The Impact of Environmental, Social, and Governance Disclosure on the Firm Value of Non-Financial Firms Listed in South Africa

Abstract

This study examines the impact of Environmental, Social, and Governance (ESG) disclosures on the firm valuation of non-financial firms listed in South Africa, using Tobin’s Q as a firm value proxy. Using a panel data approach of 642 firm-year observations from 2017 to 2022, the study applies Fixed Effects, Random Effects, and Generalized Method of Moments (GMM) estimators to address possible endogeneity concerns. The results consistently show that, for the whole sample, ESG disclosures are positively and significantly related to firm value, thus supporting the view that markets reward transparency and sustainability initiatives. Firm size and liquidity also have positive impacts, while financial leverage has an inverse relationship with firm value. Subgroup regression analysis shows significant sectoral differences: ESG disclosure in non-manufacturing companies has a positive and significant relationship with firm value, in line with stakeholder and signaling theories, emphasizing the premium for intangible assets like reputation and trust. However, in manufacturing companies, ESG disclosure is negatively and significantly associated with firm value, implying concerns among investors regarding compliance costs, strategic misalignment, or possible “greenwashing.” The study contributes to the emerging-market literature by (i) introducing a PCA-based ESG index specific to JSE-listed non-financials, (ii) triangulating results across static and dynamic specifications to ensure robustness, and (iii) uncovering sectoral heterogeneity that has been largely overlooked. The research also has practical implications for corporate managers, policymakers, and investors on the alignment of ESG practices to industry attributes for long-term value optimization.

1. Introduction

The relationship between Environmental, Social, and Governance (ESG) disclosure and firm value has gained growing attention in recent years, especially as sustainability and ethical governance have become central themes in global investment and corporate strategy. In South Africa, a country with a well-developed capital market, significant environmental challenges, and a unique socio-economic landscape, ESG disclosure plays an increasingly critical role in shaping investor confidence, corporate accountability, and long-term firm value. South African companies, particularly those listed on the Johannesburg Stock Exchange (JSE), have faced mounting pressure from stakeholders to demonstrate transparency in ESG practices (Marchbank 2016). These pressures arise from both regulatory developments, such as the King IV Report on Corporate Governance, and voluntary sustainability frameworks, including integrated reporting initiatives and the Global Reporting Initiative (GRI). As investors, regulators, and society demand clearer accountability, firms are adopting ESG disclosure not merely as a reputational exercise but as a strategic tool to enhance competitiveness, reduce risk, and attract responsible capital (Agbakwuru et al. 2024).

Globally, more than 80% of leading corporations had integrated ESG strategies by 2018, demonstrating the increasing incorporation of sustainability into corporate strategy (Gopal and Pitts 2025). Empirical findings further suggest that such integration enhances transparency and long-term value creation (Friede et al. 2015; Liang and Renneboog 2017; Agbakwuru et al. 2024). However, ESG integration is far from uniform. According to Calderon (2023), over US$41 trillion in assets under management globally are now linked to ESG strategies, underscoring the growing integration of ESG metrics into mainstream investment decision-making. Its effectiveness depends on the depth, accuracy, and credibility of disclosures across environmental, social, and governance dimensions. ESG disclosure has thus emerged as a proxy for firms’ commitment to long-term value creation, efficient resource allocation, and responsible corporate citizenship. The South African context presents unique challenges and opportunities for ESG integration. The nation grapples with persistent issues such as inequality, unemployment, climate risk, and corporate governance scandals, making ESG transparency even more vital. While some firms lead in ESG adoption, others lag behind, reflecting disparities in institutional readiness and strategic orientation. Regulatory reforms and investor activism have pushed many companies to enhance their ESG disclosures, but empirical evidence linking such disclosures to firm value remains sparse and inconclusive within the South African setting. The theoretical underpinning of this study draws on stakeholder theory and signaling theory. This study draws on stakeholder theory (Freeman 1984) to argue that transparent ESG disclosure aligns corporate actions with the interests of multiple stakeholders, enhancing legitimacy. Furthermore, signaling theory (Spence 1973) suggests that firms use ESG reporting to communicate credibility and superior management quality to investors. Empirical evidence supports these perspectives; for example, Friede et al. (2015) found a predominantly positive link between ESG and financial outcomes, Khan et al. (2016) showed that material sustainability practices improve firm performance, and Liang and Renneboog (2017) emphasized that corporate responsibility fosters stakeholder trust and long-term value creation.

This study makes three distinct contributions to ESG–firm value literature in emerging markets. First, it develops a PCA-based composite ESG disclosure index tailored to JSE-listed non-financial firms (2017–2022), ensuring a statistically grounded measure of disclosure quality. Second, it triangulates across Fixed Effects, Random Effects, and dynamic GMM estimators to account for unobserved heterogeneity and endogeneity, enhancing causal inference. Third, it disaggregates results by sector, revealing a pronounced asymmetry: ESG disclosure negatively impacts firm value in manufacturing but positively in non-manufacturing. Together, these contributions sharpen our understanding of ESG’s heterogeneous valuation effects in the South African capital market.

2. Literature Review

A clear distinction must be made between ESG disclosure and ESG performance, as prior studies often conflate the two. ESG disclosure refers to the extent and quality of information firms provide on their environmental, social, and governance practices, while ESG performance reflects the actual outcomes of those practices. Although disclosure is important for legitimacy, reducing information asymmetry, and enhancing investor confidence (Clarkson et al. 2013; Dhaliwal et al. 2011), it may not always capture the underlying substance of ESG initiatives. In contrast, ESG performance has been linked to operational efficiency, reduced environmental and social risks, and long-term value creation (Friede et al. 2015). High-performing firms may generate positive outcomes even with limited disclosure, while others may disclose extensively without implementing meaningful ESG strategies. This study focuses on ESG disclosure as the primary independent variable, while acknowledging that disclosure may only partially reflect true ESG performance. The South African regulatory environment provides a distinctive and rigorous context for examining ESG disclosure. The King IV Report on Corporate Governance (Institute of Directors in Southern Africa [IoDSA] 2016) mandates integrated reporting, requiring listed companies to provide material ESG-related information that demonstrates long-term value creation and sustainability. Complementing this, the JSE became the first stock exchange globally to mandate integrated reporting in 2010, embedding sustainability into corporate reporting requirements for all listed firms. More recently, the Financial Sector Conduct Authority (FSCA) Stewardship Code (2020) reinforced ESG integration by encouraging institutional investors to consider ESG issues in their investment decisions and to actively monitor investee companies.

Despite worldwide pressure for ESG integration, empirical studies on the effect of ESG disclosure on firm value in the South African context are similarly at the nascent stage. In contrast to a robust literature base in developed economies like the United States of America, the United Kingdom, and Western Europe (see Chua and Byun 2025; Friede et al. 2015; Quang et al. 2025), South African research in this regard is sparse and disjointed. This is particularly concerning within the unique socio-economic environment of South Africa, with its pronounced disparity, eco-susceptibility, and complex governance system. South Africa is a continental pioneer in corporate governance systems, so this is especially unexpected. According to the King IV Report on Corporate Governance, listed companies are required to create integrated reports that include financial and non-financial data, including ESG indicators, using the “apply and explain” method (Khatlisi and Enwereji 2025). Additionally, in 2010, the JSE became one of the first exchanges worldwide to implement a mandatory sustainability reporting requirement based on guidelines from the Global Reporting Initiative (GRI). While mandated by the King IV Report on Corporate Governance, whether disclosures on ESG will have a material effect on valuations based on the market continues to be poorly covered (Maroun 2022; Yeoh 2021). Much of the literature that does exist on South African companies concentrates narrowly on Corporate Social Responsibility (CSR) in isolation, and not overall ESG disclosure, and tends to confuse operating performance with valuation. Many existing local studies have tended to focus narrowly on corporate governance variables or short-term accounting-based performance indicators, such as Return on Assets (ROA) and Return on Equity (ROE), without sufficiently addressing firm valuation from a capital market perspective. For example, de Villiers and Marques (2016) focused on CSR but did not consider a composite ESG framework, while Ntim and Soobaroyen (2013) examined board structures without linking governance disclosures to stock market valuation. These studies fall short of capturing how integrated ESG practices, particularly in the post-King IV era, contribute to firm value in capital markets.

In contrast, recent international studies from 2020 to 2025 have made significant strides in establishing the connection between ESG disclosure and firm value using robust panel data and advanced econometric techniques. For instance, Alsayegh et al. (2023) conducted a comprehensive panel study on Saudi Arabia’s top 100 non-financial firms and found a strong positive association between ESG disclosure and Tobin’s Q, ROA, and ROE. Importantly, they applied a principal component analysis (PCA) to construct a statistically rigorous ESG index and used both fixed effects and Generalized Method of Moments (GMM) estimators to address endogeneity concerns, methodologies that are largely absent in South African ESG research. Their study also differentiated the impact across sectors, revealing that non-manufacturing firms benefited more from ESG disclosure, a critical insight that South African studies have yet to explore. Similarly, Almulla et al. (2025), focusing on firms across the MENA region, highlighted that ESG-firm value relationships are sector-dependent, suggesting that ESG signals are interpreted differently by investors depending on industry characteristics. This nuance is particularly relevant for the South African economy, which comprises diverse sectors including mining, financial services, telecommunications, and agriculture, yet most ESG research continues to treat these industries homogenously. Mbhalati and Masehela (2024) attempted to assess ESG practices among JSE-listed firms but limited their scope to environmental performance and excluded Tobin’s Q and other market-based outcomes. Their approach did not account for the interplay between social and governance dimensions, thus failing to reflect the holistic ESG construct increasingly adopted in practice and theory. This oversight points to a lack of comprehensive ESG measurement and valuation linkage in South African empirical work.

In developed markets, Elgharbawy and Aladwey (2025) showed that ESG performance is positively linked to market value among UK FTSE 350 companies. However, applying these findings to South Africa may be problematic due to differences in investor ESG literacy, regulatory enforcement, and corporate transparency cultures. Emerging market evidence, such as that from Aboud and Diab (2018) in Egypt, underscores this point. Their study found that social and governance dimensions had a stronger effect on firm value than environmental factors, illustrating the importance of contextual ESG weighting, a topic still under-researched in South Africa. Finally, Signori et al. (2021), through their investigation of European firms, demonstrated that ESG’s impact on firm value varies significantly by sector and ESG dimension, affirming the need for industry-level disaggregation in empirical models. South African studies, however, rarely conduct such nuanced analysis, often pooling all sectors together and thereby obscuring industry-specific ESG dynamics. This limited attention has seen the existence of ESG value relevance in the South African capital market broken down into discrete pieces.

Furthermore, methodological weaknesses diminish the strength of available evidence even more. Most of the studies use a cross-sectional approach, simple ESG proxies (e.g., having an ESG report), or simple scoring criteria that fail to consider disclosure completeness and quality. For example, de Villiers and Marques (2016) used GRI-based indices without correcting for endogeneity and dynamic firm effects through panel models. Putri et al. (2023) tested integrated reporting compliance, but not the causal mechanism of ESG disclosure and firm value. These shortcomings restrict making firm conclusions about the economic value of ESG disclosures. Conversely, international research has started using more econometric sophistication and data reduction to prove causality. For instance, Khan et al. (2016) showed ESG factors’ financial materiality in U.S. companies. But institutional, regulatory, and investor conditions in South Africa are quite different, and so it is not appropriate to extrapolate these findings without local evidence (Abor et al. 2022).

Specifically, there is a gap in the South African literature: relatively few large-scale longitudinal panel data analyses that apply sophisticated methodologies like GMM or Two-Stage Least Squares (2SLS-IV) to estimate endogeneity, unobserved heterogeneity, and reverse causality. Also, no published South African study applies PCA to estimate a composite ESG index, reducing the statistical power of available ESG–value relationships. This lack of methodological soundness and disaggregated data reduces our knowledge of whether and how ESG disclosures add value to the firm in a setting as heterogeneous and special as that of South Africa.

Building on the distinction between ESG disclosure and ESG performance, and considering the regulatory context in South Africa, this study develops the following hypotheses. Since ESG disclosure reduces information asymmetry and improves transparency, it is expected to be positively associated with firm value across the overall sample of listed non-financial firms. However, the relationship may vary across sectors due to differences in capital intensity, regulatory exposure, and stakeholder expectations. In manufacturing sectors, which are resource-intensive and subject to higher environmental scrutiny, ESG disclosure may be perceived as a cost burden or a signal of higher risk exposure. By contrast, in non-manufacturing sectors, disclosure may enhance reputation, strengthen investor trust, and improve access to capital, thereby creating stronger positive associations with firm value.

H1. 

ESG disclosure positively relates to firm value.

H2a. 

ESG disclosure positively relates to firm value in non-manufacturing firms.

H2b. 

ESG disclosure negatively relates to firm value in manufacturing firms.

3. Methodology

3.1. Research Design and Sample Size

This study investigates the impact of ESG disclosure on firm value in South Africa using a longitudinal dataset of the top 100 non-financial firms listed on the JSE from 2017 to 2022. The six-year window was chosen to capture the period following the adoption of the King IV Code on Corporate Governance, which institutionalized integrated reporting and heightened corporate responsibility expectations (Corvino et al. 2020; Institute of Directors in Southern Africa [IoDSA] 2016). An unbalanced panel of 642 firm–year observations was constructed, reflecting instances where firms did not report complete ESG or financial data. The sample covers diverse sectors, including mining, manufacturing, telecommunications, and consumer goods, while excluding financial institutions (banks, insurers, and investment firms). These were omitted due to their distinct regulatory regimes (e.g., Basel III, Solvency Assessment and Management), capital structures, and risk profiles, which could otherwise bias the estimation of ESG–value relationships.

The sampling process began with 150 non-financial firms listed on the JSE between 2017 and 2022. Firms were retained if they remained listed for at least three consecutive years during the study period, disclosed ESG-related information in their annual or integrated reports, had sufficient market capitalization and liquidity to support market-based valuation metrics such as Tobin’s Q and had complete financial data available from Bloomberg and Refinitiv. This approach emphasized quality and continuity of data rather than size alone. Firms were excluded if they exhibited substantial missing data, lacked consistent ESG reporting, or underwent delisting/mergers that disrupted continuity. The final purposive sample comprised 100 firms, ensuring representation across sectors and adequate data quality. While this approach may favor larger, more visible firms with stronger reporting practices, potentially biasing results upward, it reflects best practice in ESG research where data consistency is critical.

Data were drawn from multiple reliable and triangulated sources to enhance validity. ESG disclosure scores were obtained from Refinitiv Eikon, Bloomberg ESG, and MSCI, which provide structured environmental, social, and governance ratings widely used in both emerging and developed market studies (Matemane et al. 2024). Financial data including revenues, assets, market values, and ratios were sourced from audited annual reports, integrated reports, and the JSE’s Stock Exchange News Service (SENS). Only companies that consistently disclosed ESG information or were rated by third-party providers were included, ensuring that the dataset captured meaningful engagement with ESG reporting practices under frameworks such as the GRI and the Integrated Reporting Framework (IIRC). The inclusion of diverse industries allowed for sectoral comparisons, particularly between manufacturing and non-manufacturing firms, an underexplored area in South African ESG research despite its international relevance (Alsayegh et al. 2023).

This dataset provides a high-quality panel enabling econometric analysis of ESG disclosure and firm value. It responds to calls for longitudinal, methodologically rigorous ESG research in South Africa (Delport et al. 2024). While ESG ratings have methodological limitations, they capture a company’s sustainability activities, including social responsibility, environmental performance, and governance structures. Given that the influence of ESG pillars may differ, the study analyzes both aggregate ESG scores and individual environmental (ENV), social (SOC), and governance (GOV) sub-scores. Firm value is measured using Tobin’s Q, defined as the ratio of the market value of assets to their replacement cost. This market-based indicator captures investor expectations and long-term value potential more effectively than accounting ratios such as ROA or ROE, aligning with literature that views market metrics as forward-looking measures of sustainability performance (Mysaka and Derun 2021; Friske et al. 2023).

Control variables include financial leverage (risk and capital structure), asset efficiency (asset utilization), growth potential, and firm size (capacity to invest in ESG), which have been shown to affect the ESG–firm value relationship (Buallay 2019; Raimo et al. 2021). By incorporating these controls, the study isolates the effect of ESG disclosure on Tobin’s Q. Refinitiv Eikon provided the ESG disclosure measures, while annual reports and financial databases supplied the financial and market information for 2017–2022. This research design allows examination of whether markets reward greater ESG transparency under South Africa’s King IV governance regime (Institute of Directors in Southern Africa [IoDSA] 2016).

Table 1. Variable Definitions and Descriptions.

Category	Variable Symbol	Variable Name	Definition and Description
Dependent Variables	Q	Tobin’s Q	Tobin’s Q is the ratio of a firm’s market value to the book value of its tangible assets, serving as an indicator of firm valuation and market expectations.
Independent Variable	ESG	ESG Index	The ESG index, derived through principal component analysis (PCA), aggregates environmental, social, and governance indicators to evaluate a firm’s sustainability and governance performance.
Control Variables	SIZE	Firm Size	Calculated as the natural logarithm of total assets, firm size reflects the scale of a company’s operations and resource base during the observed period.
	AGE	Firm Age	Firm age is the number of years since the firm’s establishment, indicating its longevity and accumulated market experience.
	LEV	Financial Leverage	Total Liabilities/Total Assets
	TAN	Asset Tangibility	Net PPE/Total Assets
	LIQ	Liquidity	Liquidity assesses a firm’s ability to meet short-term obligations, generally based on the ratio of current assets to current liabilities.

presents the definitions and measurements of all variables, including Tobin’s Q.

3.2. Variable Definitions and Model Specifications

Table 1 presents the variable definitions and model specifications used to describe the scaling and econometric estimations for testing the study’s hypotheses.

The econometric model is operationalized as follows:

$$Tobin Q=f (ESG disclosures, Xi)$$

(1)

The firm value of the company is determined by one proxy server, which is Tobin Q, and $X_i$ is a vector of the control factors that significantly affect the firm value of the company:

$${Tobin Q}_{it} = {\beta }_0+{\beta }_1{ESG}_{it}+{\delta }_0{X}_{it}+{\mu }_{it}$$

(2)

The dependent variable, ${Tobin Q}_{it}$, represents company performance and is primarily assessed using Tobin’s Q as the performance metric (Butt et al. 2023). The ESG disclosure index, which reflects a firm’s environmental, social, and governance practices over time, varies depending on the chosen measurement methodology. In this study, the ESG index is constructed using PCA to ensure robustness and reduce dimensionality (see, Kurnoga et al. 2022). Additionally, the model incorporates firm-specific control variables that remain consistent regardless of the measurement approach. These include ${SIZE}_{it}$, representing firm size; ${AGE}_{it}$, denoting the number of years since a company’s establishment; and ${LEV}_{it}$, which captures the degree of financial leverage employed by the firm in a given year (Gonçalves et al. 2022). The empirical model can thus be structured as follows:

$${Tobin Q}_{it}={\beta }_0+{\beta }_1{ESG}_{it}+{\beta }_2{SIZE}_{it}+{\beta }_3{AGE}_{it}+{\beta }_4{LEV}_{it}+{\beta }_5{TAN}_{it}+{\beta }_6{LIQ}_{it}+{\mu }_{it}$$

(3)

$${Tobin Q}_{it}={\beta }_0+{\beta }_1{TobinQ}_{it-1}+{\beta }_2{ESG}_{it}+{\beta }_3{SIZE}_{it}+{\beta }_4{AGE}_{it}+{\beta }_5{LEV}_{it}+{\beta }_6{TAN}_{it}+{\beta }_7{LIQ}_{it}+{\mu }_{it}\dots$$

(4)

The control variables include SIZE, AGE, LEV, TAN, and LIQ, which collectively control for structural, financial, and operational differences influencing firm value.

Panel data methods provide advantages over purely cross-sectional or time-series approaches, as they increase degrees of freedom, reduce multicollinearity, and allow control for unobservable firm-specific heterogeneity, a key concern when analyzing diverse firms across industries and governance regimes (Hsiao 2022). To determine the most suitable baseline model, a Hausman test was applied to assess whether fixed or random effects were appropriate. The test checks whether individual effects are correlated with explanatory variables: if correlated, the fixed effects model is consistent; otherwise, the random effects model is efficient (Greene 2024). This procedure also helped identify possible endogeneity and guided model selection.

Given the prominence of endogeneity in ESG–performance studies, this research employed System-GMM estimation. Endogeneity may arise from simultaneity (reverse causality between ESG disclosure and firm value) or omitted variable bias. System-GMM addresses this by using lagged values of endogenous regressors as instruments, which are correlated with the regressors but uncorrelated with the error term (Wooldridge 2023). We implemented the Arellano–Bover/Blundell–Bond estimator with Tobin’s Q (t − 1) included as a regressor. Instruments were collapsed, with lags 2–3 applied, and AR(1) and AR(2) tests confirmed the absence of second-order autocorrelation.

Instrument validity was assessed using the Hansen J-test of overidentifying restrictions. A non-significant p-value (p > 0.05) indicates valid instruments. In this study, the Hansen J-test yielded p = 0.395, confirming the appropriateness of the instruments and correct model specification. The combined use of fixed effects, random effects, and GMM estimators strengthened robustness by accounting for unobserved heterogeneity, endogeneity, and dynamic interactions. This multi-estimator approach aligns with best practice in ESG–finance research (Clément et al. 2023).

To measure ESG disclosure consistently, a composite ESG index was developed using Principal Component Analysis (PCA). ESG indicators were extracted from company sustainability and annual reports (2017–2022) and standardized for comparability. PCA reduced dimensionality, addressed multicollinearity, and retained components explaining the maximum variance. Component retention was guided by variance explained, and Exploratory Factor Analysis (EFA) clarified factor structure. A Shelby/NC weighting scheme assigned importance to components, which were aggregated and normalized into the final ESG index.

Suitability of PCA was confirmed through the Kaiser–Meyer–Olkin test (KMO = 0.72) and Bartlett’s test of sphericity (χ² = 523.41, p < 0.001). The first principal component had an eigenvalue of 2.94, explaining 68% of the total variance, and was retained for index construction. This composite index captured sectoral variation and temporal dynamics in ESG disclosure while providing a standardized, investor-relevant measure of sustainability practices (

Table A2. PCA Loadings and Variance Explained for ESG Index.

Indicator	Factor Loading	Communality
Environmental Score (E)	0.82	0.67
Social Score (S)	0.79	0.64
Governance Score (G)	0.76	0.58
Eigenvalue	2.14	-
Variance Explained (%)	71.2%	-

KMO = 0.72; Bartlett’s Test of Sphericity p < 0.01. Indicators standardized (z-scores). First principal component retained and used as ESG index.

)

All variables were winsorized at the 1st and 99th percentiles to mitigate outlier influence. Year and industry fixed effects were included, and robust standard errors were clustered at the firm level. Sensitivity checks using the Market-to-Book ratio and disaggregated ESG pillars (ENV, SOC, GOV) produced consistent results.

The Hausman test (χ²(5) = 16.87, p = 0.004) indicated preference for fixed effects over random effects. Nevertheless, both static (OLS, FE, RE) and dynamic (System-GMM) models were estimated to strengthen inference. Static models provide baseline associations but may suffer from reverse causality and omitted variable bias. System-GMM, by including lagged dependent variables and valid instruments, corrects for these concerns, allowing for more reliable causal interpretation. Firm value was primarily measured using Tobin’s Q, a forward-looking market-based indicator. Control variables included firm size, leverage, age, liquidity, and tangibility, all of which have been shown to affect firm value. Sectoral analysis further distinguished whether ESG–value relationships differed between manufacturing and non-manufacturing firms (

Table A3. Sectoral Classification.

Sector Category	Description	Examples
Manufacturing	Firms engaged in industrial production, resource processing, or goods output	Mining, Chemicals, Food & Beverage, Industrial Equipment
Non-Manufacturing	Firms primarily in services, trade, finance, utilities, and IT	Insurance, Banking, Retail, Telecom, Energy Distribution

Note: Classification follows JSE sector taxonomy with adaptation for analytical purposes.

). The combined use of multiple estimators, a large sample of 642 firm-year observations, and a rigorously developed PCA-based ESG index enhanced the statistical reliability and contextual relevance of the findings. Together, these methodological strategies ensured a robust evaluation of ESG disclosure’s impact on firm value in South Africa.

4. Results and Discussion

This section presents and discusses the empirical results of the study, grounded in factual data and quantitative evidence. However, descriptive statistics and correlation analysis were employed to examine the value patterns and interrelationships between ESG index dimensions and firm-level accounting indicators. The baseline regression analysis examines the impact of the ESG disclosure index on firm value with a controlling variable for size, leverage, age, and liquidity. Subgroup comparisons are also made between manufacturing and non-manufacturing companies in the study to consider sectoral heterogeneity in the ESG–value relationship. Finally, a Robustness check was carried out to carried out using the GMM estimator to verify consistency. The discussion combines these results with a review of the literature to make insightful comparisons and determine the conformity or deviance from theoretical anticipations and previous empirical observations.

4.1. Descriptive Statistics

This subsection provides an overview of the key variables used in the empirical analysis, summarizing their central tendencies, dispersion, and distributional characteristics.

Table 2. Descriptive Statistics Results.

Variable	Mean	SD	Min	Max	Skewness	Kurtosis	Jarque–Bera p
ESG Disclosure (Index)	32.884	12.738	10.981	76.415	0.524	2.547	0.000
Environmental (E)	28.931	10.736	11.943	70.728	0.432	2.191	0.000
Social (S)	31.674	11.804	9.982	74.892	0.469	2.319	0.000
Governance (G)	31.122	10.042	11.128	73.011	0.427	2.188	0.000
Tobin’s Q	1.833	1.071	0.602	8.915	2.489	11.781	0.000
Firm Size (total assets, R’000)	75,934	240,116	975	2,643,521	6.743	54.882	0.000
Leverage	3.076	3.462	1.027	79.314	6.584	83.916	0.000
Firm Age (years)	2.02	1.80	1.03	46.9	2.014	8.305	0.000
Tangibility	0.527	0.819	0.000	12.107	3.853	44.739	0.000
Liquidity	2.168	2.224	0.108	28.121	4.672	54.189	0.000
Observation	642	642	642	642	642	642	642

Firm size is reported in R’000 (logged in regressions). Firm age is presented in actual years for interpretability. Leverage, tangibility, and liquidity are ratios.

presents the descriptive statistics. The mean ESG disclosure index is 32.88 with a standard deviation of 12.74, indicating a moderate level of transparency and substantial variability across firms. The minimum and maximum scores range from 10.98 to 76.41, reflecting a wide gap between low and highly comprehensive disclosures. The distribution is slightly positively skewed (0.524) and platykurtic (2.547), and the Jarque–Bera test confirms non-normality (p < 0.01). Decomposing the index, the mean scores for the environmental (28.93), social (31.67), and governance (31.12) dimensions are similar, suggesting that firms tend to emphasize all three pillars relatively evenly. Each dimension shows moderate positive skewness and platykurtic distributions, again indicating a few high-performing firms but a general trend of moderate ESG disclosure across the sample.

Table A1. Firm Counts by Year and Sector (2017–2022).

Year	Manufacturing Firms	Non-Manufacturing Firms	Total Firms	Firm-Year Obs.
2017	42	58	100	100
2018	39	55	94	94
2019	41	57	98	98
2020	38	56	94	94
2021	40	54	94	94
2022	40	56	96	96
Total	240	336	576	642

Notes: Panel is unbalanced due to missing ESG/financial disclosures in certain years. “Top 100” firms determined annually (rolling sample) by market capitalization.

in the Appendix A shows the details of the firm counts.

The mean Tobin’s Q of 1.83 implies that firms, on average, are valued above book value, likely due to investor optimism or growth expectations. Its high standard deviation (1.07) and maximum value (8.92) reveal notable heterogeneity, driven by a few high-value firms. The variable is highly right-skewed (2.489) and leptokurtic (11.78), as confirmed by the Jarque–Bera test (p < 0.01). Firm size, measured by total assets, is extremely right-skewed (6.743) and leptokurtic (54.882), consistent with the presence of a few very large firms. Financial leverage averages 3.08, suggesting debt-oriented financing, with pronounced skewness (6.584) and kurtosis (83.916), indicating significant variation across firms.

Firm age averages 0.70 (logarithmic form), representing a mix of younger and older firms with a positively skewed distribution. The mean tangibility ratio of 0.53 implies that more than half of assets are tangible, typical of capital-intensive industries. Liquidity averages 2.17, reflecting strong short-term solvency, though the high standard deviation (2.22) and skewness (4.672) suggest that a subset of firms hold exceptionally high liquid assets. However, all the variables violate the Jarque–Bera normality test (p < 0.01), which suggests that none of them are normally distributed. This should perhaps not be unexpected for firm-level observations due to outliers, sectoral heterogeneity, and firm structural heterogeneity.

4.2. Correlation Analysis

This subsection provides an overview of the correlation patterns among the key variables used in the study.

Table 3. Correlation Matrix Results.

Variables	ENV	SOC	GOV	ESG Index	Tobin’s Q	Firm Age	Firm Size	Leverage	TAN	LIQ
ENV	1
SOC	0.238	1
GOV	−0.139	0.324	1
ESG Index	0.657 *	0.731 *	0.491	1
Tobin’s Q	0.094	0.003	0.125	0.012	1
Firm Age	0.034	−0.078	0.010	0.041	0.103	1
Firm Size	0.009	−0.136	0.082	0.128	0.018	0.177	1
Leverage	−0.021	−0.192	−0.017	−0.092	0.198	−0.027	−0.030	1
TAN	−0.112	0.056	0.018	−0.037	−0.062	0.081	0.166	0.042	1
LIQ	0.074	−0.048	−0.021	0.091	0.083	0.029	−0.058	−0.079	−0.037	1

* p < 0.1 denote levels of statistical significance.

presents the correlation matrix, which highlights notable relationships among ESG dimensions, the ESG Index, firm characteristics, and firm performance. Furthermore, the ESG dimensions moderately correlate with each other, with SOC and ESG Index most strongly correlated (r = 0.731). The inter-correlation between ENV and ESG Index (r = 0.657) and GOV and ESG Index (r = 0.491) also suggests that all three pillars play an important role in overall ESG disclosure, but to different degrees. Surprisingly, the social dimension seems to have the most dominant role in driving the ESG Index, indicating companies report or invest more in social issues like employee well-being, diversity, and community involvement. Nevertheless, the relationship between ESG dimensions and firm value is relatively weak. The strongest positive correlation is between GOV and Tobin’s Q (r = 0.125), followed by Environmental (r = 0.094), with the social component having an almost zero correlation (r = 0.003). The ESG Index also has a very weak positive correlation with Tobin’s Q (r = 0.012), indicating that ESG disclosure in general does not have a strong direct correlation with firm value. These results are consistent with the arguments of Wang (2024) established that ESG investment enhances stakeholder trust but does not necessarily accrue in the short term in the form of market-based firm value increments.

In comparison to more recent literature, Madison and Schiehll (2021) in Canada note that though investors pay more attention to ESG ratings, the financial materiality of ESG disclosures is sector- and region-specific. In an emerging economy such as South Africa, where disclosure standards and enforcing institutions can be relatively weaker, the ESG-firm value relation would be statistically weaker compared to developed markets (Simbi et al. 2023). This can account for the low correlations in the present study. Furthermore, Maama (2021) discovered that governance quality impacts valuation more in African companies than in environmental and social disclosures, which aligns with the comparatively higher correlation in our findings between the GOV dimension and Tobin’s Q. For the firm-level characteristics, the correlation with Tobin’s Q is moderate for firm age (r = 0.103), indicating that firms with more aged firms can maintain more stable business and higher market valuation. Firm size has a near-zero correlation with Tobin’s Q (r = 0.018), which is unexpected because larger firms would have economies of scale and investors’ confidence. This weak relationship may reflect the heterogeneity of firm size across industries in the sample, in line with Ishaq et al. (2021), who found contradictory impacts of firm size on performance in manufacturing sector firms of Pakistan. Financial leverage, however, has a positive association with Tobin’s Q (r = 0.198), suggesting that highly leveraged firms can be seen to be aggressive growth seekers by investors, corroborated by Arhinful and Radmehr (2023) in their capital structure and firm performance in Tokyo research.

TAN and LIQ both demonstrate relatively low correlations with other variables, suggesting that they capture distinct aspects of firm characteristics. Tangibility exhibits a weak negative relationship with Tobin’s Q (r = −0.062), indicating that firms with higher levels of fixed assets relative to total assets may experience slightly lower market valuation, likely because tangible assets are less flexible and may limit growth opportunities. Conversely, liquidity shows a small positive correlation with Tobin’s Q (r = 0.083), implying that firms with stronger short-term solvency positions are more capable of meeting obligations and maintaining investor confidence, which can be beneficial to market value.

4.3. Regression Analysis

This subsection presents the baseline regression findings examining the relationship between ESG disclosure and firm value using Fixed Effects, Random Effects, and System-GMM estimators.

Table 4. Panel Regression Results.

Variable	Fixed Effects Model	Random Effects Model	GMM Model
Dependent Variable (Tobin’s Q)
ESG Index	0.030 ***	0.028 ***	0.032 ***
Firm Size	0.015 ***	0.013 ***	0.013 **
Firm Age	−0.001	−0.001	−0.001
Leverage	−0.044 ***	−0.047 ***	−0.049 ***
Liquidity	0.009 ***	0.008 ***	0.007 **
Tangibility	−0.006 ***	−0.007 ***	−0.005 ***
Constant	0.329 ***	0.348 ***	0.371 ***
Observations	642	642	536
R-squared	0.223	0.212	—
F-statistic	13.01 ***	13.33 ***	—
Number of Instruments			32
Hansen J (p-value)			0.287
AR(1) Test (p-value)			0.015
AR(2) Test (p-value)			0.412
Observations			642
Firms			100

Standard errors in parentheses. *** p < 0.01, ** p < 0.05.

shows the result of panel regression. The Fixed Effects, Random Effects, and GMM regression results show conclusive evidence regarding the determinants of firm value (captured by Tobin’s Q) in relation to ESG disclosure and financial properties at the firm level. In all three estimation models, Tobin’s Q positively and statistically significantly relates to the ESG Index. More particularly, the ESG coefficient varies from 0.028 to 0.032 and is significant at the 1% level in all the models. It is worth mentioning here that, before 2019, firm value was hindered by a variety of factors, and the spread of the COVID-19 pandemic in 2020 had a slight impact on firm performance, as indicated by the negative, yet not statistically significant, coefficient. The empirical evidence suggests that the worst negative impact of the pandemic was experienced in 2021, the year when its effects were most visible. This result indicates that higher ESG disclosure always relates to higher firm value and supports the argument that markets appreciate transparency and sustainability initiatives. These findings support more recent research like that by Quintiliani (2022), which corroborates that investors value ESG performance, especially in emerging economies where there has been a transition towards ESG integration. Indeed, firm size has also shown a positive and significant relationship with Tobin’s Q for all three models. The impact is strongest for the Fixed Effects model (0.015) and persists at the 1% or 5% level in specifications. This result suggests that larger firms are possibly more favorably evaluated by the market due to economies of scale, increased visibility, and possibly improved governance mechanisms. The results confirm Chininga et al.’s (2024) findings that firm size is significant in increasing the value of the firm in the South African setting, possibly by undertaking more diversified activities and capital access.

Firm age does not have a statistically significant impact across any of the models. The uniformly low and negative coefficients (−0.001) indicate that, holding other variables at a constant level, age is not a good predictor of firm value within this sample. This finding could be an effort to refute the case that older firms do not necessarily need to be more innovative or productive, particularly in fast-moving markets. It is well aligned with Farooq et al. (2021), who explain how firm age deteriorates when other structural controls are undertaken, especially for non-manufacturing firms. Financial leverage, however, negatively affects firm value with a significant coefficient across all models at between −0.044 and −0.049. This reverse suggests that greater levels of debt are correlated with smaller Tobin’s Q, arguably reflecting greater financial risk or long-term solvency problems. This result is consistent with pecking order and trade-off theories of capital structure and more recent empirical evidence by Margono and Gantino (2021), where high levels of leverage are found to be detrimental to firm value in Sub-Saharan Africa because of the weight of interest and shortage of credit. Liquidity is positively and strongly correlated with Tobin’s Q in all specifications, but with a lower magnitude (coefficients of 0.007 to 0.009). This evidence would indicate that companies with better liquidity profiles are viewed positively by investors, perhaps owing to the capacity of such companies to meet their short-term debts and stay liquid in a turbulent financial environment. These results are supported by Cayón and Gutierrez (2021), whose results highlighted that liquidity has an important signaling function for firm stability and operating robustness.

The results reveal that the TAN variable exhibits a small, negative, and statistically insignificant coefficient across all models, suggesting that while tangible assets contribute to operational stability, they may not enhance firm value as measured by Tobin’s Q. This finding indicates that a higher proportion of fixed assets can reduce managerial flexibility and the firm’s capacity to adapt quickly to changing market conditions, thereby exerting a marginally adverse influence on valuation. Similar evidence has been reported in prior studies. Margono and Gantino (2021), analyzing firms listed on the Indonesia Stock Exchange, found that tangibility negatively affected firm value, explaining that high asset tangibility limits liquidity and investment flexibility, which in turn restrains market performance. Likewise, Ishaq et al. (2021) observed a weak and statistically insignificant relationship between tangibility and Tobin’s Q in Pakistan’s manufacturing sector, concluding that tangible assets do not significantly drive investor perception of firm growth potential. Consistent with these findings, Mysaka and Derun (2021) reported that firms with higher levels of fixed assets tend to experience reduced market valuation due to inefficiencies in reallocating capital to more productive investments.

Regarding the model fit, the Fixed Effects model has an R-squared of 0.223, which implies that approximately 22.3% variation in firm performance is captured by the model variables. The Random Effects model gives a slightly lesser R-squared (0.212), whereas the GMM model, intended to correct for potential endogeneity, does not give an R-squared but guarantees the robustness and reliability of the findings. The F-statistics for the panel models are strongly significant (p < 0.01), bearing witness to the validity of the entire model. Hausman χ²(5) = 16.87, p = 0.004, confirming FE preference over RE. Despite such strong findings, there are vital limitations to be addressed. First, though the models reveal consistent significance to ESG disclosure, causality might not be assured in the absence of stronger instruments or exogenous shocks. Second, though the sample is complete, it only consists of those firms that possess available ESG information, generating selection bias. Third, omitted variable bias still exists, particularly since there are complicated interactions between governance, regulation, and firm performance in emerging markets.

Differences across models are marginal, with the GMM results confirming the overall robustness of the panel estimates through valid instruments and the absence of serial correlation. Collectively, these results highlight that ESG performance, liquidity, and firm scale drive firm value, while tangibility and leverage exert weak or adverse effects.

Table 5. OLS Regression (Tobin’s Q as Dependent Variable).

Variable	Tobin’s Q	Std. Error
Firm Size	0.026 ***	(0.006)
Firm Age	0.083 ***	(0.028)
Manufacturing Dummy	0.021 *	(0.015)
Leverage	−0.005 *	(0.005)
Liquidity	0.008 **	(0.004)
Tangibility	−0.004	(0.005)
ESG Disclosure Index	0.001 ***	(0.001)
Constant	0.527 ***	(0.073)
Observations	642	642
R-squared	0.298
Hansen J-Test (p)	0.395
Adjusted R2 (within)	0.312
F-statistic (p)	0.000
Observations	642
Firms	100

*** p < 0.01, ** p < 0.05, and * p < 0.1 are the standard errors, which are enclosed in parentheses.

presents the results of OLS. The regression output is insightful regarding the determinants of the value of the firm, as captured by Tobin’s Q, from 642 firm observations. When Tobin’s Q is used as a proxy variable for firm value, the ESG Disclosure Index has a positive and statistically significant coefficient (0.001, p < 0.01), showing that greater levels of ESG disclosure are positively related to greater market valuation. This result is in line with the developing consensus that investors value transparency and ethical behavior. It implies that companies dealing with ESG issues more holistically are apt to have greater investor trust and possibly superior longer-term financial performance. This result agrees with existing literature showing the potential of ESG activity to drive value. For instance, Cayón and Gutierrez (2021) determined that transparency of ESG mitigates information asymmetry and improves stakeholder trust, hence having a positive influence on firm value. Furthermore, a high positive correlation between ESG disclosure and the valuation of firms in African and emerging economy settings. In South Africa, Aydoğmuş et al. (2022) similarly observed that ESG reporting enhances investor image and financial valuation, particularly among listed firms whose businesses are working within controlled industries like finance and mining.

In addition to ESG disclosure, there are also other firm-level variables having a strong association with Tobin’s Q. Firm size is highly and positively associated (0.026, p < 0.01), showing that larger firms, possibly because of their already established market presence, diversified business, and resource access, are more valued. This agrees with Aydoğmuş et al. (2022) research, which demonstrated that firm size is the greatest value driver for Sub-Saharan Africa. Firm age is also positively and significantly correlated with firm value (0.083, p < 0.01), indicating that older firms enjoy gains in reputation, stakeholder relationships, and operational expertise. This is echoed by Farooq et al. (2021), who emphasized that firm maturity plays a key role in enhancing credibility and valuation on capital markets. Notably, the manufacturing dummy variable is also positively associated with firm value (0.021, p < 0.05), suggesting that manufacturing companies may gain more from ESG disclosures compared to other companies. This can be attributed to increased monitoring of the environment and labor practices by producing industries. It concurs with evidence of Chininga et al. (2024), whose observation was that sectoral effects strongly mediate the ESG–firm value link and environmentally concerned sectors gain greater valuation premia from ESG efforts.

In contrast, financial leverage is highly and negatively correlated with firm value (–0.005, p < 0.05), indicating that greater debt reduces firm valuation. This reflects investor concern for financial risk and solvency, particularly where the economic environment is unreliable or unsure. The findings are consistent with Liang and Renneboog (2017) and Bhatia and Tuli (2017), who indicate that high leverage undermines market confidence and limits the capacity of a firm to finance ESG investment in a sustainable way. The LIQ variable shows a positive and significant relationship with Tobin’s Q (β = 0.008, p < 0.05), supporting the view that firms with stronger short-term liquidity positions are perceived as more stable and capable of meeting financial obligations, thus improving investor confidence. Conversely, TAN exhibits a negative but statistically insignificant coefficient (β = −0.004, p > 0.10), suggesting that while tangible assets contribute to operational stability, their fixed nature may limit strategic flexibility, a finding consistent with prior studies such as Margono and Gantino (2021), Ishaq et al. (2021), and Mysaka and Derun (2021).

The R-squared value of 0.298 in the diagnostics model in this case signifies that the model variables account for the difference in Tobin’s Q to the tune of almost 30%, and that is what usually happens with firm-level data. The Hansen J-test (p = 0.395) also reveals that the model instruments are not spurious and that the model is free from over-identifying restrictions, which confirms the GMM estimation.

4.4. Robustness Tests and Subgroup Analysis

This subsection presents the robustness tests in a systematic manner by comparing the subgroup regressions with the baseline FE and GMM results. Rather than repeating patterns already established in Section 4, the discussion here focuses only on the outcomes that deviate meaningfully from the main model. The purpose of this approach is to highlight sector-specific differences that alter the relationship between ESG disclosure and firm value, thus providing a clearer understanding of heterogeneity across industries. Only statistically meaningful divergences are discussed to avoid redundancy.

As shown in

Table 6. Subgroup Regression: Manufacturing vs. Non-Manufacturing Firms.

	Tobin’s Q
Variable	Manufacturing	Non-Manufacturing
Firm Size	0.033 *** (0.008)	0.008 (0.005)
Leverage	0.019 *** (0.002)	0.032 *** (0.004)
Firm Age	0.064 (0.049)	0.085 * (0.043)
Liquidity	0.010 ** (0.004)	0.009 ** (0.004)
Tangibility	−0.008 (0.006)	−0.005 (0.005)
ESG Disclosure	−0.007 *** (0.002)	0.006 *** (0.001)
Intercept	0.961 *** (0.117)	0.303 *** (0.073)
Observations	214	428
R-squared	0.181	0.091

*** p < 0.01, ** p < 0.05, and * p < 0.1 are the standard errors, which are enclosed in parentheses.

, the he subgroup regression between manufacturing and non-manufacturing companies in South Africa reveals dramatic sectoral differences between the effect of firm characteristics and ESG disclosure on firm value, as indicated by Tobin’s Q. Firm size is positively and statistically related to firm value (coefficient = 0.033, p < 0.01) in manufacturing companies, indicating that larger manufacturing companies are more valuable to the market. This can be explained through economies of scale, better-established infrastructure, or improved access to capital. Leverage also positively covaries with firm value (coefficient = 0.019, p < 0.01), whereby manufacturing firms with greater debt are perceived positively, possibly because such firms are likely to use capital investment to expand operations and yield returns. However, firm age, though positively related to firm value (coefficient = 0.064), is not significant, and this shows that only maturity does not necessarily result in better valuation in this sector. Surprisingly, ESG disclosure in manufacturing companies has a statistically significant and negative relationship with firm value (coefficient = −0.007, p < 0.01). This being an unexpected finding, it indicates that increased ESG disclosure is no longer viewed as an asset but as a cost among manufacturing companies. This finding could be indicative of investors’ apprehension with the cost of complying with ESG, especially among capital-intensive industries with slim margins. Or, inversely, it may suggest that ESG practices are not incorporated yet in operational plans or are not yet resulting in tangible performance improvements, such that investors are wondering about their short-term impact. All these findings are consistent with Maso (2024), who stated that in manufacturing-intensive industries, ESG practices can fail if they are viewed as responsive or by-endurance rather than strategic.

In contrast, the outcome for the non-manufacturing companies reveals different dynamics. Firm size statistically insignificantly and negligibly impacts firm value (coefficient = 0.008), revealing that, unlike for manufacturing, size is not a determining factor in the value of knowledge-intensive or service-based firms. Financial leverage is still significantly positive (coefficient = 0.032, p < 0.01), again suggesting that debt financing is positively perceived by investors, perhaps because of effective capital management or future growth opportunities. Firm age is also positively associated with firm value (coefficient = 0.085) and marginally significant (p < 0.10), suggesting that experience, reputation, and credibility in the market are obtained and enjoyed by older firms in the non-manufacturing industry. The most significant difference is in the ESG disclosure role. In non-manufacturing companies, ESG disclosure is significantly and positively correlated with firm value (coefficient = 0.006, p < 0.01) relative to the negative correlation found in manufacturing companies. The finding indicates investors pay higher premiums for transparency in ESG in non-manufacturing industries where intangible assets of reputation, brand, and stakeholder trust are more pivotal in competitive success. This is consistent with Li et al. (2024), who concluded that ESG practices are able to better enhance valuation in industries where stakeholder interaction and ethical behavior are essential to firm success and survival.

The finding that TAN remains negative (β = −0.005, p > 0.10) while LIQ shows a positive and significant relationship (β = 0.009, p < 0.05) is consistent with prior empirical evidence in corporate finance and sustainability literature. Margono and Gantino (2021) found that firms with high asset tangibility often experience lower market valuation because tangible assets limit operational flexibility and reduce responsiveness to market opportunities, thereby dampening Tobin’s Q. In contrast, Ishaq et al. (2021) reported that firms with greater liquidity tend to achieve higher Tobin’s Q values, as strong short-term solvency enhances investor confidence and indicates efficient working capital management. Similarly, Aydoğmuş et al. (2022) demonstrated that liquidity positively influences firm performance, while tangibility exerts a weak or negative effect, as asset-heavy firms face slower capital turnover and higher maintenance costs. From a model performance perspective, for manufacturing firms, R-squared is at 0.181, meaning that the model could explain about 18.1% of firm-value variation. That is very solid for firm-level data. For non-manufacturing firms, the R-squared has been recorded at 0.091, suggesting that there may be other variables not included in the model that are significant in explaining firm value, such as innovation, digital competence, or customer loyalty, to name a few.

This hypothesis was validated, with the analysis uncovering sector-specific differences in the relationship between ESG disclosure and firm value. Our hypothesis posited that ESG disclosure positively influences firm value, proxied by Tobin’s Q, across both manufacturing and non-manufacturing firms. However, the subgroup regression results in Table 6 provide a more nuanced picture. Among non-manufacturing firms, ESG disclosure exhibits a positive and statistically significant effect on firm value (β = 0.006, p < 0.01), consistent with theoretical expectations and prior empirical research. This finding lends strong support to stakeholder theory and signaling theory, suggesting that in sectors where intangible assets such as brand reputation, customer trust, and service quality are paramount, transparent ESG practices are viewed favorably by investors. This is aligned with Khanchel and Lassoued (2022), who found that ESG disclosures improve market valuations, especially in consumer-driven and service-oriented sectors where ethical conduct and corporate responsibility influence investor decisions.

In contrast, the results for manufacturing firms tell a different story. The ESG disclosure coefficient is negative and statistically significant (β = −0.007, p < 0.01), indicating an inverse relationship between ESG disclosure and Tobin’s Q. This suggests that in capital-intensive and production-driven industries, ESG practices may be perceived as cost-intensive, non-core, or even distracting from profitability objectives. Investors may interpret increased ESG reporting in such firms as an indicator of rising compliance costs, regulatory burdens, or strategic misalignment. This supports the critical perspective raised by Abela (2022), who argues that in the absence of performance-based ESG implementation, sustainability reporting can appear performative or burdensome, especially in traditional sectors struggling to balance short-term returns with long-term sustainability goals.

These findings also raise important questions about the quality and authenticity of ESG disclosure. It is possible that the negative valuation effect in manufacturing firms reflects not just investor skepticism, but also the superficiality of reporting, so-called “greenwashing”, where firms disclose ESG information for image purposes rather than substantive performance improvements. Without third-party assurance, clear materiality alignment, or verifiable targets, such disclosures may fail to convince the market of their long-term value. The explanatory power of the models also provides insights worth interrogating. The R² for manufacturing firms (0.181) is stronger than that for non-manufacturing firms (0.091), suggesting that traditional financial and structural variables play a more dominant role in determining value within manufacturing. However, the relatively low R² in both models also indicates that a significant portion of firm value remains unexplained, pointing to the need for further research into other moderating or mediating variables, such as innovation capability, export intensity, board composition, or regional policy regimes.

5. Conclusions and Recommendations

The findings support the overall hypothesis that, in South Africa’s capital market, information disclosure regarding sustainable issues is being rewarded by more favorable market valuations. This bolsters the theoretical notion that ESG disclosure signals future growth potential, improves investor confidence, and lessens information asymmetry. The finding is robust to endogeneity issues solved through GMM, and remains valid after the control for size, leverage, liquidity and firm age. A more nuanced, sectoral analysis does nevertheless demonstrate that the value effects of ESG are far from homogeneous. For non-manufacturing industries, ESG disclosure continues to be a value driver. Investors seem to perceive some sustainability reporting as an indicator of reduced non-financial risk, better stakeholder relationships and better long-term opportunities for growth. For production companies, though, the relationship is the reverse: greater ESG disclosure and lower Tobin’s Q. This adverse market reaction is likely an expression of ongoing skepticism regarding the capital–resource expensive cost–benefit trade-off of ESG activities in a world where the costs of conformance or accusations of ‘greenwashing’ may override thought about short-term profitability. Orthodox financial variables act as anticipated. Increased firm size raises valuation across the board and particularly in manufacturing, implying economies of scale and increased bargaining power. Leverage, though penalized in the entire sample, is value-adding in the two subgroups, implying South African investors today reward moderate debt usage, perhaps as a substitute for growth aspirations subject to industry appropriateness. The age of the firm is a poor determinant, but it identifies reputation and organizational learning as value-contributing drivers, but not across all industries.

This study recommends that corporate managers in non-manufacturing industries need to regard ESG disclosure as a value creation and market differentiation source in the long run, not merely as compliance. By linking ESG initiatives to business core purposes and maintaining transparent, verifiable disclosure, companies can build investor trust, increase their market valuation, and strengthen stakeholder relations. Non-manufacturing companies should continue to invest in increasing the quality and depth of their ESG reporting, connecting it with corporate performance metrics, and making the disclosures material, forward-looking, and relevant material stakeholders. For production companies, the findings indicate the need for a more sophisticated approach. These companies must concentrate on integrating ESG values into their way of doing business, for example, enhancing resource productivity, supply chains that are sustainable, and compliance, as opposed to treating ESG as a cosmetic reporting exercise to the external environment. Strategic alignment of product innovation, green technology, and cost reduction with ESG can turn perceived ESG liabilities into value-capturing assets. Also, manufacturers must work diligently to communicate more effectively the business case for ESG activities and highlight the way such activities generate risk avoidance, operational resilience, and long-term profitability. Policy-wise, South African standard setters and regulators need to think about creating more industry-specific ESG reporting regulations considering the varying realities of varying industries. Policymakers can also enhance the enforcement of integrated reporting regulations and prompt companies to shift from disclosure to performance-based, value-led ESG commitments. Capacity-building exercises to train firms, particularly small and medium enterprises, in the right implementation and disclosure of ESG would improve the overall quality of disclosure in the market. Enabling third-party assurance and harmonizing ESG reporting requirements with international guidelines will increase credibility, comparability, and investor confidence.

Yet some limitations must be noted. First, ESG disclosure scores can be of widely differing quality, depth, and comparability between companies, and this might influence the accuracy of the estimates. South African ESG reporting is still improving, and differences in reporting standards between industries and companies can create noise. Second, the study only captures quantitative disclosure and not qualitative ESG performance and thus may miss capturing the real effect of ESG implementation as compared to reporting. Besides these limitations, the findings remain highly relevant and significant. They also reinforce the perception that ESG disclosure is neither a regulatory nor an ethical requirement but a strategic asset that can be utilized to create value for firms in the capital markets of South Africa.

This research extends the understanding of ESG disclosure’s impact on firm value in South Africa but leaves some limitations and avenues for future research. Aggregate ESG scores based on counts instead of quality or credibility of disclosures are used here, and it is highly important to distinguish between significant and superficial reporting. The adverse impact observed in manufacturing companies indicates sector-specific factors, and more in-depth industry studies or case studies are needed. Despite the use of GMM to control endogeneity, causality persists as an issue that can be addressed by future studies through natural experiments or regulation shocks. The sole dependence on Tobin’s Q as a measure constrains scope; wider measures such as credit spreads, stakeholder returns, or risk-adjusted returns could give a more comprehensive view. Exclusion of private companies and SMEs constrains generalizability; thus, increasing the sample might shed light on ESG impacts by ownership. Finally, longitudinal studies are suggested to track delayed or cyclical ESG effects.