The impact of environmental regulation on technological innovation of enterprises: Based on empirical evidences of the implementation of pollution charges in China

Wang, Yuxing; Ye, Wenhui; Wang, Bichun Article The impact of environmental regulation on technological innovation of enterprises: Based on empirical evidences of the implementation of pollution charges in China Economics: The Open-Access, Open-Assessment Journal Provided in Cooperation with: De Gruyter Brill Suggested Citation: Wang, Yuxing; Ye, Wenhui; Wang, Bichun (2024) : The impact of environmental regulation on technological innovation of enterprises: Based on empirical evidences of the implementation of pollution charges in China, Economics: The Open-Access, Open-Assessment Journal, ISSN 1864-6042, De Gruyter, Berlin, Vol. 18, Iss. 1, pp. 1-16, https://doi.org/10.1515/econ-2022-0068 This Version is available at: https://hdl.handle.net/10419/306083 Standard-Nutzungsbedingungen: Die Dokumente auf EconStor dürfen zu eigenen wissenschaftlichen Zwecken und zum Privatgebrauch gespeichert und kopiert werden. Sie dürfen die Dokumente nicht für öffentliche oder kommerzielle Zwecke vervielfältigen, öffentlich ausstellen, öffentlich zugänglich machen, vertreiben oder anderweitig nutzen. Sofern die Verfasser die Dokumente unter Open-Content-Lizenzen (insbesondere CC-Lizenzen) zur Verfügung gestellt haben sollten, gelten abweichend von diesen Nutzungsbedingungen die in der dort genannten Lizenz gewährten Nutzungsrechte. Terms of use: Documents in EconStor may be saved and copied for your personal and scholarly purposes. You are not to copy documents for public or commercial purposes, to exhibit the documents publicly, to make them publicly available on the internet, or to distribute or otherwise use the documents in public. If the documents have been made available under an Open Content Licence (especially Creative Commons Licences), you may exercise further usage rights as specified in the indicated licence. https://creativecommons.org/licenses/by/4.0/ Research Article Yuxing Wang*, Wenhui Ye, and Bichun Wang The Impact of Environmental Regulation on Technological Innovation of Enterprises: Based on Empirical Evidences of the Implementation of Pollution Charges in China https://doi.org/10.1515/econ-2022-0068 received May 12, 2023; accepted November 08, 2023 Abstract: Environmental protection is closely related to high-quality economic developments. Based on the matching of micro-databases from 2000 to 2008, this study used the “Regulations on the Collection and Use of Pollution Fees” policy implemented in China in 2003 as the exogenous impact to construct the intensity Difference in Differences model in order to investigate the effects of pollutant discharge fee on technological innovation of enterprises and the underlying mechanisms. The results showed that governmental environmental regulations significantly improved the level of technological innovation of enterprises, and the conclusion was still valid after a series of robustness tests. The results of the parallel trend verified the rationality of the differential setting and the dynamic effects showed that the pollutant dischargefeeshadacontinuouspromotingeffect on the technological innovation of enterprises. The results of the placebo tests rejected the original hypothesis of the mistaken model. The mechanism verifications revealed that the strengthening of environmental regulation by the government acted on the innovation level of enterprises through the two mechanisms, i.e., the promotion of enterprises’fixed asset investments and government subsidies, and finally improved the enterprises’ technological innovation levels. Keywords: environmental protection regulation, enterprise innovation, pollutant discharge fees, differences-in-differences 1 Introduction The report of the Party’s 20th National Congress proposed “promoting green development and promoting harmonious coexistence between human and nature,”which highlighted the need to further promote the prevention of environmental pollution, to carry out the treatments of new pollutants, and to promote the improvements of urban and rural living environments. As we can see, promoting economic development also needs to promote the green and healthy development of society and the ecological environments. As the saying goes, “lucid waters and lush mountains are invaluable assets.”We should stick to the path of green development and jointly build a foundation for ecological civilization. In China, the will of the government plays a major role in environmental regulations. On the one hand, the government as the designer of environmental regulations restricts the behaviors of market players. On the other hand, the central government will also send environmental inspection teams to supervise and inspect the implementations of central environmental policies by local governments. As a very important part of society, the main purpose of enterprises is to make profits. Environmental regulatory policies can affect their costs and profits, thus having a significant impact on them. In 2003, the Chinese government implemented the Regulations on the Management of the Collection and Use of Pollutant Discharge Fees. Notably, as the first related policy on the national level, the environmental regulations provide us with a good opportunity for quasi-natural experiments to identify the impact on micro market players and help us to identify the impact on enterprise technological innovations by using Difference in Differences (DID) method. According to traditional economic theories, increasing investments in environmental protection will increase the operating costs and social values, and reduce the overall competitiveness of enterprises, which may hinder their  * Corresponding author: Yuxing Wang, School of Economics, Yunnan University of Finance and Economics, Kunming 650221, China, e-mail: [email protected] Wenhui Ye: School of Economics, Yunnan University of Finance and Economics, Kunming 650221, China Bichun Wang: School of Statistics and Mathematics, Yunnan University of Finance and Economics, Kunming 650221, China Economics 2024; 18: 20220068 Open Access. © 2024 the author(s), published by De Gruyter. This work is licensed under the Creative Commons Attribution 4.0 International License. further development (Wang & Wan, 2011). However, Porter’s hypothesis (Porter & van der Linde, 1995) believes that enterprises’compliance with the government’s environmental regulations not only does not increase their operating costs but also promotes their technological innovation and brings greater benefits, thus occupying a competitive position for enterprises in the international market. Can the environmental regulations formulated by the Chinese government promote the innovation of enterprise? What is the impact of environmental regulation on micro-market players? These are the questions to be answered in this study. We took the pollutant discharge fee implemented by the Chinese government in 2003 as a quasi-natural experiment and used the relevant data of the industrial enterprise database, the pollution database, and the enterprise innovation database of China to study the impact of environmental protection rules on enterprise technological innovation, which might help to put forward relevant policy suggestions based on our study. This study identifies the causal effects of environmental protection tax from three aspects: research perspective, research method, and action mechanism. To identify the causal effects, we took the promulgation of the Regulations on the Administration of the Collection and Use of Pollutant Discharge Fee (hereinafter referred to as “Pollutant Discharge Fee”) in 2003 as a quasi-natural experiment and used policy evaluations to get the impact on enterprise innovation. In terms of research methods, we used the DID method to analyze the enterprise micro database, avoided the endogeneity and missing variable bias, and thus obtained a robust consistent estimator. In terms of the mechanisms, we explored the mechanism of enterprise investment and government subsidy, and then explored the impact on innovation from the perspective of enterprise operation decisions, which are important marginal contributions of this study. The following contents of this manuscript are the institutional background and literature review (Section 2), the mechanism analysis (Section 3), the data processing and model setting (Section 4), the basic regression and robustness tests (Section 5), the heterogeneity analysis (Section 6), the mechanism verifications (Section 7), and the conclusions and policy recommendations (Section 8). 2 Institutional Background and Literature Review 2.1 Institutional Background Since the twenty-first century, the awareness of environmental protection has been deeply rooted in various fields such as economy, culture, and politics. As early as 1979, China promulgated the “Environmental Protection Law of the People’s Republic of China (Trial)”for the first time and began to collect a certain amount of pollution discharge fees, but this is only implemented in local areas and regions and has not been really carried out throughout China. In 1982, the Interim Measures for Collecting Pollutant Discharge Fees were subsequently promulgated, which were not well implemented due to various subjective and objective limitations. Until the party’s 16 national Congress, a new generation of central leadership proposed “harmonious development of human and nature,”and paid great attention to ecological protection, and implemented the “Regulations on the collection and use of pollution fees”in 2003, which was a milestone in the field of environmental protection, marking the beginning of the nationwide environmental regulation implementation. As it was issued by the State Council, the implementation was mandatory and rigid throughout China, and the punishment for pollution emission of enterprises was very severe stipulating one–three times fines and forced suspension of production. This is the first time that legislation has been used to regulate environmental pollution and bring out remarkable outcomes in China. The JiangSu Provincial Department of Environmental Protection has formulated the “Jiangsu Province Measures for the Issuance and Management of Pollutant Discharge Permits (Trial)”to standardize the behavior of pollutant discharge permits and strengthen the management of pollutant discharge permits. Through the implementation of policies, the innovation level of enterprises in the region has been significantly improved. We collected data related to pollution discharge from 2001 to 2008 from the official website of the Ministry of Ecological and Environmental Protection of China. Figures 1 and 2 show the content of heavy metals, and the emission of sulfur dioxide and industrial dust in wastewater discharge, respectively. It can be seen from the figure that before the introduction of government environmental regulation in 2003, the emission of pollutants was on the rise. However, after the government imposed the discharge fee in 2003, the discharge of pollutants such as wastewater, waste gas, and industrial dust decreased significantly, indicating that enterprises significantly reduced the discharge of pollutants after the government imposed the discharge fee, and the discharge regulations played a significant role in reducing pollution and protecting the environment. After the implementation of the Chinese government’s environmental protection policies in 2003, the pollutant emissions of enterprises significantly decreased. However, this undoubtedly increased the cost burden on enterprises. So what impact does this have on their technological innovation. 2Yuxing Wang et al. 2.2 Literature Review 2.2.1 The Impact of Environmental Regulation on the Performance of Enterprises Optimizing the business environments of enterprises has a great impact on enterprise performance, while the impact of environmental regulation on enterprise performance remains unclear. Current research results are mainly categorized into the following two aspects. First, the implementation of environmental regulations by the government will increase the production cost of enterprises, which has a negative impact on the performance of enterprises. Jie et al. (2014) found that government environmental regulation would increase the cost of production and operation of enterprises, thus having a negative impact on enterprise performance. Second, government environmental regulation has a positive impact on enterprise performance by promoting technological innovations and increasing enterprise production efficiencies. Ma et al. (2012) found that the positive effect of environmental regulation through technological innovation exceeded the negative effect caused by increased cost, which finally led to the increase of industrial performance. Yang and Peng (2021) found that environmental regulations have a positive impact on corporate performance. Similarly, Liu Xuezhi et al. (Liu & Duan, 2021) used the DID Figure 1: Heavy metals in wastewater discharge from 2001 to 2008 in China. Figure 2: The emission of sulfur dioxide and industrial dust from 2001 to 2008 in China. Impact of Environmental Regulation on Technological Innovation of Enterprises 3 model to verify the positive impact of environmental regulations in the current period on enterprise performance in the following period. Javeed et al. (2020) studied the manufacturing industry in Pakistan and found that after considering the degree of product market competition, the promoting effect of environmental regulations on corporate performance is strengthened; Li et al. (2020) found that after considering the impact of organizational redundancy, it strengthened the effect of voluntary regulation on promoting green innovation. 2.2.2 The Impact of Environmental Regulation on the Innovations of Enterprises Technological innovations are the major content and the core driving force of the development of enterprise. Due to its high investment, long time cycle, etc., enterprise technology innovation level is greatly influenced by internal resource allocation, external social financial support, and local environmental policies. Therefore, the changes in the intensity of regional environmental regulation will have a greatimpactontheproductioncostandfinancing constraints of enterprises, which might further affect the technological innovation level of enterprises. Reportedly, environmental regulation might reduce the capital flow and increase the financing constraints by increasing the production costs of enterprises, thus inhibiting the technological innovation of enterprises (Bi & Li, 2020;Liu&Ran,2016). Domestic and foreign scholars have made rich achievements in the research on the impact of environmental regulation policies on enterprise innovations, but there is no unified opinion. Some scholars admitted that environmental regulations such as emission tax and pollution permit have a positive impact on promoting the technological innovation of enterprises (Villegas-Palacio & Coria, 2010). Langpap and Shimshack (2010) took water pollution treatment in the United States as the research object and found that public participation regulation tools played a crucial role in preventing water pollution. However, some other scholars hold opposite views on this issue. On the one hand, the imposition of pollution tax and the improvement of emission standards both reduce the expected returns of enterprises’ research and development, which makes enterprises have to reduce their investment in research and development. Therefore, strict environmental regulation will reduce enterprises’technological innovations (Antweiler et al., 2001). On the other hand, environmental regulation is equivalent to imposing new constraints on the production decision-making of enterprises, which will hinder the technological progress of enterprises (Becker, 2011; Lanoie et al., 2011). Other scholars believe that the impact of environmental regulation on technological innovation depends on the game of different forces. According to Porter’s hypothesis, the impact of environmental regulation on enterprise innovation depends on the game between “innovation compensation effect”and “compliance cost effect.”Appropriate environmental regulation policies can reduce the uncertainty of enterprises’ future expectations through strict policies, and force enterprises backward to carry out technological innovations. In this case, the production capacity of enterprises can also be improved. Specifically, Calel and Dechezlepretre (2011) took the Emissions Trading System of the European Union as the research object and found that the technological innovation level of enterprises regulated by environmental regulations had been improved. Some scholars are concerned that the intensity of environmental regulations will change in stages with different time (Tong & Zhang, 2012). Specifically, although the cost of enterprises will increase in the short term along with the increases of the intensity of environmental regulations (Miao & Su, 2019; Zhao et al., 2019), the enterprises will be forced to improve their industrial structures and technological innovation levels in the long term (Den & Wang, 2021; Lv & Huang, 2021). Therefore, scholars have conducted relevant studies on the nonlinear relationship between environmental regulation intensity and enterprise innovation and found that the relationship shows a significant U-shaped curve (Feng & Jia, 2021; Fan et al., 2021; Wang et al., 2021), while other scholars found that the relationship between environmental regulation and enterprise technological innovation presents an inverted “U” shaped curve and an inverted “N”shaped curve (Wang & Liu, 2014; Yu et al., 2019; You & Li, 2022). Kesidou et al. argue that manufacturing enterprises located in regions with stricter environmental regulations have stronger technological innovation effects and can generate more green patents. Sun et al. conducted a study using data from 132 companies in 16 highly polluting industries in China and found that environmental regulations have a promoting effect on green technology progress, and companies pay more attention to research and development investment in environmental protection; Lanoie et al. studied enterprises in seven OECD countries and found that strict regulations can promote innovation (Kesidou & Wu, 2020; Lanoie et al., 2011; Sun et al., 2019). Presently, there are many types of environmental regulations implemented in China, and different types of environmental regulations have significant heterogeneity on the innovation level of enterprises. Among them, the number of environmental legislation and the amount of pollutant discharge charges have a significant effect on the improvement of enterprises’green technology innovation 4Yuxing Wang et al. (Hua & Li, 2022; Jin et al., 2022; Zhang et al., 2021). In addition, the implementation of environmental regulations is greatly influenced by the government’sfinancial policies in China. The financial system in China is a combined system of political centralization and economic decentralization, which is influenced by the pressure of “promotion tournaments”of various regions. Local governments should not only maintain the growth of local economies but also protect local environments. Therefore, the impact of environmental regulation on the level of technological innovation of enterprises is closely related to the financial policies of local governments (Wu & You, 2019). Collectively, there is no unified conclusion yet about the relationship between environmental protection regulation and enterprise technological innovation based on current studies, and the content of causal inference was not involved in relevant discussions, which has a strong endogeneity problem. In terms of the mechanisms, the current studies have not explored the impact on enterprise technological innovation from the perspective of enterprise operation decisions. (Griffith & Reenen, 2021; Guo & Liang, 2022). 3 Mechanisms Analysis In order to pursue more profits, enterprises will constantly update technologies and improve the level of innovation. Generally speaking, government environmental regulations affect the export behaviors of enterprises mainly by encouraging enterprise innovation and promoting the increase of enterprise intermediate product input, so as to promote enterprises to increase the added values of export products and thus improve the quality of export products (Figure 3). 3.1 Enterprise Investment Mechanism The nationwide environmental regulations are equivalent to strong constraints on companies, which force companies to make relative reactions under such conditions. On the one hand, the overall costs of enterprises will increase according to government’s strict environmental regulations. Briefly, enterprises have to increase investments in sewage equipment and improve production efficiencies in order to reduce production costs, which results in the increase of the fixed assets of enterprises in the book as the fixed assets investments. On the other hand, enterprises will pay more attention to the efficiencies of capital use, especially the efficiencies of investments. Briefly, enterprises would use limited resources intensively in the production fields, reduce the unit energy consumption and emissions of products, increase the unit investment efficiencies of products, and use the least resources to create the maximum value. In this chain, the government’senvironmental regulations will increase the investment in fixed assets of enterprises. The increase of fixed assets improves the book value of enterprises and is conducive to promoting unit product investment efficiencies, improving the innovation degree and technological content of enterprises, and finally promoting the improvement of enterprise technological innovations. 3.2 Government Subsidy Mechanism Notably, the Interim Measures for Collecting Pollution Charges issued in China in 2003 provided incentives for enterprises that complied with environmental protection emission standards or voluntarily reduced pollution emissions. Moreover, the purchase of energy-saving and emission-reduction equipment can be deducted from the input tax. All these have greatly reduced the production costs of enterprises. On the one hand, the government’s subsidized policies and subsidies for enterprises’environmental protection reduced the cost of enterprises, increased the proportion of factor input, improved the production efficiencies of enterprises, and promoted the enthusiasm of enterprises to invest in the field of technological innovations. On the Figure 3: Mechanisms of environmental regulation on the level of enterprise innovation. Impact of Environmental Regulation on Technological Innovation of Enterprises 5 other hand, the funds available to enterprises have actually increased, and more funds can be used in the field of technology investment and research studies. Under such conditions, enterprises can be encouraged to improve the level of technological innovation. 4 Data Processing and Empirical Model 4.1 Data Sources In this study, the China Industrial Enterprise Database, China Enterprise Pollution Database, and China Enterprise Innovation Database from 2000 to 2008 were used including a total of 32,035 observed values. Briefly, the database of Chinese industrial enterprises contains the basic information of state-owned enterprises and private enterprises above the designated size, including financial indicators, registration information, year of establishment, sales information, and cost information, which provides detailed data for our study of micro-enterprises. The Chinese Enterprise Pollution Database contains detailed data on pollution emission indicators of micro-enterprises, including exhaust gas, wastewater, industrial dust, and other pollution indicators. The enterprise innovation database contains the innovation indicators of micro-enterprises, including the number of patents granted and the number of patent applications. 4.2 Model of Measurement In this study, the DID method is adopted to study the impact of the pollutant discharge fee implemented in 2003 on enterprise innovations, which is conducive to alleviating the possible endogeneity and missing variable bias. Because the pollutant discharge fee in 2003 was rolled out all at once without a pilot, we could not distinguish the treatment group and the control group according to the ordinary DID method. To solve this problem, we grouped the enterprises according to the median emission intensity of pollutants and then distinguished the treatment group and the control group. For this purpose, we established the DID model of intensity as follows: =+ × + ++++αα γX δσθεlnnova treat post . ict ict i c t it01 Specifically, lnnova ict represents the innovation level of the ith enterprise of city cin the year t. treat is the divided intensity based on the median amount of pollution discharge, i.e., those that are greater than or equal to the median amount belonged to the treatment group, and those that are less than the median amount belonged to the control group. post is the virtual group when the pollution discharge fee is promulgated, i.e., those that are greater than or equal to 2003 were marked as 1, and those that are less than 2003 were marked as 0. X ict represents a series of control variables at the city level and the enterprise level. δ i represents the fixed effects of individual enterprises. σ c represents the fixed effects at the city level. θ t represents the fixed effects at the control time level. ε it is the random disturbance term. α 1 is the regression coefficient of DID that we were interested in. 4.3 Explained Variable In this study, the logarithm of the number of patents and the logarithm of R&D expenditure were used as the explained variables. Specifically, the logarithm of the number of patents was used as the result of baseline regression, and the logarithm of R&D expenditure was used as the control index of robustness test. 4.4 Core Explanatory Variable In this study, the pollution emission intensity of enterprises was established as the core explanatory variable. After the standardization of wastewater, waste gas, solid pollutants, and sulfur dioxide, the overall pollution emission index was synthesized by weighting. Then, the data of pollution emission indexes were arranged, i.e., those that are greater than or equal to the median amount belonged to the treatment group, and those that are less than the median amount belonged to the control group. The relevant variable names and statistical descriptions are shown in Table 1. 5 Main Results 5.1 Baseline Results This study adopted the Chinese industrial enterprise database, pollution database, and enterprise innovation data from 2000 to 2008, including 32,035 observed values to identify the DID model. The estimated results are shown in Table 2. Specifically, the explained variable is the 6Yuxing Wang et al. innovation level of the enterprise, and (1) is the logarithm of R&D expense as the regression result of the explained variable, and (2) is the regression result of control variables at the enterprise level added on the basis of (1), and (3) is the regression result of patent logarithm as the explained variable. Results of the basic regression showed that the coefficients were significantly positive, and all passed the 1% level test. In addition, when controlling variables at the enterprise level such as asset-liability ratio, enterprise size, enterprise life, and enterprise profit rate were added, there was no significant change in the results, and the coefficient was still significantly positive, indicating the robustness of the conclusion. Collectively, we concluded that the environmental regulation of pollutant discharge fees introduced by the Chinese government in 2003 had significantly improved the innovation level of enterprises. 5.2 Dynamic Effect and Parallel Trend Test The premise of applying the DID method is that the treatment group and the control group have the same development trend before the policy shock. This study used the Event Study method reported by Jacobson et al. (1993)to verify the parallel trend of the DID model and further discussed the dynamic effects of policy shocks on this basis. The specific estimation is as follows: ∑=+ × + +++ + ≥− αβ DγX δσθ ε lnnova treat . ict kkit tkict i c t it 02 3 Table 1: Variable names and statistical descriptions Variable name Observations Mean Standard deviation Min Max lnyf 32,035 0.5963 2.079 0 14.3 lnzl 32,035 0.1949 0.6339 0 7.724 did 32,035 0.29 0.4539 0 1 size 31,995 11.6477 1.498 0 18.96 lnL 27,960 6.185 1.1838 0 11.3122 lev 31,995 0.5737 0.2913 −0.2824 9.844 finance 31,947 0.036636 1.649 −287.25 62.428 age 32,004 16.07 30.598 0 2008 lnwage 25,769 2.616 0.6998 −5.2 7.529 sub 26,237 0.0032 0.0628 −0.18 7.59 lnkl 27,909 4.313 1.26 −3.4 10.9 profit 31,995 0.1977 27.954 −2.37 5,000 kc 28,405 3.3 449.872 0 75,820 Table 2: Regression results of basic variables Variable (1) (2) (3) (4) R&D expenses Add control variable Number of patents Add control variable lnyf lnyf lnzl lnzl did 0.506*** 1.172*** 0.089*** 0.112*** (0.049) (0.110) (0.014) (0.019) size 0.698*** 0.113*** (0.089) (0.016) lnL 0.075 0.037** (0.091) (0.016) lev 0.374*** −0.015 (0.134) (0.024) finance 0.034 0.003 (0.021) (0.004) age −0.000 −0.000 (0.001) (0.000) lnwage 0.086* 0.015 (0.047) (0.009) sub −0.073 0.030* (0.106) (0.017) lnkl 0.022 0.006 (0.062) (0.011) profit 0.189* 0.016 (0.113) (0.021) kc −0.008 −0.003 (0.012) (0.002) Constant 0.473*** −8.449*** 0.173*** −1.432*** (0.015) (0.889) (0.004) (0.172) Observations 30,462 20,145 30,462 20,145 R-squared 0.476 0.592 0.669 0.681 Notes: *, **, and *** indicate statistical significance at the 10, 5, and 1% levels, respectively. Impact of Environmental Regulation on Technological Innovation of Enterprises 7 Specifically, treat it is the pollution emission of the ith enterprise in the year t. D tk is the virtual event time variable with a value of 0 or 1. When k<0, D tk is taken as 1 in the kth year before the policy impact; otherwise, it is taken as 0; when k>0, take 1 in the kth year after the policy shock occurs, otherwise take 0; when k=0, D tk is taken as 1 in the year when the policy impact occurs, otherwise, it is taken as 0. In order to simplify the analysis, the case of k≤ −2 is considered as k=−2; that is, the corresponding D tk of the 2 years before the policy impact is taken as 1; otherwise, it is taken as 0. In the regression, k=−1 is taken as the benchmark group, and the regression coefficient β k represents the difference between the treatment group and the control group 1 year before the policy impact. Figure 4 shows the test results of the parallel trend from 2001 to 2006. The point in the graph represents the β k estimate, and the dotted line passing through the point and perpendicular to the X-axis represents the 95% horizontal confidence interval. The X-axis represents the estimated value of β k from 2001 to 2006, and the policy occurred at the year 2003. As shown in Figure 4, the estimated value of β k at the year 2001 and 2002 fluctuates around 0, and the width of the corresponding 95% level confidence interval is wide and crosses 0, and the difference between the treatment group and the control group has not changed significantly. Collectively, the above results indicated that the parallel trend test has been passed. From 2003 to 2006, the enterprises in the treatment group and the control group started to open a gap after the promulgation of the Regulations on the Collection and Use of Pollution Discharge Fees, indicating that the effects were brought about by the impact of the policy. In addition, Figure 4 also reflects the impact of the promulgation of the Regulations on the Collection and Use of Pollution Discharge Fees in 2003 on the level of technological innovation of enterprises. Currently, the estimated value of β k from 2003 to 2006 are all significantly positive, indicating that the promulgation of the pollution charge policy has promoted the innovation level of enterprises. 5.3 Robustness Test Although the previous parallel trend test and DID results showed that environmental regulations have significantly improved the level of technological innovation of enterprises, it is still impossible to completely eliminate the endogenous problems caused by measurement errors and enterprise selfselection. In order to determine the reliability of the research results, we conducted a series of robustness tests. 5.3.1 Placebo Test All the enterprises in 2003 from the overall panel data were selected with randomly selected 50% of them and matching them with the overall panel data. The 50% of the enterprises selected were used as the experimental group, and the rest were used as the control group. On this basis, DID processing was performed, and the processes were repeated for 200 times. The final results with the results of 200 random processes are shown in Figure 5.TheX-axis represents the size of the estimated coefficient of the “pseudo policy dummy variable,”and the Y-axis represents the size of the density value and pvalue. The curve is the kernel density distribution of the estimated coefficient, the dot is the pvalue corresponding to the estimated coefficient, the vertical dotted line indicates the real estimated value (i.e., 0.112) of the DID model, and the horizontal dotted line is the significance level (i.e., 0.1). The estimation coefficients are mostly concentrated near zero, and most of the pstatistics are greater than 0.1. The true estimates of the DID model belong to obvious outliers, indicating that our regression results passed the placebo test. 5.3.2 Robustness Test of the Explained Variable The first method is to shrink the tail of the explained variable. We shrank the tail of the explained variable by 1%. Column (1) is the result of shrinking the tail of the enterprise’s R&D expense data, and Column (2) is the result of shrinking the number of enterprise patents. Results showed that the regression results of the explained tail shrinking treatment were significantly positive, which were consistent Figure 4: Parallel trend test chart from 2001 to 2006. Notes: The solid line plots the point estimation coefficient, and the dashed lines denote the 95% confidence interval. 8Yuxing Wang et al. Jie, M. H., Wang, J., & Liu, D. M. (2014). Environmental regulation, technological innovation and business performance. Nankai Business Review,17(6), 106–113. Jin, X., Guan, H. X., & Chen, S. (2022). How do environmental regulation tools affect green technology innovation of enterprises – Heterogeneity effect research based on dual perspectives. Frontiers of Science and Technology of Engineering Management,41(4), 62–68. Kesidou, E, & Wu, L. (2020). Stringency of environmental regulation and eco-innovation: Evidence from the eleventh Five-Year Plan and green patents. Economics Letters,190(c), 109090. Lanoie, P., Laurent-Lucchetti, J., Johnstone, N., & Ambec, S. (2011). Environmental policy, innovation and performance: New insights on the Porter hypothesis. Journal of Economics ＆Management Strategy, 20(3), 803–842. Langpap, C., & Shimshack, J. (2010). Private citizen suits and public enforcement: Substitutes or Complements. Environmental Economics and Management,59(3), 235–249. Li, D., Tang, F., & Zhang, L. (2020). Differential effects of voluntary environmental programs and mandatory regulations on corporate green innovation. Natural Hazards,103, 3437–3456. Liu, C. L., & Ran, M. S. (2016). The impact of environmental regulation on enterprise technological innovation under financing constraints. Systems Engineering,34(12), 62–69. Liu, X. Z, & Duan, D. D. (2021). Environmental regulation, technological innovation and enterprise performance: A quasi natural experiment based on VOCs pollution charge policy in petrochemical industry. Chinese Journal of Environmental Management,13(2), 94–101 + 133. Lv, P., & Huang, S. Q. (2021). Will the pressure of environmental regulation promote the transformation and upgrading of enterprises. Nankai Business Review,24(4), 116–129. Ma, H. L., Huang, D. C, & Yao, H. Z. (2012). Technological innovation, industrial performance and environmental regulation: An empirical analysis based on the Yangtze River Delta. Soft Science,26(1), 1–5. Miao, M., & Su, Y. D. (2019). The impact of environmental regulation on enterprise technological innovation –A test of intermediary effect based on financing constraints. Soft Science,33(12), 100–107. Porter, M. E., & van der Linde, C. (1995). Toward a new conception of the environment-competitiveness relationship. Journal of Economic Perspectives,9(4), 97–118. Sun, Y, Du, J, & Wang, S. (2019). Environmental regulations, enterprise productivity, and green technological progress: Large-scale data analysis in China. Annals of Operations Research,290, 369–384. Tong, W, & Zhang, J. M. (2012). Can environmental regulation promote technological innovation –A reexamination based on the data of Chinese manufacturing enterprises. Finance & Economics,11,66–74. Villegas-Palacio, C., & Coria, J. (2010). On the interaction between imperfect compliance and technology adoption: Taxes versus tradable emissions permits. Journal of Regulatory Economics,38(3), 274–291. Wang, J., & Liu, B. (2014). Environmental regulation and total factor productivity of enterprises: Empirical analysis based on data of Chinese industrial enterprises. China Industrial Economics, 3,44–56. Wang, G. Y., & Wan, D. (2011). Porter hypothesis, environmental regulation and technological innovation of enterprises: A comparative analysis of the Middle East. China Soft Science,1(1), 100–112. Wang, Z. Y., Cao, Y., & Lin, S. L. (2021). Characteristics and heterogeneity of environmental regulation’s impact on green technology innovation of enterprises –based on green patent data of Chinese listed companies. Studies in Science of Science,39(5), 909–919+929. Wu, G. Z., & You, M. D. (2019). The impact mechanism of environmental regulation on technological innovation and green total factor productivity: The regulatory role based on fiscal decentralization. Journal of Industrial Engineering and Engineering Management, 33(1), 37–50. Yang, R, & Peng, A. (2021). Environmental regulation, technological innovation and performance of heavily polluting enterprises. Journal of East China Normal University(Humanities and Social Sciences),53(1), 129–141 + 173–174. Yu, K. X., Hu, X. Q., & Song, Z. (2019). Environmental regulation, government support and green technology innovation –An empirical study based on resource-based enterprises. Journal of Yunnan University of Finance and Economics,35(4), 100–112. You, M. D., & Li, L. N. (2022). Environmental regulation intensity, frontier technology gap and enterprise green technology innovation. Soft Science,36(8), 108–114. Zhang, G. X., Feng, W. C., & Wang, A. L. (2021). Research on the heterogeneity effect of different types of environmental regulations on technological innovation of industrial enterprises. Management Review,33(1), 92–102. Zhao, L., Xue, Y, & Hu, Y. Q. (2019). Environmental regulation intensity and technological innovation: An empirical study from pollution intensive manufacturing industry. Science & Technology Progress and Policy,36(10), 59–65. Impact of Environmental Regulation on Technological Innovation of Enterprises 15 Appendix A. Variable Descriptions Table A1: Variable descriptions Variable name Data descriptions lnyf Logarithm of research and development costs lnzl Logarithm of enterprise patents did Core variable of DID model(treat*post) size Logarithm of total assets lnL Logarithm of the number of employees lev Total corporate liabilities/total corporate assets finance Interest expense/fixed assets age Current year minus year of business opening lnwage Logarithm of (Total wages payable/number of employees) sub Government subsidies/total sales lnkl Logarithm of (Total fixed assets/number of employees) profit Total profit/total corporate assets kc Fixed assets/Gross industrial output value 16 Yuxing Wang et al.