Executive remuneration, particularly the variable component tied to performance (IREM), remains a central pillar in corporate governance discourse globally. Theoretically, incentive-based remuneration is premised on agency theory, which posits that aligning executive remuneration with firm performance mitigates agency conflicts and promotes shareholder value maximisation (Jensen & Meckling 1976; Murphy 1999). However, for this alignment to be credible and legitimate, transparent disclosure is essential. Comprehensive remuneration disclosure enables stakeholders to assess whether remuneration outcomes are justifiable, performance linked and ethically grounded (Bebchuk & Fried 2004; Conyon & He 2011).

In South Africa, executive remuneration disclosure has faced heightened scrutiny because of concerns over excessive pay, opacity and weak alignment with firm performance (Mans-Kemp & Viviers 2018; Ntim, Lindop & Thomas 2013; Siwendu et al. 2024; Viviers 2015). King IV attempts to address these concerns by mandating not just disclosure but also justification and stakeholder-aligned rationale for remuneration outcomes (Delport, Oberholzer & Van Rooyen 2024; IoDSA 2016). However, the implementation of these principles has been inconsistent, with significant variation observed across firms and sectors (Scholtz et al. 2022).

Importantly, political cost theory offers insight into firms’ disclosure of executive remuneration, suggesting that those firms with high profits, high IREM or public visibility may face greater pressure and respond with increased transparency to manage legitimacy or impressions (Haniffa & Cooke 2002; Watts & Zimmerman 1978). In unequal and discontented contexts like South Africa, firms may enhance disclose to mitigate reputational and regulatory risks, particularly when remuneration packages are perceived disproportionate or unjustified.

Empirical studies offer mixed evidence on the remuneration levels–disclosure relationship. For instance, Barkemeyer et al. (2014) found that firms with complex and generous remuneration structures were more likely to disclose detailed narratives to justify such arrangements. Similarly, Harvey et al. (2020) observed that high IREM triggered stronger disclosure responses among firms seeking legitimacy and stakeholder approval. On the other hand, Bebchuk and Fried (2004) argue that firms may also engage in strategic obfuscation, where disclosure is deliberately vague or minimal, to deflect criticism or avoid regulatory backlash.

South African studies have echoed this tension. Scholtz et al. (2022) found that some JSE-listed firms responded to King IV by enhancing narrative disclosure, while others continued to rely on boilerplate statements, particularly in contentious sectors. This heterogeneity implies that the relationship between IREM and disclosure is not merely linear but mediated by governance norms, stakeholder pressure and perceived political costs. We therefore hypothesise that:

Board structure, encompassing size and gender diversity, plays a pivotal role in remuneration governance practices. From an agency theory perspective, the board functions as a central monitoring mechanism, tasked with protecting shareholder interests and overseeing executive remuneration (Fama & Jensen 1983). Larger and more diverse boards are generally assumed to strengthen oversight, mitigate agency conflicts and encourage transparency. However, the influence of board structure on RGD is complex and context dependent, warranting multidimensional theoretical and empirical consideration.

Board size is often associated with increased capacity for oversight and specialisation. Larger boards may draw from a broader pool of skills and perspectives, potentially leading to more robust deliberation and more detailed disclosures (Boateng et al. 2022). Resource dependency theory supports this view by suggesting that larger boards provide access to critical resources, including legitimacy, information and stakeholder linkages (Pfeffer & Salancik 1978). These resources are particularly relevant in environments like South Africa, where firms must navigate complex regulatory expectations and stakeholder demands. Empirical studies have supported this assertion, finding that board size is positively related with governance quality and disclosure, particularly in emerging markets (Ntim & Soobaroyen 2013).

However, a competing strand of the literature warns that excessively large boards may suffer from coordination inefficiencies, diluted responsibility and reduced effectiveness in monitoring executives (Jensen 1993). Beiner et al. (2004) and Jenter, Schmid and Urban (2023) found that beyond an optimal size, the marginal benefit of adding board members diminishes, sometimes resulting in weaker control over contentious issues such as executive remuneration. In the South African context, board size must also be understood in light of transformational imperatives, compliance with King IV and the sectoral complexity of JSE-listed firms, all of which influence how effectively board size translates into transparent disclosure.

Board gender diversity, meanwhile, has emerged as both a normative and instrumental aspect of good governance. Proponents of gender-diverse boards argue that women bring distinct ethical orientations, stakeholder sensitivity and risk aversion that enhance deliberation and improve transparency (Alkhawaja et al. 2023; Gul, Srinidhi & Ng 2011). Stakeholder theory further supports the idea that diverse boards are better positioned to consider broader societal expectations, particularly in relation to fairness and equity in remuneration practices. In line with this, King IV explicitly advocates for diversity, equity and inclusion as key governance principles (IoDSA 2016).

Empirical evidence from emerging markets supports the positive impact of gender diversity on disclosure. Studies by Alkhawaja et al. (2023) and Yilmaz et al. (2023) found that female board representation was associated with enhanced environmental, social and governance (ESG) reporting, including remuneration practices. In South Africa, Toerien, Breedt and De Jager (2023) documented a positive association between board gender diversity and ESG transparency, although the authors caution that the effect is stronger when women hold influential roles or serve on key committees such as remuneration and audit.

Nonetheless, some studies highlight limitations. Muhammad, Migliori and Mohsni (2023) and Terjesen, Sealy and Singh (2009) argue that without a critical mass or meaningful influence, female representation may be symbolic or tokenistic, limiting its impact on governance outcomes. This concern is particularly salient in South Africa, where transformation policies have led to increases in female board representation but not necessarily in decision-making authority (Abdelkader, Gao & Elamer 2024).

Taken together, board size and gender diversity represent structural dimensions of board power that can shape RGD through multiple pathways, including enhanced oversight, stakeholder legitimacy and resource mobilisation. Their effect, however, is likely contingent on broader institutional factors such as regulatory expectations, market pressures and internal governance culture. We therefore hypothesise that:

Ownership structure, particularly the presence of managerial (insider) and institutional (outsider) shareholders, has a significant bearing on corporate governance dynamics and transparency outcomes. Both ownership types exert influence over strategic decision-making, including the disclosure of executive remuneration. Their effects, however, are theoretically and empirically complex, reflecting both monitoring and entrenchment incentives.

From an agency theory perspective, managerial ownership (also referred to as director or insider shareholding) is traditionally viewed as a mechanism for aligning the interests of executives with those of shareholders. When executives hold a meaningful equity stake in the firm, they are more likely to act in certain ways that enhance firm value, potentially including better governance and disclosure practices (Fama & Jensen 1983; Morck, Shleifer & Vishny 1988). Moderate levels of managerial ownership can therefore mitigate agency costs by increasing commitment to accountability. This alignment may lead to enhanced transparency, including the justification of executive remuneration.

However, institutional theory and more recent governance literature suggest a non-linear or even inverse relationship at higher levels of insider ownership. Excessive managerial ownership may lead to entrenchment, where executives get sufficient control to resist external scrutiny, reduce board independence and supress disclosure to avoid criticism or regulatory attention (Eng & Mak 2003; Shleifer & Vishny 1989). In the South African context, characterised by concentrated ownership and relatively high levels of managerial control in some sectors, this dynamic may result in reduced transparency and weakened stakeholder engagement (Ntim 2012). Empirical evidence from emerging markets supports these concerns, as firms with high managerial ownership have been observed to under-disclose contentious governance issues, including executive remuneration (Singh et al. 2022).

Conversely, institutional ownership is often seen as a counterbalance to managerial dominance. Institutional investors such as pension funds, asset managers and insurance companies are generally regarded as sophisticated stakeholders who demand accountability, standardised reporting and sound governance practices (Hartzell & Starks 2003). From a monitoring hypothesis perspective, their presence serves as a deterrent to managerial opportunism and can incentivise firms to improve the transparency of remuneration disclosures to maintain investor confidence (Anguilera et al. 2012).

Nonetheless, the influence of institutional investors is not uniformly positive. Under the tunnelling effect, a concept derived from entrenchment and conflict-of-interest theories, dominant or affiliated institutional investors may collude with management to expropriate value from the firm at the expense of minority shareholders, particularly in markets with weak investor protections (Cheung et al. 2006; La Porta et al. 1999). In such cases, institutions may pressure firms to limit disclosure of sensitive governance matters, including remuneration, to obscure preferential arrangements or avoid triggering regulatory scrutiny. This duality highlights the complex role that institutional investors can play, as either transparency-enhancing monitors or opportunistic actors complicit in obfuscation.

South African evidence reflects both sides of this narrative. Naidu (2024) finds that institutional investors have played a critical role in improving RGD by demanding King IV compliance and integrated reporting. However, in domestically concentrated settings or in firms with crossholdings, institutional oversight has been less consistent, raising concerns about passive monitoring or selective disclosure practices.

Taken together, managerial and institutional ownership reflect the ‘people’ dimension of remuneration governance, each exerting distinct pressures on disclosure. Their effects are shaped by ownership concentration, regulatory environments and the balance of power between insiders and external stakeholders. We therefore hypothesise that:

This study employs an integrative theoretical framework combining agency theory, stakeholder theory, institutional theory, resource dependency theory and political cost theory to examine the drivers of RGD. Together, these perspectives illuminate the structural and behavioural forces that shape a firm’s disclosure practices, particularly in the emerging market context of South Africa.

At its core, agency theory provides the foundational rationale for linking executive remuneration to disclosure practices. It posits that separation between ownership and control introduces agency conflicts, which performance-linked remuneration, as a bonding mechanism, seeks to mitigate (Jensen & Meckling 1976). However, such incentive schemes must be accompanied by transparent disclosure, as a monitoring mechanism, to allow shareholders and other stakeholders to assess their fairness and effectiveness (Bebchuk & Fried 2004). Hence, ‘pay’ is not just a compensation mechanism but a behavioural signal that requires disclosure for legitimacy.

Stakeholder theory broadens this perspective by recognising the diverse and sometimes competing interests of multiple parties, including employees, regulators, the media and the general public (Freeman 1984). In this view, remuneration governance and its disclosure are not only about aligning with shareholders but about securing legitimacy across a wider network of stakeholders. Board dynamics, in particular ‘power’ variables such as board size and gender diversity, plays a central role in responding to stakeholder expectations and enhancing transparency in ethically and socially sensitive matters like executive remuneration.

Institutional theory adds a macro-level layer by acknowledging the normative and coercive pressures that firms face within their institutional environments (DiMaggio & Powell 1983). Regulatory frameworks such as the JSE listing requirements, which enforce compliance with best practice requirements such as those enshrined in King IV and other investor-led initiatives, exert pressure on firms to conform the best practice disclosure standards. These institutional forces are especially potent in dual environments like South Africa, where firms are accountable to both local shareholders and to global capital markets. Ownership structures ‘people’, particularly institutional investors, are key conduits for such pressures.

Resource dependency theory further explains how board structures enable firms to manage external pressures and dependencies (Pfeffer & Salancik 1978). Larger and more diverse boards are seen as better equipped to secure critical resources, enhance legitimacy and navigate complex governance challenges. These structural features can also improve board capacity to oversee executive remuneration and ensure transparent, stakeholder-aligned disclosures.

Finally, political cost theory provides a critical lens for understanding disclosure behaviour as a strategic response to reputational risk and public scrutiny (Watts & Zimmerman 1978). It posits that firms with high profitability, prominent market positions or controversial remuneration practices may face greater political and societal pressure, thereby incurring higher ‘political costs’. In such cases, disclosure serves as a legitimising tool used to pre-empt criticism, regulatory intervention or shareholder dissent. This is particularly relevant in South Africa, where executive remuneration is a politically charged issue amid extreme income inequality and socio-economic transformation agendas. As such, firms may voluntarily enhance RGD to manage political costs and maintain their social license to operate.

Taken together, these theories support the view that RGD is not simply a function of regulation or internal policy, but a strategic and multidimensional outcome shaped by behavioural incentives (pay), structural governance mechanisms (power) and ownership pressures (people). This theoretical pluralism provides a fertile ground for interrogating how JSE-listed firms navigate the intersection of internal governance dynamics and stakeholders’ expectations in shaping RGD. Table 1 presents a summary of study hypotheses and respective theoretical justifications.

This study adopts a cross-sectional quantitative design, drawing on firm-level data from JSE-listed firms for the 2023 financial year. A stratified random sampling approach was used to select 100 firms across key sectors to ensure representation. A defining feature of this study’s methodology is the manual hand-collection of critical variables from integrated annual reports. Specifically, data on RGD and IREM were meticulously extracted from the firms’ integrated reports. The hand-collection process enabled an in-depth evaluation of narrative disclosures, policy articulation and remuneration structures, facets often under-represented or inconsistently coded in online commercial databases. This manual approach was essential for capturing the richness and heterogeneity of remuneration governance practices among sample firms. In addition, board characteristics, specifically board size and gender diversity, were collected from both the integrated reports and Bloomberg, while ownership structure, firm size, leverage and performance indicators were sourced from the Bloomberg and Iress databases. To account for observable firm-specific differences, the study employs cross-sectional regression analysis.

The operationalisation of variables in this study follows established corporate governance and accounting literature, ensuring conceptual clarity and empirical validity.

To test the study hypotheses, we specify the following liner model² (Equation 1): RGDi=α+β1 PAYi+∑k=12β2POWERi  +∑k=12β3PEOPLEi+∑k=14δkCONTROLi+εi[Eqn 1] where PAY_i denotes the construct representing average IREM awarded to executive directors (IREM), POWER_i represents board characteristics (BSIZE and BGDIV), PEOPLE_i represents ownership variables (DSHARE and ISHARE), CONTROL_i represents control variables used in the study (SIZE, INDUSTRY, LEV and ROA) and є_i is the error term. All the variables are defined in Table 2.

To ensure that the estimated coefficients from the OLS regression are best linear unbiased estimators (BLUE), we examined whether the Gauss–Markov assumptions hold for our data. To confirm linearity of the data, scatterplots and residual-fitted value plots confirmed that the relationship between the dependent variable (RGD) and the explanatory variables is approximately linear. To confirm that the data were independent, the sample consists of firm-year observations that are independent across firms. We tested for serial correlation using Durbin–Watson test. The test statistics indicate no significant autocorrelation. To ensure that the data did not exhibit multicollinearity, variance inflation factors (VIFs) were computed for all regressions. All VIF values were below 5, indicating no serious multicollinearity problems. Although the Gauss–Markov theorem does not require normality for OLS to be BLUE, we assessed the distribution of residuals using the normal Q–Q plots. Results indicate approximate normality. Based on these diagnostics, the OLS estimators can be considered BLUE.

Table 3 presents the descriptive statistics of all the variables used in the empirical analysis. To reduce skewness in distribution, natural logarithmic transformation was applied to IREM, ROA and SIZE, although the descriptive statistics are reported in their original (pre-transformation) form.

Table 3 displays considerable cross-section variation for all variables, consistent with the study’s sampling strategy. The table reveals that the mean RGD score is significantly lower than the median score, indicating a negatively skewed distribution. This points that the majority of sample firms provide relatively high levels of disclosure. This distribution is further illustrated in Figure 2. In contrast, the mean IREM value, representing the ‘pay’ construct, is significantly higher than the median value, indicating a positively skewed distribution, with firms paying up to a maximum of R282.50m in IREM. The minimum value of R0 for IREM further confirms that some firms do not provide incentive-based remuneration to executives. In terms of the ‘power’ construct variables, BSIZE with both mean and median number of board members being 10, and a relatively small standard deviation, shows relative stability across sample firms. Board gender diversity shows variation across firms with a minimum of 28.12% and a maximum of 72.71%, reflecting growing but uneven representation of women in boardrooms. Both DSHARE and ISHARE, representing the people construct, have distributions that are also positively skewed. This suggests that while most firms have low ownership concentration, a few firms exhibit significantly higher levels of managerial and institutional ownership.

Table 4 displays the Pearson correlation matrix for the key study variables.

Multiple regression models require the absence of multicollinearity between the dependent, independent and control variables. As shown in Table 4, all correlation coefficients were below the 0.8 threshold, indicating no serious multicollinearity concerns, in line with guidelines suggested by Siavoshi (2024). The signs of all coefficients between RDG and independent variables are as expected and hypothesised.

Table 5 presents the results of the multiple regression analysis. The overall model explains 40.7% of the variation in RGD, suggesting a moderate yet meaningful explanatory power for the variables representing the study’s central constructs of pay, power and people.

Consistent with Hypothesis 1 (H₁), the results show that IREM representing the ‘pay’ construct is the most significant and influential predictor of RGD. This finding lends strong empirical support to agency theory, which posits that performance-sensitive remuneration structures require credible disclosure mechanisms to ensure alignment between executives and shareholders. It also aligns with the political cost theory, which suggests that firms offering generous IREM are more likely to enhance disclosure as a strategic tool to mitigate reputational risks and pre-empt regulatory or stakeholder backlash. The magnitude and significance of this relationship affirm that behavioural incentives embedded in executive remuneration structures are powerful drivers of transparency in remuneration governance.

Turning to the ‘power’ construct, BSIZE emerges as a significant determinant of RGD, confirming Hypothesis 2 (H₂). This result supports the dual theoretical logic of agency theory and resource dependency theory. Larger boards may enhance oversight of executive remuneration and bring a broader range of expertise and stakeholder responsiveness, thereby strengthening the board’s capacity to ensure detailed and principled disclosure. The positive effect of BSIZE also reinforces the idea that structural governance attributes play a vital role in shaping how remuneration practices are communicated to stakeholders. Board gender diversity is positively signed, and although not statistically significant, the p-value (0.112) is approaching conventional significant threshold, offering partial support for Hypothesis 3 (H₃). This result may reflect the limitations of symbolic representation in driving substantive governance outcomes. As highlighted in stakeholder and resource dependency theories, the impact of gender diversity on governance effectiveness, and by extension disclosure, may depend not merely on presence but on influence, such as participation in key committees or the possession of decision-making authority. The finding aligns with critiques in the governance literature that without sufficient voice or critical mass, diversity may fail to shift organisational disclosure behaviour meaningfully.

Within the ‘people’ domain, which captures the ownership dynamics influencing governance outcomes, ISHARE is found to have a marginally significant and positive association with RGD, offering tentative support for Hypothesis 5 (H₅). This result resonates with institutional theory, which emphasises the role of sophisticated investors in shaping corporate behaviour through normative and coercive pressures. Institutional investors, particularly those with international mandates or reputational sensitivity, may demand more transparent and structured RGD as part of their board oversight responsibilities. However, the modest strength of this relationship may reflect the heterogeneity in institutional investor activism and monitoring intensity across sample firms. In contrast, DSHARE shows no significant relationship with RGD, providing no empirical support for Hypothesis 4 (H₄). While agency theory posits that moderate managerial ownership may enhance alignment and thereby encourage transparency, the absence of significance here aligns with concerns from institutional and political cost theories that entrenched insider ownership can reduce responsiveness to external disclosure pressures, particularly in contexts where managerial control is concentrated.

Taken together, these findings affirm the relevance and explanatory strength of the ‘pay, power and people’ framework in understanding the dynamics of RGD in an emerging market context. The strong and consistent effect of IREM confirms that firms facing heightened scrutiny because of performance-linked remuneration are more inclined to disclose comprehensively. The significance of BSIZE highlights the structural governance mechanisms through which transparency is operationalised, while the results of ownership structure suggest that stakeholder influence on disclosure is contingent upon the nature and intensity of oversight. Overall, the regression results not only validate several of the study’s core hypotheses but also reinforce the integrative theoretical framing that conceptualises remuneration disclosure as a strategic, structural and stakeholder-responsive outcome rather than a purely regulatory formality.