1. Introduction
With the continuous emergence of environmental problems, the international voice for sustainable development is increasing. High-emission industries not only lead to greenhouse effects and ecological deterioration but also overdraft the long-term vitality of the national economy [
1]. The importance of green innovation has been widely accepted by governments. At the seventy-fifth session of the United Nations General Assembly, the United States proposed to achieve ‘net zero emissions’ by 2050, and the European Union proposed to become the first ‘climate neutral’ continent in the same year. China, as the largest developing country, also faces severe sustainable difficulties. According to BP World Energy Statistics Yearbook, China’s total carbon emissions in 2020 is 9899.3 million tons, accounting for about 30.7% of the total global emissions, ranking first in the world. In order to solve the environmental problem, China’s State Council issued the State Council Guidance on Accelerating the Establishment and Perfection of a Green Low-carbon Recycling Economic System in 2021, which clearly pointed out that we should adhere to the working principle of leading by innovation and improve the efficiency of energy allocation through the construction of green technology innovation, the making of laws and regulations support system and the building of green supply chain to pilot with some enterprises. The promotion from the national level makes improving the level of green innovation a popular topic discussed by scholars.
At present, scholars’ research on green innovation mainly focuses on measuring methods and influencing factors. At the level of measuring methods, green innovation is measured by efficiency, such as enterprise R&D efficiency and achievement transformation efficiency [
2,
3]. On the other hand, a very important branch of literature is to evaluate the influencing factors of green innovation including putting forward a variety of research perspectives. Based on the perspective of the market, some scholars discussed the common expected growth of market share related to new market segments and consumers’ demand preference for green products [
4,
5]. Based on the perspective of the government, some scholars proposed that the government required enterprises to promote green innovation and development through administrative and legal means. However, due to the lack of financial capacity and other reasons, most enterprises lacked the motivation to carry out innovation practice [
6]. In order to share the pressure of strict regulation, some enterprises began to turn to establishing effective political connections [
7,
8,
9]. When scholars study green innovation in the context of China’s economy, they must take into account the characteristics of China’s special ownership and weak market mechanism, which is different from Western countries: political connection has become an indispensable factor in this context. In the study of the role of political connection on green innovation, it has been divided into three categories: promotion, inhibition, and U-shaped. Scholars who promote the theory believe that political connections can effectively minimize the risk of enterprise innovation and enhance confidence in enterprise green development [
10,
11,
12]. Zhang et al. [
13] analyzed the promoting effect of political connections from the perspective of entrepreneurial strategy. However, some studies have shown that political connections can also have a negative impact because in order to establish good political connections, enterprises need to pay more rent-seeking costs; this produces the ‘crowding-out effect’ of innovation resources [
14]. At the same time, scholars have found that political connections in different types of enterprises also have different effects. For example, political connections in non-state-owned enterprises have an inverted U-shaped effect on enterprise innovation [
15,
16,
17]. There are many discussions on the relationship between political connections and green innovation in the existing literature, but no consensus has been reached. In order to enrich the theoretical research in this field, this study will find out how political connections affect green technology innovation.
Compared with the previous studies, there are three innovation points in this paper. Firstly, this study attempts to examine the correlation between political connections and green innovation. This paper uses the panel data of all A-share listed companies on the Shanghai and Shenzhen stock markets from 2008 to 2019 as the original sample. During this period, China, the largest developing country in the world, has rapid economic growth but not a perfect market system. Moreover, China is in the process of reforming and upgrading its system, and it does take a certain amount time to improve the corresponding system, which stimulates the desire of enterprises to seek political connections [
9,
18]. China’s economic system is dominated by public ownership which resulted in the government’s irreplaceable role in economic development; therefore, the role of political connection is particularly important for enterprises. A huge amount of money and time will be spent by the enterprises to establish political connections each year [
19]. The above shows that the particularity of Chinese samples during the above time period gives more profound significance to the study of political connections and green innovation than in other Asian countries.
Secondly, this study aims to solve the potential heterogeneous results brought by Chinese samples in different regions and industries. The industrial distribution is not the same in the east and west regions of China because of the current economic pattern and the unbalanced development. The industries are mainly energy, chemical industry and agriculture in the western regions, while the manufacturing and service industries in the east [
20,
21]. The study predicts that political connections have different degrees of impact on green innovation of enterprises in different regions and industries. Thirdly, the study attempts to examine the specific mediating role of R&D investment (RDexp) and excessive debt (
EXLEVBit) in the influencing mechanism, and further discuss the influencing mechanism. Most of the previous literature focuses on the role of political connections in green innovation. In contrast, there are few discussions on the impact mechanism. This paper hopes to broaden the research perspective in this field by studying the mediating effect of different dimensions.
The division of this paper is as follows. The
Section 2 reviews the literature and proposes research hypotheses. The
Section 3 introduces the research methods and establishes the research model. The
Section 4 reports the empirical research results.
Section 5 conducts robustness analysis.
Section 6 draws the conclusion.
3. Sample Selection and Empirical Design
3.1. Sample Selection and Data Source
Taking into account the availability of various indicators and the sample representation of enterprises, the paper uses all A-share listed companies on the Shanghai and Shenzhen stock exchanges from 2008 to 2019 as the original sample. The original sample data of listed companies are from GuoTaiAn CSMAR database, and the data are merged. Further screening process of the samples is as follows. Firstly, since the financial statements of enterprises in the financial industry are quite different from those of other enterprises, the paper selects them according to the CSMAR industry code and excludes the differences in the financial statements of samples in the financial industry. Secondly, it excludes the sample of ST company during the study period. Thirdly, it excludes the sample of asset–liability ratios greater than 1; fourthly, in order to eliminate the influence of extreme value on empirical analysis, all continuous variables are processed by upper and lower 1% winsorize. Finally, it deletes the enterprise samples without the needed relevant variables.
3.2. The Green Innovation of Enterprises
Patent is the most important measure of enterprise innovation activities and output. Based on this theory, the paper uses the natural logarithm lnGPall, which is the sum of green invention patent authorization and green utility model patent authorization of listed companies in the year, to measure the output of innovation activities in the green environment of listed companies. These green patent data of listed companies come from the China National Intellectual Property Administration.
3.3. Political Connections of the Enterprises
Referring to the standard practice of the existing literature, the paper measures the political connection of enterprises by the experience of executives in listed companies; it constructs a binary variable enterprise political connection (PC): if either the chairman or the general manager of the enterprise, or both, is now or has been in the government department, the PC value is 1, otherwise it is 0.
3.4. Empirical Model Setting
In order to verify the above theoretical assumptions, the paper intends to construct a two-way fixed-effect model to discuss the influence of political connection of listed companies on the development of green innovation of enterprises. The specific empirical analysis model is shown in Equation (1):
Among them, the subscripts i, t, respectively, represent the enterprise and the year. The explained variable is the natural logarithm of the green patent authorization of listed company i in the year t, which measures the green innovation output level of listed companies. The core explanatory variable PCit is a virtual variable of whether the listed companies have political connections. Therefore, β1 is the core parameter to be estimated in this paper, and it is expected that β1 is significantly negative by theoretical assumptions. In the model, Xit is the set of control variables at the enterprise level, which is set as follows. In order to control the macro external environment shock, the paper controls the year fixed effect . Considering the interference of potential unchanging factors of enterprise industry on causal inference, Equation (1) also controls the fixed effect of enterprise industry . is a random perturbation term of the model. In order to prevent heteroscedasticity from affecting the reliability of empirical results in this paper, all statistical inferences are discussed based on heteroscedasticity robust standard errors. In summary, the paper will use the panel two-way fixed-effect model of China’s A-share listed companies to make empirical analyses of the theoretical assumptions.
3.5. Selection of Control Variables
In terms of control variables, referring to the existing literature standards, in this paper,
Xit specifically includes: (1) Size is the natural logarithm of the total assets of listed companies at the end of the current year; (2) Age is measured by the company’s listed years; (3) Leverage is the ratio of total liabilities of listed companies to total assets in the current year; (4) ROA is the measurement of return on assets of listed companies in the year; (5) Fix is the measurement of fixed asset ratio of listed companies in the year; (6) Cash is the measurement of cash holdings ratio of listed companies in the year; (7) Indratio is the measurement of the proportion of independent directors of listed companies in the year; (8) Boardsize is the measurement of board of directors shareholding ratio of listed companies in that year; (9) Mshare is the measurement of management shareholding ratio of listed companies in that year; (10) Top1 is the measurement of the shareholding proportion of the largest shareholder of listed companies in the year; (11) SOE is a dummy variable. If the listed company is a state-owned enterprise, take 1, otherwise 0; (12) Growth is the income growth rate of listed companies. The variable types, names and definitions are detailed in
Table A1 (See
Appendix A. Same as below).
Table A2 shows detailed descriptive statistics of the variables.
It is necessary to analyze the correlations between the main research variables before conducting empirical regression analysis to prevent the unrecognized problem caused by model misspecification. The correlation test results of control variables are shown in
Table A3. It can be seen from
Table A3 that the correlation coefficient between the core explanatory variables and each control variable is not large, so there is no systematic error caused by the highly col-linearity problem in this paper.
6. Conclusive Comments and Discussions
This paper discusses the influence of political connections on enterprise green innovation and its influencing mechanism. As a consequence, political connections inhibit the level of green innovation, which confirms our hypothesis and is consistent with the results of previous empirical studies [
29,
51]. In addition, the mechanism analysis is in line with our hypothesis, i.e., R&D investment and excessive debt ratio play a mediating role, which can also be supported by the existing literature [
52,
53,
54,
55]. All these conclusions have passed the robustness test. Meanwhile, we conduct a series of heterogeneity analyses for regions and industries for which the results show that political association will have different impacts on green innovation in the East, West, and different industries in China.
In the existing studies, scholars from all over the world generally believe that among all the emerging markets (BRICS and other emerging markets), China is equipped with more sufficient funds and more convenient financial institutions, where fewer financing constraints are required for enterprises. Correspondingly, the whole financing process must be supported by relevant laws and regulations to ensure the good operation of capital flow. However, compared with developed countries, China’s domestic market system is not perfect at present, with some laws still in the pilot stage, giving enterprises the motivation to pursue political connections. Enterprises can enjoy efficient financing conditions and ensure profitability through political and business relations. On the contrary, the lack of political connections of SMEs, together with the difficulty of financing, is always a problem to be solved in China’s economic system. Under these circumstances, the important role of political connections breeds a series of nonmarket strategies, such as corruption. Daily capital operation and production behavior can be ensured by enterprises through sound political connections, without which the resulting ‘organizational inertia’ has a destructive impact on green innovation. Therefore, it is just because of the existence of China’s efficient financing environment that the conclusion obtained in this paper is different from that obtained [
28] when the observation object is set as other emerging markets.
Our study is subject to several limitations. (1) Based on the mediating effect of R&D investment and excessive debt ratio, this paper empirically analyzes the influence mechanism of political connections on corporate green innovation. However, the perspective of this paper mainly focuses on the internal governance of enterprises, with insufficient attention paid to the external environment. For instance, the relationship between political connections and green innovation may also be influenced by external factors such as political turnover [
56] and negotiation intentions [
57], which may have moderating effects on the two. (2) Due to the differences in political connections in different economic environments and political systems, political connections and green innovation are highly bound to China’s socialist market economic system, so this paper lacks comparative analysis with different economic environments.
Combined with the limitations of this study, the key areas of future research are as follows: (1) Enrich the research perspective, incorporate the external environment into the consideration of mechanism analysis, and test whether internal and external factors can interact with the relationship between political connections and green innovation. (2) Researchers need to compare and analyze the influence of political connections on green innovation in different economic environments and further verify the influencing mechanism of political connections on enterprise green innovation.
In general, we focus on China, an emerging economy that is undergoing industrial transformation and rising economic strength, hoping to contribute to the green innovation field. It is worth mentioning that our study found that the inhibition of political connections on enterprise green innovation was mainly reflected in the manufacturing industry rather than in the nonmanufacturing industry. However, it is precisely the manufacturing industry that urgently needs to carry out green innovation and reduce pollution levels, which suggests that countries should weaken the political connections of their manufacturing industries and reduce their protection. Such a strategy can, in return, force manufacturing enterprises to improve their green innovation ability.