Women in the top management team: bridging the gender divide in advanced technologies

Despite significant progress in the pursuit of gender equality, countless obstacles persist that hinder this journey (WEF, 2018). One area of concern is the gender gap in advanced technologies and Artificial Intelligence (AI)-related professions, particularly considering the disproportionate underrepresentation of women in education, employment, and leadership across these high-growth sectors. Extensive research conducted by esteemed institutions, such as the World Economic Forum (WEF, 2018) and the European Commission (2018), has shed light on the unprecedented scarcity of women in the high-tech sector, thereby presenting a formidable challenge to the labour market. As highlighted by prior studies (Palomares-Ruiz et al., 2021; Wang and Degol, 2017), this underrepresentation of females in technology permeates across all echelons, spanning from Science, Technology, Engineering, and Mathematics (STEM) students to upper-tier positions within organisations. Consequently, women's potential contribution to organisational dynamics through their perspectives, beliefs, and expertise is at risk of fading in the future labour market (WEF, 2018). The marginalisation of women in high-tech education and industry denies them the opportunity to cultivate advanced technical skills, which are increasingly sought-after even in fields traditionally unaffected by digitalisation, such as education, healthcare, and non-profit sectors (Hamet and Tremblay, 2017; Glass and Cook, 2020). Given the strategic relevance of technological innovation for economic competitiveness and social development, addressing this gender gap is not only a matter of equity, but also a critical imperative for ensuring inclusive and sustainable growth in the digital era (UNESCO, 2024; WEF, 2025; EIGE, 2020).

In this vein, addressing the persistent gender gap in high-tech sectors needs not only equitable access to technical education, but also organisational conditions within firms that support women's recruitment, retention, and progression. In response, scholars have increasingly paid attention to two key firm levers: gender diversity in leadership positions (Cook and Glass, 2014; Glass and Cook, 2020; Liu et al., 2024) and the implementation of work-life balance (WLB) policies (Al-Najjar and Salama, 2022; Baker et al., 2024).

Much of the existing research on gender diversity in leadership has especially focused on women's representation on boards of directors (BoDs), largely due to their central role in corporate governance (Birindelli et al., 2019; Li et al., 2018). This research interest has yielded valuable insights into how gender-diverse boards can enhance oversight, accountability, and long-term value creation (Francoeur et al., 2008; Terjesen et al., 2009). However, it overlooks a crucial limitation: BoDs typically do not engage in the day-to-day strategic and operational decision-making that shapes internal practices and workplace culture (Hillman and Dalziel, 2003; Shepherd et al., 2024). Along this line of reasoning, research has shown that while BoDs play a critical governance role, their involvement in strategic execution is often limited, particularly when compared to management teams (Cho et al., 2021; Francoeur et al., 2008; Wu et al., 2022). Indeed, Top Management Team (TMT) members – typically defined as the group of senior executives responsible for the strategic and operational management of the firm – are directly involved in shaping organisational strategy, managing resources, and influencing workplace culture (Hambrick, 2007; Glass and Cook, 2020). Their decisions have a more immediate and practical impact on internal practices concerning the advancement of gender equity (Matsa and Miller, 2011; Glass and Cook, 2020). As a result, recent research has increasingly turned to the role of female representation within TMTs (Post and Byron, 2015; Hoobler et al., 2018; Ali et al., 2021), particularly in male-dominated sectors where cultural transformation is both necessary and challenging.

Similarly, WLB policies have been widely investigated for their potential to improve women's labour market outcomes (Al-Najjar and Salama, 2022; Baker et al., 2024). These policies offer flexibility that is particularly relevant for individuals managing caregiving and professional responsibilities, and have been linked to higher retention, job satisfaction, and inclusion of women in male-dominated fields (Al-Najjar and Salama, 2022; Baker et al., 2024; Tushabe et al., 2025). Studies have shown that WLB programs can facilitate greater female participation (Gordon et al., 2012; Tellhed et al., 2023), yet their impact heavily depends on the organisational culture in which they are embedded. Women are often reluctant to fully utilise these benefits due to concerns about career penalties or perceptions of reduced commitment (Rogier and Padgett, 2004; Bertola et al., 2023). Research has consistently found that WLB policies are most effective when leaders actively endorse and legitimise their use (Burnett et al., 2010; Perrigino et al., 2018; Bertola et al., 2023).

Despite these contributions, the two research fields – namely, gender diversity in leadership positions and WLB policies – have been largely developed in parallel. Indeed, scant studies have analysed the interaction between the presence of women in strategic leadership positions and WLB policies, and in turn, their influence on the broader patterns of women's representation in organisations (Anglin et al., 2022). This interaction is especially relevant given that the effectiveness of WLB policies has been shown to depend on supportive organisational cultures – conditions that women leaders may be uniquely positioned to foster (Glass and Cook, 2020; Dezsö and Ross, 2012).

To fill this research gap, we draw on Social Role Theory (SRT) (Dawson, 1997; Eagly, 1997) as the theoretical framework for our research hypotheses development. SRT posits that individuals form gender role expectations by observing the social division of labour (Wood and Eagly, 2002). Over time, these observations give rise to widely shared stereotypes that not only describe how men and women tend to behave, but also prescribe how they are expected to behave and which roles they are deemed suited to occupy (Diekman and Goodfriend, 2006). These expectations are deeply embedded in organisational life and have particular consequences when individuals assume roles that fall outside traditional gender prescriptions (Burkhardt et al., 2020; Eagly and Carli, 2009).

When women occupy leadership roles – especially in male-dominated business sectors – they tackle a mismatch between societal expectations and their professional identity (Tokbaeva and Achtenhagen, 2023; Mistry et al., 2024). This role incongruity can heighten their awareness of the structural and normative barriers women face and prompt them to challenge these dynamics from within (Anglin et al., 2022; Burkhardt et al., 2020; Eagly and Karau, 2002). Based on this theoretical perspective, we conceptualise women managers not merely as demographic representatives, but as agents of organisational change who may fuel gender inclusion. Indeed, through role-modelling and strategic advocacy, women managers can challenge traditional gender role expectations by visibly occupying high-status positions and advocating for greater inclusion. Their presence serves as a signal that leadership is accessible to women and may inspire others, while also motivating them to support structural change (Cook and Glass, 2014; Dezsö and Ross, 2012). This mechanism underpins our expectation that greater female representation covering leadership roles will be associated with a higher share of female employees in high-tech firms.

Furthermore, SRT emphasises that women are culturally expected to assume primary caregiving responsibilities (Wood and Eagly, 2002), exacerbating role incongruity when such expectations conflict with the behavioural and temporal demands of full-time employment, particularly in male-dominated, high-pressure sectors like high-tech (Eagly and Karau, 2002; Rudman and Glick, 2001). While women managers can signal inclusion and advocate for change, their ability to increase female workforce participation may be reinforced by the presence of WLB policies. These last – especially flexible working arrangements – help reduce the friction between professional and caregiving roles and support women's employment (Gordon et al., 2012; Tellhed et al., 2023), thus enhancing the effectiveness of inclusive leadership.

In light of these cues, the aim of this research is to explore the extent to which the percentage of women managers, in interaction with WLB policies, influences the representation of women in the broader workforce within high-tech firms. To this end, we conduct a longitudinal analysis on the European listed companies belonging to the high-tech sector during the 2014–2020 timeframe. The empirical evidence spotlights that increasing women's management representation, along with more inclusive WLB policies, leads to a higher presence of female employees. To infer, we consider that organisations with more women managers are better positioned to reduce the gender gap through the dual effect of role-modelling and the introduction of supportive policies. This approach not only enhances female labour force participation, but also enables the creation of a more inclusive and innovative workforce.

The rest of the paper is structured as follows: the next section reviews the current literature, while Section 3 develops the research hypotheses. In Section 4, the methodological approach is explained. Section 5 presents the findings, while the final section provides a comprehensive picture of the implications for both firms and societal entities, as well as the limitations of the research and suggests future research avenues.

According to the “Global Gender Gap Report” released by WEF (2024), it will take an estimated 134 years to close the global gender gap at the current rate of progress. As previously evidenced by the WEF (2018), this extended timeline is further compounded by new gender disparities emerging in advanced technologies and AI-related jobs (European Commission, 2024; UNESCO, 2023; WEF, 2024), with 72% of the gap needing to be closed in professions related to advanced technologies, including AI.

What is particularly alarming regarding the gender gap is that digital skills are becoming essential across nearly all sectors. Without noteworthy female participation in advanced technologies, women risk missing the opportunity to attain the crucial knowledge and expertise for being part of the future workforce (WEF, 2018). As advanced technologies become integral across industries, women may struggle to meet the growing demand for digital competencies not only in the advanced technology sector, but also in highly digitalised sectors (WEF, 2018). As a result, women face the risk of marginalisation even in sectors traditionally dominated by the female workforce, such as education and healthcare, which are now heavily influenced by advanced technologies (Fuchs, 2023; Kumar et al., 2024). Furthermore, women could toil to achieve these digital skills during their educational path, due to their lower presence in STEM fields, which represent the foundation for cultivating digital talent. However, the root of this gender segregation in education is not merely a result of personal inclinations or competencies; it is heavily influenced by cultural and systemic factors that must be addressed when discussing gender disparities (Eagly and Steffen, 1984; Sebastián-Tirado et al., 2023; Shapiro and Williams, 2012). Indeed, cultural expectations and stereotypes play a significant role in the gender imbalance within STEM fields (Margolis, 2017; OECD, 2022; UNESCO, 2022). For instance, UNESCO (2022) highlights that gender stereotypes in educational materials and teacher expectations continue to discourage girls from engaging in science and technology. Moreover, OECD (2022) points out that parents are more likely to encourage boys over girls to pursue careers in technology, contributing to a sustained gender gap in digital skills. These cultural biases not only affect educational choices, but also have long-term impacts on career trajectories, resulting in fewer women acquiring the digital skills needed in the modern labour market (Wang and Degol, 2017).

Gender segregation is not confined to education, but extends into the labour market, narrowing women's opportunities to find employment aligned with their qualifications (WEF, 2024). This reality further deters women from pursuing STEM education, as they face not only cultural barriers, but also reduced employment prospects despite achieving similar qualifications to men (European Institute for Gender Equality EIGE, 2018; WEF, 2018). In this vein, national and international governments are working to increase the number of STEM women students and professionals by changing the narrative around women in advanced technologies and implementing policies aimed at reducing gender stereotypes and biases (Kubberød et al., 2021; Saifuddin et al., 2022). This effort underscores the relevance of role models in addressing gender segregation, as having female leaders and professionals in high-tech can inspire and motivate young girls to pursue careers in these fields (Dasgupta and Stout, 2014). Besides, women leaders and professionals are not only passive role models, but can actively contribute to closing the gender gap through policies and behaviours steered towards equality (Burkhardt et al., 2020; Eagly and Carli, 2009). These dynamics highlight the structural depth of the gender gap in high-tech and underscore the potential significance of women's presence in strategic leadership roles.

SRT posits that individuals learn and internalise gender roles by observing the distribution of tasks and responsibilities in society (Wood and Eagly, 2002). These observations form the basis of prescriptive stereotypes that define not only how men and women are expected to behave, but also which roles they are considered suited to occupy (Diekman and Goodfriend, 2006). In male-dominated sectors, such as high-tech, leadership roles are still implicitly coded as masculine – associated with traits like assertiveness, decisiveness, and emotional detachment (Mickey, 2022; Neely et al., 2023; Ezzedeen et al., 2015). Thus, when women occupy these positions, they often confront a role incongruity: a mismatch between gender-based expectations and the professional identity they embody (Eagly and Karau, 2002).

However, it is precisely this incongruity that may enhance women managers' awareness about the structural and normative barriers facing other women in the workforce. As members of a numerically and culturally underrepresented group, women in decision-making roles often develop a stronger sensitivity to gendered dynamics (Burkhardt et al., 2020; Eagly and Carli, 2009). In the specific case of women in TMT – the small, powerful group of executives responsible for shaping strategy, resource allocation, and internal culture (Hambrick, 2007) – this awareness is paired with the positional authority needed to influence organisational priorities (Glass and Cook, 2020). Notably, this operational authority distinguishes managers from board members or non-strategic executives, whose roles are often more supervisory or symbolic (Hillman and Dalziel, 2003; Shepherd et al., 2024). While BoDs primarily influence corporate governance and oversee strategic direction, the TMT is responsible for the formulation and day-to-day implementation of organisational policies, making them more directly positioned to influence internal culture and affect women's representation within the workforce (Hambrick, 2007; Carpenter et al., 2004). This perspective makes women managers uniquely positioned to challenge prevailing norms and promote diversity-oriented change (Glass and Cook, 2020; Dezsö and Ross, 2012).

Building on this theoretical grounding, we identify role-modelling and strategic advocacy as the first key mechanisms through which women managers may foster broader gender inclusion in the workforce. First, their presence in leadership positions serves as a powerful symbolic cue, signalling that advancement is not only possible for women, but also supported (Dasgupta and Asgari, 2004; Stout and Dasgupta, 2011). This signal can reduce stereotype threat, elevate internal aspirations among female employees, and foster perceptions of organisational fairness and opportunity (De Celis et al., 2015; Joshi et al., 2015). Secondly, beyond symbolism, women managers also have the formal capacity to shape human resource strategies and institutional practices – such as recruitment, retention, and promotion criteria – thereby structurally supporting gender diversity (Cook and Glass, 2014; Liu et al., 2024). Prior studies have found that greater female representation in leadership positions is positively associated with increased representation of women across organisational levels (Matsa and Miller, 2011; Hoobler et al., 2018; Ali et al., 2021). Accordingly, we hypothesise that a higher presence of women in TMT positions is positively associated with female workforce participation. Based on this theoretical foundation, we propose the following hypothesis:

Drawing on SRT, we consider that one of the most enduring cultural expectations surrounding gender roles is the assignment of primary caregiving responsibilities to women (Wood and Eagly, 2002; Eagly and Karau, 2002). In high-tech sectors – where work cultures are shaped by traditionally masculine norms, such as long hours, constant availability, and uninterrupted career paths (Rudman and Glick, 2001; Koenig et al., 2011) – this caregiving norm introduces a second, compounding layer of role incongruity for females. Women are not only underrepresented in these male-dominated environments, but are also disproportionately burdened by the tension between societal caregiving expectations and the behavioural and temporal demands of high-status professional roles. This results in symbolic and material barriers to women's entry and advancement in the workforce, constraining their participation even when qualifications and aspirations align (Blair-Loy and Jacobs, 2003; Heilman, 2012).

While women managers can help disrupt these dynamics through symbolic role-modelling and strategic advocacy (as proposed in H1), we argue that their impact is significantly reinforced by the presence of institutional supports that address the caregiving-related dimension of role incongruity. WLB policies, such as flexible scheduling, remote work, and caregiving leave (e.g. Eikhof et al., 2007), can remarkably reconcile the tension between professional and caregiving roles, making it more feasible for women to remain and thrive in high-tech environments. In male-dominated environments, where the use of flexibility measures is often stigmatised or discouraged (Perrigino et al., 2018; Bertola et al., 2023), the formal availability and organisational support for WLB policies can significantly help women leaders in enhancing female employee participation. By addressing the caregiving-related constraints that disproportionately affect women, these policies create the reinforcing conditions under which symbolic leadership and strategic advocacy can translate into broader gender inclusion.

Building on the earlier considerations, we posit the following hypothesis:

The analysis focused on European listed companies operating in the high-tech sector during the 2014–2020 time frame. We gathered data from the Refinitiv Eikon (Datastream) database by including public companies headquartered in the European Union (EU) working in the high-tech sector, as classified by Eurostat ^[1]. Table 1 sets out the process of sample selection and outlines the main steps undertaken to build our final dataset. The initial dataset comprised 1,354 companies (Stage 2), spanning 27 European countries across various high-tech sectors, such as Software and IT Services, Financial Technology (Fintech), Infrastructure, Telecommunications Services, and Technology Equipment (Table 1). Subsequently, in Stage 3, we applied data availability criteria, thereby excluding companies with missing data for the variables of interest. This resulted in a sample of 189 high-tech companies (Stage 4), yielding a maximum potential of 1,323 observations (Stage 5). Further refinement through the removal of missing values reduced the final number of observations to 512 (Stage 6). The final sample represented 38.70% of the total maximum potential observations, indicating that our study reflected a relevant proportion of European high-tech firms (Ludwig et al., 2022).

The percentage of women employees is used as a dependent variable (Khaled et al., 2021; Rojo-Suárez and Alonso-Conde, 2023). Based on Jermias and Mahmoudian (2024), this metric represents the ratio of women employees to the total number of employees, excluding information pertaining to top echelon positions, such as executive, management, and board roles. The label is: w_employees_i,t.

Previous literature has assessed the presence of women in top echelon positions using various measures, such as the critical mass of female directors (Fernandez-Feijoo et al., 2014), the percentage of women on the BoD (Birindelli et al., 2019; Romano et al., 2024), and the ratio of women on TMT (Graafland, 2020). Building on prior studies (Cole et al., 2025; Sandretto et al., 2025), the percentage of female managers (top, senior, middle, and junior) over total managers was used as our main explanatory variable. We argue that this metric offers a valuable indicator of gender diversity across management levels and a reasonable proxy for the prevalence of female leadership within the organisation. The label is: w_managers_i,t.

The presence of WLB policies was assessed using flexible working hours (Lewis and Humbert, 2010). Following prior research (Birindelli et al., 2019), we used a dummy variable coded as 1 if the company offers programs or processes designed to assist employees in achieving a balance between their work and personal lives. The earlier programs may encompass flexible work arrangements, such as telecommuting, flexible working hours, job sharing, and reduced or compressed work weeks. Conversely, the dummy variable is equal to 0 if the company does not demonstrate a commitment to WLB practices. The label is: flexible_working_hours_i,t.

We also considered a comprehensive set of board-related variables, such as the presence of women on the BoD (female_bd_i,t), the ratio of female executive directors (w_direct_executive_i,t), and board size (bd_size_i,t) (Adams and Ferreira, 2009; Farrell and Hersch, 2005). The percentage of women on the BoD and the proportion of female executive directors reflect the level of gender diversity within the company, covering crucial indicators about the inclusive nature of corporate governance. Board size, defined as the total number of directors, may significantly influence gender diversity within the boardroom (Joecks et al., 2013). Moreover, we incorporated the Return on Equity (roe_i,t) as a financial control variable, which is calculated as net profit divided by total equity (Jahmane and Gaies, 2020; Rao et al., 2023). This proxy is commonly utilised to measure firm financial performance, thus providing insights into how board composition and gender diversity may correlate with financial outcomes. By integrating both governance and financial dimensions, our analysis seeks to embrace a holistic overview of the factors influencing the recruitment of female employees within organisations.

A panel data regression model utilising the STATA command xtreg was run to analyse the temporal dynamics within cross-sectional data and incorporating individual effects to address unobservable heterogeneity (Baltagi, 2021). We applied the Hausman test to assess the suitability of using fixed or random effects. The significant p-value from the Hausman test indicated a systematic difference in the coefficients of the two models, leading us to conclude that fixed-effects estimation was more appropriate for our study (Hausman, 1978).

To address potential time effects that could influence our results, we incorporated year fixed effects into our regression models (namely, by including the STATA command i.year into the syntax) (Christodoulou and Sarafidis, 2017; Lian et al., 2023). This approach allows us to control for any unobserved year-specific factors, such as macroeconomic trends or policy changes, that might affect gender diversity across all companies in our sample. Moreover, we employed firm-level clustered standard errors to account for potential autocorrelation within firms over time. This ensures that our standard errors and p-values are robust to potential time-series dependence in the data. In particular, the regression coefficients were estimated using robust standard errors to address potential autocorrelation and heteroscedasticity issues, with adjustments made using the STATA command vce(cluster id) (Wursten, 2018).

To explore the first hypothesis (H1), the following econometric model was performed:

To test the second hypothesis, we build the interaction terms as follows (Baron and Kenny, 1986; Hayes, 2017):

1. interact_term_1_i,t = w_managers_i,t × flexible_working_hours_i,t

2. interact_term_0_i,t = w_managers_i,t × (1 – flexible_working_hours_i,t)

By splitting the dummy variable in this way, we can explore how the effects of w_managers_i,t differ under varying conditions of flexible_working_hours_i,t.

Therefore, the model for testing the moderating hypothesis is formulated as follows (H2):

Table 2 shows the descriptive statistics of the variables encompassed in our research design (Figure 1), offering insights into the distribution and central tendencies over a sample of 512 firm-year observations.

The dependent variable (w_employees_i,t) ranges from 21.50 to 44.00, resulting in a mean of 31.85 and a standard deviation of 5.40, highlighting moderate variability in workforce size across units. The maximum value of the main explanatory variable (w_managers_i,t) is 33.00, while the mean is 23.83. A standard deviation of 4.14 suggests a relatively consistent managerial presence in the observed units.

The binary moderating variable (flexible_working_hours_i,t) has a mean value of 0.56 and a standard deviation of 0.40, reflecting variability in the adoption of WLB policies.

The control variables exhibit a wide distribution of values across the sample. The percentage of female_bd_i,t ranges from 0 to 57.14, while the proportion of w_direct_executive_i,t spans from 0 to 55.56. Board size (bd_size_i,t) varies from 3.00 to 20.00, and ROE (roe_i,t) ranges from −4.57 to 1.43.

Table 3 highlights the Pearson correlation matrix among independent and control variables, alongside their corresponding Variance Inflation Factors (VIFs) to test for possible multicollinearity concerns. Since some coefficients – inter alia, statistically significant – are not very relevant (i.e. below the critical threshold of |0.40|), multicollinearity does not pose a significant weakness regarding the reliability of our empirical evidence. Moreover, VIF values range between 1.14 and 1.21, suggesting that multicollinearity is not a relevant issue (Hair, 2014).

Table 4 sets out the results of the regression models for panel data examining the relationship between women managers and the percentage of women employees, alongside the moderating effect of flexible working hours. Model 1A focuses on the basic relationship between w_managers_i,t and the w_employees_i,t, uncovering a significant positive interplay (β = 0.36; p < 0.01). Such empirical evidence supports our H1, according to which a greater gender diversity in management roles positively conditions female workforce representation.

Model 2A incorporates the interaction terms to assess the moderating effects of flexible_working_hours_i,t (i.e. interact_term_1_i,t and interact_term_0_i,t). The interaction term interact_term_1_i,t shows a substantial positive relationship (β = 0.48; p < 0.01). Therefore, the presence of WLB policies strengthens the positive relationship between women in leadership positions and the gender diversity within their workforce. Conversely, the interaction term interact_term_0_i,t does not yield significant findings. This corroborates our H2 by emphasising the relevant role that supportive WLB policies play in maximising the effectiveness of female managers.

The control variables hold similar trends across both models. The variable bd_size_i,t exhibits a significant negative coefficient (β = – 0.38; p < 0.10), indicating potential barriers to gender diversity in larger companies with industry-specific dynamics, such as high-tech companies (Brieger et al., 2019). Meanwhile, roe_i,t consistently shows a significant positive relationship (β = 0.72; p < 0.05), further substantiating the interplay between enhanced financial performance and increased gender diversity in the workplace. Besides, the control variables w_direct_executive_i,t and female_bd_i,t did not exibit statistically significant effects on the percentage of w_employees_i,t in either model, suggesting that the presence of women in executive positions and on the BoD does not directly translate into an increase in the proportion of women employees within the organisation.

In our post hoc analysis, we conducted several tests to substantiate the goodness of our empirical evidence. First, the Hausman test results are significant, confirming that the fixed-effects model was appropriate for addressing unobserved heterogeneity. Second, the LM-Poolability test outcomes indicate that pooling the data would not be suitable, reinforcing the integrity of the analytical approach employed and validating the panel data approach. Third, the Modified Wooldridge test findings showed a p-value of <0.01, highlighting the presence of serial correlation in the residuals of the regression models. Serial correlation can lead to biased standard errors, which can affect the reliability of hypothesis testing. However, we addressed this issue by using robust standard errors to address potential autocorrelation and heteroscedasticity issues in our regression analyses (Born and Breitung, 2016). Finally, the Mean VIFs (i.e. 1.13 for Model 1A and 1.18 for Model 2A) indicate that multicollinearity among the independent variables is not a relevant concern (Hair, 2014).

To further corroborate our empirical evidence, Figure 2 plots the moderating effect of flexible_working_hours_i,t on the relationship between w_managers_i,t and w_employees_i,t. Two trend lines emerge: one for companies with flexible working hours (True in pink; interact_term_1_i,t) and another for those without flexible working hours (False in dark blue; interact_term_0_i,t). As evidenced by our quantitative results (Table 4), the interaction between WLB policies and the percentage of women managers further strengthens the positive effect on the share of women employees. In other words, WLB policies may act as a catalyst that boosts the influence of female managers on the representation of women in the workforce.

Table 5 shows a series of robustness checks by employing lagged independent variables to examine the temporal dynamics between women managers, WLB policies, and female workforce representation. In more detail, we incorporated one-year (L1), two-year (L2), and three-year (L3) lagged values for the percentage of women managers (L1.w_managers_i,t-1, L2.w_managers_i,t-2, L3.w_managers_i,t-3), flexible working hours (L1.flexible_working_hours_i,t-1, L2.flexible_working_hours_i,t-2, L3.flexible_working_hours_i,t-3), and their interaction terms (L1.interact_term_1_i,t-1, L2.interact_term_1_i,t-2, L3.interact_term_1_i,t-3, and L1.interact_term_0_i,t-1, L2.interact_term_0_i,t-2, L3.interact_term_0_i,t-3) into our regression models. This approach serves two key purposes.

First, lagging exogenous variables is a typical econometric technique to estimate endogenous predictors (Kennedy, 2008). Following previous studies (Cabeza-García et al., 2018; Romano et al., 2024; Sandretto et al., 2025; Song et al., 2025), we safeguard the robustness of our findings by addressing potential reverse causality issues. In particular, endogeneity arises when the independent variable is correlated with the error term (Antonakis et al., 2014); while reverse causality is a specific type of endogeneity where the dependent variable might affect the independent variable (Bellemare et al., 2017). Lagging the independent variable(s) helps mitigate this concern by ensuring that the explanatory variable precedes the dependent variable in time, reducing the likelihood that the current value of the dependent variable is influencing the current value of the independent variable. By proving that our results hold even when using lagged independent variables, we demonstrate that our findings are not simply due to reverse causality or other forms of endogeneity.

Second, analysing different time lags (i.e. L1, L2, and L3) allows us to better grasp the temporal dynamics among these relationships, revealing the duration over which these effects unfold. Specifically, in Model 1A, L1.w_managers_i,t-1 is positive and significant (β = 0.15; p < 0.10), pointing out that an increase in the proportion of women managers in the previous period has positively influenced the percentage of women employees in the current period. In Model 1B, L1.interact_term_1_i,t-1 shows a significant and positive coefficient (β = 0.30; p < 0.01), indicating that the presence of flexible working hours reinforces the baseline relationship. By contrast, L1.interact_term_0_i,t-1 is not statistically significant, suggesting that the absence of WLB policies does not moderate the basic relationship. Focussing on Model 2B, L2.interact_term_1_i,t-2 yields a significant and positive coefficient (β = 0.17, p < 0.05), suggesting that the presence of flexible working hours policies strengthens the positive relationship between female managers and the percentage of female employees. Furthermore, this analysis indicates that there are diminishing statistically significant effects when looking at longer-term horizons (see Models 3A-3B). Thereon, we can assert that the significance of effects is maintained at the L2, but lost with L3. This hints that the more impactful and positive effect of women managers and WLB policies on female workforce representation is likely to materialise within the L1-L2 timeframe.

Taking all empirical evidence together, the robustness and reliability of our results are further corroborated.

Our findings prove that women in TMT positively sway the percentage of women employees and that WLB policies significantly strengthen this relationship. These insights are particularly relevant in advanced technologies and AI-related sectors, where the gender gap remains a critical challenge and where addressing structural barriers is crucial for fostering gender equity. We argue that this dual influence operates through distinct but complementary mechanisms.

First, the positive association between women managers and female workforce representation (H1) reflects two interrelated dynamics. On the one hand, in male-dominated environments where leadership is still implicitly coded as masculine (Koenig et al., 2011; Ezzedeen et al., 2015), the visibility of women in managerial roles serves as a powerful symbolic cue. Their presence challenges prevailing stereotypes and signals that strategic leadership is attainable for women, thereby reducing stereotype threat and fostering a sense of belonging (Stout and Dasgupta, 2011; Dasgupta and Asgari, 2004). On the other hand, women leaders – often navigating the role incongruity between traditional gender expectations and leadership roles in male-dominated sectors (Eagly and Karau, 2002) – may develop heightened awareness of the structural and normative barriers facing other women (Burkhardt et al., 2020; Eagly and Carli, 2009). This awareness can translate into strategic advocacy, whereby women in leadership positions influence recruitment, retention, and promotion practices to support gender diversity (Cook and Glass, 2014; Liu et al., 2024). These twofold mechanisms – namely, symbolic role-modelling and strategic advocacy – help explain the observed increase in female workforce representation in firms with greater management gender diversity. Interestingly, our findings also show that the presence of women in executive positions or on BoDs does not significantly affect the proportion of female employees. This aligns with prior research suggesting that management members, due to their operational and strategic responsibilities, are more directly positioned to shape internal culture and influence day-to-day practices that affect female workforce composition (Matsa and Miller, 2011; Hoobler et al., 2018; Ali et al., 2021).

Second, our findings show that the positive effect of women managers on female workforce participation is significantly increased in firms that offer WLB policies, thereby supporting H2. This moderating effect is not simply a matter of organisational support, but is theoretically grounded in SRT's emphasis on caregiving expectations as a central component of gender roles (Wood and Eagly, 2002; Eagly and Karau, 2002). In high-tech sectors – where predominantly male work cultures often valorise long hours, constant availability, and uninterrupted career paths (Blair-Loy and Jacobs, 2003; Perrigino et al., 2018) – these caregiving norms introduce a second, compounding layer of role incongruity for women. WLB policies help mitigate this specific tension by addressing the caregiving-related constraints that disproportionately affect women's ability to participate and advance in the workforce. In this way, they heighten the symbolic and strategic influence of women managers by making it more feasible for other women to reconcile professional and caregiving roles.

Our lagged analysis reveals that the positive effects of women managers and WLB policies on female workforce representation are statistically significant within a one-to two-year window, but not beyond. This temporal pattern suggests that while inclusive leadership and supportive policies can yield meaningful improvements in the short to medium term, their influence tends to plateau over time. One possible interpretation is that these firm-level interventions are effective up to a certain threshold, beyond which broader societal expectations and structural constraints begin to reassert themselves. In this vein, our findings highlight a critical boundary condition: organisational efforts, though necessary, may not be sufficient to sustain long-term gains in gender representation unless accompanied by parallel shifts in cultural norms, public policy, and institutional frameworks. For instance, the World Economic Forum (2024) highlights persistent gender disparities in digital skill acquisition, which shape how men and women engage with technological transitions and future workforce opportunities. Although the share of women with STEM skills has increased modestly – from 24.4% in 2016 to 27.1% in 2024 – this progress remains inadequate to close the gap. Parity is especially low in high-growth fields such as AI and big data (30%), programming (31%), and cybersecurity (31%). These disparities are mirrored in workforce representation and leadership trajectories. While women occupy nearly half of entry-level positions globally, their presence drops sharply at higher levels, with only one-quarter of C-suite roles held by women. Alarmingly, the rate of women's hiring into leadership positions has declined from 37.5% in 2021 to 36.4% in early 2024, falling below pre-pandemic levels. Thereon, this interconnectedness of challenges – spanning digital skills, workforce participation, and leadership access – elucidates that to achieve sustained gender equity in the workforce, firm-level initiatives must be embedded within a broader ecosystem of cultural and institutional change.

This study offers a novel contribution by integrating two research domains that have largely developed in parallel: gender representation in strategic leadership and the implementation of WLB policies. By examining their intersection within the high-tech sector – a context marked by entrenched masculine norms and heightened social role tensions (Koenig et al., 2011; Li and Chan, 2024) – we provide novel insights into how symbolic and structural levers can jointly foster gender inclusion. Furthermore, this manuscript expands the scope of SRT (Dawson, 1997; Eagly, 1997) by applying it to a sector where cultural expectations around gender roles are particularly pronounced (Wood and Eagly, 2002; Eagly and Karau, 2002). In doing so, it addresses multiple layers of role incongruity: firstly, the tension experienced by women in leadership roles within male-dominated environments (Eagly and Carli, 2009; Tokbaeva and Achtenhagen, 2023); and secondly, the additional strain imposed by societal caregiving expectations (Blair-Loy and Jacobs, 2003; Heilman, 2012). Importantly, the study enriches theoretical understanding by identifying two distinct yet complementary mechanisms through which female leadership can enhance women's workforce participation. The first is symbolic and strategic – where women in TMT act as role models and advocates for inclusion (Cook and Glass, 2014; Dezsö and Ross, 2012). The second is instrumental – where the presence of WLB policies augments the effectiveness of inclusive leadership by addressing practical barriers to participation (Perrigino et al., 2018; Bertola et al., 2023). Together, these findings contribute to a deeper understanding of how gender equity can be promoted through the interplay of leadership representation and supportive organisational practices.

From a practical perspective, our findings highlight the importance of creating organisational contexts in which female leaders can effectively exercise strategic authority and visibility to promote gender diversity. Notably, our regression analyses point out that the presence of women in executive or board-level roles – measured through variables such as female executive directors (w_direct_executive_i,t), and women on boards of directors (female_bd_i,t) – does not, in itself, significantly predict higher levels of female workforce participation (Tables 3 and 4). This underscores that representation alone is insufficient unless accompanied by meaningful opportunities to influence organisational policies and culture. Conversely, our results highlight that women in strategic positions, as in TMTs, serve as both symbolic role models and strategic advocates for inclusion, challenging prevailing gender norms and signalling that leadership is accessible to women (Cook and Glass, 2014; Dezsö and Ross, 2012). Their presence can reduce stereotype threat, elevate aspirations among female employees, and foster perceptions of fairness and opportunity (Stout and Dasgupta, 2011; Joshi et al., 2015).

Besides, our findings underscore the relevance of aligning leadership diversity with institutional supports. WLB policies – such as flexible scheduling, remote work, and caregiving leave – can amplify the effectiveness of inclusive leadership by addressing the caregiving-related dimension of role incongruity (Eagly and Karau, 2002; Blair-Loy and Jacobs, 2003). Thus, organisations may benefit from normalising the use of WLB policies to mitigate the stigma often associated with flexibility. This includes integrating WLB practices into performance management systems, ensuring that their use does not signal reduced commitment or hinder career advancement (Rogier and Padgett, 2004; Bertola et al., 2023). Leadership visibility and endorsement are critical to legitimising these policies and encouraging uptake. In this scenario, firms should create an environment where female leaders can effectively advocate for diversity initiatives and enhance female representation. Without a supportive culture and collaborative engagement across all levels, even women in high-ranking positions may struggle to exert meaningful influence. To support their efforts, companies should consider implementing diversity training programs to increase awareness about gender-specific challenges and biases. Such initiatives can foster a more inclusive culture and promote understanding of the value that women bring to leadership positions and firm performance.

Finally, sector-specific considerations are warranted. In high-tech environments – frequently dominated by a work culture where long hours, constant availability, and uninterrupted career paths are valorised (Mickey, 2022; Cooper, 2000; Neely et al., 2023; Petrucci, 2020) – addressing role incongruity is particularly urgent. Firms in these sectors may consider tailoring WLB initiatives to the specific demands of innovation-driven work cultures, ensuring that flexibility is both feasible and culturally accepted.

While this study enriches the literature on gender diversity in TMT and the organisational conditions supporting inclusion, it is not without limitations. First, the analysis focused exclusively on publicly listed companies headquartered in the European Union. This narrows the generalisability of our findings to private firms or organisations working in other institutional and cultural contexts. Even though the empirical analysis focuses on the European listed firms, the organisational and cultural dynamics shaping gender inclusion may markedly differ in non-European regions and emerging economies, where varying gender norms, institutional arrangements, cultural expectations, and policy frameworks plays a significant role (Anglin et al., 2022; Eagly and Carli, 2009; Khalaf et al., 2024; Mansour et al., 2025). Recent works from geographically different settings - for instance, research on board gender diversity in Asian firms (Saleh et al., 2025) - highlighted the strategic relevance of gender-inclusive leadership beyond Western contexts. In this vein, future research could further develop the earlier line of inquiry, exploring the role of women across multiple leadership levels in non-Western contexts. A comparative, cross-cultural research agenda would be instrumental in deepening theoretical understanding and enhancing the global relevance of gender diversity scholarship. Second, our study is confined to the high-tech sector – a context characterised by entrenched masculine norms and heightened role incongruity (Mickey, 2022; Cooper, 2000; Neely et al., 2023). Although this focus enhances theoretical relevance, it limits the applicability of our results to industries with different gender dynamics. Comparative studies across sectors could provide a more nuanced understanding of how sector-specific cultures interact with leadership diversity and structural supports. Third, the use of secondary data imposes some constraints. Our proxy for women managers is based on the percentage of female managers, which may not fully capture the strategic influence of top-level executives. Similarly, the binary coding of WLB policies does not reflect variation in policy design, implementation quality, or organisational support. Future research could benefit from more granular data, including qualitative insights into how these policies are perceived and enacted within firms (Perrigino et al., 2018). Fourth, although we employ fixed-effects models and lagged variables to mitigate endogeneity concerns, the observational nature of our data limits causal inference. Unobserved factors, such as organisational culture, leadership style, or internal advocacy networks, may influence both the adoption of WLB policies and gender diversity outcomes. Indeed, although the use of both fixed-effects models and lagged independent variables tackles key concerns regarding reverse causality and omitted variable bias (Cabeza-García et al., 2018; Song et al., 2025), future studies may consider the adoption of instrumental variable techniques, such as Two-Stage Least Squares (2SLS) or Generalised Method of Moments (GMM), to further strengthen the robustness of the results (Antonakis et al., 2014; Bellemare et al., 2017). Furthermore, mixed-methods approaches, including interviews or case studies, could help uncover these latent dynamics and enrich our understanding of causal mechanisms.

To infer, future research may explore additional moderators and mediators, such as diversity training, inclusive leadership behaviours, or external pressures (e.g. regulatory mandates or investor expectations). Investigating how macroeconomic conditions, industry shocks, or technological disruptions shape the relationship between leadership diversity and workforce composition could also yield valuable insights.