The impact of ESG performance on firm performance: the moderating role of audit quality – evidence from China

In recent years, ESG performance has evolved from a symbolic commitment to a strategic imperative. As stakeholders demand greater accountability, firms face pressure to translate sustainability initiatives into tangible value. However, the financial impact of ESG remains contested. While some studies suggest ESG improves reputation and financing access, others highlight costs, inconsistencies, and greenwashing risks. Recent comparative evidence further indicates that these outcomes vary significantly across institutional settings.

Audit quality offers a potential mechanism to reconcile these mixed findings. High-quality audits mitigate information asymmetry, signaling that ESG claims are verifiable rather than merely symbolic. Recent literature emphasizes that governance structures—including audit committees and external assurance—crucially condition ESG outcomes. For instance, Abdullah (2024) and Sahu et al. (2025a, b) demonstrate that the ESG–performance link is strengthened by robust internal and external oversight mechanisms. Complementing this, Farhan et al. (2026) highlight the synergistic role of CSR and audit committees in validating environmental practices, while Cai et al. (2024) show that media coverage further amplifies the reputational consequences of audit signals. These findings underscore the importance of examining how audit quality shapes the ESG–performance relationship.

China provides a compelling setting to address this due to its “dual carbon” goals and evolving disclosure mandates. However, reporting quality remains uneven, and ownership structures create distinct incentives. State-owned enterprises (SOEs) often prioritize political mandates, whereas non-SOEs are driven by market pressures. Institutional theory suggests these differences shape both ESG incentives and reporting credibility. Recent reviews confirm that ESG outcomes are strongly conditioned by such governance arrangements.

Despite these dynamics, empirical evidence remains limited. Prior studies in China have largely examined ESG's direct impact, but few have explicitly tested audit quality as a moderating channel. Moreover, even fewer works extend the analysis to ownership heterogeneity, despite recognition that institutional environments crucially shape ESG outcomes. Addressing these gaps, this study asks: (1) Does ESG performance enhance financial outcomes in China? (2) Is this relationship stronger under higher audit quality? (3) Does this moderation differ between SOEs and non-SOEs?

This paper makes three contributions. First, it provides updated large-sample evidence (2015–2024) utilizing Huazheng ESG data to capture recent regulatory shifts. Second, it introduces audit quality (abnormal fees and Big 4) as a moderating mechanism, situating the analysis alongside emerging evidence on governance contingencies (Abdullah, 2024; Sahu et al., 2025a, b; Cai et al., 2024; Farhan et al., 2026). Third, it incorporates ownership structure as an institutional lens, revealing divergent ESG–audit dynamics between SOEs and non-SOEs, addressing recent calls for more nuanced analyses.

Corporate sustainability has evolved into a strategic imperative. Stakeholder and legitimacy theories suggest that effective ESG practices build trust, lower risk, and align with societal expectations, thereby enhancing valuation and resilience (Freeman, 1984; Fatemi et al., 2018; Martinez et al., 2021; Okafor et al., 2021). While empirical evidence in China and globally largely supports a positive ESG–performance link (Busch and Friede, 2018), particularly in demonstrating that high ESG-performing firms exhibit greater resilience and value stability during market downturns (Broadstock et al., 2021), contrasting findings highlight compliance costs and greenwashing risks that may erode financial returns (Galant and Cadez, 2017; Boubaker et al., 2020). These divergences often stem from measurement variations and institutional maturity, with recent Chinese data using Huazheng scores showing stronger consistency than global ratings (Alahdal et al., 2023). Taken together, while ESG creates value, its benefits depend on credibility and context.

Despite potential benefits, unaudited ESG disclosures face credibility concerns due to managerial opportunism. Agency and signaling theories posit that high-quality audits—proxied by Big 4 affiliation or fee premiums—mitigate information asymmetry and validate corporate claims (DeAngelo, 1981; Spence, 1974; DeFond and Zhang, 2014). Furthermore, rigorous external financial audits tend to create positive spillover effects on the reliability of non-financial disclosures (Cohen and Simnett, 2015). Recent research indicates that audit quality amplifies the ESG–performance nexus by ensuring disclosures reflect substantive practices (Velte, 2025). Emerging evidence also highlights that audit effectiveness is synergistic with internal governance: Abdullah (2024), Sahu et al. (2025a, b), and Farhan et al. (2026) demonstrate that audit and CSR committees interact with external assurance to strengthen sustainability outcomes. In China's evolving regulatory landscape, where media coverage further reinforces audit signals (Cai et al., 2024), external audits substitute for weak formal institutions, potentially enhancing ESG's financial payoff.

Institutional theory suggests organizational strategies reflect external legitimacy pressures (DiMaggio and Powell, 1983). In China's unique institutional setting, corporate non-financial disclosures are frequently utilized as symbolic tools to align with government directives (Marquis and Qian, 2014). Consequently, SOEs often pursue ESG to satisfy political mandates rather than for immediate financial returns. Conversely, non-SOEs rely on ESG strategically to differentiate themselves in competitive markets and significantly lower their cost of capital to alleviate financing constraints (El Ghoul et al., 2011; Abdullah, 2024). Therefore, assurance mechanisms are likely more critical for non-SOEs to validate their disclosures to external stakeholders, whereas SOEs are inherently shielded by political support and state-backed resource access (Velte, 2025; Alahdal et al., 2023). Thus, ownership heterogeneity conditions the audit–ESG dynamic.

This study utilizes an unbalanced panel of Chinese A-share listed firms (2015–2024). We apply the following screening criteria: (1) excluding financial institutions; (2) removing firms under Special Treatment (ST/PT); (3) dropping observations with missing core variables; and (4) retaining firms with at least three consecutive years of data. Continuous variables are winsorized at the 1st and 99th percentiles. The final sample comprises 35,175 firm-year observations (∼5,000 unique firms). Data are integrated from Huazheng Information (ESG scores), CSMAR (financials and audit details), and annual reports (audit opinions). All analyses are conducted in Stata 18.

The definitions, measurements, and data sources for all variables used in this study—including dependent, independent, moderating, and control variables—are summarized in Table 1.

To examine the impact of ESG on firm performance, we employ a high-dimensional fixed effects model. This specification is appropriate as it controls for unobserved time-invariant firm heterogeneity (such as corporate culture) and common macroeconomic shocks.

Where $Perfit$ denotes firm performance (ROA or ROE), $ESGit$ is the ESG score, and $Xit$ is the vector of control variables. $μi$ represents firm fixed effects, and $λt$ represents year fixed effects. Standard errors are clustered at the firm level to account for serial correlation within firms over time.

Moderation Model (Audit Quality): To test whether audit quality strengthens the ESG–performance relationship, we augment the baseline model with interaction terms:

Where $AQit$ represents the audit quality proxies (Abnormal Fees, Big 4, etc.). Importantly, to mitigate multicollinearity concerns arising from interaction terms, we mean-center continuous variables (ESG and Audit Fees) before calculating the product term. In this specification, $β3$ captures the moderating effect. A significant positive $β3$ would indicate that high audit quality enhances the value-relevance of ESG activities.

To ensure the reliability of our baseline findings and address potential endogeneity, we conduct a battery of rigorous robustness checks. Specifically, we employ alternative variable measurements (ROE and a year-standardized ESG score) and incorporate high-dimensional fixed effects (Industry × Year) to absorb time-varying unobserved shocks. To mitigate reverse causality and omitted variable bias, we utilize a two-step System GMM estimator and an instrumental variable (IV) approach based on local audit market characteristics. Finally, given China's unique institutional environment, we conduct subsample analyses to examine whether the ESG–performance relationship and audit moderation exhibit structural differences across ownership regimes (SOEs vs. non-SOEs).

Table 2 presents the descriptive statistics for the variables used in this study. The mean return on assets (ROA) is 0.033 with a standard deviation of 0.068. The average ESG score is 41.104, indicating a moderate level of corporate sustainability performance within the sample. Regarding the binary variables, 6.19% of the observations are audited by Big 4 auditors, 97.09% received a standard unqualified opinion, and 33.42% of the firms are state-owned enterprises (SOEs).

Table 3 reports the Pearson correlation coefficients among the main variables. As expected, ESG performance is positively and significantly correlated with both profitability measures (ROA and ROE). The maximum correlation between independent variables is 0.750 (between LnAudit and Size). Combined with the low VIF values reported earlier, these results indicate that multicollinearity is not a material concern for our regression models.

Table 4 reports the baseline regressions examining the impact of ESG performance on firm profitability (ROA and ROE). Across all specifications, with and without firm-level controls, ESG scores are positively and significantly associated with firm performance (p < 0.01). In terms of economic magnitude, a one-standard-deviation increase in ESG translates into a meaningful rise of approximately 0.188% points in ROA and 0.678% points in ROE (Columns 2 and 4). These results provide strong empirical support for Hypothesis 1, indicating that ESG engagement significantly enhances financial outcomes in the Chinese capital market, capturing performance-enhancing mechanisms beyond standard firm characteristics.

Table 5 presents the moderating role of audit quality—proxied by abnormal audit fees, Ln(Audit fees), Big-4 auditor, and standard unqualified opinion—in the ESG–performance relationship. To reduce multicollinearity, continuous variables are mean-centered (denoted as ESG_mc) before constructing the interaction terms.

Across all specifications, the interaction terms between ESG_mc and the audit quality proxies are statistically insignificant, indicating that Hypothesis 2 is not supported. Higher audit quality does not materially strengthen the ESG–performance linkage in our sample. Unlike prior studies documenting stronger ESG–performance links under robust audit oversight (e.g. Velte, 2025), our evidence suggests these mechanisms are less effective in the Chinese audit market. A plausible explanation is that the dominance of standard unqualified opinions and relatively low Big-4 penetration reduce the incremental signaling value of external audits. Furthermore, audit fees in this context may largely capture client complexity and risk rather than pure audit effort, thereby diluting their moderating role.

To address potential endogeneity, we re-estimated the baseline models using lagged ESG and dynamic panel GMM. Tables 6 and 7 results suggest that baseline estimates may overstate the short-term benefits of ESG. Instead, these additional tests indicate that ESG advantages are more likely to emerge through long-term channels—such as reputation, financing, and risk mitigation—rather than immediate accounting returns.

To further examine whether the ESG–performance relationship varies across ownership structures, we split the sample into SOEs and non-SOEs. As shown in Table 8, ESG performance is positively associated with profitability in both subsamples, with stronger effects for SOEs than for non-SOEs. Regarding audit quality (proxied by abnormal audit fees), we find consistently negative main effects for both groups, reflecting audit frictions that reduce profitability. However, the interaction terms (ESG X AbnFees) remain insignificant in both subsamples, indicating that abnormal fees do not systematically moderate the ESG–performance link regardless of ownership. Overall, these results suggest that ESG benefits are robust across ownership structures. Institutional differences remain important: SOEs may benefit from government support that cushions audit frictions, whereas non-SOEs are more exposed to market monitoring pressures.

To validate the reliability of our findings against potential measurement artifacts, we re-estimated the baseline and moderation models using a year-standardized ESG measure (ESG_Z). As shown in Table 9, ESG_Z maintains a significant positive association with both ROA and ROE, yielding comparable economic magnitudes to our primary measure and robustly supporting Hypothesis 1. Consistent with our main findings, the interaction term (ESG_Z X AbnFees) remains statistically insignificant under this specification. Overall, these tests confirm that our main conclusions are robust to alternative variable scaling methods.

The empirical evidence of this study provides partial but meaningful support for the hypothesized relationships.

Regarding H1, our baseline regressions consistently show that ESG performance improves firm profitability (both ROA and ROE), even after controlling for firm size, leverage, board structure, and growth opportunities. This supports the argument, grounded in stakeholder and legitimacy theory, that ESG activities generate reputational benefits, strengthen stakeholder trust, and reduce financing constraints. Our findings align with recent evidence from emerging markets by Alahdal et al. (2023) who emphasize that in transitional economies, sustainability disclosure serves as a critical mechanism for reducing information asymmetry and enhancing competitive advantage. However, our endogeneity analyses indicate that the short-term benefits of ESG weaken once dynamic persistence is considered, suggesting that the positive effects of ESG are more likely to materialize through long-term mechanisms such as market reputation and risk mitigation rather than immediate accounting returns.

In addition, robustness checks using year-standardized ESG scores confirm that our results are not sensitive to scaling or coverage issues in Huazheng data. ESG_z remains significantly positive for both ROA and ROE, with comparable economic magnitudes to the baseline specification. This evidence strengthens confidence in H1 by demonstrating that the ESG–performance link holds under alternative measures and is not an artifact of data construction.

With respect to H2, the evidence on the moderating role of audit quality is nuanced. While the interaction with abnormal audit fees is statistically insignificant, we find that Big-4 auditors positively moderate the ESG–profitability relationship in specific contexts. These findings parallel the work of Sahu et al. (2025a, b), who document that specialized audit oversight (such as audit committee expertise) serves as a catalyst for advancing sustainable development goals. Similarly, our results resonate with Farhan et al. (2026), who highlight, in the context of OECD countries, that external assurance mechanisms work synergistically with internal governance (CSR committees) to validate environmental practices. The mixed evidence in our study—where Big 4 affiliation matters more than fee premiums—suggests that in the Chinese market, the reputational endorsement of top-tier auditors is a more powerful signal for ESG credibility than the resource input proxied by fees.

The heterogeneity analysis further highlights the institutional dimension. ESG performance exerts positive effects on both SOEs and non-SOEs, but the magnitude is stronger for SOEs. This divergence is rooted in China's unique institutional fabric, where state ownership often provides implicit guarantees that substitute for external monitoring. For SOEs, ESG engagement aligns with government political mandates, directly yielding resource access and legitimacy. By contrast, non-SOEs are more sensitive to audit frictions, as evidenced by the larger negative coefficient of abnormal audit fees. This supports the view that for private firms lacking political shields, strong governance and high-quality audits act as vital compensatory mechanisms to strengthen the ESG–performance link in a region characterized by lower market trust.

Overall, the results confirm that ESG investments are generally beneficial (supporting H1), and these findings remain robust under alternative measurement approaches. The role of audit quality as a moderator (H2), however, varies across proxies. The mixed evidence underscores the complexity of ESG–performance dynamics in emerging markets. As highlighted by Sahu et al. (2025a, b) regarding board diversity and Farhan et al. (2026) regarding CSR committees, the effectiveness of ESG is contingent on a constellation of governance factors. These insights contribute to ongoing debates and highlight the need for future research to employ alternative assurance proxies (e.g. audit partner characteristics) and to disentangle the long-term versus short-term performance channels of ESG.

This study investigates the relationship between ESG performance and firm profitability in Chinese A-share listed companies, with a particular focus on the moderating role of audit quality and ownership heterogeneity. Using a large panel dataset covering 2015–2024, we find that ESG engagement generally enhances firm performance, providing support for our first hypothesis. Importantly, robustness tests using standardized ESG measures confirm that this positive effect is not sensitive to scaling or coverage issues.Our conclusions are robust to alternative ESG scaling and dynamic specifications (including System- and Difference-GMM). However, dynamic panel models reveal that the short-term accounting benefits of ESG are less robust once profit persistence is controlled for, suggesting that ESG creates value primarily through long-term channels such as reputation, financing advantages, and risk mitigation.

The evidence for our second hypothesis is more nuanced. While audit quality—proxied by abnormal audit fees and Big-4 auditors—has direct effects on profitability, its moderating role on the ESG–performance nexus is weak and inconsistent across specifications. This indicates that audit quality does not universally amplify ESG outcomes, but its influence depends on institutional and firm-level contexts. The heterogeneity analysis further shows that SOEs benefit more strongly from ESG, reflecting government support and legitimacy channels, whereas non-SOEs are more vulnerable to audit frictions, consistent with institutional theory.

This study makes three contributions. First, it extends ESG research in emerging markets by providing up-to-date evidence from China through 2024. Second, it enriches the audit quality literature by demonstrating that different proxies yield divergent insights. Third, it highlights the institutional role of ownership structure in conditioning ESG outcomes, offering new perspectives on SOE versus non-SOE dynamics.

Nevertheless, several limitations remain. Audit fees may capture audit complexity rather than quality; future research could draw on partner-level attributes, inspection outcomes, or accrual-based measures. ESG scores also face challenges of coverage and comparability, underscoring the need for triangulation with alternative rating agencies. Finally, although we address endogeneity using lagged models, system GMM, and robustness checks, stronger causal identification (e.g. natural experiments, instrumental variables, or propensity score matching) would enhance inference.

Overall, the findings suggest that ESG is a valuable long-term strategic investment for Chinese firms, but its performance benefits are contingent on governance quality and institutional context. For policymakers, this highlights the importance of improving ESG disclosure standards and developing reliable assurance mechanisms. For managers, the evidence emphasizes that ESG engagement can strengthen legitimacy and long-run profitability, particularly when supported by credible governance and high-quality audits.