Gender diversity on boards of directors and corporate financial performance

(1)

1

Gender diversity on boards of directors and corporate financial performance

Empirical evidence from the Far East and Central Asia

By Lili Zhu: s2906627 Supervisor: Dr. Boudewijn de Bruin

Assessor: Dr. Halit Gonenc

Date: 9th_{June 2017}

University of Groningen Faculty of Economics and Business International Financial Management

Abstract

This paper investigates the relationship between board gender diversity and firm financial performance in the Far East and Central Asia between 2010 and 2015 using unbalanced panel data from a sample of companies in the target region. Ordinary least squares and fixed effects regression methods are applied. The overall results suggest that board gender diversity positively affects firm financial performance. Besides, shareholder governance is found to strengthen the positive relationship between board gender diversity and firm financial performance. This study suggests that companies as well as policy makers should consider more female membership on corporate boards, as greater diversity on the board may bring economic gains. Furthermore, a high-quality shareholder governance system can help to enhance the benefits of board gender diversity.

(2)

2

1. Introduction

Diverse boards are more likely to engage in initiating competitive and creative practices than boards composed of similar members, while variation in the board may cause a slower strategic decision execution process due to diverse perspectives (Hambrick, Cho,

and Chen, 1996). Jackson (1992) defines two groups of attributes that can explain

different behaviors and perspectives on boards. The first one is called "personal attributes" that are demographic characteristics of group members, such as gender, race, and personality. The second one is "task-related attributes" that designate skills and capabilities of a team member for performing a particular task. For example, gender, race and many other personal characteristics are influencing people's behaviors. Thereby, these characteristics can be managed and balanced by companies to increase board effectiveness, improve problem-solving, and accelerate the strategic decision-making process (Heidrick & Struggles, 2016).

Internationally, European Union countries are the leaders in having females’ presence on corporate boards. For example, Norway (46.7%), France (34.0%), and Sweden (33.6%) have the highest proportions of women on their boards. With a rapid development of Asian economy in the global market, Asian companies are seeking more diversified boards, top management teams, and executive committees. The regulation change nowadays encourages more diversification. Asian countries such as Japan, Malaysia, Singapore and others are pushing companies to achieve more diversified boardrooms. Meanwhile, these countries start applying corporate governance codes (Heidrick & Struggles, 2016). Nevertheless, Asian countries still fall behind Western ones in gender diversification on their boards, for example, Taiwan (4.5%), South Korea (4.1%), and Japan (3.5%) score the lowest of female participation in their boards (Catalyst, 2017).

(3)

3 example, in Belgium, in publicly traded and State-Owned Enterprises, the gender quota is set at 33%, which means at least 33% of board directors should be female. Moreover, in India, it is required that public companies should have at least one female director on their boards (Catalyst, 2017).

The purpose of this study is to investigate the relationship between gender diversity in the boardroom and firm financial performance empirically. In this research, a theoretical framework is applied to help illustrate the possible relationship between board gender diversity and firm financial performance. This framework is based on four key theories: agency theory, resource dependence theory, human capital theory, and collective intelligence. First, agency theory supports that women’s participation in the boardroom increases the board’s independence and monitoring capabilities (Campbell

and Bohdanowicz, 2015). Resource dependence and human capital theories are

complementary. The former suggests that boards need abundant human resources to deal with uncertainty in the business environment, while the latter emphasizes an individual's resources such as knowledge, education, experience and reputation, and they are necessary capital to boards (Singh, Terjesen, and Vinnicombe, 2008). From a social psychological perspective, the female participation’s positive impact on collective intelligence is proven in current psychological research findings (Woolley

and Malone, 2011). Apart from this, social problems in the workplace, such as gender

(4)

4 two propositions (see Appendix 1).

This study is performed using an unbalanced panel sample that includes 307 publicly traded companies in the Fast East and Central Asia and 1,675 observations from 2010 to 2015. Sample countries are China, Hong Kong, India, Indonesia, Japan, Malaysia, Philippines, Republic of Korea, Singapore, Sri Lanka, Taiwan, and Thailand. To avoid biased results because of endogeneity problem, fixed effects methods are used to cope with unobserved heterogeneity and reverse causality, and board gender diversity variables are lagged by one year to address the possibility that firm financial performance could also affect the choice of board composition.

The final results of this paper suggest that board gender diversity does improve the firm financial performance. Moreover, the positive effect of board gender diversity on the firm financial performance are proven to be strengthened under an environment with a strong shareholder governance.

Most of the empirical researches regarding gender diversity are conducted with companies from US, UK, and the European Union. However, there is little empirical evidence for Asian context. From a practical perspective, this study contributes by adding empirical evidence about board gender diversity effect on firm financial performance in the Far East and Central Asia context. On the academic side, this research extends board gender diversity literature by providing more insight into a positive moderating effect of shareholder governance on gender diversity on the board of directors-firm financial performance relationship.

(5)

5

2. Literature review

This section starts with an introduction of corporate governance. Then four main streams of theories are used to help predict the effect of the board of directors on corporate value. Based on the theories and a review of the previous studies concerned with board gender diversity and firm financial performance, the hypotheses are developed subsequently.

2.1. Corporate governance

Corporate governance refers to the framework of rules and regulations, in which board of directors is responsible for the governance (Cadbury Committee, 1992). From an agency perspective, in a corporate governance system, a board functions as an internal mechanism to solve the principal-agent conflicts. Shareholders are "principals" and executives as selected as so-called "agents" (Carter, Simkins, and Simpson, 2003). At the firm level, resource dependence theory emphasizes the important role of board directors within the resource exchange process between a company and outside units

(Terjesen, Sealy, and Singh, 2009). Furthermore, at the individual level, the human

capital theory describes the individual resources needed from board directors for leadership. At the board level, the research on collective intelligence reveals the benefits of board diversity on team performance. Besides, current social problems are also important to take into considerations, including gender stereotype and “Tokenism”.

2.2. Agency theory

(6)

6 one is monitoring. If principals and agents have a conflict of interests or preferences, then it will be difficult for shareholders to understand the actions of executives

(Eisenhardt, 1989; Finkelstein and Hambrick, 1996; Glorie, 2012). Under an effective

monitoring, the board can help align the interests of both parts; therefore firm financial performance can be improved because of decreased agency cost (Hillman and Dalziel,

2003). Several scholars find that board independence is a critical factor for boards to

function effectively, and they suggest that a company board needs to have a sufficient balance between outside and inside directors. Usually, outside directors own more independence and stronger monitoring capabilities than inside directors (Bøhren and

Staubo, 2016; Carter, Simkins, and Simpson, 2003).

2.3. Resource dependence theory

Although resource dependence theory is used less frequently than agency theory, the previous empirical evidence argues that resource dependence theory is superior to agency theory (Hillman, Withers, and Collins, 2009). Different from agency theory, resource dependence theory asserts that board resources are essential capital to firms. On the one hand, board capital can be human capital, such as board members' expertise, experience, and reputation. On the other hand, relational capital is another essential board capital, which creates a linkage between a company and other outside firms or organizations. Based on this dependency relationship, companies have access to more resources, and they seek linkages to exchange and acquire beneficial resources

(Hillman and Dalziel, 2003; Terjesen, Sealy, and Singh, 2009). In the corporate

(7)

7

2.4. Human capital theory

As mentioned above, human capital is one of the board capital. Human capital theory

by Becker (1964) states that everyone has their human capital, including all the personal

attributes such as skills, knowledge, and education. Moreover, firms benefit from its stock of various human capital. On the corporate boards, directors provide human capital to the board with their special experience and knowledge (Kesner, 1988).

Based on resource dependence and human capital theories, it can be deduced that board gender diversity can improve firm performance if the appointment of female board directors could offer more human and social capital to the board, for example, new perspectives, experience, knowledge, etc.

2.5. Collective intelligence

(8)

8 review the literature about different genders' impact on the team behavior and performance, and it is concluded that team collaboration improves with a higher proportion of women within the team. Collective intelligence exists in group patterns of behavior that matter in completing tasks, therefore more female participation in groups could contribute to a higher level of collective intelligence. It will be beneficial for strategic planning, execution and decision-making process of the companies.

2.6. Gender stereotype and Tokenism

(9)

9 family, and higher education for women are an attempt to increase their value beyond wives (Chung, 1994). Even in some island Asian countries, from the perspective of parents, daughters are educated only to meet the requirement of marriage(Chinn, 2002). For example, in Korea, there are still barriers for educated women in the workplace, because women are still supposed to be considered as obedient and sacrificing roles such as wives, daughters-in-law, or mothers (Cho, 1997; Johnsrud, 1995). In China, barriers and pressures like gender inequality and social perception differences still exist for women in management roles in the workplace. It is shown that women are discouraged in the pursuit of management careers because of social perceptions of women (Bowen, Wu, and Hwang, 2007; Cooke, 2005; Woodhams, Xian, and Lupton,

2015).

The gender stereotype's impact on employment could be indicated somehow through a decorative presence of females in companies that are predominated by male workforce

(Davison and Burke, 2000). The concept “Tokenism” popularized by Kanter (1977)

describes the situation that one get hired for an occupation because of differences from others, but not qualifications. Usually, such people are women or minorities. In this situation, they are chosen by organizations because of legal reasons, but the organizations are still skewed to males. Female workers are appointed, as they are put in women’s role first, and individuals later. Therefore, their capabilities cannot be fully capitalized, and the voices of females cannot be taken as seriously as males by the management (Terjesen, Sealy, and Singh, 2009). Torchia, Calabrò, and Huse (2011)

point out that many corporate boards appoint only one or a minority of female directors. However, the token status of women could be improved if the proportion of female increases (Zimmer, 1988). Kramer, Konrad, and Erkut (2008) argue that the benefits of women directors cannot be reaped if there is only one woman on the board of directors.

Erkut, Kramer, and Konrad (2008) further find that gender inequality issue in the

(10)

10

3. Hypothesis development

3.1. Board gender diversity and firm financial performance

Theoretically, gender diverse boards are beneficial to firm performance. First, from an agency theory perspective, female representation on the board can improve decision-making as the independence of the board increases with a higher proportion of females, because female directors are usually outsiders that do not belong to "old boy" network. Compared with males, female directors are more likely to ask questions that male directors would not ask (Adams and Ferreira, 2009; Campbell and Bohdanowicz, 2015;

Carter, Simkins, and Simpson, 2003). Second, based on a resource dependence rationale,

historically, females have less experience as directors and less exposure to the business environment than males. Now it is possible for women board members bringing fresh perspectives and experience to the board (Adams and Ferreira, 2009; Adams, Hermalin, and Weisbach, 2010; Carter, Simkins, and Simpson, 2003; Farell and Herch, 2005;

Fields and Keys, 2003). It is also found that compared with males, it is easier for

females to link firms with different stakeholders, such as customers and suppliers

(Carter, Simkins, and Simpson, 2003; Hillman, Shropshire, and Cannella, 2007; Randøy,

Thomsen, and Oxelheim, 2006). For example, some companies hire female to their

board because female directors can maintain good relations with female consumers (Liu,

Wei, and Xie, 2014). Therefore, gender diversity in board members may enhance the

understanding of the marketplace and encourage innovations and activism. Also, board members of different genders can generate wider variations in values, broader perspectives, which can ultimately increase the probability of a more exhaustive and efficient decision-making. In contrast, less or non-diverse board has a narrower view of the market (Carter, Simkins, and Simpson, 2003).

(11)

11

Smith, and Verner (2006) examine a sample of 2,500 Danish listed companies and find

that the female representation on board directors has a positive relationship with firm performance, and especially they mention that the level of the positive effect of women in top management is dependent on the qualifications of senior female managers. In the research of Carter, Simkins, and Simpson (2003) based on 797 Fortune 1,000 firms, they find out a significantly positive relationship between board gender diversity (percentage of women on boards of directors), and firm value (Tobin’s Q). In a Spanish context, Campbell and Mínguez-Vera (2008) also find that board gender diversity (proportion of female directors on the board) has a positive impact on the firm value measured by Tobin's Q, with a panel data analysis. In the study of Malaysian publicly listed companies between 2008 and 2009, Julizaerma and Sori (2012) find a positive relationship between board gender diversity (percentage of women directors) and return on assets (ROA) of the firms, by using ordinary least square regression method. Liu,

Wei, and Xie (2014) find similar results from Chinese publicly listed companies during

1999 to 2011, with the firm's performance measured by return on assets (ROA) and return on sales (ROS). Using a sample of companies from Hong Kong, Malaysia, Singapore, and South Korea, Low, Roberts, and Whiting (2015) report a significant positive effect of board gender diversity (percentage of female directors) on return on equity (ROE), which is another accounting-based performance indicator of a company. Besides, there is some indirect evidence that may support a positive relationship. In the financial services industry analysis by Kumar (2010), female analysts have better-than-average capabilities than male analysts. Also, Rodríguez-Domínguez, García-Sánchez,

and Gallego-Álvarez (2012) find that when the academic background and working

conditions of both genders are the same, women outperform men in the sectors which are traditionally dominated by male.

(12)

12 companies, and the results show there is no relationship between female representation on the board and firm performance. After studying Spanish listed companies on the Madrid Stock Exchange, Gallego-Álvarez, García-Sánchez, and

Rodríguez-Dominguez (2010) find no relationship between the level of board diversity and

corporate performance from both market and accounting measures. Using a sample of 255 Swedish listed companies in six financial years, Alm and Winberg (2016) find similar results for the relationship between female directors on the board and firm performance (Tobin’s Q and ROA). Based on the theories as mentioned earlier and existing empirical evidence, the first hypothesis is formulated:

H1: There is a positive relationship between board gender diversity and firm financial performance.

3.2. The moderating role of shareholder governance

Based on agency theory, shareholders are regarded as the “principal” in a corporate governance mechanism. Shareholders have the right to vote for appointing and removing of board directors as boards of directors are responsible for the shareholders’ interest. Thus, shareholders can participate in the decision-making process in a corporate governance system. Apart from this, shareholders also can be updated with corporate information on a regular basis; therefore, shareholders can participate in monitoring the company (Cadbury Committee, 1992; Nam and Nam, 2004; Shleifer

and Vishny, 1997). Shareholder governance is the framework in which power shifts

between a firm’s management, its board of directors, and its shareholders (Popadak,

2013). Adams and Ferreira (2009) suggest that the effect of board gender diversity on

(13)

13 controlled. Under conditions of a high shareholder governance, shareholders will have more power in the decision-making on the board of directors. Conversely, under conditions of limited shareholder rights, shareholders (including minority shareholders) are less influential in the appointment and removal of board directors. Therefore, board gender composition is more likely to be decided by “old boy network” in boards, and it can be expected that gender diversity benefits will be harmed. So women’s voices will be much less heard under the circumstance of “Tokenism” (see section 2.6.).

Therefore, the second hypothesis is based on the assumption that under a situation with a high quality of shareholder governance, shareholders' rights and assets are safeguarded; thus, a more effective and competitive corporate boards may be created, and ultimately realize the economic benefits of board gender diversity fully. The second hypothesis is as follows:

H2: The positive effect of board gender diversity on firm financial performance will be strengthened with a greater extent of shareholder governance.

4. Methodology

4.1. Data collection

In this research, unbalanced panel data are generated from the East and Central Asian companies. These companies are all publicly listed and from China, Hong Kong, India, Indonesia, Japan, Malaysia, Philippines, Republic of Korea, Singapore, Sri Lanka, Taiwan, and Thailand. Listed firms are chosen because of more available published financial information, such as Tobin's Q, which is a ratio between market capitalization and total assets of a company. The observations are based on six fiscal years that from 2010 to 2015. Firm characteristics data are collected from the Bureau van Dijk’s Orbis Database. Financial information is retrieved for measuring the firm financial performance. Board composition information is gathered from Thomson Reuters Datastream Economic Database.

(14)

14 to avoid deleting too many samples, firms that have at least three-year data available are retained. As the sample includes different Asian countries, U.S. $ is used as the main currency. Furthermore, companies from financial and insurance sector are excluded because these financial services companies are equipped with distinct accounting regulations, and also these companies have different operations compared with non-financial and insurance companies (Biais and Pagano, 2002; Damodaran, 2009). Then setting the restriction of the year of incorporation, the initial sample includes 4,999 listed companies. After excluding firms without ISIN number, the sample becomes 4,922 firms, after eliminating firms that have more than half of the financial year value missing, the final sample size is 307 companies and contains 1,675 observations. There is another important thing to notice about the different number of observations of ROA, ROE, and Tobin’s Q, The filter process of three measurements of firm financial performance are different. In this paper, ROA is the main dependent variable while ROE and Tobin’s Q are for the robustness check for the results, most of descriptive and correlation information will be among ROA, other independent variables, control variables, and a country-level moderator variable.

4.2. Variable definitions

4.2.1. Dependent variables

(15)

15 of the firms. However, the drawback of using Tobin's Q to measure the benefits of the diversification is due to its assumption that the financial market could capitalize the benefits; otherwise, this measure is not that suitable (Lang and Stulz, 1994). In this paper, Tobin’s Q is not appropriate for measuring the market-based performance of the sample firms because of the underdeveloped nature of Asian capital market. For example, the Malaysian capital market is more volatile than in the U.S. (Johl, Kaur, and

Cooper, 2015). Many Asian companies, for examples many Chinese companies, are

State Owned Enterprises (SOEs) and not traded in the second market. The government, as the non-tradable shareholders, normally purchase their shares at a different price compared to the price determined by initial public offering. So this may cause a gap in the market price of companies’ tradable and non-tradable shares (Haveman and Wang,

2013). Therefore, Tobin’s Q cannot perfectly reflect the market performance of Asian

companies.

Another measuring method is accounting-based indicators. There are three reasons for accounting-based measurement. The accounting-based method is for measuring the firms’ accounting returns, which is “backward-looking”, therefore investigating the effect of board gender diversity on the previous operational performance of the company. While accounting-based measurement is criticized for interference from accounting convention as well as tax laws (Wernerfelt and Montgomery, 1988). ROA, as the most popular methods in the past literature, are used as the main dependent financial performance variable. ROE and Tobin's Q are used for testing the robustness of the results. ROA is computed by dividing income by total assets, and it shows how efficiently a company by using its assets, without any change in the firm’s financing policy. ROE is defined as net income divided by shareholders' equity, and it is appropriate to measure for investors as it gives an insight of how expeditiously the company utilizes shareholders' equity (Chen, Cheng, and Hwang, 2005). Both ROA and ROE are based on self-reported data from the financial information of a firm (Pletzer

(16)

16

4.2.2. Independent variables

The independent variable in this study is board gender diversity. To measure the gender diversity on the board of directors, this study adopts “Pfemale” and “Dfemale” as two proxies, “Pfemale” is the percentage of female members of all the board of directors of the firm. As mentioned in the literature review section, to eliminate the possible "token" issue of female directors, firms with a single or zero female director are also considered as extreme cases; therefore a dummy variable of gender diversity is added as “Dfemale”. It takes two values: 1 if there are two or more female directors on the board, and 0 otherwise.

4.2.3. Control variables

To control other variables that may affect firm financial performance, several other variables are also added. On the one hand, firm size, firm age, and leverage, as firm characteristic factors, are argued to influence the financial performance. From the resource-based theory, a large firm scale could get more resources for competitive advantages (Barney, 1991). In case that firm size has no linear relationship with firm financial performance variables, in the regression models, firm size is measured by the natural logarithm of total assets. Firm age is also predicted to be related to firm financial performance. On the one hand, aged companies gain more market experience and learning capabilities (Stinchcombe and March, 1965). On the contrary, increasing firm age might damage firms' substantial value. First, it brings higher costs, obsolete resources, and rigid systems, decreasing growth rate, and less investment in R&D department. Second, older companies are more rent-seeking than younger ones because of their larger board size, a larger payout of executives, and ultimately poorer governance (Loderer and Waelchli, 2010). Similarly, it is expected that firm age might

(17)

17 make managers strive for the interest of shareholders, therefore, is beneficial for increasing firm performance (Berger and Di Patti, 2006). However, several empirical pieces of research show a negative effect of leverage on firm performance (Gleason,

Mathur, and Mathur, 2000; El-Sayed Ebaid, 2009; Simerly and Li, 2000).

Meanwhile, board characteristics factors, such as board size, and board meetings are also predicted to be related to firm performance. Board size is measured by the natural logarithm of the total number of members on the board of directors. Board size may be detrimental to firm financial performance because a larger board size demand causes coordination and communication problems (Eisenberg, Sundgren, and Wells, 1998;

Yermack, 1996). However, Van den Berghe and Levrau (2004) argue that board of

directors has an impact on the decision-making process as larger boards have more talented and dynamic ideas, skills, and knowledge, and it is positive for firm performance. Consistent with agency theory, the frequency of board meetings, one proxy of board activity, is associated with corporate governance. It is measured by the number of meetings of the board during the year (Vafeas, 1999). Board meetings can be positive or negative to firm performance, and it depends on the level of board meetings. Board meetings with high quality can advance the decision-making effectiveness, and in this case, a higher frequency of having board meetings is beneficial for firm performance (Van den Berghe and Levrau, 2004). Boards with poor performance will try to improve the quality of board meetings to improve operating performance (Vafeas, 1999).

4.2.4. Moderator variable

(18)

18 proxy for the total corporate governance that is not appropriate for measuring shareholder governance. Second, G-index is a firm-specific measuring method as based on 24 firm-specific provisions (Aebi, Sabato, and Schmid, 2012; Gompers, Ishii, and

Metrick, 2003). Because the scoring of corporate governance system is mainly based

on those internationally recognized corporate codes, such as World Bank, ICGN, OECD

(Van denBerghe and Levrau, 2003). In this paper, shareholder governance is measured

by a direct shareholder governance index available in World Bank database. Moreover, this index varies in different countries. The extent of shareholder governance is between 0 and 10. The higher the shareholder index value is, the better the shareholder governance is (World Bank, 2016).

4.3. Endogeneity

There are two sources of endogeneity: unobserved heterogeneity and reverse causality. Omitted firm and board-specific factors could affect the board member selection and also firm financial performance. If companies engage in the corporate social responsibility actions or follow the legitimacy regulations, more female directors will be hired, and firm performance may also be improved (Sila, Gonzalez, and Hagendorff,

2016). The direction of the relationship between board gender diversity and firm

(19)

19 lagged by one year.

4.4. Models and estimation methods

In order to investigate whether board gender diversity affect firm financial performance (H1), the basic model of firm financial performance as a function of female director variables and other control variables can be written like this in equation 1 and 2:

Performanceit = ai + 𝛽1 Pfemaleit + 𝛽2 Firmsize it + 𝛽3 Firmage it + 𝛽4 Leverageit + 𝛽5

Boardsize it + 𝛽6 Boardmeetingit + λi + it (1)

Performanceit = ai + 𝛽1 Dfemaleit + 𝛽2 Firmsize it + 𝛽3 Firmage it + 𝛽4 Leverageit + 𝛽5

Boardsize it + 𝛽6 Boardmeetingit + λi + it (2)

Where Performance is measured by ROA and two other alternative performance measures: Tobin’s Q and ROE to decrease the likelihood of the coincidence of the results. Board gender diversity is measured by the percentage of women directors on the board (Pfemale), or a dummy female variable of two or more women directors present on the board (Dfemale). Where i presents firm and t presents time. Where the parameter with primary interest is 𝛽1, which determines whether there is a relationship

between board gender diversity and firm financial performance, and furthermore check whether there is a positive or negative relationship between each other. Firmsize, Firmage, and Leverage are three firm-level control variables. Boardsize, as well as Boardmeeting, are two board-level control variables. Where ai is a constant component, and it refers to error term. As mentioned before, the data set in this research is a panel of companies, in able to control the potential unobservable variables bias, firm fixed effects term λi is also added. It is time invariant and varies across companies.

(20)

20 Performanceit = ai + 𝛽1 Pfemaleit *SG + 𝛽2 Firmsize it + 𝛽3 Firmage it + 𝛽4 Leverageit +

𝛽5 Boardsize it + 𝛽6 Boardmeetingit + λi + it (3)

Performanceit = ai + 𝛽1 Dfemaleit *SG + 𝛽2 Firmsize it + 𝛽3 Firmage it + 𝛽4 Leverageit +

𝛽5 Boardsize it + 𝛽6 Boardmeetingit + λi + it (4)

There are two estimation methods used, the first one is ordinary least squares (OLS) regression and the second one is fixed effects regression. However, there is a limitation of using OLS regression. It assumes the relation between variables is constant over time across sample firms. Due to the unobservable firm or board characteristics, the OLS regression method is likely to generate biased results. Appling a fixed effects or random effects model is one solution for the unobservable heterogeneity (Smith, Smith, and

Verner, 2006). A fixed effects model incorporates both time-series and cross-section

data, and it can control individual-specific effects, even possibly unobservable. Because some of the unobservable variables may be correlated with other independent variables in regression relationship, therefore if only using the cross-sectional method, it is difficult to control individual effects (Hausman and Taylor, 1981). In this study, unbalanced panel information is observed on the same sample firms during the same period, and thus it can control unobserved heterogeneity among firms.

5. Empirical results

5.1. Descriptive statistics

(21)

21

Table 1

Descriptive statistics

Variable Mean Min Median Max Std Observations

Panel A: financial performance

ROA 0.067 -0.302 0.051 0.754 0.076 1,842

Panel B: board gender diversity

Pfemale 0.074 0.000 0.071 0.375 0.082 1,779 Dfemale 0.246 0.000 0.000 1.000 0.431 1,779

Panel C: control variables

Firm size (ln) 15.903 12.522 15.966 20.593 1.381 1,842 Firm age (ln) 3.501 1.792 3.526 4.963 0.683 1,842 Leverage 0.337 0.000 0.344 0.955 0.212 1,841 Board size (ln) 2.319 1.099 2.303 3.258 0.307 1,779 Board meeting 9.098 1.000 8.000 57.000 5.035 1,676

This table presents the descriptive statistics of all the main firm and board variables. The sample includes 1,675 observations of 307 companies studied from 2010 to 2015. The definition of related variables can be found in Appendix 2.

directors on boards is 7.4%, which means around 7.4% of all the directors are women. Furthermore, 24.6% of the sample firms have two or more female directors present on the firm's board. Panel C summarizes the control variable data, including the board and firm-specific variables. The average firm size (ln) and firm age (ln) are 15.903 and 3.501. It denotes that on average, the sample firms have $8,064,643 in total assets and an around 33 years’ age. The mean value of total debt to total assets is 33.7%, which indicates the companies have 33.7% debt and 66.3% equity respectively. The average number of board size (ln) is 2.319, indicating approximately ten board members in each of the sample companies’ board. It is noteworthy that on average there are about ten board meetings in one financial year. Board meeting frequency in the sample seems small, and there is a minimum number of one board meeting per year for an Indonesian company. While in many Asian countries, such as Thailand, Malaysia, Korea, and Indonesia, board meeting frequency is determined based on a quarterly report basis; therefore a rate of four board meetings in one financial year is set as a minimum level. Therefore, ten board meetings one year is also easy to understand in this analysis (Nam

(22)

22

5.2. Correlation among variables

A correlation matrix analysis is for checking potential multicollinearity among variables. As a rule of thumb, it shows multicollinearity when the correlation is higher than 0.700. As shown in Table 2, the two measures of board gender diversity, Pfemale and Dfemale shows a strong correlation, which is 0.725 (>0.700). Table 7 and Table 8 in Appendix 4 gives the correlation matrix between Tobin’s Q, ROE, and all the firm and board-related variables used in this study. The results are similar, which show a correlation of 0.723 and 0.725 (>0.700) respectively. While multicollinearity is reasonable in this case because these two measures are used for different hypotheses goals in the regression models and not used simultaneously. Multicollinearity is not a problem here, and there is no need to fix the correlation between Pfemale and Dfemale.

Table 2 Correlation matrix 1 2 3 4 5 6 7 8 1 ROA 1 2 Pfemale 0.110 1 3 Dfemale 0.079 0.725 1 4 Firm size (ln) -0.432 -0.123 -0.004 1 5 Firm age (ln) -0.046 -0.168 -0.135 0.016 1 6 Leverage -0.401 -0.092 -0.024 0.287 -0.044 1 7 Board size (ln) -0.137 0.032 0.171 0.283 -0.001 0.233 1 8 Board meeting -0.217 -0.154 -0.076 0.234 0.128 0.161 0.050 1

This table presents the correlation matrix among Tobin’s Q and all the firm and board variables that are used in this study. The definition of related variables can be found in Appendix 2.

5.3. Regression analysis

(23)

23 financial performance (ROA), with board diversity measured by a dummy female variable. In other words, regression model 2 (see section 4.4.) for hypothesis 1 can be confirmed by taking potential “Token” status of female directors (see section 2.6.). The fact that the coefficient of dummy female variable is significant, but that of percentage female variable is insignificant supports “critical mass” (see section 2.6.) theory (Erkut,

Kramer, and Konrad, 2008). In columns 1 and 2, board gender diversity is measured by

the proportion of females on the board. The coefficient is positive and small, but not statistically significant at any level. In columns 3 and 4, female directors’ presence on the board is measured by a dummy female variable. ROA appears to be significantly positive correlated with board gender diversity, at a significant level of 1% and 5% respectively. Also, it is noteworthy that the positive relationship between the dummy female variable and ROA persist, but the significant level decreased from 1% to 5%, which implies that there is evidence for reverse causality that ROA, as a proxy for operational performance of a company, increases the companies' tendencies to appoint female directors.

For Hypothesis 2, the overall results show that shareholder governance has an insignificant and positive moderating effect on the board gender diversity-firm financial performance (ROA) relationship. As the prerequisite of analyzing the moderating effect of shareholder governance is that there does exist a relationship between board gender diversity and firm financial performance (ROA), it is only necessary to analyze columns 7 and 8, because these two models are related to the association between dummy female variable and ROA. It shows that shareholder governance has an insignificant and positive effect on the positive impact of board gender diversity (dummy female variable) on firm financial performance (ROA).

For both hypothesis 1 and hypothesis 2, the adjusted R2 _{is approximately 28%, and it}

(24)

24

Table 3

OLS and lagged OLS regression: board gender diversity and ROA

H1 H2

OLS Lagged OLS Lagged OLS Lagged OLS Lagged

(1) (2) (3) (4) (5) (6) (7) (8) Pfemale 0.020 0.018 (0.020) (0.022) Dfemale 0.010*** _0.011** (0.004) (0.004) Pfemale*SG -0.004 -0.006 (0.011) (0.012) Dfemale*SG 0.002 0.002 (0.002) (0.002) Firm size (ln) -0.019*** -0.018*** -0.019*** _-0.018*** _-0.019*** _-0.018*** _-0.019*** _-0.018*** (0.001) (0.001) (0.001) (0.001) (0.001) (0.001) (0.001) (0.001) Firm age (ln) -0.004* -0.004 -0.004* _-0.003 _-0.004* _-0.004 _-0.004* _-0.003 (0.002) (0.002) (0.002) (0.002) (0.002) (0.002) (0.002) (0.002) Leverage -0.109*** -0.112*** -0.109*** _-0.111*** _-0.109*** _-0.112*** _-0.109*** _-0.111*** (0.008) (0.008) (0.008) (0.008) (0.008) (0.008) (0.008) (0.008) Board size (ln) 0.008 0.007 0.005 0.004 0.008 0.007 0.005 0.004 (0.006) (0.006) (0.006) (0.006) (0.006) (0.006) (0.006) (0.006) Board meeting -0.001*** -0.001*** -0.001*** _-0.001*** _-0.001*** _-0.002*** _-0.001*** _-0.001*** (0.000) (0.000) (0.000) (0.000) (0.000) (0.000) (0.000) (0.000) R2 _0.281 _0.277 _0.284 _0.280 _0.281 _0.277 _0.284 _0.280 Adjusted R2 _0.279 _0.274 _0.281 _0.277 _0.278 _0.274 _0.281 _0.277 Observations 1,675 1,549 1,675 1,549 1,675 1,549 1,675 1,549 Firms 307 300 307 300 307 300 307 300 Firm fixed effects No No No No No No No No Time fixed effects No No No No No No No No

This table represents the results of OLS regression in model (1), (3), (5), and (7). Also, it reports OLS regression with lagged variables (Pfemalet-1, Dfemalet-1,

Pfemalet-1*SG, Dfemalet-1*SG, and Board size (ln) t-1) in model (2), (4), (6), and (8).

(25)

25 Table 4 shows the results of fixed effects regression and lagged fixed effects regression about the relations between board gender diversity and ROA. This regression method can capture the effect of unobservable firm and board characteristics, thereby eliminating the inconsistent and biased results of traditional OLS regression. Columns 1, 3, 5, and 7 are the baseline models, while columns 2, 4, 6, and 8 are models with one-year lagged board gender diversity variables. Compared with the results of the OLS regression in Table 3, the adjusted R2_{increases significantly from 28% to 83%,}

indicating the fixed effects model can better fit the data than OLS method.

The overall findings in Table 4 show a positive and also statistically significant (P<0.10) relationship between board gender diversity and ROA, with board gender diversity measured by the proportion of females on the boardroom (column 1). Thus, regression model 1 (see section 4.4.) for hypothesis 1 can be confirmed. This finding is consistent with that of Julizaerma and Sori (2012), which is also based on an Asian context, namely Malaysia. This result also aligns with the findings of Erhardt, Webel, and

Shrader (2003) and Adams and Ferreira (2009), which suggest that diverse boards

distribute more effort to the oversight function of the board of directors and thus supports agency theory. If the “Token” status of women directors (see section 2.6.) is taken into account, the positive relationship between dummy female variable and ROA remains but not significant anymore (column 3). It is also found that with one-year lagged board characteristics values, the positive impact of board gender diversity on Tobin’s Q still exists, but is not significant at any significant level (column 2). This finding suggests that “Tokenism” (see section 2.6) is psychologically harmful to females and this issue would undermine women’s influence on the corporate performance (Zimmer, 1988).

(26)

26

Table 4

Fixed effects and lagged fixed effects regression: board gender diversity and ROA

H1 H2

FE Lagged FE Lagged FE Lagged FE Lagged

(1) (2) (3) (4) (5) (6) (7) (8) Pfemale 0.039* _0.034 (0.020) (0.023) Dfemale 0.003 0.004 (0.003) (0.003) Pfemale*SG 0.026* _0.007 (0.015) (0.017) Dfemale*SG 0.007*** _0.005** (0.002) (0.002) Firm size (ln) -0.023*** _-0.018*** _-0.023*** _-0.018*** _-0.023*** _-0.018*** _-0.023*** _-0.019*** (0.004) (0.004) (0.004) (0.004) (0.004) (0.004) (0.004) (0.004) Firm age (ln) -0.081*** _-0.086*** _-0.076*** _-0.084*** _-0.079*** _-0.086*** _-0.075*** _-0.083*** (0.012) (0.012) (0.012) (0.012) (0.012) (0.012) (0.012) (0.012) Leverage -0.143*** _-0.149*** _-0.143*** _-0.150*** _-0.143*** _-0.149*** _-0.143*** _-0.149*** (0.013) (0.014) (0.013) (0.014) (0.013) (0.014) (0.013) (0.014) Board size (ln) 0.009 0.004 0.009 0.004 0.008 0.004 0.008 0.003 (0.007) (0.007) (0.007) (0.007) (0.007) (0.007) (0.007) (0.007) Board meeting 0.000 0.001 0.000 0.001 0.000 0.001 0.000 0.001 (0.000) (0.000) (0.000) (0.000) (0.000) (0.000) (0.000) (0.000) R2 _0.861 _0.867 _0.861 _0.867 _0.861 _0.867 _0.862 _0.867 Adjusted R2 _0.829 _0.834 _0.829 _0.834 _0.829 _0.834 _0.830 _0.834 Observations 1,675 1,549 1,675 1,549 1,675 1,549 1,675 1,549 Firms 307 300 307 300 307 300 307 300

Firm fixed effects Yes Yes Yes Yes Yes Yes Yes Yes

Time fixed effects No No No No No No No No

This table represents the results of fixed effects regression in model (1), (3), (5), and (7). Also, it reports fixed effects regression with lagged variables (Pfemalet-1,

Dfemalet-1, Pfemalet-1*SG, Dfemalet-1*SG, and Board size (ln) t-1) in model (2), (4),

(27)

27

5.4. Robustness check

Apart from ROA, the two alternative measurements of firm financial performance: Tobin’s Q and ROE are used for a robustness check of regression results with ROA. Table 9 and Table 10 in Appendix 5 provide robustness checking of the OLS regression results in Table 3. It is shown that the OLS regression results of ROA are robust to the measurement by ROE, because the findings of Table 3 and Table 10 in Appendix 5 are similar. As shown in Table 10, coefficients in Column 3 and 4 confirm the significant and positive effect of board gender diversity (dummy female variable) on firm financial performance (ROE). Also, shareholder governance again shows a positive, but not statistically significant effect on the positive relationship between board gender diversity (dummy female variable) and firm financial performance (ROE). The findings in Table 9 indicate no matter gender diversity variables are non-lagged or lagged by one year, none of the coefficients of the gender diversity variables is statistically significant. Thus board gender diversity’s positive effect on financial performance (Tobin’s Q) cannot be verified. Consequently, the positive signs of shareholder governance has no economic significance.

(28)

28

6. Conclusions and limitations

How gender diversity on corporate boards impacts firm performance is still a source of concern. The aim of this paper is to shed light on the relationship between gender diversity on the board of directors and firm financial performance in the Far East and Central Asia context, with a sample containing 307 companies and 1,675 observations from 2010 to 2015. Firm financial performance is mainly defined by ROA, and additionally, ROE and Tobin’s Q are two alternative measures for robustness check. Unbalanced panel data are collected, and adopted estimation methods are both OLS and fixed effects models.

(29)

29 Apart from the results of above two hypotheses, firm and board-related factors’ effect on company financial performance is also of interest. Firstly, firm size is negatively associated with ROA, ROE, and Tobin’s Q. Secondly, firm age is negatively related to ROA and ROE, but positively associated with Tobin’s Q. Thirdly, leverage is found to have a significantly negative effect on ROA and Tobin’s Q, but has no relationship with ROE. Fourthly, board size is positively related to Tobin’s Q only. Finally, board meeting is significantly negatively associated with Tobin’s Q and ROE but has no relationship with ROA. To summarize, five findings are concluded for control variables: (1) Firms with bigger size (the natural logarithm of total assets) generate less financial value, including both accounting returns (ROA and ROE) and stock value (Tobin’s Q). (2) Compared with younger firms, aged firms generate higher accounting returns (ROA and ROE) but have less stock value (Tobin’s Q). (3) Firms with lower leverage have higher accounting returns (ROA) and stock value (Tobin’s Q) than those with higher leverage. (4) Firms with larger board size might have higher market value (Tobin’s Q) than those with smaller-sized boards. (5) Firms with boards that meet more frequently might have higher accounting returns (ROE) and stock value (Tobin’s Q).

It is also interesting to see that the significant level of coefficients decreases with lagged firm and board-related characteristics, which implies that there exists a certain degree of reverse causation of firm operational performance (ROA and ROE) leading to the appointment of female directors on corporate boards. In other words, companies hire more women directors in the financial years with a good operational performance, which can attract more female applicants.

(30)

30 greater with the improved shareholder governance. Asian countries should be more conscious of the necessity of a good shareholder governance system for increasing firms’ value. Asian countries should be more engaged in building and developing higher-standard shareholder governance system.

There are several limitations to be mentioned in this study. First, the sample size in this research is relatively small. Compared with 2,500 Danish samples of Smith, Smith, and

Verner (2006), 307 sample size is not entirely representative of the Asian companies.

(31)

31

Appendix

Appendix 1 Conceptual model

+

Appendix 2 Variable definition

Variable Definition and calculation

Board gender diversity (Pfemale) Percentage of females on the board of directors Board gender diversity (Dfemale) Equals to 1 when the board has two or more female

directors, otherwise 0. Firm financial performance

(Tobin’s Q)

Ratio between market capitalization value and total assets

Firm financial performance (ROA) Ratio between operating income and total assets Firm financial performance (ROE) Ratio between operating income and stockholders’

equity

Firm size (ln) Natural logarithm of total assets

Firm age (ln) Natural logarithm of age of the firm since incorporation Leverage Ratio between total debt and total assets

Board size (ln) Natural logarithm of number of members on the board of directors

Board meeting Number of board meetings during the year Shareholder governance (SG) The extent of safeguarding of shareholders by

shareholder governance index

Board gender diversity Firm financial performance Shareholder governance

(32)

32

Appendix 3 Descriptive statistics of robustness variables

Table 5

Descriptive statistics with Tobin’s Q

Tobin's Q 1.348 0.000 0.703 17.947 1.934 1,812

Pfemale 0.074 0.000 0.071 0.375 0.081 1,761 Dfemale 0.244 0.000 0.000 1.000 0.429 1,761

This table presents the descriptive statistics of all the main firm and board variables. The sample includes 1,652 observations of 304 companies studied from 2010 to 2015. The definition of related variables can be found in Appendix 2.

Table 6

Descriptive statistics with ROE

ROE 0.138 -4.279 0.109 4.614 0.267 1,842

Pfemale 0.074 0.000 0.071 0.375 0.082 1,779 Dfemale 0.246 0.000 0.000 1.000 0.431 1,779

(33)

33

Appendix 4 Correlation matrix of robustness variables

Table 7

Correlation matrix with Tobin’s Q

1 2 3 4 5 6 7 8 1 Tobin’s Q 1 2 Pfemale 0.090 1 3 Dfemale 0.029 0.723 1 4 Firm size (ln) -0.527 -0.138 -0.013 1 5 Firm age (ln) 0.076 -0.169 -0.132 0.0228 1 6 Leverage -0.346 -0.087 -0.020 0.2979 -0.055 1 7 Board size (ln) -0.183 0.034 0.172 0.2815 0.006 0.237 1 8 Board meeting -0.222 -0.156 -0.074 0.2310 0.128 0.161 0.048 1

This table presents the correlation matrix among Tobin’s Q and all the firm and board variables that are used in this study. The definition of related variables can be found in Appendix 2.

Table 8

Correlation matrix with ROE

1 2 3 4 5 6 7 8 1 ROE 1 2 Pfemale 0.081 1 3 Dfemale 0.074 0.725 1 4 Firm size (ln) -0.292 -0.123 -0.004 1 5 Firm age (ln) -0.056 -0.168 -0.135 0.016 1 6 Leverage -0.119 -0.092 -0.024 0.287 -0.044 1 7 Board size (ln) -0.110 0.032 0.171 0.283 -0.001 0.233 1 8 Board meeting -0.144 -0.154 -0.076 0.234 0.128 0.161 0.050 1

(34)

34

Appendix 5 Robustness test

Table 9

OLS and lagged OLS regression: board gender diversity and Tobin’s Q

H1 H2

(1) (2) (3) (4) (5) (6) (7) (8) Pfemale 0.308 -0.398 (0.509) (0.557) Dfemale 0.124 0.140 (0.095) (0.104) Pfemale*SG 0.800*** 0.648* (0.272) (0.307) Dfemale*SG 0.137** 0.106* (0.056) (0.062) Firm size (ln) -0.644*** -0.635*** -0.645*** -0.632*** -0.631*** -0.626*** -0.639*** -0.628*** (0.032) (0.034) (0.032) (0.034) (0.032) (0.034) (0.032) (0.034) Firm age (ln) 0.261*** 0.259*** 0.265*** 0.279*** 0.250*** 0.254*** 0.260*** 0.276*** (0.060) (0.063) (0.059) (0.062) (0.060) (0.063) (0.059) (0.062) Leverage -1.776*** -1.777*** -1.770*** -1.745*** -1.760*** -1.757*** -1.761*** -1.727*** (0.203) (0.214) (0.203) (0.213) (0.203) (0.214) (0.203) (0.213) Board size (ln) -0.057 -0.078 -0.084 -0.127 -0.101 -0.107 -0.111 -0.141 (0.138) (0.143) (0.140) (0.145) (0.139) (0.143) (0.140) (0.145) Board meeting -0.038*** -0.042*** -0.038*** -0.041*** -0.034*** -0.039*** -0.036*** -0.039*** (0.008) (0.009) (0.008) (0.009) (0.008) (0.009) (0.008) (0.009) R2 _0.332 _0.322 _0.332 _0.323 _0.335 _0.324 _0.335 _0.324 Adjusted R2 _0.329 _0.319 _0.330 _0.320 _0.332 _0.321 _0.332 _0.321 Observations 1,652 1,528 1,652 1,528 1,652 1,528 1,652 1,528 Firms 304 297 304 297 304 297 304 297

Firm fixed effects No No No No No No No No

(35)

35

Table 10

OLS and lagged OLS regression: board gender diversity and ROE

H1 H2

(1) (2) (3) (4) (5) (6) (7) (8) Pfemale 0.096 0.052 (0.075) (0.083) Dfemale 0.041*** _0.039** (0.014) (0.015) Pfemale*SG 0.024 -0.010 (0.040) (0.045) Dfemale*SG 0.011 0.009 (0.008) (0.009) Firm size (ln) -0.047*** _-0.046*** _-0.048*** _-0.046*** _-0.047*** _-0.046*** _-0.047*** _-0.046 (0.005) (0.005) (0.005) (0.005) (0.005) (0.005) (0.005) (0.005) Firm age (ln) -0.014 -0.016* _-0.013 _-0.013 _-0.015* _-0.016* _-0.013 _-0.014 (0.009) (0.009) (0.009) (0.009) (0.009) (0.009) (0.009) (0.009) Leverage -0.032 -0.048 -0.030 -0.045 -0.032 -0.049 -0.029 -0.043 (0.030) (0.032) (0.030) (0.032) (0.030) (0.032) (0.030) (0.032) Board size (ln) -0.024 -0.028 -0.033 -0.037* _-0.025 _-0.027 _-0.035* _-0.038 (0.021) (0.021) (0.021) (0.022) (0.021) (0.022) (0.021) (0.022) Board meeting -0.003*** _-0.004** _-0.003*** _-0.004*** _-0.003*** _-0.004*** _-0.003*** _-0.004 (0.001) (0.001) (0.001) (0.001) (0.001) (0.001) (0.001) (0.001) R2 _0.096 _0.093 _0.099 _0.097 _0.096 _0.093 _0.100 _0.097 Adjusted R2 _0.092 _0.090 _0.096 _0.093 _0.092 _0.089 _0.096 _0.093 Observations 1,675 1,549 1,675 1,549 1,675 1,549 1,675 1,549 Firms 307 300 307 300 307 300 307 300

Firm fixed effects No No No No No No No No

(36)

36

Table 11

Fixed effects and lagged fixed effects regression: board gender diversity and Tobin’s Q

H1 H2

(1) (2) (3) (4) (5) (6) (7) (8) Pfemale 1.158*** _0.439 (0.364) (0.427) Dfemale 0.100* _0.052 (0.056) (0.062) Pfemale*SG 0.344 -0.087 (0.260) (0.308) Dfemale*SG 0.029 -0.001 (0.038) (0.043) Firm size (ln) -0.433*** _-0.420*** _-0.435*** _-0.423*** _-0.442*** _-0.419*** _-0.437*** _-0.423*** (0.074) (0.078) (0.074) (0.078) (0.074) (0.079) (0.074) (0.079) Firm age (ln) 0.430** _0.367 _0.547*** _0.398* _0.449** _0.361 _0.554*** _0.397* (0.213) (0.224) (0.209) (0.221) (0.214) (0.226) (0.209) (0.222) Leverage -1.900*** _-1.797*** _-1.894*** _-1.806*** _-1.888*** _-1.799*** _-1.891*** _-1.806*** (0.232) (0.249) (0.232) (0.249) (0.232) (0.249) (0.232) (0.249) Board size (ln) 0.279** _0.078 _0.290** _0.073 _0.272** _0.081 _0.285** _0.073 (0.117) (0.121) (0.117) (0.121) (0.117) (0.121) (0.117) (0.121) Board meeting 0.007 0.009 0.007 0.009 0.006 0.009 0.007 0.009 (0.007) (0.007) (0.007) (0.007) (0.007) (0.007) (0.007) (0.007) R2 _0.934 _0.936 _0.934 _0.936 _0.934 _0.936 _0.934 _0.936 Adjusted R2 _0.919 _0.920 _0.918 _0.920 _0.919 _0.920 _0.918 _0.920 Observations 1,652 1,528 1,652 1,528 1,652 1,528 1,652 1,528 Firms 304 297 304 297 304 297 304 297

(37)

37

Table 12

Fixed effects and lagged fixed effects regression: board gender diversity and ROE

H1 H2

(1) (2) (3) (4) (5) (6) (7) (8) Pfemale 0.123 0.028 (0.087) (0.104) Dfemale 0.011 0.004 (0.014) (0.015) Pfemale*SG 0.092 0.001 (0.063) (0.076) Dfemale*SG 0.016* _0.016 (0.009) (0.010) Firm size (ln) -0.061*** _-0.057*** _-0.061*** _-0.057*** _-0.063*** _-0.057*** _-0.063*** _-0.058*** (0.018) (0.019) (0.018) (0.019) (0.018) (0.019) (0.018) (0.019) Firm age (ln) -0.213*** _-0.176*** _-0.200*** _-0.174*** _-0.208*** _-0.176*** _-0.197*** _-0.170*** (0.051) (0.055) (0.050) (0.054) (0.051) (0.055) (0.050) (0.054) Leverage 0.023 0.022 0.023 0.022 0.026 0.022 0.025 0.024 (0.056) (0.061) (0.056) (0.061) (0.056) (0.061) (0.056) (0.061) Board size (ln) 0.002 0.032 0.003 0.031 0.000 0.032 0.000 0.030 (0.028) (0.030) (0.028) (0.030) (0.028) (0.030) (0.028) (0.030) Board meeting 0.001 0.002 0.001 0.002 0.001 0.002 0.001 0.002 (0.002) (0.002) (0.002) (0.002) (0.002) (0.002) (0.002) (0.002) R2 _0.763 _0.769 _0.763 _0.769 _0.764 _0.769 _0.763 _0.770 Adjusted R2 _0.709 _0.713 _0.709 _0.713 _0.709 _0.712 _0.709 _0.713 Observations 1,675 1,549 1,675 1,549 1,675 1,549 1,675 1,549 Firms 307 300 307 300 307 300 307 300

(38)

38

References

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

Adams, R. B., Hermalin, B. E., Weisbach, M. S., 2010. The role of boards of directors in corporate governance: A conceptual framework and survey. Journal of

Economic Literature, 48(1), 58-107.

Aebi, V., Sabato, G., Schmid, M., 2012. Risk management, corporate governance, and bank performance in the financial crisis. Journal of Banking & Finance, 36(12), 3213-3226.

Alm, M., Winberg, J., 2016. How does gender diversity on corporate boards affect the firm financial performance?. Bachelor thesis paper. University of Gothenburg.

Barney, J., 1991. Firm resources and sustained competitive advantage. Journal of Management, 17(1), 99-120.

Bear, J. B., Woolley, A. W., 2011. The role of gender in team collaboration and performance. Interdisciplinary Science Reviews, 36(2), 146-153.

Becker, G. S. Human Capital. New York: Columbia University Press (for NBER), 1964. "A Theory of the Allocation of Time." Economic Journal, 75 (1965), 493-517

Berger, A. N., Di Patti, E. B., 2006. Capital structure and firm performance: a new approach to testing agency theory and an application to the banking industry. Journal of Banking & Finance, 30(4), 1065-1102.

Biais, B., Pagano, M., 2002. New research in corporate finance and banking. Oxford University Press on Demand.

Black, B. S., De Carvalho, A. G., Khanna, V. S., Kim, W., Yurtoglu, B. B., 2016. Corporate Governance Indices and Construct Validity.

Bøhren, Ø., Staubo, S., 2016. Mandatory gender balance and board independence. European Financial Management, 22(1), 3-30.

Bowen, C., Y. Wu., C. Hwang., R. Scherer., 2007. Holding up half the sky? attitudes towards women as managers in the People’s Republic of China. The

(39)

39 Brannon, L., 2004. Gender: psychological perspectives. Psychology Press.

Cadbury Committee., 1992. Report on the financial aspects of corporate governance. GEC, London.

Campbell, K., Bohdanowicz, L., 2015. Corporate Governance and the Growing Role of Women in the Boardroom.

Campbell, K., Mínguez-Vera, A., 2008. Gender diversity in the boardroom and firm financial performance. Journal of Business Ethics, 83(3), 435-451.

Carter, D.A., B.J. Simkins., W.G Simpson., 2003. Corporate governance, board diversity, and firm performance. Financial Review, 38 (1).

Catalyst., 2017. Legislative Board Diversity. Retrieved 31 May 2017, from

http://www.catalyst.org/legislative-board-diversity

Catalyst., 2017. Women on Corporate Boards Globally. Retrieved 31 May 2017, from

http://www.catalyst.org/knowledge/women-corporate-boards-globally

Chen, M. C., Cheng, S. J., Hwang, Y., 2005. An empirical investigation of the relationship between intellectual capital and firms’ market value and financial performance. Journal of intellectual capital, 6(2), 159-176.

Chinn, P. W., 2002. Asian and Pacific Islander women scientists and engineers: a narrative exploration of model minority, gender, and racial stereotypes. Journal of Research in Science Teaching, 39(4), 302-323.

Cho, S., 1997. Korean gifted girls: the achieved and the unachieved. Gifted Education International, 12(2), 111-17

Chung, J-S., 1994. Women’s unequal access to education in South Korea. Comparative Education Review, 38(4), 487-505.

Cooke, F. L., 2005. Women's managerial careers in China in a period of reform. Asia Pacific Business Review, 11(2), 149-162.

Damodaran, A., 2009. Valuing financial service firms. Stern Business.

(40)

40 Eisenberg, T., Sundgren, S., Wells, M. T., 1998. Larger board size and decreasing firm

value in small firms. Journal of Financial Economics, 48(1), 35-54.

Eisenhardt, K. M., 1989. Agency theory: an assessment and review. Academy of Management Review, 14(1), 57-74.

El-Sayed Ebaid, I., 2009. The impact of capital structure choice on firm performance: empirical evidence from Egypt. The Journal of Risk Finance, 10(5), 477-487.

Erhardt, N. L., Werbel, J. D., Shrader, C. B., 2003. Board of director diversity and firm financial performance. Corporate governance: An international review, 11(2), 102-111.

Erkut, S., Kramer, V. W., Konrad, A. M., 2008. Critical mass: Does the number of women on a corporate board make a difference. Women on corporate boards of directors: International research and practice, 222-232.

Farrell, K. A., Hersch, P. L., 2005. Additions to corporate boards: the effect of gender. Journal of Corporate finance, 11(1), 85-106.

Fields, M. A., Keys, P. Y., 2003. The emergence of corporate governance from Wall St. to Main St.: outside directors, board diversity, earnings management, and managerial incentives to bear risk. Financial Review, 38(1), 1-24.

Finkelstein, S., Hambrick, D., 1996. Strategic Leadership: Top Executives and Their Effects on Organizations. West Publishing Company. St. Paul, Minnesota.

Freeman, R.E., 1984. Strategic Management: A Stakeholder Approach. Pitman Publishing, Boston, MA.

Gallego-Álvarez, I., García-Sánchez, I. M., Rodríguez-Dominguez, L., 2010. The influence of gender diversity on corporate performance. Revista de Contabilidad, 13(1), 53-88.

Gleason, K. C., Mathur, L. K., Mathur, I., 2000. The interrelationship between culture, capital structure, and performance: evidence from European retailers. Journal of Business Research, 50(2), 185-191.

Glorie, D., 2012. Women on the board of directors and the effect on firm value. Bachelor thesis paper. University of Amsterdam, Amsterdam.

(41)

41 Hambrick, D. C., Cho, T. S., Chen, M. J., 1996. The influence of top management

team heterogeneity on firms' competitive moves. Administrative science quarterly, 659-684.

Hausman, J. A., Taylor, W. E., 1981. Panel data and unobservable individual effects. Econometrica: Journal of the Econometric Society, 1377-1398.

Haveman, H. A., Wang, Y., 2013. Going (more) Public: institutional isomorphism and ownership reform among Chinese firms. Management and Organization Review, 9(1), 17-51.

Heidrick & Struggles., 2016. Building an effective and diverse board: Lessons from Asia. Retrieved 7 June 2017, from

http://www.heidrick.com/~/media/Publications%20and%20Reports/Building%20 an%20effective%20and%20diverse%20board%20Lessons%20from%20Asia.pdf

Hillman, A. J., Dalziel, T., 2003. Boards of directors and firm performance: Integrating agency and resource dependence perspectives. Academy of Management review, 28(3), 383-396.

Hillman, A. J., Shropshire, C., Cannella, A. A., 2007. Organizational predictors of women on corporate boards. Academy of Management Journal, 50(4), 941-952.

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

Hofstede, G., 1996. Gender stereotypes and partner preferences of Asian women in masculine and feminine cultures. Journal of Cross-Cultural Psychology, 27(5), 533-546.

Jackson, S. E., 1992. Diversity in the workplace: Human resources initiatives. Guilford Press.

Johl, S. K., Kaur, S., Cooper, B. J., 2015. Board characteristics and firm performance: Evidence from Malaysian public listed firms. Journal of Economics, Business and Management, 3(2), 239-243.

Johnsrud, L. K., 1995. Korean academic women: multiple roles, multiple challenges. Higher Education, 30(1), 17-35.