ESG Performance, Research and Development Investment and Enterprise Green Technology Innovation

. Based on the panel data of China's Shanghai and Shenzhen A-share industrial listed companies from 2014 to 2021, this paper constructs A fixed-effect model to investigate the influence of ESG performance on enterprises' green technology innovation, and constructs a mediating effect model to explore the mediating role of R&D investment in the influence of ESG performance on enterprises' green technology innovation. The results show that good ESG performance has a significant promoting effect on enterprises' green technology innovation. And there is an intermediary path for R&D investment. It means that ESG promotes enterprise green technology innovation through R&D investment. Finally, the conclusions of this paper passed the robustness test and endogeneity test. This paper opens up a new research perspective and provides theoretical and empirical experience for enterprises to achieve sustainable development in the future.


Introduction
Innovation is the source of economic development. Since the reform and opening up, China's economy has achieved rapid growth along with innovation activities. At the same time, the over exploitation and utilization of natural resources have also caused serious ecological problems. With the introduction of the two-carbon goal, China should actively promote green development to achieve both economic and environmental benefits. Traditional innovation activities don't take environmental factors into account. As an important subject and key actor to achieve green development [1], enterprises need to combine "green" and "innovation" to carry out green technology innovation. However, green technology innovation activities have higher investment risk, which make enterprises face difficulties such as information asymmetry and financing constraints. And such activities lead to insufficient investment in R&D and restrict enterprises' green technology innovation activities. Therefore, the academic circle has carried out a series of discussions and researches on how to effectively promote enterprises to carry out green technology innovation activities. At present, factors affecting enterprises' green technology innovation mainly include external factors such as government subsidies[2], environmental regulations [3], media supervision[4], green finance [5] and internal factors such as tenure of senior managers, age, gender and education level [6]. However, with the rise of the concept of sustainable development, more and more enterprises pay more attention to the building of their own image, and begin to disclose environmental, social and governance information in the social responsibility report, so as to show the public the performance of their social responsibilities. So, will ESG performance have an impact on enterprises' green technology innovation activities?
According to the existing literature in the academic circle, the impact of ESG performance on enterprise value and business performance[7] is the focus of scholars, and few studies have explored the impact of ESG performance from the perspective of green technology innovation. On the one hand, good ESG performance can help enterprises gain support from external institutional investors and obtain funds for green innovation. On the other hand, ESG information disclosure requires enterprises to pay additional costs, which will squeeze their own funds and limit their green technology innovation activities. Therefore, it is particularly important to explore the impact of ESG performance on green technology innovation. In addition, the availability of capital plays a key role in enterprises' green technology innovation activities, and continuous R&D investment helps improve enterprises' technological innovation ability. Therefore, whether R & D investment plays a role in the path of ESG performance affecting green technology innovation is worth further discussion in this paper.

Theoretical analysis and research hypothesis
Under pressure from public media and interest stakeholders, companies will actively engage in behavior that meets social expectations and peer standards. With the popularity of green concept, enterprises will take the initiative to carry out green technology innovation to reduce pollution emissions. At this point, if the enterprise has a good ESG performance, on the one hand, the debt financing cost of the enterprise will be reduced [8], and on the other hand, the expectation of stakeholders will be satisfied. Investors will be more willing to provide stable financial support for such enterprises[9] and support enterprises to carry out green technology innovation. Moreover, the quality of financial information disclosed by the company will gradually improve as the company undertakes more social responsibilities [10]. More transparency in the use of corporate funds will lead to managers' less use of corporate resources and more efficient use of funds. At this point, money is more likely to be invested in green R&D activities. Finally, as the reserve force of innovation activities, R&D personnel need to improve the performance of enterprise ESG to effectively enhance the sense of identity and belonging of employees [11], so as to reduce the damage caused by staff turnover to green technology innovation activities. Therefore, the following hypothesis is proposed.
Hypothesis H1. Good ESG performance can promote enterprises to carry out green technology innovation.
The daily operation and development of enterprises should not only pursue the maximization of benefits, but also meet the needs of stakeholders. External institutional investors provide financial resources for enterprises [12] and will actively pay attention to the ESG performance of enterprises to judge investment risks and make investment decisions. If an enterprise has a good ESG investment strategy, internal employees will also increase their sense of corporate cultural identity, actively promote internal innovation mechanism, and promote R&D investment [13]. Then if the enterprise's ESG score is high, the positive signal of good operation and sufficient capital will be released in the capital market, and the trust of institutional investors will be absorbed, so as to chase up corporate investment and improve the cash flow of R&D input [14]. Moreover, R&D investment itself is a kind of investment behavior, which inevitably needs a certain scale of funds. At this point, good ESG performance of enterprises will enhance investors' investment willingness, thus increasing the cash flow of R&D investment. Based on this, the following hypothesis is proposed.
Hypothesis H2. Good ESG performance can promote enterprises to invest in research and development.
On the one hand, a good ESG performance of an enterprise can effectively enhance its reputation, enhance its competitiveness and avoid financing difficulties, thus ensuring that an enterprise has enough funds for R&D investment. On the other hand, the increase of R&D investment is conducive to innovation performance [15]. Specifically, R&D can be regarded as "input" and patent as "output". The direct result of R&D is the acquisition of patented technology, so green technology innovation can be regarded as a function of R&D input. At this point, continuous R&D investment is conducive to obtaining a higher number of patents and improving the technological innovation ability of enterprises. The innovation performance of an enterprise is also related to its ability to obtain and utilize resources. Thus, the following hypothesis is proposed.
Hypothesis H3. Good ESG performance can promote enterprises' green technology innovation through the intermediary path of promoting R&D investment.

Research Samples and Data Sources
China's A-share industrial listed companies from 2014 to 2021 are selected as the research object, and the samples are processed as follows:ST, ST* and other companies with unstable business conditions are excluded, and samples with incomplete data information and cannot be supplemented. The main continuous variables are Winsorize at the upper and lower 1% level. Finally, 9688 observations were obtained from 1211 sample data. Data sources in this paper are as follows: patent data comes from CNRDS database, ESG rating data comes from Wind database, and other control variable data comes from CSMAR database.

Variable Design
The core variables and control variables of this paper are shown in Table 1.

G p a ten t E S G C o n tro ls YE A R R D E S G C o n tro ls YE A R G p a ten t E S G R D C o n tro ls YE A R
For the above model, i and t indicate enterprise and year. Gpatent and ESG respectively represent enterprise green technology innovation and ESG performance. RDln stands for corporate R&D investment. Controls are a series of control variables. In addition, ∑ YEAR indicates a time-fixed effect, and ξ indicates random disturbance term. As shown in Table 2, there is a big gap in green patents among enterprises. The average ESG is 4.13, and the average ESG level of the sample enterprises is generally low. Differences in R&D levels between companies are also evident. The sample enterprises selected in this paper have a wide age distribution. The standard deviations of other control variables are all around 1, indicating that the fluctuation range of data is small and the selection of sample data is relatively stable.

Baseline Results
In this paper, the fixed effect model is adopted and the benchmark regression analysis is carried out according to formulas (1) to (3). The regression results are shown in Table 3. Notes: standard error in parentheses; *p < 0.1, **p < 0.05, ***p < 0.01. The regression results in column (1) show that the regression coefficient of ESG is significantly positive at the level of 1%, indicating that the improvement of ESG performance helps promote enterprises to carry out green technology innovation. In column (2), the regression coefficient of ESG to RDln is significantly positive at the level of 1%, indicating that good corporate ESG performance can effectively promote R&D investment.
In column (3), the regression coefficients of ESG and RDln are also significantly positive. The regression coefficient of ESG in column (3) decreased by 0.0041 compared with column (1), so the influence of enterprise ESG on enterprise Gpatent decreased after the addition of RDln variable, indicating that enterprise R&D played a partial mediating role in the promotion effect of ESG performance on enterprise green technology innovation. Therefore, hypothesis 1, 2 and 3 are verified by threestep regression method.
As for the control variable,the age coefficient of enterprises is significantly negative, indicating that the green technology innovation ability of newly listed enterprises is stronger. This is because newly listed companies are more likely to learn and accept the latest science and technology. The size coefficient is significantly positive. Larger enterprises have sufficient capital base, which is conducive to green research and development. The Top10 coefficient is negative, which may be because the lower the ownership concentration, the more effectively the principal-agent conflicts among stakeholders can be alleviated. TobinQ coefficient is significantly positive, indicating that the higher the value of the company, the stronger the sense of innovation.

Sobel Test
This paper adds Sobel test to prove that R&D investment has a significant mediating effect between ESG performance and green technology innovation. Table 4 shows that Sobel's Z value is 4.940 and P value is less than 0.05, indicating that the null hypothesis is rejected. The intermediary effect of R&D investment exists, and it accounts for 16% of the total effect.

Substitution Variable Method
In this paper, the proportion of green invention patent applications is used as an alternative index for robustness test. As shown in Table 5, the regression results are basically consistent with the previous ones.  This paper refers to Benlemlih&Bitar's method[17] to calculate the mean value of ESG ratings of all listed companies in the province to which the listed company belongs except the company itself. As an instrumental variable, two-stage least square method was used for regression. The final results are shown in Table 6.It indicates that the regression results are relatively robust.

Conclusions and Suggestions
This paper explores the effect and conduction path of enterprise ESG performance on green technology innovation, and draws the conclusion that good ESG performance can significantly promote enterprises to carry out green technology innovation, and R&D investment plays a significant mediating role in this incentive path. The research in this paper enriches the relevant literature on the influencing factors of enterprise green technology innovation. It provides a new perspective for the study of green technology innovation under the background of sustainable development.
The policy suggestions are as follows: First, the government should introduce a series of preferential policies to encourage enterprises, such as tax relief. In addition, a more standardized ESG evaluation system and a more complete ESG information disclosure system should be established to promote high-quality development of enterprises. Secondly, enterprises themselves must realize the importance of green technology innovation for future sustainable development, and actively assume the responsibility of environmental protection and social responsibility. And enterprises should improve the level of ESG to attract external investment to promote innovation activities.