Navigating Environmental Tax Challenges: Business Strategies for Chinese Firms Sustainable Growth

Abstract

The environmental tax burden on Chinese firms is continuously increasing. How do firms respond to environmental tax pressure to achieve sustainability? This study aims to investigate whether environmental tax burden affects firms’ total factor productivity (TFP), an important indicator of sustainable development. Utilizing panel data of Chinese A-share listed firms from 2018–2022, we find that environmental tax burden positively impacts Chinese firms’ TFP. Interestingly, the impact of environmental tax burden on TFP is mediated through fixed asset investment rather than technological innovation. This indicates that in the short term, the pressure of environmental tax on Chinese firms has not triggered the Porter Effect. Additionally, this effect is stronger in larger firms, non-state-owned firms, and sample firms with weaker financial constraints and lower transaction costs. Therefore, different firms can flexibly adopt different business strategies to achieve sustainable development when facing environmental tax burden.

1. Introduction

China’s targets of carbon peaking by 2030 and achieving carbon neutrality by 2060 represent a commitment to global sustainability. Since the inception of the environmental tax on 1 January 2018, China has employed this fiscal mechanism as a key instrument in reducing pollution and emissions. The environmental tax, with enhanced administrative efficacy and regulatory enforceability, surpasses the previous pollution discharge fee in compelling high-polluting and energy-intensive firms to transform and upgrade, aligning with dual goals of emission reduction and high-quality economic growth.

A growing number of academic studies have examined the association of corporate environmental behavior with production efficiency and developed different opinions. The “Porter hypothesis” points out that proactive environmental regulations within firms can catalyze innovation and offer competitive advantages, thereby enhancing production efficiency and operational outcomes [1,2]. Conversely, the “restriction hypothesis” argues that strict environmental regulations impose additional cost burdens on firms, possibly hindering improvements in production efficiency and performance [3,4]. Furthermore, the “uncertainty hypothesis” suggests that the economic consequences of corporate environmental actions are uncertain [5,6].

In the past decades, China has implemented various environmental regulations measures to deal with the adverse effects of ecological pollution [7]. In the early 1970s, the Chinese government mainly adopted an administration-based ecological policy, levying environmental penalties and pollutant discharge fees. The penalty or fee system has urged polluters to cut emissions [8]. However, the pollution discharge fee system primarily imposes administrative penalties for excessive emissions, which makes no distinction between light and heavy polluters [9]. Hence, the former had no incentive to cut emissions. The pollution is not controlled as expected in high-polluting industries, and the penalty is relatively small for enterprises in lucrative production activities. Therefore, the central authorities attempted to move from a fee to a tax scheme, carrying out market-based environmental regulation. In 2007, the Ministry of Environmental Protection of China initiated a comprehensive study on environmental taxation reform measures, leading to enacting the Environmental Protection Tax Law in 2016, officially taking effect on 1st January 2018. The environmental tax has a more authoritative effect than the pollution discharge fee [10].

In terms of environmental tax in China, prior literature has utilized the event shock of transitioning pollutant discharge fee to environmental tax (abbreviated as “fee to tax”) to examine policy effects on corporate performance [11], corporate innovation [12], and corporate transformation and upgrading [13]. Existing research primarily examines policy effects at the event of “fee to tax” using methods like difference-in-differences but lacks studies on how firms choose strategies in response to environmental tax pressure.

Generally, the tax reform significantly increases compliance costs for firms [14]. Faced with the continuously increasing environmental tax burden, firms are unlikely to remain inactive. In order to reduce costs and improve productivity, firms can choose different environmental strategies to achieve sustainable development. Based on institutional theory, firms can choose a proactive environmental strategy, such as technological innovation. This is a sustainable but high-risk environmental strategy. Additionally, firms can also opt for reactive environmental strategies, such as purchasing green machinery and equipment, and developing new environmentally friendly products. Such fixed asset investments are an immediate strategy that can meet short-term environmental requirements, but they require a substantial amount of funds. Therefore, how will firms in China choose business strategies to improve their production efficiency?

To address this question, we use panel data of A-share listed firms from 2018 to 2022 in China to investigate the association of environmental tax burden on firms’ total factor productivity. We find environmental tax burden positively affects firms’ TFP, indicating that the environmental tax enhances corporate environmental efforts and forces green transformation. Moreover, the way in which environmental tax pressure affects the production efficiency of firms is through investment in fixed assets rather than technological innovation. This implies that environmental regulations force firms to invest in fixed assets to meet emission standards in the short term. Besides, we also discuss different situations about the association of environmental tax burden on firms’ total factor productivity. Results show that large-scale firms are more capable of alleviating the increase in environmental tax costs through measures such as replacing equipment, leading to a more pronounced improvement in firms’ TFP. Secondly, non-state-owned firms are more sensitive to increases in environmental tax costs and are likely to take proactive measures to enhance their TFP. In addition, in samples of firms with weaker financing constraints and lower transaction costs, the positive effect of the environmental tax burden on TFP is more significant.

This study contributes to the existing literature in two ways. First, the choice of environmental policy instruments has been extensively debated since Pigou’s seminal contribution of using environmental taxes to internalize welfare losses from externalities [15]. From a micro perspective, the famous “Porter hypothesis” suggests environmental regulations stimulate corporate innovation, improving productivity and investment. However, existing empirical evidence on corporate behavior and performance provides conflicting guidance. Using samples and data from China, this study shows environmental tax burden promotes corporate production efficiency, but its mechanism of action is not through technological innovation but through fixed asset investment. This finding enriches the literature on the relationship between corporate environmental behavior and production efficiency, providing evidence outside the “Porter hypothesis”. Second, some papers discussing this topic using Chinese samples and data are mainly based on the “fee-to-tax” reform event [7,10], using the difference-in-differences method to examine the economic consequences of environmental taxes, which makes it difficult to explain the specific environmental tax burden faced by firms. This paper takes the amount of environmental tax paid as an indicator of the firm’s environmental tax burden and, based on institutional theory, demonstrates the impact and strategies when firms face the pressure of environmental tax.

This study has significant theoretical and practical implications. Theoretically, this work supplements the literature on environmental tax reform effects and introduces the concept of environmental tax burden and a measurement method. Using empirical data from China, the study shows environmental tax burden enhances total factor productivity primarily through fixed asset investments, providing evidence outside the “Porter hypothesis” to enrich existing literature on environmental taxation and corporate performance. Practically, our findings provide managerial implications for practitioners and policymakers. It helps firms in practice to identify the best strategies for coping with the environmental tax burden, thereby assisting them in achieving the dual goals of energy conservation and stable performance. Furthermore, the research findings are also beneficial for policymakers to refine the environmental tax system, stimulate corporate innovation potential, and promote economic growth and emission reduction.

This paper is structured as follows. Section 2 introduces the institutional background of the environmental tax and develops the research hypotheses. Section 3 is the research design, which mainly introduces the sample, model, and variable measurement. Section 4 presents the main results of the empirical analysis and the robustness results. Section 5 further analyzes the mediating results and heterogeneous results on the relationship between the environmental tax burden and total factor productivity. Section 6 is the discussion and conclusion.

2. Institutional Background and Hypothesis Development

With increasing focus on climate change in recent decades, understanding the effects of environmental regulations on productivity is crucial for designing effective environmental tax reforms. This section outlines the development of environmental tax systems and presents details on the implementation of China’s environmental tax system. It also introduces the main hypotheses of this study.

2.1. Institutional Background

Environmental regulations are a major instrument to encourage firms to take environmental protection actions, mainly including command and control environmental regulations (e.g., pollutant discharge fees) and market-based environmental regulations (e.g., environmental taxes, tradable permits).

For a long time, command-and-control environmental regulations have been the primary instrument. Drawing on developed countries’ experiences, China established a pollution discharge fee system in 1979 and has since introduced a series of regulation systems to levy fees on firms for pollution discharge, aiming to control major industrial pollutants’ emission levels. Previous research has found that pollution discharge fees actively promote pollution emission reductions. However, due to low fee collection standards and enforcement issues, there are concerns and even doubts from the government and all sectors of society about the pollution discharge fee system’s effectiveness in pollution control [16,17,18].

Starting from the 1990s, environmental taxes as an important tool of market-based environmental regulations gradually gained popularity [15]. According to Pigou’s tax principle, environmental taxes target the negative externalities caused by corporate environmental pollution and can effectively encourage enterprises to reduce pollution by internalizing the unit cost of pollution emissions. Unlike other environmental policy tools, environmental taxes not only restrict environmental pollution but also provide additional revenue to offset economic losses, commonly referred to as the environmental taxes’ double dividend effect [19,20,21].

After implementing the reform and opening up policy, China has become one of the world’s leading economies with high economic growth rates and has achieved significant economic development. However, this growth has led to severe resource and environmental issues. The emissions of major pollutants have exceeded the environment’s capacity, and environmental pollution and ecological destruction have become the main bottlenecks to the sustainable development of the economy. Therefore, continuously improving the ecological environment and maintaining economic growth has become a key strategic goal for Chinese enterprises and the government [22]. Against this backdrop, the Chinese government has attempted market-based environmental regulation, and in 2016, formulated the “Environmental Protection Tax Law”, which officially came into effect on 1 January 2018, based on the good governance effects achieved by environmental taxes in some countries and regions.

China’s environmental tax, as stated in the “Environmental Protection Tax Law of the People’s Republic of China”, focuses on pollutants including air pollutants, water pollutants, solid waste, and noise. The basis for taxation is the taxable pollutant emissions of enterprises, resulting in different environmental tax burdens for enterprises with varying levels of pollutant emissions. Although China allows each region to set different environmental tax collection standards based on its own conditions, most regions apply the minimum standards for collecting environmental taxes. Therefore, whether the current standards for environmental tax collection are effective in achieving China’s double dividend is worth studying.

2.2. Hypothesis Development

Environmental taxes are a market-based regulation and an effective, preventive, and long-term environmental economic policy tool. It is a vital component of the environmental policy system. This paper examines the mechanism and impact of the environmental tax burden on total factor productivity based on theories such as the related research of institutional theory and the Porter hypothesis.

2.2.1. Environmental Tax Burden and Total Factor Productivity

As the global commitment to green development intensifies and the capacity of resources and the environment to sustain economic activities diminishes, firms are faced with increasingly stringent environmental regulations. Environmental taxation serves as a prototypical regulatory tool that significantly impacts corporate operations [23]. Typically, to comply with environmental regulations, firms must pay taxes and fees associated with their energy consumption and the emission of pollutants, incurring a cost of compliance [24]. The “restriction hypothesis” argues that strict environmental regulations impose additional cost burdens on firms, possibly hindering improvements in production efficiency and performance [3,4,25,26]. Pollution-intensive firms may even have to passively reduce their production activities in the short term to meet stringent environmental standards temporarily [27]. In addition, the “pollution haven hypothesis” claims that firms relocate to countries with weak environmental standards when environmental taxes rise, reducing profits, productivity, and inputs by limiting production possibilities [28]. Consequently, environmental tax directly increases the average fixed costs of firms, adversely affecting their total factor productivity.

Conversely, the “Porter hypothesis” suggests that appropriate environmental regulations can promote corporation innovation and the innovation will even offset the corporation loss caused by environmental regulations and improve corporation competitiveness [1,2,29,30,31]. Similarly, the “factor endowment hypothesis” suggests that employing available clean natural resources improves production possibilities and productivity [32]. So, from a dynamic perspective, firms are reluctant to witness a decline in total factor productivity (TFP) in the pursuit of sustainable development. Environmental regulations have the potential to bolster corporate environmental consciousness and provide incentives for firms to offset compliance costs through enhancements in production processes or by adopting environmentally friendly equipment [10,33]. As a result, environmental regulations are not only unlikely to diminish TFP but may also confer a competitive advantage. However, only well-designed environmental regulations can incentivize firms to enhance productivity. For instance, in systems of pollutant discharge fees, firms lack subjective motivation to improve processes and increase environmental investment, and there is spatial heterogeneity across different regions in China [34]. Since administrative penalties occur sporadically during inspections and involve relatively low amounts. Besides, the emissions spillover is not taken into the pollutant discharge fee system [9]. Compared to pollutant discharge fees, environmental tax is directly calculated based on discharge amount, with more stringent regulatory measures. Referring to the work of Kochhar and David (1996) and so on, using competing hypotheses in empirical analysis can effectively evaluate and weigh multiple possible explanations or hypotheses, reduce cognitive biases, enhance the accuracy of the analysis, and achieve multidimensional interpretations [35]. So, based on the existing literature and the analysis in this study, we propose the following competing hypotheses:

Hypothesis 1a (H1a). 

The environmental tax burden is positively associated with firms’ TFP.

Hypothesis 1b (H1b). 

The environmental tax burden is negatively associated with firms’ TFP.

2.2.2. The Mediating Roles of Technological Innovation and Fixed Asset Investment

Institutional theory has risen to prominence as a popular and powerful explanation for organizational behavior [36,37]. In the past, many excellent papers have been based on institutional theory to discuss various topics such as CSR (Corporate Social Responsibility), innovation [38], organizational culture [39], and entrepreneurship [40]. This study aims to use institutional theory to analyze the sustainable business strategies of Chinese firms in response to environmental tax challenges. In the institutional theory, organizations are influenced by normative pressures arising from the state or organization itself. Under these conditions, these pressures guide organizations to respond proactively or reactively by adhering to legitimated elements [41]. Environmental tax is a typical form of institutional pressure, which often requires firms to divert attention from operational performance to cope. So, when a firm is aware of increased production costs, it typically adopts two strategies to address the situation: On one hand, with the increase in environmental tax burdens, firms may allocate more resources to technological innovation, exploring eco-friendly materials and processes [42,43]. This is a sustainable and proactive business strategy. On the other hand, firms may purchase relatively environmentally friendly production equipment and materials, reflecting increased fixed asset investment [10,33,44]. This is a real-time and reactive business strategy. Both approaches can alter production inputs, enhancing efficiency. Therefore, we propose the following hypotheses:

Hypothesis 2 (H2). 

Technological innovation strengthens the relationship between the environmental tax burden and firms’ total factor productivity.

Hypothesis 3 (H3). 

Fixed asset investment strengthens the relationship between the environmental tax burden and firms’ total factor productivity.

3. Research Design

The research design primarily introduces the research sample, data sources, variable definitions, and the empirical model.

3.1. Sample and Data

The sample includes all A-share firms listed on the Shanghai and Shenzhen stock exchanges from 2018 to 2022, following China’s implementation of the Law of Environmental Protection Taxation on 1 January 2018 (The Shanghai and Shenzhen A-share listed markets are the largest stock trading markets in China, encompassing listed companies from a wide range of industries. Using Shanghai and Shenzhen A-share listed firms as samples can well represent the overall situation of Chinese listed firms. Besides, data on public firms are relatively more reliable because these firms are required by law to disclose accurate financial information in their annual reports and these reports must meet international standards [45,46]). Data is sourced from the China Stock Market & Accounting Research (CSMAR) database. We excluded firms in the financial industry, those categorized as ST, *ST, or PT, and any samples with abnormal or missing data. We also winsorized all continuous variables at the 1st and 99th percentile to mitigate outliers’ influence on empirical results.

3.2. Variable Definitions

Total Factor Productivity (TFP) is an indicator used to measure output growth from increased factor inputs like capital and labor. Unlike traditional indicators measuring total “growth”, this index better captures enterprise development quality. Due to simultaneity bias and sample selection bias in TFP calculated by OLS, we adopt TFP calculated by OP and LP methods as firm development quality measurement indices [47,48].

Environmental tax burden (LnEnvtax) is an indicator constructed in this paper, used to represent the environmental tax pressure faced by firms. LnEnvtax is measured by taking the natural logarithm of the amount of environmental tax payable. The increase in the amount of environmental tax payable will lead firms to bear a heavier economic burden. In addition, for the robustness check, we also adopted the ratio of the amount of environmental tax payable divided by the amount of tax payable (Envtax).

We control for the following variables based on prior research [6,8,11,12,13], mainly including TOP1 (largest shareholder’s share percentage), Growth (the growth rate of operating income), ROA (return on assets), Return (annual stock return excluding dividends), LEV (assets-liabilities ratio), FirmSize (the log of total assets), SOE (nature of equity ownership), Age (years since a firm is registered), and SA (financial constraints index).

Table A1. Variable definitions.

Variables	Description
TFP_OP TFP_LP	Calculated following Olley and Pakes (1996)
	Calculated following Levinsohn and Petrin (2003)
LnEnvtax Envtax	The natural log of the amount of environmental tax payable at the end of year t.
	The ratio of the amount of environmental tax payable for year t to the total tax payable at the end of year t.
Inv	The ratio of cash paid by a firm for the acquisition and construction of fixed assets, intangible assets, and other long-term assets to total assets at the end of year t.
R_D	The ratio of R&D investment divided by operating revenue at the end of year t.
TOP1	The percentage of shares held by the largest shareholder at the end of year t.
Growth	The ratio of the difference between the total operating income of the year t and the total operating income of year t − 1 divided by the total operating income of the year t − 1.
ROA	Return on assets, defined as net income for year t divided by total assets at the end of year t.
Return	The annual return on individual stocks excluding dividends at the end of year t.
LEV	Total liabilities divided by total assets, at the end of year t.
FirmSize	The natural log of total assets as of the end of year t.
SOE	A dummy variable that equals 1 if the ultimate controlling shareholder of the company is the state, and 0 otherwise.
Age	Years since a firm is registered
SA	The SA index calculated according to the formula by Hadlock and Pierce (2010), the larger the absolute value of the SA index, the more severe the financing constraints of the company are.
ManCost_ratio	The ratio of management expenses at the end of year t to total profit.
LifeCycle	Using the cash flow model method (Dickinson, 2011), the sample is categorized into three stages: the growth phase, the mature phase, and the decline phase, with corresponding values of 1, 2, and 3, respectively.

in Appendix A summarizes the variable definitions.

3.3. Empirical Model

In order to test the impact of environmental tax burden on firms’ TFP, we estimate the following ordinary least squares (OLS) regression model:

$$\begin{gathered} {TFP\_LP}_{i,t}={\alpha }_0+{\alpha }_1{LnEnvtax}_{i,t}+{\alpha }_2{TOP1}_{i,t}+{\alpha }_3{Growth}_{i,t}+{\alpha }_4{ROA}_{i,t}+{\alpha }_5{Return}_{i,t}+{\alpha }_6{LEV}_{i,t} \\ +{\alpha }_7{FirmSize}_{i,t}+{\alpha }_8{SOE}_{i,t}+{\alpha }_9{Age}_{i,t}+{\alpha }_{10}{SA}_{i,t}+Year+Industry+{\epsilon }_{i,t } \end{gathered}$$

(1)

where TFP_LP measures total factor productivity, LnEnvtax measures environmental tax burden. Control variables include TOP1, Growth, ROA, Return, LEV, FirmSize, SOE, Age, and SA. The ${\alpha }_i$ represents regression coefficients, and ${\epsilon }_{i,t}$ is an error term. We also control industry and year.

To test the mediating effects of Hypothesis 2 and Hypothesis 3, we followed the methodology of Wen and Ye (2014) and constructed a three-step mediation effect model [49]:

$${Mediator}_{i,t}={\beta }_0+{\beta }_1{LnEnvtax}_{i,t}+{\beta }_k{Controls }_{i,t}+Year+Industry+{\epsilon }_{i,t }$$

(2)

$${TFP\_LP}_{i,t}={\gamma }_0+{\gamma }_1{LnEnvtax}_{i,t}+{\gamma }_2{Mediator}_{i,t}+{\gamma }_k{Controls }_{i,t}+Year+Industry+{\epsilon }_{i,t }$$

(3)

where the Mediator measures mediating variables, i.e., technological innovation and fixed asset investment. Controls indicate control variables, which are the same as in Equation (1). The ${\beta }_i$ represents regression coefficients, and ${\epsilon }_{i,t}$ is an error term. We also control industry and year.

4. Empirical Results

To examine the correlation between environmental taxes and firms’ total factor productivity, we conducted an empirical study. In this section, we present the results of descriptive statistical analysis, correlation coefficient analysis, basic regression analysis, and robustness tests.

4.1. Descriptive Statistics

Table 1. Descriptive Statistics for the Variables.

Variables	n	Mean	Standard Deviation	Min	Max
TFP_LP	4089	10.583	0.817	8.900	12.693
LnEnvtax	4089	10.168	3.030	2.344	16.961
Inv	4089	0.058	0.050	0.001	0.242
R_D	3893	0.039	0.032	0.000	0.200
TOP1	4089	0.333	0.143	0.087	0.726
Growth	4089	0.153	0.313	−0.454	1.657
ROA	4089	0.051	0.076	−0.279	0.274
Return	4089	0.076	0.475	−0.603	2.060
LEV	4089	0.407	0.184	0.067	0.859
FirmSize	4089	9.739	0.547	8.764	11.261
SOE	4089	0.277	0.447	0	1
Age	4089	12.987	8.330	1	30
SA	4089	−3.933	0.226	−4.534	−3.397
ManCost_ratio	4089	0.797	1.833	−4.058	12.411
LifeCycle	4081	1.789	0.727	1	3

reports the descriptive statistical results of the variables. The mean value of TFP_LP is 10.583, and the standard deviation of 0.817. It indicates that the differences between businesses are relatively minor. The mean value of LnEnvtax is 10.168, maximum value 16.961, minimum value 2.344, and standard deviation 3.030, suggesting the firms’ environmental tax burden is significantly different. The results for the control variables are similar to those in prior literature [11,12,13].

4.2. Correlation Analysis of Variables

Table A2. Correlations for the Variables.

	TFP_LP	LnEnvtax	TOP1	Growth	ROA	Return	LEV	FirmSize	SOE	Age
TFP_LP	1
LnEnvtax	0.470 ***	1
TOP1	0.158 ***	0.144 ***	1
Growth	0.196 ***	0.045 ***	0.024	1
ROA	0.217 ***	0.049 ***	0.159 ***	0.295 ***	1
Return	0.071 ***	0.016	−0.006	0.250 ***	0.209 ***	1
LEV	0.403 ***	0.304 ***	0.013	0.054 ***	−0.342 ***	0.015	1
FirmSize	0.767 ***	0.600 ***	0.171 ***	0.089 ***	0.060 ***	0.050 ***	0.500 ***	1
SOE	0.247 ***	0.352 ***	0.230 ***	−0.035 **	−0.048 ***	−0.013	0.210 ***	0.382 ***	1
Age	0.339 ***	0.415 ***	−0.005	−0.086 ***	−0.093 ***	0.020	0.279 ***	0.468 ***	0.508 ***	1

Note: The table presents a matrix of Pearson correlation coefficients. *** p < 0.01; ** p < 0.05.

in Appendix B presents correlations for the variables. In response to prior literature [11,12,13] shedding light on the effects of the environmental tax burden on firms’ TFP, we first analyzed the relationship between the environmental tax burden and firms’ TFP, measured by the amount of environmental tax paid by firms. We find that the coefficient of environmental tax burden on TFP is 0.470, which is significantly positive at the 1% level, indicating that environmental tax burden is positively associated with firms’ TFP, without considering control variables. Besides, the results indicate that the majority of the bivariate correlations are lower than 0.3. Moreover, we further check for collinearity with Variance Inflation Factors (VIF). The highest VIF is far less than the cut-off points of 10 for all the models. These results indicate that potential multicollinearity could not impact the coefficients of the independent variable.

4.3. Basic Regression Analysis

Table 2. Basic Regression Results.

Dependent Variable	(1)	(2)	(3)	(4)
	TFP_LP	TFP_LP	TFP_LP	TFP_LP
LnEnvtax	0.134 ***	0.014 ***	0.134 ***	0.014 **
	(35.18)	(4.39)	(18.91)	(2.44)
TOP1		0.217 ***		0.217 **
		(3.93)		(2.30)
Growth		0.225 ***		0.225 ***
		(8.69)		(7.62)
ROA		2.001 ***		2.001 ***
		(17.37)		(10.26)
Return		−0.056 ***		−0.056 ***
		(−3.02)		(−3.75)
LEV		0.505 ***		0.505 ***
		(9.70)		(5.13)
FirmSize		1.022 ***		1.022 ***
		(49.66)		(26.23)
SOE		−0.019		−0.019
		(−0.93)		(−0.50)
Age		0.002 *		0.002
		(1.75)		(0.91)
SA		0.024		0.024
		(0.63)		(0.38)
_cons	9.403 ***	0.149	9.403 ***	0.149
	(84.34)	(0.56)	(56.22)	(0.32)
n	4089	4089	4089	4089
Adjusted R2	0.290	0.683	0.290	0.683
Year FE	Yes	Yes	Yes	Yes
Industry FE	Yes	Yes	Yes	Yes
Firm FE			Yes	Yes

Notes: Standard errors are in parentheses. *** p < 0.01; ** p < 0.05; * p < 0.10.

shows the results of the effect of environmental tax burden on firms’ TFP. In columns (1) and (2), the influence coefficient of the environmental tax burden on TFP is significantly positive at 1%, meaning that the pressure of environmental protection tax stimulates firms to improve TFP, thus verifying Hypothesis 1a. Columns (3) and (4) are fixed-effect results, and standard errors are clustered at the firm level. The influence coefficients of the environmental tax burden on TFP are significantly positive at 1% and 5%, indicating that the conclusion is robust. Judging from the results, prior literature suggests that market-based regulations, compared to administrative command-and-control regulations, are more effective in achieving the double dividend effect, that is, reducing pollution while increasing production capacity [15]. China’s environmental tax burden is positively associated with firms’ TFP. This suggests that China’s environmental tax system has a certain “double dividend” effect [19,20,21]. To some extent, this provides some support for the “Porter hypothesis” and the “factor endowment hypothesis”.

4.4. Robustness Checks

We test the robustness of our findings using several methods. First, we replace the independent variable with the environmental tax rate (Envtax) and find that the Envtax is significantly positively correlated with firms’ TFP, with a coefficient of 3.664 at the 1% level. Second, we replace TFP_LP with TFP_OP and find that the impact coefficient of the environmental tax burden on firm TFP_OP is 0.029, significantly positive at the 1% level, indicating a robust conclusion. Third, we use a fixed effects model to control for firm-level factors that may affect both the environmental tax burden and TFP. The results are consistent with previous results; model setting doesn’t affect regression results in this study. Fourth, we consider the potential lag effect of the environmental tax burden on firms’ TFP and test the model with a one-year lag, but the conclusion remains unchanged. Finally, following Mugerman et al. (2022) [50,51], we examined the effect of the environmental tax burden on firms’ TFP with double-clustering at the firm and time levels, and the results are reported in the robustness checks section. All robustness tables are displayed in Appendix C.

5. Further Analysis

The previous results have shown that the environmental tax burden promotes firms’ total factor productivity. This section will delve into how the environmental tax burden stimulates total factor productivity and examine whether its impact varies.

5.1. Response Strategy: Fixed Asset Investment or Technological Innovation?

The environmental tax represents a common institutional pressure pushing firms to shift their focus from operational performance to compliance. To address this, firms typically adopt two strategies. One is an immediate and passive strategy, where they purchase more environmentally friendly production equipment and materials to reduce pollution and tax costs, reflected in increased fixed asset investment. This improves capital quality and TFP. The other is a more sustainable and proactive strategy, where they invest in longer-term technological innovation in response to increased environmental protection expenditures, advancing TFP. Therefore, the question is how do Chinese enterprises choose strategic operational decisions to improve production efficiency?

We represent the fixed asset investment by the ratio of cash paid by enterprises for the purchase and construction of fixed assets, intangible assets, and other long-term assets to the total assets at the end of the year. We represent the level of technological innovation by the ratio of R&D investment to operating revenue. Subsequently, we introduce fixed asset investment and technological innovation as mediating variables into the model to examine the mediating roles of technological innovation and fixed asset investment in the relationship between environmental tax burden and TFP.

Table 3. Mediating Analysis Results.

Dependent Variable	(1)	(2)	(3)	(4)
	Inv	TFP_lp	R_D	TFP_lp
LnEnvtax	0.001 ***	0.013 ***	−0.002	0.004
	(3.41)	(4.57)	(−1.28)	(1.14)
Inv		1.308 ***
		(8.08)
R_D				5.797 ***
				(21.66)
TOP1	0.011 **	0.232 ***	−0.012 ***	0.154 ***
	(2.04)	(4.48)	(−3.74)	(3.16)
Growth	0.006 **	0.233 ***	−0.001	0.205 ***
	(2.20)	(8.07)	(−0.77)	(7.27)
ROA	0.071 ***	2.094 ***	−0.058 ***	1.661 ***
	(6.69)	(15.04)	(−5.78)	(12.98)
Return	−0.000	−0.056 ***	0.001	−0.047 ***
	(−0.08)	(−3.14)	(0.87)	(−2.77)
LEV	0.031 ***	0.545 ***	−0.030 ***	0.359 ***
	(6.13)	(9.42)	(−8.37)	(6.47)
FirmSize	0.005 **	1.028 ***	0.003 ***	1.036 ***
	(2.21)	(47.69)	(2.58)	(50.13)
SOE	−0.009 ***	−0.031	−0.003 ***	−0.028
	(−5.14)	(−1.52)	(−3.13)	(−1.38)
Age	−0.002 ***	−0.000	−0.001 ***	−0.000
	(−14.68)	(−0.15)	(−6.76)	(−0.23)
_cons	0.011	0.165	0.063 ***	0.478 **
	(0.38)	(0.66)	(3.68)	(2.03)
n	4089	4089	3895	3895
adj. R2	0.175	0.688	0.281	0.728
Year FEs	Yes	Yes	Yes	Yes
Industry FEs	Yes	Yes	Yes	Yes

Notes: Standard errors are in parentheses. *** p < 0.01; ** p < 0.05.

presents the results of the mediating analysis for Hypothesis 2 and Hypothesis 3, elucidating. Columns (1) and (2) illustrate the mediating effect of fixed asset investment. The results indicate that the environmental tax burden positively influences fixed asset investment at the 1% significance level. Both the environmental tax burden and fixed asset investment have a positive impact on TFP, with the coefficient of the environmental tax burden’s effect on TFP being 0.013, which is a decrease compared to the baseline regression results. This suggests that fixed asset investment partially mediates the enhancement of total factor productivity by the environmental tax burden. Columns (3) and (4) display the mediating effect of technological innovation. The findings show no significant relationship between the environmental tax burden and technological innovation, indicating that the mediating effect of technological innovation on the impact of the environmental tax burden on enterprise production efficiency has not yet been demonstrated.

Unlike previous studies [29,30,31,32,52], although we found that the environmental tax burden enhances total factor productivity, upon closer examination of the impact pathway, the influence of China’s environmental tax burden on TFP does not occur through technological innovation but is primarily through fixed asset investments. Therefore, this is an important theoretical contribution of the paper, as we are providing evidence that diverges from the “Porter Hypothesis”.

5.2. Heterogeneous Effects

We also considered that the impact of environmental tax burden on a firm’s total factor productivity may be influenced by various factors, so we conducted a heterogeneous analysis of their relationship under different conditions.

On one hand, the relationship between environmental tax burden and firms’ TFP is influenced by explicit firm characteristics such as size, enterprise nature, and lifecycle stage. It is also affected by implicit firm characteristics like financing constraints and information costs. On the other hand, the relationship is also affected by regional characteristics. Therefore, we conducted a relevant heterogeneous analysis of the relationship between the environmental tax burden and the firms’ TFP.

5.2.1. The Firm Size and Nature of Ownership Heterogeneous

According to existing research [53,54], firm size is posited to exert a significant influence on fixed asset investment. The firm improves on production efficiency as it grows. Large firms may have more funds and greater flexibility for R&D or fixed asset investment, thereby promoting TFP. Columns (1) and (2) of

Table 4. The Firm Size and Nature of Ownership Heterogeneous.

	(1) Large Firms	(2) Small Firms	(3) State-Owned Firms	(4) Non-State-Owned Firms
LnEnvtax	0.020 **	0.011 ***	0.010	0.014 ***
	(2.50)	(3.07)	(1.44)	(3.74)
TOP1	0.184 *	0.251 ***	0.321 ***	0.177 ***
	(1.79)	(4.08)	(3.52)	(2.83)
Growth	0.249 ***	0.207 ***	0.281 ***	0.191 ***
	(5.65)	(5.74)	(4.95)	(5.68)
ROA	1.410 ***	2.118 ***	1.624 ***	2.122 ***
	(4.21)	(14.16)	(6.36)	(13.46)
Return	−0.044	−0.046 **	−0.057	−0.041 **
	(−1.34)	(−2.16)	(−1.55)	(−1.99)
LEV	0.429 ***	0.538 ***	0.484 ***	0.525 ***
	(3.16)	(8.45)	(4.39)	(7.78)
FirmSize	0.980 ***	0.991 ***	1.058 ***	1.003 ***
	(12.92)	(32.03)	(24.01)	(38.17)
SOE	0.024	−0.042 *	0.000	0.000
	(0.64)	(−1.67)	(.)	(.)
Age	0.001	0.002	0.005 *	−0.000
	(0.43)	(1.01)	(1.92)	(−0.27)
SA	0.110	−0.020	0.066	−0.006
	(0.98)	(−0.47)	(0.80)	(−0.14)
_cons	0.879	0.299	−0.177	0.276
	(0.83)	(0.99)	(−0.29)	(0.96)
n	1024	3065	1133	2956
adj. R2	0.547	0.482	0.739	0.635
Year Fes	Yes	Yes	Yes	Yes
Industry Fes	Yes	Yes	Yes	Yes
p-value	0.000 ***		0.060 *

Note: “p-value” is used to test the significance of the difference in LnEnvtax coefficient between groups, which is obtained by bootstrap sampling 1000 times. *** p < 0.01; ** p < 0.05; * p < 0.10.

show the relationship between environmental tax burden and TFP differs by firm size. Compared to small firms, large firms show a more significant positive effect of environmental tax burden on TFP.

Additionally, compared to state-owned firms, non-state-owned firms typically have a weaker awareness of environmental and social responsibilities, but they exhibit better policy sensitivity [55]. Columns (3) and (4) of Table 4 show the relationship differs by ownership. In non-state-owned firms, the environmental tax burden is significantly positively correlated with TFP at the 1% level. This relationship is insignificant in state-owned firms. Bootstrap p-values confirm the significance of these differences: 0.000 and 0.060 for the two groups, respectively, indicating significance at the 1% and 10% levels.

5.2.2. Financing Constraints and Transaction Costs Heterogeneous

Fixed asset spending may be sensitive to the availability of internal finance [55]. We assume that financing constraints are negatively associated with fixed asset investment, thereby reducing the effect of the environmental tax burden on TFP. Based on existing research [56,57], there are various measurement variables for financing constraints, such as the Z-score, SA index, KZ index, WW index, etc. This study adopts the SA index as a substitute for financing constraints, and the results of other measurement variables are not significantly different. The results in

Table 5. Financing Constraints and Transaction Costs Heterogeneous.

	(1) Strong Financing Constraints	(2) Weak Financing Constraints	(3) High Transaction Costs	(4) Low Transaction Costs
LnEnvtax	0.005	0.025 ***	0.000	0.027 ***
	(1.15)	(5.33)	(0.09)	(5.42)
TOP1	0.202 ***	0.255 ***	0.274 ***	0.135 *
	(2.64)	(3.54)	(3.77)	(1.83)
Growth	0.245 ***	0.191 ***	0.237 ***	0.201 ***
	(5.87)	(5.07)	(4.86)	(5.55)
ROA	2.042 ***	2.110 ***	3.616 ***	1.962 ***
	(11.79)	(10.06)	(9.44)	(12.58)
Return	−0.060 **	−0.049 *	−0.048 *	−0.067 ***
	(−2.39)	(−1.91)	(−1.78)	(−2.79)
LEV	0.501 ***	0.525 ***	0.517 ***	0.540 ***
	(6.43)	(6.17)	(6.33)	(6.47)
FirmSize	0.997 ***	0.995 ***	1.076 ***	0.971 ***
	(28.62)	(32.37)	(35.16)	(30.90)
SOE	0.012	−0.050	0.013	−0.042
	(0.47)	(−1.46)	(0.44)	(−1.45)
Age	0.002	0.005 *	0.005 ***	0.000
	(1.31)	(1.80)	(2.80)	(0.04)
SA	0.080	−0.116	−0.027	0.095 *
	(1.08)	(−1.37)	(−0.56)	(1.85)
_cons	0.707	−0.265	−0.601 *	0.857 **
	(1.55)	(−0.59)	(−1.71)	(2.27)
n	2044	2041	2045	2044
adj. R2	0.616	0.737	0.675	0.690
Year Fes	Yes	Yes	Yes	Yes
Industry Fes	Yes	Yes	Yes	Yes
p-value	0.000 ***		0.000 ***

Note: “p-value” is used to test the significance of the difference in LnEnvtax coefficient between groups, which is obtained by bootstrap sampling 1000 times. *** p < 0.01; ** p < 0.05; * p < 0.10.

, columns (1) and (2), show that the environmental tax burden significantly promotes TFP more for firms with weaker financing constraints, indicating that it forms a constraint on capital for those with stronger constraints, but those with weaker constraints can easily overcome this.

Similarly, fixed assets spending may be sensitive to the transaction costs. Transaction costs are the expenses required to obtain accurate market information, as well as the costs associated with negotiation and ongoing contractual agreements. Although the transaction costs in the investment market are implicit, they may significantly erode the expected excess returns on investment [58]. Referring to existing literature [59,60], we measure transaction costs by the proportion of management expenses to examine the interactive relationship between transaction costs, environmental tax burden, and TFP. Columns (3) and (4) show that the relationship between environmental tax burden and TFP differs under different transaction cost levels. In firms with low transaction costs, the environmental tax burden is significantly positively correlated with TFP at 1%, while not significant in those with high transaction costs. Bootstrap empirical p values confirm the significance of these differences, both of which at 0.000, which are significant at the 1% level.

5.2.3. Heterogeneous in Corporate Life Cycle

Based on the literature of Dickinson (2011) and others [61,62], we use the cash flow model method to categorize the sample into growth, mature, and decline stages.

Table 6. Heterogeneous in Corporate Life Cycle.

	(1) Growth	(2) Mature	(3) Decline
LnEnvtax	0.018 ***	0.007 *	0.022 ***
	(2.97)	(1.65)	(2.63)
TOP1	0.137	0.189 **	0.427 ***
	(1.59)	(2.53)	(3.20)
Growth	0.197 ***	0.272 ***	0.220 ***
	(4.18)	(6.52)	(2.75)
ROA	2.167 ***	2.036 ***	1.718 ***
	(7.66)	(11.69)	(5.46)
Return	−0.048	−0.054 **	−0.084 *
	(−1.59)	(−2.10)	(−1.86)
LEV	0.568 ***	0.667 ***	0.291 **
	(5.21)	(8.29)	(2.15)
FirmSize	1.017 ***	1.020 ***	1.024 ***
	(27.19)	(33.54)	(18.11)
SOE	0.033	−0.042	−0.050
	(0.84)	(−1.58)	(−0.99)
Age	0.005 **	0.003	−0.009 ***
	(2.25)	(1.40)	(−3.00)
SA	0.067	0.055	−0.211 **
	(1.14)	(1.09)	(−2.14)
_cons	0.248	0.299	−0.642
	(0.58)	(0.80)	(−1.07)
n	1599	1741	737
adj. R2	0.645	0.741	0.626
Year Fes	Yes	Yes	Yes
Industry Fes	Yes	Yes	Yes
p-value	0.015 **	0.000 ***	0.000 ***

Note: “p-value” is used to test the significance of the difference in LnEnvtax coefficient between groups, which is obtained by bootstrap sampling 1000 times. *** p < 0.01; ** p < 0.05; * p < 0.10.

illustrates that the impact of the environmental tax burden on TFP in the mature stage is less than that in the growth and decline stages. This suggests that stable and profitable firms, having more abundant capital, are not significantly affected by the environmental tax burden on TFP. The bootstrap method confirms the significance of these differences by comparing the significance of the LnEnvtax coefficient differences between growth and non-growth, mature and non-mature, and recession and non-recession stages. The p-values are 0.015, 0.000, and 0.000, respectively, indicating significance at the 5%, 1%, and 1% levels.

5.2.4. Heterogeneous in Regional

As the prior literature pointed out, locational advantages and governance institutions, as economic factors, will affect the production and investment environment of various regions [22].

Table 7. Heterogeneous in Regional.

	(1) East	(2) Middle	(3) West
LnEnvtax	0.017 ***	0.031 ***	−0.007
	(4.28)	(4.24)	(−0.79)
TOP1	0.197 ***	0.211 *	0.197
	(3.15)	(1.81)	(1.32)
Growth	0.222 ***	0.305 ***	0.137 *
	(6.43)	(4.96)	(1.90)
ROA	2.031 ***	1.721 ***	2.476 ***
	(12.15)	(5.93)	(8.33)
Return	−0.052 **	−0.049	−0.047
	(−2.31)	(−1.28)	(−1.05)
LEV	0.514 ***	0.344 **	0.553 ***
	(7.32)	(2.51)	(3.72)
FirmSize	1.046 ***	0.764 ***	1.134 ***
	(39.90)	(13.07)	(21.45)
SOE	−0.063 **	0.010	0.168 ***
	(−2.34)	(0.26)	(3.35)
Age	0.002	0.019 ***	−0.001
	(1.15)	(5.78)	(−0.29)
SA	−0.017	0.253 ***	0.031
	(−0.39)	(2.87)	(0.31)
_cons	−0.227	3.142 ***	−0.822
	(−0.75)	(4.65)	(−1.14)
n	2798	753	536
adj. R2	0.678	0.706	0.760
Year FEs	Yes	Yes	Yes
Industry FEs	Yes	Yes	Yes
p-value	0.165	0.000 ***	0.000 ***

Note: “p-value” is used to test the significance of the difference in LnEnvtax coefficient between groups, which is obtained by bootstrap sampling 1000 times. *** p < 0.01; ** p < 0.05; * p < 0.10.

illustrates the regional differences in the impact of the environmental tax burden on TFP. From the point of regional differences, the east and middle regions have a weak but not strong Porter hypothesis phenomenon while that of the west region is not significant. Our result is similar to the prior literature [52].

6. Discussion and Conclusions

This section presents empirical conclusions and proposes practical suggestions for firms to address environmental taxes based on the main research findings. It also discusses the limitations of the study and suggests future research directions.

6.1. Discussion

This study examined the relationship between environmental tax burden and firms’ total factor productivity using panel data from listed companies on the Shanghai and Shenzhen Stock Exchanges from 2018–2022. This study has the following findings.

First, the environmental tax burden positively affects firms’ TFP. This indicates that although environmental taxes increase compliance costs for firms, they also enhance their environmental protection efforts and promote green transformation.

Second, the way in which environmental tax pressure affects the production efficiency of firms is through fixed asset investment rather than technological innovation. This implies that environmental regulations force firms to invest in fixed assets to meet emission standards in the short term. We believe that, based on institutional theory, technological innovation is more aligned with proactive strategies and represents a long-term and sustainable environmental strategy. In contrast, fixed asset investment is more of a reactive strategy, serving as a short-term, real-time environmental strategy. Therefore, Chinese firms facing the pressure of environmental taxes have only demonstrated short-term and immediate responses.

Third, when facing the pressure of environmental taxes, firms can enhance production efficiency through fixed asset investment, but such investments require substantial capital. We also discuss the heterogeneous effects on the association between environmental tax burden and firms’ total factor productivity. Results show that large-scale firms are more capable of alleviating the increased costs of environmental taxes through measures such as replacing equipment, leading to a more pronounced improvement in firms’ TFP. Secondly, non-state-owned firms are more sensitive to the increase in environmental tax costs and will take proactive measures to enhance firms’ TFP. In addition, in samples of firms with weaker financing constraints and lower transaction costs, the positive effect of the environmental tax burden on TFP is more significant. This indicates that firms need to choose business strategies according to the specific situation to improve their production efficiency.

6.2. Theoretical Contributions

Our findings contribute to the literature in the following two ways. First, previous literature has primarily discussed the economic consequences of pollution discharge fees and the economic impacts of the “fee-to-tax” event. This study uses the amount of environmental tax payment as a proxy for firms’ environmental tax burden and examines the micro-effects of environmental tax pressure on firms. This study primarily focuses on the impact of the ‘quantity’ of environmental taxes on corporate production efficiency. This work supplements the literature on environmental tax reform effects and introduces the concept of environmental tax burden and a measurement method.

Second, using empirical data from China, this study shows that environmental tax burden enhances total factor productivity primarily through reactive business strategies, such as fixed asset investments, rather than proactive business strategies, like technological innovation. This finding enriches the literature on the relationship between corporate environmental behavior and production efficiency, providing evidence outside the “Porter hypothesis”. This study complements the literature on the effectiveness of environmental tax reforms.

6.3. Managerial Implications

Practically, our findings provide managerial implications for practitioners and policymakers.

China’s implementation of the environmental tax in 2018 was a significant policy decision in achieving the 2030 carbon peak and 2060 carbon neutrality goals. Our findings provide important policy implications for policymakers. Firstly, environmental taxes can effectively enhance firms’ total factor productivity, so policymakers should actively promote the implementation of such taxation. Based on thorough justification, the level of environmental tax collection could be appropriately increased. Secondly, the intermediary mechanism through which environmental taxes affect firms’ total factor productivity is fixed asset investment. From this perspective, policymakers should focus on providing convenience for the renewal of business assets. Thirdly, policymakers should also consider that during the asset replacement process, certain small and medium-sized enterprises (SMEs), private enterprises, and enterprises facing financing difficulties may face excessive tax pressure due to the inability to complete asset replacement smoothly. Countries should actively consider providing a more convenient policy environment to facilitate financing for such entities. Finally, the research findings are also beneficial for policymakers to refine the environmental tax system, stimulate corporate innovation potential, and promote economic growth and emission reduction.

Our findings also provide important managerial implications for enterprise leaders. When facing environmental tax burdens, firms must respond with either proactive or reactive strategies in order to maintain sustainable development. Besides, firms can seize this opportunity to update their fixed assets through the imposition of environmental taxes, thereby achieving both economic and social benefits. In practice, it helps firms to identify the best strategies for coping with the environmental tax burden, thereby assisting them in achieving the dual goals of energy conservation and stable performance.

6.4. Limitations and Future Research

This paper has the following limitations. First, according to institutional theory, we categorize business strategies into proactive strategies and reactive strategies to cope with the environmental tax burden. Among them, proactive strategies such as technological innovation require a longer period of development. However, the scope of this article covers the period from 2018 to 2022, which may not be long enough. In the future, extending the sample period could allow for continued observation of innovative strategies. Second, this study does not analyze the impact of different fixed asset investments and types of innovation on the relationship between environmental tax burdens and TFP. Future research can further deepen the study and analyze more detailed scenarios in which environmental tax burdens enhance TFP.