlands since the female participation on Dutch boards is too low, as indicated by Dutch Female Board Index (2014). However, Dutch research into this subject is still lacking. In a sample of listed companies on the AEX, AMX, ASCX in the period 2004-2014, I provide robust evidence that gender diversity between non-executives has significant effects on their monitoring and advising duties. I find, after controlling observable board and firm characteristics, that gender diversity is positively related to the total amount and the equity-based part of CEO compensation, which suggests that the interests of the CEO and shareholders are more aligned. Furthermore, I find that gender diversity is positively related to the quality of investments, which suggests that gender diversity between non-executives results in more selective decision-making regarding investment decisions. Finally, I find that gender diversity between non-executives is positive (not) related to the long-term performance (short-term performance).

(3)

[3]

1. Introduction

n 2005, female directors occupied around 3 and 5 percent of the total executive and non-executive positions of listed companies in the Netherlands. These proportions increased to 6 and 19 percent in 2014 (The Dutch Female Board Index, 2014). Furthermore, 34 percent (30/87) of the listed companies still do not have female directors on board in 2014 in the Netherlands (The Dutch Female Board Index, 2014). This points out clearly: there is still a lack of sufficient representation by both executive and non-executive female directors in Dutch boardrooms.1_{The topic of gender diversity within the board of directors has garnered attention}

from academics (e.g. Erhardt, Werbel & Shrader, 2003; Hillman & Dalziel, 2003; Adams & Ferreira, 2009; Carter D’Souza, Simkins & Simpson, 2010; Gul, Srinidhi & Ng, 2011; Lückerath-Rovers, 2013; Ben-Amar, Francoeur, Hafsi & Labelle, 2013). However, Dutch research into the subject is still lacking. Therefore, I focus in this study on gender diversity across the boardrooms of companies listed on the Dutch stock market for the years 2004–2014, and I investigate the effects of gender diversity on the monitoring and advising duties of boards. Gender diversity and other classes of diversity such as age, tenure, expertise and ethnicity (Van der Walt & Ingley, 2003; Harrison & Klein, 2007) within boards have become an international corporate governance research subject, due to demographic and political developments and governance shortcomings in companies exposed to the financial crisis of 2007 (ICAEW, 2014). As a result, governance codes have defined best practice regulations or mandatory requirements concerning gender diversity because increasing “female representation on boards of directors may link firms to different customers, current and potential employees and important suppliers including investors” (Hillman, Shropshire & Cannella, 2007: 944). Hence, it is a way to strengthen board capital,2 which is expected to have direct effects on board responsibilities such as understanding the business, fostering social acceptability, maintaining independence, monitoring of operations and establishing a good control environment (Cravens, & Wallace, 2001; Corporate Governance Code Monitoring Committee, 2008; ICAEW, 2014). The Dutch corporate governance code, section III 3-1, reads that board diversity is an important way to guarantee the independence and effectiveness of the board of directors (Corporate Governance Code Monitoring Committee, 2008). Furthermore, section III point 38 specifies that “the Committee has included an explicit provision that companies are expected to apply and disclose a specific objective in relation to board diversity” (Corporate Governance Code Monitoring Committee, 2008: 55). Companies in other European countries, such as Norway, Finland, France, Italy and Spain, are subjected to legal requirements regarding female directors.3

1_{The most common governance structure in the Netherlands is a two-tier board where “the management board is}

entirely composed of executive directors” (Maassen & van den Bosch, 1999: 33) and “the supervisory board is composed entirely of non-executive directors, which secures an independent composition” (Maassen & van den Bosch, 1999: 31). To avoid confusion, I comply with the international terminology; therefore I use the following definitions: non-executives (supervisory board) and executives (management board).

2 _{Board capital involves breadth and depth. “Board capital breadth is defined as the portfolio of directors’}

functional, occupational, social, and professional experiences and extra-industry ties and captures the heterogeneity of the directors’ human and social capital.” (Haynes & Hillman; 2010: 1145). “Board capital depth

refers to the embeddedness of directors in the firm’s primary industry through interlocking directorships, managerial positions, or occupational experience in the primary industry of the firm, and is the sum of the director’s intra-industry human and social capital” (Haynes & Hillman; 2010: 1145).

3_{Norway’s mandatory requirement of at least 40% female non-executive directors on board is effective for all}

public limited firms since 2006. Non-complying firms will be dissolved by court (The Norwegian Corporate Governance Code, 2014). Boards with more than nine members have to comply with the 40% quota. For smaller boards there are different rules(Bøhren & Staubo, 2015). In Finland a comparable gender quota of 40% regarding the board of directors is required for state-owned companies, effective from 2006. Since 2008, listed companies in Sweden have to strive for equal positions (males and females) on boards. In France, companies have to strive

(6)

[6]

Furthermore, a proposal of a 40% EU gender quota regarding large publicly listed companies4

was approved in 2013 (the underrepresented gender has to make up to at least 40%).5 The legal implemented quotas are often justified by the business and economic case for diversity “based on the notion that given changing demographic trends and skill shortages, the effective use of diverse skills within an organization makes good business sense” (Cassell, 2000: 269). Dutch companies that meet at least two of the three requirements of art. 2.397 paragraph 1 of the Civil Code6 have to comply or explain regarding a directive target of 30% for female directors in the Netherlands since 2009. Non-complying companies have to disclose information concerning their representation of directors on a board and how companies should achieve the target in the future. From 2016 on, companies will be required to comply with the 30% gender quota (Lückerath-Rovers, 2011). Issues could arise concerning the proportion of expected female representation. Female executives are expected to hold 8.6 percent of the total executive positions, where female executives are expected to hold 23.5 percent of the total non-executive positions, as of January 2016 (The Dutch Female Board Index, 2014). Therefore, this is an opportune moment to analyze the impact of increasing gender diversity on monitoring and advising functions has had so far, in order to set the stage for what to expect from a mandatory requirement.

In this study I question whether there is a relationship concerning gender diversity between

non-executives and their advising and monitoring duties, for listed firms at the Dutch stock market.

The following sub-questions are settled: (Q1) Why should gender diversity between non-executives be essential for listed companies at the Dutch stock market? (Q2) Are there differences between companies with and without female directors in the Netherlands? (Q3) Is there a relationship concerning gender diversity between the non-executive directors and their duties related to the quality of financial reporting? (Q4) Is there a relationship concerning gender diversity between the non-executive directors and CEO compensation? (Q5) Is there a relationship concerning gender diversity between the non-executive directors and their duties related to the quality of investments?

This study is relevant for a number of reasons. First, it contributes to the existing literature on gender diversity. Adams and Fereirra (2009) investigated the effects of gender diversity within boards on corporate governance and financial performance. They provided evidence, based on U.S. companies, suggesting that gender-diverse boards have better attendance behavior, hold CEOs more accountable to stock performance, and compensate CEOs more equitably. The results presented by Adams and Fereirra (2009) are convincing, however, the study by Carter et al. (2010) suggests that more research is warranted. Secondly, this study complements the work of Lückerath-Rovers (2013), who state that more investigation regarding gender diversity after the normative target in the Netherlands since 2009 is justified. Thirdly, understanding the effects of gender diversity within boards on monitoring and advising duties is important for regulators in the Netherlands. If gender diversified boards are better protectors of share- and stakeholders due to better monitoring and advising, a requirement might be mandatory rather than the existing “comply or explain” regulation.

for a 40% gender balanced board since 2010. As of 2017, large listed companies are expected to have to comply with a 40% mandatory gender quota. Listed and state-owned firms have had to comply with a 33% gender quota in Italy since 2012 (Bianco et al., 2015).

4_{Small and mid-sized companies, which have less than 250 employees and a turnover below 50 million EUR, are}

excluded (The Dutch Female Board Index, 2014).

5 _{Large public companies have to comply with the gender quota by 2020 (}_{The Dutch Female Board Index, 2014}_),

and the underrepresented gender has to make up 30% of the total executive and non-executive positions.

6_{The requirements within art. 2.397 paragraph 1 Civil Code are divided into three standards: (a) the value of assets}

presented in the balance sheet is at least 17.5 million EUR; (b) the net sales are at least 35 million EUR for the fiscal year and (c) the average number of employees is at least 250 for the fiscal year.

(7)

[7]

I provide new robust evidence that gender diversity between non-executives, even after controlling for observable firm and board characteristics, does have significant effects on the group performance of executives. In my analysis, I find that gender diversity between non-executives is positively related to the equity-based compensation of the CEO, which suggests that gender diversity between non-executive directors results in keeping a CEO more accountable with long-term incentives (e.g. growth in shareholder value). Furthermore, I find that the total compensation of the CEO is higher in companies with gender diversified executives. I also find a negative relationship concerning gender diversity between non-executives and over- or underinvestment. These findings suggest that gender diversity could result in less information asymmetry, as managers of firms are more selective concerning investment decisions due to better monitoring and advising by gender-diversified non-executives. Shareholders could take this fact into account concerning their investment decisions. Finally, I find a positive (no) relationship between gender diversity and the firm’s long-term (short-term) performance. These results suggest that gender diversity between non-executives should enhance shareholder wealth in the long term.

The remainder of this paper is structured as follows: section 2 gives an overview of the existing literature on gender diversity. Section 3 points out the theoretical framework and hypothesis development. Section 4 then presents the sample, variable overview and data. Section 5 details the results. Finally, sections 6 and 7 present the discussion and conclusions.

(8)

[8]

2. Literature Review

2.1 Corporate Governance in the Netherlands

The most common governance structure in the Netherlands is a two-tier board which consists of a supervisory board and a management board (Maassen & van den Bosch, 1999; Lückerath-Rovers, 2013). The Dutch governance structure where non-executives (supervisory board) and executives (management board) are separated (Maassen & van den Bosch, 1999) is different from the international one-tier structure, where all directors are together on one board (Lückerath-Rovers, 2013). The expected benefit of a two-tier structure is that non-executives are more independent and that the positions of CEO and chairman are separated (Rouyer, 2013). In fact, under a two-tier structure, non-executives face an independence paradox: “in obtaining adequate information non-executives are dependent on the executives they are expected to supervise and to be independent from” (Hooghiemstra & van Manen; 2004: 314).

The non-executives in the Netherlands’s duties are to “supervise the policies of the management board and the general affairs of the company and its affiliated enterprise, as well as to assist the management board by providing advice. In discharging its role, the supervisory board shall be guided by the interests of the company and its affiliated enterprise, and shall take into account the relevant interests of the company's stakeholders” (Corporate Governance Code Monitoring Committee, 2008: 19). The Dutch corporate governance code section III 1.3 mentions that characteristics such as gender, age, function(s), date of appointment, tenure and nationality for every single non-executive have to be disclosed to guarantee the expertise and independence of the non-executives (Corporate Governance Code Monitoring Committee, 2008).

The executives have “to manage the company, which means, among other things, that it is responsible for achieving the company’s aims, the strategy and associated risk profile, the development of results and corporate social responsibility issues that are relevant to the enterprise. The management board is accountable for this to the supervisory board and to the general meeting” (Corporate Governance Code Monitoring Committee, 2008: 11).

2.2 The Different Cases and Gender Diversity

There have been various understandings concerning the question of why gender diversity should be effective or occur within boards. The moral case for gender diversity is based on a view of social responsibility (Mattis, 2000; Fondas, 2000). In this case, members on boards try to signal that they operate as a credible, integrated and non-discriminating board. Social responsibility goes beyond the idea of value maximization and involves all relevant interests of the company’s stakeholders (Van der Walt & Ingley, 2003). Carver (2002) claims that board diversity should occur due to a moral obligation that is tucked away in the board’s stewardship role, by representing all diversified shareholders. The business case suggests that gender diversity within boards should occur as reaction to demographical developments of both labor and consumer markets, due to aging, urbanization and multicultural influences (Mattis, 2000). Gender diversity within boards is also expected to occur for commercial reasons (Brennan & McCafferty, 1997; Hillman et al., 2007). Brammer, Millington and Pavelin (2009) suggest that female directors are favorable for companies who operate close to customers, which suggests that female directors could be nominated because of their commercial skills. A political perspective suggests that diversified boards bring in diverse social capital (Hillman & Canella, 2003), which should be essential as boards have to engage with the civil (democratic) society (Putnam, 1995). Therefore, Bilimoria (2000a) argues, female directors have different views on

(9)

[9]

markets, the environment and ethical developments, which should be beneficial for organizations. Van der Walt and Ingley (2003) give an overview of the possible motivators of (gender) diversity on boards for listed companies which are for example: companies benefit from expertise and knowledge, get access to strategic resources, have to comply with statutory or governance requirements or by their professional disciplines.

Still, there has been a visible lack of representation of female directors as indicated by the Dutch Female Board Index (2014) in the Netherlands, a multicultural and aging country (CBS, 2015), where companies operate in an open and international economy (Van Veen & Elbertsen, 2008). Burke (1997) argues that a lack of female directors occurs due to uncertainty about their corporate experience. The relative power over the board by a CEO during the appointment process suggests that the new director tends to be demographically similar to a CEO and vice versa (Westphal & Zajac, 1995). As a result of this, board performance would be harmed if boards were ensuring that the most suitable director possible was not selected (Burke, 1997; Carver, 2002).

2.3 Monitoring Duties and Gender Diversity

Earlier studies provided evidence suggesting that gender diversity is positively related to board independence (Van der Walt & Ingley, 2003; Carter et al., 2003; Adams & Ferreira, 2009; Bøhren & Staubo, 2015). A review paper by Cravens and Wallance (2001) gives a clear analysis regarding board independence as measured by outside directors. Increasing board independence is related to a better quality of monitoring, an increase in financial performance and shareholder wealth, a greater likelihood of removing poor performing CEOs and a decrease in financial statement fraud. Gender-diversified boards are expected to be better monitors for a firm because “directors with diverse skills, experiences and backgrounds are more likely to raise questions than simply echo the voice of management” (Selby, 2000: 239).

Earlier literature suggests a positive association between female directors who are seated in the boardroom and financial reporting (Gul et al., 2011; Srinidi, Gul & Tsui, 2011; Abbott, Parker & Presley, 2012; Gul, Hutchinson & Lai; 2013). Gul et al. (2011) suggest, based on firms in the U.S., that gender diversity on boards is positively related to the informativeness of stock prices. Furthermore, they found evidence that gender diversity on boards is positively associated with disclosure of private information. Another study based on a sample of 2200 U.S. firms in the period of 2001–2007 suggests that gender diversity is related to better quality of earnings forecasts (Gul et al., 2013). Abbott et al. (2012) found a negative relationship between financial restatements and female board representation based on U.S. firms during the years 1997–2002.

The results that Adams and Ferreira (2009) report, suggest a positive association between female directors and monitoring activities due to better board attendance behavior and a demand for greater accountability from poorly performing CEOs. Furthermore, their findings suggest that gender-diverse boards put more effort in to monitoring activities (e.g. manager compensation is more sensitive to performance) and hold more monitoring positions (e.g. in audit or governance committees). Upadhyay (2014) provided evidence, based on S&P firms over the years 2000–2003, suggesting that gender diversity on boards improves corporate governance systems, where this improvement is especially valuable for opaque firms or firms in complex environments. This should be rewarded by a lower required cost of capital from investors.

(10)

[10]

2.4 Advising Duties and Gender Diversity

Gender diversity in the boardroom could be of value for strategic decision-making by providing different perspectives based on different backgrounds and expertise (Westphal & Milton, 2000). Hence, a lack of diversity on boards could harm critical thinking (Mattis, 2000). Hillman et al. (2007) provided evidence, based on 1.000 U.S. large companies, which suggests that size, industry types and network effects have a significant impact on the likelihood on the appointment of female directors. These outcomes are consistent with the view that female directors should be appointed because of the different human and social capital they bring in (e.g. gains from different resources, knowledge, expertise and skills). Evidence from the U.K. suggests that female directors often have board functions if they have a close relation with final customers (e.g. Retail, Media and Banking) (Brammer, Millington, & Pavelin, 2007). Miller and Triana (2009) examined the effects of gender diversity on reputation and innovation. They found a positive mediating relationship between gender diversity on boards and innovation, reputation and financial performance. One of their findings is that innovation, measured by R&D expenses as percentage of sales, positively mediates the relationship between gender diversity and financial performance. This finding suggests that firms with gender diversified boards are more creative during the decision-making process by bringing in different human and social capital. Riley and Chow (1992) suggest that gender-diverse boards are better advisors of the management because females are on average more risk averse regarding investment decisions than men, which should result in less homogeneity. Welbourne and Cycyota (2007) examined 543 IPO firms, and they came up with evidence suggesting that IPO’s were significantly more successful when females held executive positions. Their results suggest that female executives are beneficial within the innovation and problem-solving processes.

2.5 Legal Requirements and Gender Diversity

As mentioned before, political measures, e.g. mandatory or normative quotas), are expected to lead to an increase in female representation on boards. Recently, Bøhren and Staubo (2015) have investigated the impact of the mandatory 40% gender quota on board independence and firm value in Norway. Non-complying firms will be dissolved by court. Their findings suggest that this legal gender quota strongly increases board independence;7 however, it reduces firm value. The gender quota has the most impact on small, young, non-listed firms who have powerful stockholders (Bøhren & Staubo, 2015). A comparable study in Italy suggests that companies with family-affiliated female directors – more dependent boards – are smaller and have concentrated ownership (Bianco et al., 2015). These authors predict that a legal gender quota would affect these kinds of companies the most in Italy. A study by Ben-Amar et al. (2013) based on Canadian firms facing M&A decisions suggests that board diversity has, overall, neither a positive nor negative effect. High demographic diversity affects family-owned firms in a negative way and does not affect institutionally owned firms; low demographic

7_{Bøhren and Staubo}₍₂₀₁₅₎_{found evidence that a legal gender quota shifts up the average fraction of independent}

directors. They define an independent director as follows: “the board member is neither a full-time employee in the firm, a former employee, an employee of a closely related firm, related to a member of management, nor has business relationships with the firm” (Bøhren & Staubo, 2015:4). Furthermore, they distinguish grey (affiliated) directors with relation to members of the management. A possible explanation could be that female directors seldom belong to the old boys network. In contrast to this, it’s possible that firms appoint females as tokens since they represent their category. The results reported by Bøhren and Staubo (2015) suggest that alternative measures concerning board independence, e.g. fraction outside directors, less fraction directors or fraction grey and outside directors, do have a negative and significant relationship with financial performance.

(11)

[11]

diversity affects institutionally and family-owned firms positively, where statutory diversity has no influence on strategic decisions (Ben-Amar et al., 2013).

Legal gender quotas are expected to guarantee board independence or improve firm value by board capital. However, the literature suggests that legal gender quotas have some unintended effects. A tradeoff between independent (outside) directors with a monitoring role and dependent (inside) directors with an advising role exists (Adams, & Ferreira, 2007; 2009; Baldenius, Melumad, & Meng, 2014), and a legal gender quota can force independence above the optimal level, resulting in a decrease in firm value (Drymiotes, 2007; Ahern & Dittmar, 2012; Bøhren & Staubo, 2015).

2.6 Financial Performance and Gender Diversity

Previous studies have suggested that gender diversity within corporate boards should improve firm value or financial performance since it is expected that heterogeneous boards are more effective regarding their duties than homogeneous boards. Nevertheless, earlier studies came up with mixed evidence. Lückerath-Rovers (2013) provided evidence within the Netherlands. Her results, based on 99 listed companies on the AEX from 2005–2007, show that firms with female directors perform significantly better than those without female directors based on ROE. Ntim (2015) provided evidence from listed companies in South-Africa, a multicultural country (Andreasson, 2013). His results suggest that gender (and ethnic) diversity are positively associated with market valuation. Erhardt et al. (2003)found a significant relationship, using 127 large U.S. companies from 1993–1998, between the percentage of women (and ethnic minorities) on board and the ROA and ROI. In addition, Carter, Simkins and Simpson (2003) provided evidence from the U.S. that the proportion of women and ethnic minorities in the board increases firm value. Evidence from Spain suggests a positive relationship between gender diversity on boards and financial performance measured by Tobin’s Q (Campell & Minguez-Vera, 2008). Adams and Ferreira (2009) used U.S. firms, they found a negative (positive) relationship between financial performance and gender diversity when governance is strong (weak), based on the period of 1996–2003. Evidence from German listed firms in 2000– 2005 suggests that gender diversified boards at first have a negative effect on firm performance, however, a after a threshold of 30%, gender diversity should affect financial performance positively (Joecks, Pull & Vetter, 2013). Ahern and Dittmar (2012) investigated the relationship between gender diversity and Tobin’s Q, using Norwegian public limited firms from 2001– 2009. They found a negative relationship. Rose (2007) did not find a significant relationship between gender diversity on boards and the Tobin’s Q, using a sample of listed Danish firms during the period of 1998–2001. In addition, the study by Carter et al. (2010) did not find a significant relationship between female directors on boards (or board committees) and firm performance measured by the Tobin’s Q and ROA, using U.S companies over a five-year period (1998–2002). Carter et al. (2010) note in their study that the body of research is not easily comparable because used data, methods and time periods are different. Therefore, I summarized earlier financial performance-based studies, conforming to the study by Joecks et al. (2013), presented in Appendix A.

(12)

[12]

3. Theoretical Framework and Hypotheses

3.1

Decision-Making Behavior

Diversity can be described as the sum of differences between group members (Solonas, Nelvam, Navarro & Leiva, 2012) and can be divided into non-demographic and demographic diversity (Harrison, Price & Bell, 1998). Harrison and Klein (2007) distinguish diversity as separation, disparity and variety. Each category has its own maximum value, which reflects the optimal heterogeneity of a group (Solonas et al., 2012).

Separation. The members within a group differ concerning their personal beliefs, attitudes or

opinions (Harrison & Klein, 2007). Harrison and Klein (2007) predict that this diversity category has systematic consequences such as distrust and more interpersonal conflict or cohesion (if individual members are less or more different).

Disparity. The members of a group differ concerning their status, decision-making power and

income (Harrison & Klein, 2007). Predicted outcomes suggest that this kind of diversity between group members is related to their socially valued assets or resources, which should result in more competition between members(Harrison & Klein,2007).

Variety. In this category, group members are different concerning their biometric factors, e.g.

gender. Diversity-as-variety assumes that all members within a group have unique or distinctive information with regard to their individual industry experience and network ties. Harrison and Klein (2007) argue that this should result in creativity and better decision quality.

Gender diversity can be categorized as diversity-as-variety and is expected to be beneficial to organizations because diverse groups are able to process information better without losing qualitatively useful information; this should result in better group performance, improved decision-making behavior and more creative thinking (Jackson, May & Whitney, 1995; Harrison & Klein, 2007). Furthermore, males and females have different views with regard to financial decisions; females are more risk averse (Riley & Chow, 1992), females formulate different questions by keeping different perspectives in mind (Burke, 1997; Van der Walt & Ingley, 2003), and finally, females are less aggressive with regard to investment strategies than males (Charness & Gneezy, 2012).

3.2

Critical Mass Theory

Gender diversity can be seen as a contribution within information processing and creative thinking (Harrison & Klein,2007), but there is uncertainty regarding the exact composition of gender-diversified groups and their impact on decision quality and information processing. Kanter (1997a,b) has introduced the critical mass theory, which express that there are different effects on group interaction within different group categories. These categories are discussed below.

Uniform groups. The members within uniform groups are the same with regard to their gender

(Kanter, 1997a).

Skewed groups. The dominant type of members within a group controls the other gender (e.g.

the males control the females). The underrepresented gender, called tokens (Kanter, 1997a), “are not treated as individuals, but as representatives for their category” (Joecks et al., 2013: 62).

Tilted groups. The members within tilted groups are less extremely distributed than in skewed

(13)

[13]

and knowledge within the interaction process (Kanter, 1997a); therefore creative thinking and information processing should be improved.

Balanced groups. The composition of females and males within groups becomes equally

distributed, so both females and males contribute their different skills and knowledge during the interaction process and gender-based difference becomes less (Kanter, 1997a).

“In Sum, critical mass theory postulates that until a certain threshold or critical mass of women in a group is reached, the focus of the group members is not on the different abilities and skills that woman bring into the group” (Joecks et al., 2013: 64).

3.3 Resource Dependence-Based View

The resource-based view expresses that the board of directors acts as a link between the company and the company’s environment (Pfeffer & Salancik, 1978): “A key insight of this perspective is that organizations are open systems, depending upon external entities for survival, and that the resulting uncertainties pose significant challenges and costs to the organizations” (Hillman et al., 2007: 942). The resource-based view includes three proposed benefits: (1) advice and counsel, (2) legitimacy and (3) communication. Advice and counsel refer to the greater range of perspectives within heterogeneous groups as compared to homogenous groups; this greater range should result in more conflict, alternative solutions, creative thinking and less cohesion (Hillman et al., 2007). An increase in gender diversity between non-executives could lead to better advice and council by better generating alternative solutions and keeping different environmental perceptions in mind. Legitimacy refers to “the pressure of stakeholders for gender diversity that firms depend upon, and few organized interests argue against such board appointments” (Hillman et al., 2007: 944). Carter et al. (2010) suggest that diverse boards have more access to talent because board diversity links the firm to a broader range of talent in the pool. Communication and commitment refer to the various capabilities of female directors, which are useable by communicating with the firm’s environment, e.g. investors or suppliers.

Diversity between directors should result in more human and social capital, which are both preconditions for effective monitoring and advising, according to Hillman and Dalziel (2003). The human capital theory suggests that each person has an unique view because each person has different skills, education, reputation and experience (Hillman et al., 2007;Tserjesen, Sealy & Singh, 2009). Social capital refers to the difference between resources available through the different business contacts and network from an individual director. An increase of female representation should positively affect human and social capital, which could in turn have positive effects on board performance (Tserjesen et al., 2009), since female directors have access to different strategic resources, business contacts, and expertise and bring with them along with different skills and reputation, which could be valuable for organizations, as mentioned in the literature review.

3.4 Agency Theory

The agency theory expresses that separation between ownership and the management of listed firms exists (Jensen & Meckling, 1976). This theory suggests that problems could occur since the interests of the principal (owner) and agent (manager) are not completely aligned. According to the agency theory, agency costs8_{could arise as a result of adverse selection and}

8_{“Agency costs include the costs of structuring, monitoring, and bonding a set of contracts among agents with}

conflicting interests. Agency costs also include the value of output lost because the costs of full enforcement of contracts exceed the benefits” (Fama & Jensen, 1983: 33).

(14)

[14]

moral hazard.9_{The board of directors is expected to reduce agency costs by supervising}

managers’ behavior (Fama & Jensen, 1983). The board of directors’ supervising and advising duties may include: representing shareholders, monitoring organizations’ wealth, hiring, firing and compensating managers (Erhardt et al., 2003).

3.5 Hypothesis Development

Financial reporting. Managers are able to use their judgements in financial accounting

practices (Fields, Lys & Vincent, 2001); thus, “earnings management occurs when managers use judgement in financial reporting and in structuring transactions to alter financial reports to either mislead some stakeholders about the underlying economic performance of the company or to influence contractual outcomes that depend on reported accounting numbers” (Healy & Wahlen, 1999). The use of judgement in financial accounting practices could lead to different earnings management patterns, which are outlined by Scott (2012): (1) taking a bath, (2) income minimization, (3) income maximization and (4) income smoothing. If a company has to report a loss during times of stress or reorganizations the management could clear the desks and choose to report a large loss, which is described as “taking a bath” (Scott, 2012). A less extreme, but quite similar manner of reporting loss is income minimization, where the earnings are deliberately adjusted downwards. Income maximization is the opposite and could be used for bonus or debt-covenants purposes. Income smoothing occurs when managers would like to report constant earnings over time.

Using a survey, Dichev, Graham, Harvey and Rajgopal (2013) investigated the motivators for earnings management. They found that managers of listed companies will use earnings management to achieve or beat earnings forecasts or benchmarks, to influence share prices and to comply with debt-covenants (Amstrong, Guay & Weber, 2010). Healy (1985) suggests that earnings management could occur when managers would like to achieve targets to catch bonuses.

Based on the resource-based view, gender diversity should improve the advice and council function because a more heterogeneous composition of non-executives should lead to more conflict and less cohesion by asking different questions and thinking more creatively. Furthermore, as stated before, females and males behave differently since females are less opportunistic and are more likely to comply with accounting regulations. Based on the critical mass theory, a tilted composition should be positively related to earnings quality because the underrepresented gender is able to ally and make use of their knowledge during the group process. For this reason, gender diversity between non-executives should reduce the agency problem concerning the misleading judgements by managers in financial accounting. Based on these considerations, I formulate the following two-part hypothesis:

Hypothesis 1a: Gender diversity between non-executive directors is positively related with

the quality of the earnings.

Hypothesis 1b: The tilted composition of non-executive directors is positively related with

the quality of the earnings.

9_{“Adverse selection is a type of information asymmetry whereby one or more parties to a business transaction, or}

potential transaction, have an information advantage over other partiers” (Scott, 2011: 21). “Moral Hazard is a type of information asymmetry whereby one or more parties to a business transaction, or potential transaction, can observe their actions in fulfillment of the transactions but other parties cannot” (Scott, 2011: 22).

(15)

[15]

Compensation. As mentioned before, conflicting interests between the principal and agent

could arise, which should result in agency costs. A way to avoid these conflicting interests is to establish a remuneration package (Jensen & Meckling, 1976) to prevent shareholders from opportunistic behavior by aligning the interests of CEOs and shareholders (Fama & Jensen, 1983; Gray & Canella, 1997). Compensation packages are complex and heterogeneous and could be based on performance (Adams & Ferreira, 2009) and information from peer groups (Shin, 2013). Remuneration packages could include a fixed, variable- (cash) and equity-based (stock options and shares) component. Variable compensation is related to short-term performance (short-term incentives) and equity compensation to long-term performance (long-term incentives).

Based on the differences in decision making between males and females, gender diversity between non-executives should improve the process of establishing a CEO’s remuneration package (e.g. females have longer investment horizon and keep their accounts for a longer period than males do). As mentioned in the literature review, females hold CEOs more accountable for performance, so the compensation package should include a higher equity-based portion to align the interests of shareholders with those of the CEO. From the critical mass perspective, I expect that the underrepresented gender within a tilted composition is able to improve the interaction process with their skills, as they are not regarded as tokens. With these considerations in mind, I formulate the following hypotheses:

Hypothesis 2a: Gender diversity between non-executive directors is positively related to

equity-based CEO compensation.

Hypothesis 2b: The tilted composition of non-executive directors is positively related to

equity-based CEO compensation.

Hypothesis 3a: Gender diversity between non-executive directors is negatively related to

the total amount of CEO compensation.

Hypothesis 3b: The tilted composition of non-executive directors is negatively related to

the total amount of CEO compensation.

Investments. Opportunistic behavior could arise from the choice of CEOs to make investments

that are not aligned with the interests of the shareholders (Jensen & Meckling, 1976): “Models of moral hazard use this intuition to suggest that managers will invest in negative net present value projects when there is divergence in the principal-agent incentives” (Biddle, Hilary & Verdi, 2009: 114). For example, CEOs should not invest in risky projects if their rewards are based on short-term incentives: “Models of adverse selection suggest that if managers are better informed than investors about a firm’s prospects, they will try to time capital issuances to sell overpriced securities” (Biddle et al., 2009: 114).

From a resource-based view, gender diversity between non-executives should lead to better advising and council by better group performance due to less cohesion, creative thinking and the posing of more varied questions. Furthermore, females and males bring in different information, captured by the firm’s environment since they both have different business contacts. From the critical mass perspective, I expect that the tilted composition of non-executives is able to use their expertise and knowledge as the underrepresented gender is able to ally. I expect that both benefits should reduce the problems of adverse selection and moral hazard due to gender diversity between the non-executives. I formulate the following hypothesis:

(16)

[16]

the quality of investments.

Hypothesis 4b: The tilted composition of non-executive directors is positively related to the

quality of investments

Performance. Monitoring by the non-executives should decrease the costs as a result of

opportunistic behavior, consequently this should improve firm value (Fama & Jensen, 1983). Gender diversity between non-executives should lead to better executive monitoring as it is expected that increasing gender diversity between non-executives is related to increasing independence, which should result in better performance (Van der Walt & Ingly, 2003;Adams & Fereirra, 2009). From a resource-based perspective, gender diversity could “help provide a better link with the organization’s reputation and value” (Ntim, 2013) due to, for example, gains from critical strategic resources, provision of more relevant information and differences in decision quality, as outlined above. Therefore, I formulate the following hypotheses:

short-term performance.

long-term performance.

(17)

[17]

4. Methodology, Sample and Data Collection

4.1 Data

The data contains all companies listed on the AEX, AMX or ASCX on march, 26, 2015, as these companies are expected to have to comply with the legal gender quota from 2016 on, if this quota will be approved by the government (Lückerath-Rovers, 2013; Joecks et al., 2013). For the 75 companies that meet this criteria, I collected data over the years 2003-2014. Financial data was collected by means of Thomson Reuters Eikon, data about non-executives was manually gathered by annual reports, data about CEO characteristics and compensation was manually gathered from the information provided by the Vereniging Effecten Bezitters (VEB).10 This results in 900 unique firm-year observations.

The collected financial data11 of 2003 is only used to calculate delta’s and growth rates, after that, these 75 observations are deleted. Financial companies are deleted due to difference in accounting regulations and standards. Furthermore 6 companies from the sample are excluded due to a lack of relevant data within Thomson Reuters Eikon. 29 observations do not include relevant control variables, so these observations are excluded too. For other adjustments with regard to firm-year observations, I refer to table 1.

Table 1: Sample

Description Firm-Year Observations

Gross Sample 900

Less 2003 (75)

Less financial companies (66)

Less inaccurate data specific companies (66)

Less missing firm-year observations (29)

Net useable sample 664

Deleted firm-years: Financial Reporting

Deleted firm-years: Compensation

Deleted firm-years: Investments

Deleted firm-years: Financial Performance

10_{The VEB maintains the interests of shareholders by juridical actions, visit shareholder meetings and debates}

with regulators, relevant parties and the Dutch government.

11_{Reported amounts are in millions of EUR. Amounts reported in USD are calculated to EUR by using the}

(18)

[18]

4.2 Independent Variables

As mentioned earlier, gender diversity is categorized as variety. The Blau Index (BlauNex) and the Teachman Index (TeachNex) are useful measures regarding calculating heterogeneity between group members on categorical data (Harrison & Klein,2007; Campell, & Minguez-Vera, 2008; Miller & Triana, 2009; Solonas et al., 2012; Joecks et al., 2013). Miller and Triana (2009) point out that the Blau index is an optimal measure of diversity as it fulfills the four criteria for a good measurement: “it has a zero point to represent complete homogeneity, larger numbers indicate greater diversity, the index does not assume negative values and the index is not unbounded” (Miller & Triana, 2009: 766). The Teachman index is largely similar to “the Blau Index although it will always yield a larger number than the Blau index and is more sensitive to small differences in the gender composition of boards since it is a logarithmic measure of diversity” (Campell & Minguez-Vera, 2008: 442). I use the normalized Blau and Techman indexes as researchers are able “to compare the values obtained to their suitable maximum values and thus make proper conclusions for specific conditions” (Solonas et al., 2012: 412) such as group size. The normalized formulas are outlined below, for a more detailed calculation I refer to Appendix C. Additionally, I use %FemNex which is calculated as the female non-executive directors to total non-executive directors.

(1) 𝐵_𝑁= 𝐵 𝐵_𝑚𝑎𝑥 𝑇𝑁= 𝑇 𝑇𝑚𝑎𝑥

With respect to the critical mass, I use four dummy variables developed by Joecks et al. (2013). The first dummy, UniNex represents a uniform board and is equal to 1 if there is no female non-executive director, 0 otherwise. The second dummy, SkewNex represents a skewed board, is equal to 1 if there is at least 1 female non-executive on board but the total representation of female non-executive directors is less than 20%, 0 otherwise. The third dummy, TiltNex represents a tilted board, is equal to 1 if there the total representation by female non-executives is at least 20% but less than 40%, 0 otherwise. The fourth dummy, BalNex represents a balanced board and is equal to 1 if female executives make up at least 40% of the total non-executives, 0 otherwise.

Table 2: Variables and Empirical Proxy’s

Variable Proxy Description Gender Diversity between non-executive directors

BlauNex Blau Index See appendix C.

TeachNex Teachman Index See appendix C.

%FemNex % Female

non-executive directors

# female non-executive directors to # non-executive directors.

UniNex Uniform composition Dummy variable: equal to 1 if there is no female

non-executive director; 0 otherwise.

SkewNex Skewed composition Dummy variable: equal to 1 if there is at least 1 female

non-executive director but the total representation of females is below 20%; 0 otherwise.

(19)

[19]

Table 2: Continued

Variable Proxy Description

TiltNex Tilted composition Dummy variable: equal to 1 if the representation of female

non-executive directors is between 20% and 40%; 0 otherwise.

BalNex Balanced

composition

Dummy variable: equal to 1 if representation of female non-executive directors is at least 40%; 0 otherwise.

Firm Characteristics

Size Company size Logarithm of year-end total assets.

Lev Leverage Book value of total liabilities divided by total assets at the end

of the year.

MB Market-to-book ratio Market value of equity divided by book value of equity at the end of the year.

RevGrowth Growth in revenues Revenues (t) divided by the revenues (t-1).

Cash Operating cash flow Operating cash flow divided by total assets at the end of the

year (t-1)

Industry Industry

classification

Created dummy variables based on the NACE Rev. 2 main Section. 1 = Mining and Quarrying, 2 = Manufacturing, 3 = Construction, 4 = Wholesale and Trade, 5 = Transport and Storage, 6 = Information and Communication, 7 = Financial Services, 8 = Real Estate, 9 = Science and Technic Activities, 10 = Administrative and Support, 11 = Human Health, 12 = Other Activities.

|Da| Discretionary

accruals

Absolute value of the total discretionary accruals as result of eq. 2, 3 and 4.

InvQ Investment quality Dummy variable: 1 if the company falls in the quartiles 1 or 4

(representing over/under investing); 0 if the company falls in the quartiles 2 or 3. Quartiles are based on eq. 5.

Financial Performance

ROA Return on assets Income after taxes divided by the average value of total assets during the year.

ROE Return on equity Net income after taxes divided by the average value of common equity during the year.

Q Tobin’s Q Book value of liabilities plus market value of equity divided

by total assets at 31 December of each year.

Board / CEO characteristics

TotalNex Total non-executive

directors

# non-executives.

One-Two Board structure Dummy variable: 1 if the company uses a one-tier structure; 0

if the company uses a two-tier structure.

CEOTen CEO tenure The number of years since the CEO hold the CEO position.

CEOAge CEO age The number of years since the CEO was born.

CEONat CEO nationality Dummy variable: equal to 1 if the CEO has a Dutch

nationality; 0 otherwise.

CEOGend CEO gender Dummy variable: equal to 1 if the CEO is a male; 0 if the

CEO is a female.

CFOGend CFO gender Dummy variable: equal to 1 if the CFO is a male; 0 if the CFO

is a female.

CEO compensation

FixComp Fixed CEO

compensation

% Fixed compensation of the total CEO compensation. Fixed compensation includes fixed salary and pensions as reported by the VEB.

VarComp Variable CEO

compensation

% Variable compensation of the total CEO compensation. Variable compensation includes bonuses and other cash related fees as reported by the VEB.

(20)

[20]

Table 2: Continued

Variable Proxy Description

EQComp Equity-based

compensation

% Equity-based compensation of the total CEO compensation. Equity compensation includes options and shares as reported by the VEB. The value of options is calculated with the Black Scholes model.12

TotComp Total CEO

compensation

Logarithm of the total CEO compensation.

4.3 Dependent Variables

4.3.1 Financial Reporting

The first dependent variable is the absolute value of discretionary accruals, calculated by the Modified Jones model (Dechow & Sloan, 1995), which is used by Burara, Davidson, Rama and Thiruvadi (2010) to examine the relationship between the presence of a female CFO and earnings quality.13_{“The Modified Jones Model is designed to eliminate the conjectured}

tendency of the Jones Model to measure discretionary accruals with error when discretion is exercised over revenue recognition” (Bartov, Gul & Tsui, 2000: 9). The Modified Jones Model for non-discretionary in the event year for firm x is:

(2) 𝑇𝑎𝑐_𝑡 𝑇𝑎𝑡−1= 𝛼1( 1 𝑇𝑎𝑡−1) + 𝛼2( ∆𝑅𝑒𝑣_𝑡− ∆𝑅𝑒𝑐_𝑡 𝑇𝑎𝑡−1 ) + 𝛼3( 𝑃𝑃𝐸_𝑡 𝑇𝑎𝑡−1) + 𝜀𝑡 Where,

𝑇𝑎𝑐_𝑡 = Total accruals14 in year 𝑡 for firm 𝑥 𝑇𝑎_𝑡 = Total assets for firm 𝑥 in year 𝑡

∆𝑅𝑒𝑣_𝑡 = Change in revenues for firm 𝑥 from year 𝑡 − 1 to 𝑡 ∆𝑅𝑒𝑐_𝑡 = Change in receivables for firm 𝑥 from year 𝑡 − 1 to 𝑡

𝑃𝑃𝐸_𝑡 = Gross value of property plant and equipment for firm 𝑥 in year 𝑡 𝛼1, 𝛼2, 𝛼3 = OLS estimates, representing the firm specific parameters for each year

𝜀𝑡 = Residual item, represent the firm-specific discretionary accruals

Firm-specific parameters for each year (𝛼1, 𝛼2 and 𝛼3) are obtained using the estimation model

as described below. (3) 𝑇𝑎𝑐𝑡 𝑇𝑎_𝑡−1= 𝑘1( 1 𝑇𝑎_𝑡−1) + 𝑘2( ∆𝑅𝑒𝑣𝑡 𝑇𝑎_𝑡−1) + 𝑘3( 𝑃𝑃𝐸𝑡 𝑇𝑎_𝑡−1) + 𝜀𝑡

12_{The values of options and shares are estimated at their issue time, so (un)realized share returns are not included.}

The value of options is calculated by using Black Scholes, where the following is taken into account: the current share price, exercise price, maturity, dividend yield, interest rate and volatility of the share. The VEB has made the following assumptions with regards to the calculation of options values: volatility is set up to 40%, risk free interest rate is set up to 3.7% and the dividend yield is set up to 4%.

13_{Burara et al.}₍₂₀₁₀_{) estimate the absolute value of abnormal discretionary accruals by each industry-year, based}

on two-digit SIC Code. Given the sample size, I estimate the accruals for each specific year.

14_{Total accruals are the difference between net income before extraordinary items minus discounting operations}

(21)

[21] Where,

𝑘₁, 𝑘₂, 𝑘₃ = OLS estimates of 𝛼₁, 𝛼₂, 𝛼₃

𝜀_𝑡 = Residual item

After that, I calculate the discretionary accruals, see equation 4. I use their absolute values. A large (small) value of |Dat| represents more (less) earnings management by discretionary

accruals. (4) 𝜀_𝑡 = 𝐷𝑎_𝑡 = 𝑇𝑎𝑐𝑡 𝑇𝑎_𝑡−1− [𝛼1( 1 𝑇𝑎_𝑡−1) + +𝛼2( ∆𝑅𝑒𝑣_𝑡− ∆𝑅𝑒𝑐_𝑡 𝑇𝑎_𝑡−1 ) + 𝛼3( 𝑃𝑃𝐸_𝑡 𝑇𝑎_𝑡−1)] |𝐷𝑎_𝑡| Where,

𝐷𝑎_𝑡 = Is the error term, total discretionary accruals in year 𝑡 for firm 𝑥 |𝐷𝑎𝑡| = Total absolute value of discretionary accruals in year 𝑡 for firm 𝑥

4.3.2 CEO Compensation

The following two dependent variables are TotComp and EQComp. TotComp is calculated as the logarithm of the total CEO compensation during a year and EQComp is defined as the equity-based part to the total CEO compensation, which is consistent with previous studies (e.g. Adams & Ferreira, 2009;Francis, Hasan & Wu, 2014; Kim et al., 2014; Masulis, Wang & Xie, 2014). For a complete view, I also include VarComp and FixComp in this study, which are the variable and fixed parts of the total CEO compensation.

4.3.3 Investments

Consistent with previous studies (e.g. Chen, Hope, Li & Wang, 2010; Kim, Mauldin & Patro, 2014), I use the model of Biddle et al. (2009) to measure the quality of investments. This model predicts investment as function of sales growth and is described in equation 5. This model estimates the over- or underinvestment using the residuals for each company for each year. 15 After calculating the residuals, I divide the companies for each year in to quartiles, conform the study by Kim et al. (2014). Quartiles 1 and 4 representing over- or underinvestment, quartiles 2 and 3 representing the normal investment level. I create a dummy variable (InvQ) which is equal to 1 if the firm falls into quartile 1 or 4 and is equal to 0 if the company falls into 2 or 3 given the year of interest.

(5) 𝐼𝑛𝑣_{𝑥,(𝑡+1)} = 𝛼₀+ 𝛼₁𝑆𝑎𝑙𝐺𝑟𝑜𝑤𝑡ℎ + 𝜀_(𝑡+1)

15_{Kim et al.}₍₂₀₁₄_{) estimate the residuals for each industry-year based on the 48-industry classification of Fama}

& French, given the size of the total sample in this study, residuals are estimated for each year of all companies included in the total sample.

(22)

[22] Where, 𝐼𝑛𝑣_{𝑥,(𝑡+1)} = (𝑅&𝐷 + 𝐶𝐴𝑃𝐸𝑋 + 𝐴𝐶𝑄 − 𝑆𝑎𝑙𝑒 𝑃𝑃𝐸)_𝑡+1/𝑇𝑎_𝑡] × 100 And, 𝑆𝑎𝑙𝐺𝑟𝑜𝑤𝑡ℎ_𝑡 = 𝑆𝑎𝑙𝑒𝑠_𝑡/𝑆𝑎𝑙𝑒𝑠_𝑡−1 And,

𝐼𝑛𝑣𝑥,(𝑡+1) = Investments for firm 𝑥 in year 𝑡 + 1

𝑇𝑎_𝑡 = Total assets 𝑥

𝑅&𝐷 = R&D expenditure for firm 𝑥 𝐶𝐴𝑃𝐸𝑋 = Capital expenditure for firm 𝑥

𝑃𝑃𝐸 = Plant, Property and Equipment for firm 𝑥 𝐴𝐶𝑄 = Acquisition expenditure for firm 𝑥

4.3.4 Financial Performance

I measure long term financial performance by Q which is Tobin’s Q, this is consistent with previous studies (e.g. Rose, 2007; Adams & Ferreira, 2009). The Tobin’s Q represents the capability of a company to produce shareholder wealth (Rose, 2007). If the Tobin’s Q exceeds 1, it reflects strong growth opportunities (Himmelberg, Hubbard & Darius, 1999). I measure the short term performance by an accounting-based measure (ROE) which reflect the success of a firm given the year of interest, which is consistent with earlier studies (e.g. Haslam et al., 2010; Lückerath-Rovers2013).

4.4 Control Variables

I include diverse control variables. I do not include every control variable in each regression model (see the section: empirical results). The following control variables are settled. Size which is a proxy for the size of a company and is measured by the logarithm of total assets (e.g. Rose, 2007; Adams & Fereirra, 2009; Lückerath-Rovers, 2013). Leverage is used as proxy for the financial structure of the firm and is calculated by dividing the total liabilities to the total value of assets. MB is calculated as the market value of equity to the book value of equity and represents the firms growth opportunities (Biddle et al., 2009;Burara et al., 2010; Srinidi et al., 2011)RevGrowth contains the change in revenues during a year. Cash contains the value of the

operating cash flow to the beginning value of assets (Burara et al., 2010; Srinidi et al., 2011). For controlling the total non-executive positions, I use TotNex (# non-executives) (Adams & Fereirra, 2009; Lückerath-Rovers, 2013). CEOGend and CFOGend both are dummies. These variables are equal to 1 the position is hold by a male, 0 the position is hold by a female CEO or CFO. For other CEO characteristics and industry variables, I refer to table 2.

(23)

[23]

5. Empirical Results

5.1 Descriptive Statistics

5.1.1 The facts of gender diversity between non-executive directors in the Netherlands

Table 316 shows the means and standard deviations among the study variables for the total sample and for firms with (without) female non-executive directors. The means of BlauNex and

TeachNex are 0.3174 and 0.3495 respectively (1 represents a full heterogeneous board, 0 a full

homogeneous). The mean of %FemNex over the total sample is about 0.1055, which is far below the gender target of 30%. The uniform composition of non-executives on Dutch boards is the most common type with an average of 0.53 followed by a tilted composition (0.31). The means of the skewed composition and balanced composition are 0.14 and 0.02 respectively. In figure 1, I give an overview of the growth in %FemNex. Since 2010, there was a strong growth of %FemNex. In 2004, the participation of female executive directors to total non-executive directors was about 5% on average. This percentage increased up to 10% in 2009 and to 22% in 2014, which is consistent with the indicated numbers by the Dutch Female Board Index (2014). Not reported in table 2, 22% (11/49) of the included companies do not have a female non-executive director on board. Furthermore there are 12 companies17_{that do comply}

with the normative gender target of 30% in 2014.

In figure 2, I present the means of %FemNex across different industries. On average, female non-executive directors hold 15% of the total non-executive positions in industries 5 (Transport and Storage) and 6 (Information and Communication) during 2004-2014. Furthermore, female non-executive directors are underrepresented in industries 3 (Construction), 8 (Real Estate) and 9 (Science and Technic Activities) which is consistent with the findings of previous studies in the U.K. (e.g. Brammer et al., 2007; 2009).

The means of CEOGend and CFOGend are 0.97 and 0.96, respectively. This implicates that females hold, on average, the CEO and CFO positions for 3% and 4%. With respect to the overall executive-positions, these findings suggests that the representation of females on executive-positions is far below the normative target of 30%, which is consistent with the findings by the Dutch Female Board Index (2014).

5.1.2 Differences between companies with and without female non-executive directors

In table 3, I compare the means of different firm characteristics across the firm-year observations in which firms have at least one female non-executive (N = 310) on board or do not have a female director on board (N = 354). As reported in table 3, firms with female non-executive directors are larger, more leveraged and perform better in terms of Tobin’s Q. With respect to board and CEO characteristics, firms with female non-executive directors have more non-executive positions, a lower CEO tenure and more female executive positions in terms of the CEO and CFO. Furthermore, the compensation of the CEO is higher and more (less) equity (fixed) based at firms with female directors. These comparisons suggests that the choice to nominate female non-executives could be influenced by firm or board characteristics. Hence, it is essential to include firm characteristics as control variables, as I do.

16_{Pearson correlations are presented in Appendix B, there is no multicollinearity issue because VIF is below 10.} 17_{7 companies listed at the AEX, 3 companies listed at the AMX and 2 companies listed at the ASCX do comply}

(24)

[24]

Table 3: Descriptive statistics

Total Sample Firms with at least one female non-executive

director

Firms without female non-executive directors

N Mean St. Dev N Mean St. Dev N Mean St. Dev Difference in

means (t-value) Gender diversity between non-executive directors

BlauNex 664 0.3174 0.3595 310 0.6799 0.1731 354 0 0 TeachNex 664 0.3495 0.3867 310 0.7486 0.1452 354 0 0 %FemNex 664 0.1055 0.1275 310 0.2262 0.0865 354 0 0 UniNex 664 0.53 0.499 310 0 0 354 1 0 SkewNex 664 0.14 0.347 310 0.30 0.459 354 0 0 TiltNex 664 0.31 0.461 310 0.65 0.476 354 0 0 BalNex 664 0.02 0.144 310 0.05 0.208 354 0 0 Firm Characteristics Size 664 3.1288 0.9044 310 3.5579 0.8783 354 2.7531 0.7468 0.804 (12.759)*** Lev 664 0.5646 0.1772 310 0.5928 0.1873 354 0.5398 0.1640 0.052 (3.881)*** MB 606 2.0737 1.4278 286 2.0631 1.4563 320 2.0833 1.4041 -0.020 (-0.174) RevGrowth 606 1.3830 3.3366 286 1.4876 4.2553 320 1.2896 2.2175 0.197 (0.729) Cash 606 0.0943 0.0922 286 0.0923 0.0858 320 0.0961 0.0976 -0.003 (-0.513) |Da| 606 0.0696 0.0038 286 0.0631 0.0830 320 0.0753 0.1041 -0.012 (-1.588)* InvQ 581 0.49 0.500 250 0.44 0.498 331 0.53 0.500 -0.088 (-2.098)** Financial Performance ROA 664 0.0492 0.0831 310 0.0500 0.0859 354 0.0485 0.0807 0.01 (-0.229) ROE 664 0.1121 0.1851 310 0.1088 0.1992 354 0.1149 0.1720 -0.01 (-0.422) Q 664 1.4717 0.6956 310 1.5155 0.7389 354 1.4334 0.6539 0.022 (2.517)***

Board / CEO Characteristics

TotalNex 664 5.74 2.403 310 6.95 2.637 354 4.68 1.538 2.262 (13.257)*** One-Two 664 0.14 0.349 310 0.19 0.396 354 0.10 0.295 0.098 (3.558)*** CEOTen 664 6.5327 5.1247 310 5.5620 4.4711 354 7.3828 5.5023 -1.820 (-4.701)*** CEOAge 664 53.67 6.547 310 53.87 5.702 354 53.50 7.209 0.377 (0.751) CEONat 664 0.79 0.407 310 0.68 0.467 354 0.89 0.317 -0.206 (-6.567)*** CEOGend 664 0.97 0.175 310 0.95 0.222 354 0.99 0.118 -0.037 (-2.665)*** CFOGend 664 0.96 0.198 310 0.94 0.240 354 0.98 0.149 -0.039 (-2.453)***

Gender Diversity in Dutch Boardrooms: Impact on the Monitoring and Advising Duties

Gender Diversity in Dutch Boardrooms:

Impact on the Monitoring and

Gender Diversity in Dutch Boardrooms:

Impact on the Monitoring and

Advising Duties

Table of Contents

1. Introduction

2. Literature Review

2.1 Corporate Governance in the Netherlands

2.2 The Different Cases and Gender Diversity

2.3 Monitoring Duties and Gender Diversity

2.4 Advising Duties and Gender Diversity

2.5 Legal Requirements and Gender Diversity

2.6 Financial Performance and Gender Diversity

3. Theoretical Framework and Hypotheses

Decision-Making Behavior

Critical Mass Theory

3.3 Resource Dependence-Based View

3.4 Agency Theory

3.5 Hypothesis Development

4. Methodology, Sample and Data Collection

4.1 Data

4.2 Independent Variables

4.3 Dependent Variables

4.4 Control Variables

5. Empirical Results

5.1 Descriptive Statistics