How Do Corporate Environmental, Social, and Governance (ESG) Factors Affect Financial Performance?

Abstract

Responsible investments are becoming increasingly relevant in stakeholder decision-making, propelled by the emphasis on sustainable development. Specifically, enterprises should acknowledge the significance of creating value for multiple shareholders based on the environment, society, and corporate governance. In this article, we contribute to the theoretical and empirical literature on corporate environmental, social, and governance (ESG) performance in China. This paper employs a two-way fixed-effect model to examine the influence of ESG activities on financial performance, focusing on 3268 Shanghai and Shenzhen A-share companies that have consistently participated in such activities from 2011 to 2022. The findings indicate that improvements in ESG practices positively influence corporate financial performance, with property rights and industry categorization moderating this relationship. Furthermore, agency cost, financing cost, social reputation, market power, and enterprise innovation partially mediate ESG performance and financial performance. This study encourages enterprises to integrate sustainable value creation into the national development strategy, thereby achieving harmonious economic and social development.

1. Introduction

In 2006, the United Nations Principles for Responsible Investment (UN PRI) advocated environmental, social, and governance (ESG) factors as part of investment decisions and required investors to align their responsible investment practices with the UN Sustainable Development Goals [1]. Recently, as people have tended to attach importance to non-financial risks, the concept of ESG has become mainstream worldwide and gradually gained widespread attention in China. From 2016 to 2020, ESG investment has developed rapidly in China, during which investors have started incorporating ESG into their investment strategies, and ESG’s impact on the capital markets has progressively increased. The primary reason behind this phenomenon is that the concept of responsible investment with the themes of environment (E), society (S), and governance (G) aligns with China’s new development stage and new development concept. Specifically, the 19th National Congress of the Communist Party of China (CPC) put forward the three critical battles against potential risk, poverty, and pollution in 2017. Firstly, preventing financial risks requires attention to corporate governance, which is consistent with the principle of governance. Secondly, the battle against poverty means enterprises should shoulder more social responsibilities, aligning with societal needs. Thirdly, pollution prevention and control aims to protect the environment, which aligns with sustainable environmental protection. Additionally, the successive promulgation of policies related to the concept of ESG-responsible investment marks the continuous development and improvement of China’s green industry transformation and carbon financial market system [2]. Therefore, by further exploring the intrinsic transmission mechanism of ESG and financial results from different perspectives, the listed companies will be guided to assume more social responsibilities and internalize the externalities caused by the pursuit of profit maximization into corporate costs and thus promote their financial health. In the long run, it is beneficial to enhance the corporate governance system and promote the implementation of the national development strategy.

Recently, more and more scholars began to pay attention to the impact of ESG performance on corporate financial performance. For example, Shen and Li (2024) [3], Inamdar (2024) [4], Jucá et al. (2024) [5], and Giannopoulos et al. (2022) [6] pointed out that ESG performance can significantly improve corporate financial performance but did not further explore the reasons behind the significant improvement of corporate financial performance after the improvement of ESG performance. I addition, there is literature that further tries to study the mechanism of ESG on corporate financial performance. For example, Du and Kong (2024) [7] pointed out that the executives with good ESG performance have a higher shareholding ratio, which is easier to improve financial performance. Based on the research of heavy polluting enterprises, Chen and Luo (2024) [8] found that good ESG performance can significantly improve the green innovation ability of enterprises and then improve financial performance. Su and He (2024) [9] believe that ESG performance improves corporate reputation and thus financial performance. Marie et al. (2024) [10] believe that CEO political connections will weaken the role of ESG performance in improving financial performance. Luo et al. (2024) [11], Li and He (2024) [12] believe that ESG performance can effectively alleviate financing constraints and improve corporate financial performance. Although the above literature has studied how ESG performance affects corporate financial performance, the research on the mechanism is still scattered, lacking a systematic theoretical framework and empirical support.

Theoretically, ESG can significantly reduce the agency cost and capital cost of enterprises on the cost side. Through green production, energy conservation, and emission reduction, enterprises can effectively reduce environmental risks, decrease fines and compensation due to environmental violations [11], and then reduce agency costs. At the same time, enterprises that actively fulfill their social responsibilities and improve the level of corporate governance can enhance investor confidence and attract more socially responsible capital, so as to obtain financing at a lower cost and optimize the capital structure [12]. On the sales side, enterprises with excellent ESG performance can significantly improve social reputation, enhance the trust and loyalty of consumers and partners [13,14], expand market share, and enhance market power. On the development side, the ESG concept also urges enterprises to continuously explore innovation paths [15], such as the greening of product design, the improvement of production processes, and the sustainability of business models. This will open up new growth points for enterprises and further improve business performance and competitiveness.

Given the above background, this paper selects 3268 Shanghai and Shenzhen A-share companies that have been continuously involved in ESG activities from 2011 to 2022 as a research sample. Further, it explores the intrinsic mechanisms of corporate ESG and financial performance through the cost side, sales side, and development side. Specifically, we construct a two-way fixed-effect model to clarify the correlation between corporate involvement in ESG and their financial performance, presented as multiple linear regression. The conclusion remains valid after rigorous robustness and endogeneity tests, such as substituting variables, employing multidimensional fixed effects, adjusting sample intervals, and using an instrumental variable approach. Furthermore, we verify the mediating effect of agency cost, financing cost, social reputation, market power, and enterprise innovation. Finally, the heterogeneity analysis is conducted based on the individual characteristics of enterprises to provide a theoretical basis for how ESG performance affects financial performance in different types of enterprises.

Our study contributes to research on ESG in China and advances the field in two significant ways. First, previous studies either lacked in-depth research on the interaction mechanism between ESG and corporate financial performance or only focused on a single dimension to explore how ESG can improve corporate financial performance. From the perspective of the cost side, sales side, and development side, this study explores the internal mechanism of ESG from the aspects of management cost, capital cost, social reputation, market power, and technological innovation, which helps to further explore the internal influence mechanism of ESG, broaden the existing research boundary, and enrich the relevant research content.

Second, promoting the ESG performance of enterprises can use market means to deal with the externalities that bring barriers to sustainable development and guide enterprises to adjust corresponding sustainable investment strategies at the micro-level. In addition, the fulfillment of ESG responsibility can establish a good reputation for enterprises, promote long-term value appreciation by reducing financing costs, and maximize the benefits of enterprises. At the macro-level, this study encourages enterprises to combine sustainable value creation with the national development strategy, conform to the principal strategy of the national development of green finance, and achieve the industrial and economic goals set by the state, which is of practical significance.

2. Theoretical Analysis and Research Hypotheses

In this part, this essay will make a research hypothesis on the main effect related to the relationship between corporate involvement in ESG and financial outcomes. Then, this paper will further make a research hypothesis about the mediate effect from the perspective of the cost side, sales side, and development side.

2.1. ESG Performance and Financial Performance

ESG is an index system to comprehensively evaluate the performance of enterprises in environmental protection, social responsibility, and corporate governance. The score of ESG can fully reflect the performance of enterprises in environmental protection, social responsibility, and internal governance. First of all, in order to obtain a higher ESG score, enterprises usually purchase environmental treatment equipment or update more advanced production equipment, reduce energy consumption and pollution emissions in the production process, and then actively respond to the national green low-carbon strategy. In this process, enterprises will increase the production of green products and reduce the output of polluting products, improve market competitiveness by adjusting product production strategies, and then improve the economic benefits of enterprises with differentiated green products [16]. Secondly, ESG can fully reflect the performance of enterprises’ social responsibilities. Enterprises that actively fulfill social responsibilities are usually more likely to establish a good image among stakeholders, such as suppliers and consumers, and then form close business ties with upstream and downstream enterprises [14], enrich the supply sources of raw materials, and increase the sales channels of enterprises’ products, so as to improve the financial performance of enterprises. Thirdly, enterprises with good ESG performance usually have more perfect and effective internal governance processes, which can improve the decision-making efficiency and execution efficiency, so that when facing new development opportunities, enterprises can quickly gain insight into market demand, accelerate the strategic deployment and production progress of product production, and put products into the market at a faster speed, so as to effectively improve financial performance. Finally, knowledge management strategies refer to how an enterprise acquires, organizes, shares, and uses knowledge resources to facilitate effective communication and collaboration between the enterprise and its stakeholders. It also has a positive impact on improving the corporate governance structure [17,18]. Integrating knowledge management strategies with ESG factors has the potential to catalyze technological innovation and optimize processes, thereby improving operational efficiency and product quality [19]. These changes will ultimately be reflected in the company’s financial statements and contribute to financial performance.

Hypothesis 1. 

Good ESG performance can improve corporate financial performance.

2.2. The Mediating Effect of Cost Side

ESG can improve financial performance by reducing agency costs. The disclosure of ESG information by enterprises can convey more internal information to investors, help investors form an effective supervision mechanism for the management decision-making and daily operation of enterprises, effectively alleviate the information asymmetry between enterprises and external investors, and greatly reduce the agency costs of enterprises [13,20]. Moreover, the principal-agent theory believes that the separation of enterprise ownership and management rights will lead to the improper behavior of enterprise management in the conflict of interests with shareholders, which may damage the brand image of the enterprise and reduce the competitiveness of the enterprise, which is not conducive to the improvement of financial performance. ESG performance has significantly improved the transparency of enterprise information, which is conducive to shareholders’ timely monitoring and restricting the improper behavior of the management, and preventing the management from reducing the agency cost of the enterprise through financial manipulation or shortsighted behavior for their own interests [21]. In addition, a perfect corporate governance structure can significantly improve the transparency of internal information, ensure the accurate and timely transmission of information across departments, and reduce the cost of internal communication. No matter what kind of reduction in agency costs mentioned above, it will help enterprises put more resources into their production and operation activities and enhance market competitiveness by focusing on enterprise production and product innovation, so as to enhance enterprise profitability and financial performance.

ESG can improve financial performance by reducing capital costs. Investors tend to focus on ESG out of social preferences and altruistic motives [22]. Good ESG performance not only reflects the positive response of enterprises to the national energy conservation and emission reduction policies but also shows that enterprises have sustainable development potential and long-term investment value [23], which can attract more green investment and reduce financing costs [24]. At the same time, the disclosure of ESG information by enterprises can establish a better long-term relationship with stakeholders and form “social capital”. On the one hand, it can obtain direct or indirect financing at a lower cost through a third-party guarantee, especially debt financing [25]. On the other hand, it can overcome internal and external information asymmetry and reduce financing costs by promoting information flow [26]. The reduction in capital cost means that enterprises can obtain the funds needed for production and operation at a lower cost, ease the capital constraints of enterprises, improve the liquidity of funds, and help enterprises use more abundant funds to engage in core businesses such as technology research and development and market expansion, so as to promote the long-term high-quality development of enterprises and improve financial performance.

Hypothesis 2. 

ESG performance can significantly improve financial performance through decreasing management cost and financing cost.

2.3. The Mediating Effect of Sales Side

ESG can improve corporate financial performance by improving social reputation. Existing studies believe that good corporate reputation stems from the comprehensive embodiment of product and service quality, consumer satisfaction, and voluntary social responsibility [27]. A good social reputation can be regarded as an intangible asset of an enterprise, which can maintain the competitive advantage of the enterprise in the market and is conducive to the improvement of financial performance. First, by improving ESG performance, enterprises can establish a high-quality image of environmental protection, a strong sense of social responsibility, and a perfect corporate governance structure among suppliers and customers so as to promote enterprises to obtain priority in purchasing raw materials and providing product sales activities so as to improve financial performance. Secondly, the good ESG performance of enterprises can also send the signal of their efforts in environmental protection and social responsibility sharing to the government, enhance the reputation of enterprises in government agencies, and make it easier to obtain government support [28]. They can not only enjoy various government subsidies and preferential policies to increase cash flow but also rely on government endorsement to improve the competitiveness of their products, which helps to improve financial performance. Thirdly, the information on charitable corporate donations and public welfare activities in the ESG report can effectively convey the commitment of enterprises to social responsibility to external stakeholders, cultivate social trust capital, improve corporate reputation [25], attract more consumers and partners, and promote the improvement of corporate financial performance.

ESG can improve the financial performance of enterprises by increasing market power. First of all, with global climate change and the increasing shortage of resources, consumers’ demand for environmental protection products is growing, and enterprises producing green and low-carbon products can occupy an obvious competitive advantage in the market. By optimizing the production process and improving the production technology, enterprises can improve the cleanliness of products, not only significantly reduce environmental pollution and energy consumption, but also effectively meet the needs of consumers [29], which helps enterprises to obtain a higher market position in the fierce market competition so as to improve financial performance. Secondly, to a certain extent, enterprises’ active social responsibility also shows that they attach importance to after-sales service and consumer rights protection, which helps to establish a stable relationship with consumers. By providing high-quality products and services to consumers, enterprises can win the long-term trust and support of consumers and further enhance brand loyalty [30], so as to consolidate their market position and improve their financial performance. Thirdly, the efficient and orderly internal governance process of enterprise helps to reduce the friction between departments, improve the ability of coordination and cooperation between departments [31], reduce the internal control cost in the production activities of the enterprise, and then enable the enterprise to provide products to the market at a lower price, occupy a greater cost advantage in the market competition, so as to improve its financial performance.

Hypothesis 3. 

ESG performance can significantly improve financial performance through improving social reputation and increasing market power.

2.4. The Mediating Effect of Development Side

ESG can improve the financial performance of enterprises by promoting technological innovation. Good ESG performance can effectively convey to the capital market the positive practice of enterprises in energy conservation and emission reduction policies and social responsibility, which helps enterprises obtain trust and support from banks, institutional investors, and individual investors so as to help enterprises invest sufficient funds in R&D activities and promote their own technological innovation [12]. Enterprises can obtain diversified and abundant knowledge resources about innovation and overcome information asymmetry [32]. Good ESG performance also usually means that enterprises can provide employees with higher salaries, benefits, and better career development opportunities, which is conducive to attracting a large number of labor forces to gather in enterprises, especially highly skilled labor forces. This will provide sufficient reserves of technical talents for enterprise technological innovation [33], maintain competitive advantage, and improve enterprise value through technological innovation. In addition, a good internal management structure can effectively reduce the shortsightedness of enterprise management, prevent enterprises from focusing only on short-term profits due to the high cost of innovation investment, improve the management’s innovative thinking orientation, and encourage enterprises to actively innovate [34], so as to win market competitiveness and improve their financial performance with advanced production technology and high value-added products.

Hypothesis 4. 

ESG performance can significantly improve financial performance through promoting enterprise innovation.

3. Materials and Methods

3.1. Sample and Data

This research conducts a series of empirical analyses to test the above hypothesis based on the theoretical analysis. This paper selected enterprises continuously involved in ESG performance from 2011 to 2022 as the sample. In order to make the sample more standardized and the study’s results closer to authenticity, the data are selected and processed based on the following three principles: Firstly, samples of ST companies and ST* companies are eliminated. Secondly, the financial and insurance enterprises are removed. Thirdly, samples with incomplete data are excluded, which means if a listed company did not receive an ESG rating in the year or did not disclose indicators such as corporate assets, it will be excluded from this sample [35]. Furthermore, this paper tends to minimize the impact of outliers on data analysis by performing winsorization on all variables at 1% and 99% levels. Ultimately, 3268 samples of companies continuously involved in ESG performance were selected with a total of 18,438 firm-year observed values.

Meanwhile, the data on ESG performance are sourced from the Huazheng index, the financial data of the listed companies are retrieved from the CSMAR database, and the data on patents are obtained from the CNRDS database. Additionally, this paper calculates financing costs according to the Ohlson–Juettner model and uses Stata16.0 to run regression and analyze data.

3.2. Variable Design and Measurement

3.2.1. Dependent Variable

The dependent variable of this essay is the financial performance of each enterprise. Referring to the treatment of Peng et al. [36], this paper adopts return on asset (ROA) to measure the enterprise’s financial performance, which can be calculated from the ratio of net income to total assets.

3.2.2. Independent Variable

The independent variable of this essay is the ESG score of the companies. Referring to the variable setting model of Song et al. [37], this paper selects the relevant data of companies’ ESG scores from the Hua Zheng database. This paper adopted the Hua Zheng ESG index and assigned points (from 9 to 1) to the enterprise’s financial performance rating from AAA to C.

3.2.3. Control Variable

Considering the impact of non-ESG factors on firms’ financial performance, this paper introduces the following control variables: company size, cash flow, company growth, company age, asset–liability ratio, the proportion of independent directors, and ownership concentration.

3.2.4. Mediating Variable

The first mediating variable of this essay is the agency cost (Ag), which is measured by the proportion of enterprise management fee to operating revenue. The second mediating variable is the financing cost (Fc). In addition, referring to Sheng’s research [38], this paper adopted the internal rate of return (IRR), constructed the Ohlson–Juettner model to measure equity financing cost, and defined debt financing cost as financial expenses/total amount of debt. Through the Ohlson–Juettner model, this paper sets the price for each share:

$$P_0={\displaystyle \frac{{EPS}_1}{r_e}}+{\displaystyle \frac{{EPS}_1+{EPS}_2-r_e\left({EPS}_1-{dps}_1\right)}{r_e\left(r_e-g\right)}}$$

(1)

and this paper set:

$$\mathrm{A}={\displaystyle \frac{1}{2}}\left[\left(r-1\right)+{\displaystyle \frac{{dps}_1}{p_0}}\right]$$

(2)

and equity financing cost:

$$r_e=A+\sqrt{A^2+{\displaystyle \frac{{EPS}_1}{P_0}}\times \left[{\displaystyle \frac{{EPS}_2-{EPS}_1}{{EPS}_1}}-\left(r-1\right)\right]}$$

(3)

where r_e is the cost of equity financing; EPS₁ is the expected earnings per share for the next year; EPS₂ is the expected earnings per share for the second year; dps₁ is the expected dividends per share for the next year; p₀ is the share price per share; and r − 1 is the dividend growth rate.

The third mediating variable of this essay is social reputation. This paper refers to Guan and Zhang (2019) [39] to measure corporate reputation (REP) by building a reputation evaluation system. First, on the basis of existing research, considering the evaluation of corporate reputation by various stakeholders and adhering to the principles of operability, hierarchy, effectiveness, and relative completeness, 12 corporate reputation evaluation indicators are selected. Then, the enterprise reputation score is calculated by the factor analysis method for 12 indicators. Finally, according to the corporate reputation score from low to high, it is divided into ten groups, and each group is assigned REP from 1 to 10 in turn. The fourth mediating variable of this essay is market power. Market power (Power) is measured by the proportion of enterprise operating revenue to industry operating revenue. The last mediating variable of this essay is innovation. Enterprise innovation (Rd) is measured by the natural logarithm of the number of patents/1000. More detailed variable definitions are provided in

Table 1. Definitions of variables.

	Variable Name	Symbol	Definition
Explained variable	Return on total assets	ROA	Net profit as a percentage of average total assets for the year
Explanatory variable	ESG score	ESG	ESG score according to rated categories: Highest score is 9. Minimum score is 1
Control variable	Enterprise scale	Size	The natural pair value of total assets at the end of the yer
	Asset–liability ratio	Lev	Total liabilities as a percentage of total assets
	Cash flow	Cashflow	Cash flow generated by operation as a percentage of total assets for the year
	Enterprise growth	Growth	Percentage increase in revenue
	Ownership concentration	Top1	The shareholding ratio of the largest shareholder
	The proportion of independent directors	Board	Number of independent directors as a percentage of the total number of board members
	Year	Firmage	Ln (Year of the year—year of establishment +1)
Mediating variable	Agency cost	Ag	The proportion of enterprise management fee to operating revenue
	Financing cost	Fc	Equity financing cost + debt financing cost
	Social reputation	REP	Referring to Guan and Zhang (2019) [39]
	Market power	Power	The proportion of enterprise operating revenue to industry operating revenue
	Enterprise innovation	Rd	the number of patents/1000

.

3.3. Model Specifications

Firstly, this paper constructs a regression model to verify the first hypothesis, which is mainly used to verify the correlation between ESG performance and financial performance.

$${ROA}_{it}={\alpha }_0+{\alpha }_1{ESG}_{it}+{\alpha }_2{Control}_{it}+INDFE+YearFE+{\epsilon }_{it}$$

(4)

Here, $ROA$ represents the financial performance, and ‘it’ represents the T-year of the I-enterprise. ${ESG}_{it}$ represents the ESG score of I-enterprise in the year of T. Control represents control variables, namely enterprise scale, asset–liability ratio, cash flow, enterprise growth, ownership concentration, the proportion of independent directors and years. $INDFE$ and $YearFE$ represent fixed effects of industry and year, respectively. $\epsilon$ represents random interference. If ${\alpha }_1$ is significant, then ESG performance is associated with financial performance. On this basis, it can be further investigated whether agency cost, financing cost, social reputation, market power, and enterprise innovation mediate the effect or not. Otherwise, the correlation coefficient of ESG performance and financial performance has no remarkable relation and does not have the prerequisites of the intermediary effect.

Secondly, this paper tends to verify the second hypothesis about the mediate effect by constructing a regression model.

$$M_{it}={\beta }_0+{\beta }_1{ESG}_{it}+{\beta }_2{Control}_{it}+INDFE+YearFE+{\epsilon }_{it}$$

(5)

$${ROA}_{it}={\lambda }_0+{\lambda }_1{ESG}_{it}+{\lambda }_2{M}_{it}+{\lambda }_3{Control}_{it}+INDFE+YearFE+{\epsilon }_{it}$$

(6)

$${ROA}_{it}={\gamma }_0+{\gamma }_1{ESG}_{it}+{\gamma }_2{M}_{it}+{\gamma }_3{ESG}_{it}\times M_{it}+{\gamma }_4{Control}_{it}+INDFE+YearFE+{\epsilon }_{it}$$

(7)

where $M_{it}$ represents five mediating variables. Take financing cost as an example; if ${\beta }_1$ is significant, then ESG performance is associated with financing cost. On this basis, whether ${\lambda }_2$ is significant can be further tested later. Otherwise, the correlation coefficient of ESG performance and financing cost does not have a remarkable relation and does not have the prerequisites of the mediating effect at the same time. That means that financing cost does not play an intermediary role in corporate ESG on financial success, and there is no need to verify further. However, if ${\beta }_1$ and ${\lambda }_2$ are both significant, then ESG performance is associated with financing cost. On this basis, the existence verification of the intermediary effect of financing costs can be further carried out later. In addition, it could be found that the impact of ESG performance on financial performance is completely realized through the intermediary effect of financing cost by testing the significance of ${\lambda }_1$. That means that if ${\beta }_1$ and ${\lambda }_2$ are both significant but ${\lambda }_1$ is not significant, thus the mediating variable financing cost plays a complete mediating role between ESG performance and financial performance. If ${\beta }_1$, ${\lambda }_2$, and ${\lambda }_1$ are all significant, thus financing costs play a partial mediating role between ESG performance and financial performance. In addition, this paper uses Equation (7) to further test that the financing cost is indeed a mechanism of action for ESG to improve firms’ financial performance, and if γ₃ is significantly positive, then it indicates that ESG can improve firms’ financial performance by reducing financing cost.

4. Results

4.1. Descriptive Statistics

The results of the descriptive analysis of the variables are shown in

Table 2. Descriptive statistics.

Variable	Obs	Mean	SD	Max	p50	Min
ROA	18438	0.0594	0.0532	0.2473	0.0520	−0.3730
ESG	18438	4.3133	1.0324	8.0000	4.0000	1.0000
Size	18438	22.5206	1.3238	26.4523	22.3192	19.6286
Lev	18438	0.4155	0.1958	0.9079	0.4104	0.0319
Cashflow	18438	0.0564	0.0676	0.2669	0.0546	−0.1994
Growth	18438	0.2139	0.3907	4.0242	0.1426	−0.6576
Top1	18438	0.3544	0.1486	0.7578	0.3361	0.0802
Board	18438	2.1346	0.1961	2.7081	2.1972	1.6094
Firmage	18438	2.8736	0.3435	3.6109	2.9444	1.3863
Ag	18438	0.0811	0.0623	0.0068	0.0664	0.6414
Fc	18438	0.0794	0.0309	0.2884	0.0760	−0.0161
REP	16129	5.4971	2.8725	1.0000	5.0000	10.0000
Power	18438	0.0454	0.1141	0.0000	0.0091	1.0000
Rd	11946	0.0026	0.0015	0.0000	0.0026	0.0093

. In addition, the maximum value of financial performance (ROA) is 0.2473, the minimum value is −0.3730, and the mean value is 0.0594. These figures indicate that there is a significant difference in the financial performance of selected listed companies, and some firms are making losses. Therefore, it can be found that the overall financial performance of selected enterprises may not be ideal. Additionally, the maximum value of enterprise ESG performance is 8, the minimum value is 1, and the average value is 4.3133, which indicates that enterprise ESG performance is good, but there are some slight variations among ESG performance of selected enterprises.

4.2. Multicollinearity Test

Before conducting regression, this paper uses the Pearson correlation coefficient to investigate the correlation between selected variables and whether serious multicollinearity exists.

As shown in

Table 3. Correlation analysis.

Variable	ROA	ESG	Size	Lev	Cashflow	Growth	Top1	Board	Firmage
ROA	1.000
ESG	0.100 ***	1.000
Size	−0.142 ***	0.196 ***	1.000
Lev	−0.394 ***	−0.003	0.590 ***	1.000
Cashflow	0.463 ***	0.056 ***	0.026 ***	−0.191 ***	1.000
Growth	0.215 ***	−0.056 ***	−0.007	0.048 ***	−0.012 *	1.000
Top1	0.051 ***	0.059 ***	0.214 ***	0.108 ***	0.085 ***	−0.044 ***	1.000
Board	−0.048 ***	0.021 ***	0.236 ***	0.143 ***	0.030 ***	−0.048 ***	0.028 ***	1.000
Firmage	−0.048 ***	0.044 ***	0.266 ***	0.186 ***	0.078 ***	−0.049 ***	−0.076 ***	0.055 ***	1.000

Note: * Significance at the 10% level. *** Significance at the 1% level.

, the correlation coefficient between ESG performance and financial performance is 0.100, and it is significant at the level of 1%, indicating a significant positive correlation between ESG performance and financial performance. Also, the result preliminarily proved that improving ESG performance can promote financial health. Furthermore, it noted that the correlation coefficients of the selected control variables and financial performance are all significant at the level of 1%, indicating that the selected control variables are all significantly correlated with financial performance and can serve to control the characteristics of the enterprises. Meanwhile, the correlation coefficient among variables is below 0.8, indicating that there is no significant multicollinearity in the above-selected variables, and the variables are relatively reasonable.

Moreover, this model was also assessed further for multicollinearity by using the variance inflation factor (VIF), and the results obtained from the test are shown in

Table 4. Results of the multicollinearity test.

Variable	VIF	1/VIF
Size	1.86	0.538250
Lev	1.69	0.593406
Firmage	1.11	0.897402
Cashflow	1.09	0.918211
Top1	1.08	0.923476
ESG1	1.07	0.937108
Board	1.06	0.9339457
Growth	1.01	0.985656
Mean VIF	1.25

. It can be seen that the mean value of VIF is 1.25, and each value of VIF is significantly less than 5, so it shows that there is no severe problem of multicollinearity.

4.3. Regression Analysis of Involving in ESG Performance and Financial Performance

This part constructs a fixed-effect model and uses stepwise regression to examine the relationship between ESG performance and financial performance, and the comparison between without and with control variables can be seen in

Table 5. Regression results of main effect variables.

	(1)	(2)
Variables	ROA	ROA
ESG	0.00568 ***	0.00341 ***
	(14.2984)	(10.3815)
Size		0.00288 ***
		(8.0138)
Lev		−0.09954 ***
		(−40.3757)
Cashflow		0.31037 ***
		(48.3017)
Growth		0.03232 ***
		(26.1564)
Top1		0.02313 ***
		(10.0503)
Board		−0.00109
		(−0.6360)
Firmage		−0.00047
		(−0.4582)
Constant	0.03489 ***	−0.00774
	(19.6163)	(−0.9654)
IND	YES	YES
YEAR	YES	YES
Obs	18438	18438
R-squared	0.108	0.417

Note: *** Significance at the 1% level.

. Before adding control variables in column (1), the estimated coefficient of ESG performance on financial performance is 0.00568 and significant at the 1% level; after the control variable is added in column (2), the estimated coefficient of corporate ESG performance on financial performance is 0.00341 and significant at the 1% level. The estimated coefficients of ESG are found to be attenuated, indicating that the inclusion of control variables serves to effectively mitigate the interference of other observed factors. Taking column (2) as an example, every improvement in ESG performance by one standard deviation will lead to a 0.06-fold increase in financial performance (0.00341 × 1.0324/0.0594). The regression result shows that involvement in ESG performance has a significant positive effect on enterprise performance, which effectively proves hypothesis 1. To sum up, the better ESG performance is, the greater the positive impact on the enterprise’s financial performance.

4.4. Mechanism Analysis

4.4.1. Agency Cost and Financing Cost

This part examines how ESG performance improves financial performance based on agency costs and financing costs. As shown in

Table 6. Mediating effects of agency cost and financing cost.

	(1)	(2)	(3)	(4)	(5)	(6)	(7)
Variables	ROA	Agency Cost	ROA	ROA	Financing Cost	ROA	ROA
ESG	0.00341 ***	−0.00186 ***	0.00314 ***	0.00234 ***	−0.00037 **	0.00339 ***	−0.00055
	(10.3815)	(−4.8653)	(9.7133)	(4.2993)	(−2.0686)	(10.3446)	(−0.6530)
Ag			−0.14393 ***	−0.18458 ***
			(−19.0318)	(−7.1097)
ESG × Ag				0.01017 *
				(1.7192)
Fc						−0.03843 **	−0.25316 ***
						(−2.5564)	(−5.2563)
ESG × Fc							0.05019 ***
							(4.8016)
Size	0.00288 ***	−0.00755 ***	0.00179 ***	0.00184 ***	−0.00104 ***	0.00284 ***	0.00306 ***
	(8.0138)	(−19.9253)	(5.0452)	(5.1601)	(−5.4541)	(7.9118)	(8.5051)
Lev	−0.09954 ***	−0.05620 ***	−0.10763 ***	−0.10763 ***	−0.08769 ***	−0.10291 ***	−0.10336 ***
	(−40.3757)	(−19.7440)	(−43.6526)	(−43.6740)	(−68.6157)	(−36.4118)	(−36.5611)
Cashflow	0.31037 ***	−0.05314 ***	0.30272 ***	0.30275 ***	−0.00230	0.31028 ***	0.30996 ***
	(48.3017)	(−8.2203)	(47.7803)	(47.7858)	(−0.8051)	(48.3283)	(48.3015)
Growth	0.03232 ***	−0.01277 ***	0.03048 ***	0.03049 ***	−0.00055	0.03230 ***	0.03223 ***
	(26.1564)	(−11.3262)	(25.4073)	(25.4093)	(−1.1655)	(26.1741)	(26.0815)
Top1	0.02313 ***	−0.01933 ***	0.02035 ***	0.02039 ***	−0.01448 ***	0.02258 ***	0.02246 ***
	(10.0503)	(−7.5619)	(8.9513)	(8.9674)	(−11.6001)	(9.8397)	(9.7899)
Board	−0.00109	−0.00259	−0.00146	−0.00149	−0.00284 ***	−0.00119	−0.00133
	(−0.6360)	(−1.3034)	(−0.8642)	(−0.8860)	(−3.0261)	(−0.7000)	(−0.7819)
Firmage	−0.00047	−0.00275 **	−0.00087	−0.00088	−0.00277 ***	−0.00058	−0.00078
	(−0.4582)	(−2.0955)	(−0.8642)	(−0.8759)	(−4.7402)	(−0.5611)	(−0.7575)
Constant	−0.00774	0.30838 ***	0.03664 ***	0.03897 ***	0.16030 ***	−0.00158	0.01140
	(−0.9654)	(35.9566)	(4.5272)	(4.7383)	(37.4094)	(−0.1917)	(1.3138)
IND	YES	YES	YES	YES	YES	YES	YES
YEAR	YES	YES	YES	YES	YES	YES	YES
Observations	18438	18438	18438	18438	18438	18438	18438
R-squared	0.417	0.397	0.434	0.434	0.448	0.417	0.418

Note: * Significance at the 10% level. ** Significance at the 5% level. *** Significance at the 1% level.

below, column (1) is the result of the benchmark regression of this paper, and it is significant at the 1% level. Thus, it shows that corporate involvement in ESG could have a significant impact on finance performance. Moreover, the estimated coefficient of ESG performance on agency cost in column (2) is −0.00186, and it is significant at the 1% level, indicating a negative correlation between ESG performance and agency cost. Moreover, the estimated coefficients of ESG performance on financial performance through agency costs in column (3) are significant. However, the estimated coefficient of corporate ESG performance is 0.00314 and smaller than 0.00341 in the benchmark regression, which suggests that agency cost plays a partial mediating effect in the process of ESG performance improvement of financial performance. In addition, the multiplier coefficient of ESG and agency cost in column (4) is significantly positive, indicating that ESG has a stronger role in promoting financial performance in enterprises with higher agency costs, which further verifies the establishment of agency cost mechanisms.

As shown in column (5), the estimated coefficient of ESG performance on financing cost is −0.00037, and it is significant at the 5% level, indicating a negative correlation between ESG performance and financing cost. Moreover, the estimated coefficients of ESG performance on financial performance through financing costs in column (6) are significant. However, the estimated coefficient of corporate ESG performance is 0.00339 and smaller than 0.00341 in the benchmark regression, which suggests that financing costs play a partial mediating effect in the process of ESG performance improvement of financial performance. In addition, the multiplier coefficient of ESG and financing cost in column (7) is significantly positive, indicating that ESG has a stronger role in promoting financial performance in enterprises with higher financing costs, which further verifies the establishment of financing cost mechanisms. Therefore, hypothesis 2 is effectively assessed, and agency cost and financing cost should be considered an incomplete intermediary mechanism of ESG performance on financial performance.

4.4.2. Social Reputation and Market Power

This part examines how ESG performance improves financial performance based on social reputation and market power. As shown in

Table 7. Mediating effects of social reputation and market power.

	(1)	(2)	(3)	(4)	(5)	(6)	(7)
Variables	ROA	Social Reputation	ROA	ROA	Market Power	ROA	ROA
ESG	0.00341 ***	0.14571 ***	0.00113 ***	0.00223 ***	0.00227 ***	0.00339 ***	0.00393 ***
	(10.3815)	(10.5175)	(4.0745)	(3.2288)	(3.8081)	(10.3298)	(11.0017)
REP			0.01403 ***	0.01485 ***
			(68.7224)	(29.5302)
ESG×REP				−0.00020 *
				(−1.8466)
Power						0.00722 *	0.05907 ***
						(1.8203)	(5.1099)
ESG×Power							−0.01177 ***
							(−4.8401)
Size	0.00288 ***	2.11144 ***	−0.02630 ***	−0.02617 ***	0.03537 ***	0.00262 ***	0.00277 ***
	(8.0138)	(138.6442)	(−46.1058)	(−45.7745)	(39.8859)	(6.7164)	(7.0485)
Lev	−0.09954 ***	−0.76054 ***	−0.08884 ***	−0.08897 ***	−0.00248	−0.09952 ***	−0.09980 ***
	(−40.3757)	(−7.7029)	(−42.5153)	(−42.4902)	(−0.6193)	(−40.3618)	(−40.4564)
Cashflow	0.31037 ***	8.36825 ***	0.17213 ***	0.17243 ***	0.04184 ***	0.31007 ***	0.30998 ***
	(48.3017)	(36.7307)	(30.4099)	(30.5105)	(4.2747)	(48.2341)	(48.2122)
Growth	0.03232 ***	0.67364 ***	0.02224 ***	0.02220 ***	0.00063	0.03231 ***	0.03225 ***
	(26.1564)	(16.6667)	(22.5680)	(22.5456)	(0.4131)	(26.1594)	(26.1585)
Top1	0.02313 ***	1.09569 ***	0.00598 ***	0.00594 ***	0.03564 ***	0.02288 ***	0.02282 ***
	(10.0503)	(11.5910)	(3.0865)	(3.0664)	(8.6320)	(9.9390)	(9.9165)
Board	−0.00109	−1.47611 ***	0.02067 ***	0.02065 ***	−0.00103	−0.00108	−0.00119
	(−0.6360)	(−20.1601)	(13.2071)	(13.2027)	(−0.3006)	(−0.6320)	(−0.6990)
Firmage	−0.00047	0.14418 ***	−0.00160 *	−0.00161 *	−0.01240 ***	−0.00038	−0.00046
	(−0.4582)	(3.1759)	(−1.8424)	(−1.8547)	(−5.7208)	(−0.3710)	(−0.4481)
Constant	−0.00774	−40.79535 ***	0.55198 ***	0.54464 ***	−0.73713 ***	−0.00242	−0.00727
	(−0.9654)	(−122.1206)	(48.6746)	(45.8752)	(−39.4498)	(−0.2814)	(−0.8381)
IND	YES	YES	YES	YES	YES	YES	YES
YEAR	YES	YES	YES	YES	YES	YES	YES
Observations	18438	16128	16128	16128	18438	18438	18438
R-squared	0.417	0.686	0.602	0.602	0.545	0.417	0.417

Note: * Significance at the 10% level. *** Significance at the 1% level.

below, column (1) is the result of the benchmark regression of this paper, and it is significant at the 1% level. Thus, it shows that corporate involvement in ESG could have a significant impact on finance performance. Moreover, the estimated coefficient of ESG performance on social reputation in column (2) is 0.14571, and it is significant at the 1% level, indicating a positive correlation between ESG performance and social reputation. Moreover, the estimated coefficients of ESG performance on financial performance through social reputation in column (3) are significant. However, the estimated coefficient of corporate ESG performance is 0.00113 and smaller than 0.00341 in the benchmark regression, which suggests that social reputation plays a partial mediating effect in the process of ESG performance improvement of financial performance. In addition, the multiplier coefficient of ESG and social reputation in column (4) is significantly negative, indicating that ESG has a stronger role in promoting financial performance in enterprises with lower social reputation, which further verifies the establishment of social reputation mechanisms.

As shown in column (5), the estimated coefficient of ESG performance on market power is 0.00227, and it is significant at the 1% level, indicating a positive correlation between ESG performance and market power. Moreover, the estimated coefficients of ESG performance on financial performance through market power in column (6) are significant. However, the estimated coefficient of corporate ESG performance is 0.00339 and smaller than 0.00341 in the benchmark regression, which suggests that market power plays a partial mediating effect in the process of ESG performance improvement of financial performance. In addition, the multiplier coefficient of ESG and market power in column (7) is significantly negative, indicating that ESG has a stronger role in promoting financial performance in enterprises with lower market power, which further verifies the establishment of market power mechanisms. Therefore, hypothesis 3 is effectively assessed, and social reputation and market power should be considered an incomplete intermediary mechanism of ESG performance on financial performance.

4.4.3. Enterprise Innovation

The realization path of ESG performance promoting enterprise financial performance is further tested based on enterprise innovation. As shown in

Table 8. Mediating effects of enterprise innovation.

	(1)	(2)	(3)	(4)
Variables	ROA	Rd	ROA	ROA
ESG	0.00341 ***	0.00014 ***	0.00352 ***	0.00489 ***
	(10.3815)	(10.8624)	(9.0395)	(6.2851)
Rd			1.73041 ***	4.00959 ***
			(6.3408)	(3.5020)
ESG × RD				−0.50942 **
				(−2.0755)
Size	0.00288 ***	0.00039 ***	0.00289 ***	0.00295 ***
	(8.0138)	(24.3875)	(6.2985)	(6.4307)
Lev	−0.09954 ***	0.00011	−0.10678 ***	−0.10674 ***
	(−40.3757)	(1.2854)	(−36.6384)	(−36.6188)
Cashflow	0.31037 ***	0.00137 ***	0.33994 ***	0.34039 ***
	(48.3017)	(6.9551)	(43.7850)	(43.8580)
Growth	0.03232 ***	−0.00014 ***	0.03961 ***	0.03960 ***
	(26.1564)	(−4.2565)	(22.7513)	(22.7913)
Top1	0.02313 ***	0.00036 ***	0.02503 ***	0.02484 ***
	(10.0503)	(3.9846)	(8.8412)	(8.7639)
Board	−0.00109	0.00033 ***	0.00073	0.00070
	(−0.6360)	(4.7052)	(0.3594)	(0.3437)
Firmage	−0.00047	0.00000	−0.00198	−0.00195
	(−0.4582)	(0.0363)	(−1.5508)	(−1.5340)
Constant	−0.00774	−0.00757 ***	−0.01238	−0.01960 *
	(−0.9654)	(−22.1455)	(−1.2364)	(−1.8651)
IND	YES	YES	YES	YES
YEAR	YES	YES	YES	YES
Observations	18438	11940	11940	11940
R-squared	0.417	0.283	0.461	0.461

Note: * Significance at the 10% level. ** Significance at the 5% level. *** Significance at the 1% level.

below, column (1) still shows the benchmark regression results of this paper and proves hypothesis 1. As for column (2), the estimated coefficient of ESG performance on innovation is 0.0.00014, which is significant at the 1% level. The results indicate that ESG performance can help promote enterprise innovation. In column (3), the estimated coefficients of enterprise ESG performance on financial performance through enterprise innovation are significant. This means that enterprise innovation partially mediates the process of ESG performance and financial performance. In addition, the multiplier coefficient of ESG and innovation in column (4) is significantly negative, indicating that ESG has a stronger role in promoting financial performance in enterprises with lower innovation, which further verifies the establishment of innovation mechanisms. Therefore, hypothesis 4 is proved, and corporate ESG performance can improve financial performance by promoting corporate innovation.

5. Discussion

5.1. Single Aspect of ESG Performance and Financial Performance

This paper further conducts a multi-dimensional regression based on the environmental responsibility rating, social responsibility rating, and corporate governance rating of ESG performance to test the impact of each sub-dimension of ESG performance on corporate financial performance. As shown in

Table 9. Regression results of the single aspect of ESG on financial performance.

	(1)	(2)	(3)
Variables	ROA	ROA	ROA
E	−0.00002
	(−0.0676)
S		0.00351 ***
		(10.9004)
G			0.00292 ***
			(10.1512)
Size	0.00379 ***	0.00286 ***	0.00311 ***
	(10.5123)	(7.9872)	(8.7200)
Lev	−0.10324 ***	−0.09965 ***	−0.09672 ***
	(−42.1071)	(−40.5156)	(−38.4801)
Cashflow	0.31220 ***	0.31027 ***	0.30983 ***
	(48.5892)	(48.3135)	(48.3027)
Growth	0.03193 ***	0.03230 ***	0.03239 ***
	(25.6529)	(26.1635)	(26.0811)
TOP1	0.02359 ***	0.02317 ***	0.02170 ***
	(10.2273)	(10.0720)	(9.4709)
Board	−0.00135	−0.00114	−0.00013
	(−0.7909)	(−0.6707)	(−0.0768)
Firmage	−0.00065	−0.00047	−0.00062
	(−0.6294)	(−0.4528)	(−0.6056)
Constant	−0.01101	−0.00775	−0.01665 **
	(−1.3461)	(−0.9675)	(−2.0683)
IND	YES	YES	YES
YEAR	YES	YES	YES
Observations	18438	18438	18438
R-squared	0.413	0.417	0.417

Note: ** Significance at the 5% level. *** Significance at the 1% level.

below, the estimated coefficient of corporate environmental rating on financial performance in column (1) is −0.00002 but insignificant. In column (2), the estimated coefficient of CSR on financial performance is 0.00351 and significant at the level of 1%. The estimated coefficient of corporate governance on financial performance in column (3) is 0.00292 and significant at the level of 1%. The above estimation results show that improving corporate ESG performance in different dimensions of social responsibility and corporate governance can promote the improvement of corporate financial performance, while the different dimensions of environmental responsibility have no significant impact on financial performance.

5.2. Robustness Tests

First, considering that variable measurement errors may lead to biased estimation results, this paper uses return on equity (ROE), gross profit margin on sales (GPM), and earnings per share (EPS) to replace the explained variables for robustness testing, respectively. As shown in columns (1)–(3) of

Table 10. Variable changes and model changes.

	(1)	(2)	(3)	(4)	(5)	(6)	(7)	(8)
Variables	ROE	GPM	EPS	ROA	ROA	ROA	ROA	ROA
ESG	0.00471 ***	0.00375 **	0.04717 ***	0.00015 **	0.00339 ***	0.00180 ***	0.00341 ***	0.00341 ***
	(7.6174)	(2.3064)	(5.9151)	(2.0843)	(10.1487)	(4.4952)	(7.8928)	(5.9030)
Size	0.00706 ***	0.01195 ***	0.21184 ***	−0.00399 ***	0.00321 ***	0.00543 ***	0.00288 ***	0.00288 ***
	(9.5108)	(5.9348)	(10.2456)	(−2.6869)	(8.6727)	(4.8458)	(5.1096)	(4.9081)
Lev	−0.02992 ***	−0.30833 ***	−0.97021 ***	−0.11780 ***	−0.10045 ***	−0.10559 ***	−0.09954 ***	−0.09954 ***
	(−5.4745)	(−27.7459)	(−11.1669)	(−12.2924)	(−40.1327)	(−22.3026)	(−28.9669)	(−29.1206)
Cashflow	0.50670 ***	0.42088 ***	3.36605 ***	0.37770 ***	0.30300 ***	0.19924 ***	0.31037 ***	0.31037 ***
	(43.2022)	(5.9913)	(18.9984)	(15.3736)	(47.0210)	(26.6692)	(33.6347)	(13.0612)
Growth	0.05957 ***	0.03689 ***	0.42110 ***	0.03874 ***	0.03078 ***	0.02990 ***	0.03232 ***	0.03232 ***
	(25.5247)	(5.1272)	(13.8298)	(6.5481)	(25.3465)	(25.8237)	(24.2666)	(10.9520)
Top1	0.04230 ***	0.01825	0.17612 ***	0.01737 **	0.02576 ***	0.03193 ***	0.02313 ***	0.02313 ***
	(9.8287)	(1.3659)	(4.0049)	(2.3258)	(11.2382)	(5.2543)	(6.5015)	(5.3882)
Board	−0.00180	−0.01517 ***	−0.21321 ***	−0.00647	−0.00114	0.00455	−0.00109	−0.00109
	(−0.5463)	(−2.8475)	(−3.4970)	(−1.3568)	(−0.6658)	(1.3888)	(−0.4288)	(−0.3486)
Firmage	0.00244	0.00201	−0.11793 ***	−0.00832 **	−0.00112	0.00548	−0.00047	−0.00047
	(1.3190)	(0.5346)	(−4.6945)	(−2.2300)	(−1.0548)	(1.1010)	(−0.2957)	(−0.3035)
Constant	−0.12588 ***	−0.06005	−3.51223 ***	0.21646 ***	−0.01277	−0.08117 ***	−0.00774	−0.00774
	(−7.6682)	(−1.1328)	(−10.8794)	(5.7793)	(−1.5601)	(−2.8483)	(−0.6089)	(−0.6018)
IND	YES	YES	YES	YES	NO	NO	YES	YES
YEAR	YES	YES	YES	YES	NO	YES	YES	YES
PRO × YEAR	NO	NO	NO	NO	YES	NO	NO	NO
IND × YEAR	NO	NO	NO	NO	YES	NO	NO	NO
FIRM	NO	NO	NO	NO	NO	YES	NO	NO
Observations	18438	18438	18437	1532	18363	18143	18438	18438
R-squared	0.273	0.121	0.191	0.653	0.472	0.658	0.417	0.417

Note: ** Significance at the 5% level. *** Significance at the 1% level.

below, the estimated coefficients of ESG performance on financial performance are significantly positive, indicating that after replacing the explained variables, ESG performance still positively impacts improving financial performance, strengthening the robustness of the previous conclusions. Additionally, this paper uses ESG from MSCI to replace the explanatory variable for robustness testing. As shown in column (4) of Table 10 below, the estimated coefficients of ESG performance on financial performance are significantly positive, indicating that after replacing the explanatory variables, ESG performance still positively impacts improving financial performance, strengthening the robustness of the previous conclusions.

Secondly, considering that there are large differences in regional characteristics such as economic development status and traffic [40,41], and that it is difficult to control all regional characteristics by adding a few variables, thus this paper adds fixed effects to control regional characteristics that change over time. As shown in column (5) of Table 10, the estimated coefficient of corporate ESG performance on financial performance is significantly positive, which indicates that after controlling for regional characteristics, ESG performance still has a significant effect on improving financial performance and strengthens the robustness of the previous conclusion. The paper also adds fixed effects to control firm characteristics that do not change over time. As shown in column (6) of Table 10, the estimated coefficient of corporate ESG performance on financial performance is significantly positive, which indicates that after controlling for firm characteristics, ESG performance still has a significant effect on improving financial performance and strengthens the robustness of the previous conclusion.

Thirdly, in this paper, the robust standard error in the benchmark regression is replaced by the robust standard error of clustering at the enterprise and industry levels, respectively. The results are shown in columns (7)–(8) of Table 10. The estimated coefficients of corporate ESG performance on financial performance are significantly positive, which indicates that after changing the cluster method, ESG performance still has a significant effect on improving financial performance and strengthens the robustness of the previous conclusion.

Fourth, considering the impact of the epidemic on the production and operation activities of enterprises, which may lead to large fluctuations in financial performance, this paper excludes the year of the epidemic, and the sample in 2020 is excluded. As shown in column (1) of

Table 11. Excluded policy interference.

	(1)	(2)	(3)
Variables	ROA	ROA	ROA
ESG	0.00350 ***	0.00333 ***	0.00208 ***
	(10.0043)	(10.1780)	(4.0001)
Polu_18		0.01370 ***
		(10.9697)
Size	0.00292 ***	0.00291 ***	0.00312 ***
	(7.7537)	(8.1366)	(5.2062)
Lev	−0.09896 ***	−0.09912 ***	−0.11790 ***
	(−39.1860)	(−40.3483)	(−25.5764)
Cashflow	0.30390 ***	0.30925 ***	0.35779 ***
	(45.4490)	(48.3075)	(31.9079)
Growth	0.03095 ***	0.03173 ***	0.05414 ***
	(24.7719)	(25.7892)	(23.4557)
Top1	0.02270 ***	0.02290 ***	0.02529 ***
	(9.4187)	(9.9953)	(6.6705)
Board	−0.00078	−0.00121	−0.00493 *
	(−0.4406)	(−0.7089)	(−1.6793)
Firmage	0.00066	−0.00065	−0.00429 **
	(0.6261)	(−0.6329)	(−2.1095)
Constant	−0.01249	−0.00952	0.01300
	(−1.4916)	(−1.1899)	(0.9670)
IND	YES	YES	YES
YEAR	YES	YES	YES
Observations	16738	18438	6409
R-squared	0.412	0.420	0.508

Note: * Significance at the 10% level. ** Significance at the 5% level. *** Significance at the 1% level.

below, the estimated coefficient of ESG performance on financial performance is significantly positive, which indicates that after excluding the impact of the epidemic, the ESG performance of enterprises still has a significant effect on financial performance.

In order to eliminate the possible mixed impact of the promulgation of environmental protection law on the benchmark results, this paper adds the policy variable of environmental protection law (polu_18) to the benchmark regression model. Specifically, based on the Guidelines for Industry Classification of Listed Companies revised by the China Securities Regulatory Commission in 2012, the Management Directory of Industry Classification for Environmental Verification of Listed Companies formulated by the Ministry of Environmental Protection in 2008, and the Guidelines for Disclosure of Environmental Information of Listed Companies, the paper identifies thermal power, iron and steel, cement, electrolytic aluminum, coal, metallurgy, chemical industry, petrochemicals, building materials, papermaking, brewing, pharmaceuticals, fermentation, textile, tannery, and mining industries as heavy polluters. Secondly, the year 2018, when the environmental protection tax law was passed, is taken as the policy time. If the enterprise is a heavy polluter and the year is 2018 or later, then polu_18 is taken as 1, otherwise it is 0. The regression result is shown in column (2) of Table 11. The estimated coefficient of ESG performance on financial performance is still significantly positive, which indicates that after excluding the impact of the promulgation of environmental protection law, the ESG performance of enterprises still has a significant effect on financial performance.

Additionally, in September 2018, the CSRC revised the “Code of Governance for Listed Companies”, which clearly stipulates that enterprises need to disclose their environmental information and fulfill their social responsibility in accordance with the regulations, in which their own or important subsidiaries belonging to the key emission units need to be mandatorily disclosed, and the other listed companies implement the policy of “no disclosure, no explanation”. In order to exclude the impact of this policy on the benchmark results, this paper selects the sample after 2018 to re-regress; the results are shown in column (3) of Table 11. The estimated coefficient of ESG performance on financial performance is still significantly positive, which indicates that after excluding the impact of this policy, the ESG performance of enterprises still has a significant effect on financial performance.

5.3. Endogeneity Test

5.3.1. 2SLS Model

Since there may be a reverse causality between ESG performance and financial performance, which leads to endogeneity problems, this paper uses the mean ESG performance of other enterprises in the same province as the instrumental variable to solve the endogeneity problems. As shown in

Table 12. Endogeneity test.

	(1)	(2)	(3)
Variables	ESG	ROA	ROA
ESG		0.00932 **
		(2.2350)
IV	0.35591 ***
	(9.6555)
L.ESG			0.00144 ***
			(3.5487)
Size	0.26120 ***	0.00132	0.00428 ***
	(33.4971)	(1.1207)	(9.7950)
Lev	−1.06625 ***	−0.09312 ***	−0.10461 ***
	(−21.1982)	(−17.8107)	(−33.7329)
Cashflow	0.54955 ***	0.30728 ***	0.32974 ***
	(4.8280)	(45.0348)	(42.5048)
Growth	−0.10765 ***	0.03298 ***	0.03537 ***
	(−5.4217)	(25.2990)	(22.6618)
Top1	0.13429 **	0.02232 ***	0.02371 ***
	(2.5722)	(9.3136)	(8.5984)
Board	−0.07125 *	−0.00061	−0.00241
	(−1.8197)	(−0.3472)	(−1.1902)
Firmage	−0.04099 *	−0.00018	0.00107
	(−1.7266)	(−0.1695)	(0.8413)
Constant	−2.44684 ***	0.00558	−0.03287 ***
	(−10.4914)	(0.4768)	(−3.3247)
IND	YES	YES	YES
YEAR	YES	YES	YES
Observations	18438	18438	13323
R-squared	0.157	0.406	0.428
Kleibergen–Paaprk LM statistic	92.680 ***
Kleibergen–Paaprk Wald F statistic	93.230

Note: * Significance at the 10% level. ** Significance at the 5% level. *** Significance at the 1% level.

below, the value of the Kleibergen–Paap rk LM statistic is 92.680 and significant at the level of 1%. The value of the Kleibergen–Paap rk Wald F statistic is 93.230 and greater than 16.38, which indicates that the instrumental variable passes the underidentification test and the weak instrumental variable test. The estimated coefficient of the instrumental variable in column (1) is significantly positive, and the estimated coefficient of the ESG performance on the financial performance in column (2) is significantly positive, which indicates that after using the instrumental variable method to deal with the endogeneity problem, the ESG performance still helps to promote the improvement of the enterprise financial performance, which confirms the robustness of the previous conclusions.

5.3.2. Explanatory Variables with a One-Period Lag

This paper uses the core explanatory variables with a one-period lag for regression to further avoid the endogeneity problem caused by reverse causality. As shown in Table 12 below, the estimated coefficient of the one-period-lagged ESG performance on the financial performance of enterprises in column (3) is significantly positive, which indicates that the one-period-lagged ESG performance of enterprises can still promote the improvement of the financial performance of enterprises, which strengthens the robustness of the previous conclusions.

5.4. Heterogeneity Test Using Different Ownership Nature

First of all, considering that enterprises with different ownership natures have significant differences in financial performance and ESG performance, this part takes property rights as the moderating variable to verify their moderating effect on the relationship between ESG performance and financial performance. According to the property right, the sample is divided into state-owned and non-state-owned enterprises for heterogeneity analysis.

As shown in

Table 13. Results of the heterogeneity test using property rights.

	(1)	(2)
	State-Owned	Non-State-Owned
Variables	ROA	ROA
ESG	0.00147 ***	0.00457 ***
	(2.88222)	(11.02883)
Size	0.00427 ***	0.00409 ***
	(8.40378)	(7.77880)
Lev	−0.10468 ***	−0.09896 ***
	(−28.86119)	(−29.13747)
Cashflow	0.25844 ***	0.32985 ***
	(25.24481)	(40.85975)
Growth	0.02623 ***	0.03416 ***
	(13.79498)	(22.09220)
Top1	0.00894 ***	0.03589 ***
	(2.59483)	(11.95007)
Board	−0.00199	0.00228
	(−0.84189)	(0.98609)
Firmage	0.00439 **	−0.00133
	(2.38990)	(−1.00876)
Constant	−0.03673 ***	−0.04742 ***
	(−2.82654)	(−4.06374)
IND	YES	YES
YEAR	YES	YES
Observations	6388	12044
R-squared	0.452	0.406
p-Value	0.000 ***

Note: ** Significance at the 5% level. *** Significance at the 1% level.

below, the estimated coefficient of ESG performance on financial performance in column (1) is 0.00147 and significant at 1%, and the estimated coefficient of ESG performance on financial performance in column (2) is 0.00457 and is significant at 1% level. The above estimation results show that ESG performance can improve financial performance. However, it noted that the estimated coefficient of ESG performance in column (1) is much smaller than that in column (2) by comparing the correlation coefficients, and the p-value of the inter-group coefficient difference test by using the Fischer combination test (sampling 1500 times) is 0.000. This indicates that there is an intergroup coefficient difference at the level of 1% and the improvement of ESG performance of non-state-owned enterprises has a stronger role in promoting financial performance than state-owned enterprises.

In other words, the ESG performance of non-state-owned enterprises has more significant effects on the financial performance. On one hand, China’s national situation determines that non-state-owned enterprises face greater market competition pressure, so their income volatility is more vulnerable to systemic risks and negative events, so the positive effect of increasing ESG performance on financial performance is more obvious. In addition, according to the principal-agent theory and stakeholder theory, non-state-owned enterprises are more likely to have interest conflicts between large and small shareholders, so the establishment of a good corporate governance structure will have a positive impact on enterprises’ healthy and long-term financial health.

5.5. Heterogeneity Test Using Different Industry Category

In addition, not only does the nature of property rights affect the relationship between ESG performance and financial performance, but the type of industry the enterprise belongs to also affects it. Compared with non-high-tech enterprises, high-tech industries have better ESG-related investment performance due to the leading application of high-tech, which makes the financial performance fluctuate to some extent. Therefore, referring to the methodology of existing studies [42], this paper divides the sample into high-tech and non-high-tech industries for heterogeneity analysis.

As shown in

Table 14. Results of the heterogeneity test using different industry categories.

	(1)	(2)
	High-Tech	Non-High-Tech
Variables	ROA	ROA
ESG	0.00429 ***	0.00203 ***
	(9.8586)	(4.1105)
Size	0.00216 ***	0.00391 ***
	(4.4544)	(7.2564)
Lev	−0.09767 ***	−0.10166 ***
	(−31.0936)	(−25.4098)
Cashflow	0.34633 ***	0.26069 ***
	(40.4610)	(27.5063)
Growth	0.03752 ***	0.02532 ***
	(22.4994)	(14.3995)
Top1	0.02399 ***	0.02312 ***
	(8.1168)	(6.3492)
Board	0.00111	−0.00465 *
	(0.4892)	(−1.8085)
Firmage	−0.00071	0.00045
	(−0.5158)	(0.2898)
Constant	−0.00190	−0.01750
	(−0.1816)	(−1.4151)
IND	YES	YES
YEAR	YES	YES
Observations	10947	7491
R-squared	0.420	0.416
p-Value	0.000 ***

Note: * Significance at the 10% level. *** Significance at the 1% level.

below, the estimated coefficient of enterprises’ ESG performance on financial performance in column (1) is 0.00429 and significant at 1% level; the estimated coefficient of enterprises’ ESG performance on financial performance in column (2) is 0.00203 and significant at 1% level, which suggests that ESG performance can significantly promote enterprises’ financial performance. By comparing the correlation coefficients, it shows that the estimated coefficient of firms’ ESG performance in column (1) is much larger than that in column (2), and the empirical p-value of the between-group coefficients difference test using Fisher’s Combined Test (with 1500 samples) is 0.000, which means that there is a difference in coefficients between the groups at the 1% level. The above results suggest that the ESG performance of high-tech enterprises is a stronger financial performance driver than non-high-tech enterprises.

6. Conclusions

This paper takes the A-share listed companies continuously involved in ESG performance from 2011 to 2022 as the research object and empirically examines the intrinsic mechanism between ESG performance and financial performance from the perspective of cost side, sales side, and development side through a bidirectional fixed-effect model. Then, this study’s results show a positive correlation between ESG and financial performance. Furthermore, agency cost, financing costs, social reputation, market power, and enterprise innovation play an intermediary role in the relationship between ESG performance and financial performance. Moreover, the relationship between corporate involvement in ESG and financial health is also affected by a series of internal and external factors, such as the property rights of the enterprise and the industry category of industry. Therefore, this study draws the following conclusions:

Firstly, enterprises can promote their financial performance by improving their ESG performance. Good ESG performance can help enterprises build a positive brand image and accumulate a good reputation to reduce financial risks when faced with negative events. Also, enterprises with good ESG performance can obtain external financial support through diversified financing channels, which is conducive to long-term financial health. Specifically, the effect of involvement in ESG performance on corporate financial performance is affected by two factors: the property rights of the enterprise and the industry category. On one hand, in terms of property rights, non-state-owned enterprises have a more significant relationship between ESG and financial performance due to greater market competition, more prominent principal-agent problems, and a more explicit objective of maximizing shareholders’ interests. On the other hand, in terms of the industry category, this study finds that, compared with non-high-tech firms, the relationship between ESG performance and financial performance is more significant for high-tech firms. The primary reason may be that high-tech enterprises are more sensitive to market changes and have a stronger willingness to make full use of technological innovation to reduce energy consumption and pollutant emissions in their production processes, thereby improving environmental performance and financial health.

Secondly, corporate involvement in ESG performance can improve financial health based on the partial mediating effect of the cost side, sales side, and development side. ESG can effectively mitigate information asymmetry between firms and external investors, greatly reducing firms’ agency costs. According to the resource-dependence theory, enhancing ESG performance is conducive to fostering a favorable corporate image and facilitating enterprises to secure loans from financial institutions at a lower cost. Furthermore, companies with good ESG performance that produce green and low-carbon products can have a clear competitive advantage in the market competition and thus significantly improve their financial performance. Moreover, enterprises with good ESG performance can effectively combine high-quality resources within the enterprise, improving the enterprise’s innovation capacity and avoiding being out of the market due to insufficient innovation.

Based on the above, combined with the characteristics of the Chinese market, this paper proposes the following measures to address. From the perspective of the enterprise, enterprises should attach more importance to involvement in corporate ESG activities and integrate social responsibility into the corporate culture, thus enhancing financial performance. Furthermore, enterprises could strengthen the development of innovations oriented to ESG concepts to build long-term competitive advantages. In addition, enterprises are supposed to use various channels, such as media, to improve the exchange of information with stakeholders and focus on the stakeholders’ actual needs.

From the perspective of policymakers, although ESG performance is significantly improved by companies themselves through engaging in environmentally friendly behaviors, actively assuming social responsibility, and improving governance, government departments can still encourage companies to improve their ESG levels through the implementation of relevant policies, which will in turn improve their financial performance. The policymakers should provide policy incentives to enterprises with good ESG performance and increase support for enterprises that actively fulfill their social responsibilities. For example, more favorable tax policies should be given to enterprises that actively improve their ESG performance, and policymakers should increase penalties for enterprises that ignore their social responsibilities and provide timely guidance. On the one hand, policymakers should improve the financing environment and provide policy support for enterprises to reduce financing costs from the external level. On the other hand, enterprises should construct a more scientific and comprehensive ESG evaluation system so that partners can more truly understand the level of corporate social responsibility fulfillment. In addition, it should be acknowledged that, limited by the availability of ESG rating data, the Hua Zheng database mainly provides ESG ratings of listed companies, which are relatively characterized by high assets, which limits the generalizability of the findings of this paper to a certain extent, pending subsequent studies using a wider range of ESG rating data of companies.