To address the potential endogeneity between ESG environmental performance and corporate performance, this study employs analyst attention as an instrumental variable. Specifically, analyst attention (Analyst) is measured as the natural logarithm of one plus the number of financial analysts following a firm (ln(1 + number of analysts following the firm)).

From the perspective of relevance, analyst attention serves as an important external monitoring mechanism and is closely related to a firm’s ESG environmental performance. Through information interpretation and supervision, analysts enhance the transparency of corporate environmental governance and strengthen compliance pressure, thereby motivating firms to adopt more proactive and standardized environmental practices. From the perspective of exogeneity, analyst attention affects corporate performance only indirectly through its influence on ESG environmental performance, without exerting a direct impact on firm outcomes.

This logic resonates with classical Chinese philosophical thought, particularly the Mohist ideas of jian ai (“impartial care”), fei gong (“non-aggression”), and he er bu. tong (“harmony without uniformity”). The principle of jian ai advocates impartial and independent treatment of all entities, while fei gong emphasizes non-interference and the avoidance of imposing one’s will or creating conflicts of interest. By analogy, analyst attention functions as an external stimulus influencing ESG environmental behavior but does not directly intervene in corporate performance.

From an economic standpoint, analyst attention primarily reflects capital market interest in a firm, driven by external factors such as industry characteristics, firm size, and reputation. For instance, firms in emerging industries may attract extensive analyst coverage due to market attention, even though their financial performance remains unstable. Conversely, mature firms in traditional industries may demonstrate strong and stable performance while attracting fewer analysts. This asymmetry embodies the notion of he er bu. tong, mei mei yu gong—harmony in diversity—implying that while analyst attention influences environmental performance, it does not directly affect corporate performance; the two remain harmoniously independent.

Table 4 reports the regression results using the instrumental variable method. Column (1) shows that the coefficient of Analyst is 0.0095, which is statistically significant at the 1% level. Combined with the results of weak instrument and identification tests, the instrumental variable satisfies the required statistical standards. Column (2) shows that the coefficient of EP is 1.7820, also significant at the 1% level, indicating that even after controlling for potential endogeneity, ESG environmental performance continues to significantly enhance corporate performance, further reinforcing the conclusions of the baseline regression.

To address the potential endogeneity between ESG environmental performance and corporate performance, this study employs a Difference-in-Differences (DID) approach, utilizing the revision of the Environmental Protection Law (EPL) as a quasi-natural experiment. The rationale for using the new EPL is that its implementation significantly strengthened legal constraints and regulatory enforcement, thereby increasing firms’ environmental compliance pressure and, consequently, improving their ESG environmental performance. This policy reform provides an exogenous shock, enabling causal identification of the effect of ESG environmental performance on corporate performance.

To analyze the impact of this policy, two variables are defined: Treat and Post. Following the Guidelines for Industry Classification of Listed Companies revised by the China Securities Regulatory Commission (CSRC) in 2012, firms in high-pollution industries are categorized using industry codes (e.g., B06, B07, B08, B09, C17, C19, C22, C25, C26, C28, C29, C30, C31, C32, D44). The variable Treat is assigned a value of 1 if a firm belongs to one of these high-pollution industries, and 0 otherwise. Post represents the post-policy period, taking a value of 1 for the years 2015 and beyond, and 0 for earlier years. The interaction term Treat × Post is used to capture the policy’s exogenous effect.

To evaluate the robustness of the core findings in response to the policy shock, Model (6) is introduced. In this model, the interaction term β 1 ⋅ Trea t it ⋅ Pos t it captures the differential impact of the policy on high-polluting firms. The term ∑ γ k × Control k , it , it represents the linear combination of control variables, while α i and λ t denote firm and year fixed effects, respectively. Here, i refers to the firm and t refers to time.

Table 5 presents the regression result_s under this quasi-natural experimental setting. In Column (1), without including control variables, the estimated coefficient of Treat × Post is 0.0175 and is statistically significant at the 1% level. In Column (2), after incorporating control variables, the coefficient decreases to 0.0115 but remains significant at the 1% level, indicating that even after considering the exogenous policy shock, the main conclusion of this study remains robust.

As shown in Figure 2, the parallel trends assumption holds, as there is no significant difference between the treatment and control groups before the policy implementation. Post-policy, the treatment group shows a clear upward trend, confirming the policy’s effective intervention.

To further verify the robustness of the policy shock and eliminate potential omitted-variable bias, Figure 3 presents the placebo test results. A 500-round random sampling process generates pseudo-policy variables, keeping all control variables constant. The randomly assigned policy implementation time yields a coefficient of 0.0115, within the range of low-probability events, suggesting that omitted-variable bias has a negligible impact on the core findings.

Tobin’s Q, by comparing the sum of a firm’s market value and total liabilities with its total assets, not only reflects the firm’s growth potential but also provides key information to stakeholders regarding its market valuation. Specifically, the market value of non-tradable shares is estimated by multiplying net asset value per share by the number of non-tradable shares, while net debt is calculated by deducting items such as employee compensation payable, taxes payable, dividends payable, other payables, and deferred income tax liabilities from total liabilities.

Because Tobin’s Q emphasizes the evaluation of long-term corporate performance, and Return on Equity (ROE) serves as an important indicator of a firm’s capital utilization efficiency, reflecting its ability to generate profits from shareholders’ equity, this study employs Tobin’s Q and ROE as alternative measures of corporate performance.

As shown in Table 6, from the perspectives of shareholder return (ROE), market value (Tobin’s Q), and profitability (ROA), the coefficient of ESG environmental performance (EP) remains significantly positive across all models. This finding suggests that environmental performance not only enhances firms’ current performance but also exerts a persistent dynamic effect, indicating that the optimization of environmental governance can comprehensively strengthen corporate performance.

In the financial field, the long-tail effect is an important phenomenon that highlights the potential impact of low-probability tail events on overall economic activities. To eliminate the influence of the exogenous shock caused by the COVID-19 pandemic, this study shortens the sample period to 2006–2019 and re-conducts the empirical tests. The results indicate that ESG environmental performance (EP) continues to exert a significantly positive effect on corporate performance, confirming the robustness of the main findings even after excluding pandemic-related distortions (Table 7).

This section conducts empirical tests based on Equations 4 and 5. The regression results are reported in Columns (1)–(3) of Table 9.

As shown in Column (1), the coefficient of the interaction term EP × MC is −0.3156, which is statistically significant at the 5% level. This indicates that market competitive position significantly weakens the positive effect of environmental performance on green innovation. In highly competitive markets, firms tend to prioritize short-term financial outcomes and reduce resource allocation to long-term, high-risk green innovation projects, thereby limiting the potential of environmental performance to translate into green innovation. These results provide empirical support for Hypothesis H_2a.

In Column (2), the coefficient of the interaction term Patent × MC is 0.0048, significant at the 5% level. This finding suggests that market competitive position significantly strengthens the mediating effect of green innovation in the relationship between environmental performance and corporate performance. Consistent with the dynamic capability theory, firms with stronger competitive positions can achieve technological breakthroughs and accumulate intellectual property through green innovation. Such firms not only enhance the conversion efficiency of green innovation but also optimize resource allocation, thereby expanding the pathways through which environmental performance improves corporate performance. Hence, Hypothesis H_2b is supported.

In Column (3), the coefficient of the interaction term EP × MC is 0.0501, which is statistically significant at the 1% level. This result indicates that the positive impact of environmental performance on corporate performance is more pronounced among firms with higher market competitive positions. Furthermore, this finding implies that the positive moderating effect of market competitive position on the relationship between green innovation and corporate performance (H_2b) exceeds its negative moderating effect on the relationship between environmental performance and green innovation (H_2a). Therefore, Hypothesis H_2c is also verified.

To analyze whether the effect of environmental performance on corporate performance (ROA) differs across ownership types, firms are divided into state-owned enterprises and non-state-owned enterprises (non-SOEs). The group regression results are presented in Table 10.

The findings reveal that the positive impact of ESG environmental performance on firm performance is more pronounced among non-SOEs. This may be because non-state-owned enterprises tend to be more market-oriented and sensitive to sustainability-driven reputation effects. By actively fulfilling environmental responsibilities, these firms can enhance their corporate image, attract investment, and strengthen their market competitiveness, thereby converting environmental performance into performance gains more efficiently.

In contrast, while state-owned enterprises also exhibit a positive relationship between environmental performance and corporate performance, the effect is relatively weaker. The reason may lie in their relatively abundant access to resources and policy support, which may reduce market-driven incentives for proactive environmental responsibility. Consequently, the conversion efficiency of environmental performance into tangible performance outcomes is lower in SOEs compared with non-SOEs.

Overall, this divergence reflects the market-driven advantages of non-SOEs in green transformation while highlighting the need for SOEs to improve the efficiency of translating environmental performance into performance outcomes. These findings provide valuable implications for optimizing firm-specific strategies and policy frameworks to promote differentiated yet coordinated green development.

The influence of internal driving forces and external constraints on ESG environmental performance is particularly significant. In this study, data related to executives’ environmental backgrounds were manually extracted from publicly available résumé information on the Sina Finance website. The specific identification criteria are as follows: if the résumé of a top executive contains keywords such as “environment,” “environmental protection,” “low carbon,” or “green,” the executive is considered to possess an environmental background. In addition, based on the data released by the Ministry of Ecology and Environment (MEE) and the Institute of Public and Environmental Affairs (IPE), we determine whether a firm has a record of environmental penalties. The environmental background of executives and the existence of penalty records serve as indicators of internal driving and external constraints, respectively, and both are assigned a value of 1 if present and 0 otherwise.

As shown in Table 11, the promoting effect of ESG environmental performance on corporate performance is strongest in firms whose chairpersons have a green background. The coefficient is 0.0553, significant at the 1% level. In firms without such a background, the coefficient decreases to 0.0267 and remains significant only at the 5% level. The results of the Fisher test further confirm the significant difference between the two groups, indicating that chairpersons with green backgrounds can more effectively promote the implementation of environmental responsibility strategies, enhance firms’ resource integration capabilities, and improve market recognition—thereby amplifying the positive impact of ESG environmental performance on corporate performance.

For firms without environmental penalty records, ESG environmental performance significantly improves firm performance, with a coefficient of 0.0296, significant at the 1% level. However, in firms that have been penalized, this effect is substantially weakened, with the coefficient dropping to 0.0164 and becoming statistically insignificant. The Fisher test again shows a significant difference between the two groups, suggesting that firms penalized for environmental violations suffer reputational damage, which undermines their credibility in capital markets and reduces their ability to secure policy support. Furthermore, such firms are forced to divert resources toward fines and remediation rather than long-term performance improvement. The market’s negative perception of penalized firms further diminishes their ability to enhance firm value through environmental performance, highlighting the critical role of environmental compliance in the transformation of environmental performance into economic outcomes.

Given that significant differences exist across industries in terms of environmental governance, the impact of ESG environmental performance on corporate performance may vary depending on industry characteristics. To more comprehensively examine the relationship between ESG environmental performance and firm performance, this study further divides firms into high-tech versus non–high-tech enterprises and high-pollution versus non–high-pollution enterprises for comparative analysis.

The classification of high-tech enterprises follows the Administrative Measures for the Recognition of National High-Tech Enterprises and the Guidelines for Industry Classification of Listed Companies issued by the China Securities Regulatory Commission (CSRC). Firms that meet these criteria are identified as high-tech enterprises (assigned a value of 1), while others are assigned 0. Similarly, the definition of high-pollution enterprises is based on the CSRC’s Guidelines for Industry Classification of Listed Companies and the Catalogue for Environmental Verification of Listed Companies by Industry released by the Ministry of Ecology and Environment. Firms belonging to industries listed in these documents are classified as high-pollution enterprises (assigned a value of 1), and the rest are assigned 0.

As shown in Table 12, the impact of ESG environmental performance on corporate performance is insignificant among high-tech enterprises, with the coefficient of EP being 0.0150 and statistically insignificant. This suggests that the marginal improvement effect of environmental performance on high-tech firms’ performance is limited. The possible reason is that high-tech enterprises already possess strong technological capabilities and competitive advantages, making the incremental benefits of environmental improvements less evident.

In contrast, among non–high-tech enterprises, the coefficient of EP is 0.0372 and statistically significant, indicating that firms in traditional industries can more easily obtain capital market support and policy resources through improved environmental performance, thereby enhancing their overall performance.

For high-pollution enterprises, the promoting effect of environmental performance is weak and statistically insignificant, with the coefficient of EP being 0.0087. This may be because high-pollution firms allocate more resources toward compliance costs and pollution control, which diminishes the direct positive effect of environmental performance on firm performance. Furthermore, as the market holds higher environmental expectations for high-pollution firms, marginal improvements in environmental performance are less likely to translate into significant performance gains.

Conversely, among non–high-pollution enterprises, the coefficient of EP is 0.0348 and statistically significant, suggesting that these firms are more capable of converting environmental improvements into market trust and performance returns.

Overall, these results demonstrate that industry characteristics significantly influence the relationship between ESG environmental performance and corporate performance. The industry context plays a critical role in shaping the effectiveness of firms’ environmental management practices and their ability to convert environmental performance into economic outcomes.