The Moderating Role of Board Gender Diversity on Safety Performance - Firm Financial Performance Nexus: Evidence from JSE- Listed Firms

1. Introduction

The mining industry plays a vital role to country’s economic growth, development and decent labour (Dikgwathe & Mulenga, 2022; Anwar, Sari, Islamiya, Ahmad & Alias, 2024). Despite increasing regulatory, environmental and social pressures, South African mining industry continues to be a substantial contributor to gross domestic product (GDP) and national economic development (Cawood, Fenn, Grobler & McDougall, 2024). However, mining activity remains one of the most hazardous in workplace that expose workers to injuries and fatalities (Gruz-Ausejo, Cama-Ttito, Solano, Copez-Lonzoy & Vera-Ponce, 2024). Occupational injuries and fatalities can have detrimental impact on firm financial performance (FFP) (Hinsberg, Nadesan, Parrish & Lamanna, 2025). Workplace accidents create increased production disruptions, compensation claims, regulatory sanctions, litigation costs, thereby reducing firm performance (Hinsberg et al., 2025). Integrating occupational safety, especially in mining sector is important strategic priority toward ensuring safe working conditions and improving firm performance (Milea (Parvu), Moraru & Cioca, 2025). This also facilitates the achievement of the United Nations sustainable development Goal (SDG 8: Decent work and economic growth).

Organisations that integrate safety practices into their operation experience improved financial performance (Ying & Allaqtta, 2025). Minimising workplace injuries and fatalities reduce direct costs such as compensation, death benefit, and legal and litigation costs that can negatively affect FFP (Asiedu, Appiagyei, Amfo-Out, Parku & Obuobisa-Darko, 2023). However, no consensus has been reached on relationship between safety performance (SP) and FFP. While some scholars found a positive relationship between SP and FFP (Adu-Gyamfi, He, Nyame, Boahen & Frempong, 2021; Yang, Lin & Maresova, 2021; Xuanya, Suxia, Qiang & Jingjing, 2023; Shabani, Jerie & Shabani, 2023; Bautista-Bernal, Quintana-Garcia & Marchante-Lara, 2024; Madwe & Nzuza, 2026), others report no or negative association (Kalemba, 2017; Forteza, Carretero-Gomez & Sese, 2017; Oswald, Ahiaga-Dagbui, Sherratt & Smith, 2020; Estudillo, Carretero-Gomez & Forteza, 2024). This suggests that firm characteristics such as corporate governance (CG) mechanisms may impact the strength of relationship.

Board gender diversity (BGD) on corporate boards has emerged as an important mechanism for improving board effectiveness, strategic oversight, risk management and for improved FFP (Saktiawan, Setiyono & Rachmadi, 2026). Increasing the proportion of female representation on boards enhance decision-making, improving risk management and improving firm performance (Islam, French & Ali, 2022; Paolone, Pozzoli, Chhabra & Di Vaio, A, 2024). Occupational safety is a critical governance and risk management issue in mining sector (Milosevic, Stojanovic, Nikolic, Mihajlovic, Brkic, Perisic & Spasojevic-Brkic, 2025). BGD on corporate board may provide more effective oversight of occupational safety policies (Lin, Chien, Chang, Liao & Nagata, 2026), enabling organisation to realise financial benefits associated superior SP.

Prior research has extensively examined occupational safety, BDG, and FFP, however these variables have been explored in isolation. As a result, little attention has been paid how BGD shapes the association between firm profitability and SP, especially in mining industry in South Africa. Many studies have primarily investigated the direct impact of BGD on firm performance (Mohsni, Otchere & Shahriar, 2021; Kabir, Ikra, Saona & Azad, 2023; Zeng, Huang, Lu, Wu & Yin, 2025) or the effect of occupational safety practices on FFP (Bautista-Bernal et al., 2024; Moon, Ikeda-Araki & Mun, 2025). Therefore, there is limited understanding on how BGD shapes the extent to which workplace SP impact FFP.

Consequently, the study aims to investigate the impact of SP on FFP of JSE-listed mining companies over the period 2015 to 2025. Furthermore, the study explores whether BGD influence the degree to which improvements in occupation safety practices improved FFP. The study contributes to the body of knowledge by illustrating how BGD influences the association between FFP and SP in JSE-listed mining companies. The study’s findings could assist regulators, investors and policymakers in improving workplace safety and encouraging diverse corporate boards to improve firm FFP. This study extends existing literature by providing evidence on how BGD and occupational safety could jointly support the SDG 8. Moreover, the present study highlights how these constructs facilitate the reduction of workplace accidents and improving FFP.

2. Theoretical Framework

This study employs resource dependency theory (RDT) and stakeholder theory (ST) to explore the extent to which BGD influences relationship between SP and FFP. RDT posits that sustainable responsible practices, such as green innovation initiatives and occupational risk management strategies can be utilised to provide organisation with long-term firm value (Almaqtari, Elmashtawy, Farhan, Almasria & Alhajri, 2024; Correia, Shahzad, Martins & Baheer, 2024). Firms utilise available resources and skills to facilitate social sustainable development and financial performance (Koczias & Demirel, 2025). CG mechanisms have emerged as critical in obtaining resources, offering guidance and organising connection between external stakeholders and companies (Bendickson, Gur & Taylor, 2018). In the context of this study, female directors bring alternative perspectives and skills to address problems of workplace accidents in high-risk sector such as mining, contributing to better management of stakeholder demands of safe working conditions. This study argues that the presence of female on board brings new ideas that lead to connecting FFP and social sustainable practices (SP). Therefore, women directors provide unique networks and resources in the organisation, thereby reducing associated costs of occupational injuries and fatalities.

On the other hand, ST postulates that many stakeholders shape the firm operation activities, that should be take into account in decision making process (Sulemana, Cheng, Agyemang, Osei & Nagriwum, 2025). Firms that consider stakeholders’ interests and expectations often invest in socially responsible practices such as workplace safety practices (Awa, Etim & Ogbonda, 2024), thereby improving firm competitive advantage and FFP. BGD enables companies to respond to employee well-being by promoting effective corporate board oversight on occupational safety practices (Alhosani & Nobanee, 2023). Therefore, firms that integrate CG mechanisms with safety management create wealth and attaining firm sustainability to foster enhanced FFP (Almaqtari et al., 2024).

3. Literature Review and Hypothesis Development

This section reviews the theoretical and empirical literature on SP, BGD and FFP. It synthesises existing evidence to identify gaps in the literature and establish the conceptual basis for the proposed relationships.

Safety Performance and Firm Financial Performance

Workplace safety has emerged as serious concern for companies operating in high-risk sectors such as mining due to employees’ exposure to dangerous working conditions (Blanco-Juarez & Buele, 2025). Occupational safety is widely recognised as a strategic organisational capability that enhance operational efficiency, employee well-being and firm performance (Mtikitiki, Madonsela, Maphanga, Grangxabe, Baloyi & Phungela, 2025). Prior research commonly adopts injury frequency rate (IFR) and fatality frequency rate (FFR) to measure safety performance (Melchior & Zanini, 2019; Madwe & Nzuza, 2026). Occupational accidents have adverse impacts on employee well-being and translate into monetary losses through increased compensation claims, legal costs, production disruptions and regulatory fines (Asiedu et al., 2023; Hinsberg et al., 2025). As a result, organisations prioritise workplace safety to facilitate safe working conditions in support of SDG 8, while improving FFP through reducing costs associated with occupational accidents.

Consequently, effective occupational safety management minimises accident costs, reduce operational disruptions and improving stakeholder confidence (Al-Dmour, AlKhawaldeh, Al-Dmour, Obidat & Al-Dmour, 2025). In line with this argument, prior research reveals that companies that integrate occupational safety practices into their corporate strategies enhance their profitability and firm value (Adu-Gyamfi et al., 2021; Yang et al., 2021; Xuanya et al., 2023; Bautista-Bernal et al., 2024; Madwe & Nzuza, 2026). These scholars emphasis that firms that invest in safety practices improve operational efficiency and reduce financial losses associated with occupational accidents, contributing to FFP.

Despite claimed benefits of occupational safe practices, existing literature reveals mixed results on the impact of safety performance on financial performance. While some researchers found a positive relationship between safety performance and financial performance (Adu-Gyamfi, He, Nyame, Boahen & Frempong, 2021; Yang, Lin & Maresova, 2021; Xuanya, Suxia, Qiang & Jingjing, 2023; Shabani, Jerie & Shabani, 2023; Bautista-Bernal, Quintana-Garcia & Marchante-Lara, 2024; Madwe & Nzuza, 2026), others report no relationship or even negative effects(Kalemba, 2017; Forteza et al., 2017; Oswald et al., 2020; Estudillo et al., 2024). This suggests that the implementation of workplace safety practices may have significant short term operating costs without realising immediate financial benefits. These inconsistencies highlight that financial benefits of occupational safety practices may depend on firm characteristics that shape how companies translate workplace safety practices into financial value. This perspective offers a strong justification for exploring potential CG mechanisms that influence this association. Based on this argument, the study formulates the following hypothesis:

H1: 

Lower injury frequency rate is positively associated with financial performance of JSE-listed mining companies.

H2: 

Lower fatality frequency rate is positive associated with firm financial performance.

Board Gender Diversity, Occupational Safety and Firm Financial Performance

BGD has become an important topic in CG due to its ability to improve board effectiveness and firm performance (Marquez-Cardenas, Gonzalez-Ruiz & Duque-Grisales, 2022). In the context of this study, BGD is defined the percentage of women directors on boards, and it shows how companies encourage inclusivity and diversity in their CG (Zulvina & Setiawan, 2025). Yami, Alshurafat, Shaaban (2025) argue that increasing female representation on corporate boards strengthen CG. Female directors contribute diverse perspectives and professional background and cognitive viewpoints that improve board oversight and FFP (Wu, Gao, Luo, Xu & Shi, 2024). Consequently, BGD is widely recognised as a critical CG mechanism that enhance firm sustainability and create firm value.

Existing research reports female directors as risk-aware, ethically oriented and stakeholder-focused that their male counterparts (Mohsni et al., 2021; Kabir et al., 2023). Moreover, increasing women serving on corporate boards foster accountability, enhances board independence and improves the quality of CG (Ojeyinka, Matemane, Moraka & Molaoa, 2025), leading to long-term FFP.

However, there is no consensus in the existing literature on the relationship between BGD and FFP. Several researchers report that report that increase in women serving on corporate boards enhance firm profitability and improve FFP (Mohsni et al., 2021; Alodat, Salleh, Nobanee & Hashim, 2024; Aziz & Cek, 2026). These authors contend that BGD improves CG quality, support board oversight, contributing to improved FFP. On the other hand, other studies found that BGD negatively influence FFP (Kabir et al., 2023; Pandey, Kumar, Post, Goodell & Garcia-Ramos, 2023; Wang, Ma, Xue & Zhang, 2024; Zeng et al., 2025;). These mixed results suggest that the financial value of BGD depends on other CG mechanisms and environment in which board operate.

One important contingent variable in which BGD may create firm value is workplace safety (Marheni, Sherry & Yulfiswandi, 2024). Occupational accidents expose organisation to significant financial and reputational risks. BGD may encourage to invest in occupational safety practices, which may eventually positively impact FFP (Widyasari & Marheni, 2022). Prior research indicates that the BDG influences the relationship between sustainable responsible practices and FFP (Albitar et al., 2020; Duppati et al., 2020; Fernando et al., 2020; Zhu et al., 2022). BGD improves working conditions and reduce injuries and fatalities by promoting the implementation of safety practices in workplace (Wu et al., 2024; Yami et al., 2025). The investment in occupational safety reduces costs associated with accident, thereby improving FFP.

Although many studies have explored the relationship between occupational safety and BGD, limited studies investigated their combined effect on FFP, especially in JSE-listed mining sector. Therefore, the present study addresses this research gap by explaining how BGD shapes the association between FFP and safety performance in JSE-listed mining industry. As a result, the following hypothesis is formulated:

H3: 

Board gender diversity positively moderates the relationship between safety performance and corporate financial performance.

H3a: 

Board gender diversity is positively moderating the association between injury frequency rate and firm financial performance.

H3b: 

Board gender diversity positively moderates the fatality frequency rate- financial performance relationship.

4. Materials and Methods

Research Design, Data Collection and Sampling

Quantitative explanatory research design was adopted to examine the role of BGD on SP and FFP. The study employs secondary data obtained from JSE-listed mining companies’ annual reports, integrated reports and sustainability reports of JSE-listed mining companies, spanning 2015 to 2022. The study period of 2015 to 2025 was selected to ensure an up-date safety performance data after the introduction of the United Nations SDGs programs. The study selected JSE-listed mining industry as this sector is susceptible to higher scrutiny from stakeholder due to their potential risk sustainability and have occupational safety information in their integrated and sustainability reports. Therefore, the present research employed a purposive sampling technique. The research sample comprised 46 JSE-listed mining companies. The unbalanced panel data structure was used to address missing observation resulted from firm listing and delistings during study period, generating 438 firm-year observation retained for final analysis. Many accounting and finance researchers adopt Stata for their studies (Tawiah, Zakari & Khan, 2021; Agyemang, Yusheng, Kongkuah, Musah & Musah, 2023). In the same vein, this study utilised Stata.

Variable Measurement

Flowing previous studies, IFR and FFR were utilised to measured Safety performance (Melchior & Zanini, 2019; Ramos, Afonso & Rodrigues, 2020; Madwe & Nzuza, 2026). The study employed accounting-based and market-based indicators to measure FFP, namely return on assets (ROA) and Tobin’s Q (TQ), respectively. The percentage of women directors on the board measured BGD. The total number of women serving on corporate boards has commonly used in prior research to measure BGD (Wasiuzzaman & Wan Mohammad, 2020; Brahma, Nwafor & Boateng, 2021). The study controlled Firm size (FSIZE) and board independence (BI), each control variable providing insight into varying aspects of FFP and SP. BI was measured by the percentage of non-executive directors serving on the corporate board, the natural logarithm of total assets measured FSIZE. Table 1 shows the measurements of the variable of this study.

Econometric Model

Following Almaqtari et al. (2020), the study four models, and each model representing dependent variable, and moderating effect of BGD. These models are more appropriate in examining how BGD shapes the relationship between SP and FFP (Agyemang et al., 2023). These models are presented as follows:

Model 1: ROA and Injury frequency rate

${ROA}_{it}$ = ${\beta }_0$ + ${\beta }_1{IFR}_{it}$ + ${\beta }_2{BGD}_{it}$ + ${\beta }_3$ (${IFR}_{it}x{BGD}_{it}$) + ${\beta }_3{FSIZE}_{it}$ + ${\beta }_4{BI}_{it}$ + ${\mu }_{it}$ + ${\epsilon }_{it}$…. Equa (1)

Model 2: ROA and Fatality frequence rate

${ROA}_{it}$ = ${\beta }_0$ + ${\beta }_1{FFR}_{it}$ + ${\beta }_2{BGD}_{it}$ + ${\beta }_3$ (${FFR}_{it}x{BGD}_{it}$) + ${\beta }_4{FSIZE}_{it}$ + ${\beta }_5{BI}_{it}$ + ${\mu }_{it}$ + + ${\epsilon }_{it}$…. Equa (2)

Model 3: TQ and Injury frequency rate

${TQ}_{it}$ = ${\beta }_0$ + ${\beta }_1{IFR}_{it}$ + ${\beta }_2{BGD}_{it}$ + ${\beta }_3$ (${IFR}_{it}x{BGD}_{it}$) + ${\beta }_4{FSIZE}_{it}$ + ${\beta }_5{BI}_{it}$ + ${\mu }_{it}$ + ${\epsilon }_{it}$ ….. Equa (3)

Model 4: TQ and Fatality frequency rate

${TQ}_{it}$ = ${\beta }_0$ + ${\beta }_1{FFR}_{it}$ + ${\beta }_2{BGD}_{it}$ + ${\beta }_3$ (${FFR}_{it}x{BGD}_{it}$) + ${\beta }_4{FSIZE}_{it}$ + ${\beta }_5{BI}_{it}$ + ${\mu }_{it}$ + ${\epsilon }_{it}$ ….. Equa (4)

In equations, ${ROA}_{it}$ and ${TQ}_{it}$ represent firm financial performance (FFP), i and t represent firm and year. ${FFR}_{it}$ and ${IFR}_{it}$ measure safety performance (SP). (${FFR}_{it}x{BGD}_{it}$) and (${IFR}_{it}x{BGD}_{it}$) represent the moderating effect of BGD on SP-FFP relationship. ${\beta }_0$ is a constant (intercept), while ${\beta }_1$ -${\beta }_5$ represent parameters to be estimated. ${\mu }_{it}$ represents unobserved firm-specific effect that capture time-invariant characteristics unique to firm and ${\epsilon }_{it}$ is error term.

Data Analysis

The descriptive statistics was used to summarise the characteristics of the study variables, thereafter Pearson correlation analysed relationships among the variables. To assess the presence of multicollinearity among explanatory variables the variance inflation factor (VIF) was employed.

Fixed Effects (FE) and Random Effects (RE) regression models were estimated. The Hausman test was utilised to determine the most suitable regression model. Based on the Hausman tests results, the FE was chosen as the estimator to analyse the association between SP and FFP. FE regression model controls for all time-invariant differences between the individuals, and its coefficients cannot be biased due to time-invariant characteristics (Ren & Allison, 2025).

Robustness Test

The two-step system generalised method of moments (System GMM) was adopted to check robustness of the study’s findings. System GMM addresses potential endogeneity arising from reverse causality and omitted variables (Tubik & Herberger, 2026). The Hanse Test for over-identifying restriction was used to assess the validity of the instrument in System GMM regression models. Furthermore, the Arellano-Bond first-order [AR(1)] and second-order [AR(2)] were adopted serial correlation.

5. Results and Discussion

Descriptive Statistics

Table 2 illustrate the descriptive statistics for the dependent variables (ROA & TQ), the independent variable (IFR & FFR), moderating variable (BGD) and control variables (FZIZE & BI). Table 2 show that occupational injuries are relatively common in JSE-listed companies (mean=3,29; min=0.65), however fatalities happen less regularly (mean=0.063; min=0.000). Table 2 indicates while some sampled JSE-listed firms were profitable (max=42.3%), other firms experienced significant losses (-29.5%).

The results show that an average of 28.5% of female directors serving on corporate boards in JSE-listed firms, while an average of 72% are non-executive board of directors.

Correlation Analysis

Table 3 presents the Pearson correlation coefficients of all study’s variables. The results a positive statistically relationship SP and FFP (r = -0.281, p < 0.01; r = -0.254, p < 0.01). This suggests that companies with better safety and health practices to prevent occupational accidents achieve higher profitability. However, findings report weak and statistically insignificant association between fatality accidents and FFP (r = -0.072, p > 0.01; r = -0.058, p > 0.01), suggesting that fatality accidents may not immediately affect financial performance.

The results show that BGD is positively associated with FFP (r = 0.321, p < 0.01; r = 0.286, p < 0.01).

Multicollinearity Test

The study utilised variance inflation factors (VIF) to determine the presence of serious multicollinearity among the study’s explanatory variables. Kim (2019) indicates that multicollinearity is present when the VIF is higher than 5 to 10. Table shows that ALL VIF values are between 1.29 and 1.68, and the mean VIF is 1.46. Table 4 confirm the absence of serious multicollinearity as all values are below the threshold of 10.

Hausman Specification Test

The study performed the Hausman test to determine most appropriate regression model between FE and RE model. The results rejected the null hypothesis (${\mathsf{{\rm X}}}^2$ =24.71, p =0.003), suggesting FE as the more suitable than RE model.

Fixed Effects Regression Results

In model 1 (ROA & IFFR) Table 5 shows a positive statistically significant relationship between IFR) and ROA (β = 0.021, p < 0.05). The results indicate that mining firms with lower occupational injuries generate higher profits. As a result, these findings support Hypothesis 1 (H1). These results also in line with RDT which argues that achieving socially sustainable performance is contingent on external resources and CG mechanisms such as BGD play a critical role in obtaining these resources.

The findings also support ST that postulates that firm that have safe working conditions are more open to employee well-being and other stakeholder expectations, leading to improved firm competitive advantage and improved FFP.

These findings are consistent with those of Adu-Gyamfi et al. (2021), Yang et al. (2021), Xuanya et al. (2023), Shabani et al. (2023), Bautista-Bernal et al. (2024) and Madwe and Nzuza (2026), who reported a positive association workplace safety performance and FFP. However, the study’s results challenge Kalemba (2017), Forteza et al. (2017) and Estudillo et al. (2024), who that there is revealed no relationships between SP and FFP.

In the model 2 (ROA & FFR), no significant relationship was found between SP and FFP (β = 0.009, p > 0.10). This indicates that occupational accidents may not immediately affect FFP in JSE-listed mining firms. Therefore, the results refute Hypothesis 2 (H2). While preventing occupational fatal accident is important, this may indicate that fatal accident happen relatively infrequently and this reduce their significant impact on FFP.

In term of RDT, findings indicate that CG mechanisms may contributes to firm performance by facilitating access to resources, supporting entire workplace safety practices, instead of only solving fatal accidents in isolation. Similarly, ST argues that stakeholders assess firms based on their overall commitment to workplace safety practices, rather than individual fatal incidences.

These findings are consistent with those of Kalemba (2017), Forteza et al. (2017), Oswald et al. (2020) and Estudillo et al. (2024), who found no insignificant association between SP and FFP.

Model 3 (TQ & IFR) shows that lower IFFR positively impact FFP (β = 0.049, p < 0.01), suggesting that firms with better workplace safety performance also achieve higher market valuations. Accordingly, The results support Hypothesis 1 (H1). This suggests that investors consider occupational safety practices as evidence of good CG, lower workplace risks and improved firm performance. The results support RDT which indicates that sound CG ensure access to valuable resources that improved FFP.

The research findings also confirm ST which assumes that investors value companies that show responsible CG and safeguard the interests of all stakeholders.

The findings are in line with those of Yang et al. (2021), Bautista-Bernal et al. (2024), Shabani et al. (2023) and Madwe and Nzuza (2026), who reported that occupational safety practices positively influence and FFP.

Model 4 (TQ & FFR) shows no statistical relationship between SP and FFP (β = 0.022, p > 0.10), highlighting that investors do not adjust market valuations based solely on fatal accident, but consider broader workplace health and safety management strategies

Confirming Model 2, these results are in line previous studies that no insignificant relationship between SP and FFP (Oswald et al., 2020; Estudillo et al., 2024).

Table 5 shows that BGD statistically and positively moderates the SP (IFF)-FFP nexus across all models, thereby confirming Hypothesis 3a (H3a). These results highlight that firm with well-represented women in the boards generate higher profitability. Although the direct impact of FFR ON BGD is statistically insignificant, the interaction relationship is significant. This suggests that BGD improve effectiveness of occupational safety, translating to improved FFP. Therefore, the results support Hypothesis 3b (H3b). These results are consistent with RDT which claims that CG mechanisms such as BGD serve as important channels through which firm gain access to valuable resources. The research findings also confirm ST which assumes firms that encourage gender-diverse corporate boards are more responsive to employee well-being and safety practices. that increasing female representation in corporate boards improve occupational safety, contributing to FFP.

The findings are consistent with those of Albitar et al. (2020), Fernando et al. (2020), Zhu et al. (2022) and Duppati et al. (2020), who revealed that BGD facilitates the sustainability-related initiatives-firm performance relationship through improved governance quality and stakeholder engagement. However, this study contributes to sustainability and FFP literature by highlighting that BGD also enhances the financial value associated with occupation SP in the South African mining industry.

Robustness Test: System GMM

The study used two-step System GMM to assess whether the FE results were sensitive to potential endogeneity. Table 6 shows that the sign, magnitudes and significance level of coefficients remain unchanged compared to FE model. Table 6 indicates no overidentifying restriction problem and the instrument is valid (Hansen test: p> 0.05). Arellano-Bond AR(2) test shows the absence of second-order serial correlation, suggesting that all models were correctly specified.

The consistency of the findings between FE and System GMM regression models offer evidence that the study’s conclusions are robust and are free from estimation bias or endogeneity.