Board gender diversity and CSR performance: A French study

Abstract

The current paper examines how gender diversity influences the corporate social responsibility (CSR) performance of French companies. Specifically, we investigate the impact of certain attributes of female directors, such as their age, education, and nationality, on CSR performance. To analyze this, we utilize the generalized method of moments (GMM) on a sample of 53 French firms listed on the SBF 120 index over the period 2008–2017. The findings reveal a positive association between the presence of women on the board and CSR performance. Furthermore, we discover that the age of female directors has a significant and positive effect on CSR performance. Additionally, the existence of a CSR committee is found to have a significant and positive influence on CSR. However, our study does not find any correlation between the education and nationality of female directors and CSR performance.

Keywords:

• gender diversity
• age diversity
• nationality diversity
• education diversity
• CSR performance

1. Introduction

In a highly competitive landscape, companies are compelled to confront social and environmental challenges as significant obstacles. The trust of key stakeholders such as employees, consumers, and citizens are more readily gained by socially responsible companies. Consequently, senior management places a particular emphasis on addressing these concerns (Madime & Gonçalves, 2022; Michelon et al., 2013). To this end, governance mechanisms, particularly the composition of the board of directors, play a crucial role in ensuring the quality of extra-financial information and in matters of corporate social responsibility (CSR) regarding social and environmental issues (Kassinis & Vafeas, 2002; Maon et al., 2009). This CSR practice would be influenced by the characteristics and attributes of directors. Indeed, the upper echelon theory states that the knowledge of directors added to their values and experiences impact the quality of the disclosure of CSR information (Hambrick & Mason, 1984). Also, a diversity of gender, training, nationality, and experience of directors improves the performance of the company (Hillman et al., 2000).

Several empirical studies have examined the relationship between governance and corporate social responsibility (CSR), particularly in developed countries (Crifo & Rebérioux, 2015; Fordham & Robinson, 2018; García-Sánchez & Martínez-Ferrero, 2016).However, there is a relatively limited number of studies that specifically focus on the impact of board diversity on CSR (Hoang et al., 2018; Ibrahim & Hanefah, 2016). Consequently, the influence of factors such as gender diversity (Khan et al., 2019), age (Beji et al., 2020; Katmon et al., 2019), nationality (Katmon et al., 2019; Khan et al., 2019), and training (Beji et al., 2020; Harjoto et al., 2015), on the social and environmental behaviour of companies has been analyzed. The findings indicate a lack of consensus and mixed results, failing to define an optimal board composition in terms of these four axes of diversity (gender, age, nationality, and training) that would effectively enhance CSR practices. Moreover, studies that have examined the link between gender diversity and CSR practice have essentially analyzed the effect of the presence of female gender in Boards but have not looked into the impact of female executives’ attributes on CSR policy. Therefore, this issue requires further research. And to the extent that female attributes could have a relevant impact on CSR performance, we decided to analyze the following question research: Do female executives’ attributes influence CSR performance?

Due to its role in the decision-making process, the board of directors is described by the literature as the main group responsible for CSR policies (Kruger, 2009). And considering that the diversity of board members is reflected by their different characteristics, specificities, skills, and that these individual differences can have an impact on the quality of the decision-making process and its effectiveness (Carter et al., 2010), a better understanding of the relationship between board characteristics and corporate social responsibility is still of major interest. Thus, the main objective of this research is to understand how female directors behave towards social and environmental dimensions by considering their characteristics and attributes, namely age, nationality, and education. Therefore, the primary aim of this study is to gain insight into the behaviors of female executives concerning social and environmental aspects, taking into account their specific characteristics and attributes.

This study makes several key contributions. Firstly, it goes beyond analyzing the impact of gender diversity on CSR and enriches the CSR literature by specifically focusing on the attributes of female executives, an aspect that remains relatively less exploited in the literature and which the scope of results could optimize the CSR policies of companies. Secondly, as the governance literature has shown the usefulness of the existence of specialized committees in companies, our research also examines the influence of the presence of a CSR committee on companies’ commitment to social and environmental issues. Thirdly, the study investigates how the attributes of female executives affect two dimensions of CSR: the social dimension and the environmental dimension. Finally, the French context serves as an advantageous empirical framework due to existing laws, such as the Copé-Zimmerman law, which imposes quotas for women on corporate boards. This study delves deeper into this context by evaluating the impact of female directors’ attributes beyond the mere question of quota representation on the board of directors.

The generalized method of moments (GMM) in a dynamic panel was applied to a sample of 53 French companies listed on the SBF 120 index, during the period from 2008 to 2017. According to the results, the presence of female executives on the board is positively linked to CSR performance. The age variable shows a positive and significant effect. We also note that the existence of a CSR committee has a positive and significant effect on CSR performance. However, there is no significant correlation observed between the educational background and nationality of female executives, and the overall score reflecting the quality of CSR performance.

The rest of this article is organized as follows. The second section presents a review of the literature with the formulation of hypotheses. The third section presents the characteristics of the sample and the methodology. The results are discussed in a fourth section, and finally the conclusion.

2. Background

Over the last years, social and political expectations have gradually become structured and institutionalized around issues of equality, diversity and of combating discrimination. Their fields of application are numerous, and one of the more active focuses on the governance of companies and organizations.

Nowadays, it is well known that public authorities, investors, civil society, exert many pressures on companies to introduce more diversity in their governance bodies; the gender criterion is the most used, now subject to binding regulations in various countries, in the meaning of strengthening the presence of women in management and governance bodies.

In France, the law of 2011¹ (Copé-Zimmermann law) on parity on the board of directors applies to public limited companies and limited partnerships by shares listed on the stock exchange and provides quantified quotas: before the beginning of the year 2014, the boards of directors and supervisory boards had to have a minimum of 20% of women among their members and 40% within six years after the establishment in application of the law. The situation has progressed in recent years but the diversity remains contrasted from one company to another and the objectives set by the law rarely achieved: according to the European Women on Boards network’s 2016 survey of 600 companies² representing to 12 countries, between 2011 and 2015, the proportion of women on the Boards of European companies almost doubled, rising from 13.9% to 25% on average.

After doubling of the ratio of women in their governance bodies, it was then possible to wonder about the consequences of the presence of women in boards on others dimensions as firm performance, firm market value, or CSR. The impact of gender diversity on firm performance or company market value has been the subject of several studies and showed different results concerning the sign of the impact (Martın-Ugedo et al., 2019; Simkins & Simpson, 2003; Terjesen et al., 2015). However, the effect of gender diversity on CSR performance, which empirical studies are relatively less abundant, is still relevant especially for the French market which lends itself to this empirical examination given that the regulatory framework in this country imposes the presence of a minimum ratio of women in boards. Moreover, studies that have examined the link between gender diversity and CSR practice have essentially analyzed the effect of the presence of female gender in Boards but have not looked into the impact of the attributes of these female directors on CSR policy. This issue therefore requires further research.

If a positive effect of the women on boards on CSR performance is confirmed, gender diversity on boards will gain interest for the literature of governance in general, and more precisely for companies concerned about the effectiveness of their CSR policy.

3. Theoretical literature review

In general, the relationship of gender diversity with business performance can be explained through different theories in several academic fields such as social psychology and management (Maniyalath & Narendran, 2016). In management, researchers appeal to theories related to corporate governance, such as agency and resource dependence theories (Joecks et al., 2013).

Moreover, and in connection with the theme we are analyzing, the concept of board diversity would align with the framework of the upper echelon theory. According to this theory, the demographic characteristics of directors (such as age, education, values, etc.) influence their decision-making (Hambrick & Mason, 1984). This theory is based on two fundamental premises. Firstly, leaders’ decisions are driven by their personal interpretations of previous strategic situations they have encountered. Secondly, these interpretations are shaped by the leaders’ individual personalities, values, and experiences. In essence, the cognitive and behavioral models of leaders play a pivotal role in shaping corporate strategies.

Hafsi and Turgut (2013) and Boulouta (2013) believe that demographic characteristics of directors could influence the company’s strategy. Specifically, age, gender, nationality, and education could shape the decision-making process. The more the cognitive bases of the leaders are diversified, the more the management team is predisposed to meet the requirements of stakeholders (Autissier & Ben Lahouel, 2014). Governance practice then significantly influences the decisions of a company to respond to different issues. And as an internal mechanism, the board of directors impacts decisions relating to companies’ CSR strategy (Kassinis & Vafeas, 2002; Maon et al., 2009).

4. Empirical literature review and hypotheses development

4.1. Gender diversity and CSR

Contrary to their male counterparts who focus more on economic and shareholder objectives (Adams et al., 2011), female directors tend to take more into consideration a wider range of stakeholders. Upper echelon theory argues that the representation of women on boards impacts the cognitive input of directors (Byron & Post, 2016). In this sense, the decisions taken by the board of directors in terms of social responsibility depend on the gender representation in its composition.

Many studies have shown that women prioritize CSR issues (Alonso-Almeida et al., 2015; Hur et al., 2016 Female directors were consistently more caring and inclusive than their male counterparts in improving environmental performance (Adams & Funk, 2012). Thus, a significant proportion of women in boards implies better disclosure of CSR information (Shu & Chiang, 2020; Yaseen et al., 2019). However, other evidence reports a negative link between the two previous elements (Bruna et al., 2021).

H1:

The presence of female directors influences CSR performance.

4.2. Presence of foreign directors and CSR

Foreign directors are strongly attached to transparency, responsibility, and the reputation of the company in the market (Oxelheim & Randey, 2003). The knowledge, skills and values accumulated by an individual derive from their professional background and education (Becker, 1964). The combination of these values and the skills of foreign directors enable the company to achieve its objectives. Thus, a board made up of foreign directors would be a more efficient, independent board with a higher level of control and monitoring (Zaid et al., 2020).

The attributes of foreign directors allow boards to effectively manage and solve complex major problems and achieve superior results, compared to companies with more homogeneous boards of directors (Adams et al., 2015). Consequently, this diversity of foreign directors within the Board gives it a better ability to decide on the strategic orientations relating to CSR. In terms of empirical findings, certain studies indicate a positive influence of multinational management teams on CSR, attributing it to their global exposure, diverse experiences, skills, and knowledge (Khan et al., 2019). Conversely, other studies suggest a negative correlation (Katmon et al., 2019), or even a neutral effect (Barako & Brown, 2008) on CSR.

H2:

The diversity of nationality of female directors influences CSR performance.

4.3. Education diversity and CSR

The diversity of directors’ education is an asset for the company. This diversity promotes the ability of directors to exploit data, process complex information, and deal with the uncertainty involved in engaging in international operations (Hsu et al., 2013). The knowledge and the professional and personal qualifications of directors affect their choices as well as their strategic decisions. These accumulated assets stem from their academic backgrounds (Becker, 1964). Previous research suggests that academic background can have a strong effect on business outcomes (Finkelstein et al., 2009).

The theoretical perspective of the upper echelon theory can be mobilized to examine the relationship between the diversity of education of directors and the level of commitment of the company in the disclosure of CSR information. Also, the Resource-Based Theory (RBT) states that the diversity of the nature of the directors’ education gives rise to alternative ideas concerning the strategic decisions of the company such as the disclosure of CSR (Katmon et al., 2019). Studies that have analyzed the impact of the diversity of education on CSR are generally limited (Lewis et al., 2014). Adopting a qualitative approach, Ho (2005) notes the importance of directors’ qualifications on CSR commitments.

H3:

The nature of the education of female directors influences CSR performance.

4.4. Age diversity and CSR

Age diversity is a governance variable that impacts the social and environmental performance of companies. On the one hand, age is seen as an asset for the board of directors and part of human capital (Sonnenfeld, 2002). Its diversity can enhance the experiences, resources, and knowledge of board members. Age diversity helps dissipate skills and experiences from older directors to younger directors who could later contribute to critical decision-making (Ali et al., 2014). On the other hand, this diversity can create cognitive conflicts and prevent group cohesion and therefore harm the performance of the company (Talavera et al., 2018).

Regarding the empirical experiences related to age diversity, the existing results do not show a consensus. While Goergen et al. (2015) find a positive association between age diversity and board effectiveness and performance, Roitto (2013) reports a negative relationship between the average age of directors and CSR disclosure, while Katmon et al. (2019) find a neutral effect between the two elements. Other studies argue that age diversity leads to more balanced decision-making and therefore influences CSR performance (Beji et al., 2020; Ferrero-Ferrero et al., 2013).

H4:

The age diversity of female directors influences CSR performance.

5. Research design

5.1. Sample selection and data sources

The sample includes companies that are part of the SBF 120 index for the period between 2008 and 2017 (10 years). This market index accounts for three quarters of the market capitalization of the Paris marketplace. We excluded financial institutions, real estate companies and insurance corporations because of their specificities. We also eliminated organizations whose financial and governance data were unavailable. As a result, our final sample consists of 53 companies and 530 annual observations. Companies are from nine different sectors of activity according to the sector classification used which is of the ICB “Industry Classification Benchmark”. Details concerning the selection of the sample are summarized in Tables 1 and 2.

Table 1. Selection of the sample

Description	Numbers
Initial sample	120
Eliminated financial institutions, real estate companies and insurance firms	21
Excluded firms with incomplete governance and/or financial data	46
Final sample	53
Firm year observation (i.e., 53*10 years)	530

Table 2. Sample composition by industry

ICB industry classification	Number of firms	Percentage
Consumer goods	11	20%
Basic materials	2	3,77%
Consumer services	10	18,87%
Technology	4	7,55%
Industrials	19	35,85%
Oil and Gas	1	1,89%
Health care	3	5,66%
Utilities	2	3,77%
Telecommunications	1	1,89%
Total	53	100%

The CSR scores are taken from the Thomson Reuters Asset4 database. The financial data are extracted from the Thomson one Banker database. The Information describing the composition and characteristics of the boards of directors are collected from a documentary study by exploiting the annual reports and the reference documents of companies available on their official websites.

5.2. Model and measurement of variables

The dependent and independent variables of our research model, which is quantitative, are described below.

5.2.1. Dependent variable: CSR performance

This variable shows the overall score relating to the disclosure of CSR information out of 100 points. Each dimension is evaluated between 0 for companies that disclose no information and 100 for companies who disclose all of Thomson Reuters Asset4 criteria.

5.2.2. Independent variables

Gender diversity: Three variables are used to calculate the number of women on board.

First, the proportion of female directors is computed as the ratio of the number of women to the total number of directors. Second, it is determined by the number of female board members. Finally, we employ the Blau index (1977). This index is calculated based on the following formula:

$\mathrm{D}=1-\sum _{\mathrm{i}=1}^{\mathrm{N}}{\mathrm{P}}_{\mathrm{i}}^2$

P = share of individuals in a category.

N = category number.

When directors are precisely balanced between these two categories (50% females and 50% males), the maximum and minimum values of this variable are 0.5 and 0 respectively (there are only men or women).

5.2.2.1. Age diversity

It is divided into five age groups: less than 40-years old, 40 to 49, 50 to 59, 60 to 69, and 70- years old and older. When women directors are precisely balanced throughout the five categories, this score has a maximum value of 0.8.

5.2.2.2. Education diversity

Falls into five categories: Management (MBA or similar), Technical (Schools of engineering or institutes of technology), Administration (ENA, Sciences Po, IEP Paris etc.), Law (ENM, Faculty of Law), and other (diplomas not included in the previous categories). When women administrators are precisely balanced throughout the five categories, this score has a maximum value of 0.8.

5.2.2.3. Nationality diversity

It is calculated by categorizing nationality into two categories: women others nationality and women French nationality. When women directors are precisely balanced between the two categories (50% of women of French nationality, 50% of women of foreign nationality), this score has a maximum value of 0.5.

5.2.3. Control variables

We introduce a set of variables related to the composition of the board, the existence of a CSR committee and the specific characteristics of the company that can influence on CSR performance. To take into account the potential effects of board characteristics, we introduce variables such as female CEO, board size, duality of functions, and board independence. With regard to firm-specific characteristics, we include the following variables: firm size, level of debt, and return on assets (ROA). Finally, we introduced two binary variables (years and industry) into the regression model. The definition of diversity variables and control variables is presented in the table below:

5.3. Specification model

We offer the following empirical model to estimate the effect of female executives’ attributes on CSR performance.

${CSRP}_{i,t}={\alpha }_0+{\beta }_1{CSRP}_{t-1}{+\beta }_2{\mathrm{G}\mathrm{e}\mathrm{n}\mathrm{d}\mathrm{e}\mathrm{r}}_{it}+{\beta }_3{\mathrm{A}\mathrm{G}\mathrm{E}}_{it}+{\beta }_4{\mathrm{E}\mathrm{D}\mathrm{U}\mathrm{C}}_{it}+{\beta }_5{\mathrm{F}\mathrm{O}\mathrm{R}\mathrm{\_}\mathrm{N}\mathrm{A}\mathrm{T}}_{it}+{\beta }_6{CSRC}_{it}+{\beta }_7{\mathrm{C}\mathrm{E}\mathrm{O}\mathrm{F}}_{it}+{\beta }_8\mathrm{I}\mathrm{n}\mathrm{d}\mathrm{e}\mathrm{p}+{\beta }_9{\mathrm{D}\mathrm{u}\mathrm{a}\mathrm{l}\mathrm{i}\mathrm{t}\mathrm{y}}_{it}+{\beta }_{10}{\mathrm{B}\mathrm{S}}_{it}+{\beta }_{11}{\mathrm{R}\mathrm{O}\mathrm{A}}_{it}+{\beta }_{12}{Size}_{it}+{\beta }_{13}{LEV}_{i,t}+\sum _{t=1}^9{\beta }_{t}industry+\sum _{t=1}^{2008-2017}{\beta }_{t}year+{\epsilon }_{i,t}$

A problem of endogeneity between CSR performance and the representation of female on board is raised by our model. Indeed, this endogeneity can have an impact both on the motivation of women to join the board and on the motivation of the board to recruit women (Adams and Ferreira, 2009). Traditional econometric methods, such as OLS (least squares method) or fixed effects or random effects model, cannot effectively solve this endogeneity problem, as pointed out by Martín-Ugedo and Mínguez-Vera (2014). We use the method of generalized moments on panel data (GMM) proposed by Arellano and Bond (1991) and developed by Arellano and Bover (1995) to solve this difficulty. According to these authors, this method solves the problems of simultaneity bias, inverse causality (for example, between CSR performance and the representation of women on board), as well as the possibility of omitted variables. It is based on orthogonality conditions between lagged variables and error term.

Some authors have used the two-stage least squares (2SLS) method to consider endogeneity, but the choice of instruments is an issue (Campbell & Minguez-Vera, 2008). Thus, we use the GMM estimator in the first differences of Arellano and Bond (1991) and therefore estimate the model that variables appear for each period in first differences, which is likely to eliminate the specific individual effects.

To ensure the efficiency of the estimation by GMM method in a dynamic system, a series of tests are carried out: AR (2): the second-order autocorrelation test for errors; and the Hansen test (J test), which validates lagged variables as instruments.

6. Empirical results and discussion

6.1. Descriptive results

Table 4 presents descriptive statistics of variables used in our study. The average CSR score of our sample is 82.25 with a maximum of 89.59 and a minimum of 4.27. The average value of the Blau index is 0.17 which is low. This result is similar to the study of Kahloul et al. (2022). Furthermore, results also show that 11% of board members are female. Only 3% of women chair the board. As for the attributes of female administrators, according to five identified categories: Management (MBA or similar), Technical (Engineering schools or technology schools) Administration (ENA, Sciences Po, etc.), Lawyer and others, the average Blau index is around 0.14 showing low heterogeneity in the profiles. The Blau age index varies between 0 and 0.8. Nevertheless, the average of this index is relatively low (0.12). We also observe an index of the diversity of nationality of female directors (according to two categories: French nationality and other nationalities) is very low (0.06). Table 3 also indicates that the percentage of companies with a CSR committee is 30%. Concerning the composition of the board, its size is composed of an average of 13 members, 53% are independent members. Our sample shows that 34% of companies combine the functions of President and CEO. Finally, with regard to the other control variables, the average level of return on assets is 5%, while the level of indebtedness is around 30%.

Table 3. Definitions of variables

Variable	Signification	Measure
Dependent variable
CSR performance	CSRP	Total score, calculated as the yearly average of environmental score and social score
Independent variables
Number of females on the board	NFB	Number of females on the board of directors
Proportion of female directors	PFD	Number of female directors divided by total number of directors on the board
Board gender diversity	BGD	In the formula in Blau (1977), pi represents two categories the proportion of men and the proportion of women in the board. $D=1-{\sum }_{i=1}^N{P}_i^2$
Age of female directors	AGE	Blau index of board age as $1-{\sum }_{i=1}^N{P}_i^2$ Where pi is the proportion of female directors in categories i (less than 40-years old, 40 to 49,50 to 59, 60 to 69, and 70- years old and older)
Educational diversity of female	EDUC	Blau index of education diversity as$1-{\sum }_{i=1}^N{P}_i^2$ Where pi represents four categories: Management (MBA or similar), Technical (Schools of engineering or institutes of technology), Administration (ENA, Sciences Po, IEP Paris etc.), Law (ENM, Faculty of Law), and other (diplomas not included in the previous categories)
Proportion of female foreign directors	FOR_NAT	Blau index of female foreign directors as$1-{\sum }_{i=1}^N{P}_i^2$ pi represents two categories i the proportion of women of foreign nationality and the proportion of women of French nationality
Control variables
Female CEO	CEO_F	Dummy variable equal to 1 when the CEO is female; 0 otherwise
CSR committee	CSRC	Dummy variable equals 1 if the company has a CSR committee and 0 otherwise
Board size	BS	Number of directors on the board
Board independence	Indep	Board independence Ratio of number of independent directors to total board size
CEO duality	Duality	Dummy variable that takes the value of 1 if the CEO is also the chair of the board; 0 otherwise
Leverage	LEV	Ratio of total debt to total assets
Return on assets	ROA	Net income divided by assets
Firm Size	Size	Natural logarithm of the total assets

Table 4. Descriptive statistics

Variables	Mean	Median	Maximum	Minimum	Standard deviation
CSRP	82,25	89,52	96,59	4,27	18,37
PFD	0,11	0,08	0,55	0	0,13
NFB	1,52	1	9	0	1,74
BGD	0,17	0,14	0,5	0	0,16
AGE	0,12	0	0,80	0	0,24
EDUC	0,14	0	0,80	0	0,25
FOR_NAT	0,06	0	0,5	0	0,16
CEO_F	0,03	0	1	0	0,17
CSRC	0,30	0	1	0	0,46
BS	13,02	13,00	21	5	3,09
Duality	0,34	0	1	0	0,47
Indep	0,53	0,53	1	0	0,19
ROA	0,05	0,04	0,43	−0,20	0,05
LEV	0,30	0,25	2,45	0,000	0,27
Size	10,09	10,16	11,45	8,84	0,59

Table 5 presents the correlation matrix between all variables. Kervin (1992) indicates that a correlation greater than 0.7 is generally considered to be a problem of multicollinearity. We note a strong correlation between the Blau index of gender and the Blau index of education (0.84), between the Blau index of gender and the Blau index of age (0.80), and between the Blau index of age and the Blau index education (0.81). In order to estimate the regression model, we estimate these variables separately.

Table 5. Correlation matrix

Variables	1	2	3	4	5	6	7	8	9	10	11	12	13	14	15
1.CSRP	1.00
2.BGD	0,12***	1.00
3.NFB	0,17***	0,95***	1.00
4.PFD	0,13***	0,98***	0,96***
5.AGE	0,07	0,8***	0,84***	0,8***	1.00
6.EDUC	0,11**	0,84***	0,86***	0,84***	0,81***	1.00
7.FOR_NAT	0,19***	0,63***	0,70***	0,64***	0,62***	0,68***	1.0
8.CEO_F	−0,14***	0,28***	0,28***	0,28***	0,28**	0,27***	0,29***	1.00
9.Duality	−0,03	0,08*	0,19***	0,13***	0,32***	0,35***	0,23***	−0,06	1.00
10.CSRC	0,19***	0,20***	0,26***	0,22***	0,24***	0,18***	0,11***	−0,09**	0,14***	1.00
11.Indep	0,11	0,27***	0,09**	0,2***	−0,06	−0,01	0,05	0,03	−0,30	0,02	1.00
12.LEV	−0,2***	−0,04	−0,07	−0,07	−0,11	−0,09**	−0,02	−0,01	−0,05	0,09**	0,15**	1.00
13.ROA	−0,15***	−0,04	−0,08*	−0,06	−0,06	−0,08*	−0,11**	−0,13***	−0,06	0,02	0,14***	0,13***	1.00
14.Size	0,55***	0,11**	0,18***	0,09*	0,09*	0,16***	0,24***	−0,07*	−0,03	0,15***	−0,01	−0,11**	−0,32***	1.00
15.BS	0,38***	0,10**	0,23***	0,07*	0,18***	0,16***	0,20***	−0,01	0,01	0,23***	−0,23***	−0,04	−0,18***	0,63***	1.00

*, **, *** represent the significance level of 10%, 5% and 1%, respectively.

6.2. Regression results

Table 6 presents the results of the study. Gender diversity is positively and significantly linked to CSR performance, as expected (models 1, 2, and 3). This positive relationship could be attributed to several factors, including the fact that female directors exhibit more prosocial business behavior than men do (Kabongo et al., 2011; Williams, 2003), encourage ethical behavior (Flynn & Adams, 2004), and take into account the requirements of a broader range of stakeholders than only male directors (Kramer & Konrad, 2008. Female directors are also more ecologically aware and protective of the environment than men are (Braun, 2010). Thus, a better balance of representation between female and male on the board of directors promotes the company’s commitment to CSR actions and investment in sustainable development strategies. This diversity of the board can bring diverse CSR perspectives in various approaches and improve the complexity of problem-solving and the quality of decision-making relating to CSR. Our results corroborate those of Oino and Liu (2022), Beji et al. (2020) and Francoeur et al. (2017) Moreover, the estimated coefficient of female CEO variable is negatively and significantly associated with CSR performance. This result does not confirm the conclusions of Aabo and Giorici (2022), Lim and Chung (2021) who find that female CEOs are more favorable to decisions relating to CSR investments. In terms of age diversity, the results reveal a positive and significant relationship with the CSR performance. The finding is consistent with upper echelon theory (Hambrick & Mason, 1984) and means that older female administrators demonstrate superior moral reasoning and that they have a lot of work experience and wisdom in making decisions in terms of CSR. Our results are consistent with those of Beji et al. (2020), and Ferrero-Ferrero et al. (2013) who argue that an age-diverse board of directors leads to more balanced decision-making which improves CSR performance.

Table 6. Relation between CSR and gender (Arellano–Bond GMM dynamic panel data)

Variables	(1)	(2)	(3)	(4)	(5)	(6)
Lag CSRP	0,14*** (6,43)	0,14*** (6,21)	0,14*** (6,65)	0,15*** (6,41)	0,15*** (6,52)	0,16*** (6,33)
PFD			9,74*** (4,41)
BGD		8,33*** (3,64)
NFB	0,60*** (4,16)
AGE				1,62** (2,32)
EDUC					−0,76 (−0,85)
FOR_NAT						−0,79 (−0,82)
CEO_F	−4,23*** (−2,98)	−4,11*** (−3,08)	−4,36*** (−3,07)	−2,99*** (−3,06)	−2,63*** (−2,79)	−2,66*** (−2,99)
CSRC	3,16*** (5,96)	3,19*** (5,84)	3,07*** (5,95)	3,71*** (7,25)	3,92*** (7,19)	3,86*** (7,50)
BS	0,32*** (2,99)	0,35*** (3,47)	0,38*** (3,87)	0,38*** (3,48)	0,36*** (3,45)	0,37*** (3,47)
Indep	−2,28 (−1,03)	−2,99 (−1,29)	−2,52 (−1,16)	−1,03 (−0,47)	−0,24 (−0,11)	−0,57 (−0,27)
Duality	−3,57*** (−4,49)	−3,25*** (−3,69)	−3,54*** (−4,36)	−3,25*** (−3,87)	−3,01*** (−3,64)	−3,11*** (−3,93)
LEV	−3,64** (−2,13)	−3,01* (−1,72)	−2,51 (−1,16)	−4,15** (−2,23)	−3,75* (−1,85)	−3,34* (−1,97)
ROA	17,62*** (7,01)	18,05*** (7,28)	17,74*** (7,56)	17,11*** (6,98)	16,45*** (6,24)	17,53*** (6,36)
Size	21,21*** (12,87)	20,97*** (12,89)	20,80*** (13,24)	22,31*** (14,98)	23,12*** (14,05)	23,13*** (13,35)
Country FEs	Yes	Yes	Yes	Yes	Yes	Yes
Year EFs	Yes	Yes	Yes	Yes	Yes	Yes
Observations	424	424	424	424	424	424
AR (1)	−3,05 0,002	−3,03 0,002	−3,03 0,002	−3,08 0,002	−3,07 0,0002	−3,07 0,002
AR (2)	2,81 0,005	2,83 0,004	2,80 0,005	2,81 0,005	2,78 0,01	2,82 0,005
Hansen test (Chi-square, p-value)	43,58 (0,15)	43,11 (0,16)	43,43 (0,15)	44,87 (0,14)	43,04 0,16	42,65 (0,17)

*, **, *** represent the significance level of 10%, 5% and 1%, respectively.

Concerning the nationality of female directors, we noted the absence of significant relationships with CSR performance. It appears that the presence of foreign female directors, in the context we analyzed, does not provide new resources and different perspectives to develop CSR performance (skills, and experiences, political connections, access to networks). The finding is not consistent with studies that prove a significant effect of foreign female directors (Beji et al., 2020; Dardour et al., 2018). As for the variable of the diversity of schools at the origin of the education path of female administrators, our results do not reveal any significant links to CSR performance. This result is consistent with certain empirical studies (Dardour et al., 2018; Khan et al., 2019) and can be explained by the strong accumulation of knowledge relating to CSR, in particular after the introduction of NRE law in France in 2001 (The law on New Economic Regulations) and the European directives in favor of CSR in the early 2000s. Female administrators are made aware of these issues independently of their training.

In terms of control variables, we discover that having a CSR committee has a positive and significant effect on CSR performance. Evidence suggests that French firms which have a CSR committee outperform companies that do not have one. This positive relationship could be attributed to the functioning of a CSR committee which is efficient and respects the CSR transparency practices (Liao et al., 2014). The findings are consistent with those of Baraibar-Diez and Odriozola (2019), and Velte and Stawinoga (2020). The variable board size is also positively associated with CSR performance, implying that a bigger board size can improve CSR performance. At this level, we corroborate the study of Nguyen et al. (2021), and Lin and Nguyen (2022). A large size might bring a wider diversity of perspectives, opinions, and experiences around CSR issues. Our results suggest that French companies with larger boards tend to be efficient towards their CSR policy compared to smaller counterparts. Furthermore, our results confirm a negative relationship between the duality of functions and CSR performance. This result rather corroborates those of previous studies (Beji et al., 2020; Naciti, 2019). The findings further support prior research indicating that company size has a favorable and significant effect on CSR performance (Beji et al., 2020; Bruna et al., 2021; Chebbi et al., 2018; Dardour et al., 2018; Nguyen et al., 2023; Tran & Nguyen, 2022). The larger its size the more the company identifies and discloses many CSR-related issues. The level of indebtedness is negatively and significantly correlated with CSR performance, which is consistent with the conclusions of Oino and Liu (2022). Finally, the positive and significant link between the return on assets (ROA) and the quality of CSR performance suggested that profitable companies are more likely to devote resources to CSR investment (Nguyen et al., 2023; Oino & Liu, 2022).

6.3. Robustness test

Bond (2002) recommends testing the robustness of results, which can differ depending on the methodology utilized (the fixed effects method). Table 7 shows that the presence of female directors has a positive and significant effect. The table also demonstrates a negative and significant effect of female CEO variables on CSR performance. It can be seen from Table 7 that in general the results of all models yielded similar estimations proofing the robustness of the study.

Table 7. Panel fixed effects regression on board diversity and CSR disclosure

Variables	(1)	(2)	(3)	(4)	(5)	(6)
Constant	−338,13*** (−7,23)	−336,41*** (−7,17)	−336,38*** (−7,22)	−350,11*** (−7,52)	−353,84*** (−7,58)	−353,6*** (−7,58)
PFD			14,87** (2,33)
BGD		12,30** (1,98)
NFB	0,92** (1,94)
AGE				2,44 (0,71)
EDUC					0,42 (0,11)
FOR_NAT						0,68 (0,14)
CEO_F	−11,01*** (−3,17)	−11,21*** (−3,20)	−11,34*** (−3,29)	−9,05*** (−2,71)	−8,55** (−2,47)	−8,61** (−2,45)
CSRC	5,09*** (3,89)	5,21*** (4,04)	5,98*** (3,83))	5,59*** (4,31)	5,79*** (4,52)	5,79*** (4,57)
BS	1,86*** (4,14)	1,88*** (4,21)	1,91*** (4,32)	1,97*** (4,42)	1,99*** (4,50)	1,99*** (4,47)
Indep	12,56* (1,85)	11,33 (1,62)	11,85* (1,76)	15,07** (2,26)	15,59** (2,33)	15,64** (2,36)
Duality	−3,76 (−1,59)	−3,54 (−1,15)	−3,77 (−1,61)	−3,26 (−1,37)	−2,97 (−1,24)	−2,96 (−1,26)
LEV	−25,78*** (−5,77)	−25,45*** (−5,69)	−25,45*** (−5,70)	−26,12*** (−5,83)	−26,29*** (−5,86)	−26,27*** (−5,84)
ROA	−8,57 (−0,65)	−7,34 (−0,57)	−7,76 (−0,59)	−6,86 (−0,52)	−6,00 (−0,46)	−6,01 (−0,45)
Size	39,25*** (8,29)	39,02*** (8,21)	38,99*** (8,25)	40,23*** (8,50)	40,54*** (8,55)	40,52*** (8,55)
Country FEs	Yes	Yes	Yes	Yes	Yes	Yes
Year EFs	Yes	Yes	Yes	Yes	Yes	Yes
Observations	530	530	530	530	530	530
Prob. > F	0,00	0,00	0,00	0,00	0,00	0,00
R2	0,75	0,75	0,75	0,75	0,74	0,74

*, **, *** represent the significance level of 10%, 5% and 1%, respectively.

7. Summary and conclusion

This paper studies the impact of gender diversity on CSR performance. It specifically examines the features of female directors on the CSR score. The study is conducted on a sample of 53 companies listed on the SBF 120 index over a period from 2008 to 2017. First, the findings indicate a positive and significant relationship between the presence of female and CSR performance. This result suggests that the higher the proportion of female directors on the board, the higher score of CSR. Thus, we can conclude that companies with more female on board are more responsible socially, and that female directors play an important role in CSR practice. This finding is consistent with previous studies (Beji et al., 2020; Boulouta, 2013; Francoeur et al., 2017; Oino & Liu, 2022). Second, a positive and significant relationship between the age diversity of female directors and CSR performance is shown. This result is consistent with Beji et al. (2020), Ferrero-Ferrero et al. (2013), and Third, the diversity of education and the diversity of nationality of female directors have no effect on CSR performance. Finally, a positive and significant effect of the existence of a CSR committee is observed. This result is consistent with Baraibar-Diez and Odriozola (2019) and Velte and Stawinoga (2020). These findings are robust across various econometric estimations with panel data and dynamic GMM. This mitigation of potential endogeneity or reverse causality affecting previous results is provided.

From a managerial point of view, our study provides a better understanding of mechanisms affecting CSR performance and how female executives’ attributes improve its development. Our study refutes the business case of gender diversity. Indeed, our results lead to a positive relationship between the representation of female on board and CSR performance. Companies that have a particular interest in being seen as socially responsible, should incorporate female members in their boards.

This study has limitations that could be explored as future research directions. First, the sample size is small. Further studies using the CAC All Tradable index are needed to generalize our results. Second, the study exclusively looks at French companies. However, other countries may have different cultures or different institutional contexts. Our study might be reproduced by integrating other countries to confirm or refute our results. Indeed, Boulouta (2013) points out that the relationship between gender diversity and CSR performance depends on cultural gender stereotypes. This could help in observing the impact of country characteristics on gender diversity for example inflation, legal system, and political factors. Thirdly, the study primarily concerned with CSR. Future research would include new CSR measurements, such as social score, environmental score or ESG score.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Notes

1. Law n°2011–103 of January 27, 2011 (known as Copé-Zimmermann law).

2. Those taken into account in the STOXX 600 stock market index.