Corporate Cash Dividends and the Environmental Protection Tax: Evidence from China

Abstract

Cash dividends, as a tangible form of monetary distribution, serve as a fundamental mechanism for remunerating investors for their capital commitments. Beyond manifesting a firm’s commitment to fulfilling its social responsibilities toward shareholders, such distributions potentially shape corporate deliberations regarding accountability toward a broader spectrum of stakeholders. Drawing on behavioral explanations of corporate decision-making, this study examines the association between cash dividend payouts and environmental protection tax burdens among Chinese A-share listed companies from 2018 to 2023. The empirical results indicate a significant and robust negative association between corporate cash dividend payouts and environmental protection tax burdens. Mechanism analysis suggests that this cross-domain behavioral consistency is primarily channeled through the proactive fulfillment of corporate environmental responsibilities. Further inquiry reveals that both government environmental subsidies and media coverage exert positive moderating effects on this relationship. Notably, this observed negative association is particularly pronounced in firms characterized by lower executive environmental awareness, those operating in regions with lenient environmental regulations, companies navigating economic downturns, and those situated within low-pollution industries. This research provides novel evidence for the “governance complementarity” hypothesis, suggesting that financial accountability and environmental stewardship are mutually reinforcing rather than mutually exclusive. Furthermore, it offers a pioneering micro-behavioral perspective on how firms in emerging economies can harmonize shareholder wealth distribution with green transition objectives.

1. Introduction

The increasing prominence of global environmental challenges, such as rising temperatures, extreme weather events, glacier melting, and sea level rise, has fostered a broader international consensus on sustainable development strategies [1,2]. According to the United Nations 2024 SDG (Sustainable Development Goal) Progress Report, only 17% of the SDGs are currently on track, while nearly 50% have deviated from their expected trajectories [3]. Environmental issues have currently ascended to the vanguard of global discourse, representing a formidable impediment to sustainable socio-economic development [4,5]. In the context of China, the rapid pace of economic expansion has frequently coincided with intensified environmental pressures. Concurrently, the escalating calls from both the public and the government for ecological stewardship have positioned pollution abatement and the cultivation of environmental consciousness as central pillars of the national strategic agenda [6]. From the macro perspective down to the micro level, the growing sustainability pressures highlight a significant challenge of balancing financial decision-making with environmental and social responsibilities faced by corporations [7]. Cash dividends are widely regarded as the most common mechanism through which firms fulfill their responsibilities to shareholders, representing a critical financial decision [8,9,10,11]. However, few studies have examined the effect of this rewarding method on corporate environmental issues [12,13,14]. Addressing this effect can provide valuable insights for aligning corporate financial strategies with broader social and environmental imperatives.

Using data from A-share listed firms in China from 2018 to 2023, this study conducts an empirical analysis and finds that increases in cash dividend payouts in a given year are significantly associated with lower environmental protection tax payments in the subsequent year. Mechanism analysis suggests that the observed negative association between proactive dividend policies and environmental protection tax liabilities is potentially channeled through the intensified fulfillment of corporate environmental responsibilities. Further analyses reveal that government environmental subsidies and media coverage significantly moderate the relationship between cash dividends and environmental protection tax. Moreover, the effect of dividend payouts on reducing environmental protection tax is particularly pronounced for firms whose executives have low environmental awareness, firms located in regions with relaxed environmental regulations, businesses experiencing economic downturns, and companies in low-pollution sectors.

This study offers several contributions. First, this study advances the micro-foundations of Corporate Social Responsibility (CSR) research by synthesizing Moral Consistency Theory and Upper Echelons Theory into a cohesive analytical framework. Unlike prior CSR studies that predominantly focus on financial performance [15,16,17,18], firm value [19,20,21], or CSR ratings [22,23], this study sheds light on the intrinsic moral motivations driving executives to exhibit consistent CSR behavior across multiple domains, thereby deepening our understanding of CSR decision-making mechanisms. Second, it expands research on the relationship between environmental protection tax and corporate behavior by proposing a novel causal pathway based on financial policy. While previous studies have primarily examined the direct effects of environmental protection tax on green innovation [24,25], environmental investment [26], and the green transformation of high-pollution firms [27], few studies have explored how internal financial policies influence firms’ environmental protection tax burdens and emission control efforts. This study offers an innovative perspective on how dividend policy, functioning as a mechanism for fulfilling shareholder obligations, may systematically align with and indirectly shape corporate environmental performance through the lens of moral consistency. Third, this study pioneers the use of firms’ environmental protection tax payments as an objective and quantitative measure of corporate environmental responsibility, thereby overcoming the limitations of subjective CSR scores or self-reported environmental investments. This approach enables more accurate assessments of firms’ actual emission control behaviors and environmental performance, offering valuable insights for sustainability research. Given China’s dynamic market economy and comprehensive environmental taxation reforms, the findings offer practical implications for other emerging economies facing similar green transition challenges. The mechanisms revealed in this study provide a transferable framework for policymakers worldwide to balance financial incentives with environmental sustainability goals.

2. Literature Review

2.1. Dividend Policy and Corporate Social Responsibility

Dividend policy is a core topic in corporate finance and has traditionally been explained by agency theory and signaling theory. Agency theory argues that cash dividends reduce managerial discretion over free cash flow and thus help mitigate overinvestment and agency conflicts [28,29]. Signaling theory suggests that dividend payouts convey positive private information about a firm’s future cash flows and profitability to the market [30,31,32]. Although these perspectives explain why firms pay dividends, they provide limited insight into how dividend policy relates to non-financial objectives such as CSR.

As ESG and sustainability concerns have become more important in corporate decision-making, the relationship between dividend policy and CSR has received growing attention. Existing studies mainly offer two contrasting views. The first is the resource crowding-out view, which argues that dividend payouts consume internal funds and may reduce resources available for environmental and social investment. For example, large dividend commitments may weaken long-term investment capacity, especially in firms with limited financial flexibility [33,34,35]. The second is the governance complementarity view, which sees dividend policy and CSR as mutually reinforcing governance mechanisms that reduce information asymmetry and help firms respond to multiple stakeholder demands [11]. Supporting this view, prior studies find that stronger CSR performance is associated with a greater likelihood of cash payouts and more stable dividend policies [9,13,36,37,38].

Despite these advances, the literature remains focused largely on aggregate CSR or ESG measures, with limited attention to environmental responsibility as a distinct dimension. As a result, it is still unclear whether dividend payouts weaken firms’ environmental commitment by tightening internal resources or strengthen it through better governance and responsibility alignment. This unresolved issue suggests the need for a more integrated explanation that goes beyond financial constraints and incorporates managerial values and behavioral logic.

2.2. Environmental Protection Tax and Corporate Environmental Responsibility

China’s Environmental Protection Tax Law (EPTL), implemented on 1 January 2018, replaced the previous pollutant discharge fee system and marked an important shift in environmental regulation. By taxing pollutant emissions, the EPTL internalizes environmental externalities and encourages firms to incorporate pollution costs into production and investment decisions [39,40]. Existing studies show that the EPTL has influenced green innovation, environmental investment, green transformation, and ESG performance, although its effects are not always uniform across firms [24,25,26,27,41,42].

Corporate environmental responsibility (CER), as a key dimension of CSR, is shaped by both internal governance factors and external pressures [43,44]. Internally, board characteristics and ownership structure can affect firms’ willingness and ability to fulfill environmental responsibilities [45,46,47,48]. Externally, regulatory enforcement, customer expectations, competition, environmental norms, and media attention all influence corporate environmental behavior [49,50].

Although research on the EPTL and CER has expanded rapidly, most studies focus on the consequences of environmental taxation itself rather than on the firm-level decisions associated with tax outcomes. In particular, limited attention has been paid to whether routine financial policies, such as dividend payouts, are related to environmental protection tax payments and whether this relationship operates through CER. This gap is important because it connects corporate finance decisions with firms’ environmental behavior under formal regulatory constraints.

2.3. Behavioral Explanations of Corporate Decision-Making

Understanding the link between dividend payouts and environmental responsibility requires a perspective that goes beyond standard rational-choice models. Traditional research often assumes that firms allocate resources purely on the basis of financial optimization [51]. However, this assumption does not fully explain why firms facing similar constraints may differ in balancing shareholder returns and environmental responsibilities. The Behavioral Theory of the Firm offers a more suitable framework by emphasizing bounded rationality, selective attention, and organizational routines [52,53,54]. From this perspective, corporate decisions reflect not only resource constraints but also managerial cognition and internal organizational processes. Moral Consistency Theory further suggests that decision makers tend to maintain consistency between prior responsible actions and subsequent behavior in order to preserve a coherent ethical image and social reputation [55,56].

To translate this logic from the individual to the organizational level, this study draws on Upper Echelons Theory (UET). UET argues that executives’ cognitive bases, values, and personal characteristics shape corporate strategies and outcomes [57]. Prior studies show that managerial traits and ethical orientations influence environmental practices and sustainability-related decisions [58,59]. When top managers view environmental responsibility as part of long-term firm value and social legitimacy, environmental issues are more likely to receive strategic attention and organizational support. Yet such influence only translates into stable corporate behavior once it is embedded within organizational processes. Individual-level ethical preferences must be internalized into organizational operations through top management team deliberation, formal decision-making procedures, and governance arrangements, thereby giving rise to organization-level sustainable strategies [60,61]. Existing studies find that managerial environmental attitudes significantly promote the adoption of proactive environmental strategies, and this effect typically operates via transmission through internal governance and implementation mechanisms [61,62]. Accordingly, once environmental responsibility is integrated into corporate governance practices, firms are more likely to exhibit a consistent responsibility orientation across distinct decision domains. Therefore, the relationship between cash dividends and environmental responsibility should not be simplistically interpreted as a resource substitution or crowding-out effect; rather, it reflects whether sustainability has been internalized as part of the firm’s governance logic and long-term strategic commitment. Corporate-level moral consistency, in this sense, is not an anthropomorphic interpretation of the organization, but the persistent manifestation of institutionalized value hierarchies across varying decision contexts.

2.4. Literature Review and Research Gap

The above review identifies three main gaps in the existing literature. First, studies on dividend policy and CSR have produced mixed findings, and most of them rely on broad CSR or ESG measures rather than environmental responsibility specifically. As a result, the relationship between dividend payouts and CER remains unclear. Second, research on the EPTL has concentrated mainly on the economic and innovation effects of environmental taxation, with limited attention to whether internal financial decisions are associated with firms’ environmental tax burden. Third, although previous studies acknowledge the role of managerial characteristics and governance structures in responsible behavior, they offer limited explanation of how responsibility-oriented managerial values are translated into consistent organizational actions across shareholder and environmental domains.

Based on these considerations, this study integrates dividend policy, corporate environmental responsibility, and the environmental protection tax into a unified analytical framework. Taking the combination of executives’ behavioral logic and governance mechanisms as its primary perspective, it examines the mechanism through which dividend payouts affect corporate environmental responsibility and environmental tax burdens. This study contributes to revealing the underlying mechanism through which dividend payouts influence corporate environmental behavior, and provides new empirical evidence for understanding the relationship between corporate financial decisions and sustainable development.

3. Theoretical Background and Hypothesis Development

3.1. Corporate Cash Dividends and Environmental Protection Tax

The environmental protection tax is an important institutional arrangement designed to internalize the negative externalities generated by corporate pollution emissions [63,64]. In the Chinese context, however, the actual tax burden may vary with regulatory enforcement intensity, regional policy differences, and industry-specific environmental standards. Therefore, a lower environmental protection tax payment does not necessarily indicate superior environmental performance in every case. Nevertheless, because the tax is mainly levied on taxable pollutant emissions, it generally reflects lower emission intensity, stronger pollution control capability, or better environmental compliance [39,42]. Against this background, this study examines whether corporate cash dividend payouts are systematically associated with firms’ environmental protection tax burdens.

To maintain theoretical focus, this study treats moral consistency theory as the primary explanatory mechanism. The core argument is straightforward. When top management chooses to distribute cash dividends, it publicly demonstrates responsibility toward shareholders through a visible financial decision. Once such a responsibility-oriented stance is expressed in one important governance domain, managers may become more inclined to maintain a consistent responsible posture in other domains, especially environmental governance. In this way, cash dividend payouts may activate a cross-domain responsibility consistency. This consistency may encourage firms to strengthen environmental management, reduce pollutant emissions, and ultimately bear lower environmental protection tax burdens. Upper echelons theory further explains how this behavioral logic is translated into firm-level outcomes through managerial cognition, value orientation, and strategic decision-making [57].

From the perspective of moral consistency theory, decision makers tend to align subsequent actions with prior morally or socially desirable choices. They do so to preserve self-consistency and maintain a coherent ethical image [65]. At the organizational level, this does not mean that firms themselves possess morality. Rather, firms’ strategies and policies are shaped by top executives’ cognition, values, and priorities. Recent studies show that executives’ ESG awareness and green cognition can significantly improve corporate green innovation and ESG performance [66,67]. These findings indicate that managerial value orientations can indeed be translated into firm-level responsible behavior through organizational decision-making processes. Accordingly, when managers demonstrate responsibility toward shareholders through cash dividend payouts, they may be less willing to tolerate an inconsistent pattern in which shareholder-oriented responsibility is emphasized while environmental responsibility is neglected. As this consistency motive is transmitted through executive decision-making, resource allocation, and internal governance arrangements, firms may become more likely to enhance environmental governance and reduce pollution-related tax burdens. Therefore, moral consistency, operating through managerial cognition and organizational processes, constitutes the dominant theoretical mechanism of this study.

This internal mechanism may be further reinforced by the firm’s external evaluative environment. In particular, signaling theory, stakeholder theory, and legitimacy theory are introduced as supporting perspectives rather than parallel core mechanisms. First, dividend payouts are often interpreted as positive signals of profitability, stability, and managerial confidence [68,69]. In increasingly ESG-sensitive capital markets, however, the credibility of this signal may weaken if firms reward shareholders while performing poorly in environmental governance [70,71,72]. Second, firms that distribute cash dividends may face broader expectations regarding responsible environmental conduct from regulators, communities, consumers, employees, and institutional investors [73,74,75,76,77]. Third, under rising environmental scrutiny, firms are also under pressure to align their actions with prevailing social norms and institutional expectations in order to maintain legitimacy [78,79,80,81,82]. Taken together, these perspectives help explain why consistency between shareholder-oriented financial decisions and environmental responsibility becomes more visible and more consequential in practice. However, they do not replace the primary moral consistency mechanism.

Overall, the theoretical logic of this study centers on a responsibility-consistency mechanism. Cash dividend payouts represent a visible shareholder-oriented responsible action. Managers may then strengthen environmental governance to maintain consistency in responsible conduct. External market and institutional pressures further reinforce this tendency. Accordingly, firms with higher cash dividend payouts are expected to bear lower environmental protection tax burdens.

Hypothesis 1: 

Corporate cash dividend distribution is negatively associated with corporate environmental protection tax payments.

3.2. Corporate Cash Dividends, Environmental Responsibility, and Environmental Protection Tax

CER may serve as an important channel through which cash dividends are associated with environmental protection tax. In this study, CER refers to a firm’s substantive commitment to environmental governance, environmental protection investment, environmental information disclosure, and environmental compliance management [83,84]. This definition is consistent with the empirical design of the present study and with the broader ESG literature, which regards CER as an important component of long-term value creation and corporate risk management [70]. If cash dividend payouts are associated with stronger CER, and stronger CER in turn leads to better pollution control and lower taxable emissions, then CER may mediate the relationship between cash dividends and environmental protection tax.

Consistent with the theoretical logic developed in Section 3.1, this study treats CER as the organizational manifestation and transmission channel of the responsibility-consistency mechanism activated by cash dividend payouts. The core idea is that when managers use cash dividends to demonstrate responsibility toward shareholders, moral consistency theory suggests that they may become more inclined to maintain a similarly responsible posture in environmental decision-making [65]. Upper echelons theory further explains how this value orientation is translated into firm-level strategies through managerial cognition, strategic priorities, and governance arrangements [57]. Accordingly, once top executives adopt a shareholder-oriented responsible action through dividend distribution, they may also strengthen CER to avoid inconsistency between financial responsibility and environmental conduct. As this responsibility orientation becomes embedded in organizational routines and governance practices, firms may become more likely to increase environmental investment, improve environmental management, enhance environmental disclosure, and strengthen compliance with environmental regulations.

The development of CER may also be reinforced by the firm’s external evaluative environment. As discussed in Section 3.1, signaling theory, stakeholder theory, and legitimacy theory are not treated here as separate core mechanisms. Instead, they are supporting perspectives that help explain why responsibility consistency becomes more consequential in practice. In increasingly ESG-sensitive capital markets, firms are expected not only to reward shareholders but also to demonstrate credible environmental responsibility. Likewise, broader stakeholder scrutiny and legitimacy pressures make it more costly for firms to maintain shareholder-oriented financial policies while neglecting environmental obligations. Under these conditions, managers may have stronger incentives to translate the responsibility signal conveyed by cash dividends into substantive CER fulfillment.

Stronger CER, in turn, may lead to lower environmental protection tax payments by improving pollution control and reducing taxable emissions. Firms with greater CER are more likely to adopt cleaner production processes, improve pollution treatment capacity, reduce pollutant emissions, and comply more effectively with environmental regulations [85,86,87]. Because environmental protection tax is largely tied to the level of taxable emissions [42], these improvements should be reflected in a lower tax burden. In this way, CER provides a plausible and observable organizational mechanism linking a shareholder-oriented financial decision to a firm’s environmental tax outcome.

Overall, cash dividend payouts may promote CER by activating a responsibility-consistency logic within the firm and by intensifying the market, stakeholder, and legitimacy pressures for such consistency. Improved CER may subsequently reduce taxable pollutant emissions and environmental protection tax payments. This leads to the following hypothesis:

Hypothesis 2: 

Corporate environmental responsibility mediates the negative association between corporate cash dividend distribution and environmental protection tax payments.

4. Data and Methods

4.1. Data and Sample

This paper initially selects A-share listed companies on the Shanghai and Shenzhen stock exchanges that disclosed environmental protection tax payments between 2019 and 2023, matched with their corresponding cash dividend data from 2018 to 2022. The sample excludes financial companies due to their distinct financial statement structures, firms experiencing abnormal operational conditions (e.g., ST and *ST), and observations with missing data. Additionally, all continuous variables undergo winsorization at the 1% and 99% percentiles to mitigate the potential influence of outliers. After the above screening and preprocessing steps, a final sample of 6018 firm-year observations was obtained. All data utilized in this study were obtained from the China Stock Market and Accounting Research (CSMAR) database, which is acknowledged by academic journals on Chinese economic, accounting, and financial data.

4.2. Variables

4.2.1. Dependent Variable

The environmental protection tax is a tax imposed or waived on individuals and organizations based on their levels of pollution or environmental stewardship in relation to resource development and utilization, environmental degradation, or conservation efforts. This tax can be categorized into broad and narrow definitions. The Organization for Economic Co-operation and Development (OECD) defines the broad environmental protection tax as encompassing all tax types associated with environmental protection. Meanwhile, the narrow definition pertains specifically to independent taxes that are directly related to environmental protection initiatives. In the context of China, the narrowly defined environmental protection tax pertains to the tax officially instituted in 2018, which replaced the earlier pollution discharge fee. This tax represents the sole green tax in the country and is applicable exclusively to enterprises that directly emit taxable pollutants, which include air and water pollutants, solid waste, and industrial noise. Additionally, public treatment facilities that exceed established emission standards are also subject to this tax. The dependent variable analyzed in this study, denoted as LnET, specifically addresses the narrowly defined environmental protection tax. It is calculated according to the formula presented in model (1), where $Emissio{n}_{itk}$ signifies the quantity of the $k$ type of taxable pollutant discharged by enterpris $i$ in year $t$, and $Local\_Rat{e}_k$ refers to the applicable tax rate for the enterprise’s location. This dependent variable effectively quantifies the environmental cost intensity incurred by enterprises as a result of their actual emissions.

$$LnE{T}_{it}=\ln \left(1+{\displaystyle \sum _{k=1}^K\left(Emissio{n}_{itk}\times Local\_Rat{e}_k\right)}\right)$$

(1)

4.2.2. Independent Variables

Cash Dividend (Div) is operationalized as the ratio of cash dividends per share to total assets per share [88,89]. This measure effectively captures the return on investment from an accounting perspective. Considering that Chinese enterprises typically distribute cash dividends in the first half of each fiscal year, the independent variable is lagged by one period to fully capture its impact on subsequent environmental protection tax payments.

4.2.3. Control Variables

This paper selects the following seven control variables that may affect corporate environmental protection tax: (1) Company fundamentals: firm size (Size), leverage (Lev), profitability (ROA), cash flow (Cashflow), Tobin’s Q (TobinQ). (2) Corporate governance: independent director (Indep), equity concentration (Top1). Additionally, industry (Industry) and year (Year) fixed effects are incorporated into the models. Moreover, considering that environmental protection tax may capture institutional and regulatory differences rather than solely firm-driven environmental performance, this study further includes province (Province) fixed effects to account, to some extent, for regional heterogeneity in regulation and policy. Province fixed effects absorb province-specific environmental policy baselines and enforcement intensity, thereby mitigating, to some degree, omitted variable bias stemming from regional differences in regulation or policy. To comprehensively capture the effects of corporate cash dividend distribution on environmental protection tax payments, all control variables are also lagged by one period. Detailed definitions of all variables are provided in Table A1 in Appendix A.

4.3. Model Design

To empirically investigate the impact of corporate cash dividends on environmental protection tax payments, this study establishes the following regression model:

$$\begin{array}{cc}LnE{T}_{i,t}& ={\alpha }_0+{\alpha }_{1}Di{v}_{i,t-1}+{\alpha }_{2}Control{s}_{i,t-1}+YearFixedEffect\\ & +IndustryFixedEffect+ProvinceFixedEffect+{\epsilon }_{i,t}\hfill \end{array}$$

(2)

In model (2), the dependent variable (LnET) denotes the environmental protection tax payments of firm i in year t. The independent variable (Div) represents the cash dividends of firm i in year t − 1, thereby offering a more comprehensive reflection of how corporate cash dividend payments are associated with environmental protection tax payments. In addition, the model controls for year, industry, and province fixed effects, while ${\epsilon }_{i,t}$ represents the random error term.

To further examine H2, this study follows the methodological framework of Baron and Kenny [90] and specifies Models (3) and (4) to investigate the potential mechanism through which corporate cash dividends are associated with environmental protection tax:

$$\begin{array}{cc}CE{R}_{i,t}& ={\beta }_0+{\beta }_{1}Di{v}_{i,t-1}+{\beta }_{2}Control{s}_{i,t-1}+YearFixedEffect\\ & +IndustryFixedEffect+ProvinceFixedEffect+{\epsilon }_{i,t}\hfill \end{array}$$

(3)

$$\begin{array}{cc}LnE{T}_{i,t}& ={\delta }_0+{\delta }_{1}Di{v}_{i,t-1}+{\delta }_{2}CE{R}_{i,t}+{\delta }_{3}Control{s}_{i,t-1}\\ & +YearFixedEffect+IndustryFixedEffect\hfill \\ & +ProvinceFixedEffect+{\epsilon }_{i,t}\hfill \end{array}$$

(4)

Specifically, Model (3) is employed to test the effect of Div on the mediating variable (CER). Model (4) is then used to examine, after controlling for the mediating variable, the direct effect of Div on LnET and the effect of CER on environmental protection tax. $CE{R}_{i,t}$ represents the environmental responsibility performance of firm i in year t, and all other variables are defined consistently with those in the main regression Model (2).

5. Empirical Results

5.1. Descriptive Statistics

Table 1 reports the descriptive statistics of the main variables used in this study. The mean and median values of environmental protection tax (LnET) are 11.46 and 11.47, respectively, suggesting that the distribution of this variable is broadly centered and does not exhibit pronounced asymmetry. At the same time, the relatively large standard deviation (2.986), together with the wide range from 0 to 18.180, indicates substantial cross-firm heterogeneity in environmental tax burdens. This variation is economically meaningful, as it suggests that sample firms differ markedly in pollutant emissions, environmental governance, or compliance conditions. The mean value of cash dividends (Div) is 0.016, implying that, on average, firms distribute cash dividends equivalent to 1.6% of total assets. This magnitude suggests that dividend payouts are non-trivial in economic terms. The median value (0.009) is lower than the mean, indicating a right-skewed distribution in which a relatively small number of firms pay comparatively large dividends. Such dispersion provides a relevant basis for examining whether differences in dividend policy are systematically associated with differences in environmental protection tax outcomes. The descriptive patterns of the remaining control variables are broadly consistent with those reported in prior studies.

Table 1. Summary Statistics of Main Variables.

Variable	N	Mean	p50	SD	Min	Max
LnET	6018	11.460	11.470	2.986	0	18.180
Div	6018	0.016	0.009	0.021	0	0.105
Size	6018	22.410	22.270	1.255	20.050	26.060
Lev	6018	0.411	0.410	0.186	0.065	0.886
ROA	6018	0.060	0.056	0.071	−0.203	0.266
Cashflow	6018	0.059	0.057	0.064	−0.115	0.241
TobinQ	6018	1.853	1.467	1.148	0.796	7.101
Indep	6018	37.710	36.360	5.285	33.330	57.140
Top1	6018	0.334	0.308	0.146	0.087	0.746

Appendix B reports the subgroup comparison results based on the amount of cash dividends distributed. We divide the full sample into two subsamples according to the mean value of dividends and then test the differences in means and medians between these two groups (t-test and z-test). The results show that both tests yield significantly negative values (−2.01/−2.64), indicating that the environmental protection taxes for the high-dividend group are lower than those for the low-dividend group, initially supporting H1. Additionally, Appendix C presents the correlation coefficients for all variables and shows a negative correlation between LnET and Div.

5.2. Baseline Regression Results

Table 2 presents the baseline regression results on the relationship between cash dividends and corporate environmental protection tax. To examine the stability of the estimates, we adopt a stepwise specification strategy by progressively adding sets of control variables. Column (1) includes only industry, year, and province fixed effects, and the coefficient on Div is negative (−4.204) but statistically insignificant. After adding firm-level financial controls in Column (2), the coefficient on Div becomes −8.053 and is significant at the 1% level. Column (3) further incorporates corporate governance controls, and the coefficient on Div remains negative and statistically significant at −8.200.

Table 2. Baseline Regression Analysis.

	(1)	(2)	(3)
	LnET	LnET	LnET
Div	−4.204	−8.053 ***	−8.200 ***
	(−1.508)	(−2.886)	(−2.944)
Size		1.210 ***	1.208 ***
		(24.594)	(24.539)
Lev		0.056	0.051
		(0.165)	(0.149)
ROA		1.665 **	1.649 **
		(2.240)	(2.217)
Cashflow		2.411 ***	2.403 ***
		(3.430)	(3.421)
TobinQ		−0.251 ***	−0.252 ***
		(−5.161)	(−5.167)
Indep			0.008
			(0.918)
Top1			0.135
			(0.385)
_cons	11.383 ***	−15.462 ***	−15.759 ***
	(165.245)	(−14.561)	(−14.330)
Year	Yes	Yes	Yes
Industry	Yes	Yes	Yes
Province	Yes	Yes	Yes
Observations	6018	6018	6018
Adjusted R2	0.320	0.487	0.487

Note: The symbols **, and *** denote statistical significance at the 10, 5, and 1% levels, respectively. All reported t-statistics are based on standard errors adjusted for clustering at the firm level [91].

Taken together, these results indicate a stable negative association between cash dividend payments and environmental protection tax burdens after accounting for a richer set of observable firm characteristics. This finding directly relates to our first research question, namely whether firms with higher cash dividend payouts tend to face lower subsequent environmental protection tax burdens. The consistently negative coefficient across specifications suggests that such a relationship is not driven solely by a parsimonious model setting.

The estimated magnitude is also economically meaningful. In Column (3), the coefficient on Div is −8.200. Because Div is measured as cash dividends scaled by total assets, a 1 percentage point increase in Div (that is, an increase of 0.01) is associated with a 0.082 decline in LnET. Since the dependent variable is in logarithmic form, this coefficient can be interpreted approximately as an 8.2% reduction in environmental protection tax payments. Put differently, if two otherwise similar firms differ in dividend payouts by 1 percentage point of total assets, the firm with the higher dividend payout is expected to have an environmental protection tax burden that is about 8.2% lower on average. This effect is also sizable relative to the variation in the data. The sample mean of Div is 0.016, and its standard deviation is 0.021. Thus, a one-standard-deviation increase in cash dividends represents a meaningful shift in payout policy rather than a trivial fluctuation. Multiplying 0.021 by the estimated coefficient of −8.200 implies a 0.172 decline in LnET. In exact terms, this corresponds to an approximately 15.8% lower environmental protection tax burden. This magnitude suggests that the documented relationship is economically relevant, not merely statistically significant. To make the economic interpretation more intuitive, the results imply that relatively moderate differences in dividend policy are associated with non-trivial differences in firms’ environmental tax burdens. In other words, the estimated effect is large enough to matter in practical terms, rather than reflecting only a small statistical relationship detectable in a large sample.

At the same time, the baseline results should not be interpreted as establishing causality. Because the analysis is based on observational firm-level data, concerns such as omitted variables, selection effects, and reverse causality may still remain. Accordingly, the evidence in Table 2 is more appropriately viewed as documenting a statistically and economically meaningful association rather than a definitive causal effect. Nevertheless, the negative and significant coefficient on Div is consistent with our theoretical expectation that firms distributing higher cash dividends are more likely to exhibit lower subsequent environmental protection tax burdens. In this sense, the baseline evidence provides initial support for H1, while the subsequent robustness and endogeneity analyses further examine whether this relationship survives more demanding empirical tests.

5.3. Mechanism

Environmental responsibility represents an essential dimension of CSR [92,93,94,95,96]. After deciding to distribute cash dividends and meet their obligations to shareholders, corporate executives, guided by moral consistency, allocate more managerial resources to environmental responsibilities. This results in reduced environmental protection tax. To test H2, this study follows Cheng and Liu [97] to construct an environmental responsibility measurement system that captures firms’ substantive environmental responsibility performance. Specifically, this index covers multiple dimensions of environmental responsibility, including Environmental Management System Certification, Environmental Governance Performance Disclosure, Environmental Protection Activity, and Environmental Incident. For ease of exposition, a brief description of the main components is provided in the main text, while the detailed scoring criteria are reported in Appendix D. To examine whether CER constitutes an important channel through which cash dividends are associated with environmental protection tax, this study estimates Models (3) and (4), with the results reported in Table 3. This analysis speaks directly to the second research question of the paper: whether higher cash dividend payments are associated with stronger CER, and whether such improvement in CER is further associated with lower environmental protection tax burdens.

Table 3. Mechanical Test.

	(1)	(2)	(3)
	LnET	CER	LnET
Div	−8.200 ***	2.078 **	−8.067 ***
	(−2.944)	(2.212)	(−3.940)
CER			−0.064 *
			(−1.701)
Size	1.208 ***	0.224 ***	1.223 ***
	(24.539)	(11.926)	(33.693)
Lev	0.051	−0.971 ***	−0.011
	(0.149)	(−8.408)	(−0.049)
ROA	1.649 **	1.636 ***	1.754 ***
	(2.217)	(6.078)	(2.792)
Cashflow	2.403 ***	−0.190	2.391 ***
	(3.421)	(−0.803)	(3.847)
TobinQ	−0.252 ***	0.005	−0.251 ***
	(−5.167)	(0.347)	(−7.720)
Indep	0.008	0.005 *	0.008
	(0.918)	(1.653)	(1.286)
Top1	0.135	0.199	0.148
	(0.385)	(1.624)	(0.622)
_cons	−15.759 ***	−1.099 ***	−15.829 ***
	(−14.330)	(−2.643)	(−19.958)
Year	Yes	Yes	Yes
Industry	Yes	Yes	Yes
Province	Yes	Yes	Yes
Observations	6018	6018	6018
Adjusted R2	0.487	0.176	0.487

Note: The symbols *, **, and *** denote statistical significance at the 10, 5, and 1% levels, respectively. All reported t-statistics are based on standard errors adjusted for clustering at the firm level [91].

Column (2) shows that the coefficient on Div is 2.078 and significant at the 5% level, suggesting that firms with higher cash dividend payments tend to exhibit stronger CER. The estimated magnitude is also economically meaningful. Because Div is measured as cash dividends scaled by total assets, a 1 percentage point increase in Div (i.e., an increase of 0.01) is associated with an approximately 0.0208 increase in CER. Put differently, if two otherwise similar firms differ in dividend payouts by 1 percentage point of total assets, the firm with the higher dividend payout is expected to have a CER level that is higher by about 0.0208. Using the sample standard deviation of Div (0.021), a one-standard-deviation increase in cash dividends is associated with an approximately 0.0436 increase in CER. This indicates that the documented relationship is economically relevant rather than merely statistically significant. Column (3) further includes both Div and CER in the regression for LnET. The coefficient on CER is −0.064 and significant at the 10% level, indicating that stronger CER is associated with lower environmental protection tax burdens. In economic terms, a one-unit increase in CER is associated with a 0.064 decline in LnET. Given the log specification of the dependent variable, this corresponds approximately to a 6.4% reduction in environmental protection tax payments. Thus, higher CER is associated not only with a statistically significant reduction in environmental protection tax, but also with a non-trivial reduction in practical terms. Meanwhile, the coefficient on Div remains negative and significant at the 1% level, and its absolute value declines from 8.200 to 8.067 after CER is included. Although this attenuation is modest, it is consistent with the interpretation that CER may partially account for the negative association between cash dividends and environmental protection tax. The Bootstrap test further reports an indirect effect of −0.786, with a 95% confidence interval of [–1.268, −0.374], which excludes zero. The Sobel test yields similar evidence (Z = −3.508, p < 0.01). Taken together, these results suggest that CER is a statistically meaningful and economically relevant potential transmission channel.

Nevertheless, caution is warranted in interpreting these results. Given the observational nature of the data, the evidence should be understood as supporting an indirect association consistent with the proposed mechanism, rather than as establishing a definitive causal mediation effect. Overall, the mechanism analysis provides supportive evidence for H2 by showing that cash dividends are positively associated with CER, and that stronger CER is in turn associated with lower environmental protection tax burdens.

6. Robustness Test

6.1. Alternative Measures

Following Deng et al. [98] and Sinha & Damle [89], this study replaces the primary explanatory variable (Div) in model (2) with alternative measures, namely, willingness to pay cash dividends (Dumdiv), total cash dividends paid (Dividend), and dividends per share (DPS), to verify robustness. Specifically, Dumdiv is a dummy variable that equals 1 if the listed company distributes cash dividends in the accounting period, and 0 otherwise. Dividend is calculated as the natural logarithm of total cash dividends distributed plus one. DPS is calculated by multiplying earnings per share by the dividend payout ratio. These alternative explanatory variables (Dumdiv, Dividend, and DPS) are substituted into model (2) to perform robustness tests. The coefficients on Dumdiv, Dividend and DPS are negative and significant at the 1% level (−0.289 with t = −2.744; −0.016 with t = −2.712; −0.180 with t = −3.890) in Column (1)–(3) of Table 4.

Table 4. Robustness And Endogeneity Tests.

	(1)	(2)	(3)	(4)	(5)	(6)	(7)	(8)	(9)
	LnET	LnET	LnET	ESG	EPI	ET_OR	LnET	LnET	LnET
Div				2.101 **	4.488 *	−0.001 **	−8.164 ***	−8.428 ***	−6.994 **
				(2.360)	(1.712)	(−2.003)	(−2.866)	(−2.989)	(−2.272)
Dumdiv	−0.289 ***
	(−2.744)
Dividend		−0.016 ***
		(−2.712)
DPS			−0.180 ***
			(−3.890)
Controls	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes
Year	Yes	Yes	Yes	Yes	Yes	Yes	No	Yes	Yes
Industry	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes
Province	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes
Year × Province	No	No	No	No	No	No	Yes	No	No
N	6018	6018	6018	6018	6003	6018	6017	6018	4454
Adjusted R2	0.487	0.487	0.487	0.196	0.256	0.331	0.482	0.126	0.492

Note: The symbols *, **, and *** denote statistical significance at the 10, 5, and 1% levels, respectively. All reported t-statistics are based on standard errors adjusted for clustering at the firm level [91].

It is important to note that a lower environmental protection tax burden does not necessarily imply superior environmental performance. In practice, lower environmental taxes may also reflect reduced production, tax avoidance, differences in regional regulatory intensity, or industry-specific compliance standards. To capture firms’ environmental performance more directly, we replace environmental protection tax in Model (2) with two alternative proxies: ESG performance (ESG) and environmental protection investment (EPI). Following Xie and Lyu [99] and Lai et al. [100], ESG performance is measured using the Huazheng ESG rating. This rating comprises nine ordered categories—C, CC, CCC, B, BB, BBB, A, AA, and AAA—which are coded from 1 to 9 in ascending order; the firm-year ESG score is constructed as the average of the four quarterly ratings. Following Cui et al. [101] and Huang and Lei [102], environmental protection investment is measured as the natural logarithm of annual environmental investment expenditure plus one. The re-estimation results are reported in Table 4. When ESG is used as the dependent variable, the coefficient on Div is 2.101 and significant at the 5% level, suggesting that higher cash dividend payments are associated with better ESG performance. Economically, a 0.01 increase in Div is associated with an approximately 0.021-point increase in the ESG score. A one-standard-deviation increase in Div is associated with an approximately 0.044-point increase in ESG. When EPI is used as the dependent variable, the coefficient on Div is 4.488 and significant at the 10% level, indicating that higher cash dividend payments are also associated with greater environmental investment. In economic terms, a 0.01 increase in Div is associated with an approximately 0.0449 increase in the log of environmental investment, which corresponds to about a 4.6% increase in environmental investment in level terms. At the same time, it is important to consider an alternative explanation for the baseline results. Environmental protection tax payments may be influenced by firms’ production scale, such that a lower tax burden may simply reflect reduced output rather than improved environmental performance. To address this concern and better align the analysis with the paper’s core research question, we replace LnET with the ratio of environmental protection tax to operating revenue (ET_OR) and re-estimate the model. This measure captures the environmental tax intensity per unit of output and therefore helps reduce the confounding effect of fluctuations in production scale. The results are reported in Table 4, Column (6). The coefficient on Div is −0.001 and significant at the 5% level, indicating that firms with higher cash dividend payouts tend to exhibit lower environmental tax intensity relative to operating revenue. Economically, this suggests that the negative association documented in the baseline analysis is not solely attributable to lower production levels, but is also reflected in a lower environmental tax burden per unit of business activity. In this sense, the result provides additional support for the paper’s main argument that higher cash dividends are associated with better environmental outcomes rather than merely smaller production scale.

Taken together, these results suggest that the positive association between cash dividends and firms’ sustainability-related performance remains robust when alternative dependent variables are employed. Nevertheless, given the observational nature of the data, the findings should be interpreted as economically relevant associations rather than definitive causal effects.

6.2. Cross-Fixed Effect

Tax liabilities are influenced not only by financial and non-financial factors but also by regional variations and temporal differences within industries or regions. To mitigate potential omitted variable bias, this study controls for industry and regional fixed effects and further incorporates region-year interaction fixed effects. The regression coefficient on Div in Column (7) of Table 4 is −8.164 and significant at 1% level, supporting the original conclusions.

6.3. Alleviate Endogeneity Tests

To further mitigate potential endogeneity concerns, this study employs three additional methods for robustness testing.

First, to examine the robustness of the baseline results, we re-estimate the model using a Tobit specification. As shown in the descriptive statistics, the environmental protection tax variable (LnET) has a minimum value of 0, indicating a left-censored distribution. In this case, OLS estimation may be less appropriate. Therefore, the Tobit model is adopted as an alternative estimation method. The results in Column (8) of Table 4 show that the coefficient on Div remains significantly negative (−8.428, p < 0.01), suggesting that the negative association between cash dividend payments and environmental protection tax is robust.

Second, the propensity score matching (PSM) method was employed using a nearest-neighbor 1:1 matching approach. All control variables from the main regression model were used in a logistic regression to estimate the propensity scores. Subsequently, propensity scores obtained from logistic regression were used to match the treatment and control groups in a 1:1 ratio. The standardized mean differences (%) of all matched variables were below 10%, indicating an adequate balance between treatment and control groups. Finally, the primary regression model was re-estimated using the matched sample, and the regression results are reported in Table 4. Column (9) shows that the coefficient of Div is −6.994 and statistically significant at the 5% level.

Third, an instrumental variable (IV) approach was adopted to further address potential endogeneity issues. Specifically, this study utilizes the market institutional environment (Institution) as an instrumental variable. This variable is calculated as the marketization index for Chinese regions compiled by Fan Gang and Wang Xiaolu divided by 1000. This index system, constructed based on international marketization measurement frameworks and adapted to China’s institutional reforms, comprises five dimensions: the relationship between government and market, non-state economic development, development of product and factor markets, market intermediary organizations, and the legal environment. Compiled using principal component analysis, this index provides an authoritative measure of China’s economic transition for empirical analysis. Following practices from Nee & Matthews [103] and Arin et al. [104], this study employs the marketization index corresponding to the region and year of each sample firm as a proxy for the market institutional environment.

The market institutional environment has a significant impact on corporate cash dividends. Improvements in the market institutional environment strengthen capital market regulation and standardize dividend distribution. A mature institutional environment, characterized by sound investor protection, helps enhance the transparency and stability of corporate dividend policies. In addition, as the institutional environment matures, investors increasingly favor stocks that provide stable cash flows, which encourages firms to adopt prudent dividend policies to attract investors. Beyond satisfying the relevance condition, this study further argues for the validity of the exclusion restriction. First, by its institutional nature, the marketization process represents a macro-level external environment. It mainly functions by regulating capital markets and strengthening corporate governance norms, rather than directly intervening in firms’ micro-level decisions regarding the operation of pollution-control equipment or the adoption of emission-reduction technologies. Therefore, it lacks a direct physical channel through which it could affect environmental protection tax burdens. Second, the influence of the market institutional environment is mainly transmitted indirectly through firms’ internal financial decisions, such as the moral consistency motive associated with cash dividends, and subsequently affects environmental performance. Finally, the province fixed effects included in the model effectively absorb the direct interference arising from regional differences in environmental regulation and tax enforcement intensity. Taken together, this instrumental variable can only affect environmental protection tax burdens indirectly through the single channel of dividend policy, thereby satisfying the exclusion restriction assumption. Therefore, this study uses the market institutional environment as an instrumental variable in the endogeneity test of model (2). The results are presented in Table 5. Column (1) displays the first-stage estimation results. The significant coefficient (0.542 with t = 4.31) on Institution confirms a strong correlation between the selected instrumental variable and cash dividends. Column (2) presents the second-stage estimation results. The coefficient on Div_P is −104.8 significant at 10% level, which is consistent with the benchmark regression findings. The Cragg-Donald Wald F-statistic (17.238) and the Kleibergen-Paap Wald rk F-statistic (18.548) both exceed the Stock-Yogo critical value at the 10% level, thus passing the weak instrument test. Overall, these results provide supplementary support for a negative association between corporate cash dividends and environmental protection tax, although they do not by themselves establish definitive causality, consistent with the previous conclusions. However, it should be acknowledged that, despite the use of alternative model specifications, propensity score matching (PSM), and instrumental variable (IV) methods, endogeneity cannot be entirely ruled out in observational studies.

Table 5. Instrumental variable approach.

	(1)	(2)
	Div	LnET
Div_P		−104.8 *
		(−2.39)
Institution	0.542 ***
	(4.31)
Controls	Yes	Yes
Year	Yes	Yes
Industry	Yes	Yes
Cragg-Donald Wald F statistic	17.238
Kleibergen-Paap Wald rk F statistic	18.548
N	6013	6013

Note: The symbols *,and *** denote statistical significance at the 10, 5, and 1% levels, respectively. All reported t-statistics are based on standard errors adjusted for clustering at the firm level [91].

7. Further Analysis

7.1. Moderating Effect

7.1.1. The Moderating Effect of Government Environmental Subsidies

Government environmental subsidies refer to a suite of fiscal policies implemented by the Ministry of Finance and local governments at various administrative levels, aimed at facilitating firms’ upgrading of environmental protection equipment and green production processes. As a form of governmental intervention [105], such subsidies constitute a positive incentive mechanism that can encourage firms to assume greater CER and orient their strategies toward sustainable transformation [106,107,108]. By providing additional financial resources, government environmental subsidies can alleviate the fiscal burden associated with environmental governance initiatives [25].

From the stakeholder theory perspective, firms that distribute cash dividends to satisfy shareholder expectations must simultaneously respond to the environmental expectations of other key stakeholders—including regulators and institutional investors [77]. In this context, government environmental subsidies may serve a dual function: first, it enhances firms’ awareness of and alignment with governmental environmental governance objectives [109]; second, it mitigates the resource trade-off between dividend payouts and environmental investment, thereby enabling firms to fulfill CER without compromising shareholder returns. Consequently, higher levels of government environmental subsidies may strengthen the negative association between cash dividends and environmental protection tax—i.e., the extent to which dividend payments are linked to lower tax liabilities may be amplified under more generous subsidy regimes.

From the legitimacy theory lens, environmental responsibility has become a critical source of social acceptance and institutional legitimacy in an era where green development is increasingly codified as a normative institutional expectation [81,82]. Firms that simultaneously distribute dividends while neglecting environmental governance—and consequently incur high environmental protection taxes—may be perceived as prioritizing shareholder interests over broader public responsibilities, thereby undermining their legitimacy. Government environmental subsidies, by signaling explicit policy preferences and institutional expectations, intensifies the legitimacy-related pressures on recipient firms. As a result, such firms may be more inclined to strengthen environmental governance and reduce emissions—not only to access subsidies, but also to reaffirm their social legitimacy. In this sense, government environmental subsidies not only provide resource support but also reinforces external institutional constraints that encourage alignment between dividend policy and CER. This dual role may potentiate the observed association whereby cash dividends are linked to lower environmental protection tax burdens.

To support the aforementioned inference, this study references the work of Han et al. [105] and quantifies government environmental subsidies (GES) as the total value of environmental subsidies provided by the government to publicly listed companies, expressed as a percentage of the companies’ total assets. Column (2) of Table 6 shows that the coefficient on Div × GES is −32.412 and significant at the 10% level (t = −1.90). This result indicates that the negative association between cash dividends and environmental protection tax payments becomes stronger when firms receive higher government environmental subsidies. The economic magnitude of this moderating effect is also meaningful. A one-standard-deviation increase in GES strengthens the negative marginal association between Div and LnET by an additional 4.149 log points. Put differently, government environmental subsidies amplify the extent to which higher cash dividends are associated with lower environmental protection tax burdens. When both Div and GES increase by one standard deviation, LnET is associated with an additional decline of approximately 0.087 log points. This corresponds to about 2.9% of the sample standard deviation of LnET, or roughly an 8.3% reduction in environmental protection tax payments in level terms. Therefore, the moderating effect is not only statistically detectable but also economically relevant in practical terms. Nevertheless, given the observational nature of the data, this finding should be interpreted as evidence of association rather than a definitive causal effect.

Table 6. Moderating Effect Analysis.

	(1)	(2)	(3)
	LnET	LnET	LnET
Div	−8.200 ***	−8.353 ***	−7.735 ***
	(−2.94)	(−3.00)	(−2.67)
GES		1.740 ***
		(3.24)
Div × GES		−32.412 *
		(−1.90)
Media			0.046
			(0.88)
Div × Media			−2.871 **
			(−2.22)
Controls	Yes	Yes	Yes
Year	Yes	Yes	Yes
Industry	Yes	Yes	Yes
Province	Yes	Yes	Yes
N	6018	6018	5666
Adjusted R2	0.497	0.498	0.505

Note: The symbols *, **, and *** denote statistical significance at the 10, 5, and 1% levels, respectively. All reported t-statistics are based on standard errors adjusted for clustering at the firm level [91].

7.1.2. The Moderating Effect of Media Attention

Media serve not merely as channels of information dissemination, but also play a pivotal role in shaping public perceptions and amplifying the external pressures faced by firms, thereby potentially prompting executives to take more decisive actions. Firms that attract media attention are often subject to closer scrutiny, particularly in the domain of environmental protection [50,110,111,112]. From the perspective of stakeholder theory, media attention may intensify the monitoring pressure exerted by external stakeholders—including the public, regulators, and investors—on firms’ environmental conduct. Under conditions of high media attention, dividend-paying firms that neglect CER may become more vulnerable to skepticism and pressure from multiple stakeholder groups. In response to these stakeholder expectations, such firms may have stronger incentives to advance environmental governance alongside shareholder payout policies. From the perspective of signaling theory, media attention enhances the transparency and verifiability of corporate behavior [113,114]. As a result, the credibility of the signal conveyed by cash dividend payments may depend more heavily on the consistency between dividend policy and environmental performance. If dividend distributions are accompanied by a relatively high environmental protection tax burden, the market may question the authenticity of the governance signal embedded in such payouts. Under these circumstances, firms may be more inclined to improve their environmental performance so as to reinforce the credibility of dividend payments, which may in turn be associated with lower environmental protection tax burdens.

To support the aforementioned argument, this study references the findings of Zheng et al. [50] and operationalizes media attention (Media) by calculating the total media coverage of publicly listed companies divided by 1000. In addition, the interaction term Div × Media is introduced into Model (2) to examine the moderating effect of media attention. Column (3) of Table 6 shows that the coefficient on Div × Media is −2.871 and significant at the 5% level (t = −2.22). This result indicates that the negative association between cash dividends and environmental protection tax payments becomes stronger when firms receive greater media attention. The economic magnitude of this moderating effect is also meaningful. A one-standard-deviation increase in Media strengthens the negative marginal association between Div and LnET by an additional 1.777 log points. Put differently, stronger media attention amplifies the extent to which higher cash dividends are associated with lower environmental protection tax burdens. When both Div and Media increase by one standard deviation, LnET is associated with an additional decline of approximately 0.037 log points. Given the log specification of the dependent variable, this corresponds to roughly a 3.7% reduction in environmental protection tax payments in level terms. Therefore, the moderating effect of media attention is not only statistically significant but also economically relevant in practical terms. Nevertheless, given the observational nature of the data, this finding should be interpreted as evidence of association rather than as a definitive causal effect.

In sum, both government environmental subsidies and media attention appear to strengthen the negative association between cash dividend payments and environmental protection tax—though they operate through distinct pathways. Government environmental subsidies primarily functions by enhancing firms’ capacity to fulfill CER, via institutional endorsement and financial resource supplementation. In contrast, media attention operates predominantly through external monitoring and informational amplification, thereby tightening the constraints on firms to maintain consistency between dividend signaling and environmental performance. These findings suggest that the extent to which firms can effectively reconcile shareholder returns with environmental responsibilities is not solely determined by internal governance structures, but is also significantly shaped by the external institutional context—including policy incentives and stakeholder-induced pressures.

7.2. Heterogeneity Analyses

7.2.1. Heterogeneity of Executives’ Green Perception

Executives’ green perception refers to senior managers’ understanding of resources and environmental challenges and mainly encompasses perceptions of green competitive advantage, CSR consciousness, and external environmental pressure. CSR consciousness is particularly reflected in executives’ proactive actions and contributions to environmental protection and resource conservation [66,67,115]. Wu et al. [67] conducted a study examining firms in China that are characterized by high levels of pollution. Their findings suggest that senior management plays a pivotal role in enhancing corporate environmental performance by integrating environmental initiatives with broader business strategies. Executives with substantial environmental knowledge are proficient in monitoring industry developments and environmental regulations [116]. They consistently collect and analyze data related to environmental protection, thereby deepening their understanding of ecological issues [117]. Moreover, these leaders engage in proactive measures to conserve energy and reduce emissions, while also incorporating environmental considerations into the design of new products and production processes [67,118]. Such practices enable companies to more effectively meet their environmental obligations. Therefore, in firms with high levels of executives’ green cognition, managers typically do not rely on dividend distribution to incentivize environmental initiatives. Whether or not cash dividends are issued, these executives proactively invest in pollution control and green innovation driven by intrinsic motivation [119]. According to the behavioral explanation of corporate decision-making developed in Section 2.3, the moral-consistency effect potentially triggered by dividend payouts may be offset in such firms, thereby weakening the marginal incentive effect of dividend distribution. By contrast, in firms where executives exhibit relatively low levels of green perception, managers may lack proactive environmentally oriented awareness [61,120]. This suggests that environmental responsibility has not yet been incorporated into executives’ default hierarchy of values, and that the firm has not formed a stable green moral identity. Under such circumstances, the payment of cash dividends may serve as an explicit signal that the firm is fulfilling its responsibilities to shareholders, thereby activating a moral-consistency mechanism at the organizational level. To avoid the perceived inconsistency of being accountable to shareholders while neglecting environmental responsibilities, managers may become more inclined to increase subsequent environmental governance investment to preserve the coherence of the firm’s overall responsibility image. In this context, the moral-consistency mechanism associated with dividend payouts may become an important driver of CER fulfillment, making the negative association between cash dividends and environmental protection tax more pronounced.

To validate the above reasoning, this study employs textual analysis methodology following Duriau et al. [121] and Osborne et al. [122]. We measure executive green perception (EGP) through three dimensions: green competitive advantage, CSR, and perceived external environmental pressure. Relevant keywords were selected, and their frequency in annual reports of listed companies (2018–2023) was quantified. The full sample was dichotomized at the median EGP into high-EGP and low-EGP subgroups for separate regression analyses. Table 7 shows that the coefficient on Div is −3.147 and insignificant for the high-EGP subgroup, whereas it is −11.495 and significant at the 1% level (t = −2.791) for the low-EGP subgroup. This pattern suggests that the negative association between cash dividends and environmental protection tax payments is concentrated among firms with lower executive green perception. In other words, cash dividend payouts appear to be more strongly associated with lower environmental protection tax burdens when executives have not yet formed a relatively strong green value orientation. The economic magnitude of this difference is also meaningful. In the low-EGP subgroup, a 1 percentage point increase in Div is associated with a 0.115 decrease in LnET. Given the log specification of the dependent variable, this corresponds to approximately a 10.9% reduction in environmental protection tax payments. By contrast, the coefficient in the high-EGP subgroup is not statistically distinguishable from zero, suggesting that once executives already possess stronger green perception, the additional association between dividend payouts and lower environmental protection tax becomes much weaker. Therefore, the heterogeneity result is not only statistically informative but also economically relevant. Nevertheless, this evidence should be interpreted as an association rather than as a definitive causal effect, given the observational nature of the data. Overall, these results support the behavioral logic proposed in Section 2.3: executive green perception moderates the extent to which cash dividends can activate moral consistency at the organizational level, such that the tax-reducing effect of dividend payouts is more pronounced in firms where environmental responsibility has not yet been internalized into executives’ default value ordering.

Table 7. Heterogeneity empirical results.

	(1)	(2)	(3)	(4)	(5)	(6)
	LnET	LnET	LnET	LnET	LnET	LnET
	High Executives’ Green Perception	Low Executives’ Green Perception	Strong Environmental Regulation	Weak Environmental Regulation	Pollute = 1	Pollute = 0
Div	−3.147	−11.495 ***	−4.449	−12.585 ***	−2.332	−10.430 ***
	(−0.901)	(−2.791)	(−1.411)	(−3.095)	(−0.664)	(−2.580)
Controls	Yes	Yes	Yes	Yes	Yes	Yes
Year	Yes	Yes	Yes	Yes	Yes	Yes
Industry	Yes	Yes	Yes	Yes	Yes	Yes
Province	Yes	Yes	Yes	Yes	Yes	Yes
N	3199	2815	3161	2842	2121	3545
Adjusted R2	0.559	0.351	0.475	0.497	0.602	0.301

Note: The symbols *** denote statistical significance at the 10, 5, and 1% levels, respectively. All reported t-statistics are based on standard errors adjusted for clustering at the firm level [91].

7.2.2. Heterogeneity of Environmental Regulation

Environmental regulation represents governmental measures to standardize corporate environmental conduct through policies and legislation. Its objectives are to reduce pollution emissions, promote rational resource utilization, and advance the green economy [123,124]. Stringent environmental regulation has a significant influence on corporate environmental behavior. As an external constraint, environmental regulation plays a critical role in shaping firms’ ecological practices. When enterprises operate in regions with rigorous environmental regulation, they typically face stricter monitoring systems and external constraints, including more frequent environmental inspections, elevated non-compliance costs, and well-developed incentive mechanisms for environmental protection [125]. These institutionalized pressures foster stable environmental governance patterns within firms, rendering environmental practices increasingly routinized. Consequently, the marginal effect of cash dividends as an external incentive tool on environmental governance becomes comparatively attenuated. Conversely, in regions with lax environmental regulation, corporations experience reduced external pressure for environmental compliance. Their ecological conduct relies more heavily on internal initiative and external reputational incentives [126,127]. From the perspective of signaling theory, in regions with weaker environmental regulation, the transparency of information concerning firms’ environmental performance tends to be relatively limited [128]. Under such conditions, external stakeholders may find it more difficult to directly assess whether firms are actively fulfilling CER. As a result, cash dividends, as an observable financial decision, may be more readily interpreted as a positive signal of operational soundness and responsibility orientation. To preserve the credibility of this signal, firms may be more inclined to further strengthen environmental governance and reduce pollutant emissions, which may in turn be associated with a lower environmental protection tax burden. Meanwhile, according to legitimacy theory, weaker environmental regulation implies relatively less stringent formal institutional constraints and weaker coercive legitimacy pressures on firms [129]. In such contexts, the fulfillment of environmental responsibility may depend more heavily on firms’ voluntary responsiveness to broader societal expectations. Under these circumstances, cash dividend payments may become an important opportunity for firms to cultivate an image of responsibility and, correspondingly, may encourage greater investment in environmental governance. By contrast, in regions with stronger environmental regulation, environmental compliance already constitutes a basic requirement for maintaining organizational legitimacy, and the additional role of cash dividends may therefore be comparatively limited. Accordingly, the negative association between cash dividends and environmental protection tax may be more pronounced in regions characterized by weaker environmental regulation.

To validate the above reasoning, this study employs the methodology of Chen et al. [130] to measure environmental regulation intensity (ER). Specifically, ER is constructed as the frequency density of environment-related terms in local government work reports, calculated as the ratio of the number of relevant word occurrences to the total word count. The full sample is then divided at the median value of ER into strong-regulation and weak-regulation subsamples for separate regression analyses. Table 7 shows that the coefficient on Div is −4.449 and statistically insignificant for firms located in strong-regulation regions, whereas it is −12.585 and significant at the 1% level (t = −3.095) for firms in weak-regulation regions. This pattern suggests that the negative association between cash dividends and environmental protection tax payments is concentrated in regions with weaker environmental regulation. In other words, dividend payouts appear to be more strongly associated with lower environmental protection tax burdens when firms face less stringent external environmental constraints. The economic magnitude of this heterogeneity is also meaningful. In the weak-regulation subgroup, a 1 percentage point increase in Div is associated with a 0.126 decrease in LnET. Given the log specification of the dependent variable, this corresponds to approximately an 11.8% reduction in environmental protection tax payments. By contrast, the coefficient in the strong-regulation subgroup is not statistically distinguishable from zero, suggesting that when environmental regulation is already stringent, the additional association between dividend payouts and lower environmental protection tax becomes much weaker. Therefore, the heterogeneity result is not only statistically informative but also economically relevant. Nevertheless, given the observational nature of the data, this evidence should be interpreted as an association rather than as a definitive causal effect. A plausible explanation is that in regions with weaker environmental regulation, cash dividends may more effectively serve signaling and legitimacy-building functions. As a result, firms may be more likely to strengthen environmental governance after dividend payouts, which is reflected in a stronger negative association with environmental protection tax.

7.2.3. Heterogeneity of Pollution Industry

This study classifies industries into heavily polluting and non-heavily polluting categories based on the 2012 secondary industry classification issued by the China Securities Regulatory Commission (Heavily polluting industries are defined as: B06 coal mining and washing, B07 petroleum and natural gas extraction, B08 ferrous metal mining and selection, B09 non-ferrous metal mining and selection, C17 textile industry, C19 leather, fur, feather and its products and footwear industry, C22 paper and paper products industry, C25 petroleum processing, coking and nuclear fuel processing industry, C26 chemical raw materials and chemical products manufacturing industry, C28 chemical fiber manufacturing industry, C29 rubber and plastic products industry, C30 non-metallic mineral products industry, C31 ferrous metal smelting and rolling processing industry, C32 non-ferrous metal smelting and rolling processing industry, D44 power, heat production and supply industry, a total of 15 industries. Non-heavily polluting industries are: B mining industry, C manufacturing industry, D electricity, heat, gas and water production and supply industry—the remaining industries in the same category). Enterprises in non-heavily polluting industries generally face fewer environmental challenges. From the perspective of stakeholder theory, following cash dividend distributions, firms are expected not only to meet shareholders’ demands for financial returns, but also to address the environmental responsibility expectations of other key stakeholders, including governmental authorities, the general public, and local communities [80]. Relative to heavily polluting industries, firms in non–heavily polluting sectors typically face more tractable environmental challenges and incur lower adjustment costs in adopting environmentally responsible practices [131]. Consequently, such firms are better positioned to concurrently enhance their environmental performance post-dividend distribution—thereby sustaining constructive relationships with a diversified stakeholder base and, ceteris paribus, reducing their exposure to environmental protection tax. Meanwhile, from the perspective of legitimacy theory, firms in heavily polluting industries, given their pronounced environmental externalities, already operate under a baseline expectation of environmental compliance as a prerequisite for maintaining social legitimacy [129]. In this context, cash dividend payments are unlikely to substantially alter the structural constraints that compel these firms to bear elevated environmental responsibilities; hence, the marginal effect of dividends on environmental governance may be comparatively muted. By contrast, firms in non–heavily polluting sectors face relatively weaker external legitimacy pressures regarding environmental conduct. For them, dividend payouts may serve as a salient, visible demonstration of normative governance and responsible stewardship. This symbolic act may, in turn, catalyze more deliberate environmental management interventions—such as emission reductions and process upgrades—to reinforce institutional credibility. As a result, the negative association between cash dividends and environmental protection tax may be more pronounced in non–heavily polluting industries.

To empirically examine this proposition, the full sample was divided into two subgroups: heavily polluting industries (Pollute = 1) and non-heavily polluting industries (Pollute = 0) [132,133]. Columns (5) and (6) of Table 7 show that the coefficient on Div is −2.332 and statistically insignificant for firms in heavily polluting industries, whereas it is −10.430 and significant at the 1% level (t = −2.580) for firms in non-heavily polluting industries. This pattern suggests that the negative association between cash dividends and environmental protection tax payments is concentrated among firms in non-heavily polluting industries. The economic magnitude of this heterogeneity is also meaningful. In the non-heavily polluting subgroup, a 1 percentage point increase in Div is associated with a 0.104 decrease in LnET. Given the log specification of the dependent variable, this corresponds to approximately a 9.9% reduction in environmental protection tax payments in level terms. By contrast, the coefficient in the heavily polluting subgroup is not statistically distinguishable from zero, suggesting that the association between dividend payouts and environmental protection tax burdens is much weaker for firms operating in heavily polluting industries. One possible explanation is that firms in heavily polluting industries are already subject to stricter environmental scrutiny and compliance requirements, which may reduce the additional signaling or legitimacy-building role of cash dividends. Nevertheless, given the observational nature of the data, this result should be interpreted as evidence of association rather than as a definitive causal effect. Overall, the findings suggest that, relative to firms in heavily polluting industries, firms in non-heavily polluting industries exhibit a stronger negative association between cash dividend payouts and environmental protection tax payments.

8. Conclusions

Against the backdrop of global sustainable development and China’s transition toward a formal environmental taxation system, this study examines the association between corporate financial payout policies and explicit environmental liabilities. Using a comprehensive sample of Chinese A-share listed companies from 2018 to 2023, we find a significant and robust negative association between cash dividend distributions and corporate environmental protection tax burdens. This finding suggests that fulfilling financial responsibilities to shareholders does not necessarily crowd out environmental commitment. Instead, it is associated with stronger environmental compliance. Mechanism analysis further indicates that this cross-domain consistency operates through the fulfillment of corporate environmental responsibilities and is strengthened by external monitoring, such as media attention, as well as by resource support, such as government environmental subsidies.

This study contributes to several important theoretical discussions. First, it provides new evidence for the “governance complementarity” view in corporate finance and challenges the traditional “resource crowding-out” argument. While conventional agency-based reasoning often suggests that financial payouts reduce the resources available for non-financial objectives, our findings show that shareholder-oriented financial decisions and environmental responsibility may move in the same direction rather than compete with each other. Second, this study advances the micro-foundations of corporate environmental responsibility by linking moral consistency theory with upper echelons theory. In doing so, it explains how executives’ internal value orientation and desire for ethical self-consistency may be translated into firm-level environmental behavior. This perspective moves beyond treating the firm as a passive entity that merely responds to external pressure and instead highlights the cognitive and ethical foundations of corporate sustainability.

The findings also carry practical implications for corporate boards and executive teams. In many firms, dividend policy and environmental strategy are still managed separately. Our results suggest that this separation may overlook important governance spillovers. Boards may therefore benefit from adopting a more integrated governance framework in which financial decisions are evaluated not only in terms of investor response, but also in terms of their broader implications for corporate responsibility and long-term legitimacy. For executives, the evidence suggests that cash dividend payouts need not be viewed solely as short-term financial distributions. Under certain conditions, they may also be associated with a more responsible organizational posture, which can reinforce environmental responsibility and improve compliance outcomes.

More importantly, the results offer explicit implications for policymakers, especially in emerging economies where green transition goals must often be pursued alongside capital-market development. First, the findings suggest that environmental regulation should not be designed under the assumption that shareholder-oriented financial behavior and environmental governance are inherently in conflict. In emerging markets, where firms often face both financing constraints and rising sustainability pressure, policy design may be more effective when it recognizes the potential complementarity between sound corporate governance and environmental compliance. Second, the moderating results indicate that external institutions matter. Media attention and government environmental subsidies both strengthen the negative association between cash dividends and environmental protection tax burdens, implying that policy effectiveness can be enhanced when formal regulation is supported by information transparency and targeted resource incentives. Third, the weaker effect observed in regions with stronger environmental regulation suggests that environmental policy should not rely exclusively on command-and-control enforcement. Instead, regulators may consider a more coordinated policy framework that combines regulatory discipline with governance-based incentives, public scrutiny, and subsidy support. Such a framework may be particularly valuable for other emerging economies seeking to balance investor confidence, industrial upgrading, and environmental protection during institutional transition.

At the same time, these policy implications should be interpreted with appropriate caution. Because our evidence is based on observational data, it supports a robust pattern of association rather than a definitive causal claim. Accordingly, the results should not be read as implying that increasing dividend payouts will mechanically reduce environmental tax burdens. Rather, they suggest that in a broader governance setting, firms with stronger shareholder-oriented financial commitment may also be more likely to engage in environmental responsibility and compliance.

Despite these contributions, this study has several limitations that point to promising directions for future research. First, although we employ multiple econometric approaches to mitigate endogeneity concerns, the empirical results should still be interpreted as associations rather than definitive causal effects. Second, while environmental protection tax payments provide a relatively objective and statutory measure of environmental burden, they capture only one dimension of overall environmental performance. Future studies may further examine whether similar patterns hold for more granular outcomes, such as specific pollutant emissions, carbon intensity, or third-party environmental ratings. Third, due to data constraints and the difficulty of directly measuring managerial psychological processes in archival settings, the moral-consistency mechanism proposed in this paper is inferred theoretically rather than directly observed. Future research could refine the measurement of executive value orientation and test the proposed behavioral mechanism more directly through survey-based, experimental, or mixed-method approaches.

Overall, this study shows that corporate dividend policy and environmental responsibility need not be viewed as separate or competing domains. In the context of China’s environmental tax reform, the evidence suggests that shareholder-oriented financial behavior may be associated with stronger environmental responsibility and lower environmental tax burdens. By highlighting this relationship, the study contributes to a more integrated understanding of how corporate finance, executive values, and environmental governance interact in emerging economies.