Does ESG Performance Promote High-Quality Development of Enterprises in China? The Mediating Role of Innovation Input

Abstract

Environmental, Social and Governance (ESG) concept has internal consistency with the high-quality development of China’s economy theoretically, and implementing ESG concept is an important way to achieve sustainable economic development. However, whether ESG performance can promote the high-quality development from the perspective of micro enterprises is rarely studied. Thus, we explore the impact and mechanism of ESG performance on enterprises’ high-quality development. The empirical results indicate that good ESG performance is conducive to promoting high-quality development of enterprises. Environmental performance and social performance can promote high-quality development of enterprises more than corporate governance performance. The mechanism results show that innovation input plays a medicating role between ESG performance and enterprises’ high-quality development. The additional analysis suggest that the promoting effect is more obvious in state-owned enterprises, environmentally sensitive enterprises and enterprises with less financing constraints. This study has enlightenment significance for enterprises to value ESG performance and government departments to formulate relevant policies.

1. Introduction

ESG, short for Environmental, Social and Governance, is an important concept in corporate management and financial investment that has emerged in recent years. Although ESG-related concepts such as corporate social responsibility and socially responsible investment have been put forward in the early 1960s, it was not until 2010s that ESG issues were gradually taken seriously by enterprises [1]. Since ESG is used in different contexts, there is no uniform definition [2]. It is typically used to evaluate the long-term sustainability of corporate investments [3]. In the United Nations Principles for Responsible Investment (UN-PRI), Environmental, Social and Governance factors refer to three different but related areas of social consciousness, which respectively represent the environmental responsibility, social responsibility and corporate governance [1]. The environmental (E) relates to the performance of a company as a manager of the natural environment, such as reducing greenhouse gas emissions, resource exploitation and waste discharge [4]. The second dimension (S) examines how a company handles and balances its relationship with its stakeholders including employees, suppliers, customers and communities [5]. The corporate governance (G) standard involves the company’s leadership, executive compensation, audit, internal control and shareholders’ rights [6]. According to the statistical Research Report on ESG Information Disclosure of A-Share Listed Companies in 2020 released by Syntao, the number of ESG reports issued by A-share listed companies has continued to grow since 2011. A total of 1092 A-share listed companies released ESG reports in 2020, accounting for 25.3%. Valuing ESG performance has become a trend in corporate development.

In recent years, China’s economy has shifted from a high-speed growth stage to a high-quality development stage [7]. As a new development concept, the high-quality development is characterized by innovation, green, coordination, openness, and sharing, which provides an important guarantee for the sustainable development of the economy and forms a consensus on global development [7,8]. The report of the 19th National Congress of the Communist Party of China clearly stated that we should “promote the transformation of the quality, efficiency and power of economic development, and improve total factor productivity”. The total factor productivity of enterprises is an indicator of the overall efficiency that reflects the transformation of factor resources into final products [9]. Improving total factor productivity is an important path for China to achieve higher quality, more efficient and more sustainable economic development. In September 2020, the proposal of “double carbon” policy-carbon neutralization and carbon peak, offered a clear goal for China’s high-quality economic development [10]. Hence, the “double carbon” policy is the main driving force advancing the development of ESG, and ESG is an important starting point to promote the implementation of the “double carbon” policy.

Theoretically, the ESG concept has internal consistency with the high-quality development of China’s economy from the macro level. The practice of ESG concept will speed up the high-quality development of China’s economy. While in terms of enterprises, can the practice of ESG concept by enterprises, that is, whether the ESG performance of enterprises can promote the high-quality development of enterprises? How does an enterprise’s ESG performance affect its quality development? And whether the impact of ESG performance on the high-quality development of enterprises changes with the heterogeneity of enterprises? The solution of these problems needs the support of empirical evidence.

Previous studies have failed to provide consistent results concerning how ESG performance affects financial performance or firm value. Some scholars have argued that good ESG performance can bring many positive impacts to enterprises. For example, ESG activities can protect stakeholder interests, thus meeting their needs and minimizing potential transaction costs [11]. Meanwhile, focusing on ESG performance can enable a company to better mobilize its internal resources [12], enhance employee loyalty and enthusiasm [13], thereby improving employee productivity and work efficiency [14]. Consequently, companies with high ESG performance have better financial performance and higher firm value [15,16,17]. However, some researches have proposed that ESG activities go against the goal of maximizing shareholders’ wealth. The cost of practicing ESG concept or disclosing ESG information is large than revenues [18,19]. Also, managers may value ESG performance or even greenwash ESG information for reasons of managerial utility and personal reputation [20,21], affecting adversely the company’s financial performance and overall firm. Besides, some findings have concluded that there is no obvious relationship between ESG performance and financial performance [22].

Most of the ESG-related researches has been focused on the impact on the cost of capital [23,24,25], the financing constraints [26], innovation level [27,28], bond default rates [29] besides the impact on financial performance or firm value. There is little literature on the influence of high-quality development of enterprises from the perspective of sustainability. At the same time, there have been few studies on how ESG performance exert an influence. Zhang et al. [30] reached the conclusion that the interaction effect between green innovation and social disclosure on firm value is a substitution effect. Thus, the economic consequences of ESG performance are uncertain and paradox. Related studies including mechanism research needs to be further explored.

High-quality development is a transition from denotative growth to connotative growth, which is reflected in the improvement of total factor productivity [7]. Based on stakeholder theory, signal transmission theory and resource dependence theory, this paper examines the impact of ESG performance on the high-quality development of enterprises. For micro-enterprises, the improvement of total factor productivity mainly depends on R & D and technological innovation [31]. In other words, innovation input is the intrinsic driver of firm productivity improvement and long-term economic growth [32,33]. Thus, this study explores the intermediary mechanism of ESG performance affecting the high-quality development of enterprises from the perspective of innovation input. Moreover, considering the heterogeneity of different Chinese enterprises, this paper further distinguishes the impact of ESG performance on high-quality development in enterprises with different property rights, and financing constraints.

Taking the Chinese A-share listed companies that publicly traded in the Shanghai Stock Exchange and Shenzhen Stock Exchange from 2011 to 2019 as the research sample, this paper empirically examined the influence of ESG performance on high-quality development of enterprises, as well as the specific impact mechanism from the perspective of corporate innovation input. The results were as follows: first, Chinese A-share listed companies that value ESG performance would see a big increase in total factor productivity. Our results remained stable after using alternative measurements for the high-quality development of enterprises and excluding some sample periods. To mitigate the endogenous brought by the reverse causality and omitting variables, we adopted the robustness checks including adding the control variable at the regional level, lag regressions and instrumental variable regressions. Second, corporate innovation input played a mediating role in explaining the positive effect of the ESG performance on enterprises’ total factor productivity. The stepwise regressions and bootstrap tests indicated that enterprises with good ESG performance would further improve their total factor productivity by increasing innovation input. The results of additional tests showed that the property rights, environmental sensitivity and financing constraints moderated the impact of ESG performance on the total factor productivity. The positive relationship became more pronounced for state-owned enterprises, environmentally sensitive enterprises and enterprises with less financing constraints.

This paper studies the relationship between corporate ESG performance and high-quality development and its impact mechanism. It contributes to the existing literature in the following ways. First, under the context of the vigorous development of ESG, this paper explores the impact of ESG on the high-quality development of enterprises from both theoretical and empirical aspects. The research on high-quality development mainly focuses on the macro level, such as countries and regions. And there is little literature on the high-quality development of enterprises. Based on the signal transmission theory, stakeholder theory and resource dependence theory, this paper theoretically analyzes the impact of ESG on the high-quality development of enterprises, and expands the research on the economic consequences of ESG performance. Second, limited attention has been paid to the specific mechanism of ESG performance. This research examines the impact mechanism of ESG performance in promoting the high-quality development of listed firms from the perspective of innovation input, providing empirical evidence for the improvement of firms’ innovation capability brought by ESG performance. This paper examines the specific mechanism of ESG performance promoting the high-quality development of listed enterprises from the perspective of innovation input, and provides empirical evidence for the improvement of enterprise innovation ability brought by ESG performance. Third, combined with the heterogeneity of Chinese enterprises, this study further discusses the impact of ESG performance on the high-quality development of enterprises from the three aspects of property rights, environmental sensitivity and financing constraints, which provides support and practical inspiration. theories for different types of Chinese enterprises to achieve high-quality development goals.

The rest of the paper is organized as follows. We make a description of institutional background and literature review in Section 2. The hypotheses development is presented in Section 3. Section 4 provides the methodology, including sample selection, variable definition, and research methods. Section 5 is the empirical results and analysis. Section 6 offers discussion and the study conclusions are drawn in Section 7.

2. Institutional Background and Literature Review

2.1. Institutional Background

ESG was a product of Socially Responsible Investment (SRI). The concept of SRI can be dated back to the 1960s. After the Second World War, with the rapid growth of industrial economy in Europe and the United States, a series of environmental and social problems emerged. While the public environmental protection movements were launched, the capital market began to attach importance to environmentally responsible investment [1]. The term “environment, society and governance” first appeared in the 2004 UN report entitled “Who Cares Who Wins”. In 2006, UN-PRI was formally established. And this organization initially proposed the environmental, social and governance framework and listed some factors to be considered. Since then, ESG has developed rapidly and has been widely used for investment decisions in the developed countries [34]. At present, there are some ESG rating systems in the world, such as the Dow Jones Sustainability Index, Thomson Reuters, FTSE Russell, Goldman Sachs, etc. [35].

In order to comply with the global ESG development trend, relevant government departments and regulatory agencies in China have successively promulgated policies to promote the development of ESG. In 2015, the “Environmental, Social and Governance Reporting Guidelines” issued by the Hong Kong Stock Exchange made semi-mandatory requirements for corporate ESG information disclosure. In 2018, CSRC revised the “Governance Guidelines for Listed Companies” to introduce ecological and environmental protection requirements into the corporate governance framework. In the same year, Shanghai Stock Exchange and Shenzhen Stock Exchange drafted ESG Information Disclosure Guidelines and held many consultation forums [30]. In June 2021, CSRC again revised the periodic reporting formats and guidelines for listed companies, adding two chapters specifically for ESG disclosure requirements. It can be seen that governments and enterprises attach increasing importance to ESG system construction and information disclosure. Moreover, the number of ESG rating agencies is rising, such as Synao green finance, CASVI, Sino-Securities Index etc.

2.2. Literature Review

Since the concept of ESG was introduced, ESG performance has been concerned widely and studied extensively by scholars. Early studies related to the economic consequences of ESG performance focused on financial performance and firm value but provided mixed results. Some scholars found a positive relationship between ESG performance and financial performance [15,16,17]. Brogi and Lagasio [15] used ROA as the proxy for financial performance and further concluded that in the ESG three dimensions, S was the most related factor compared with E and G. Duque-Grisales et al. [36] confirmed that ESG performance and financial performance was significantly statistically negative. Another findings suggested that there was no obvious relationship between them [22,37]. In the researches related to firm value, most affirmed the positive relationship between ESG performance and firm value [38,39]. Li et al. [38] further indicated that higher CEO power could enhance the ESG disclosure effect on firm value. However, Velte [40] found that ESG performance had a positive impact on ROA but no impact on firm value.

In the past three years, the research on the impact of ESG on enterprises has been expanding. A company’s ESG performance or its commitment to ESG responsibility can reduce capital costs [23,24,25], ease financing constraints [26], improve innovation level [27,28], and affect bond default rates [29]. In terms of the impact on the cost of capital, Eliwa et al. [23] held the idea that firms with stronger ESG performance have a lower cost of debt on the basis of legitimacy and institutional theories, taking firms in 15 EU countries as the research object. Ok and Kim [24] regarded ESG grades as the total CSR performance and their research conclude that Korea firms with CSR performance generally exhibit cheaper equity financing. Chouaibi et al. [25] found that CSR activities lower the cost of equity capital from French ESG Data. As for financing constraints, Zhang et al. [26] used DID to assesses the effects of COVID-19-associated shocks on financial constraints and found that strong firm ESG performance could alleviate the financial constraints caused by the pandemic. Besides, under the background of the COVID-19 Pandemic, ESG factor could explain industry returns and environmental and social dimensions are the main drivers of the ESG impact on different industries [41]. In the field of the innovation performance, corporate environmental initiatives, social initiatives and governance initiatives have direct positive impacts on it [27]. Especially, ESG performance can increase the number of green invention patents [28]. Also, the study by Li et al. [29] indicated that there is a positive relation between environmental performance and the bond default rate, which is negatively correlated with social responsibilities and corporate governance.

Some scholars have also studied ESG from other perspectives. Daugaard [42] explored the literature related to ESG including the heterogeneous nature of ESG investing, its costs and motivations, and its management literature origins. While Li et al. [43] used the literature analysis tool Citespace and reviewed ESG related research from the following five aspects: the theoretical basis of ESG research, the interaction between the dimensions of ESG, the impact of ESG on the economic consequences, the risk prevention role of ESG, and ESG measurement. Additionally, based on the differences between rating agencies on ESG ratings of individual companies, Christensen [35] found that emphasizing ESG disclosure will usually exacerbate ESG rating differences. The greater the ESG disagreement, the higher the volatility of returns and the larger the price movements. Therefore, this paper will explore the economic consequences of ESG performance from the perspective of high-quality development of enterprises and study its mechanism.

3. Hypotheses Development

3.1. ESG Performance and High-Quality Development of Enterprises

The goal of enterprises with good ESG performance is not only to maximize profits, but to optimize the allocation of resources and consider multi stakeholders, so as to promote their sustainable development [1].

First of all, from the perspective of the relationship between environmental responsibility performance and high-quality development of enterprises, good environmental responsibility performance can promote high-quality development of enterprises. On the one hand, green is one of the new development concepts, and the Chinese government has adopted a variety of environmental regulations to strengthen the construction of ecological civilization [44]. Based on the perspective of external pressure and legitimacy, under the hard constraints of external environmental regulations and the soft constraints of social public environmental requirements, the expected pollution control costs and production costs of enterprises have greatly increased [45]. In order to reduce the costs of enterprises, their willingness and motivation to improve their environmental performance will be enhanced. On the other hand, environmental pollutants, as a kind of waste discharged by enterprises in the production process, reflect the inefficiency of resource utilization in the production process [12]. According to the viewpoint of sustainable development theory, it is beneficial for enterprises to take environmental responsibility to reduce the emission of environmental pollutants, thereby effectively saving energy [46]. Therefore, the emphasis and practice of enterprises on environmental protection will help enterprises to use appropriate environmental management methods, increase pollution prevention and control efforts, improve resource utilization efficiency, and promote high-quality development.

Secondly, from the perspective of the relationship between social responsibility performance and high-quality development of enterprises, enterprises can improve their development quality by undertaking social responsibility. First, according to stakeholder theory and signal transmission theory, good corporate social performance is conducive to create corporate image and enhance corporate reputation [20,21]. The social responsibility report issued by enterprises provides a kind of non-financial information, which can alleviate the information asymmetry between investors and managers [47], reduce the financing constraints, and expand the source of funds [48], so as to lay a foundation for the high-quality development of enterprises. Second, based on resource dependence theory, having scarce, valuable and imitable resources can help companies improve their product quality and service ability, thereby gaining and maintaining competitive advantages [47]. In essence, corporate social responsibility is an investment rather than a cost. It can help employees, investors and other stakeholders bring positive returns [12]. As an invisible resource, good social responsibility performance is conductive to obtain the value identity of the employees, attract excellent human resources to participate in the production activities [13,20], and thus achieve high-quality development of enterprises. Besides, from the perspective of new institutional economics, fulfilling social responsibility contributes to guiding enterprises to weigh the marginal income and marginal cost of social capital and optimize its internal value, so as to realize the Pareto optimal state of resource input and output [49].

Finally, from the perspective of the relationship between corporate governance performance and high-quality development of enterprises, improving corporate governance is beneficial to achieve high-quality development of enterprises. Good corporate governance can coordinate the interest relationship among shareholders, the board of directors, and managers and alleviate the principal-agent problems [50]. On the one hand, effective corporate governance is conducive to strengthening the supervision of shareholders over managers and reducing the first type of agency costs. An efficient board of directors has stronger supervision and can play a better role in promoting the total factor productivity of enterprises [51]. On the other hand, good corporate governance performance can alleviate the hollowing out behavior of major shareholders, reduce the second type of agency costs, and further improve production efficiency [52]. When the major shareholders hollow out and occupy the company’s resources, various investment activities will have a crowding-out effect, which is not conducive to speeding up production efficiency. Meanwhile, a sound incentive mechanism is good for improving the enthusiasm of the managers [19,20], thereby improving the level of organizational production. Especially equity incentive system, a performance-oriented compensation structure system can improve the company’s efficiency to a certain extent. Thus, the existence of equity incentives will make managers worker hard and thus advance enterprises’ development. Based on the above analysis, we propose the following hypotheses:

Hypothesis 1 (H1).

Good ESG performance is beneficial to promoting high-quality development of enterprises.

Hypothesis 1a (H1a).

Good environmental performance is beneficial to promoting high-quality development of enterprises.

Hypothesis 1b (H1b).

Good social performance is beneficial to promoting high-quality development of enterprises.

Hypothesis 1c (H1c).

Good governance performance is beneficial to promoting high-quality development of enterprises.

3.2. The Mediating Role of Innovation Input

High-quality development is a transition from denotative growth to connotative growth, which is reflected in the improvement of total factor productivity [7]. And technology progress is a direct channel that affects the total factor productivity of enterprises. The level of total factor productivity and profit rate of enterprises with R & D investment is about 14% higher than that of enterprises without R & D [53,54]. Therefore, companies that focus on implementing ESG concepts will promote high-quality development by increasing investment in innovation. The specific analysis is as follows:

In terms of the relationship between environmental responsibility performance, innovation input and high-quality development of enterprises, enterprises taking environmental responsibility will increase innovation input, thereby improving the total factor productivity of enterprises. First, good environmental responsibility performance means that enterprises need to use raw materials that meet environmental protection standards, save energy consumption in all aspects of the production process, and then produce green products [55]. Thus, companies will be forced to increase R & D investment and improve the efficiency of resource utilization from the beginning of production to the completion of production [28]. Second, enterprises’ commitment to environmental responsibility and disclosure of environmental information brings a good image of a green responsibility practitioner to the outside world. Enterprises may not only receive more financial subsidies from the government, but also gain the favor of investors [33], so as to provide financial support for the R & D investment of enterprises and help them optimize production processes and improve production efficiency. Third, enterprises that focus on environmental performance tend to maintain close ties with environmental organizations, which will accelerate the sharing of environmental knowledge, and make internal and external knowledge combine, thereby triggering the innovation potential of enterprises [56]. Thus, enterprises will accelerate the pace of R & D, and then promote their high-quality development.

In terms of the relationship between social responsibility performance, innovation input and high-quality development of enterprises, enterprises that focus on undertaking social responsibility can promote the high-quality development of enterprises by increasing their innovation input. First, the funds needed by enterprises for R & D investment mainly come from equity financing and bond financing [48]. The disclosure of relevant information about the fulfillment of social responsibilities can enhance shareholders’ willingness to invest and seek debt financing opportunities by providing guarantees and repaying debts in a timely manner, reducing capital costs and providing financial support for R & D investment [28], thereby improving corporate resource allocation. Second, corporate R & D personnel can not only accumulate practical experience and gain functional benefits from participating in various social responsibility activities, but also cultivate the feelings between team members and improve work efficiency, which can be effectively transformed into innovative outputs [57]. At the same time, corporate social responsibility can convey its values and norms to the technical talent market, attract more potential high-quality R & D talents, improve R & D efficiency, and gain competitive advantages [58]. Third, companies that value social responsibility performance are more likely to establish extensive and strong relationships with multiple stakeholders. These relationship networks are conducive to obtaining diversified innovation resources and accelerating the pace of innovation for enterprises to achieve high-quality development [38].

From the perspective of the relationship between corporate governance performance, innovation input and high-quality development of enterprises, sound corporate governance is oriented to promote corporate innovation and value creation, and advance its sustainable development. First, since large shareholders are more willing to pay more costs to supervise the innovation process of enterprises, while small and medium-sized shareholders focus on short-term interests and ignore the long-term performance brought by innovation [59], mutual restriction can not only prevent major innovation decision-making risks, but also reduce the coordination cost and achieve the high-quality development of enterprises. Second, the appropriate incentive mechanism is one of the most important factors affecting the R & D activities of enterprises. The existence of equity incentives will promote the alignment of the interests of managers and core technical personnel with shareholders, thereby further increasing the R & D investment of enterprises [60], promoting technological upgrading and improving total factor productivity. Third, good corporate governance performance is also reflected in the fact that independent directors can play their substantive roles. Independent directors actively perform their due responsibilities and obligations, not only supervising the management and avoid short-sighted behavior, but also providing valuable reference for the company’s innovative decision-making, so as to promote the company to increase R & D efforts and integrate existing resources, achieve high-quality development [44].

Thus, we propose the following hypothesis:

Hypothesis 2 (H2).

Innovation input mediates the impact of ESG performance on firms’ high-quality development.

Hypothesis 2a (H2a).

Innovation input mediates the impact of environmental performance on firms’ high-quality development.

Hypothesis 2b (H2b).

Innovation input mediates the impact of social performance on firms’ high-quality development.

Hypothesis 2c (H2c).

Innovation input mediates the impact of governance performance on firms’ high-quality development.

4. Methodology

4.1. Sample Selection

Bloomberg provides data for Chinese listed companies from 2006 onward, but there are many missing scores from 2006 to 2010. Therefore, this paper takes 2011 as the starting year of the sample and selects Shanghai and Shenzhen A-share listed companies from 2011 to 2019 as the research object. To enhance measurement validity, we eliminated the initial sample as follows: (1) enterprises marked with ST and *ST with abnormal data; (2) firms operating in the financial sector and insurance sector; (3) samples with incomplete data. Finally, we obtained a total of 4985 observations. Then we winsorized all continuous variables at the 1% and 99% percentiles of their distribution to alleviate the effect of outliers. ESG data were obtained from the ESG index published by Bloomberg, and financial data were extracted from the Wind Info Database and the CSMAR database.

4.2. Variable Definition

(1) Dependent variable: high-quality development of enterprises. As for enterprises, high-quality development refers to the pursuit of a high level and high efficiency of economic value and social value creation. Following Zhao et al. [7], the research took the total factor productivity (TFP) of enterprises as a suitable proxy for high-quality development. The existing literature on the measurement of enterprise total factor productivity mainly are OLS method, semi-parametric method and parametric method. Among them, semi parametric method including LP method and OP method is the most widely used. And compared with OP method, LP method considers more proxy variables and has a more scientific calculation process. Thus, we adopted the method proposed by Levinsohn and Petrin to estimate the TFP [61]. The calculation model is as follows:

$$ln(Y_{i,t})={\theta }_0+{\theta }_{l}ln(L_{i,t})+{\theta }_{k}ln(K_{i,t})+{\theta }_{m}ln(M_{i,t})+{\epsilon }_{i,t}$$

(1)

where, Y is the output of enterprises, measured by operating income; L is labor input, measured by the total number of employees, K is capital input, measured by capital expenditure; M is the input of intermediate products, measured by the cash actually paid for purchasing goods and receiving services.

(2) Independent variables: ESG performance and its three dimensions. The ESG data provided by Bloomberg contain a composite, firm–year measure of ESG disclosure score and individual disclosure scores for the component parts of ESG [23]. Bloomberg’s ESG evaluation system is comprehensive and complete, as detailed in Appendix A

Table A1. Bloomberg’s ESG evaluation system.

Pillar	Topic	Weight
Environmental (33.33%)	Air Quality	4.78%
	Climate Change	4.70%
	Ecological & Biodiversity Impacts	4.79%
	Energy	4.73%
	Materials & Waste	4.74%
	Supply Chain	4.79%
	Water	4.79%
Social (33.33%)	Community & Customers	5.53%
	Diversity	5.49%
	Ethics & Compliance	5.57%
	Health & Safety	5.58%
	Human Capital	5.55%
	Supply Chain	5.54%
Governance (33.33%)	Audit Risk & Oversight	4.17%
	Board Composition	4.16%
	Compensation	4.16%
	Diversity	4.17%
	Independence	4.18%
	Nominations & Governance Oversight	4.18%
	Sustainability Governance	4.18%
	Tenure	4.18%

. Since Bloomberg’s ESG rating of Chinese listed companies started earlier and covered a relatively wide range, this paper selected ESG disclosure scores to measure ESG performance. ESG comprehensive performance is the arithmetic mean of the three dimensions.

ESG performance refers to the actual ESG-related activities conducted by the firm [62]. The more ESG activities companies carry out, the higher their disclosure scores, and the better ESG performance. For example, reducing pollutant emissions and water consumption, implementing energy efficiency and policies biodiversity policies and minimizing environmental fines belong to environmental activities, represent good environmental performance. Valuing human rights, supporting minorities and disabled in workforce and charitable contribution are usually social activities carried out by enterprises. The good governance performance means that the company has a relatively complete governance structure such as audit committee, board, compensation committee and nomination committee. And these enterprises also pay attention to strengthen employee trainings.

(3) Mediator variable: Innovation input. The innovation input refers to the resources and routines put into the efforts to create innovation [61]. Following Jiang [63], we used R & D expenses to estimate innovation input. In the model, we take the natural logarithm of R & D expenses.

(4) Control variables: Based on the related topic researches and in view of the factors affecting the total factor productivity of enterprises, we select nine sub-indicators as control variables, including some enterprise attributes and financial attributes. In addition, the industry fixed effects and the year fixed effects are controlled in the research models. The specific definitions of each variable are shown in

Table 1. Definitions of variables.

Type	Variable	Symbol	Definition
Dependent variable	High-quality development of enterprises	TFP	The total factor productivity of enterprises calculated through Levinsohn–Petrin (LP) method
Independent variables	ESG comprehensive performance	ESG	ESG disclosure score in Bloomberg
	Environmental performance	E	Environmental disclosure score in Bloomberg
	Social performance	S	Social disclosure score in Bloomberg
	Corporate governance	G	Governance disclosure score in Bloomberg
Mediator variable	Innovation input	Innovation	Natural logarithm of total R & D expenses
Control variables	Firm size	Size	Natural logarithm of total assets
	Total debt ratio	Lev	Total liabilities/total assets
	Return on total assets ratio	ROA	return/total assets
	Firm growth	Growth	Operating income growth rate
	Cash flow return on assets ratio	CFROA	Net cash flow from operating activities/total assets
	Investment expense ratio	Invt	Cash paid for purchase and construction of fixed assets, intangible assets and other long-term assets /total assets
	Tobin Q	Tobin	Market value/replacement cost
	Intangible assets ratio	Intang	Net intangible assets/total assets
	Listing period	Age	Natural logarithm of the year minus listing year
	Industry fixed effects	Industry	Industry dummy
	Year fixed effects	Year	Year dummy

.

4.3. Research Method

4.3.1. Baseline Model

In order to verify the relationship between ESG performance and the high-quality development of enterprises, we propose the baseline model as follows:

$$TF{P}_{i,t}={\mathsf{\alpha }}_0+{\mathsf{\alpha }}_{1}ES{G}_{i,t}+{\mathsf{\alpha }}_j{\displaystyle \sum _{j}Control{s}_{i,t}}+{\displaystyle \sum Industry}+{\displaystyle \sum Year}+{\mathsf{\mu }}_{i,t}$$

(2)

In model (2), the dependent variable is TFP_i,t, the total factor productivity of the firm i in year t. The independent variable is ESG_i,t, which measures ESG performance of the firm i in year t. Our primary interest is the coefficient α₁ because it captures the impact of the ESG performance on TFP. If α₁ is significantly greater than 0, H1 and its subhypotheses (H1a, H1b, H1c) are supported.

4.3.2. The Mediating Role of Innovation Input

To test the mediation effect of innovation input, we build model (3) and model (4) according to the mediation effect test process of Wen and Ye [64]. The process is based on the stepwise regression proposed by Baron and Kenny [65] and integrates the bootstrap test method. includes stepwise regression method and Bootstrap method, thereby improving the explanatory power of the results.

$$Innovatio{n}_{i,t}={\mathsf{\beta }}_0+{\mathsf{\beta }}_{1}ES{G}_{i,t}+{\mathsf{\beta }}_j{\displaystyle \sum _{j}Control{s}_{i,t}}+{\displaystyle \sum Industry}+{\displaystyle \sum Year}+{\mathsf{\mu }}_{i,t}$$

(3)

$$TF{P}_{i,t}={\mathsf{\gamma }}_0+{\mathsf{\gamma }}_{1}ES{G}_{i,t}+{\mathsf{\gamma }}_{2}Innovatio{n}_{i,t}+{\mathsf{\gamma }}_j{\displaystyle \sum _{j}Control{s}_{i,t}}+{\displaystyle \sum Industry}+{\displaystyle \sum Year}+{\mathsf{\mu }}_{i,t}$$

(4)

In the above two models, β₁ is the influence coefficient of ESG performance on innovation input; γ₁ is the influence coefficient of ESG performance on the high-quality development of enterprises, reflecting the direct effect; γ₂ is the influence coefficient of innovation input on the high-quality development of enterprises, and β₁γ₂ reflects the indirect effect.

5. Empirical Results and Analysis

5.1. Descriptive Statistics

The descriptive statistics results of the full sample are presented in

Table 2. Descriptive statistical results of variables.

Variable	N	Mean	Std	Median	Min	Max
TFP	4985	8.716	1.008	8.642	5.866	11.747
ESG	4985	22.126	6.347	20.661	11.983	45.454
E	4985	11.198	7.840	9.302	2.326	42.636
S	4985	24.896	9.103	22.807	8.772	57.895
G	4985	45.058	5.192	44.643	33.929	58.929
Size	4985	23.090	1.347	22.966	20.434	27.040
Lev	4985	0.466	0.196	0.474	0.065	0.869
ROA	4985	0.048	0.059	0.039	−0.149	0.218
Growth	4985	0.159	0.413	0.104	−0.631	5.830
CFROA	4985	0.057	0.065	0.053	−0.118	0.244
Invt	4985	0.048	0.041	0.036	0.001	0.199
Tobin	4985	1.950	1.212	1.537	0.850	7.543
Intang	4985	0.051	0.053	0.037	0.001	0.327
Age	4985	2.427	0.585	2.565	0.693	3.258

. As can be shown, the mean TFP is 8.716, the minimum value and the maximum value of TFP are 5.866 and 11.747 respectively, indicating that the total factor productivity of sample enterprises varies greatly, but the overall level is high. Also, there is a large difference between the maximum and minimum ESG performance scores, which means the overall performance is not good enough. Among the three dimensions of ESG performance, compared with governance performance, the range and standard deviation of environmental and social performance are larger, and their mean and median are obviously lower. Therefore, the environmental and social performance of the sample enterprises needs to be improved. In terms of control variables, the large standard deviation of Size, Growth and Tobin reveals the differences in the scale, growth and development prospects of enterprises. Negative values of Cflow and Roa indicate that samples’ operating level and profitability are quite different.

5.2. Baseline Regression Results

Table 3. Baseline regression results.

	(1)	(2)	(3)	(4)
Variable	TFP	TFP	TFP	TFP
ESG	0.007 ***
	(5.51)
E		0.006 ***
		(6.08)
S			0.003 ***
			(3.48)
G				0.003 *
				(1.90)
Size	0.552 ***	0.551 ***	0.556 ***	0.559 ***
	(71.02)	(72.03)	(73.64)	(73.23)
Lev	0.880 ***	0.880 ***	0.877 ***	0.868 ***
	(16.34)	(16.35)	(16.27)	(16.15)
ROA	2.925 ***	2.936 ***	2.908 ***	2.912 ***
	(16.84)	(16.90)	(16.77)	(16.77)
Growth	0.087 ***	0.087 ***	0.086 ***	0.087 ***
	(4.73)	(4.71)	(4.69)	(4.75)
CFROA	0.828 ***	0.819 ***	0.849 ***	0.853 ***
	(6.14)	(6.06)	(6.28)	(6.29)
Invt	−1.407 ***	−1.401 ***	−1.409 ***	−1.395 ***
	(−7.46)	(−7.42)	(−7.47)	(−7.39)
Tobin	−0.038 ***	−0.038 ***	−0.038 ***	−0.038 ***
	(−5.08)	(−5.11)	(−5.13)	(−5.14)
Intang	−0.965 ***	−0.960 ***	−0.960 ***	−0.957 ***
	(−5.88)	(−5.86)	(−5.83)	(−5.84)
Age	0.041 ***	0.044 ***	0.042 ***	0.037 ***
	(3.13)	(3.37)	(3.23)	(2.75)
Constant	−4.670 ***	−4.596 ***	−4.729 ***	−4.835 ***
	(−24.87)	(−24.27)	(−25.26)	(−25.98)
Industry	Yes	Yes	Yes	Yes
Year	Yes	Yes	Yes	Yes
N	4985	4985	4985	4985
R2	0.811	0.811	0.811	0.810

Note: t-statistics in parentheses. * p < 0.1 *** p < 0.01.

provides the baseline regression results between ESG performance and its three dimensions and TFP. As can been seen from column (1), the coefficient between ESG performance and TFP is significantly positive at the level of 1%, which suggests that ESG performance is conducive to promoting the high-quality development of enterprises and H1 is supported. The results in column (2) and column (3) indicate that both environmental performance and social performance are positively correlated with TFP at the 1% level. This shows that both environmental performance and social performance contribute to the high-quality development of enterprises, so H1a and H1b are verified. The influence coefficient of governance performance on TFP in column (4) is significant at the level of 10%. It can be revealed that environmental performance and social performance can promote the high-quality development of enterprises better than governance performance. The results can be explained by the following reasons.

On the one hand, among the three dimensions of environment, social responsibility and corporate governance, ESG concept pays more attention to environmental and social responsibility. Under the background of the emphasis on corporate ESG information disclosure, in order to convey a responsible and good image, companies will practice environmental responsibility and social responsibility by increasing R & D efforts, updating production equipment, and establishing good relationships with suppliers so as to promote the sustainable development of enterprises. On the other hand, corporate governance mainly measures the board of directors, general meeting of shareholders, board of supervisors and other governance structures, involving the relationship between internal stakeholders, and the score difference for governance performance of sample enterprises is small, so it is likely that the promotion effect of governance performance on the high-quality development of enterprises is not as obvious as that of environmental performance and social performance.

5.3. Robustness Checks

5.3.1. Alternative Measurement for High-Quality Development of Enterprises

In order to ensure the robustness of the previous results, we use the OLS method to calculate the total factor productivity, and regress the baseline model again. The results are presented in

Table 4. Alternative Measurement for total factor productivity.

	(1)	(2)	(3)	(4)
Variable	TFP	TFP	TFP	TFP
ESG	0.005 ***
	(4.36)
E		0.004 ***
		(4.83)
S			0.003 ***
			(3.55)
G				0.003 ***
				(2.87)
Size	0.773 ***	0.773 ***	0.781 ***	0.782 ***
	(96.74)	(98.00)	(100.82)	(99.46)
Lev	0.880 ***	0.880 ***	0.877 ***	0.868 ***
	(16.34)	(16.35)	(16.27)	(16.15)
ROA	2.611 ***	2.625 ***	2.589 ***	2.596 ***
	(14.39)	(14.46)	(14.26)	(14.31)
Growth	0.073 ***	0.073 ***	0.072 ***	0.074 ***
	(3.92)	(3.89)	(3.87)	(3.96)
CFROA	1.327 ***	1.315 ***	1.357 ***	1.359 ***
	(9.59)	(9.50)	(9.76)	(9.75)
Invt	−0.792 ***	−0.783 ***	−0.793 ***	−0.774 ***
	(−4.01)	(−3.96)	(−4.01)	(−3.92)
Tobin	−0.042 ***	−0.043 ***	−0.043 ***	−0.043 ***
	(−5.45)	(−5.50)	(−5.53)	(−5.51)
Intang	−0.869 ***	−0.863 ***	−0.860 ***	−0.859 ***
	(−5.16)	(−5.14)	(−5.09)	(−5.12)
Age	0.061 ***	0.065 ***	0.063 ***	0.055 ***
	(4.50)	(4.79)	(4.57)	(3.92)
Constant	−6.884 ***	−6.788 ***	−6.979 ***	−7.114 ***
	(−35.57)	(−34.75)	(−36.13)	(−37.14)
Industry	Yes	Yes	Yes	Yes
Year	Yes	Yes	Yes	Yes
N	4985	4985	4985	4985
R2	0.879	0.877	0.876	0.874

Note: t-statistics in parentheses. *** p < 0.01.

. Not only the coefficient of ESG performance on TFP is significantly positive, but also there are correlated between performance of each dimension and TFP significantly and positively. Thus, our results for H1 and its subhypotheses (H1a, H1b, H1c) are robust after using an alternative measure of the total factor productivity.

5.3.2. Excluding Some Sample Periods

Bloomberg’s ESG performance data was based on a small sample size before 2015. In order to reduce the impact of sample bias on the final results, we eliminated the research samples before 2015 for further regression.

Table 5. The results of excluding some sample periods.

	(1)	(2)	(3)	(4)
Variable	TFP	TFP	TFP	TFP
ESG	0.005 ***
	(4.36)
E		0.004 ***
		(4.83)
S			0.003 ***
			(3.55)
G				0.003 *
				(1.90)
Size	0.550 ***	0.551 ***	0.557 ***	0.561 ***
	(58.64)	(59.46)	(61.48)	(61.01)
Lev	0.864 ***	0.861 ***	0.868 ***	0.850 ***
	(12.96)	(12.92)	(12.97)	(12.74)
ROA	2.823 ***	2.830 ***	2.809 ***	2.799 ***
	(13.52)	(13.55)	(13.45)	(13.43)
Growth	0.090 ***	0.090 ***	0.089 ***	0.089 ***
	(4.10)	(4.08)	(4.06)	(4.08)
CFROA	1.014 ***	1.011 ***	1.042 ***	1.049 ***
	(6.12)	(6.09)	(6.25)	(6.28)
Invt	−1.642 ***	−1.626 ***	−1.646 ***	−1.633 ***
	(−6.86)	(−6.78)	(−6.88)	(−6.80)
Tobin	−0.040 ***	−0.040 ***	−0.040 ***	−0.040 ***
	(−4.82)	(−4.83)	(−4.79)	(−4.80)
Intang	−1.084 ***	−1.074 ***	−1.077 ***	−1.062 ***
	(−5.16)	(−5.14)	(−5.09)	(−5.09)
Age	0.046 ***	0.051 ***	0.050 ***	0.036 **
	(2.81)	(3.12)	(3.01)	(2.10)
Constant	−4.923 ***	−4.859 ***	−5.048 ***	−5.221 ***
	(−22.43)	(−21.84)	(−23.40)	(−24.40)
Industry	Yes	Yes	Yes	Yes
Year	Yes	Yes	Yes	Yes
N	4985	4985	4985	4985
R2	0.816	0.814	0.814	0.810

Note: t-statistics in parentheses. * p < 0.1 ** p < 0.05 *** p < 0.01.

provides the test results. The result of column (1) confirms the positive correlation between ESG performance and TFP. The results of other columns indicate that there are related positively between each dimension performance and the high-quality development of enterprises. In addition, when the research sample from 2011 to 2014 is removed year by year, the results are equally robust.

5.3.3. Endogeneity Mitigation

To alleviate the endogeneity problems, we adopted three methods: adding the control variable at the regional level in the baseline model, lag regressions and instrumental variable regressions.

First, in view of the possible endogeneity of omitting variables in the baseline model and the influence of the total factor productivity of enterprises at the regional level, we added the regional marketization process (Market) into the control variables. According to the marketization index proposed by Wang et al. [66], we got the level of marketization process in the region where the company is located and reexamined the impact of ESG performance and each dimension performance on TFP. The results are presented in

Table 6. The results of adding the control variable at the regional level.

	(1)	(2)	(3)	(4)
Variable	TFP	TFP	TFP	TFP
ESG	0.004 ***
	(3.63)
E		0.004 ***
		(4.25)
S			0.002 ***
			(2.81)
G				0.002 *
				(1.70)
Size	0.550 ***	0.550 ***	0.555 ***	0.557 ***
	(70.88)	(71.84)	(73.28)	(72.96)
Lev	0.898 ***	0.899 ***	0.895 ***	0.888 ***
	(16.69)	(16.71)	(16.62)	(16.55)
ROA	2.879 ***	2.890 ***	2.864 ***	2.867 ***
	(16.64)	(16.69)	(16.58)	(16.58)
Growth	0.100 ***	0.100 ***	0.099 ***	0.100 ***
	(5.01)	(4.99)	(4.97)	(5.03)
CFROA	0.812 ***	0.802 ***	0.829 ***	0.832 ***
	(6.01)	(5.93)	(6.12)	(6.13)
Invt	−1.467 ***	−1.461 ***	−1.470 ***	−1.460 ***
	(−7.79)	(−7.74)	(−7.80)	(−7.75)
Tobin	−0.037 ***	−0.037 ***	−0.038 ***	−0.038 ***
	(−5.04)	(−5.06)	(−5.09)	(−5.09)
Intang	−0.956 ***	−0.953 ***	−0.951 ***	−0.949 ***
	(−5.80)	(−5.80)	(−5.76)	(−5.78)
Age	0.051 ***	0.054 ***	0.052 ***	0.048 ***
	(3.98)	(4.18)	(4.06)	(3.66)
market	0.032 ***	0.032 ***	0.032 ***	0.033 ***
	(8.22)	(8.20)	(8.33)	(8.48)
Constant	−4.670 ***	−4.596 ***	−4.729 ***	−4.835 ***
	(−24.87)	(−24.27)	(−25.26)	(−25.98)
Industry	Yes	Yes	Yes	Yes
Year	Yes	Yes	Yes	Yes
N	4985	4985	4985	4985
R2	0.813	0.812	0.813	0.810

Note: t-statistics in parentheses. * p < 0.1 *** p < 0.01.

. The results in the first column confirm the positive relationship between ESG performance and TFP. The results of the last three columns suggest that the performance of three aspects can advance the high-quality development of enterprises and the promotion effect of governance performance is relatively weaker. The research conclusions were supported.

Second, since there may be a reverse causal relationship between the comprehensive performance including environmental, social and governance performance and the high-quality development of enterprises, that is, the high-quality development of enterprises may improve the comprehensive ESG performance and three dimensions performance. We retested the baseline model with a one-period lag of all explanatory and control variables.

Table 7. The lag regression results.

	(1)	(2)	(3)	(4)
Variable	TFP	TFP	TFP	TFP
ESG	0.007 ***
	(5.03)
E		0.006 ***
		(5.22)
S			0.004 ***
			(4.42)
G				0.003 *
				(1.71)
Size	0.538 ***	0.539 ***	0.544 ***	0.549 ***
	(55.65)	(56.98)	(57.42)	(57.29)
Lev	0.903 ***	0.902 ***	0.901 ***	0.884 ***
	(13.67)	(13.67)	(13.60)	(13.35)
ROA	2.780 ***	2.790 ***	2.759 ***	2.751 ***
	(12.94)	(13.01)	(12.83)	(12.78)
Growth	0.087 ***	0.087 ***	0.085 ***	0.086 ***
	(3.61)	(3.59)	(3.53)	(3.58)
CFROA	0.758 ***	0.746 ***	0.785 ***	0.795 ***
	(4.66)	(4.59)	(4.82)	(4.86)
Invt	−1.252 ***	−1.244 ***	−1.253 ***	−1.236 ***
	(−5.30)	(−5.26)	(−5.31)	(−5.23)
Tobin	−0.021 **	−0.022 **	−0.022 **	−0.022 **
	(−2.28)	(−2.32)	(−2.36)	(−2.39)
Intang	−0.903 ***	−0.897 ***	−0.898 ***	−0.897 ***
	(−4.78)	(−4.76)	(−4.73)	(−4.75)
Age	0.017	0.021	0.019	0.012
	(1.11)	(1.36)	(1.24)	(0.80)
Constant	−4.517 ***	−4.437 ***	−4.578 ***	−4.705 ***
	(−20.04)	(−19.55)	(−20.34)	(−21.22)
Industry	Yes	Yes	Yes	Yes
Year	Yes	Yes	Yes	Yes
N	4985	4985	4985	4985
R2	0.791	0.792	0.791	0.790

Note: t-statistics in parentheses. * p < 0.1 ** p < 0.05 *** p < 0.01.

provides the lag regression results. The coefficients of ESG performance, environment performance and social performance on TFP are significantly positive at the level of 1%, the coefficient of corporate governance performance on TFP is significantly positive at the level of 10%. The lag regression results indicated that our baseline results were relatively robust.

Third, following someone’s research, we selected the mean of ESG performance in the industry minus its own ESG performance as the instrumental variable of ESG performance and adopted the two-stage least square method to further alleviate the endogeneity problem. As can be seen from

Table 8. The results of 2sls IV regression results.

	(1)	(2)
	First Stage	Second Stage
Variable	ESG	TFP
ESG(IV)	3.859 ***
	(4.86)
ESG		0.009 ***
		(4.11)
Size	1.502 ***	0.540 ***
	(20.02)	(63.65)
Lev	−1.350 ***	0.916 ***
	(−2.66)	(18.44)
ROA	−0.997	3.025 ***
	(−0.62)	(19.29)
Growth	−0.208	0.086 ***
	(−1.24)	(5.28)
CFROA	2.862 **	0.810 ***
	(2.29)	(6.62)
Invt	−1.409	−1.370 ***
	(−0.77)	(−7.72)
Tobin	−0.177 **	−0.039 ***
	(−2.42)	(−5.48)
Intang	0.901	−0.971 ***
	(0.59)	(−6.55)
Age	0.055	0.044 ***
	(0.41)	(3.40)
Constant	−30.766 ***	−4.530 ***
	(−14.14)	(−24.44)
Industry	Yes	Yes
Year	Yes	Yes
N	4892	4892
R2	0.507	0.809

Note: t-statistics in parentheses. ** p < 0.05 *** p < 0.01.

, the first column shows the first stage regression result of the 2sls regressions, while the second column shows the second stage IV regression result. The instrumental variable in the first stage was significantly positively correlated with ESG performance at the 1% level. Meanwhile, in the weak instrumental variable test, the value of the statistic F was 118.498, much larger than 10, which proved that this instrumental variable is ideal, and the number of instrumental variables was equal to the number of endogenous variables, and no over-identification test was required. The results of the second stage shows that the positive correlation between ESG performance and enterprise high-quality development still exists significantly. The 2sls IV regression results were basically consistent with the above after controlling for endogeneity.

5.4. Mediating Mechanism Results

The benchmark regression results have shown that the comprehensive ESG performance and the three-dimensional performance can promote the high-quality development of enterprises. To explore the mechanism, we further tested model (3) and model (4). The odd-numbered columns in

Table 9. Mechanism test results based on innovation input.

	(1)	(2)	(3)	(4)	(5)	(6)	(7)	(8)
Variable	Innovation	TFP	Innovation	TFP	Innovation	TFP	Innovation	TFP
ESG	0.0199 ***	0.0035 ***
	(6.99)	(3.00)
E			0.0183 ***	0.0029 ***
			(8.03)	(3.15)
S					0.0059 ***	0.0020 ***
					(3.17)	(2.88)
G							0.0133 ***	0.0016
							(3.72)	(1.13)
Innovation		0.0803 ***		0.0798 ***		0.0812 ***		0.0815 ***
		(11.86)		(11.78)		(12.04)		(12.08)
Size	0.8182 ***	0.4860 ***	0.8154 ***	0.4863 ***	0.8464 ***	0.4876 ***	0.8432 ***	0.4902 ***
	(37.91)	(49.92)	(37.88)	(50.26)	(40.26)	(50.11)	(39.70)	(50.00)
Lev	−0.4324 ***	0.9143 ***	−0.4282 ***	0.9143 ***	−0.4596 ***	0.9143 ***	−0.4780 ***	0.9066 ***
	(−3.32)	(17.33)	(−3.29)	(17.32)	(−3.52)	(17.30)	(−3.65)	(17.19)
ROA	1.6332 ***	2.7939 ***	1.6818 ***	2.8018 ***	1.5626 ***	2.7811 ***	1.5886 ***	2.7825 ***
	(3.78)	(16.50)	(3.90)	(16.54)	(3.62)	(16.45)	(3.69)	(16.43)
Growth	−0.0049	0.0876 ***	−0.0060	0.0874 ***	−0.0084	0.0871 ***	−0.0041	0.0876 ***
	(−0.13)	(4.80)	(−0.16)	(4.78)	(−0.23)	(4.77)	(−0.11)	(4.81)
CFROA	1.0467 ***	0.7439 ***	1.0029 ***	0.7384 ***	1.1426 ***	0.7558 ***	1.1401 ***	0.7601 ***
	(3.04)	(5.66)	(2.92)	(5.61)	(3.31)	(5.74)	(3.31)	(5.76)
Invt	1.4019 ***	−1.5198 ***	1.4284 ***	−1.5146 ***	1.4040 ***	−1.5233 ***	1.4575 ***	−1.5138 ***
	(2.84)	(−8.48)	(2.89)	(−8.44)	(2.83)	(−8.51)	(2.94)	(−8.45)
Tobin	0.0277	−0.0398 ***	0.0270	−0.0399 ***	0.0254	−0.0401 ***	0.0262	−0.0402 ***
	(1.34)	(−5.29)	(1.30)	(−5.31)	(1.23)	(−5.33)	(1.27)	(−5.34)
Intang	0.2567	−0.9857 ***	0.2743	−0.9822 ***	0.2892	−0.9831 ***	0.2851	−0.9801 ***
	(0.63)	(−6.23)	(0.67)	(−6.22)	(0.70)	(−6.20)	(0.70)	(−6.20)
Age	−0.2055 ***	0.0574 ***	−0.1937 ***	0.0592 ***	−0.2027 ***	0.0587 ***	−0.2261 ***	0.0550 ***
	(−6.52)	(4.48)	(−6.16)	(4.63)	(−6.40)	(4.58)	(−7.00)	(4.23)
Constant	−1.6656 ***	−4.5360 ***	−1.3456 ***	−4.4890 ***	−1.9862 ***	−4.5676 ***	−2.3618 ***	−4.6429 ***
	(−6.52)	(4.48)	(−6.16)	(4.63)	(−6.40)	(4.58)	(−7.00)	(4.23)
Industry	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes
Year	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes
N	4985	4985	4985	4985	4985	4985	4985	4985
R2	0.625	0.819	0.626	0.819	0.622	0.817	0.622	0.818
Bootstrap 95% confidence interval	(0.0208, 0.0289)		(0.0189, 0.0270)		(0.0104, 0.0167)		(0.0079, 0.0131)

Note: t-statistics in parentheses. *** p < 0.01.

reflect the impacts of ESG comprehensive performance and the three-dimensional performance on corporate innovation input. All coefficients are significantly positive, indicating that good ESG comprehensive performance and three-dimensional performance help to promote the increase of enterprise innovation input. The double-number columns in Table 9 provides the test results of model (4).

When the explanatory variable is ESG performance, the impact coefficients β₁ and γ₂ of ESG performance and innovation input on TFP are significantly positive. After 1000 repeated samplings using the Bootstrap method, the confidence interval for β₁γ₂ at the 95% level did not cover 0. Moreover, γ₁ is significantly positive, and β₁γ₂ has the same sign as γ₁, suggesting that innovation input plays a mediating effect between ESG comprehensive performance and TFP. The results supported hypothesis H1.

When the explanatory variables are environmental performance and social performance, the results in columns (3)−(6) also show that corporate innovation input mediates the effect between environmental performance and TFP as well as social performance and TFP. Thus, our hypothesis 1 H2a and H2b were verified. Table 9 columns (7) and (8) show the results of model (3) and model (4) for governance performance as the explanatory variable respectively. Although the coefficient γ₁ was not significant, the confidence interval corresponding to the direct effect and indirect effect did not contain 0. Therefore, we can conclude that there is a mediation effect of innovation input between governance performance and the high-quality development of enterprises, which supports the hypothesis H2c.

5.5. Additional Tests

Based on the specific institutional background of Chinese enterprises, we further explore the relationship between ESG expression and TFP as follows:

First, we consider the impact of corporate property rights. The results are shown in

Table 10. Additional test based on property rights.

	(1)	(2)	(3)	(4)	(5)	(6)	(7)	(8)
	SOE	PE	SOE	PE	SOE	PE	SOE	PE
Variable	TFP	TFP	TFP	TFP	TFP	TFP	TFP	TFP
ESG	0.005 ***	0.003 **
	(3.35)	(2.04)
E			0.006 ***	0.002
			(4.69)	(1.47)
S					0.001 **	0.003 **
					(2.05)	(2.45)
G							0.003	0.002
							(1.59)	(0.86)
Size	0.539 ***	0.584 ***	0.535 ***	0.586 ***	0.549 ***	0.585 ***	0.548 ***	0.588 ***
	(50.74)	(50.34)	(50.77)	(51.04)	(53.17)	(51.47)	(54.21)	(50.54)
Lev	0.817 ***	0.945 ***	0.826 ***	0.944 ***	0.797 ***	0.949 ***	0.792 ***	0.946 ***
	(11.37)	(13.94)	(11.53)	(13.92)	(11.10)	(14.00)	(11.08)	(13.94)
ROA	3.203 ***	2.598 ***	3.233 ***	2.603 ***	3.151 ***	2.593 ***	3.171 ***	2.599 ***
	(13.02)	(13.15)	(13.16)	(13.17)	(12.81)	(13.13)	(12.87)	(13.14)
Growth	0.122 ***	0.061 ***	0.121 ***	0.060 ***	0.122 ***	0.060 ***	0.123 ***	0.060 ***
	(4.78)	(3.00)	(4.77)	(2.98)	(4.79)	(2.98)	(4.81)	(2.97)
CFROA	0.613 ***	1.099 ***	0.588 ***	1.100 ***	0.643 ***	1.117 ***	0.645 ***	1.113 ***
	(3.37)	(7.00)	(3.23)	(7.00)	(3.53)	(7.13)	(3.55)	(7.09)
Invt	−1.566 ***	−1.358 ***	−1.557 ***	−1.354 ***	−1.563 ***	−1.350 ***	−1.530 ***	−1.352 ***
	(−5.53)	(−6.07)	(−5.51)	(−6.06)	(−5.50)	(−6.04)	(−5.39)	(−6.04)
Tobin	−0.054 ***	−0.022**	−0.055 ***	−0.022**	−0.055 ***	−0.023**	−0.054 ***	−0.023 **
	(−4.71)	(−2.51)	(−4.73)	(−2.53)	(−4.72)	(−2.54)	(−4.66)	(−2.56)
Intang	−0.658 ***	−1.297 ***	−0.666 ***	−1.298 ***	−0.622 ***	−1.304 ***	−0.630 ***	−1.310 ***
	(−3.16)	(−5.97)	(−3.20)	(−5.97)	(−2.98)	(−6.01)	(−3.01)	(−6.03)
Age	0.137 ***	−0.050 ***	0.137 ***	−0.048 ***	0.135 ***	−0.048 ***	0.134 ***	−0.054 ***
	(5.99)	(−2.88)	(6.01)	(−2.76)	(5.91)	(−2.75)	(5.88)	(−3.04)
Constant	−4.394 ***	−5.364 ***	−4.252 ***	−5.349 ***	−4.517 ***	−5.388 ***	−4.637 ***	−5.435 ***
	(−14.41)	(−20.69)	(−13.85)	(−20.59)	(−14.78)	(−20.79)	(−15.43)	(−20.88)
Industry	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes
Year	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes
N	2427	2558	2427	2558	2427	2558	2427	2558
R2	0.821	0.803	0.822	0.803	0.820	0.803	0.820	0.803
Mean difference test	chi2(1) = 2.98		chi2(1) = 3.38		chi2(1) = 1.72		chi2(1) = 0.30
	Prob > chi2 = 0.0889		Prob > chi2 = 0.0661		Prob > chi2 = 0.1896		Prob > chi2 = 0.5830

Note: t-statistics in parentheses. ** p < 0.05 *** p < 0.01. SOE refers to state-owned enterprises, PE refers to private enterprises.

. No matter what the property rights of enterprises, the coefficient between ESG performance and TFP is significantly positive. But the inter-group coefficient difference is significant at the level of 10%. This indicates that compared with private enterprises, the ESG performance of state-owned enterprises can better promote high-quality development. Meanwhile, this promoting effect is mainly reflected in the environmental performance. The reasons for the results include two aspects. On the one hand, most of Chinese state-owned enterprises are from heavily polluting and environmentally sensitive industries such as petroleum, chemical, mining, metallurgy, machinery manufacturing and so on [67]. They are under more pressure than private companies to be environ-mentally responsible [28]. And they generally lack innovation capacity [54]. Thus, Good ESG performance will encourage them to focus on improving their innovation capabilities, so as to achieve high-quality development. On the other hand, compared with private enterprises, the natural property right relationship between state-owned enterprises and the government is conducive to long-term loans, preferential policies and other support for state-owned enterprises [29]. It means that they face less funding constraints and have more funds to practice ESG activities and increase investment in innovation. Accordingly, the promotion effect of ESG performance on high-quality development is more obvious.

Second, we conduct a grouping test based on their environmental sensitivity. According to the classified management directory of environmental protection verification industry of Listed Companies formulated by Chinese Ministry of environmental protection in 2008, the sample are divided into environmentally sensitive enterprises and environmentally insensitive enterprises. From the results in

Table 11. Additional test based on environmental sensitivity.

	(1)	(2)	(3)	(4)	(5)	(6)	(7)	(8)
	High ES	Low ES	High ES	Low ES	High ES	Low ES	High ES	Low ES
Variable	TFP	TFP	TFP	TFP	TFP	TFP	TFP	TFP
ESG	0.011 ***	0.001
	(5.25)	(1.03)
E			0.009 ***	0.002
			(5.46)	(1.41)
S					0.003 **	0.002 **
					(2.26)	(2.35)
G							0.010 ***	0.002
							(3.95)	(1.02)
Size	0.521 ***	0.575 ***	0.521 ***	0.574 ***	0.538 ***	0.574 ***	0.535 ***	0.581 ***
	(40.54)	(61.60)	(41.08)	(62.40)	(42.30)	(63.92)	(42.65)	(63.39)
Lev	0.551 ***	1.089 ***	0.547 ***	1.089 ***	0.518 ***	1.091 ***	0.509 ***	1.088 ***
	(6.11)	(16.18)	(6.10)	(16.19)	(5.70)	(16.22)	(5.71)	(16.15)
ROA	2.555 ***	3.051 ***	2.570 ***	3.056 ***	2.531 ***	3.044 ***	2.511 ***	3.035 ***
	(8.32)	(14.39)	(8.37)	(14.41)	(8.20)	(14.41)	(8.19)	(14.31)
Growth	0.114 ***	0.073 ***	0.113 ***	0.073 ***	0.116 ***	0.072 ***	0.114 ***	0.071 ***
	(3.46)	(3.10)	(3.45)	(3.10)	(3.61)	(3.08)	(3.51)	(3.03)
CFROA	0.649 ***	0.939 ***	0.611 ***	0.937 ***	0.701 ***	0.941 ***	0.706 ***	0.953 ***
	(2.81)	(5.89)	(2.64)	(5.88)	(3.01)	(5.90)	(3.03)	(5.96)
Invt	−0.193	−2.084 ***	−0.182	−2.082 ***	−0.235	−2.087 ***	−0.163	−2.076 ***
	(−0.57)	(−9.33)	(−0.53)	(−9.32)	(−0.69)	(−9.36)	(−0.47)	(−9.26)
Tobin	−0.031 **	−0.030 ***	−0.031 **	−0.030 ***	−0.034 **	−0.029 ***	−0.030*	−0.030 ***
	(−2.03)	(−3.72)	(−1.99)	(−3.73)	(−2.13)	(−3.68)	(−1.95)	(−3.76)
Intang	−1.040 ***	−0.795 ***	−1.041 ***	−0.793 ***	−1.032 ***	−0.805 ***	−1.050 ***	−0.787 ***
	(−4.11)	(−3.68)	(−4.13)	(−3.68)	(−4.04)	(−3.72)	(−4.21)	(−3.66)
Age	0.031	0.045 ***	0.038	0.046 ***	0.033	0.046 ***	0.016	0.048 ***
	(1.19)	(3.03)	(1.46)	(3.11)	(1.26)	(3.09)	(0.61)	(3.10)
Constant	−3.197 ***	−5.180 ***	−3.256 ***	−5.150 ***	−3.479 ***	−5.175 ***	0.535 ***	0.581 ***
	(−10.57)	(−24.83)	(−10.98)	(−24.34)	(−11.57)	(−25.18)	(−13.41)	(−25.19)
Industry	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes
Year	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes
N	1749	3236	1749	3236	1749	3236	1749	3236
R2	0.788	0.839	0.788	0.839	0.785	0.839	0.786	0.839
Mean difference test	chi2(1) = 14.24		chi2(1) = 14.10		chi2(1) = 0.36		chi2(1) = 14.62
	Prob > chi2 = 0.0002		Prob > chi2 = 0.0002		Prob > chi2 = 0.5468		Prob > chi2 = 0.0001

Note: t-statistics in parentheses. ** p < 0.05 *** p < 0.01. ES is the abbreviation of environmental sensitivity.

, the promotion effect of ESG comprehensive performance on the high-quality of environmentally sensitive enterprises is more obvious, and this promotion effect is mainly reflected in the environmental performance and corporate governance. The reason is that environmentally sensitive face stricter environmental regulation and industry regulation. This will inevitably prompt them to reduce pollutant emissions by purchasing environmental protection facilities, improving environmental protection technology and, thus promoting high-quality development of enterprises [45,55]. Moreover, their internal governance mechanism will organically cooperate with external environmental policies, thus promoting enterprise green technology innovation and laying a foundation for the improvement of TFP.

Third, we compare the influence of financial constraints. Following Zhang et al. [26], SA index is used to measure financial constraints. We divide the sample into high and low groups based on mean value. The results in

Table 12. Additional test based on financing constraints.

	(1)	(2)	(3)	(4)	(5)	(6)	(7)	(8)
	High FC	Low FC	High FC	Low FC	High FC	Low FC	High FC	Low FC
Variable	TFP	TFP	TFP	TFP	TFP	TFP	TFP	TFP
ESG	0.003 *	0.008 ***
	(1.86)	(4.32)
E			0.003 **	0.006 ***
			(2.45)	(4.49)
S					0.001	0.004 ***
					(0.64)	(3.93)
G							0.002	0.004 **
							(1.13)	(2.05)
Size	0.540 ***	0.543 ***	0.539 ***	0.543 ***	0.546 ***	0.548 ***	0.544 ***	0.551 ***
	(49.49)	(38.27)	(50.18)	(38.34)	(52.45)	(39.09)	(50.27)	(38.97)
Lev	0.844 ***	0.895 ***	0.848 ***	0.893 ***	0.839 ***	0.898 ***	0.839 ***	0.878 ***
	(10.14)	(12.50)	(10.17)	(12.48)	(10.09)	(12.50)	(10.11)	(12.20)
ROA	3.014 ***	3.047 ***	3.027 ***	3.061 ***	3.003 ***	3.038 ***	3.014 ***	3.035 ***
	(11.38)	(13.56)	(11.41)	(13.62)	(11.38)	(13.50)	(11.38)	(13.46)
Growth	0.080 ***	0.089 ***	0.079 ***	0.089 ***	0.080 ***	0.087 ***	0.081 ***	0.086 ***
	(3.07)	(3.60)	(3.03)	(3.58)	(3.07)	(3.51)	(3.14)	(3.49)
CFROA	0.769 ***	0.980 ***	0.755 ***	0.980 ***	0.786 ***	1.004 ***	0.782 ***	1.007 ***
	(3.66)	(5.73)	(3.59)	(5.72)	(3.75)	(5.84)	(3.72)	(5.84)
Invt	−1.171 ***	−1.632 ***	−1.170 ***	−1.607 ***	−1.178 ***	−1.639 ***	−1.156 ***	−1.633 ***
	(−4.16)	(−6.39)	(−4.15)	(−6.28)	(−4.18)	(−6.41)	(−4.11)	(−6.34)
Tobin	−0.041 ***	−0.043 ***	−0.040 ***	−0.044 ***	−0.041 ***	−0.042 ***	−0.041 ***	−0.044 ***
	(−3.45)	(−4.56)	(−3.42)	(−4.67)	(−3.50)	(−4.52)	(−3.45)	(−4.70)
Intang	−0.452 **	−1.300 ***	−0.446 **	−1.292 ***	−0.457 **	−1.306 ***	−0.471 **	−1.249 ***
	(−2.06)	(−5.34)	(−2.04)	(−5.33)	(−2.09)	(−5.32)	(−2.15)	(−5.17)
Age	0.104 ***	0.035*	0.105 ***	0.041 **	0.102 ***	0.040 **	0.100 ***	0.029
	(5.21)	(1.84)	(5.30)	(2.14)	(5.12)	(2.09)	(5.02)	(1.48)
Constant	−4.203 ***	−4.883 ***	−4.139 ***	−4.777 ***	−4.264 ***	−4.964 ***	−4.321 ***	−5.100 ***
	(−15.89)	(−15.05)	(−15.55)	(−14.64)	(−16.26)	(−15.23)	(−16.59)	(−15.33)
Industry	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes
Year	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes
N	2374	2611	2374	2611	2374	2611	2374	2611
R2	0.857	0.742	0.858	0.742	0.857	0.741	0.857	0.740
Mean difference test	chi2(1) = 3.64		chi2(1) = 2.24		chi2(1) = 6.63		chi2(1) = 0.44
	Prob > chi2 = 0.0563		Prob > chi2 = 0.1342		Prob > chi2 = 0.0100		Prob > chi2 = 0.5082

Note: t-statistics in parentheses. * p < 0.1 ** p < 0.05 *** p < 0.01. FC refers to financing constraints.

indicate that the comprehensive ESG performance of enterprises with small financing constraints has a stronger promoting effect on TFP. And this effect is mainly reflected in the social performance. The reason is that enterprises subject to large financing constraints aren’t able to have sufficient funds to undertake environmental and social responsibilities [24,25]. And compared with environmental performance and corporate governance, the reputation effect of performance social responsibility can better help them obtain certain funds, which is conducive to promoting high-quality development through purchasing equipment and hiring high-technical personnel [54,56].

6. Discussion

ESG activities have great influence on corporate risk-taking, firm performance, and even the healthy development of capital market. From the perspective of economic and business sustainability, attaching importance to ESG practice or engaging in ESG activities is a key factor for the high-quality development of enterprises. In this research, we examined how an enterprise’s ESG performance affected its quality development in the context of the growing popularity of the ESG concept. Using the data of Chinese A-share listed companies in Shanghai and Shenzhen from 2011 to 2019, we found that ESG performance could promote the high-quality development of Chinese listed companies. our results were still robust after replacing the measures of high-quality development of enterprises and excluding some samples. To alleviate the endogenous issue, we adopted three methods, adding the control variable at the regional level, lag regressions and instrumental variable regressions. The results of the mediation effect analysis proved that enterprises attaching importance to ESG performance can increase innovation input, thereby promoting their high-quality development. In addition, for state-owned enterprises, environmentally sensitive enterprises and enterprises with small financing constraints, ESG performance was more conducive to the high-quality development of enterprises.

This study enriches the literature on the factors influencing the high-quality development of enterprises, and extends the literature on the economic consequences of ESG performance. This research provides the empirical evidence for the importance of paying attention to ESG practice or engaging in ESG activities. Overall, ESG performance is an indispensable factor for enterprises to achieve high-quality development. Additionally, from the perspective of enterprise innovation input, this paper provides insight for how to accelerate the realization of high-quality enterprise development. Our research results will also encourage companies to implement ESG concepts and persuade government departments to formulate and improve relevant policies.

There are some research limitations. First of all, we use the total factor productivity as a proxy for high-quality development of enterprises. When calculating specific value, we adopt the LP method in the main test and the OLS method in the robustness check. Although the two calculation methods have a certain role of substitution, the indicator of total factor productivity cannot fully measure the high-quality development of enterprises. In the future research, a more appropriate indicator should be explored to make the conclusions more convincing. Secondly, in this paper, we only study the impact mechanism of ESG performance on enterprise high-quality development from the perspective of innovation input. There are other channels which ESG performance could influence the high-quality development of enterprises. Thus, future research could explore more mediation mechanisms. Besides, the empirical results are based on Chinese enterprises, which may not be applied to other countries and regions. A comparative study between different countries and regions is an explorable direction in the future.

7. Conclusions

Based on the Shanghai and Shenzhen A-share listed companies from 2011 to 2019 as the research object, this paper empirically studies the impact of ESG performance on the high-quality development of enterprises and explores the mediating effect of innovation put in such relationship. Meanwhile, we further study whether this impact is different for companies with different property rights, environmental sensitivity and financing constraints. The research conclusions are as follows:

(1)

There is a positive relationship between ESG performance and the high-quality development of enterprises. Compared with governance performance, the positive impacts of environmental performance and social performance on the high-quality development of enterprises are more significant. Therefore, ESG performance is conductive to the high-quality development of enterprises, and environmental performance and social performance have a stronger role in promoting the high-quality development of enterprises than governance performance.

(2)

The results of the mechanism analysis shows that the increased investment in corporate innovation brought by good ESG performance is a specific internal channel for high-quality development of enterprises. Thus, good ESG performance can increase the innovation input, thereby further improving the total factor productivity and promoting their high-quality development of enterprises.

(3)

The impact of ESG performance on the high-quality development of enterprises varies among enterprises with different property rights, environmental sensitivity and financing constraints. Specifically, ESG performance has a more obvious role in promoting high-quality development in state-owned enterprises, environmentally sensitive enterprises and enterprises with less financing constraints.