Board Gender Quota, Female Labor Market
Outcomes, and Firm Performance in France
Abstract:
This paper investigates the downstream effects of mandatory gender quota in France of SBF 120 firms for the period 2010-2018. Specifically, by analyzing the effect of quota on director characteristics, committee membership, proportion of female employees, and firm performance. The results demonstrate that subsequent the quota women were at least as, arguably even more qualified and experienced than their predecessors. The quota did not reduce the gender-gap regarding earnings. Furthermore, the quota positively affects the number of average committees a female director is a member of, the converse relationship is found for male directors. The proportion of female employees is also positively affected by the quota, thus improving the labor market outcomes for women. The quota has a neutral effect on firm performance for the sample as a whole, a positive effect for firms that are most affected by the quota, and a negative effect for least affected firms.
Key Words: Board gender quota, female directors, committees, female employees, firm performance
Study Program: MSc International Financial Management Name: Laurens Bergsma
2 1. Introduction
Despite the increasing focus on gender equality over the last decades, women remain heavily underrepresented in high-status occupations (Bertrand et al., 2019). As a result, governments and regulators have dedicated substantial efforts towards achieving gender equality within corporate boards. One particular way has been the implementation of regulations such as gender quota. Such quotas obligate corporations to increase the female representation within the board of directors. Countries such as Denmark, the Netherlands, and Australia implemented a quota with a ‘comply or explain’ method of disclosure (Terjesen et al., 2014). The first European country to implement a mandatory quota was Norway (2003), followed by countries such as Spain (2007), Iceland (2010), and France (2011) (see Table 1 in Appendix A).
Following these legislations, several studies have investigated the effect of quota on individual director characteristics. Ahern and Ditmar (2012) find that female directors subsequent the quota were younger and less experienced. Bertrand et al. (2019) found that in Norway, after the implementation of a gender quota, female directors were more qualified than their predecessors. Moreover, that the gender-gap regarding earnings decreased substantially within boards. This is in line with Boyallian et al. (2018), who argue that salaries of female directors could increase due to a strengthened bargaining position. Ferreira et al. (2017) find that director-matching in France improved due to the quota, and consequently, so did the tenure of female directors. Choudhury (2015) argues that the gender quota will only benefit a select number of female directors who will be appointed multiple corporate boards.
3 (Adams and Ferreira, 2009). Bertrand et al. (2019) argue that an increase in female representation in corporate boards could improve opportunities for women throughout the firm. Firstly, due to increased awareness of qualified women that can be appointed to top management positions. Secondly, the direct actions of female directors, such as the recommendation of female candidates. Thirdly, the development of policies beneficial for women (e.g., stringent pay policies, paid maternity leave) could improve attractiveness to work at a specific organization. Based on the arguments mentioned above, this paper will focus on the following research question: How does the gender quota in France affect the labor market outcomes for women?
4 The empirical analysis is conducted using a sample of 99 SBF 120 firms, which are listed on the Euronext Paris stock exchange. The results of the study show that female directors were more experienced, and at least as, perhaps even more qualified than their predecessors. There were sufficient female directors for organizations to meet the quota threshold. This paper shows, however, that the quota failed to reduce the gender gap in earnings. Furthermore, a reduction in the distance to the quota threshold (from 0.4 to 0) would increase the average number of committees per female director with 0.27. While simultaneously decreasing the average number of committees per male director with 0.20. The board gender quota, therefore, has a significant and positive (negative) effect on the labor market outcomes for female (male) directors. Additionally, this paper finds that the board gender quota positively and significantly affects the proportion of female employees in a firm, and hence the labor market outcomes for these women. The results demonstrate that a reduction (from 0.4 to 0) would increase the share of female employees in a firm with 6.98%. Lastly, when investigating the effect of the board gender quota on firm performance, the results show find a neutral effect.
5 organizations as women contribute to the quality of governance, and could attract other talented women. Finally, this paper can be used to further develop research in the field of gender quota legislation.
This study contributes to the limited literature on gender quota, as it is one of few that focuses on the downstream effects of board gender quota measured as labor market outcomes for women. Additionally, where previous literature focused on countries with an immediate implementation of a mandatory gender quota, this paper investigates this relationship in France, a country that opted for a gradual implementation of a mandatory gender quota.
The relevance of this paper regarding the aspects international, financial, and management flows forth from the consequences of implementing gender quotas. Mandatory quotas are generally applicable to publicly listed firms, with a substantial amount of employees and high revenues. Therefore most often large international organizations headquartered or listed in countries are subject to such quota. For these international firms, it determines how they manage board composition. Consequently, it has implications for how such boards, due to their composition, manage the organization as a whole. Hence, it also determines the financial performance of the organization. Moreover, determining whether such quotas are effective could have implications for legislators on an international level, and consequently, the firms that reside in countries of these legislators.
6 2. Literature Review
2.1 Gender Quota
7 2.2 Gender Quota in France
8
was a three-year transition period, after which each gender should at least make up 20% of the board composition. At the beginning of the year 2017, companies were obligated to have a board in which no gender was represented by less than 40%. The sanctions, in case of non-compliance, are twofold. Firstly, non-compliant organizations were only allowed to execute changes that positively contributed towards complying with the gender quota. Secondly, compensation for board members in the form of attendance could be suspended until the organization met the quota. The law applies to listed companies that trade on regulated markets, and non-listed companies that either have revenues or assets exceeding 50 million euros or that employ at least 500 employees (Law n° 2011-103 January 27th 2011).
2.3 Hypotheses development
2.3.1 Board gender quota and director characteristics
9
quota. If there is a lack of female directors, as argued by Choudchury (2015), the number of boards that female directors are appointed to will increase. If there are sufficient female directors, this number will decrease. Additionally, by including the nationality of the directors, this paper documents whether female directors are sourced locally or internationally.
Ahern and Ditmar (2012) find that the implementation of gender quota in Norway led to younger and less experienced female directors. Whereas Bertrand et al. (2019) find that female directors, after the implementation of the gender quota in Norway were more qualified than female directors before the implementation. If there are experienced and qualified female directors, the differences regarding the characteristics age, as a proxy for experience (Ahern and Ditmar, 2012), and education (qualifications and MBAs), as a proxy for how qualified female directors are (Bertrand et al., 2019), would decrease.
Ferreira et al. (2017) investigate whether the quota decreases the turnover rates for female directors and find that in France, director matching induced by the quota are more stable than before the quota. Therefore the tenure (time in role) of female directors should increase.
10 Hypothesis 1:
The board gender quota will positively affect the reduction of differences in director characteristics observed between female and male directors.
2.3.2 Board gender quota and labor market outcomes for board directors
The board of directors is a key corporate governance system as it is ultimately responsible for the performance and the success of the firm (Bhagat and Bolton, 2013). According to Green and Homroy (2018), proposals to increase the proportion of female directors are premised upon the notion to improve governance. Moreover, boards execute most of their responsibilities through committees, which implies that female directors improve the governance of a firm through committees. In line with Adams and Ferreira (2009), who find that female directors are more likely to be appointed to monitoring committees and have greater monitoring power. They also find that female directors have better attendance records and that a more gender-diverse board improves the attendance records for male directors.
11
This paper argues that an increase in gender diversity through the board gender quota will increase the number of committees that female directors join. Specifically, as the governance of an organization improves by assigning female directors to committees, due to the improved monitoring and independence. Additionally, as female directors hold more positions within committees, men are more likely to lose positions. Consequently, this paper investigates the following hypotheses:
Hypothesis 2:
A reduction of the distance to the board gender quota threshold positively affects the number of committees in which female directors participate.
Hypothesis 3:
A reduction of the distance to the board gender quota threshold negatively affects the number of committees in which male directors participate.
2.3.3 Board gender quota and labor market outcomes for women within the firm
12
management positions. Due to an increased awareness of qualified women, encountered during the search for female board directors. Secondly, female directors can take direct action to improve the outcomes for other women in the firm. By being more favorably inclined towards and recommending more female candidates for top management positions. Thirdly, female directors can be more outspoken regarding the adaptation of human resource policies that favor other female employees. Examples can include policies such as tighter controls on pay, or flexible work options for women with children. As a consequence of these policies, employment at these firms could become more attractive and lead to an increase in the share of employees that are women. Based on the previous arguments, this paper tests the hypothesis:
Hypothesis 4:
A reduction of the distance to the board gender quota threshold will positively affect the share of female employees within the firm.
2.3.4 Board gender quota and firm performance
Several studies investigated the contribution of female directors to the board of an organization. Matsa and Miller (2013) argue that women improve the quality of the decision-making process. Female directors are more independent (Bart and McQueen, 2013), and improve monitoring (Adams and Ferreira, 2009), which should improve corporate governance and hence firm performance (Green and Homroy, 2018). Women adopt a more long-term view (Rosener, 1990), and provide different resources and strategic inputs (Bilimoria, 2000).
13
directors appointments, show a robust and positive effect of female board representation on firm performance. Dezso and Ross (2002) find that female representation improves firm performance for firms that are innovation-focused. Specifically, as the information and social benefits of gender diversity are important for managerial task performance.
In contrast, Lee and James (2007) find that investors react significantly more negative towards the appointment of women, thereby decreasing firm value. Adams and Ferreira (2009) demonstrate that the average effect of gender diversity, so the appointment of more women, is negative on firm performance. Ryan and Haslam (2005) reveal that during a period of stock-market decline, companies that appointed female directors experienced relative worse performance for a period of five months in comparison to firms that appointed male directors. Moreover, the initially positive effect of female directors, the improved quality of governance, can also decrease shareholder value due to excessive monitoring (Almazan and Suarez, 2003; Adams and Ferreira, 2007).
Studies also find a neutral relationship between gender diversity and firm performance. Farrell and Hersch (2005) do not find significant abnormal returns due to the announcement of female directors being added to the board. Francoeur et al. (2008) find that female directors have no significant effect on the performance of a firm. Lastly, Chapple and Humphrey (2014) do not observe evidence of a relationship between the gender diversity of a board and firm performance.
14 Hypothesis 5:
A reduction of the distance to the board gender quota threshold will positively affect the operating profit of a firm.
3. Methodology
3.1 Data
The primary database used for the analyses is BoardEx, accessed through Wharton Research Data Services. Provding information on individual director’s characteristics on a year by year basis related to gender, age, nationality, tenure, salary, and education. The data is collected for a nine-year period, specifically for the period between the 1st of January 2010, which is approximately one year before the law was enacted, and the 31st of December 2018, which is two years after the gender quota was enforced. The board gender quota affects the largest publicly listed and non-listed firms. Data on non-listed firms is not widely available, therefore this paper focuses on publicly listed firms. Specifically, the constituents of the SBF 120 or Société des Bourses Françaises 120 Index, which is a French stock market index based on the 120 most actively traded stocks listed in Paris. It includes all firms from the CAC 40 and CAC Next 20 indexes and an additional 60 stocks listed on the Euronext Paris. The sample utilized to perform the univariate analysis consists of a balanced panel of 99 firms with 12282 director-year observations.
15 examine the effect of the board gender quota on female labor market outcomes. Similarly, Thomson Reuters’ Datastream also provides the data on the operating profit of a firm, to examine the effect of the board gender quota on firm performance. Additionally, data for firm-level control variables is drawn from Thomson Reuters’ Datastream and Worldscope. The final sample used to perform the multivariate analyses consists of a balanced panel of 99 firms with 891 firm-year observations.
3.2 Descriptive statistics and correlation
16
Table 1
Descriptive statistics.
Variables N Mean Std. Dev. Minimum Maximum
Director Characteristics Age 12,282 58.13 9.47 34 81 Nationality 12,269 0.76 0.43 0 1 Time in Role 12,282 4.89 4.44 0 25.1 Ln_Salary 4972 4.66 1.31 1.95 8.49 Number of Boards 12,282 5.02 4.99 1 33 Qualifications 12,282 1.94 1.23 0 6 MBA 12,282 0.12 0.32 0 1 Downstream effects Quota 891 0.12 0.11 0 0.4 CommitteesWomen 891 1.07 0.66 0 4.4 CommitteesMen 891 1.09 0.52 0 4.33 FemaleEmployees (%) 725 37.49 16.94 9.7 83 OperatingProfit (%) 891 12.55 13.35 -58.64 69.98 AdjustedLeast 891 0.54 0.49 0 1 ROA (%) 891 4.56 5.04 -49.6 37.61 Ln_Assets 891 8.25 1.30 4.96 11.30 Leverage (ratio) 891 0.417 0.207 0.003 0.941 CapEx (%) 891 4.15 3.84 0 36.2
Note: See Appendix A, table 2 for variable definitions. The sample is winsorized at the 1% level.
female employees within a firm has a mean value of 37.49 percent and varies between 9.7 and 83.0 percent.
17 quota and the proportion of female employees within the firm. A small negative correlation exists between the variables Quota and FemaleEmployees, which takes a value of -0.06. Lastly, table 3 demonstrates the correlation matrix regarding the regression of the board gender quota and firm performance. The variable Quota and OperatingProfit have a correlation of -0.04. There are no other large correlations with any other variables.
3.3 Research Methods
In order to examine the differences between female and male directors within boards, this paper employs a univariate analysis in the form of a mean comparison t-test, also known as the Student’s t-test. It can be used to test whether the mean value of characteristics for female and male directors are significantly different from one another. Thus, to establish that significant differences between female and male directors exist. Moreover, to test whether the mean value of characteristics is significantly different before and after the enforcement of the board gender quota, so to investigate whether the board gender quota improved the conditions for female directors. There are two econometric issues regarding the t-test. Firstly, the sample is not random, which is due to data availability. The sample consists of the largest and most traded publicly listed firms on the Euronext Paris stock exchange, which reduces the generalizability and, thus, the external validity. The second issue arises as the characteristics do not all follow a normal distribution, which could lead to type I and type II errors. However, the latter concern is mostly resolved due to the size of the sample, which is rather large.
18 effect of gender diversity, this paper can test the hypotheses as to what the effect of the board gender quota is on the dependent variables. By employing panel regressions that include firm-level control variables, year and industry fixed effects, this paper can reduce the likelihood of omitted variable bias. One assumption of the multivariate analysis is that the variables follow a normal distribution. Where possible, this paper utilizes the natural logarithm of variables to create a more normally distributed sample. Additionally, to reduce the effect of spurious outliers, the sample is winsorized at the 1% level. Furthermore, variables are tested for multicollinearity. There are no major correlations between variables, therefore multicollinearity is not a concern. All variables are tested for heteroscedasticity, and although not all variables are perfectly homoscedastic, there are no real obvious signs of heteroscedasticity. To ensure that heteroscedasticity will not pose an issue, and to ensure robust estimates, this paper utilizes robust standard errors in all multivariate analyses.
3.3.1 Board gender quota and director characteristics
19
3.3.2 Board gender quota and committee membership
This section is aiming to test the second and third hypotheses of the paper. It examines whether the board gender quota influences the likelihood that a female director is a member of a committee. Similarly, the effect of the board gender quota for male directors. To distinguish the effect of the quota from the effect of gender ratio, the variable quota is measured as follows:
𝑄𝑢𝑜𝑡𝑎𝑖𝑡= 0.4 − 𝐺𝑒𝑛𝑑𝑒𝑟𝑅𝑎𝑡𝑖𝑜𝑖𝑡 (1)
Where Quota is the absolute percentage from the quota threshold of 0.4, GenderRatio is the proportion of women in a board as a percentage, i is a given firm, and t is a given year. After computing Quota, all negative values, meaning above the threshold, are set equal to zero. Specifically, as the quota has no more effect after the proportion of board directors that are female has met the prescribed minimum of forty percent.
The average membership of committees within a single firm in a given year is calculated as:
𝐶𝑜𝑚𝑚𝑖𝑡𝑡𝑒𝑒𝑠𝑥𝑡 = 𝑁𝐶𝑥𝑡/𝑁𝑥𝑡 (2)
Where Committees is the average number of committees per director, NC is the total number of committees, where N is the total number of directors, x refers to a group which is either women or men, and t is a specific year.
The regression that is utilized to test the hypothesis whether the board gender quota affects the number of committees for female directors is a multivariate regression and is the following:
𝐶𝑜𝑚𝑚𝑖𝑡𝑡𝑒𝑒𝑠𝑥𝑖𝑡= 𝛽0+ 𝛽1𝑄𝑢𝑜𝑡𝑎𝑖𝑡+ 𝛽2𝑅𝑂𝐴𝑖𝑡+ 𝛽3𝐿𝑛𝐴𝑠𝑠𝑒𝑡𝑠𝑖𝑡+ 𝛽4𝐿𝑒𝑣𝑒𝑟𝑎𝑔𝑒𝑖𝑡+ 𝛽5𝐶𝑎𝑝𝑒𝑥𝑖𝑡+
20 Where Committees is the average number of committees, where 𝛽 is the coefficient for the cohering variable, x is referring to the group of either men or women, i is referring to a firm observation, t is a given year, 𝜂 is the industry fixed effects, 𝜇 is the year fixed effects, and 𝜖 is the error term.
3.3.3 Board gender quota and proportion of female employees within the firm
This section is aimed to analyze the effect of board gender quota on the proportion of female employees within the firm. The following multivariate regression is used:
𝐹𝑒𝑚𝑎𝑙𝑒𝐸𝑚𝑝𝑙𝑜𝑦𝑒𝑒𝑠𝑖𝑡= 𝛽0+ 𝛽1𝑄𝑢𝑜𝑡𝑎𝑖𝑡+ 𝛽2𝑅𝑂𝐴𝑖𝑡+ 𝛽3𝐿𝑛𝐴𝑠𝑠𝑒𝑡𝑠𝑖𝑡+ 𝛽4𝐿𝑒𝑣𝑒𝑟𝑎𝑔𝑒𝑖𝑡+
𝛽5𝐶𝑎𝑝𝑒𝑥𝑖𝑡+ 𝜂𝑖𝑡+ 𝜇𝑡+ 𝜖𝑖𝑡 (4)
Where FemaleEmployees is the proportion of female employees, where 𝛽 is the coefficient for the cohering variable, i is referring to a firm observation, t is a given year, 𝜂 is the industry fixed effects, 𝜇 is the year fixed effects, and 𝜖 is the error term.
3.3.4 Board gender quota and firm performance
To test the final hypothesis, whether the board gender quota affects the operating profit of a firm, the following multivariate regression is used:
𝑂𝑝𝑒𝑟𝑎𝑡𝑖𝑛𝑔𝑃𝑟𝑜𝑓𝑖𝑡𝑖𝑡= 𝛽0+ 𝛽1𝑄𝑢𝑜𝑡𝑎𝑖𝑡+ 𝛽2𝐿𝑛𝐴𝑠𝑠𝑒𝑡𝑠𝑖𝑡+ 𝛽3𝐿𝑒𝑣𝑒𝑟𝑎𝑔𝑒𝑖𝑡+ 𝛽4𝐶𝑎𝑝𝑒𝑥𝑖𝑡+ 𝜂𝑖𝑡+ 𝜇𝑡+
𝜖𝑖𝑡 (5)
21
3.3.5 Endogeneity and robustness tests
One form through which endogeneity could arise is due to simultaneity bias between the variable Quota and FemaleEmployees. The proportion of female directors in the corporate board could determine the share of female employees in the firm, and vice versa. As it is more likely that a firm with a higher percentage of female employees would also have a larger proportion of female directors. This could bias the observed coefficient estimates. In order to test whether the results that will be obtained from performing the regression of equation (4) are robust, another regression is performed that includes one year lagged values of the variable Quota. Thereby ensuring that the observed relationship is not affected by simultaneity bias. The multivariate regression to do so is as follows:
𝐹𝑒𝑚𝑎𝑙𝑒𝐸𝑚𝑝𝑙𝑜𝑦𝑒𝑒𝑠𝑖𝑡= 𝛽0+ 𝛽1𝑄𝑢𝑜𝑡𝑎𝑖𝑡−1+ 𝛽2𝑅𝑂𝐴𝑖𝑡+ 𝛽3𝐿𝑛𝐴𝑠𝑠𝑒𝑡𝑠𝑖𝑡+ 𝛽4𝐿𝑒𝑣𝑒𝑟𝑎𝑔𝑒𝑖𝑡+
𝛽5𝐶𝑎𝑝𝑒𝑥𝑖𝑡+ 𝜂𝑖𝑡+ 𝜇𝑡+ 𝜖𝑖𝑡 (6)
Where FemaleEmployees is the proportion of female employees, where 𝛽 is the coefficient for the cohering variable, i is referring to a firm observation, t is a given year, 𝜂 is the industry fixed effects, 𝜇 is the year fixed effects, and 𝜖 is the error term.
22
𝑂𝑝𝑒𝑟𝑎𝑡𝑖𝑛𝑔𝑃𝑟𝑜𝑓𝑖𝑡𝑖𝑡= 𝛽0+ 𝛽1𝑄𝑢𝑜𝑡𝑎𝑖𝑡−1+ 𝛽2𝐿𝑛𝐴𝑠𝑠𝑒𝑡𝑠𝑖𝑡+ 𝛽3𝐿𝑒𝑣𝑒𝑟𝑎𝑔𝑒𝑖𝑡+ 𝛽4𝐶𝑎𝑝𝑒𝑥𝑖𝑡+ 𝜂𝑖𝑡+
𝜇𝑡+ 𝜖𝑖𝑡 (7)
Where OperatingProfit is the operating profit of a firm, where 𝛽 is the coefficient for the cohering variable, i is referring to a firm observation, t is a given year, 𝜂 is the industry fixed effects, 𝜇 is the year fixed effects, and 𝜖 is the error term.
Secondly, talented female directors, due to their scarcity, have the privilege to choose positions at better performing firms, hence firms that least have to adjust to the quota could already be the better performing firms (Farrell and Hersch, 2005). Therefore interpreting the relationship could pose an issue. In order to test whether there is a difference between the firms that least have to adjust in comparison to firms that most have to adjust, an additional regression is performed that includes an interaction term. Specifically, to distinguish the effect of the board gender quota on firm performance for firms that are least and most adjusted. Therefore the following multivariate regression is used:
𝑂𝑝𝑒𝑟𝑎𝑡𝑖𝑛𝑔𝑃𝑟𝑜𝑓𝑖𝑡𝑖𝑡= 𝛽0+ 𝛽1𝑄𝑢𝑜𝑡𝑎𝑖𝑡+ 𝛽2𝐿𝑒𝑎𝑠𝑡𝐴𝑑𝑗𝑢𝑠𝑡𝑒𝑑𝑖𝑡+ 𝛽3𝑄𝑢𝑜𝑡𝑎 ∗ 𝐿𝑒𝑎𝑠𝑡𝐴𝑑𝑗𝑢𝑠𝑡𝑒𝑑𝑖𝑡+ 𝛽4𝐿𝑛𝐴𝑠𝑠𝑒𝑡𝑠𝑖𝑡+ 𝛽5𝐿𝑒𝑣𝑒𝑟𝑎𝑔𝑒𝑖𝑡+ 𝛽6𝐶𝑎𝑝𝑒𝑥𝑖𝑡+ 𝜂𝑖𝑡+ 𝜇𝑡+ 𝜖𝑖𝑡 (8)
Where OperatingProfit is the operating profit of a firm, where 𝛽 is the coefficient for the cohering variable, i is referring to a firm observation, t is a given year, 𝜂 is the industry fixed effects, 𝜇 is the year fixed effects, and 𝜖 is the error term.
3.3.6 Foreign ownership
23 However, after an extensive search for foreign ownership variables for the sample, the data ended up noisy. In fact, the data was so distorted that even the construction of a dummy variable was not possible. Therefore this paper was forced to exclude this variable.
4. Results
4.1 Proportion of female directors within boards
Below in Figure 1 the proportion of female directors of the sample firms between the periods 2010 and 2018 are presented. As shown, the proportion increases fairly constant over time. The quota substantially increased the proportion of female directors, within nine years, the proportion of female directors increased from 12.2% in 2010 to 41.9% in 2019. Each vertical line in the figure represents a specific event. Firstly, in 2011 the enactment of the law, when the
Figure 1
Proportion of female directors in corporate boards for sample firms.
24 proportion of female directors was 16.9%. Secondly, the year 2014 in which the first threshold of 20% was instated. The proportion of female directors in 2014 was 29.8%. Lastly, when the second threshold of 40% was enforced in the year 2017 when the proportion of female directors is 41.4%. The graph of figure 1 suggests that the quota successfully increased the representation of female directors within boards of affected firms.
4.2 Board gender quota and director characteristics
26
Table 2
Director characteristics before and after the gender quota for female and male directors. Before 2017 After 2017 Diff. T stat.
Panel A: All Female Directors
Age 54.37 55.34 -0.97 -3.00 Nationality 0.75 0.71 0.04 2.50 Time in Role 3.64 3.75 -0.11 -0.87 Ln_Salary 4.16 4.24 -0.08 -2.03 Number of Boards 3.70 3.41 0.29 2.89 Qualifications 2.03 2.07 -0.04 -0.96 MBA 0.109 0.124 -0.015 -1.35
Panel B: All Male Directors
Age 59.46 59.94 -0.48 -1.89 Nationality 0.77 0.76 0.01 1.18 Time in Role 5.43 5.27 0.16 1.19 Ln_Salary 4.84 5.01 -0.17 -2.80 Number of Boards 5.67 5.49 0.18 1.16 Qualifications 1.92 1.85 0.07 1.97 MBA 0.113 0.135 -0.022 -2.56
Note: This table reports the differences in director characteristics for the period before and after the board gender quota. In panel A the results regarding the differences for board characteristics for female directors are presented, where in panel B the differences for board characteristics for male directors are presented. The sample is composed of firms listed on the Euronext Paris that are constituents of the SBF 120 as of December 2018. The sample period is 2010-2018. Cohering to each variable is the mean value for the given period of time. Diff. denotes the observed difference for the mean values. T stat. denotes the t-statistic of the difference between the mean values.
27
Table 3
Director characteristics between female and male directors before and after the gender quota.
Female Male Diff. T stat.
Panel A: Before 2017 Age 54.40 59.46 -5.06 -23.29 Nationality 0.75 0.77 -0.02 -2.55 Time in Role 3.65 5.42 -1.77 -16.99 Ln_Salary 4.17 4.84 -0.67 -14.37 Number of Boards 3.74 5.62 -1.88 16.11 Qualifications 2.01 1.92 0.09 3.26 MBA 0.109 0.113 -0.004 -0.50 Panel B: After 2017 Age 55.39 59.94 -4.55 -13.07 Nationality 0.71 0.76 -0.05 -3.03 Time in Role 3.77 5.26 -1.49 -9.36 Ln_Salary 4.25 5.01 -0.76 -10.48 Number of Boards 3.45 5.44 -1.99 -10.90 Qualifications 2.06 1.85 0.21 4.40 MBA 0.124 0.135 -0.011 -0.85
Note: This table reports the differences in director characteristics for female and male directors. In panel A the results regarding the differences in board characteristics between female and male directors before the enforcement of the board gender quota are presented, where in panel B the differences in board characteristics for female and male directors after the enforcement of the board gender quota are presented. The sample is composed of firms listed on the Euronext Paris that are constituents of the SBF 120 as of December 2018. The sample period is 2010-2018. Cohering to each variable is the mean value for the given gender. Diff. denotes the observed difference for the mean values. T stat. denotes the t-statistic of the difference between the mean values.
28 quota, this difference became more substantial, namely 18%. One explanation for this difference in earnings, could be that female directors, on average, are paid less for the same work. The results imply that the board gender quota has failed to address the issue of unequal pay between women and men. For the number of boards that directors are appointed to, the difference increased from 1.88 to 1.99. Although this statistic initially might seem negative, in this case it actually implies that the quota had a beneficial effect for a larger group of women. As more individual female directors were appointed to the board of directors, and not just a select number of female directors who are appointed to a larger number of boards. Regarding the qualifications that directors have acquired, the difference actually increased from 0.09 to 0.21, meaning the difference more than doubled. This difference is also more significant, with a t-statistic of 4.40. This would mean that based on this variable, female directors are actually more qualified than their male counterparts, both before and after the quota. Lastly, the difference regarding the variable MBA increased from 0.004 to 0.011, which would mean that based on this variable, male directors are more qualified than their female counterparts, however, the difference still remains not significant. In conclusion, based on the results in tables 2 and 3, enough experienced and qualified women were available to be appointed to boards for firms to meet the quota. Although it must be noted that a proportion of these newly appointed female directors were sourced internationally. This could suggest that although global supply is sufficient, local supply might not be. Additionally, the quota has failed to address the issue regarding the differences in salary for women and men. Moreover, the quota also increased the tenure for female directors.
4.3 Board gender quota and committee membership
29 CommitteesWomen is regressed on Quota. The estimated coefficient is -0.74, which is negative and significant at the 1% level. Meaning that when the distance to quota decreases, the number of committees that a female director is a member of increases.
In column 2, after including the control variables, the coefficient of the distance to quota is -0.67 and significant at the 5% level. Confirming the hypothesis that the board gender quota has a positive effect on the number of committees that female directors are a member of. Therefore the board gender quota positively affects the labor market outcomes for female directors. Specifically, as female directors on average are more likely to be appointed to a committee. The variable Ln_Assets demonstrates a positive relationship of 0.05, which is also significant at the 1% level. The results of columns 1 and 2 provide support to conclude that the board gender quota has a positive and significant effect on the number of committees in which female directors participate. The 𝑅2 of the regression in column 1 takes a value of 0.33, which indicates that the model is a moderately good fit. When including the firm-level control variables, the 𝑅2 increases to 0.34, indicating that the control variables add minimal explanatory power.
In column 3, regarding the effect of the Quota on the variable CommitteesMen results in a positive coefficient of 0.51, which is significant at the 5% level. Meaning that when the distance to the quota decreases, so does the number of committees that male directors are a member of.
30
Table 4
Board gender quota and committee membership. Variables Committees Women(1) Committees Women(2) Committees Men(3) Committees Men(4) Quota -0.7359*** (0.2854) -0.6738** (0.2869) 0.5057** (0.2402) 0.5115** (0.2384) ROA -0.0038 (0.0044) -0.0116*** (0.0036) Ln_Assets 0.0517*** (0.0193) 0.0492*** (0.0160) Leverage -0.1031 (0.0951) 0.0990 (0.0790) CapEx -0.0023 (0.0056) -0.0111** (0.0046)
Industry Yes Yes Yes Yes
Year Yes Yes Yes Yes
Constant 0.4217 -0.0317 0.5251 0.0884
𝑅2 0.3337 0.3419 0.2342 0.2626
N 891 891 891 891
Note: This table reports the coefficient estimates of equation (3). CommitteesWomen is the average number of committees that a female director is a member of. CommitteesMen is the average number of committees that a male director is a member of. Quota is the distance to the gender quota threshold of forty percent. The sample is composed of firms listed on the Euronext Paris that are constituents of the SBF 120 as of December 2018. The sample period is 2010-2018. In column (1) and (3) the regression is performed with only the independent variable, whereas in column (2) and (4) the control variables are included. The standard errors are reported in parentheses. ***, **, * denote the significance of the results at 1%, 5%, and 10% level.
31 good fit for the model. The results in columns 3 and 4, demonstrate that the board gender quota has a negative effect on the number of committees that male directors are a member of.
4.4 Board gender quota and proportion of female employees within the firm
In this section the results are presented regarding the effect of the board gender quota on the proportion of female employees within the firm. In Table 5 column 1, FemaleEmployees is regressed on Quota. The coefficient is negative, namely -17.44, and is significant at the 1% level. Column 1 provides initial evidence indicating that when the distance to the quota threshold decreases, the proportion of female employees within the firm increases. The 𝑅2 of 0.87 suggests the model is a good fit. However, a large proportion of the 𝑅2 is caused by the industry fixed effects, as the industry in which a firm operates is correlated with the proportion of female employees in the firm. For instance, in the sample, manufacturing firms have lower percentages of female employees, while firms that provide healthcare have higher percentages of female employees.
32
Table 5
Board gender quota and proportion of female employees.
Variables FemaleEmployees (1) FemaleEmployees (2)
Quota -17.4408*** (3.6525) -17.4616*** (4.1272) ROA 0.0206 (0.0618) Ln_Assets -0.3971 (0.2561) Leverage 0.2249 (1.1432) CapEx 0.1435** (0.0699)
Industry Yes Yes
Year Yes Yes
Constant 24.9072 28.3881
𝑅2 0.8714 0.8730
N 725 725
Note: This table reports the coefficient estimates of equation (4). FemaleEmployees is the proportion of female employees in a firm. Quota is the distance to the gender quota threshold of forty percent. The sample is composed of firms listed on the Euronext Paris that are constituents of the SBF 120 as of December 2018. The sample period is 2010-2018. In column (1) the regression is performed with only the independent variable, whereas in column (2) the control variables are included. The standard errors are reported in parentheses. ***, **, * denote the significance of the results at 1%, 5%, and 10% level.
conclude, based on the results of column 1 and 2, the hypothesis that the board gender quota increases the proportion of female employees within the firm is accepted.
4.5 Board gender quota and firm performance
33
Table 6
Board gender quota and operating profit.
Variables OperatingProfit(1) OperatingProfit(2)
Quota -5.3517 (5.3281) -4.9653 (5.2845) Ln_Assets 0.7116** (0.3564) Leverage 1.7024 (1.7580) CapEx 0.4914*** (0.1019)
Industry Yes Yes
Year Yes Yes
Constant 8.1829 -0.1247
𝑅2 0.4240 0.4407
N 891 891
Note: This table reports the coefficient estimates of equation (5). OperatingProfit is the operating profit of a firm as a ratio. Quota is the distance to the gender quota threshold of forty percent. The sample is composed of firms listed on the Euronext Paris that are constituents of the SBF 120 as of December 2018. The sample period is 2010-2018. In column (1) the regression is performed with only the independent variable, whereas in column (2) the control variables are included. The standard errors are reported in parentheses. ***, **, * denote the significance of the results at 1%, 5%, and 10% level.
directors’ contribution is too small for it to demonstrate significant results, or the contribution of female directors needs a longer period of time to translate into improved firm performance.
34 returns. The results in columns 1 and 2, demonstrate that the quota has neither a positive nor negative effect on the operating profit of a firm.
4.6 Robustness
4.6.1 Robustness lagged board gender quota and proportion of female employees
In order to test whether the baseline results are robust and endogeneity does not affect the estimated coefficient of the variable Quota on FemaleEmployees, a multivariate regression with lagged values for the variable Quota is employed. In table 7 the results regarding this
Table 7
Robustness test lagged board gender quota and proportion of female employees. Variables FemaleEmployees (1) FemaleEmployees (2)
𝑄𝑢𝑜𝑡𝑎𝑡−1 -22.5650*** (5.2263) -21.7752*** (5.2161) ROA 0.2826*** (0.0975) Ln_Assets 0.1314** (0.1474) Leverage -0.4895 (1.7089) CapEx 2.84e-06 (0.0002)
Industry Yes Yes
Year Yes Yes
Constant 26.7821 24.4863
𝑅2 0.7445 0.7483
N 651 651
35 multivariate analysis are presented. In column 1, FemaleEmployees is regressed on 𝑄𝑢𝑜𝑡𝑎𝑡−1. The coefficient is -22.57 and is significant at the 1% level. Meaning that when the distance to the quota decreases, the proportion of female employees increases.
In column 2, the same regression is performed, this time however, the control variables are included. The coefficient for the variable Quota remains negative, namely -21.78, and is significant at the 1% level. Again, confirming that when the distance to the quota threshold decreases, the proportion of female employees in a firm increases. The results remain similar to those of table 5, after including the lagged values for the variable Quota. However, the size of the estimated coefficient increases, which is likely due to the excluded data of the year 2010. The results in Table 7 provide evidence that the observed relationship in Table 5 is robust and significant. Meaning that a reduction in the distance to threshold quota increases the proportion of female employees in the firm.
4.6.2 Robustness lagged board gender quota and operating profit
Furthermore, in order to test whether the baseline results of table 6 are robust and endogeneity does not bias the estimated coefficient of the variable Quota, a multivariate regression that includes lagged values for the variable Quota is performed. In table 8 the results regarding this analysis are presented. In column 1 OperatingProfit is regressed on 𝑄𝑢𝑜𝑡𝑎𝑡−1, the estimated coefficient is -4.56 and is not significant.
36
Table 8
Robustness test lagged board gender quota and operating profit.
Variables OperatingProfit(1) OperatingProfit(2)
𝑄𝑢𝑜𝑡𝑎𝑡−1 -4.5605 (5.4123) -3.8111 (5.3930) Ln_Assets 0.7119* (0.3813) Leverage 2.1881 (1.8585) CapEx 0.4269*** (0.1066)
Industry Yes Yes
Year Yes Yes
Constant 8.1995 -0.7690
𝑅2 0.4311 0.4448
N 792 792
Note: This table reports the coefficient estimates of equation (7). OperatingProfit is the operating profit of a firm as a ratio. Quota is the distance to the gender quota threshold of forty percent. The sample is composed of firms listed on the Euronext Paris that are constituents of the SBF 120 as of December 2018. The sample period is 2011-2018. In column (1) the regression is performed with only the independent variable, whereas in column (2) the control variables are included. The standard errors are reported in parentheses. ***, **, * denote the significance of the results at 1%, 5%, and 10% level.
size, sign, and significance levels. Therefore the results of table 6 are robust, and overall the quota has no significant effect on firm performance.
4.6.3 Robustness differential response of firms to board gender quota
37
Table 9
Robustness test differential response of firms regarding board gender quota and operating profit.
Variables OperatingProfit(1) OperatingProfit(2)
Quota -5.3517 (5.3281) -9.8151* (5.8865) AdjustedLeast -2.2142* (1.1497) Quota * AdjustedLeast 18.9119*** (6.7456) Ln_Assets 0.6861* (0.3593) Leverage 1.7294 (1.7521) CapEx 0.4821*** (0.1025)
Industry Yes Yes
Year Yes Yes
Constant 8.1829 0.3923
𝑅2 0.4240 0.4459
N 891 891
Note: This table reports the coefficient estimates of equation (8). OperatingProfit is the operating profit of a firm as a ratio. Quota is the distance to the gender quota threshold of forty percent. The sample is composed of firms listed on the Euronext Paris that are constituents of the SBF 120 as of December 2018. The sample period is 2010-2018. In column (1) the regression is performed with only the independent variable, whereas in column (2) the control variables and interaction term are included. The standard errors are reported in parentheses. ***, **, * denote the significance of the results at 1%, 5%, and 10% level.
38 18.91 more than firms that had to adjust most. Therefor complying with the gender quota actually had a negative effect on the operating profit of the firms that least had to adjust. The results in column 2 suggest there is indeed a differential response between firms that had to least and most adjust as a consequence of the quota. Furthermore, as the result for firms that had to least adjust is actually negative, this implies that the threshold of the gender quota, for these firms was suboptimal.
5. Conclusion
On January 27th 2011, France enacted a mandatory quota for corporate boards, with a
39 average number of committees that female directors are a member of. Conversely, a reduction in the distance to the quota threshold significantly decreases the average number of committees that male directors are a member of. Therefore, the quota improves the labor market outcomes for female directors, as the likelihood of appointment to committees significantly increases. Adams and Ferreira (2009) also find that female directors are more likely to be appointed to committees, specifically monitoring committees.
The results this paper yields also demonstrate that a reduction in the distance to gender quota significantly increases the proportion of female employees in the firm. The proportion of female employees could have only increased if more women were appointed to management positions, or if women were on average hired more often. Hence, the quota improved the labor market outcomes for female employees in the firm. When controlling for simultaneity, the results remain significant, thus are robust.
Moreover, regarding the analysis of the quota on firm performance, no significant effect is found. Hence, the observed effect of the quota on firm performance is neutral, which is in line with the results of Francoeur et al. (2008). After controlling for simultaneity, the results remain similar, thus indicating the results are robust and that the effect for the overall sample is neutral. However, when including an interaction term that distinguishes between firms that had least to adjust subsequent to the enactment of the quota in comparison to firms that had to adjust most, the quota has a significant effect on firm performance. Firstly, for the group that adjusted most, a reduction in the distance to the quota threshold significantly improves the firm performance. Secondly, for the group that adjusted least, a reduction in the distance to the quota threshold significantly decreases firm performance. Indicating that for the latter group, the quota threshold was suboptimal.
40 female directors. The evidence suggests that enough qualified and experienced women are present to foster the increase of female representation on corporate boards. Moreover, that the quota leads to the appointment of female directors for other high-status positions, such as committees. Additionally, the labor market outcomes for women throughout the firms improved subsequent to the implementation of the quota. Thus country regulators can utilize gender quota as a tool, not only to improve female representation, but also to improve labor market outcomes for women on a broader scale. However, in order to address gender-gap earnings issues, country regulators must look beyond gender quota in order to achieve equality. Furthermore, the results regarding firm performance heed caution when determining a threshold. As organizations with higher gender diversity on the board experienced a negative effect on firm performance. Conversely, organizations with lower levels of gender diversity in the board, demonstrated an increase in firm performance. Suggesting the optimal threshold might be a compromise in between these two groups. Secondly, for corporate organizations the results demonstrate that enough qualified and experienced women are present. The inclusion of these women could be beneficial for organizations as women contribute to the quality of governance, and could attract other talented women to the organization . Finally, this paper can be used to further develop research in the field of gender quota legislations.
The paper has several limitations. Firstly, sample selection bias might pose an issue. As this paper focuses on only the SBF 120, thus publicly listed firms, the generalizability of this study could be limited. Secondly, the sample size is rather small, therefor analyses could overstate significance levels that could lead to type II errors. Thirdly, due to data availability constraints, variables that more comprehensively measure labor market outcomes, such as gender-gap in earnings for female employees in firms, are not included.
42 References
[1] Adams, R. B. (2016). Women on boards: The superheroes of tomorrow? The Leadership
Quarterly, vol. 27(3), pp. 371-386
[2] Adams, R. B. and Ferreira, D. (2004). Gender diversity in the boardroom. Unpublished Working Paper.
[3] Adams, R. B. and Ferreira, D. (2009). Women in the boardroom and their impact on governance and performance. Journal of Financial Economics, vol. 2, pp. 291-309
[4] Ahern, K. R. and Dittmar, A. K. (2012). The Changing of the Boards: The Impact on Firm Valuation of Mandated Female Board Representation. Quarterly Journal of Economics, vol. 127(1), pp. 137-197
[5] Almazan, A. and Suarez, J. (2003). Entrenchment and Severance Pay in Optimal Governance Structures. The Journal of Finance, vol. 58(2), pp. 519-547
[6] Association Francaise des Enterprises Privees, and the Mouvement des Enterprises de France. (2013). Code of corporate governance for listed companies, article 6.3
[7] Bart, C. and McQueen, G. (2013). Why women make better directors. International Journal
of Business Governance and Ethics, vol. 8(1), pp. 874-879
[8] Bertrand, M., Black, S. E., Jensen, S., and Lleras-Muney, A. (2019). Breaking the Glass Ceiling? The Effect of Board Quotas on Female Labour Market Outcomes in Norway. The
Review of Economic Studies, vol. 86(1), pp. 191-239
[9] Bhagat, S. and Bolton, B. (2013). Director Ownership, Governance, and Performance. The
43
[10] Bilimoria, D. (2000). Building the business case for women corporate directors. In R. J.
(eds.), Women on Corporate Boards of Directors (pp. 25-40). The Netherlands: Kluwer
Academic Publishers.
[11] Bondy, K., Matten, D., and Moon, J. (2004). The Adoption of Voluntary Codes of Conducts in MNCs: A Three-Country Comparative study. Business and Society Review, vol. 109(4), pp. 449-477
[12] Boyallin, P., Dasgupta, S., and Homroy, S. (2018). Supply and demand side determinants of board gender imbalance: the U.S. evidence. Unpublished Working Paper.
[13] Burgess, Z., and Tharenou, P. (2002). Women board directors: Characteristics of the few.
Journal of Business Ethics, vol. 37(1), pp. 39–49
[14] Carter, D. A., Simkins, B. J., and Simpson, G. W. (2003). Corporate Governance, Board Diversity, and Firm Value. The Financial Review, vol. 38(1), pp. 33-53
[15] Chapple, L., and Humphrey, J. E. (2014). Does board gender diversity have a financial impact? Evidence using stock portfolio performance. Journal of Business Ethics, vol. 122(4), pp. 709-723.
[16] Choudhury, B. (2015). Gender diversity on boards: Beyond quotas. European Business
Law Review, vol. 26(1), pp. 229-243
[17] Constitution de la Ve République, 1958
[18] Costa, D. L. and Kahn, M. E. (2003). Civic Engagement and Community Heterogeneity: An Economist’s Perspective. Perspectives on Politics, vol. 1(1), pp. 103-111
44
[20] Lépinard, E. (2015). The Adoption and Diffusion of Gender Quotas in France (1982-2014). European University Institute
[21] Engelstad , F. and Teigen , M. (2012). Firms, boards and gender quotas: Comparative perspective (Comparative social research, v29). Bingley, UK: Emerald
[22] European Parliament. (2015). The policy on gender equality in France. Directorate-General for internal policies.
[23] Farrell, K. A., and Hersch, P. L. (2005). Additions to corporate boards: the effect of gender.
Journal of Corporate Finance, vol. 11(1), pp. 85-106
[24] Ferreira, D., Ginglinger, E., Laguna, M. A., and Skalli, Y. (2017). Board Quotas and Director-Firm Matching. Unpublished Working Paper. European Corporate Governance Institute.
[25] Foust-Cummings , H . (2013) . Research and Considerations Regarding Women on Boards. In Machold , S. et al . (Eds.), Getting women on corporate boards: A snowball starting in
Norway. Massachusetts: Edward Elgar Publishing
[26] Francoeur, C., Labelle, R., and Sinclair-Desgagné, B. (2008). Gender diversity in corporate governance and top management. Journal of Business Ethics, vol. 81, pp. 83-95
[27] Glick, P. (2006). Ambivalent sexism, power distance, and gender inequality across cultures. In S. Guimond (Ed.), Social comparison and social psychology: Understanding
cognition, intergroup relations, and culture (pp. 283–302). New York: Cambridge University
Press.
45
[29] Green, C.P., and Homroy, S. (2018). Female Directors, Board Committees and Firm Performance. European Economic Review, vol. 102(2), pp. 19-38
[30] Guo, L. and Masulis, R.W. (2015). Board Structure and Monitoring: New evidence from CEO turnovers. Review of Financial Studies, vol. 28(10), pp. 2770-2811
[31] Hermalin, B. E. and Weisbach, M.S. (2003). Boards of Directors As An Endogenously Determined Institutions: A Survey Of The Economic Literature. Economic Policy Review, vol. 9(1), pp. 7-26
[32] Higgs, D. (2003). Review of the role and effectiveness of non-executive directors. London, United Kingdom: The Stationery Office.
[33] Hofstede, G. (2001). Cultures Consequences (2nd eds). Thousand Oaks: Sage.
[34] Hwang, S., Schivdasani, A., and Smintzi, E. (2019). Mandating Women on Boards: Evidence from California. Unpublished Working Paper. University of North Carolina.
[35] Jianakoplos, N. A. and Bernasek, A. (2007). Are women more risk averse? Economic
Inquiry, vol. 36(4), pp. 620-630
[36] Kirsch, A. (2018). The gender composition of corporate boards: A review and research agenda. The Leadership Quarterly, vol. 29(2), pp. 346-364
[37] Klettner, A., Clarke, T., and Boersema, M. (2014). Strategic and regulatory approaches to increasing women in leadership: Multi-level targets and mandatory quotas as levers for cultural change. Journal of Business Ethics, vol. 122(1), pp. 145-165
46
[39] Labelle, R., Francoeur, C., and Lakhal, F. (2015). To Regulate Or Not To Regulate? Early Evidence on the Means Used Around the World to Promote Gender Diversity in the Boardroom.
Journal of Gender, Work and Organization, vol. 22(4), pp. 339-363
[40] French Law n° 2011-103 January 27th 2011
[41] Lee, P. M., and James, E. H. (2007). She'-E-OS: Gender effects and investor reactions to the announcements of top executive appointments. Strategic Management Journal, vol. 28(3), pp. 227-241
[42] Leszczynska, M. (2018). Mandatory Quotas for Women on Boards of Directors in the European Union: Harmful to or Good for Company Performance? European Business
Organization Law Review, vol. 19(1), pp. 35-61
[43] Lewellyn, K. B. and Muller-Kahle, M. I. (2019). The Corporate Board Glass Ceiling: The Role of Empowerment and Culture in Shaping Board Gender Diversity. Journal of Business
Ethics.
[44] Luttmer, E. F. P. (2001). Group Loyalty and the Taste for Redistribution. Journal of
Political Economy, vol. 109 (3), pp. 500-528
[45] Matsa, D. A. and Miller, A. R. (2013). A Female Style in Corporate Leadership? Evidence from Quotas. American Economic Journal: Applied Economics, vol. 5(3), pp. 136-169
[46] Oakley, J. (2000). Gender-based barriers to senior management positions: Understanding the scarcity of female CEOs. Journal of Business Ethics, vol. 27(4), pp. 321–334.
[47] Oliveira, A. and Gondek, M. (2014). Women on Company Boards – An Example of Positive Action in Europe perspectives . Bingley : Emerald.
47
[49] Rosener, J. B. (1990). Ways women lead. Harvard Business Review, vol. 68(6), pp. 119-125
[50] Ross, S. A. (1973). Return, risk and arbitrage. Unpublished Working Paper. University of Pennsylvania
[51] Ryan, M. K., and Haslam, S. A. (2005). The glass cliff: Evidence that women are over represented in precarious leadership positions. British Journal of Management, vol. 16(2), pp. 81-90
[52] Sealy, R., Doldor, E., and Vinnicombe, S. (2009). Increasing diversity on public and private sector boards. Part 1: How diverse are boards and why? Cranfield School of Management Report commissioned by the UK Government Equalities Office, October (p. 64) [53] Seierstad, C., Gabaldon, P., and Mensi-Klarbach, H. (2017). Gender Diversity in the Boardroom Volume 2 Multiple Approaches Beyond Quotas. Switzerland: Palgrave Macmillan [54] Simpson, W. G., Carter, D., and D’Souza, P. F. (2010). What Do We Know About Women on Boards? Journal of Applied Finance, vol. 20(2), pp. 60-63
[55] Szydlo , M . (2015). Gender equality on the boards of EU companies: Between economic efficiency, fundamental rights and democratic legitimization of economic governance.
European Law Journal, vol. 21(1), pp. 97-115
[56] Terjesen, S., Aguilera, R., and Lorenz, R. (2014). Legislating a Woman’s Seat on the Board: Institutional Factors Driving Gender Quotas for Board Directors. Journal of Business
Ethics, vol. 50 (2), pp. 233-251
48
[58] Van der Vegt, G. S., Van de Vliert, E., and Huang, X. (2005). Location level links between diversity and innovative climate depend on national power distance. Academy of Management
49 Appendix A
Table 1
Countries with mandatory gender quotas on corporate boards.
Country Quota Passage date Year(s) in which to
comply
Israel At least 1 woman April 19, 1999 2010
Norway 40% December 19, 2003 2008 Spain 40% March 22, 2007 2015 Iceland 40% March 4, 2010 2013 France 40% January 13, 2011 2015-2017 Malaysia 30% June 27, 2011 2016 Italy 33% June 28, 2011 2012-2015 Belgium 33% June 30, 2011 2017-2019
India At least 1 woman August 30, 2013 2015-2020
Germany 30% March 6, 2015 2016
Note: This table is constructed based on information provided by Terjesen et al. (2014).
Legislating a Woman’s Seat on the Board: Institutional Factors Driving Gender Quotas for Board Directors; and the European Commission, 2016. Gender Balance on Corporate Boards - Europe is Cracking the Glass Ceiling. A Fact Sheet from Directorate-General for Justice and
50 Table 2
Variable description.
Variables Description Additional information
Quota The distance to the quota threshold, all values lower than zero are replaced by zero. 0.4 - % of female directors in the board
Age The age of a director.
Nationality The nationality of a director. 1 = French 0 = all else
Time in Role The number of years a director holds a position.
Ln_Salary The natural logarithm of the salary of a director. Salary is in 000s of dollars
NumberofBoards The number of listed and unlisted boards on which a director occupies a position.
Qualifications The number of education degrees a director has.
MBA Whether a director has a MBA. 1 = yes 0 = no
CommitteesWomen The average number of committees that a female director attends within a firm. Total no. of committees of female directors for a firm in a given year/ Total no. of female directors
CommitteesMen The average number of committees that a male director attends within a firm. Total no. of committees of male directors for a firm in a given year/ Total no. of male directors
Chairwomen The average number of chair positions that a female director attends within a firm. Total no. of chair positions of female directors for a firm in a given year/ Total no. of female directors
Chairmen The average number of chair positions that a male director attends within a firm. Total no. of chair positions of male directors for a firm in a given year/ Total no. of male directors
FemaleEmployees (%) The percentage of female employees within a firm.
OperatingProfit (%) The operating profit of a firm as a percentage of revenue.
AdjustedLeast Classification of firms that either had to adjust their composition of the board of directors more or less due to the board gender quota.
Organizations that in the year 2011 had a distance to the gender quota threshold of 0.25 or less take the value 1, otherwise 0
ROA (%) The ratio of return on assets.
Ln_Assets The natural logarithm of the total assets of a firm. Total assets is in 000s of dollars
Leverage (ratio) The debt of a firm as a ratio of total capital.
51 Appendix B
Table 1
Correlation matrix regarding the board gender quota and committee membership.
Table 2
Correlation matrix regarding the board gender quota and proportion of female employees.
52 Table 3
Correlation matrix regarding the board gender quota and firm performance.
53 Appendix C
Table 1
Board gender quota and chair positions.
Variables Chairwomen(1) Chairwomen(2) Chairmen(3) Chairmen(4)
Quota -0.3750*** (0.1049) -0.3858*** (0.1045) -0.0370 (0.0658) -0.0436 (0.0652) ROA 0.0040** (0.0016) -0.0033*** (0.0010) Ln_Assets -0.0181 (0.0070)*** 0.0055 (0.0044) Leverage 0.0372 (0.0346) 0.0016 (0.0216) CapEx 0.0050** (0.0020) -0.0049*** (0.0013)
Industry Yes Yes Yes Yes
Year Yes Yes Yes Yes
Constant 0.0918 0.2279 0.1965 0.1681
𝑅2 0.2564 0.2777 0.1486 0.1822
N 891 891 891 891
54 Table 2
Correlation matrix regarding the board gender quota and chair positions.
