The gender pay gap in the board of directors: a group-specific analysis Open Access

Gender diversity and equal pay within corporate boardrooms have emerged as pivotal issues with far-reaching implications for organizational performance, governance and societal equity (Adams and Ferreira, 2009; Bravo and Alcaide-Ruiz, 2019). The board of directors, central to corporate decision-making and strategy formulation, stands under the scrutiny of scholars, policymakers and stakeholders alike. Amidst the multifaceted dimensions of board dynamics, the gender pay gap (GPG) within this influential echelon calls for a meticulous analysis and understanding (Aavik et al., 2023; Cook et al., 2018; Kulich et al., 2011; Yanadori et al., 2016).

While GPG research has proliferated in recent years, particularly in the context of corporate organizations (Adams and Ferreira, 2009; Goh and Gupta, 2016), studies focusing specifically on the board of directors remain scarce. Existing research predominantly addresses GPGs at the employee level, shedding light on disparities across various organizational tiers (Bertrand and Hallock, 2001; Grund, 2015; Jiang et al., 2019). However, only some studies have ventured into the intricacies of this phenomenon within the upper echelons of corporate governance, identifying salary gaps in certain managerial positions (García Martín and Herrero, 2019; Glass and Cook, 2018; Hillman and Dalziel, 2003; Pucheta-Martínez and Bel-Oms, 2015). The lack of comprehensive investigations in this area underscores a critical gap in understanding gender disparities in corporate leadership, needing further exploration.

This paper aims not only to identify the existence of a gender pay gap among board members, but also to examine whether such disparities are linked to systematic differences in the allocation of executive responsibilities, performance expectations and compensation structures between male and female directors. By doing so, the study sheds light on persistent forms of gender inequality that go beyond representation and reflect subtle imbalances in how professional roles and rewards are distributed within corporate leadership. The aim of this study is to investigate the GPG among directors in Spanish companies listed on the Madrid Stock Exchange between 2013 and 2021. In Spain, the Royal Legislative Decree 1 / 2010 of 2 July categorizes directors into executive directors (those who perform top management duties within the company); proprietary directors ^[1] (appointed as shareholders or their representatives); independent directors (engaged in external advisory roles to provide supervision, experience and expertise to the company, without being conditioned by their relationships with the company, its shareholders or its managers) and other directors (those who do not fit into the previous categories). Hence, based on prior research, the analysis differentiates between director’s categories (García Martín and Herrero, 2019). In addition, in line with recommendations in prior research, this study disaggregates director compensation (García Martín and Herrero, 2019; Pucheta-Martínez and Bel-Oms, 2015), offering a holistic view of the GPG and contributing to sustainable corporate success and broader societal progress.

In relation to salary components, these vary depending on the director’s type and role. Executive directors typically receive a mixed remuneration, consisting of a fixed salary and a variable component linked to performance and the achievement of strategic objectives. Proprietary directors, who represent shareholders, may also receive a variable component, though to a lesser extent, as their main role is oversight. In contrast, independent directors focus on supervision and mitigating agency conflicts, resulting in predominantly fixed to maintain their objectivity and independence, as recommended by the CNMV’s Code of Good Governance (CNMV, 2020).

In Spain, gender quotas in corporate governance have transitioned from voluntary recommendations (soft law) to legally binding regulations (hard law). Organic Law 3/2007 was a turning point, promoting 40% female representation in decision-making bodies by 2015. Also, the CNMV’s Code of Good Governance (CNMV, 2020) encouraged companies to voluntarily increase the presence of women on their boards setting a target of 40% female representation by 2022 (30% by 2020 in the previous Code). However, limited progress under these recommendations prompted the implementation of legislative measures. More recently, Directive (EU) 2022/2381 has introduced mandatory quotas requiring at least 40% representation of each gender on boards and executive committees by 2026. Spain must transpose this directive into its legislation by the end of 2024. This transition reflects Spain’s commitment to addressing structural barriers to gender equality in leadership by combining soft law principles with enforceable measures to achieve tangible results. In addition, the 2020 Code of Good Governance also recommends that the majority of board members should be external (including both proprietary and independent directors). Specifically, it suggests that at least 50% of the board members should be independent directors, except in the case of companies that are not of high capitalization, or those that, while being of high capitalization, have shareholders holding more than 30% of the capital. This recommendation aims to ensure external oversight and balance in the decision-making processes of companies.

This approach will not only advance academic understanding but also offers practical guidance for promoting gender equality in corporate leadership, which ultimately will foster innovation and more effective decision-making (Adams and Ferreira, 2009; Bravo and Alcaide-Ruiz, 2019; Kulich et al., 2011; Yanadori et al., 2016). The results demonstrate the existence of a GPG among executive directors in terms of fixed, variable and total compensation, although there is no gap observed among proprietary or independent directors.

These findings contribute to the ongoing research on the relationship between GPG and director compensation in several ways. First, unlike prior studies that often examine gender pay disparities as a singular phenomenon, this analysis disaggregates compensation into fixed and variable components and analyses these differences across executive, proprietary and independent directors. This approach allows to identify whether GPG stems from structural inequities in fixed salaries, performance-based incentives or both. Second, while the relationship between GPG and director compensation has been extensively studied in Anglo-Saxon contexts (Bertrand and Hallock, 2001; Carter et al., 2017; Cook et al., 2018; Geiler and Renneboog, 2015; Yanadori et al., 2016), limited research has addressed this issue in Spain, a country with a unique regulatory framework promoting actively gender diversity on corporate boards. This study contributes to the literature by exploring the GPG within this context, shedding light on whether such policies effectively address disparities in director remuneration. Third, this study demonstrates that GPG is not uniform across director categories. While executive directors exhibit a pronounced pay gap in variable compensation, proprietary and independent directors show no significant differences, suggesting that independence from firm performance reduces gender biases. This nuanced view emphasizes the importance of director roles in shaping pay structures and highlights areas where further interventions to reduce GPG could have the greatest impact.

The rest of the paper is structured as follows: Section 2 focuses on the literature review and hypothesis development. Section 3 describes the methodology used and the research design. Section 4 presents and discusses the main findings. Finally, Section 5 includes the conclusions reached.

The study of salary differences between male and female directors has become a crucial issue in understanding the broader dynamics of corporate governance and board diversity (Cole and Mehran, 2016; Schneider et al., 2021). Resource dependency theory offers a valuable framework for exploring how the skills, experiences and relationships that directors contribute to the board can influence compensation. This theory involves selecting candidates who bring diverse factors to the board to enhance the effectiveness of the company. In this context the diversity of the board is essential to obtain different inputs in terms of formation, expertise and relationships with the environment and other organizations (Hillman and Dalziel, 2003).

In relation to salaries, this theory is linked to variable components of compensation, where directors will receive higher salary depending on their contributions to the board, and therefore, the firm. Consequently, studying the determinants of salary differences is relevant, especially when compensation has a fixed and variable component dependent on the characteristics of each individual.

Women, due to gender stereotypes, have faced limited access to the resources necessary for relevant leadership roles, such as education in business management (Olson, 2013), a solid network of contacts (Rudman and Glick, 1999), and negotiation skills (Kulik and Olekalns, 2011). This limited access translates into a reduced capacity to negotiate higher variable salaries, as according to the resource dependency theory those who bring more resources – expertise, training and relationships – to the company receive higher variable compensation (Gaertner et al., 1989; Kulik and Olekalns, 2011; Olson, 2013; Rudman and Glick, 1999). Considering women’s backgrounds it is expected that salary differences between men and women exist (Bell, 2005; Bertrand and Hallock, 2001; Carter et al., 2017; Elkinawy and Stater, 2011; Hutchinson et al., 2017; Muñoz-Bullón, 2010; Vieito and Khan, 2012), especially in those categories of directors where the variable component of the salary carries an important weight (Geiler and Renneboog, 2015; Kulich et al., 2011).

Compensation of executive directors often includes a variable salary component. Executive directors have a direct impact on the strategy and financial performance of the entities, which, combined with their access to networks of power, strategic information and control over critical resources, makes them perceived as more valuable and thus eligible for higher incentives. Given gender disparities in negotiation, access to resources and other aspects related to the resource dependence theory, it is expected that men will receive a higher variable component in their salaries. It is also important to investigate whether these salary disparities also extend to the fixed component of compensation. Generally, empirical evidence suggests that male executive directors tend to receive higher compensation in board of directors (Bertrand and Hallock, 2001; Cook et al., 2018; Schneider et al., 2021). Nonetheless, it is still an open question whether this disparity can be solely attributed to individual characteristics where gender plays an indirect role, or whether gender itself constitutes a direct and significant determinant of directors’ remuneration. Accordingly, examining the fixed salary component, where individual skills, experience and other personal characteristics are not expected to exert a significant influence, is of critical importance. Therefore, the following hypothesis is proposed:

Non-executive directors, particularly proprietary and independent directors are other relevant figures on corporate boards. Proprietary directors play a significant role in the Spanish context, holding 31.6% of the board seats in companies listed on the Madrid Stock Exchange in 2021 (CNMV, 2021). As shareholders, proprietary directors are perceived as more committed to the company’s performance and actively engaged in supervisory and control functions (Acero and Alcalde, 2023). Their dual role as both investors and overseers positions them as key contributors to the board’s monitoring capacity (Garcia-Sanchez et al., 2014). Given their unique position as representatives of significant shareholders, proprietary directors typically receive fixed salaries as remuneration for their board responsibilities (García Martín and Herrero, 2019), which reduces the scope for discretionary or incentive-based pay schemes (Arrondo et al., 2008). Their role emphasizes collective oversight and aligns closely with shareholder interests, minimizing the potential impact of individual differences, such as varying levels of skill or experience, on their compensation.

From a gender perspective, the limited discretion in proprietary directors’ compensation design and the collective nature of their oversight responsibilities imply that the potential for gender-based pay disparities is lower compared to executive roles. Because their appointment is primarily linked to their ownership position, and their remuneration is less sensitive to individual performance, any gender differences in pay are expected to be minimal or non-existent (Geiler and Renneboog, 2015). Consequently, GPG is less expected in this group, at least not due to personal attributes or performance outcomes. Therefore, the following hypothesis is proposed:

In contrast, independent directors are appointed based on their expertise, experience and the resources they bring to the board, which aligns with the resource dependence theory (Goh and Gupta, 2016). Their primary function is to mitigate agency conflicts by acting as impartial intermediaries between management and shareholders, ensuring that managerial actions align with the interests of the broader shareholder base (Jensen and Meckling, 1976). Unlike executive directors, whose compensation often includes significant variable components tied to financial performance, independent directors are typically compensated through fixed fees. This structure aims to preserve their independence and objectivity by minimizing any potential conflict of interest that could arise from financial incentives tied to company performance. Given their primary oversight role rather than operational management, the compensation of independent directors is designed to focus on governance rather than short-term financial outcomes.

From a gender perspective, the absence of variable compensation and the standardized nature of remuneration for independent directors suggests that the GPG is less likely to manifest in this group. In fact, their selection is often based on objective criteria such as experience, qualifications and reputation, which reduces the influence of subjective biases that may contribute to gender-based disparities (Hillman et al., 2007; Pucheta-Martínez and Bel-Oms, 2015; Terjesen et al., 2009). Consequently, while GPGs are well documented among executive directors, particularly in variable compensation linked to resource control and negotiation dynamics (Cook et al., 2018; Elvira et al., 2023; Muñoz-Bullón, 2010; Yanadori et al., 2016), we assert the absence of a significant GPG among proprietary and independent directors. This distinction underscores the need to differentiate between executive and non-executive roles when analysing gender-based disparities in boardroom compensation. Therefore, the following hypothesis are proposed:

Our database comprises a sample of panel data that includes information on all directors who belong to the board of directors of Spanish companies listed on the Continuous Market of the Madrid Stock Exchange during the period 2013–2021. 2013 was the first year in which the National Stock Market Commission published a standardized remuneration document for listed companies (Order ECE/461/2013, of 20 March).

Companies belonging to the financial sector – namely, banks, insurance companies and investment companies – are not considered in this study due to their special characteristics, as they could affect their remuneration policies. These companies are subject to special scrutiny by financial authorities that constrains the role of their board of directors and their special accounting practices. In addition, companies undergoing liquidation process were not considered, as liquidation could lead to abnormal behaviour in remuneration policies and board composition. Finally, listed companies lacking information on their consolidated annual accounts, which is necessary to know their real situation, were also not included in this study.

Table 1 presents this data-cleaning process at firm level. After this process, the final sample consists of 9,941 director-year observations.

Data related to the compensation variables were manually collected from annual remuneration reports, while information corresponding to individual director characteristics – such as gender, factor time, tenure, committee membership, CEO or chairperson position, PhD studies and relationships – and board of directors in general – board size, ownership held by the board, independence of the nomination and compensation committee, presence of women on the nomination and compensation committee, and application of good remuneration practices index – were extracted from annual corporate governance reports. For both documents, which are part of the annual report, consolidated data were chosen. Finally, firm characteristics – firm size, leverage and performance – were obtained from the SABI database.

Directors’ individual characteristics (e.g. tenure and education) were complemented with board- and firm-level variables to account for the structural and organizational context in which compensation decisions are made. Each director was assigned the characteristics of their corresponding board and firm for the year of observation. This multi-level approach acknowledges that remuneration outcomes are shaped by both individual and contextual factors, aligning with prior research in corporate governance that integrates individual and organizational determinants of compensation (Adams et al., 2010; Goh and Gupta, 2016).

Regarding the variable of director’s compensation, the total remuneration of directors was first examined to distinguish between fixed and variable compensation. However, total compensation includes a third remuneration component called other compensation.

We have chosen not to include this compensation in this study for two main reasons. First, it is rarely used, with fewer than 10% of directors in the sample receiving it, which limits its representativeness and makes it less suitable for a comprehensive analysis of director compensation. Second, the discretionary nature of this component means it varies significantly across companies and years, as it includes non-standard payments such as redundancy payments and in-kind compensation. This variability makes difficult to identify consistent patterns and could introduce considerable noise into the analysis, compromising the robustness of the results. Therefore, we decided to exclude this variable to focus on more consistent and measurable components of compensation, which enhance the validity and comparability of the findings.

All compensation variables were measured in thousands of euros and were also winsorized at 1% to neutralize the effect of outliers:

The Fixcomp variable, which represents fixed compensation, consists of four remunerative concepts: base salary, fixed wages, attendance fees and remuneration for membership. Base salary represents the compensation earned by directors for their executive duties, excluding remuneration received due to their status as director. Fixed wages include the amount of cash compensation earned by directors for belonging to the board, regardless of their actual attendance at board meetings. Attendance fees include compensation derived from attending board meetings and, where applicable, from their committees. Remuneration for membership takes into account earned amounts according to the number of committees in which the director participates.

The Varcomp variable, which presents the variable compensation, is also computed as the sum of four components: short-term bonus, long-term bonus, equity-based pay and long-term incentive plans. Short- and long-term bonus include the variable wages accrued for a period of one year or less – for short-term bonus – or more than one year – for long-term bonus – which are linked to the performance or the achievement of certain objectives. Equity-based pay represents the amounts accrued through remuneration plans involving delivery of shares, stock options or payments related to the value of the shares. Long-term incentive plans include retirement compensation and any other survival benefits partially or fully financed by the company.

The Othcomp variable includes redundancy payments and other forms of compensation, such as in-kind wages.

The main variable under study is the gender of directors. A dichotomous variable takes the value of 1 if the director is a woman, and 0 otherwise.

The control variables in this study are related to the individual characteristics of directors, as well as characteristics of the board of directors’ and the firm. Individual characteristics include factor time (factime), tenure (tenure), committee membership (committees), CEO position (CEO), chairperson position (chairperson), educational qualifications (PhD) and relationships (Relationships). Board of directors characteristics chosen were board size (Board_Size), ownership structure (Board_Own), independence of the nomination and compensation committee (NCC_Indep), presence of women in the nomination and compensation committee (NCC_Women), good remuneration practices index (GRP_Index) and CEO duality (CEO_duality). Firm characteristics include firm size (Firm_Size), leverage (Leverage) and performance (performance).

Table 2 summarizes the set of variables used in the analysis, as well as the measurement used.

The first control variable, factime, measures the time the director has been on a board for a given year. This variable takes values between 0 and 1, where 1 is assigned to the directors who remain in the company during the entire year (365 days). As far as we know, research has not yet considered this factor. However, most of the compensation directors receive is directly related to the time they spent in the company. In addition, 40% of directors do not remain on the board for the entire year. Therefore, controlling the results using this variable is essential. Considering this, a positive relationship between factime and director’s compensation is expected.

Tenure includes, among other skills, the accumulated experience of the director within the company, and was computed through the logarithm of the number of years on the board. Directors with prior board experience can provide better advice to the board, as they have knowledge learned from previous experience in high-level decision-making (Beckman, 2006). Board experience provides familiarity with the company’s strategies and operations. In addition, it is a key factor in the promotion of directors, so it is expected that a longer tenure in a firm will provide access to positions of responsibility. Therefore, longer tenure is expected to have a positive relationship with compensation, as the director’s value is increased through within-firm experience and loyalty (Goh and Gupta, 2016).

These analyses also include committees, a variable indicating a director’s participation in board committees – namely, the executive committee, the nomination and remuneration committee, the audit committee and the corporate social responsibility committee. This variable was measured as the logarithm of the number of committees a director belongs to. These carry additional responsibilities to the basic function of a director and, therefore, they can provide additional remuneration in the form of meeting fees (Adams and Ferreira, 2008; Brick et al., 2006). Consequently, a positive relationship between participation in committees and remuneration is expected.

Following previous research on executive compensation (Carter et al., 2017; Elkinawy and Stater, 2011; Graham et al., 2011; Muñoz-Bullón, 2010; Vieito and Khan, 2012), two dummy variables that represent CEO (CEO) and chairperson positions (chairperson) are included. In addition to assuming additional tasks on the board, the positions of CEO and chairperson should require directors with extensive experience, training and skills to advise the company on decision-making and properly supervise the board’s tasks. Therefore, a positive relationship between these two variables and director remuneration is expected.

PhD is a binary variable equal to 1 if the director holds a PhD and 0 otherwise. Education, a human capital variable positively related to the director ability, has a clear effect on pay (Coelho Duarte et al., 2010). Knowledge can also provide more structured or critical approaches to decision-making or performance evaluation, suggesting that a director’s qualifications are associated with higher compensation (Goh and Gupta, 2016).

Relationships was computed as the logarithm of the number of boards a director belongs to. Relational capital in directors is a highly valued resource by companies. It provides additional contacts and business opportunities, and access to additional information and finance (Hillman and Dalziel, 2003). A director’s network enables them to establish more managerial influence-oriented connections while also bringing additional skills and knowledge. According to Renneboog and Zhao (2011), who found a positive relationship between CEO networks and remuneration, and Goh and Gupta (2016), who also found a positive relationship between networks of non-executive directors and remuneration, a positive association between director relationships and compensation is expected.

Relating to board characteristics, the first control variable was Board_Size, measured as the logarithm of the number of directors in the boardroom. The board of directors must have an appropriate size to effectively perform its functions with sufficient depth and range of opinions. In this vein, the Unified Code of Good Governance of 2020 (CNMV, 2015) recommends that a board of directors comprise between 5 and 15 members. On the one hand, smaller boards generate lower monitoring costs, as they tend to be more cohesive groups than larger ones, where costs are higher, and monitoring is more complex and less effective (Andreas et al., 2012). In addition, large boards may face free-riding problems in decision-making and control, diluting the monitoring incentives for their board members (Boone et al., 2007; Jensen, 1993). Geiler and Renneboog (2015) found a positive relationship between director compensation and board size. On the other hand, due to the major coordination problems existing in large boards, the CEO wields greater power and limits the directors’ compensation to discourage monitoring (Ryan and Wiggins, 2004). In line with this second argument, Ryan and Wiggins (2004), Brick et al. (2006) and Adams and Ferreira (2009) found a negative relationship between board size and total director compensation.

Ownership structure (Board_Own), defined as the proportion of shares held by the board of directors, is also included in this study. It can be argued that a higher percentage of ownership among director members may align their objectives with those of the shareholders and reduce the agency costs (Arrondo et al., 2008; Jensen and Meckling, 1976). Moreover, directors with significant ownership might have sufficient vested interests that they do not need to be compensated for their time, as changes in the value of their ownership interests outweigh the potential compensation received. Ozkan (2011) linked non-executive director ownership to restraint in CEO pay. Therefore, it is expected that the higher the shareholding of directors, the lower the compensation they receive. However, significant ownership may favour director entrenchment and raise their remuneration (Arrondo et al., 2008; Holderness and Sheehan, 1988).

Regulatory bodies propose that boards of directors should be composed of a large percentage of independent directors and, in addition, advise their presence in various committees. In Spain, the Unified Code of Good Governance (CNMV, 2020) recommends that at least half of the board of directors be independent and that committees mostly consist of independent directors. Thus, the independence of the nomination and compensation committee (NCC_Indep) is included in this study, measured as the percentage of independent directors in this committee, because it is responsible for designing the remuneration policy of the board. Ryan and Wiggins (2004) suggested that independent directors possess a bargaining advantage over the CEO that results in compensation more closely aligned with shareholders’ objectives. Therefore, and following Fernández Méndez et al. (2012) and Pucheta-Martínez and Narro-Forés (2014), it is expected that nomination and compensation committees with a higher proportion of independent directors will experience lower agency costs and, consequently, a lower remuneration received by directors. In line with this argument, Arrondo et al. (2008) and Andreas et al. (2012) observed a negative relationship between the percentage of independent directors and the total compensation received by directors.

This study also includes a binary variable (NCC_Women) equal to 1 if there is a woman in the nomination and compensation committee, and 0 otherwise. Shin (2012) indicated that the GPG in executive pay narrows when a higher percentage of women take part on the group responsible setting compensation, the compensation committee. Accordingly, a negative association between this variable and director’s compensation was expected, because one method to address the GPG is to reduce the compensation of male directors, who form the majority of the sample. Moreover, the presence of women on the board contributes to enhance corporate governance practices (Burgess and Tharenou, 2002; Nielsen and Huse, 2010), which means, among others, avoiding the excessive remuneration that some male directors receive.

An index of good remuneration practices (GRP_Index) was generated following Melón-Izco et al. (2020). The Spanish Codes of Good Governance provide a series of recommendations linked to the remuneration policies of companies. This variable measures the proportion of these specific recommendations that a company has fully or partially complied with, weighted by their importance. Recommendations fully complied with were assigned a weight of 1, while those partially complied were assigned a weight of 0.5. The index is computed as follows:

It was expected that companies with greater compliance with compensation policy recommendations would remunerate their directors in a stricter way, avoiding overly high compensations. Therefore, a negative relationship between good remuneration index and director compensation is expected.

CEO_duality is also a binary variable that takes the value of 1 for those board directors in which the CEO and the chairperson are the same person. This accumulation of power in a single person within the boardroom increases agency costs and the possibility of entrenchment, which can translate into higher remuneration levels (Brick et al., 2006; García-Meca, 2016). On the contrary, as previously suggested, a powerful CEO may exercise greater control over the board and therefore reduce the efficacy of directors’ monitoring through a lower remuneration (Ryan and Wiggins, 2004).

Regarding firm characteristics, the first control variable is Firm_Size, computed as the logarithm of the average number of employees (Renner et al., 2002). Larger companies tend to require more monitoring and are willing to offer much higher wages to hire and retain the best directors in the labour market (Brick et al., 2006). In addition, directors of large companies receive higher compensation due to the increased complexity of their tasks, the potentially higher value placed on their decisions, and, hence, the greater reward from making those decisions (Andreas et al., 2012; Brick et al., 2006). Therefore, and in line with the research of Ryan and Wiggins (2004), Arrondo et al. (2008), Andreas et al. (2012), Fernández Méndez et al. (2012), Amin et al. (2014), García-Meca (2016) and Goh and Gupta (2016), a positive relationship of company size and director compensation is expected.

Leverage, defined as the quotient between total liabilities and total assets, is also controlled for through the leverage ratio. On the one hand, higher levels of debt reduce agency costs associated with free cash flow by reducing the cash flow available for spending at the discretion of managers (Jensen, 1986). This argument allows to establish a negative relationship between indebtedness and director compensation (Andreas et al., 2012; Bryan et al., 2000). On the other hand, the most indebted companies require more monitoring due to the greater challenges involved in their management, resulting in higher director remuneration (Brick et al., 2006; López-Iturriaga et al., 2015). Depending on which of these two arguments carries more weight, the relationship of this variable with director compensation can be either negative or positive.

Performance was measured as the return on assets (ROA), defined as earnings before interest and taxes (EBIT) divided by total assets. According to agency theory, director compensation aligns the interests of directors with those of shareholders, thus avoiding the extraction of benefits. Following previous studies that found a positive relationship between performance and director compensation (Amin et al., 2014; Andreas et al., 2012; Arrondo et al., 2008; Fernández Méndez et al., 2012; Kulich et al., 2011), a positive relationship between firm performance and director remuneration was expected.

Finally, sector and year dummies were included as control variables to measure the industry and temporary effects in all proposed relationships.

To analyse the GPG in boards of directors, after conducting the Hausman test, a random effect panel data estimation model was used for the regression analysis, where compensation is regressed on explanatory and control variables. The model is represented as follows:

Where Compensation_it comprises total compensation, fixed compensation and variable compensation – measured as the log (1 + compensation variable) – for director i in year t. Gender_it represents the sex of director i in year t, throughout a dummy variable. CVj_it refers to the corresponding control variable j of director i in year t, as previously described. Finally, ɛ_it is the error term, which is split into three components: the individual effect (η_i), the temporal effect (d_t) and white noise or random disturbance (ν_it).

Figure 1 graphically represents the proposed model.

To examine the GPG in each compensation component, a panel data model with a censored dependent variable of 0 for the lower limit is proposed (Kulich et al., 2011) because the directors’ compensation has a lower limit of 0 when compensation is not received. The panel data methodology was used to avoid biased estimates, due to the issues of unobservable heterogeneity and potential endogeneity of the regressors. These models with censored dependent variables were estimated using random effects.

Finally, to ensure the robustness of the results previously obtained in the regression models, an alternative approach to the propensity score matching was used to pair female directors in the sample with their most similar male counterparts based on firm-level and director-level characteristics. This procedure mitigates selection problems by matching treated and untreated observations based on a set of observable characteristics.

Following Rosenbaum and Rubin (1983), a single propensity score is calculated to reduce the number of dimensions to one. The propensity score method uses first logit models to estimate the probability of a board having a female director given some observable covariates. Thus, the propensity score indicates how closely a female director (treated group) can be matched to a male director (control group) given the set of observed characteristics. To reduce selection bias estimating the treatment effect, a whole vector of a director’s characteristics is considered, including the factor time, tenure, committee membership, CEO and chairperson positions, educational qualifications and relationships, in addition to firm size and time variable. Then each female-year observation is matched with the male-year observation with the closest score. This allows not only to analyse whether there are differences in compensation between women and men with similar observed characteristics, except gender, but also to quantify the average difference in these compensation variables.

We assume that all variables that affect both the treatment and the outcome (unconfoundedness assumption) are observed, and equally that both treated and controls with similar values concerning the observed characteristics (overlap assumption) are observed.

Nearest-neighbour matching was chosen as the applicable matching method, which is the standard procedure to minimize bias in the estimations. Both matching without replacing and with replacement were used to provide additional robustness. Finally, only a single control (male director) is matched with, which ensures the smallest distance in propensity scores between treated and control and consequently yields the lowest bias.

Table 3 presents the descriptive statistics for independent and control variables. Women’s representation is higher among independent directors (26.45%) than among executive (4.06%) and proprietary directors (15.44%). This distribution aligns with existing literature, which indicates that women tend to be more frequently represented in non-executive roles, such as independent directors, compared to executive roles (Adams and Ferreira, 2009; García Martín and Herrero, 2019). However, in global terms, these figures are far from meeting the 40% target for women representation on boards by 2027 (capital companies act), as only 17.93% of the directors in our sample are women. The lower percentage of women in executive positions reflects the well-documented gender disparity in leadership roles within corporate boards, which has been attributed to factors such as gender stereotypes and lack of access to leadership opportunities (Goh and Gupta, 2016; Kulich et al., 2011).

In terms of the distribution of directors in our sample, the data shows the following proportions: executive directors make up 16.1%, proprietary directors account for 40.1% and independent directors represent 43.8% of the total board members. These results show that independent directors are close to the 50% recommended for high-capitalization companies. While the proportion of independent directors does not fully meet the 50% threshold in our sample, it is relatively high and suggests alignment with the spirit of the recommendations for ensuring external oversight on boards. Moreover, for companies that are not of high capitalization or those with significant shareholders, the requirement is generally set at one-third of the board. Therefore, the 43.8% of independent directors in our sample indicates that these companies are likely meeting or exceeding this recommendation, further reinforcing the idea that a strong external oversight is in place. In addition, the proportion of external directors (comprising both independent and proprietary directors) makes up 83.9% of the total board, thus meeting the recommendation that the majority of directors should be external.

With respect to the rest of individual characteristics of directors, it should be noted that executive directors are the ones who have been on the board the longest and chair it most frequently, as only executive directors can hold a CEO position. On the contrary, independent directors participate in a higher number of committees, possess higher qualifications and belong to a higher number of boards.

In relation to the board and firm characteristics, it should be highlighted that the executive directors belong to boards that are a bit smaller with a higher incidence of CEO and chairperson duality. In addition, the companies associated with these directors are larger and more indebted. On the contrary, independent directors belong to boards with a higher proportion of independent directors and a higher presence of women on the nomination and compensation committees.

Table 4 presents the GPG for each type of compensation and director. The data reveal that this gap is mainly present among executive directors for all compensation variables (i.e. total, fixed and variable compensation). Among proprietary directors, there is also a negative gap in fixed compensation, which, in turn, affects total compensation. However, in case of independent directors, the data indicate that female directors receive a higher amount of fixed compensation, whereas male directors earn more in variable compensation. This may be influenced, as explained earlier, by women’s greater risk aversion. It is important to note that the sample sizes for men and women are some unbalanced across all director categories, with a lower representation of women. This imbalance may have an impact on the statistical analysis and should be considered when interpreting the results. These mean difference tests between women’s and men’s compensation for each director category represent a first approximation to studying the GPG, which must be further analysed by controlling for several variables, as previously indicated

Figure 2 provides an initial overview of the annual evolution of the GPG across all directors, aggregating all types of compensation, between 2013 and 2021. This preliminary analysis reveals that while the GPG persists throughout this period, there is no clear upward or downward trend. These results serve as a starting point for the more detailed analysis of compensation components and director roles presented in the subsequent sections.

Finally, Table 5 provides the correlation matrix and the variance inflation factors (VIFs) for the variables used to explain the directors’ compensation. This allows to address potential multicollinearity problems among these explanatory variables. Considering that significant correlations are all well below 0.7 (Tabachnick and Fidell, 1996) and the VIFs are close to 1 (Besley et al., 2013; Kutner et al., 2005), it can be said that there are no collinearity issues among the explanatory variables. Specifically, correlation coefficients range from 0.001 to 0.551 in absolute terms, and the highest VIF is 1.62.

Table 6 presents the regression results for director compensation among executive directors, as well as the GPG. The results of these regressions indicate that gender is negatively associated with compensation, meaning that female directors receive lower compensation than their male counterparts. This suggests that gender may play a role in shaping compensation disparities among executive directors. In particular, women earn less in both fixed and variable pay, and consequently, receive lower total compensation. Therefore, H1a and H1b are supported.

Most control variables are in line with the results found in previous literature. For instance, holding CEO and chair positions positively influenced compensation as individual characteristics. Larger boards (Geiler and Renneboog, 2015) and less controlled boards (Arrondo et al., 2008) provide higher remuneration to directors. Similarly, larger and more profitable firms provide higher remuneration (Amin et al., 2014; Andreas et al., 2012; Arrondo et al., 2008; Fernández Méndez et al., 2012; García-Meca, 2016; Goh and Gupta, 2016). However, there is an exception in the tenure variable (for fixed and variable compensation) and committees (for fixed and total compensation), where the sign is negative, contrary to expectations based on existing literature. In addition, the independence and the presence of women on the nomination and compensation committee indicate a positive relationship with compensation.

Table 7 presents the regression results for proprietary directors. The findings indicate no evidence of GPG for this group, as the gender variable is not statistically significant for any of the remuneration components. Consequently, H2a and H2b are supported.

This result can be understood in light of the unique characteristics of proprietary directors. These directors typically sit on the board as representatives of substantial ownership stakes, often on behalf of institutional investors, family businesses or controlling shareholders. Their selection and presence on the board are primarily determined by their role as ownership representatives rather than by individual professional or executive qualifications. Unlike executive directors, whose remuneration may reflect negotiations, market benchmarks or performance evaluations, proprietary directors’ compensation is more directly tied to their shareholder representation. As a result, their remuneration structures tend to be less influenced by individual attributes, such as gender, and more standardized within this category. Furthermore, proprietary directors often serve as intermediaries for broader shareholder interests, which can dilute the potential for individualized disparities, including gender-based differences.

The last group of directors to analyse is that of independent directors. Table 8 presents the regression results for this group. Similar to proprietary directors, there is no evidence of a GPG for independent directors, as the gender variable is not statistically significant for any of the remuneration components. Consequently, H3a and H3b are also supported. This outcome can be explained by the unique characteristics of independent directors and the nature of their remuneration. Independent directors are typically appointed for their expertise, experience, skills and professional networks. They are expected to bring valuable insights, strategic guidance and independent judgement to the board, and their compensation is primarily designed to reflect these attributes rather than personal or gender-based characteristics. Their remuneration is usually determined by the value they bring to the company through their professional background, experience and industry contacts, rather than by factors such as gender. Moreover, independent directors are often subject to rigorous corporate governance standards that ensure their remuneration is consistent, fair and aligned with market norms. Therefore, the lack of a significant gender effect in this group suggests that the remuneration structures in place for independent directors are driven by the value they add to the company, effectively reducing the potential for gender-based pay disparities.

A potential concern when analysing directors’ characteristics and compensation is a possible selection bias in hiring decisions within the board of directors because companies hire directors with certain characteristics according to their needs. According to Adams et al. (2010), the composition of the board of directors and their actions are jointly endogenous. Thus, female board representation and remuneration policies may also be endogenous.

To ensure the robustness of the results previously obtained in the regression models, we use propensity score matching as an alternative approach. This method captures the average treatment effects of individual director characteristics (a treatment) on their remuneration (a treatment effect), compared to a sample of nontreatment directors in the same or a similar firm.

Table 9 presents the results of the propensity score matching method. Regarding executive directors, women receive significantly lower compensation than men with the same tenure, responsibilities, position, qualifications and relationships, within similar firms and during the same year. Specifically, female executive directors receive between 43.50% and 55.35% less than their male counterparts, depending on the estimator used.

When analysing different remuneration components, the most pronounced GPG is observed in variable compensation, where female executive directors earn between 76.68% and 81.90% less than men. As indicated above, variable compensation represents a very large portion of the total compensation of executive directors, especially in male executive directors.

However, the results indicate that there is no GPG in proprietary and independent directors with similar profiles. These results are in line with those previously obtained in the regression analysis and reflect that female executive directors receive less compensation than their male counterparts. These disparities can be attributed to individual characteristics such as experience, education and responsibilities in the firm. However, they also appear to be influenced by gender per se, suggesting that beyond the legitimate factors of differentiation, gender biases may play a role in the compensation disparities observed. This points to the potential presence of systemic gender-based inequalities in the compensation of executive directors.

The studies that have examined this topic obtained mixed results due to a great heterogeneity in three areas: individuals of the sample, remuneration concepts and requirement or not of gender diversity. First, some research has tried to study whether there is a GPG among top managers of companies (Walton and Tribbitt, 2023); while other studies have focused their efforts on studying a smaller group of these managers, such as CEOs, or have focused on one specific industry (Chen et al., 2022; Grund, 2015; Jin, et al., 2023; Mohan, 2014). Another group of studies have investigated the GPG in boards of directors, analysing executive directors or external directors (Carter et al., 2017; Goh and Gupta, 2016; Maoret et al., 2023). Second, most research has studied GPG in terms of total compensation (García Martín and Herrero, 2019; Pucheta-Martínez and Bel-Oms, 2015; Shin, 2011), even though the remuneration policy is different for each of the remuneration components, which consequently follow different behaviours. Finally, most of the related studies have not considered the existing gender diversity in their sample groups, instead studying the gender gap both in companies with gender diversity and in companies employing only men or only women (Grund, 2015).

Academically, this research incorporates relevant factors into a more compact investigation, so that the obtained results shed light on the existing literature. Thus, it serves as a turning point addressing, on the one hand, the need to study GPG by remuneration components, and on the other hand, updating the data in the Spanish context (García Martín and Herrero, 2019; Pucheta-Martínez and Bel-Oms, 2015). First, following previous research (García Martín and Herrero, 2019; Geiler and Renneboog, 2015; Kulich et al., 2011), this study indicates that GPG would only exist among executive directors. By disaggregating salary types, this study finds that variable components are present where the GPG is most pronounced. These results are in line with Muñoz-Bullón (2010) and Geiler and Renneboog (2015), who noted that much of the GPG among top executives is due to differences in variable compensation. According to previous research, women are offered contracts that are less performance-sensitive and more suited for risk-averse managers (Bertrand, 2011; Brenner, 2015; Carter et al., 2017; Graham et al., 2013; Jin et al, 2023b).

Nonetheless, a substantial amount of literature highlights the role of women and diversity on the board to enhance company results and performance, and even their reputation (Hogan and Huerta, 2018; Terjesen et al., 2016; Martínez et al., 2022; Walton and Tribbitt, 2023). Amore and Garofalo (2021) observed that there is a more pronounced GPG when there is a higher turnover rate among women, reflecting the lack of salary–performance relationship and a problem to retain women on the board.

Even so, this does not explain the GPG observed in fixed compensation for executive directors. This research takes a step further and makes a critical contribution by exposing the unjustified GPG in fixed remuneration, an area that should be entirely free of gender-based disparities. Unlike other compensation components, fixed salaries should not be influenced by subjective factors such as experience, qualifications or individual performance, and yet, this study reveals that a GPG in fixed remuneration persists. This finding underscores that these pay differences are not the result of legitimate differences in skills, experience or education, but rather that they are rooted in gender discrimination. Therefore, we encourage future research to rigorously investigate the causes and determinants of these inequities.

Second, for proprietary and independent directors, there are no gender-motivated differences in the likelihood of receiving these types of remuneration. These findings align with previous studies that emphasize the structural determinants of proprietary directors’ roles on boards and suggest that ownership representation is a key driver of their remuneration. Therefore, the absence of a significant gender effect in this group is consistent with their role as shareholder representatives, where gender is less likely to be a determining factor in their selection, participation or compensation. Conversely, this contradicts previous studies (Goh and Gupta, 2016). In the context of UK-listed firms, the later study found that non-executive female directors are paid less than their male counterparts. The authors claimed that the board does not see the value that a female can offer to the decision-making process, and that they only respond to an external pressure for diversity. Companies should hire independent directors based on their expertise, skills and relationships. So, it seems clear that their remuneration should not be conditioned by gender, but rather by these individual characteristics that contribute a greater value to their work.

Finally, this research improves the promotion of gender equality on boards of directors and female leadership, as developed in Sustainable Development Goal (SDG) 5. The continuous implementation of the gender perspective in corporate governance is enhanced by the discussion of this type of studies, providing an important support for governors and organizations (Eveline and Todd, 2009). Therefore, we advocate for the inclusion of women on boards, accepting and expanding the same discourse that previous literature already defended (Adams and Ferreira, 2009; Bravo and Alcaide-Ruiz, 2019; Kulich et al., 2011; Yanadori et al., 2016).

The GPG remains a persistent issue globally, even though gender equality is a priority for the Organisation for Economic Co-operation and Development (OECD, 2019). The GPG extends even to highly qualified and well-paid positions, as is the case of the board of directors in listed companies. Furthermore, women not only earn less than their male counterparts, but they also attain these positions much less frequently (Mohan, 2014).

This research has some theoretical implications derived from the analysis of homogeneous groups within the board of directors. This is the first study to analyse the GPG within homogeneous groups of people and remuneration jointly. Each type of director is studied separately and, in addition, their total compensation is broken into different remuneration components. Therefore, an important academic implication is that boards of directors cannot be considered as a homogeneous group, just like total compensation cannot be considered as a whole.

Our findings corroborate different behaviour both between individuals (i.e. between types of directors) and between compensations (i.e. between remuneration components). First, executive directors receive significantly higher compensation than the rest of directors, while independent directors are remunerated higher than proprietary directors. Second, regarding variable compensation, which is predominantly discretionary, is mainly present among executive directors, according to the recommendations of the Good Governance Code (CNMV, 2020). This type of remuneration represents, on average, more than half of the total compensation for male executive directors and about 35% for their female counterparts. Third, when analysing the GPG through regression models, controlling by director, board and firm characteristics, the results suggest that a GPG exists for executive directors in fixed and variable compensation. The finding of a significant difference in the fixed salary component suggests the presence of an inequity component that could be related to discriminatory practices in salary decisions. Despite the fact of both men and women potentially possessing comparable skills, training and experience, the differences in fixed salary suggest the persistence of cultural norms and implicit biases that undervalue women in leadership roles. This underscores the urgent need to promote salary transparency policies and diversity in boardrooms to mitigate gender inequalities and ensure fair compensation.

From a theoretical standpoint, the results regarding variable compensation are consistent with resource dependence theory. Executive directors, as key individuals within the company, are highly valued for their ability to manage critical resources and leverage strategic networks. This explains their higher levels of remuneration and the presence of a GPG, as access to and control over these resources tend to favour men due to systemic inequalities in network inclusion and resource allocation. Variable compensation, being tied to perceived contributions and discretionary assessments, may further exacerbate these disparities, reflecting underlying gender biases.

In contrast, no GPG was found for proprietary or independent directors. Proprietary directors hold a place on the board by owning a part of the company, entitling them to participate in its decisions. Therefore, all of them are uniformly remunerated in each board and, consequently, there should be no gender gap between male and female directors. This ensures consistency in the compensation structure for proprietary directors and reduces the influence of internal company dynamics or gender-related biases, which helps explain the lack of GPG in this group. This consistency aligns with the resource dependence theory, as their value to the company stems from their equity stakes rather than their access to external resources. Independent directors are external directors of the company hired for their skills, expertise and qualification, among others. Accordingly, their remuneration should be based on these individual characteristics that add value to the company, and not on gender. Furthermore, their independence from the company’s internal dynamics and resource control mechanisms diminishes the relevance of gender as a determinant of pay disparities in this group.

This research also has different managerial and practical implications for addressing the GPG. Firstly, it should be considered by policymakers and regulators to promote laws or regulations that lead to effective participation and equal pay between men and women on the boards of directors. Equality between male and female directors has still not been achieved despite the objective of reaching 40% women directors by 2015, proposed by Organic Law 3/2007 for the effective equality of women and men, which was later reduced to 30% women by 2020 through the Unified Code of Good Governance (CNMV, 2020). Secondly, these results can be useful for companies when designing remuneration policies for board members to avoid these significant wage gaps. Thirdly, female directors should be aware of the GPG to defend their rights and to negotiate their remunerations when they break the glass ceiling. Fourthly, these results may also be of interest to investors concerned about good corporate governance practices of listed companies. Finally, although the study is based on the Spanish context, the insights and implications are applicable to other international settings, where similar governance structures and gender challenges persist. Moreover, the way in which the evidence is presented, analysed and discussed may be of relevance to both local and international companies seeking to structure board compensation policies that are fair, transparent and aligned with gender equality goals.

This research has certain limitations. The results should be interpreted with caution due to the low proportion of female directors in our sample, which reflects the existing gender imbalance in corporate boards. In addition, the study was conducted in Spain, where a one-tier board system is used (i.e. all directors, both executive and non-executive, are part of the board) and which lacks strict gender quota regulations. Future analysis should consider the inclusion of variables for other individual director characteristics, such as expertise. This should help to shed light whether the GPG arises from of gender per se or specific characteristics. Further research should be conducted to compare the GPG in different contexts (i.e. two-tier board system, countries with a gender quota in the board, common-law countries). Moreover, research on the GPG in homogeneous groups should take a further step analysing the determinants that may influence GPG, considering moderation analysis for context, firm characteristics, board characteristics (Alcaide-Ruiz and Bravo-Urquiza, 2023; Lassoued and Khanchel, 2023) and individual characteristics (e.g. tenure) because it can deepen understanding and present more conclusive evidence. Based on this, studying the compliance with remuneration standards (soft law) in the generation or mitigation of GPGs could be particularly relevant, as we have observed that the GRP index variable significantly influences directors’ compensation. In addition, a promising path for future research would be the use of qualitative methods, such as interviews with market participants, remuneration consultants or board members, to gain further insights into the underlying drivers of the GPG. This approach could help identify non-observable or perception-based barriers and provide a richer understanding of the mechanisms behind gendered pay disparities at the board level. Finally, an important avenue for future research would be to reassess this study in 2026, when Spanish legislation mandates that at least 40% of board members be women. This regulatory change should lead to a more balanced gender representation, allowing for a more robust and equitable analysis of the GPG.