Board Gender Diversity and Firm Performance: The
Effect of National Culture
Written by
A.A.J. Scheppink
S2354306 (UoG)
T-271 (UU)
January 12
th, 2018
Supervisor: prof. dr. C.L.M. Hermes
Co-Assessor: dr. V. Purice
MSc International Financial
Management
MSc Business and Economics
Faculty of Economics and Business
Department of Business studies
University of Groningen
Uppsala University
Abstract
This paper examines the moderating effect of national culture on the relationship between board gender diversity and corporate financial performance. To test the hypotheses, Fixed Effects regression is used in combination with a sample of 1,499 firms from 23 countries and 7,125 firm-year observations over a time frame of seven years. This paper provides evidence for a significant positive effect of board gender diversity on firm performance if there are at least three females seated on the board. Furthermore, a significant moderating effect of national culture on the relationship between board gender diversity and firm performance has been found.
Keywords: Board gender diversity, firm performance, national culture, corporate governance, board of directors
1. Introduction
About two months ago, November 2017, the European Commission advocated for a new series of legislations to improve gender equality in the senior positions of public companies (The Guardian, 2017). Following countries such as Norway, Iceland, France, and Spain, this is the latest and one of the most far-reaching attempts to increase gender diversity in corporate boards. The purpose to provide everyone with the same opportunities is based on the principle of equality of treatment and should be pursued by everyone. But, is there also an economic explanation why increased gender diversity should be pursued? Or is it a merely social goal? In other words, are these gender quotas scientifically justified based on economic reasons rather than intuition or social desires? Over the last two decades, the interest of researchers in the link between Board Gender Diversity (BGD) and Corporate Financial Performance (CFP) has grown considerably. With mixed results and relatively small country sets and time frames, there is no consensus yet in the debate whether women in the board of directors truly enhance firm performance.
The argumentation for the effect of BGD on firm performance will be developed through the lenses of the Resource Dependence Theory and the Agency Theory. Until now, empirical evidence for the relationship between BGD and CFP has been mixed. In addition, research suggests that the influence of diversity on firm performance may be moderated by contextual factors (e.g. Schneid et al., 2015). Since the impact of contextual factors on the BGD–CFP relationship has not yet been explored on a wider scale, the examination of the moderating effect of national culture on the relationship between board gender diversity and firm performance is vital to this research. Culture can create gender stereotypes and influences the process of assigning different roles to men and women. This might influence the eventual perception and acceptance of people towards women in the top of the business environment. Furthermore, Brodbeck et al. (2004) found that societal culture has the strongest effect on organizational culture, which is perceived to be one of the successors in a firm’s financial performance. Therefore, the objective of this research is to provide empirical evidence on the moderating effect of national culture on the BGD-CFP link as well as assessing the sole effect of gender diversity in the board on firm financial performance.
of females. Firm performance is measured using an approximation of Tobin’s Q. Furthermore, the moderating effect of national culture is studied through three dimensions of the GLOBE study by House et al. (2004): Gender Egalitarianism, Humane Orientation, and Institutional Collectivism. These dimensions will be described in detail in the next section.
This paper expects a positive effect of gender diversity in the board on financial performance. Furthermore, this study believes the relationship between BGD and CFP to be positively moderated by national culture since the characteristics belonging to a high score on Gender Egalitarianism, Humane Orientation and Institutional Collectivism will likely mitigate the negative effects of social categorization while firms might still benefit from the positive effects of board gender diversity drawn in the Agency and Resource Dependence theory.
The remainder of this paper is organized as follows. In the next section, applicable theory around the effect of board gender diversity on firm performance and the moderating effect of culture will be presented, followed by the hypotheses and a conceptual model. The sample, data, variables and theoretical model will be discussed in Section 3. The results of the statistical analysis will be presented in section four and conclusions will be drawn in the last section after which the theoretical contribution, its limitations and suggestions for further research will be discussed.
2. Literature Review
2.1 Board Gender Diversity and Corporate Financial Performance
Research on gender diversity often falls under the larger scope of diversity work and it is believed that the general diversity arguments can also be applied for the specific case of gender diversity. This method of argumentation has been applied extensively in previous research (e.g. Carter et al., 2003; Carter et al., 2010; Francoeur, Labelle, & Sinclair-Desgagné, 2008; Marinova, Plantenga, & Remery, 2015; and Schneid et al., 2015). Gender diversity in the workplace can be classified as the acceptance of both men and women and is accomplished when there is a greater balance between the number of men and women on the work floor.
board and corporate financial performance. The diversity arguments and both theories will be discussed first, before the first hypothesis will be developed. Then, the focus shifts to a more specific aspect of board gender diversity: the critical mass theory. After discussing this theory and its arguments, the second hypothesis will be presented. Please refer to Figure 1 for a visual representation of the conceptual model, which can be found at the end of this section.
Agency theory
The Agency theory describes the role of a corporate board in aligning the interest of management and shareholders through monitoring and controlling managers (Jensen & Meckling, 1976; Marinova et al., 2015). This theoretical framework is also often used to describe the link between board characteristics and firm performance (Carter et al., 2003). Since boards are perceived to be crucial in the view of agency theory, recent literature puts more emphasis upon the different characteristics that portray a ‘successful board’. It is believed that gender diversity is, amongst others, one of these characteristics (Hillman & Dalziel, 2003).
Central to the argumentation for a gender-diverse board is the statement that “board independence is critical for boards to function in the best interests of shareholders” (Carter et al., 2003: p. 37). They argue that outside board members are less likely to collude with inside directors at the expense of the respective shareholders, since they have the incentive to build a reputation as experts in monitoring. One line of argumentation states that gender diversity can increase board independence by increasing the variety of perspectives. People with a different gender might ask other questions than their male counterparts with a more traditional background, thereby creating a more activist board (Carter et al., 2003). In a similar vein, Adams & Ferreira (2009) and Low, Roberts & Whiting (2015) found that female board members tend to use tougher monitoring on management because they are not one of the “old boys club”, and greater gender diversity on boards might also improve monitoring due to better managerial accountability. Also, there is evidence that women tend to take their role in the boardroom more seriously which might lead to better governance of an organization (Singh & Vinnicombe, 2004).
Resource dependence theory
Resource dependence theory suggests that boards provide the linkage to an organization’s vital resources and its external environment (Pfeffer & Salancik, 1978). Even more important, these linkages can be improved by an increase in size and diversity of the board of directors through several factors (Goodstein, Gautam, & Boeker, 1994; Liu, Wei, & Xie, 2014; Pfeffer, 1973). First, Carter et al. (2010) argue that female board directors may broaden the human capital by providing unique information due to their different perspective, which is in line with the arguments proposed by Robinson & Dechant (1997). They state that diversity promotes a better understanding of the market. Since approximately half of the total human population are women, it makes sense to match an organization to the diverse consumer- and supplier base in order to provide the organization with a better ability to penetrate a certain market. Furthermore, they argue that incorporating a broader perspective into the process stimulates creativity and innovation that ultimately enhances the quality of leadership, leading to enhanced performance. Since “attitudes, cognitive functioning, and beliefs are not randomly distributed in the population, but tend to vary systematically with demographic variables such as age, race and gender”, a wider set of perspectives and resources are exploited that might lead to a better understanding and ability to evaluate the complexities in the external environment and take actions accordingly (Robinson & Dechant, 1997: p. 27). This view is supported by Carter et al. (2003), Yi (2011), and Francoeur et al. (2008) who state that more diverse boards lead to a variety of perspectives who are more likely to raise questions or challenge the current strategy, solve complex issues and correct informational biases, thereby benefitting the ultimate solution-making. This could imply that diverse teams might be better in producing solutions of higher quality than homogenous groups. Although diversity can also create difficulties in communication and a higher chance of conflicts, they will ultimately outperform less diverse teams in their solutions.
Furthermore, Pfeffer & Salancik (1978) argue that increased diversity creates legitimacy and provides commitment or support for an organization in its external environment, which is supported by Hillman & Dalziel (2003) and Pfeffer & Salancik (2003). Moreover, Hillman, Cannella & Harris (2002) argue that an increase of female board representation enhances an organization’s legitimacy since gender equality is becoming one of the widely accepted social norms.
corporate governance was about the relationship between management and shareholders, nowadays firms are increasingly pressured by various stakeholders including politicians, the media and NGOs to increase its female representation in senior management. Williams (2003), for example, has shown that female directors are generally more sensitive to social and environmental issues compared to their male counterparts, which might positively affect a firm’s reputation among a wider group of stakeholders (Bear et al., 2010; Branco & Rodrigues, 2008). In this line, Robinson & Dechant (1997) argue that a more diverse board helps to reach a greater and more diverse group of people through a better representation of the different groups present in a society. This might lead to a feeling of affirmation with the relevant board members that would not exist in the case of a homogenous board of directors. The increased affirmation with gender-diverse organizations will create a differentiating advantage compared to their competitors not characterized by a gender-diverse board. Similarly, Cox & Blake (1991) state that the improved reputation of a gender-diverse organization will outperform their competitors by attracting the best female talents. Since the composition of the labour force has been changed and the competition increased, this might be a vital advantage for an organization to win the battle of talent. Additionally, the number of female managers in senior management might also have a positive influence on the lower-positioned women’s visions towards their career development and future chances, leading to a boost in productivity and firm performance (Bell, 2005; Burke & McKeen, 1996; Ely, 1990; Smith, Smith, & Verner, 2006).
Based on the Resource dependence theory it could be argued that gender diversity in the boardroom provides different resources to the firm that can lead to improved information processing, higher quality decisions, more valuable resources, better stakeholder management, and additional strategic insights that may ultimately lead to better financial performance (Carter et al., 2010; Dezsö & Ross, 2012; Hillman, Cannella, & Peatzold, 2000; and Rose, 2007).
Based on the argumentation developed in light of the Agency- and Resource dependence theory, BGD is expected to positively influence the financial performance of a firm. Therefore, I hypothesize the following:
Hypothesis 1: Board gender diversity has a positive effect on corporate financial performance.
The critical mass
Previous research suggests it does (Joecks, Pull, & Vetter, 2013; Kramer et al., 2006). The general consensus is that the threshold for a critical mass occurs when there are at least three women seated on a board together. Kramer et al. (2006) found that a critical mass leads to a different perception by the other board members where women are treated as an equal and better able to influence the process and content of board discussions. They are no longer seen as outsiders. In addition, they found several advantages of having a critical mass that might constitute the enhanced financial performance. Delicate or tough issues will less likely be ignored in discussions, which might enhance the decision-making. Due to a more open and collaborative attitude, concerns are better heard and taken care of. And having more women on the board leads to the inclusion of a broader perspective where other stakeholders are also included. The financial performance of a firm might be damaged, if they are not heard.
In addition to hypothesis 1 in which a positive relation between BGD and CFP is expected, it is believed that after a threshold value of three females on the board is reached, the performance of the firm increases. Therefore, the following hypothesis is developed:
Hypothesis 2: If at least three females are present on the board of directors, firm performance will significantly increase
2.2 The Effect of a Country’s National Culture
financial performance. This effect will be examined through three cultural dimensions proposed in the Global Leadership and Organizational Behaviour (GLOBE) study by House et al. (2004). The three cultural dimensions are: Gender Egalitarianism, Humane Orientation, and Institutional Collectivism. They will be discussed separately after which a hypothesis will be presented for every individual dimension.
The underlying concept among the cultural dimensions is social categorization, which makes a short explanation of the concept of social categorization vital. This study uses the terms social categorization and gender categorization interchangeably since they have the same meaning in the context of this specific research. Social categorization is based on social perception and refers to the process by which people are placed into groups based on characteristics like race, gender or ethnicity. An individual mentally groups people that share important characteristics (e.g. demographic features, interests, occupation). This means that people tend to make assumptions about certain characteristics of an individual based on the group they belong to. Social categorization has the risk of stereotyping complete groups, which might change the perception and attitude towards people from specific groups. This is also a fundamental aspect of the traditional debate about female participation in society, which is why it is important for this research. In short, this study believes the relationship between BGD and CFP to be positively moderated by national culture since the characteristics belonging to a high score on Gender Egalitarianism, Humane Orientation and Institutional Collectivism will likely mitigate the negative effects of social categorization while firms might still benefit from the positive effects of board gender diversity such as the arguments developed in the Resource dependence theory.
Gender Egalitarianism
An important aspect of gender roles in society is an individual’s work-life balance and the stereotypical beliefs about men and women. Gender Egalitarianism can play a pivotal role in improving the individual work-life balance and simultaneously decreasing the stereotypical beliefs of the different roles men and women have in society (Lyness & Judiesch, 2014). Low gender egalitarian countries are generally more traditional in their gender roles while high gender egalitarian countries have less differences in roles and expected priorities by gender (McDaniel, 2008). Taking into account the aspects of low gender egalitarian countries, it might be much more difficult for women to achieve a work-life balance due to a different perception of gender-specific roles. However, if a culture promotes more gender equality, it could increase women’s educational level and involvement in paid work while some of the caretaking responsibilities are shifted towards men. If the gap in the levels of work-life balance between men and women can be diminished, women will be better able to focus on their personal as well as their corporate development. Furthermore, women will be better accepted due to their equal role in society. These characteristics, belonging to high gender egalitarian countries, will therefore not only lead to more females in senior management positions, but they are also better equipped with improved firm performance as a result (Lyness & Judiesch, 2014).
Based on the arguments presented above, it could be inferred that there is less social categorization in high gender egalitarian countries, which mitigates the possibly negative effects on firm performance coming from intra-organizational frictions between genders. Also, typical male and female occupations in organizations will be blurred and gender stereotypes will be diminished (Schneid et al., 2015). Rather than be regarded as a threat, women in boardroom positions will now be valued as an equal and might be better equipped to boost financial performance of the firm. Based on the above discussion, the following hypothesis is developed:
Hypothesis 3a: Gender Egalitarianism positively moderates the relationship between board gender diversity and corporate financial performance.
Humane Orientation
humane-oriented cultures and the role of women in corporations (Bajdo & Dickson, 2001; Conelly & Rhoton, 1998; Grant, 1998). Not only will women’s advancement in organizations be encouraged, but their perspectives and input are also better heard. This might strengthen the relation between board gender diversity and firm performance.
Humane Orientation refers to “the degree to which an organization or society encourages and rewards individuals for being fair, altruistic, friendly, generous, caring and kind to others” (House et al., 2004: p. 569). While low humane oriented societies are characterized by self-interest and motivated by power and material possessions, countries with a high score on Humane Orientation are motivated by an urge of affiliation and belonging and sensitivity for discrimination (Brodbeck & Frese, 2007; Kabasakal & Bodur, 2004).
This study believes that Humane Orientation diminishes the negative influence of the traditional gender categorization, which might strengthen the effect of board gender diversity on firm performance. First, individuals in low humane oriented countries mainly care about their self-interest and are more likely to develop sub-groups which could hamper the overall group-process. On the other hand, people in countries with a high score on Humane Orientation tend to be more altruistic and caring, are more likely to form one group regardless of gender, and focus on cooperation which boosts the overall performance (Kabasakal & Bodur, 2004). Second, humane oriented societies are characterized by less discrimination which means there will be less friction and negativities within a group of individuals belonging to different social categories (House et al., 2004; Schneid et al., 2015). Lastly, Lundberg (2007) showed that individual behaviours in countries with a high score on Humane Orientation are similar to behaviours that characterize good teamwork, which could enhance interactions within a team. By providing a safe, respectful and open place to discuss important issues, companies provide women the opportunity to flourish and add value to the discussion by bringing in different perspectives, with increased financial performance as a subsequent result.
Overall, countries with a high score on humane orientation are more open to others and will likely benefit more from diversity and its different pool of resources while the negative effects of diversity – the social categorization – will be minimized. Therefore, the following hypothesis is developed:
Institutional Collectivism
Collectivism consists of two separate dimensions: Institutional Collectivism and In-group Collectivism. Institutional Collectivism is “the degree to which organizational and societal institutional practices encourage and reward collective distribution of resources and collective action” (House et al., 2004: p. 12). In-group collectivism refers to “the degree to which individuals express pride, loyalty, and cohesiveness in their organizations or families” (House et al., 2004: p. 12).
At first sight, the In-group Collectivism seems the most relevant dimension for this study. Closer investigation by Brewer & Venaik (2011), however, shows differently; more than half of the questions used to investigate the In-group Collectivism construct are highly family-related rather than group oriented. Although the definition used by House et al. (2004) incorporates ‘organizations and families’, none of the items asks about organizations but rather about families. Moreover, the dimension of Institutional Collectivism incorporates – next to the elements of organizations, societies, businesses, economies and institutions – also the element of groups in their measurement (Brewer & Venaik, 2011). Therefore, this study focuses on the dimension of Institutional Collectivism only.
Collectivist countries are characterized by individuals who feel affiliated with their employer and its respective organization with loyalty to the group being more important than pursuing your own goal. Individualistic countries on the other hand, consist of independent individuals that strive for their self-interest rather than pursuing the collective’s interest (Gelfand et al., 2004; Schneid et al., 2015). As is also the case for Gender Egalitarianism and Humane Orientation, it is believed that Institutional Collectivism can diminish the negative effects of social categorization through higher group loyalty and commitment to pursue team goals (Gelfand et al., 2004; Oyserman, Kemmelmeier, & Coon, 2002). In other words, the strong commitment towards a group and the overall goal that characterizes collectivistic societies enhances team cohesion and hence its performance (House et al., 2004).
However, these arguments could also be applied in the opposite direction. Stangor et al. (1992) found gender is one of the most important bases of social categorization, which might imply that a ‘group’ means ‘people within the same gender’ whereas people from the other gender are seen as out-group. This might hamper the effectiveness of the group since collectivistic societies are characterized by high group loyalty while they show a negative bias towards individuals who do not belong to the same group (Gelfand et al., 2004).
this study believes that by creating cohesion and a focus on the overall group interest, Institutional Collectivism eliminates the negative effects of gender categorization while it may benefit from the additional knowledge, skills and perspectives brought to the table by a gender-diverse group (van Knippenberg et al., 2004). Therefore, the following hypothesis is developed:
Hypothesis 3c: Institutional Collectivism positively moderates the relationship between board gender diversity and corporate financial performance.
Figure 1 - Conceptual model including proposed hypotheses
3. Methodology
3.1 Sample and Data
In order to investigate the relationship between BGD and CFP and the subsequent moderating effect of national culture on this relationship, this analysis is focused on a broad range of publicly traded companies across the globe.1 In line with previous research, this research focuses on public firms due to its extensive data availability (e.g. Marinova et al., 2015). Companies in the financial and utilities sector are neglected in this sample because of their specific accounting method and high correlation amongst each other. Data on the board of
directors as well as financial performance and relevant controls are obtained through DataStream from the Thomson Reuters’ Asset4 ESG database and converted to US$ where applicable. The data for the cultural dimensions are obtained from the database of the GLOBE study developed by House et al. (2004). Table 1 provides a short definition of each of the variables. Data is collected for 1,499 firms from 2010 until 2016. In the end, the sample consists of a balanced panel of 7,125 complete observations. While most studies are predominantly US based with a few others focusing on a small set of specific countries, this study takes a broader perspective by incorporating a relatively large set of countries coming from all regions. Moreover, there are only very few studies using a seven-year timeframe after the start of the most recent financial crisis.
3.2 Description of Variables
Corporate financial performanceIn the area of corporate governance research, there are several ways of measuring financial performance with the most predominant indicators being Tobin’s Q, ROE and ROA. While the first indicator is a market-based measure the latter two are based on accounting data. Although these measures are commonly used next to each other, they are not identical because they measure significantly different aspects of CFP. Whereas accounting-based profitability measures such as ROA or ROE are based on historical performance, Tobin’s Q also incorporates market expectations and thus focusses on future performance expectations (Campbell & Minguez-Vera, 2007).
opportunity to strengthen its competitive advantages, which will in turn impact long-term performance (Wernerfelt & Montgomery, 1988). Furthermore, women in the board are generally characterized by a focus on long-term goals (Marinova et al., 2015). Whereas ROA and ROE focus on past operational performance, Tobin’s Q measures long-term performance leading this paper to believe that Tobin’s Q would be a better indicator for firm performance in the context of this research. Furthermore, Tobin’s Q is not prone to reporting biases due to tax laws and accounting conventions, which makes it easier to compare between firms from different countries (Lindenberg & Ross, 1981; Campbell & Minguez-Vera, 2007). In line with previous research, Tobin’s Q is transformed into its natural logarithmic form to mitigate the effect of potential outliers (e.g. Nguyen, Locke, & Reddy, 2014).
Board gender diversity
The independent variable of interest is gender diversity on the board. Following the previous strands of literature, this research makes use of two proxies for BGD (Campbell & Minguez-Vera, 2007; Carter et al., 2003; Carter et al., 2010; Nguyen et al., 2014). First, I use %FEMALE, which is the percentage of women on the board in relation to total board size. This measure helps to understand whether the number of women relative to the total size of the board have an impact. In other words, whether a higher presence of women on the board strengthens the positive impact on CFP. The second measure, CRITICAL, links to the critical mass theory described earlier. Both Joecks et al. (2013) and Kramer et al. (2006) found that the threshold value of women on the board should be at least three. Therefore, a dummy variable is created with a value of one if there are three or more female board members present and zero otherwise. National culture
Although there are nine dimensions in total this study only uses the dimensions that are expected to be most relevant for the specific subject. Moreover, by using only three of the dimensions, the critique proposed by Hofstede (2006) and Peterson & Castro (2006) that the nine dimensions of the GLOBE study “may be too highly correlated to be considered distinct at the nation level” can be directly countered (Peterson & Castro, 2006: p. 513).
The GLOBE study is also the only study that does not only measure values, but also actual practices (Venaik & Brewer, 2008). While values represent the score that an individual would like to see in the ideal situation, the practice score is based on the perception of the actual situation. So, whereas the values score represents a score that countries wish to have, the practice score shows the actual situation as it is. In line with previous literature, this study wants to examine the real influence of culture rather than cultural ideals (Brock et al., 2008; Estrin, Baghdasaryan, & Meyer, 2009; Schneid et al., 2015). Therefore, I only use the practice scores to analyse the moderating effect of national culture. All three cultural indicators were measured on a seven-point scale ranging from 1 to 7. A low score on the respective dimension indicates poor performance on this dimension while a high score indicates good performance (House et al., 2004). Meaning that if a country shows a low score on Gender Egalitarianism, Humane Orientation, and Institutional Collectivism, it would be typified as a society with high gender inequality, a strong hierarchy and power structure that promotes self-interest rather than the common good.
Control variables
In order to investigate the real effect of BGD on CFP and the moderating effect of national culture, several controls are incorporated. Next to several corporate governance controls, some firm-level control variables are included as well. Furthermore, to control for other unobserved factors, firm- and year effects are included in the model.
the total board size. The effect of board independence on firm performance has been thoroughly investigated in the area of corporate governance but with mixed results (Carter et al., 2010). Furthermore, this study accounts for a dual function of CEO/Chairman of the Board. The Agency theory literature argues that, due to conflicts of interest, a dual function might reduce the board’s ability to function properly in terms of its governance since its prime function is to monitor the performance of its top management (Rechner & Dalton, 1991).
3.3 Theoretical Model
To test the effect of BGD on CFP and the moderating effect of national culture on this relationship, Fixed Effects regression is used.
A major concern in the diversity–performance research area is the possibility of endogeneity. There are two sources of endogeneity: unobserved heterogeneity and reverse causality. Although the focus in this research is on a unidirectional relation between BGD and CFP, several authors express their concern about the reverse causality issue in this relationship (e.g. Campbell & Minguez-Vera, 2007; Carter et al., 2010; Hermalin & Weisbach, 2003). They argue that board characteristics are often not random exogenous variables. An often-heard
TABLE 1 Variable Definitions
Independent variable (BGD)
%FEMALE Percentage of females in the board
CRITICAL Dummy variable with value 1 if at least three women in the board; 0 otherwise
Dependent variable (CFP)
TOBINQ (Market Value of Equity + Preferred Stock + Debt) / Total Assets
Moderating variable (National Culture)
GE Cultural index score for Gender Egalitarianism reflecting the degree to which a collective minimizes gender inequality ranging on a scale from 1 to 7 HO Cultural index score for Humane Orientation reflecting
the degree to which an organization or society encourages and rewards individuals for being fair, altruistic, friendly, generous, caring and kind to others ranging on a scale from 1 to 7
IC Cultural index score for Institutional Collectivism reflecting the degree to which organizational and societal institutional practices encourage and reward collective distribution of resources and collective action ranging on a scale from 1 to 7
Control variables
BOARDSIZE Natural logarithm of number of people on the board INDEPENDENT % of independent board members
DUALCEO Dummy variable with value 1 if the chairman is also CEO or has been the CEO; 0 otherwise
FIRMSIZE Natural logarithm of total assets
AGE Years of existence since its incorporation LEV Debt-to-equity ratio
argument is that firms might attract more diversified board members due to its sound operations and performance (e.g. Smith et al., 2006). To mitigate the reverse causality issue, the independent variables are lagged by one year. Furthermore, the use of the fixed-effects method automatically diminishes the unobserved heterogeneity issue. To further reduce the unobserved heterogeneity, this study follows Wintoki et al. (2012) by including a lagged performance variable as a control. The fixed effects estimates are important since they help mitigating omitted variables and address unobserved changes over time (Carter et al., 2010). Subsequently, a Hausman specification test is performed. The null hypothesis states a random effects regression is preferred while the alternative hypotheses states the fixed effects model is most consistent. Returning a p-value of 0.000, we fail to accept the null hypothesis, which means the fixed effects model is most consistent.
One of the drawbacks of the fixed effects model is the fact that time-invariant variables are dropped in the regression. Therefore, this study is, in line with the research of Elkhuizen et al. (2017), primarily interested in the coefficients of the interaction terms.
Tests for heteroskedasticity and autocorrelation have been conducted. To test for heteroskedasticity, a White test is used which fails to accept the null hypothesis (p=0.000). In addition, a Woolridge test is conducted to check for autocorrelation in panel data. This test also fails to accept the null hypothesis (p=0.000), which means that the standard errors have to be corrected for both heteroskedasticity and serial correlation. Therefore, standard errors that are robust to heteroskedasticity and serial correlation need to be included.
All models (1, 2, 3, 4) show the Fixed Effects regression model including the lagged variables as well as firm and year effects. While model (1) and (2) are used to test the first and second hypotheses, model (3) and (4) are used to test hypotheses 3a, 3b, and 3c.
1 #$%&'(),+ = - + /0%234563),+70+ /8%$59:;&<3),+70+ /=&':3>3':3'#),+70 + /?:@56A3$),+70+ /B2&94;&<3),+70+ /C5D3),+70+ /E63F),+70 + /G#$%&'( − 1),++ I),++ J)+ J+
2 #$%&'(),+ = - + /0A9&#&A56),+70+ /8%$59:;&<3),+70+ /=&':3>3':3'#),+70 + /?:@56A3$),+70+ /B2&94;&<3),+70+ /C5D3),+70+ /E63F),+70 + /G#$%&'( − 1),++ I),++ J)+ J+
3 #$%&'(),+ = - + /0%234563),+70+ /8 %234563),+70∗ D3)
4 #$%&'(),+ = - + /0A9&#&A56),+70+ /8 A9&#&A56),+70∗ D3)
+ /= A9&#&A56),+70∗ N$) + /? A9&#&A56),+70∗ &A) + /B%$59:;&<3),+70 + /C&':3>3':3'#),+70+ /E:@56A3$),+70+ /G2&94;&<3),+70+ /O5D3),+70 + /0P63F),+70+ /00#$%&'( − 1),++ I),++ J)+ J+
In these models, TOBINQ represents the financial performance of a firm, %FEMALE indicates the degree of gender diversity on the board of directors measured by the percentage of females, CRITICAL is a dummy variable that has a value of 1 if a critical mass of females is present on the board, GE, HO, and IC capture either one of the three relevant cultural dimensions; Gender Egalitarianism, Humane Orientation and Institutional Collectivism. Also, I),+ is included as the error term and J) and J+ are incorporated to capture the entity- and time-fixed effects.
4. Empirical Results
4.1 Descriptive Statistics
Table 2 provides descriptive statistics for this sample. The value of Tobin’s Q indicates that the sample firms were relatively successful over the time period, but there was also wide variation in the performance measure. The mean value of Tobin’s Q is 1.26, which is greater than one. This indicates that the market value of the firm is greater than the book value of the assets. With a standard deviation of 1.052, a minimum value of .0478 and a maximum value of 6.003 the variation is significant.
Furthermore, approximately twelve per cent of the firm’s board of directors were female with a maximum of 58.33 per cent. From all sample firms, only 14.79 per cent of them had three or more females seated on the board of directors. On average, people tend to assign lower values to Gender Egalitarianism compared to the other two cultural dimensions.
TABLE 2 Descriptive statistics
Variable N Mean SD Median Min Max
Dependent TOBINQ 10,173 1.256 1.052 .9444 .0478 6.003 Independent %FEMALE 8,994 .1194 .1125 .1111 0 .5833 CRITICAL 8,994 .1479 .3550 0 0 1 GE 8,994 3.383 .2131 3.340 2.900 3.930 HO 8,994 4.110 .3357 4.170 3.290 4.960 IC 8,994 4.436 .4257 4.270 3.680 5.220 Moderators %FEMALE * GE 8,994 .4103 .3927 .3711 0 2.240 %FEMALE * HO 8,994 .4821 .4540 .4633 0 2.392 %FEMALE * IC 8,994 .5131 .4888 .4666 0 3.045 CRITICAL * GE 8,994 .5089 1.225 0 0 3.390 CRITICAL * HO 8,994 .5826 1.406 0 0 4.960 CRITICAL * IC 8,994 .6271 1.512 0 0 5.220 Controls BOARDSIZE 8,994 2.249 .3315 2.303 .6931 3.497 INDEPENDENT 8,970 .5500 .2963 .6000 0 1 DUALCEO 8,994 .4168 .4931 0 0 1 FIRMSIZE 8,994 15.40 1.494 15.38 7.975 19.83 AGE 7,474 3.881 .7037 4.007 -.105 5.236 LEV 8,988 .7935 1.457 .4792 -3.99 9.641 TOBINQ-1 8,691 1.247 1.039 .9405 .0478 6.003
Note: Table 2 presents the number of observations, mean, standard deviation, median, minimum value and maximum value for each variable. The definition of each variable is presented in Table 1. LEV is winsorized at the 1st and 99th percentile.
TABLE 3
Descriptive statistics %FEMALE per country
Country N Mean SD Median Min Max
Australia 917 .1205 .1220 .1250 .0000 .5000 Brazil 133 .0753 .0793 .0769 .0000 .2857 Canada 721 .1173 .1098 .1111 .0000 .5455 China 105 .0299 .0590 .0000 .0000 .3750 Denmark 105 .1810 .1009 .1818 .0000 .4167 Finland 119 .2504 .0969 .2500 .1111 .5000 France 175 .2923 .1024 .3000 .0000 .5455 Germany 350 .1712 .1079 .1667 .0000 .4167 Great Britain 1,190 .1605 .1098 .1429 .0000 .5000 India 196 .0775 .0684 .0833 .0000 .3000 Indonesia 112 .0497 .0803 .0000 .0000 .3333 Ireland 98 .1650 .0832 .1667 .0000 .3846 Italy 84 .1918 .0943 .2143 .0000 .4375 Japan 1,890 .0189 .0481 .0000 .0000 .5000 Malaysia 112 .1197 .0800 .1111 .0000 .2308 Mexico 98 .0532 .0953 .0000 .0000 .4545 Singapore 140 .0724 .0782 .0801 .0000 .3333 South-Africa 182 .1966 .0935 .2000 .0000 .4286 Spain 140 .1321 .0893 .1111 .0000 .4000 Sweden 175 .2995 .1172 .2727 .0667 .5833 Switzerland 224 .1081 .1034 .1111 .0000 .4000 The Netherlands 161 .1980 .1319 .2222 .0000 .5000 United States 3,066 .1593 .0980 .1538 .0000 .5455
TABLE 4
Descriptive statistics DCRITICAL per country
Country N Mean SD Median Min Max
Australia 917 .0567 .2314 .0000 .0000 1.000 Brazil 133 .0150 .1222 .0000 .0000 1.000 Canada 721 .1345 .3415 .0000 .0000 1.000 China 105 .0095 .0976 .0000 .0000 1.000 Denmark 105 .1810 .3868 .0000 .0000 1.000 Finland 119 .2437 .4311 .0000 .0000 1.000 France 175 .8057 .3968 1.000 .0000 1.000 Germany 350 .4657 .4995 .0000 .0000 1.000 Great Britain 1,190 .1756 .3807 .0000 .0000 1.000 India 196 .0306 .1727 .0000 .0000 1.000 Indonesia 112 .0089 .0945 .0000 .0000 1.000 Ireland 98 .1939 .3974 .0000 .0000 1.000 Italy 84 .5357 .5017 1.000 .0000 1.000 Japan 1,890 .0021 .0460 .0000 .0000 1.000 Malaysia 112 .2500 .4523 .0000 .0000 1.000 Mexico 98 .1327 .3409 .0000 .0000 1.000 Singapore 140 .0214 .1453 .0000 .0000 1.000 South-Africa 182 .4670 .5003 .0000 .0000 1.000 Spain 140 .2286 .4214 .0000 .0000 1.000 Sweden 175 .6229 .4861 1.000 .0000 1.000 Switzerland 224 .0670 .2505 .0000 .0000 1.000 The Netherlands 161 .2422 .4298 .0000 .0000 1.000 United States 3,066 .2009 .4007 .0000 .0000 1.000
Note: Table 4 presents the number of observations, mean, standard deviation, median, minimum value and maximum value per country for the critical mass variable. The definition of the variable is presented in Table 1.
Table 3 and 4 display the respective statistics per country of the BGD variables since these are of central interest to this research. The United States, together with Japan and Great Britain cover a relatively large part of the sample. Looking at Table 3, European and North-American firms score higher with respect to the percentage of females on the board than their Asian and South-American counterparts. Especially, the Scandinavian countries (Denmark, Finland, and Sweden) score relatively high. Moreover, Finland and Sweden are the only two countries in the sample with no single firm consisting of solely male board members, while Sweden also scores best regarding the percentage of females in a single year (58.33%) and shows the highest mean value (29.95%). Also, there is no country consisting of firms with only male board members.
average Board of Directors in France consists of 14 people while a board in Sweden, Finland, and Denmark consists of respectively 10, 9, and 8 people. Furthermore, there is no country in the sample that only consists of firms with boards characterized by a critical mass nor firms with no critical mass of females in the board.
4.2 Correlation Matrix
Table 5 reports the Pearson’s correlation matrix between the dependent, independent, moderating and control variables. The correlation coefficients of .145 and .078 indicate that the percentage of female board members and the presence of a critical mass of female board members are positively related to CFP. At first sight, this would support the study’s first and second hypothesis. Furthermore, all cultural interaction terms show positive coefficients except for %FEMALE*GE (-.01) with regard to Tobin’s Q.
Recently, there has been discussion about multicollinearity in interaction terms due to high correlations with its independent predictor variables (see e.g. Dawson, 2014). With correlation coefficients of higher than .90, potential multicollinearity was also present in this study. One way to deal with this problem is by applying the mean-centring procedure on the predictor variables before creating the interaction term (Dawson, 2014; Shieh, 2011). This procedure has also been conducted in this research.
TABLE 5 Correlation matrix [1] [2] [3] [4] [5] [6] [7] [8] [9] [10] [11] [12] [13] [14] [15] [16] [17] [18] [19] Dependent [1] TOBINQ 1.00 Independent [2] %FEMALE .145 1.00 [3] CRITICAL .078 .652 1.00 [4] GE .044 .200 .087 1.00 [5] HO .035 -.18 -.19 -.12 1.00 [6] IC -.13 -.19 -.13 .110 .445 1.00 Moderators [7] %FEMALE * GE -.01 .267 .193 .143 .008 .313 1.00 [8] %FEMALE * HO .005 -.12 -.22 .020 .170 .041 -.15 1.00 [9] %FEMALE * IC .113 -.09 -.18 .272 .071 -.04 .228 .425 1.00 [10] CRITICAL * GE .024 .192 .259 .123 .046 .211 .694 -.07 .237 1.00 [11] CRITICAL * HO .013 -.18 -.35 .059 .209 .156 -.05 .695 .361 -.01 1.00 [12] CRITICAL * IC .064 -.17 -.44 .190 .195 .151 .240 .420 .730 .368 .614 1.00 Controls [13] BOARDSIZE .009 .220 .353 -.13 -.19 -.11 .082 -.10 -.19 .073 -.21 -.26 1.00 [14] INDEPENDENT .162 .221 .029 .205 .252 -.21 -.11 .185 .201 -.01 .176 .170 -.10 1.00 [15] DUALCEO .077 .016 .030 -.17 .135 -.07 -.05 .044 -.06 -.00 .034 -.03 .150 .204 1.00 [16] FIRMSIZE -.15 .214 .260 -.15 -.14 -.01 .106 -.08 -.19 .081 -.14 -.21 .576 .002 .156 1.00 [17] AGE -.02 .025 .066 -.01 -.06 .06 .046 -.05 -.09 .036 -.05 -.07 .085 -.03 .026 .085 1.00 [18] LEV -.05 .058 .051 -.01 -.05 -.03 .009 .038 -.04 -.01 .001 -.04 .108 .003 -.01 .147 -.00 1.00 [19] TOBINQ-1 .583 .066 .034 .019 .026 -.06 .037 -.02 .072 .050 -.00 .050 -.05 .053 .014 -.23 -.00 -.07 1.00
Note: The predictor variables for the interaction terms are first mean-centred before the interaction was created. Even though the GE, HO and IC variables are not included in the fixed effects regression due to their time-invariance, they are included in the correlation matrix to show there are no multicollinearity problems among the cultural
4.3 Fixed Effects Regression Analysis
This section provides the results of the fixed effects regression analysis concerning the relationship between BGD and CFP and the moderating effect of national culture. The regression does not include dummy variables to account for countries or industries since these are time-invariant and are already incorporated due to accounting for firm and year fixed effects. The results of the effects between BGD and Tobin’s Q as well as the moderating effect of national culture are reported in Table 6. Model (1) shows the regression of BGD, which is measured by the percentage of females, on CFP including control variables while Model (2) shows the regression between the critical mass of females on the board and Tobin’s Q. Model (3) shows the regression results of the percentage of females on the board on Tobin’s Q including the moderating effect of national culture. Model (4) follows the same procedure with a different measure for BGD: a dummy variable that indicates the presence of a critical mass. The interaction terms for culture are adapted accordingly.
In Table 6, it can be observed in Model (1) that the percentage of females on the board has no effect on Tobin’s Q. The critical mass dummy in Model (2), however, is positively and significantly related to Tobin’s Q. This suggests that greater female board representation does affect firm performance, but only after a certain threshold is reached. If there are at least three women seated on the board, the negative aspects will be outweighed by the positive ones and better financial performance will be achieved.
terms for Gender Egalitarianism and Humane Orientation are both negative and strongly significant while the Institutional Collectivism coefficient is positive and strongly significant. Also, the coefficient for the critical mass variable becomes stronger and is now significant at the one per cent level instead of ten per cent.
Since the control variables show identical results in all models, I will discuss these variables at once instead of separately per model. Firm size is significant and positively related to CFP. This is in line with the argumentation provided in the previous section regarding the economies of scale and increased market power that characterizes larger firms. It could also be related to the Resource dependence theory since larger firms generally possess a larger variety of resources, which might enhance firm performance. Second, firm age is negatively related to CFP and significant at the five per cent level. Implying that the benefit of having more experience does not outweigh the downside of becoming more prone to bureaucracy and corporate slack. Furthermore, Table 6 shows the significant and negative effect of a firm’s degree of financial leverage on CFP. This supports the view that corporate governance mechanisms and financial leverage might be substitutes to reduce the agency costs. The proxy for board independence is positive in all models but only significant in Model (3). In relation to the arguments provided by the agency theory, this result was expected since independent boards are better able to monitor management which might increase financial performance. Lastly, board size and the dual function of CEO/Chairman have no effect on firm performance.
TABLE 6
Fixed Effects Regression Estimates of the relationship between BGD and CFP, and the subsequent influence of National Culture. CFP is measured by the natural logarithm of
Tobin’s Q. Independent variables (1) TOBINQ (2) TOBINQ (3) TOBINQ (4) TOBINQ %FEMALE .004 (0.06) .186* (2.28) CRITICAL .039* (2.49) .071*** (3.95) %FEMALE * GE -1.317*** (-4.12) %FEMALE * HO -1.229*** (-7.88) %FEMALE * IC 1.159*** (5.96) CRITICAL * GE -.279*** (-7.38) CRITICAL * HO -.143*** (-5.81) CRITICAL * IC .169*** (7.45) BOARDSIZE -.013 (-0.52) -.022 (-0.86) -.012 (-0.46) -.019 (-0.70) INDEPENDENT .061 (1.89) .061 (1.85) .072* (2.40) .064 (1.98) DUALCEO -.014 (-1.73) -.013 (-1.71) -.014 (-1.64) -.015 (-1.81) FIRMSIZE .049* (2.17) .050* (2.22) .054* (2.44) .050* (2.30) AGE -.160** (-3.13) -.158** (-3.11) -.147** (-3.01) -.157** (-3.05) LEV -.007*** (-6.30) -.007*** (-6.44) -.006*** (-5.97) -.007*** (-6.66) TOBINSQ–1 .311*** (9.40) .311*** (9.43) .310*** (9.41) .311*** (9.41) Constant -.561* (-2.32) -.562* (-2.39) -.710** (-2.93) -.585** (-2.56) Within R2 .197 .198 .200 .199 Between R2 .665 .669 .676 .671 Overall R2 .545 .547 .553 .549 Observations 7,125 7,125 7,125 7,125 F-statistic 231.25 354.30 56.33 368.82
Year effects yes yes yes yes
Firm effects yes yes yes yes
t-Statistics in parentheses: ***indicates p < .01, **indicates p < .05, and *indicates p < .10 Standard errors are robust to heteroskedasticity and serial correlation.
4.4 Robustness Check
An additional test is conducted to validate the results of the main analysis. Next to Tobin’s Q, there are two other commonly used performance measures: ROE and ROA. The results can be found in Table A.1 and A.2 in the appendix.
Looking at Model (1) and (2) of Table A.1 and A.2, it can be concluded that the results for the first two hypotheses are similar for all performance indicators. No effect on firm performance can be found for the percentage of females present on the board, while the presence of a critical mass of females on the board shows a significant and positive effect on firm financial performance. These results indicate that the regression results of Tobin’s Q are robust to the other two performance measures with respect to Hypotheses 1 and 2. From these results, it can be concluded that the relationship between BGD and CFP is robust to both market-based as well as accounting-based performance measures.
In contrary to the main analysis, the moderating effect of culture shows different results for the other performance indicators. Taking a look at Model (3) of Table A.1 and A.2, it can be found that Gender Egalitarianism has no effect on the BGD–CFP relationship while there was a negative relationship found in the main analysis. Also, the moderating effect of Institutional Collectivism became negative when CFP is measured by both ROE and ROA instead of the positive effect shown in the main analysis. In contrast, the moderating effect of Humane Orientation shows to be robust to all performance measures. However, caution is required by interpreting these results. Since there appears to be no effect of BGD on CFP, the moderating effect of culture can be hard to interpret and might not make much sense.
Furthermore, by assessing the moderation of culture regarding the critical mass, results also deviate from the main analysis. Model (4) of Table A.1 shows no moderation effect of culture for any of the dimensions in terms of ROE as the performance indicator. Looking at Model (4) of Table A.2, opposing results are found for the moderation of national culture. Whereas the effect of Gender Egalitarianism was negative in the main analysis, it shows a positive effect when performance is measured by ROA. Also, the moderation effect of Institutional Collectivism switched signs. While the effect was positive in the main analysis, the effect becomes negative if ROA is used as the indicator for performance. Humane Orientation changed from a negative effect to no effect.
effect of national culture diverges significantly when different performance indicators are investigated.
5. Conclusion and Discussion
The purpose of this paper is to provide empirical evidence on the moderating effect of national culture on the BGD-CFP link as well as assessing the sole effect of gender diversity on the board on firm financial performance. The dependent variable, CFP, is measured through the natural logarithm of Tobin’s Q. Two indicators are used to measure the BGD construct: the percentage of females on the board and the presence of a critical mass of females on the board. The moderating effect of culture is measured through three cultural dimensions developed in the GLOBE study. These dimensions are Gender Egalitarianism, Humane Orientation, and Institutional Collectivism. Besides, there has been controlled for several corporate governance mechanisms and firm characteristics as well as previous performance. The models are tested using a fixed effects regression model including both year- and firm effects. The hypotheses were developed from a few theories. Hypotheses 1 and 2 are built upon the Resource dependence theory and Agency theory, while hypotheses 3a, 3b, and 3c are using cultural dimensions from the GLOBE study and are based upon the social categorization process.
Taking a step back to the conceptual model presented earlier in this research, a positive sign for all hypotheses was expected. The first and second hypotheses expect BGD to have a positive effect on CFP. While hypothesis 1 measures BGD by using the percentage of females present on the board, the second hypothesis investigates the effect of a critical mass of females seated on the board. It appears the hypothesized positive effect of BGD on CFP is partly supported by the empirical results. More specifically, no relation has been found between the percentage of females present on the board of directors and firm performance, but a significant positive effect of the critical mass of females on CFP was present. This indicates that women on the board of directors do affect CFP, but only after a certain threshold value is reached. If a critical mass of at least three women on the board of directors is reached, they will positively affect a firm’s financial performance. These results are in line with Joecks et al. (2013), Kramer et al. (2006), Rose (2007) and others.
expected to have a positive moderating effect. While the first two dimensions show significantly negative effects on the BGD-CFP relationship, Institutional Collectivism suggests a significantly positive effect. This means the results for Gender Egalitarianism and Humane Orientation are contradictory to the expected results. Albeit there are arguments that might explain the negative effect of Gender Egalitarianism and Humane Orientation, no theory has been developed yet to support these explanations. A positive moderation effect was expected based on the belief that cultures possessing specific characteristics of Gender Egalitarianism and Humane Orientation were able to diminish the social or gender categorization present in societies, which would enable better cooperation and hence increase performance. However, in high gender egalitarian countries there is almost no presence of gender categorization and consequently no difference in gender. This might imply that gender diversity does not matter anymore if everyone is equal since there exists no difference between groups. The same argument holds for Humane Orientation. Furthermore, high humane oriented societies are characterized by their openness, which might enhance performance at first due to the acceptance of new resources and perspectives brought by increased BGD. However, too many different perspectives in the board might hamper the group process and distort the balance in the group, subsequently decreasing performance. Also, high humane oriented societies tend to value cooperation and care for others most. Although this could enhance the group process, it could also lead to a reduced sense of one’s own responsibilities. People in low humane oriented societies, on the other hand, value achievements and assertiveness and are primarily interested in achieving their goal, thereby increasing CFP. In contrast to the first two dimensions, Institutional Collectivism shows to mitigate the gender categorization process, thereby enhancing an effective group process leading to increased performance. The moderating effects of culture show similar results for both models, indicating the effect of culture is robust to different measures of board gender diversity.
In summary, while hypothesis 1 needs to be rejected, hypothesis 2 can be accepted. Although the percentage of females on the board does not affect firm performance, a critical mass of at least three females positively affect CFP. Hypotheses 3a and 3b need to be rejected since Gender Egalitarianism and Humane Orientation appear to negatively influence the BGD-CFP relationship, while hypothesis 3c can be accepted since Institutional Collectivism positively moderates the relationship between BGD and CFP.
a critical mass of females on the board significantly enhances corporate performance, an almost equal presence of men and women on the board might help to boost the economy. However, firms should not make the mistake that simply appointing one or two women in the board will increase their financial performance. This study did not found any effect on CFP if the number of females on the board is below the threshold value of three.
The contribution of this study to the extant literature is present in several ways. While the relationship between BGD and CFP has been investigated thoroughly, existing research is either US based or focused on a very small set of particular countries. This study presents a broader perspective by incorporating a relatively large set of countries across all regions. In addition, only very few studies have investigated data on a time frame of seven years. Furthermore, the effect of national culture on the BGD–CFP link has, to the best of my knowledge, not yet been investigated. Even though existing research mentions the importance of contextual factors in the relationship between diversity and performance, this effect has not been investigated on a large scale. In short, this study offers new insights in the BGD–CFP area by studying the effects of one of most prevalent contextual factors: national culture. In addition, by showing that the effect of BGD on CFP is dependent on national culture, this study provides empirical evidence for the call to further investigate the influence of contextual factors on the BGD–CFP relationship.
Despite the contributions, this study has some limitations too. First, culture is assumed to be stable over time and represents general norms and values that apply to each and every individual within the same culture. However, the sample firms are assigned to the national culture where their headquarters is situated. This does not mean that all directors possess the same nationality, and even if so, individuals with foreign experience due to expat positions or international contacts might act differently from the general norms and values of a particular country. Second, one of main criteria in the sample selection process was the availability of data for the independent variables. There is a plausible chance that the sample firms were the ones that are generally better firms. This could impede the interpretation and generalisation of the results (Nguyen et al., 2014). Lastly, although a positive effect is found, there is no clarity through which channels BGD has an effect on CFP.
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.
Bain, J. S. 1951. Relation of profit rate to industry concentration: American manufacturing, 1936– 1940. The Quarterly Journal of Economics, 65(3), 293-324.
Bajdo, L. M., & Dickson, M. W. 2001. Perceptions of organizational culture and women's advancement in organizations: A cross-cultural examination. Sex Roles, 45(5-6), 399-414.
Bear, S., Rahman, N., & Post, C. 2010. The impact of board diversity and gender composition on corporate social responsibility and firm reputation. Journal of Business Ethics, 97(2), 207-221.
Bell, L. A. (2005). Women-led firms and the gender gap in top executive jobs.
Branco, M. C., & Rodrigues, L. L. 2008. Factors influencing social responsibility disclosure by Portuguese companies. Journal of Business Ethics, 83(4), 685-701.
Brewer, P., & Venaik, S. 2011. Individualism–collectivism in Hofstede and GLOBE. Journal of International Business Studies, 42(3), 436-445.
Brock, D. M., Shenkar, O., Shoham, A., & Siscovick, I. C. 2008. National culture and expatriate deployment. Journal of International Business Studies, 39(8), 1293-1309.
Brodbeck, F. C., Hanges, P. J., Dickson, M. W., Gupta, V., & Dorfman, P. W. 2004. Societal culture and industrial sector influences on organizational culture. Culture, leadership, and organizations: The GLOBE study of, 62, 654-668.
Brodbeck, F. C., & Frese, M. 2007. Societal culture and leadership in Germany. Culture and leadership across the world: The GLOBE book of in-depth studies of, 25, 147-214.
Brooks, C. 2014. Introductory Economics for Finance. 3rd edition. Cambridge University Press, Cambridge.
Burke, R. J., & McKeen, C. A. 1996. Do women at the top make a difference? Gender proportions and the experiences of managerial and professional women. Human Relations, 49(8), 1093-1104.
Campbell, K., & Mínguez-Vera, A. 2007. Gender diversity in the boardroom and firm financial performance. Journal of business ethics, 83(3), 435-451.
Carter, D. A., Simkins, B. J., & Simpson, W. G. 2003. Corporate governance, board diversity, and firm value. Financial review, 38(1), 33-53.
Carter, D. A., D'Souza, F., Simkins, B. J., & Simpson, W. G. 2010. The gender and ethnic diversity of US boards and board committees and firm financial performance. Corporate Governance: An International Review, 18(5), 396-414.
Chung, K. H., & Pruitt, S. W. 1994. A simple approximation of Tobin's q. Financial management, 70-74.
Cox, T. H., & Blake, S. 1991. Managing cultural diversity: Implications for organizational competitiveness. The Executive, 45-56.
Dawson, J. F. 2014. Moderation in management research: What, why, when, and how. Journal of Business and Psychology, 29(1), 1-19.
Eagly, A. 1987. Sex differences in social behavior: A social-role interpretation, Lawrance Erlbaum Associates. Hillsdale, NJ.
Elkhuizen, L., Hermes, N., Jacobs, J., & Meesters, A. 2017. Financial development, financial liberalization and social capital. Applied Economics, 1-21.
Ely, R. J. 1990. The Role of Men in Relationships among Professional Women at Work. In Academy of Management Proceedings (Vol. 1990, No. 1, pp. 364-368). Academy of Management.
Estrin, S., Baghdasaryan, D., & Meyer, K. E. 2009. The impact of institutional and human resource distance on international entry strategies. Journal of Management Studies, 46(7), 1171-1196.
Fama, E. F., & Jensen, M. C. 1983. Separation of ownership and control. The journal of law and Economics, 26(2), 301-325.
Finegold, D., Benson, G. S., & Hecht, D. 2007. Corporate boards and company performance: Review of research in light of recent reforms. Corporate Governance: an international review, 15(5), 865-878. Francoeur, C., Labelle, R., & Sinclair-Desgagné, B. 2008. Gender diversity in corporate governance and top management. Journal of business ethics, 81(1), 83-95.
Gelfand, M. J., Bhawuk, D. P., Nishii, L. H., & Bechtold, D. J. 2004. Individualism and collectivism. Culture, leadership, and organizations: The GLOBE study of, 62, 437-512.
Goodstein, J., Gautam, K., & Boeker, W. 1994. The effects of board size and diversity on strategic change. Strategic management journal, 15(3), 241-250.
Grant, J. 1988. Women as managers: What they can offer to organizations. Organizational Dynamics, 16(3), 56-63.
Hausmann, R., Tyson, L. D., Bekhouche, Y., & Zahidi, S. 2013. The global gender gap index 2012. In World Economic Forum.
Hermalin, B. E., & Weisbach, M. S. 2003. Boards of directors as an endogenously determined institution: A survey of the economic literature (No. w8161). National Bureau of Economic Research. Hillman, A. J., Cannella, A. A., & Paetzold, R. L. 2000. The resource dependence role of corporate directors: Strategic adaptation of board composition in response to environmental change. Journal of Management studies, 37(2), 235-256.
Hillman, A. J., Cannella Jr, A. A., & Harris, I. C. 2002. Women and racial minorities in the boardroom: How do directors differ?. Journal of management, 28(6), 747-763.
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.
Hofstede, G. 1983. National cultures in four dimensions: A research-based theory of cultural differences among nations. International Studies of Management & Organization, 13(1-2), 46-74.
Hofstede, G. 2003. Culture's consequences: Comparing values, behaviors, institutions and organizations across nations. Sage publications.
Hofstede, G. 2006. What did GLOBE really measure? Researchers’ minds versus respondents’ minds. Journal of international business studies, 37(6), 882-896.
House, R. J., Hanges, P. J., Javidan, M., Dorfman, P. W., & Gupta, V. (Eds.). 2004. Culture, leadership, and organizations: The GLOBE study of 62 societies. Sage publications.
Huse, M., & Rindova, V. P. 2001. Stakeholders' expectations of board roles: The case of subsidiary boards. Journal of Management and Governance, 5(2), 153-178.
Hutten, E. 2014. The influence of leverage on firm performance: A corporate governance perspective. Bachelor's thesis, University of Twente.
Javidan, M., House, R. J., Dorfman, P. W., Hanges, P. J., & De Luque, M. S. 2006. Conceptualizing and measuring cultures and their consequences: a comparative review of GLOBE's and Hofstede's approaches. Journal of international business studies, 37(6), 897-914.
Jensen, M. C. 1986. Agency costs of free cash flow, corporate finance, and takeovers. The American economic review, 76(2), 323-329.
Jensen, M. C., & Meckling, W. H. 1976. Theory of the firm: Managerial behavior, agency costs and ownership structure. Journal of financial economics, 3(4), 305-360.
Joecks, J., Pull, K., & Vetter, K. 2013. Gender diversity in the boardroom and firm performance: What exactly constitutes a “critical mass?”. Journal of business ethics, 118(1), 61-72.
