This study employs the Weighted Average Cost of Capital (WACC) as the primary measure of the cost of capital, consistent with Gupta and Aggarwal (2024) and Gholami (2022). WACC serves as a comprehensive metric reflecting the average return required by both equity and debt holders and is widely recognized as a key tool for pricing risk (Pratt & Grabowski, 2008). Consistent with IRESS (2024), the weighted average cost of capital (WACC) employed in this study was sourced directly from the IRESS database. WACC is computed by IRESS using two fundamental components: the cost of equity and the after-tax cost of debt. The cost of equity (CoE) is estimated using the Capital Asset Pricing Model (CAPM), incorporating the firm’s equity beta and the market risk premium; The cost of debt, denoted as Cost of Debt (CoD), is the after-tax cost of interest-bearing liabilities. In line with the prevailing South African corporate tax framework applied within the database, the effective tax rate declined from 34.5% in an earlier period, remained unchanged for most of the sample period from 2011 to 2022, and was further reduced to 27% in the period from 2023 to 2024 (South African Revenue Service, (2024) (SARS). Thus, the weighted average cost of capital (WACC) is the aggregate cost of a firm’s capital, with each segment given a weighting according to its proportionate share of the firm’s capital structure. While the baseline analysis focuses on WACC, we also decompose it into its underlying components — the cost of equity and the after-tax cost of debt — for robustness purposes, using inputs sourced from IRESS.

To improve the model’s fit and interpretation, the Weighted Average Cost of Capital (WACC) is expressed as its natural logarithm. The main explanatory variable is ESG performance, captured through the overall ESG score as well as its three disaggregated components: environmental (E), social (S), and governance (G) in line with Gupta and Aggarwal (2024). Additional explanatory variables include BPRGrowth (Price-to-Book Ratio), lnMarketCap (log of market capitalisation), lnSize (log of total assets), lnCAPEX (log of capital expenditure) and leverage (debt-to-equity ratio).

All financial and ESG data were sourced from Refinitiv Eikon and IRESS. ESG scores are based on global disclosure standards and range from 0 to 100, indicating a firm’s level of transparency and commitment to ESG practices (LSEG, 2023; WEF, 2019). The study focuses on a panel of 75 firms listed on the Johannesburg Stock Exchange (JSE), covering large-, mid- and small-cap firms over the period 2011 to 2024. The initial dataset comprised 1,062 firm-year observations. However, after addressing missing values, the final balanced panel used for the analysis consisted of 902 observations. This reflects the wider issue of the limited availability of ESG data in developing markets, where ESG disclosure is still in its infancy (Chen et al., 2024). Firms are classified as large, medium or small based on their year-end market capitalisation, in line with the threshold commonly employed in South African studies. Large-cap firms exceed ZAR 10 billion, mid-cap firms range between ZAR 2 billion and ZAR 10 billion, and small-cap firms are below ZAR 2 billion (IRESS, 2024). This segmentation enables us to examine whether the cost of capital is affected differently by ESG performance depending on size and to reflect variations in disclosure quality, market visibility, and access to finance. Clear size thresholds ensure comparability and facilitate the interpretation of heterogeneity in ESG effects.

Table 2 presents the descriptive statistics for the dependent and independent variables covering the period between 2011 and 2024. The average lnWACC is -2.385, suggesting relatively low financing costs across the sample. The mean ESG score is 50.105, with individual pillar means of 46.46 (E), 51.8 (S) and 53.9 (G), indicating generally moderate ESG engagement with slight imbalances across dimensions. The average leverage is 1.156, the average BPR growth is 2.778, the average log-size is 22.754 and the average log-market cap is 20.388. These figures collectively suggest that the sample predominantly consists of large listed firms. The wide range shown by lnCAPEX also reflects variation in investment intensity.

Summary statistics disaggregated by firm size further reveal important insights. Large firms have an average lnWACC of -2.37 with a standard deviation of 0.53, and report a relatively high mean ESG score of 59.86, with E, S, and G pillar scores of 61.38, 64.65 and 54.75 respectively, indicating a strong institutional commitment to sustainability practices. Medium-sized firms show a lower lnWACC mean of -2.43 (SD = 0.50) and an average ESG score of 44.29, with lower pillar values of 36.35 (E), 45.71 (S), and 51.92 (G), suggesting comparatively limited ESG adoption. In contrast, small firms record a mean lnWACC of -2.35 with slightly greater variability (SD = 0.62), and a lower ESG score average of 38.39, with corresponding E, S, and G values of 30.72, 34.01 and 50.70. These figures suggest that smaller firms attach less importance to ESG issues, particularly with regard to the environmental and social dimensions. However, governance efforts appear to be more consistent across different firm sizes.

Table 3 presents the correlation matrix among the key variables. The overall ESG score demonstrates a strong positive correlation with its constituent elements, namely the environment (0.796), society (0.877), and governance (0.579). A notable correlation of 0.71 between environmental and social scores suggests that firms performing well environmentally often also show strong social responsibility. In contrast, governance appears to be more distinct, with weaker correlations with environmental (0.167) and social (0.250) factors, implying that governance practices may vary independently of environmental and social initiatives. It is interesting to note that there is a positive correlation between the ESG score and the cost of capital (LnWACC). This finding is consistent with the conclusions of Nazir et al. (2022), who argue that some investors may view ESG investments as an additional financial burden, which could potentially raise the cost of financing.

Governance (G) shows weak correlations with environmental (E) and social (S) components (0.167 and 0.250), while E and S are strongly correlated with the aggregate ESG score. This reflects G’s distinct role: it captures structural mechanisms, such as the oversight and reporting functions of boards, which facilitate the implementation of ESG initiatives, albeit without direct overlap with environmental or social activities. Conversely, E and S frequently collaborate in the implementation of integrated sustainability programmes.

In accordance with the fundamental aims of this study, the presentation of results and discussion of findings is structured into three main sections. This approach is adopted to facilitate a more detailed understanding of the relationship between ESG and cost of capital. Section 4.4 investigates the overall relationship between ESG performance and the cost of capital for the entire sample of JSE-listed firms, without accounting for differences in firm size or sector. Building on this, Section 4.5 disaggregates firms based on their size (large, medium or small). The aim is to establish whether the relationship between ESG and the cost of capital varies depending on the scale of operation. Section 4.6 further deepens the analysis by evaluating the relationship from a sectoral perspective, recognizing that ESG impacts and financing conditions may differ across industries due to sector-specific risks, regulatory environments and stakeholder expectations.

The empirical results presented in Table 4 show that there is a positive association between the aggregate ESG score and the cost of capital (coefficient = 0.006). However, this relationship is not statistically significant (p = 0.153). From an economic perspective, this suggests that ESG performance does not significantly influence the pricing of capital among JSE-listed firms on average. This finding is consistent with previous studies, including those by Arslan et al. (2025), Nazir et al. (2022) and Wang (2025), which maintain that high ESG engagement, especially when extensive resources are allocated, may be viewed by investors as costly or dilutive of value in the short term. Consequently, rather than reducing financing costs, ESG efforts may be viewed as burdensome or speculative investments with uncertain returns, particularly in emerging market contexts where ESG integration is still in its infancy.

The disaggregated analysis provides additional insights. The environmental pillar (E) has no measurable effect on the cost of capital (coefficient = 0.000), a result that is consistent with that of Ramirez et al. (2022). This suggests that activities such as reducing emissions, adopting renewable energy and improving resource efficiency, are not yet recognised by investors as relevant factors in pricing firm-level risk or financial resilience. The social pillar (S) displays a small, statistically insignificant positive coefficient (0.002). This suggests that socially oriented initiatives, such as labour practices, community engagement and diversity, may be perceived as having little or no impact on value creation. This aligns with the findings of Khanchel and Lassoued (2022), who observed that social strategies tend to have longer gestation periods, rendering them less effective in influencing short-term investor sentiment. In line with legitimacy theory, the social pillar may be perceived by investors as a mechanism for minimising the cost of capital as firms engage more extensively in social initiatives. Similarly, the governance pillar (G) yields an insignificant coefficient (0.001), providing no conclusive evidence that improvements in governance structures, such as greater board independence, increased transparency or stronger shareholder rights, result in lower financing costs, despite these structures being expected to mitigate agency conflicts. From a broader economic perspective, these findings highlight a potential mismatch between ESG performance and investor perception in South Africa. The capital market does not yet fully reward firms for ESG engagement, possibly due to concerns over greenwashing, inconsistent disclosure standards or uncertainty about the materiality of ESG actions. This implies that ESG efforts alone, in the absence of clear and credible links to risk mitigation or profitability, are insufficient to reduce financing costs in this setting.

Regarding the control variables, leverage exhibits a negative and statistically significant relationship with the cost of capital. This suggests that higher debt levels may signal financial discipline, stronger investor monitoring or optimized capital structures, leading to lower perceived risk and thus reduced capital costs. The book-to-price ratio shows a positive and significant effect, indicating that firms with higher BPR values often associated with undervaluation or financial distress tend to face higher financing costs. Meanwhile, variables such as firm size (lnSize) and capital expenditure (lnCAPEX) are statistically insignificant, suggesting that neither firm scale nor investment intensity play a decisive role in determining the cost of capital in this context.

These findings have several important implications. Firstly, the results highlight the need for corporate managers to integrate ESG activities more effectively with financial strategy and investor communication. Raising ESG scores alone may not generate tangible financial benefits unless accompanied by transparency, relevance and credibility. Secondly, policymakers and regulators should improve ESG reporting frameworks and promote clearer standards that link ESG performance to financial outcomes. Thirdly, investors, particularly institutional investors, may need to evaluate ESG metrics in a more nuanced way, taking into account not only disclosure, but also the financial materiality of ESG efforts. In summary, although there is a theoretical basis for the idea that ESG performance can reduce capital costs by improving stakeholder trust and risk management, these benefits remain unrealised or unrecognised in South Africa’s capital markets. Closing this gap will require greater alignment between ESG practices, financial performance and investor understanding.

Table 5 presents empirical estimates that explore whether the nexus between ESG and the cost of capital varies by firm size on the JSE. For large firms, the aggregate ESG score exhibits a positive, albeit statistically insignificant, correlation with the cost of capital (coefficient = 0.005, p = 0.305). This suggests that stronger ESG performance does not result in lower financing costs, possibly because investors view ESG engagement as a regulatory obligation or a cost centre rather than a source of strategic value. These findings are consistent with those of Dwomor et al. (2024), Ernst & Woithe (2024b) and Gigante & Manglaviti (2022), who document the fact that investors tend to respond less to ESG factors when returns are not clearly visible or immediate. The institutional theory suggests that companies adopt ESG practices to satisfy institutional pressure rather than to facilitate market mechanisms that reduce the cost of capital. Among the control variables, leverage has a negative but only marginally significant effect (coefficient = -0.104, p = 0.090), indicating that higher debt levels may serve as a disciplinary mechanism or provide tax shields, thus lowering capital costs. The book-to-price ratio indicates a positive and statistically significant effect (coefficient = 0.001, p = 0.000), implying that firms that are considered to be undervalued or risky tend to incur higher capital costs. However, market capitalisation and capital expenditure do not appear to have a significant influence on financing costs in large firms.

Disaggregating ESG into its pillars provides additional nuance. For large firms, the environmental and governance components are negatively signed but statistically insignificant (E: -0.001, p = 0.720; G: -0.01, p = 0.756), which suggests that investors do not respond much to such initiatives. The social pillar exhibits a positive but insignificant effect (coefficient = 0.003, p = 0.524), suggesting that markets may view social programmes as costly or unnecessary. These results mirror earlier findings by Ramirez et al. (2022) and Priem and Gabellone (2024), reinforcing the idea that ESG pillars, particularly social and governance, are not yet factored into capital costs in South Africa.

For medium-sized firms, the overall ESG score is also positively signed and statistically insignificant (coefficient = 0.009, p = 0.103). This could indicate investor concerns that implementing ESG policies would place a financial burden on smaller firms, echoing arguments by Khanchel & Lassoued (2022). The environmental component likewise shows a small, positive, but insignificant effect (coefficient = 0.005, p = 0.309), consistent with Gjergji et al. (2021b), who note that ESG compliance costs can be disproportionately high for smaller enterprises. Social scores are negative but insignificant (coefficient = -0.005, p = 0.356), which may suggest that investors appreciate social efforts, albeit without much conviction. Similarly, governance scores are weakly positive (coefficient = 0.004, p = 0.341) and show no significant link to capital cost.

Unlike ESG variables, leverage is a more robust determinant in medium firms, with a statistically significant and negative relationship (coefficient = -0.028, p = 0.001). This suggests that moderate levels of debt may signal credibility or efficient capital use, thereby reducing financing costs. The book-to-price ratio maintains a positive association (coefficient = 0.002, p = 0.099), further supporting the idea that valuation gaps or perceived firm risk influence investor pricing decisions. Other controls, including size, market cap, and capital expenditure, remain insignificant.

Overall, the findings suggest a consistent disparity between ESG performance and the cost of capital for large and medium-sized South African firms. Investors appear to prioritise traditional financial indicators, such as leverage and valuation ratios, over ESG performance when determining financing conditions. ESG activities may still be perceived as either reputational or compliance costs, rather than mechanisms for mitigating risk or creating value. From a practical standpoint, large firms may have reached a plateau of ESG effectiveness, beyond which additional disclosures no longer reduce perceived risk. Meanwhile, medium-sized firms may lack the capacity to implement ESG initiatives cost-effectively. These dynamics require firm-specific ESG strategies, greater ESG transparency and more investor education. Ultimately, to improve the financial benefits of ESG, regulators and firms must collaborate to enhance the credibility, comparability and material relevance of ESG information. This will align market perceptions with long-term sustainability outcomes. Furthermore, the resource-based view suggests that ESG activities do not constitute valuable or distinctive resources; rather, they represent additional costs without providing any competitive differentiation.

To explore sectoral heterogeneity in the ESG cost of capital relationship, we disaggregate firms listed on the JSE into four major sectors: basic materials, consumer services, financials and industrials. This approach aligns with prior studies (such as Johnson, 2020; Ka kogiannis et al., 2023) that underscore the importance of industry-specific ESG materiality for benchmarking and comparability. As shown in Table 5, in the basic materials sector, the aggregate ESG score exhibits a negative but statistically insignificant effect on the cost of capital (coefficient = -0.005, p = 0.598), suggesting a weak and inconclusive link between ESG performance and financing costs. Of the control variables, leverage, firm size, capital expenditure and the book-to-price ratio all exhibit negative but insignificant relationships. Market capitalisation, meanwhile, is positive but also insignificant. When the ESG dimensions are disaggregated, it is found that the environmental pillar has a positive and marginally significant effect (coefficient = 0.039, p = 0.068). This implies that environmental investments may be perceived as costly in this sector. In contrast, the social pillar shows a negative and marginally significant relationship (coefficient = -0.035, p = 0.063), suggesting that social engagement could reduce capital costs. This supports Hypothesis and is consistent with stakeholder and legitimacy theory. Governance, however, has no measurable effect.

For the consumer services sector, the aggregate ESG score also shows a negative but statistically insignificant association with the cost of capital (coefficient = -0.005, p = 0.254). Among the control variables, leverage has a statistically significant negative effect (coefficient = -0.028, p = 0.000), suggesting that higher debt levels may be associated with reduced capital costs. Firm size and capital expenditure are negatively signed but not statistically significant. Pillar-wise ESG analysis reveals that environmental, social, and governance scores all have statistically insignificant effects on financing costs, indicating weak investor responsiveness to ESG differentiation in this sector. In the financial sector, the results show a positive and statistically significant association between ESG performance and the cost of capital (coefficient = 0.039, p = 0.003), suggesting that ESG activities may be viewed as cost-inducing rather than risk-lowering. This finding is consistent with the work of Wang et al. (2025), who highlight the possibility of diminishing returns on ESG investments in capital-intensive sectors. Consequently, Hypothesis is rejected in this context. Among the control variables, only market capitalisation is statistically significant and negatively associated with the cost of capital (coefficient = -0.354, p = 0.048). This means that larger financial firms may benefit from economies of scale in accessing capital. From a legitimacy perspective, this suggests that the ESG pillars may be viewed as a means of maintaining legitimacy and a positive public image in highly scrutinised sectors.

The disaggregated ESG analysis revealed that the governance pillar was the only dimension with a statistically significant effect (coefficient = -0.006, p = 0.076). This suggests that improvements in governance, such as increased transparency and accountability, could significantly reduce financing costs. The environmental and social dimensions, however, remain insignificant. Leverage, unusually, shows a positive and statistically significant effect (coefficient = 0.002, p = 0.022), which may reflect heightened risk perceptions in highly leveraged financial firms.

For the industrial sector, the overall ESG score is statistically significant (coefficient = 0.019, p = 0.021). This suggests that, although ESG investments strengthen stakeholder relations, they incur short-term costs that lead investors to demand higher returns, highlighting a trade-off between stakeholder-oriented sustainability and immediate financial efficiency in capital-intensive industries. None of the control variables exhibits statistical significance. Similarly, disaggregated ESG results reveal no significant effects: the environmental and governance pillars are negatively associated with the cost of capital, while the social pillar has a positive association. However, none of these relationships achieves statistical significance. These findings suggest a muted investor response to ESG practices in industrial firms, perhaps due to low ESG visibility or weak sectoral ESG integration. Taken together, the sectoral analysis reveals important nuances. ESG performance appears to reduce capital costs only marginally and selectively, mainly through the social pillar in the basic materials sector and the governance pillar in the financial sector. Nevertheless, in most sectors, ESG engagement is not yet consistently priced into capital market decisions. This reinforces the idea that investor valuation of ESG is sector-specific and contingent on perceived materiality. The results suggest the need for ESG strategies tailored to specific sectors, reflecting their risks, investor expectations and regulatory pressures. Moreover, improvements in ESG standardisation and disclosure could increase the financial visibility and impact of sustainability practices across sectors.

To assess the robustness of our baseline findings based on the weighted average cost of capital (WACC), we re-estimate the models using the cost of Debt (CoD) and cost of equity (CoE) separately. While WACC captures the firm’s overall financing cost, it aggregates equity and debt and may mask heterogeneous effects across capital providers. ESG performance may affect debtholders and shareholders differently, depending on their distinct risk exposures and pricing mechanisms.

In line with the baseline WACC estimates, the disaggregated analyses using the cost of debt and cost of equity show that the impact of ESG is selective rather than universal. At the aggregate level, ESG performance is positively and significantly associated with the cost of equity (coefficient = 0.007, p = 0.021; see Appendix Table A1), indicating higher equity financing costs for firms with stronger ESG performance. Furthermore, the effect is concentrated among medium sized firms where the governance pillar exhibits a positive and statistically significant relationship with the cost of equity (coefficient = 0.004 p = 0.003; Appendix Table A2). Sectoral analysis further indicates that the social pillar reduces the cost of debt in the basic materials sector (coefficient = -0.026, p = 0.049), while the governance slightly increases the cost of debt in the industrial sector (coefficient of 0.008 p = 0.043; Appendix Table A3). Overall, these results suggest that ESG is not fully priced, as only certain dimensions, firm sizes and industries exhibit statistically significant effects. This reinforces the idea that ESG-related risks and opportunities are only partially reflected in firms’ financing costs.

The results imply that ESG disclosure and engagement though crucial for corporate responsibility and stakeholder legitimacy do not yet guarantee reduced financing costs in South Africa. To this end, policymakers, regulators and institutional investors must collaborate to improve ESG reporting standards, enhance the transparency and comparability of ESG metrics, and align them more closely with financially material outcomes. Furthermore, educating investors is crucial for bridging the perception gap between ESG performance and financial risk mitigation.

Although the study uses a robust dynamic panel approach, it is limited by the availability and quality of ESG data in emerging markets and the relatively short time span for some firms. Future research could benefit from longer time series, market-based ESG ratings and an exploration of how ESG interacts with corporate innovation, risk disclosure and investor sentiment to affect firm value.

In conclusion, ESG engagement, which is important for legitimacy and long-term strategy, has not yet become a strong determinant of capital cost among JSE-listed firms. To unlock its economic value, firms must go beyond compliance-driven ESG and adopt sector-specific, performance-linked sustainability strategies and transparent communication aligning investor expectations with ESG outcomes, rather than focusing solely on compliance.