Heliyon 9 (2023) e21492

Contents lists available at ScienceDirect

Heliyon
journal homepage: www.cell.com/heliyon

Corporate environmental information disclosure and tax

avoidance: Evidence from China
Yu Gu, Shanhui Wang *
School of Business, NingboTech University, Ningbo, 315100, China

A R T I C L E I N F O A B S T R A C T

Keywords: This paper examines the impact of environmental information disclosure (EID) on corporate tax
Environmental information disclosure avoidance. By exploiting China’s EID policy as a natural experiment, we employ a difference-in-
Tax avoidance differences (DID) estimation approach to reach the following findings: (1) EID can promote
China
corporate tax avoidance, and this result remains robust after conducting a battery of tests; (2) the
research identifies two distinct channels through which EID influences corporate tax avoidance:
green innovation and capital investments; and (3) the heterogeneity analysis reveals that the
impact of EID on tax avoidance is notably heightened in highly polluting industries, competitive
sectors, and regions in China characterized by weak tax efforts and strong government compe­
tition. Through the application of a scientific approach, we reveal novel insights into the corre­
lation between EID and tax avoidance, thereby providing significant implications for the
development of environmental regulatory policies. In conclusion, we recommend that govern­
ments employ a targeted approach to tackle tax avoidance in specific enterprises or industries.
Simultaneously, governments should consider implementing subsidies and tax incentives to
mitigate the underlying incentives for enterprises to engage in tax avoidance.

1. Introduction

During the past few decades, an increasing emphasis on environmental sustainability has emerged, leading governments, stake­
holders, and society at large to place greater emphasis on demanding increased transparency and accountability from businesses
regarding their environmental impact. As a result, the practice of EID has received notable scholarly attention [1,2]. A substantial body
of research has examined the association between EID and firm behavior, encompassing various perspectives, such as green inno­
vation, debt financing, stock price crash risk, and financial performance [3–6]. However, a noticeable gap persists in our under­
standing of the interplay between EID and corporate tax avoidance behavior. This knowledge gap impedes our thorough
comprehension of the economic ramifications of EID and the underlying motivations driving corporate tax behavior.
The impact of EID on corporate tax avoidance is influenced by the interplay of positive and negative effects. On the one hand, EID
has the potential to increase the likelihood of corporate tax avoidance. To comply with policy requirements, firms incur costs asso­
ciated with information disclosure, leading to reduced productivity and profitability [7]. Additionally, EID may result in the disclosure
of negative information, triggering unfavorable market reactions [8]. In such scenarios, firms may seek to mitigate the additional costs
imposed by EID policies by reducing their financial expenditures through tax avoidance. On the other hand, EID has the potential to
diminish firms’ incentives for tax avoidance. The increased disclosure assists firms in mitigating the information asymmetry problem,

* Corresponding author.
E-mail address: wangshanhui@nbt.edu.cn (S. Wang).

https://doi.org/10.1016/j.heliyon.2023.e21492
Received 6 February 2023; Received in revised form 5 September 2023; Accepted 23 October 2023
Available online 30 October 2023
2405-8440/© 2023 The Authors. Published by Elsevier Ltd. This is an open access article under the CC BY-NC-ND license
(http://creativecommons.org/licenses/by-nc-nd/4.0/).
Y. Gu and S. Wang Heliyon 9 (2023) e21492

enabling them to secure improved financing conditions and lower financing costs [9]. Furthermore, EID policies can enhance the
reputation and brand value of companies, resulting in higher costs associated with reputational damage when tax avoidance is
exposed. Therefore, firms are more likely to adhere to tax regulations. In conclusion, the relationship between EID and corporate tax
avoidance is an empirical issue.
When making decisions, firms must weigh the costs and benefits of EID while also considering the associated risks and opportunity
costs of engaging in tax avoidance behavior. In this study, we address the existing research gap by investigating the potential inter­
active relationship between the disclosure of environmental information and corporate tax compliance. Our specific objective is to
examine the influence of EID on corporate tax avoidance and its underlying mechanisms. Additionally, we explore the moderating
effects of industry characteristics and the regulatory environment. By analyzing these factors, we seek to offer valuable guidance and
recommendations for policymakers and businesses alike.
Through this article, we make three primary contributions. (1) We contribute to the advancement of the literature on the factors
influencing corporate tax avoidance by furthering our understanding of the effects of EID on corporate financial behavior. Doing so
improves our comprehension of how firms strategically respond to environmental pressures and regulatory environments. (2) We
further enrich the existing body of research on corporate social responsibility (CSR). By investigating the link between EID and tax
avoidance, evaluating whether firms align their environmental responsibilities with their tax behaviors becomes feasible, thereby
providing a more comprehensive assessment of their social responsibility performance. Our findings indicate that firms tend to exhibit
an increase in tax avoidance behavior in response to heightened EID. We deepen the comprehension of the interplay between CSR and
financial considerations. (3) In our paper, we yield crucial policy insights. From a practical standpoint, the research findings of this
study offer valuable insights to policymakers and regulators. These insights can assist them in formulating effective environmental
regulations and implementing tax enforcement measures that effectively address the challenges associated with tax avoidance and
environmental sustainability.
The remainder of this paper is organized as follows. Section 2 introduces the literature review, hypotheses and policy background.
Section 3 presents the research design. Section 4 presents the empirical results, including the benchmark regression results, robustness
tests, underlying mechanisms and heterogeneity analysis. Section 5 presents the empirical results and policy recommendations, as well
as the limitations and future directions of this study.

2. Literature review, hypothesis development and policy background

2.1. Literature review

2.1.1. Literature related to EID

Numerous countries worldwide have recognized the significance of EID as a crucial tool of government environmental regulatory
practices. Extensive research has been dedicated to examining the impact of EID and its economic implications. The existing body of
literature has emphasized the significant and diverse effects of EID on corporate behavior and has highlighted the potential for EID to
stimulate green innovation [10], mitigate firms’ risk of future stock price collapses [5], lower the cost of debt financing [9] and
improve long-term corporate economic performance [11].
However, worth noting is that the effects of EID on firms are not universally positive. The disclosure of environmental information
requires companies to allocate resources for activities such as data collection, monitoring, reporting, and compliance, which can result
in additional costs for firms [12]. Meng and Zhang [8] found that firms with poor environmental performance may experience negative
investor reactions when compelled to disclose environmental information. Furthermore, Fonseka et al. [13] found that when gas,
thermal power generation, and hydro firms increase their EID levels, their cost of debt increases.
We aim to contribute to the existing research by examining the relationship between EID and tax avoidance. Specifically, we seek to
investigate whether firms with higher EID are more likely to adopt responsible tax practices or, conversely, are more prone to engaging
in tax avoidance strategies. Through our study, we pioneer the examination of the association between EID and tax avoidance by
employing the DID methodology, which helps address the challenge of accurately measuring the extent of corporate EID. By shedding
light on the interplay between firms’ EID and their tax avoidance behavior, we improve the understanding of the complex dynamics
underlying corporate decision-making processes.

2.1.2. Literature related to tax avoidance

Tax avoidance can have implications for the government’s capacity to deliver essential services [14]. Tax avoidance is a pervasive
global phenomenon, and extensive research has been conducted to examine the determinants of tax avoidance from multiple per­
spectives in the literature.
On the one hand, internal characteristics, such as firm size and profitability [15], industry type [16], top management [17],
ownership structure, and corporate governance [18], as well as financial transparency and information disclosure, have been found to
directly influence corporate tax avoidance behavior. On the other hand, external factors also impact tax avoidance. These include the
tax system and tax rates of a country [19], auditing practices [20], market competition [21], financial crises [22], and others.
Through this study, we make a valuable contribution to the current body of research on the determinants of tax avoidance by
examining two perspectives: environmental regulations and CSR. Prior studies have extensively examined the influence of command-
and-control environmental regulations and carbon emission trading on firms’ tax avoidance. These studies have validated various
mechanisms, such as financial constraints and operational risks [23–25]. We contribute to the current body of research by specifically
emphasizing the informational aspect. We also make a notable contribution to the literature on the association between CSR and tax

2
Y. Gu and S. Wang Heliyon 9 (2023) e21492

avoidance. Previous research has investigated the influence of CSR on corporate tax avoidance; however, the results have been
contradictory and inconclusive [26,27]. In this context, EID can be classified as a form of CSR. The findings of this study provide
support for the notion that CSR is associated with an increase in tax avoidance [27], thereby strengthening the understanding of this
relationship.

2.2. Hypothesis development

EID may increase corporate tax avoidance. First, firms may view tax avoidance as a strategic measure to mitigate the unfavorable
consequences of EID. The disclosure of environmental information not only intensifies companies’ commitment to compliance but also
entails substantial investments in technology, remediation, and compensatory measures, consequently potentially bolstering brand
value and incurring reputational costs [6,8]. China faces the issue of a lenient tax enforcement framework, which consequently leads to
diminished opportunity costs for firms operating in the country in relation to tax avoidance. This situation leads them to heavily
depend on such strategies to minimize the adverse consequences from policy interventions [27,28]. Second, from an agency theory
perspective, managers may perceive EID as a risk management tool to conceal opportunistic behavior, which could increase tax
avoidance practices. Using data from listed companies, Lin et al. [29] empirically found that CSR is likely to remain a form of window
dressing in China. Thus, firms’ perceptions of EID as a means to bolster their corporate reputations and exert an insurance effect may
increase tax avoidance [30]. We propose hypothesis 1:
EID significantly promotes corporate tax avoidance.
EID also has the potential to reduce tax avoidance. First, EID plays a crucial role in alleviating the information asymmetry between
firms and investors, resulting in reduced corporate debt financing costs and increased access to credit support [4]. Furthermore,
improving the quality of corporate EID has been shown to positively impact corporate profitability [31]. When firms have better access
to financing and higher profitability, they have less incentive to engage in tax avoidance activities [32]. Second, EID has the potential
to improve the comparability of accounting information, which reduces corporate tax avoidance. Although EID may not be directly
focused on the disclosure of corporate tax information, it can provide tax authorities and analysts with additional details that aid in
cross-checking the information provided by companies in the same industry.1 Consequently, environmental disclosure has the po­
tential to increase the likelihood of corporate tax avoidance being detected, effectively discouraging firms from reducing their
engagement in such practices. Third, from a risk-averse perspective, we expect firms to reduce tax avoidance when they increase their
EID. Austin and Wilson [33] posit that a noteworthy negative correlation exists between reputation and corporate tax avoidance. In
situations in which firms face greater risk of reputational damage, they tend to engage in lower tax avoidance to mitigate the un­
warranted scrutiny that could harm their reputations. The consistent disclosure of high-quality environmental information by com­
panies serves to enhance their reputational images and contributes to a further reduction in corporate tax avoidance. Thus, we propose
hypothesis 2:
EID can reduce corporate tax avoidance.

2.3. Policy background

Numerous countries and regions have implemented diverse EID programs. Prominent examples include the U.S. Toxics Release
Inventory and the European environmental label program [34]. Providing information to the public has emerged as an effective
strategy employed by governments to mitigate environmental externalities [35]. In the context of China, tension exist between
environmental governance and economic interests [36] This conflict often hampers the implementation of effective environmental
regulatory instruments by the central government, particularly due to the presence of information asymmetry [37]. Hence, the central
government is dedicated to incentivizing EID [2], recognizing it as an essential and fundamental component of environmental
governance.
In 2003, China’s EID requirements were initiated through the issuance of the Announcement on Corporate Environmental Infor­
mation Disclosure by the State Environmental Protection Administration (SEPA). This action was based on China’s Cleaner Production
Promotion Law and aimed to stimulate public awareness and concern over environmental behavior. Subsequently, in 2007, the SEPA
further advanced environmental governance by promulgating the Environmental Information Disclosure Measures (Trial). These
measures established clear and specific obligations for both firms and environmental protection administrative departments to disclose
environmental information. This development signifies the beginning of a new era in environmental governance, where “information”
is recognized as a crucial tool in the process.
Since 2008, the Institute of Public and Environmental Affairs (IPE) and the Natural Resources Defense Council (NRDC) have jointly
published the Pollution Information Transparency Index (PITI) annually to evaluate the implementation of Environmental Information
Disclosure Measures (Trial). The evaluation index is designed to evaluate the EID status of 113 cities, with a range of 0–100 points and
8 evaluation items of different weights. Local governments attach great importance to monitoring and disclosing firms’ daily ex­
ceedance rates and violation records. Among all evaluation items, “disclosure of daily exceedance records of pollution sources” has the
highest weighting of 28 points. Subsequently, “public disclosure of information on investigated and verified cases of public complaints
about environmental problems or environmental pollution by firms” and “public disclosure on application” are given a weightage of 18

1
In practice, tax authorities employ data analysis systems to compare various information, such as the logical proportional relationship among an
enterprise’s electricity consumption, water consumption, refueling expenses, and corporate tax expenditures, with those of their industry peers.

3
Y. Gu and S. Wang Heliyon 9 (2023) e21492

points. In this paper, the term “EID policy"” is used to specifically denote the evaluation of the PITI implemented in 113 cities.
Following the implementation of the EID policy, the local governments of the 113 affected cities intensified their supervision in the
fields of “disclosure of monitoring data”, “disclosure of information on violations and complaints”, and “regular publication of
pollutant emission data of firms”. Other cities have not taken similar measures. This situation provides an opportunity for this paper to
employ the DID method to identify the net effect of EID on tax avoidance.

3. Research design

3.1. Data and sample selection

Our sample consists of A-share firms listed on the Shanghai and Shenzhen Stock Exchanges from 2004 to2012,2 and the data are
obtained from the China Stock Market and Accounting Research (CSMAR) database. To be consistent with the prior literature, we also
exclude (1) firms in the financial sector; (2) ST, *ST firms; (3) firm-years with negative pretax income; (4) instances in which the firm’s
effective tax rate is below zero or above one, and (5) firm-years missing the necessary data. We also winsorize all continuous variables
at the 1 % and 99 % levels. After excluding these firms, our final sample comprises 7848 firm-year observations. Another basic set of
data, namely, the list of cities under the EID reform, is manually collected from government decrees. For robustness tests, we acquire
city-level macroeconomic data from the China Center for Economic Research (CCER) database.

3.2. Estimation framework

3.2.1. Baseline regression model

To assess the impact of the EID policy on corporate tax avoidance, we employ the DID identification strategy. Unlike multiple
regression analysis, the DID method addresses endogeneity concerns that arise from the inability to randomly assign policies or
treatments, which can introduce estimation biases. By comparing the pre and postimplementation periods and contrasting them with
those of an unaffected control group, the DID method enables us to ascertain the actual effects of the policy or treatment on the
phenomenon under investigation, thereby capturing the causal effects more effectively. The implementation of the EID policy in
specific cities provided the necessary conditions for employing the DID approach. We use the following specification:
DDBTDit = α + βEIDit + γXit + μi + ηpt + ωit (1)

where DDBTDit is the level of corporate tax avoidance of firm i in year t. EIDit is a dummy variable indicating whether firm i is in a pilot
EID reform region in year t. The coefficient β is of interest, as it shows the impact of EID on corporate tax avoidance. If the coefficient β
is significantly positive, then EID increases firms’ tax avoidance. In contrast, if the coefficient β is significantly negative, EID suppresses
corporate tax avoidance. Xit represents a series of firm characteristic variables, μi represents a firm-level fixed effect, and ηpt represents
province-year fixed effects. The province-year fixed effects capture time-varying provincial shocks. By adding firm fixed effects, we
control for time-invariant characteristics that may affect tax avoidance in different firms. ωit is the error term. In all regressions,
standard errors are clustered at the city-year level.

3.2.2. Propensity score matching and differences-in-differences techniques

Importantly, the selection of the 113 cities for the implementation of EID was not entirely random. This nonrandom selection might
impact the comparability between the cities subjected to EID policy intervention and those that were not. Therefore, we perform
robustness tests by applying the PSM-DID method. In comparison to conventional DID methods, the PSM-DID approach exhibits
notable advantages. First, PSM-DID employs the propensity score matching technique, which assists in controlling for selection bias by
establishing a comparison group that closely resembles the treatment group. This ensures a more balanced comparison between the
treatment and control groups, thereby alleviating potential biases arising from self-selection or nonrandom assignment. Second, PSM-
DID results in an improved covariate balance between the treatment and control groups by matching individuals with similar pro­
pensity scores. This reduction in the influence of confounding variables augments the internal validity of the causal estimation.
The specific PSM procedure is as follows. First, the propensity scores are calculated using logistic regression to estimate the
probability of each firm based on firm size, return on assets, cash holdings, cost of debt ratio, urban economic development, and
financial pressure. Second, using the calculated propensity scores obtained from logit regression, we estimate the PSM model using the
one-to-many nearest neighbor matching technique.3

2
In 2013, the evaluation criteria for the PITI were changed, seven new cities were included, and the disclosure of the urban PITI was expanded to
120 cities. Therefore, to ensure the consistency of the evaluation method, we select the original 113 cities as the study population and set the sample
period from 2004 to 2012.
3
We conduct balancing tests to ensure the credibility of the matching procedure. The results show no significant difference between the ob­
servables (covariates) of the treatment and control groups after matching.

4
Y. Gu and S. Wang Heliyon 9 (2023) e21492

3.3. Event study

To examine whether the time trends exhibit similar patterns between cities in the control and treatment groups during the period
before the implementation of the reform, we extend the DID analysis to an event-time specification [38] to estimate annual changes in
tax avoidance before and after the EID reform. Specifically, we apply the following specification:
∑3+
DDBTDit = α + βk Dt + γXit + πi + δpt + ωit
k≥− 4 i0+k
(2)

The dummy variable, Dti0+k , is a series of event-time dummies that equal one when the EID reform is k years away in a firm. In
particular, ti0 denotes the year when firm i implemented the EID reform, k = − 4, − 3, − 2, 0,1,2,3. The omitted time category is k = − 1,
so the posttreatment effects are relative to the period one year before the start of the reform. The definitions of the other variables are
the same as those in model (1). By focusing on our variable of interest, βk, the specification provides evidence to check for any pre­
existing trends, which in turn is an efficient way to evaluate whether there is selection bias in our main results.

3.4. Measurement of variables

3.4.1. Dependent variable

The dependent variable in this paper is corporate tax avoidance. The tax avoidance proxies widely used in the academic literature
can generally be divided into two categories. First, tax avoidance is measured by the effective tax rate (ETR) and its variants. Since
listed companies in China enjoy a wide range of tax incentives and have different nominal tax rates [39], this indicator tends to result in
a horizontal incomparability of the degree of corporate tax avoidance. Second, Manzon and Plesko [40] used the book-tax difference
(BTD) to measure tax avoidance. Desai and Dharmapala [41] argued that it is more reasonable to use DDBTD to describe the degree of
tax avoidance after deducting the effect of accrued profits. In light of the relative advantages of both approaches, we employ DDBTD as
the primary measure of firms’ tax avoidance and use BTD and ETR as alternative measures for robustness checks (see section 4.2.4. for
more details). A higher BTD or DDBTD indicates higher tax avoidance, while a higher ETR indicates lower tax avoidance.
The DDBTD indicator is measured as follows:
BTDit = α0 + α1 TAit + μi + εit (3)
∑3+
DDBTDit = α + βk k≥− 4
Dti0+k + γXit + πi + δpt + ωit (4)

where i and t denote firm and year, respectively. BTDit is measured as the difference between the pretax accounting profit and taxable
income as a percentage of total assets. TAit denotes total accruals, which is equal to the difference between net profits and cash flows
from operating activities as a percentage of total assets. μi is the mean residual for firm i over the sample period; εit is the deviation from
the mean residual of firm i in year t.

3.4.2. Independent variables

The key independent variable, EID, is the interaction of Treati and Postt. Treati is a dummy variable that equals one if firm i is located
in one of the 113 cities implementing the EID reform. Postt is a dummy variable that equals one if year t falls in 2008 or later and zero
otherwise.

3.4.3. Control variables

Following prior studies [21,24,42,43], we control for the following variables: firm size (Size), return on assets (ROA), asset-liability
ratio (Leverage), capital intensity ratio (Capital), inventory intensity ratio (Inventory) and cash holdings (Cash) to rule out alternative
explanations. table A1in the online Appendix provides detailed definitions of the variables, and summary statistics for the key variables
used in the regression analyses are provided in Table 1.

3.4.4. Other variables

In section 4.3 of the mechanism analysis and section 4.4 of the heterogeneity analysis, we introduce new variables. Here, we
provide the meaning of these variables.

3.4.4.1. Investment. Drawing on Richardson’s [44] measure of intrafirm investment, we define investment as (cash paid for the pur­
chase of fixed assets, intangible assets, and other long-term assets - net cash recovered from the disposal of fixed assets, intangible
assets, and other long-term assets - depreciation) as a proportion of total assets.

3.4.4.2. Industry competition level. The Herfindahl–Hirschman Index (HHI5), measured as the sum of the squared market shares of the
top five companies in the industry, represents industry competition [45]. The HHI5 is negatively correlated with industry competition:
the higher its value is, the weaker is industry competition.

3.4.4.3. Local government tax effort. We refer to extant studies [46–48] and develop the following regression model to characterize the
tax effort of tax authorities:

5
Y. Gu and S. Wang Heliyon 9 (2023) e21492

Table 1
Summary statistics.
Variables Observations Mean SD Min Max

DDBTD 7848 0.0004 0.0268 − 0.0996 0.1110

EID 7848 0.5437 0.4981 0.0000 1.0000
Size 7848 21.7625 1.1306 19.4399 25.3807
ROA 7848 0.0515 0.0401 − 0.0110 0.1954
Leverage 7848 0.4658 0.1909 0.0462 0.8715
Capital 7848 0.2657 0.1768 0.0030 0.7555
Inventory 7848 0.1779 0.1515 0.0000 0.7399
Cash 7848 0.1881 0.1304 0.0107 0.7383

TEit = α0 + α1 ln (GDP)it + α2 IND2i,t + α3 IND3i,t + α4 INCOMEit + α4 INVESTit + εit (5)

where i and t index city and year, TEit denotes the city’s level of tax effort measured as tax revenue as a share of GDP, ln (GDP)it
represents the logarithm of GDP, and IND2i,t and IND3i,t represent the ratios of the secondary and tertiary industries to GDP, respec­
tively. INCOMEit represents the logarithm of urban per capita disposable income, and INVESTit represents the proportion of fixed asset
investment to GDP. εit is the error term.4 The predicted values of the explanatory variables obtained by the regression model (5)
represent the estimated tax capacity, followed by the calculation of the ratio of the actual tax burden to the estimated tax capacity,
which gives us the required tax effort variable.

3.4.4.4. Local government competition. Drawing on Li et al. [49], we define the minimum number of cities that constitute 70 % of the
total GDP of a province as the number of effectively competitive cities. The higher the number of effectively competing cities is as a
percentage of the total number of cities, the higher is the degree of homogeneity of prefecture-level cities in that province, implying
that the competition faced by local governments is more intense. We use the share of effectively competitive cities to measure the level
of government competition.

4. Empirical analysis

4.1. Baseline regression results

Table 2 shows the results of our baseline specification (1) using DDBTD as the dependent variable. We first control for firm and year
fixed effects in column (1) to run the regressions while clustering at the firm-year level. The result shows a positive and statistically
significant relationship between EID reform and firms’ tax avoidance, implying that environmental regulations may have increased tax
avoidance activities in the affected firms. Column (2) reflects replacing the standard errors by clustering at the city-year level.
Province-year fixed effects are included in the estimates in column (3). In column (4), a further is taken to add firm size (Size), return
on assets (ROA), asset-liability ratio (Leverage), capital intensity ratio (Capital), inventory intensity ratio (Inventory), and cash
holdings (Cash) as additional control variables. We consistently find that EID has a positive and statistically significant effect on tax
avoidance, indicating that our previous results are quite robust across these alternative specifications. Thus, we test hypothesis 1.
This conclusion is in line with the literature that suggests a positive association between environmental regulations and corporate
tax avoidance [23–25]. Using the 11th Five-Year Plan in China as a quasinatural experiment, Geng et al. [23] stated that environmental
regulations contribute to an increase in tax avoidance. Diverging from prior research that primarily examined command-and-control
environmental regulations, we offer empirical evidence on the association between environmental regulations and corporate tax
avoidance from an informational standpoint.

4.2. Robustness checks

4.2.1. Event study

Table 3 shows the results of our specification (2). The estimates in Table 3 show that the coefficient of βk was statistically
nonsignificant before the EID reform implementation, indicating no significant differences between the treatment and control groups
in the pretreatment period. This result gives us more confidence in the validity of our identification strategy and indicates, to some
extent, that the existence of unobservable factors does not cause our baseline results.

4.2.2. Placebo test

To further explore whether the results are affected by unobservable factors [50], we conduct a placebo test by randomly assigning
the implementation of the EID reform to cities [51]. We randomly select 113 cities from the full sample as the false treatment group,
False_treat, and then re-estimated our benchmark model to perform a placebo test. Since the data generation process is completely
random, the false EID variables should not produce significant estimates. To increase the identification power, we repeat this process

4
The explanatory variables we chose in our model follow those used in the traditional tax effort literature.

6
Y. Gu and S. Wang Heliyon 9 (2023) e21492

Table 2
Effects of EID on tax avoidance.
Dependent variable DDBTD

(1) (2) (3) (4)

EID 0.0050** 0.0050** 0.0046** 0.0050**

(0.0023) (0.0024) (0.0024) (0.0023)
Size 0.0011
(0.0015)
ROA 0.1424***
(0.0208)
Leverage − 0.0199***
(0.0067)
Capital − 0.0114**
(0.0053)
Inventory − 0.0144*
(0.0076)
Cash − 0.0163**
(0.0067)
Constant − 0.0023* − 0.0023* − 0.0022 − 0.0167
(0.0013) (0.0014) (0.0013) (0.0313)
Observations 7848 7848 7848 7848
R-squared 0.0504 0.0504 0.1031 0.1244
Firm FE Y Y Y Y
Year FE Y Y Y Y
Province–Year FE N N N Y

Note: *** denotes significance at 1 %, ** at 5 %, and * at 10 %. All regressions control for year, firm, and province-year fixed effects. Robust standard
errors are reported in parentheses, column (1) represents clustering at the firm-year level and columns (2)–(4) represent clustering at the city-year
level.

Table 3
Event study.
Dependent variable DDBTD

t-4 0.0033
(0.0038)
t-3 0.0045
(0.0032)
t-2 − 0.0002
(0.0032)
Year of EID(t) 0.0104**
(0.0045)
t+1 0.0107***
(0.0038)
t+2 0.0035
(0.0037)
t+3 0.0048*
(0.0029)
Constant 0.2094***
(0.0304)
Observations 7848
R-squared 0.1219
Control Variable Y
Firm FE Y
Province–Year FE Y

Note: *** denotes significance at 1 %, ** at 5 %, and * at

10 %. All regressions control for firm and province-year
fixed effects. Robust standard errors are reported in pa­
rentheses, clustered at the city-year level.

500 times.
Fig. 1 shows the estimated coefficient values of the false EID variables for the 500 simulation tests. The distribution of the placebo
test estimates is centered around zero, indicating that the randomly assigned EID variable does not significantly affect tax avoidance.
According to the baseline estimates in column (4) of Table 2, labeled with red auxiliary lines in Fig. 1, the EID coefficient estimates are
outside the distribution of the placebo coefficients estimated in our simulation exercise. These observations suggest that a positive and
significant impact of EID on tax avoidance exists.

7
Y. Gu and S. Wang Heliyon 9 (2023) e21492

4.2.3. PSM-DID test

Column (2) of Table 4 reflects the results of robustness tests by applying the PSM-DID method. We still find a positive and sta­
tistically significant (at the 5 % level) correlation between the EID reform and tax avoidance, and the coefficient is larger than that of
the benchmark result in Table 2.

4.2.4. Alternative measurement of tax avoidance

In columns (2) and (3) of Table 4, we replace the dependent variable to examine whether the measure of tax avoidance mainly
drives our basic results. Regarding the alternative measures of firm tax avoidance, columns (2) and (3) show the results from using BTD
and ETR as dependent variables, respectively. The higher the value of BTD is, the higher the degree of tax avoidance and the value of
ETR are, and the lower the degree of tax avoidance is. As shown, our results are consistent with previous findings that show a positive
effect of EID on tax avoidance.

4.2.5. Extra control variables included

We also perform robustness tests by controlling for additional variables. Following prior studies [23], we further add
prefecture-level control variables, including the proportion of secondary industry to gross domestic product (GDP), the proportion of
tertiary industry to GDP, social fixed assets investment as a share of GDP, and the logarithm of GDP. As shown in column (4) of Table 3,
the coefficient of the EID reform is persistently positive and statistically significant at the 1 % level. We also add industry-year fixed
effects based on model (1) to exclude time-varying industrial characteristics. As reported in column (5) of Table 4, the result remains
robust and show that the EID reform is positive and significant at the 10 % level.

4.2.6. Excluding municipalities

Considering the large differences between municipalities and prefecture-level cities in China, we use the subsample excluding the
municipalities (Beijing, Shanghai, Chongqing, Tianjin) in the regression of column (6) of Table 4. The result shows that the positive
impact of EID on tax avoidance is still valid, which is consistent with the benchmark regression result in column (4) of Table 2.

4.2.7. Concurrent event exclusion

During the sample period, China also implemented several policies that may bias the estimates, such as the SO2 emissions trading
scheme, low-carbon city pilot and mandatory CSR disclosure. To further validate the effectiveness, we consider contemporaneous
policy factors and conduct robustness tests. The results remain robust.5

4.3. Mechanism analysis

4.3.1. Green innovation

To meet EID requirements, firms may undertake efforts to improve their green innovation. However, this response can impose an
increased burden on firms, thereby stimulating their incentives to engage in tax avoidance. To test this channel, we replace the
explanatory variable with the financing constraint and rerun regression model (1). The level of green innovation is measured as the
logarithm of green patents. Column (1) of Table 5 reports the estimates of the effect of corporate EID on green innovation. The
regression coefficients are significantly positive, suggesting that EID improves firms’ green innovation capabilities. This outcome is in
line with the existing body of literature, as supported by the research conducted by Ding et al. [52]. Moreover, the observed increase in
corporate green innovation may serve as a mechanism through which environmental disclosure facilitates corporate tax avoidance.

4.3.2. Capital investment

EID has the potential to stimulate firms to make substantial investments in upgrading their production facilities, while corporate tax
avoidance might be driven by the necessity for corporate capital investments. To examine this mechanism, we substitute the
explanatory variable with Investment and re-estimate the regression using model (1). Column (2) of Table 5 presents the estimation
results, and the significant positive coefficient indicates that EID has a direct and positive impact firms’ investments, supporting the
literature on environmental regulations and corporate investments [12]. Another mechanism through which EID reform may posi­
tively affect tax avoidance is an increase in capital investments. This indicates that EID primarily affects firms’ tax avoidance through
the channel of increased capital investments.

4.4. Heterogeneity analysis

4.4.1. Heterogeneity of industry type

EID is more likely to bring additional cost increases to heavy polluters, leading to greater tax avoidance by high-polluting firms.
Ding et al. [52] found that EID significantly increases the green patenting activity of high-polluting firms, respectively. To explore the

5
Tables will be made available on request.

8
Y. Gu and S. Wang Heliyon 9 (2023) e21492

Fig. 1. Distribution of the estimated coefficients of a falsification test. Note: The figure shows the estimated coefficient densities obtained by
randomly assigning the EID status to cities 500 times. The vertical line represents the estimated treatment effect reported in column (4) of Table 2.

Table 4
Robustness check.
Dependent variable DDBTD BTD ETR DDBTD DDBTD DDBTD

(1) (2) (3) (4) (5) (6)

EID 0.0063** 0.0053** − 0.0145** 0.0063** 0.0051* 0.0051*

(0.0031) (0.0025) (0.0068) (0.0030) (0.0028) (0.0028)
Constant − 0.0246 0.0179 0.4093*** − 0.1412* − 0.0404 − 0.0577*
(0.0390) (0.0303) (0.0812) (0.0749) (0.0322) (0.0324)
Observations 6542 7455 7620 7597 7619 7012
R-squared 0.1218 0.4432 0.5257 0.1229 0.1220 0.1278
Control Variable Y Y Y Y Y Y
Firm FE Y Y Y Y Y Y
Province–Year FE Y Y Y Y Y Y
Industry–Year FE N N N N Y N

Note: *** denotes significance at 1 %, ** at 5 %, and * at 10 %. All regressions control for firm and province-year fixed effects. The control variables
include firm size, return on assets, asset-liability ratio, capital intensity ratio, inventory intensity ratio, and cash holdings. Robust standard errors are
reported in parentheses, clustered at the city-year level.

Table 5
Mechanistic analysis.
Dependent variable green innovation Investment

(1) (2)

EID 0.0986*** 0.0071*

(0.0252) (0.0039)
Constant − 1.9604*** 0.1833***
(0.4096) (0.0478)
Observations 7848 7572
R-squared 0.7366 0.5711
Control Variable Y Y
Firm FE Y Y
Province–Year FE Y Y

Note: *** denotes significance at 1 %, ** at 5 %, and * at 10 %. All regressions control for firm and
province-year fixed effects. The control variables include firm size, return on assets, asset-liability
ratio, capital intensity ratio, inventory intensity ratio, and cash holdings. Robust standard errors
are reported in parentheses, clustered at the city-year level.

9
Y. Gu and S. Wang Heliyon 9 (2023) e21492

impact of industries, we divide firms into high-polluting and nonpolluting industries.6 Columns (1) and (2) of Table 6 show the effect of
EID on corporate tax avoidance in highly and nonhighly polluting industries, respectively. The impact of EID reform on tax avoidance
is significant in high-pollution industries but is not obvious in other industries.

4.4.2. Heterogeneity of industry competition level

Generally, firms in highly competitive markets have more difficulty passing on costs than do their peers in less competitive markets,
which may lead to a greater incentive to evade taxes. To explore the influence of the industry competition level, we divide the firms
into two groups of high and low industry competition levels according to the median of the HHI5. The regression results are presented
in columns (3) and (4) of Table 6. The positive correlation between EID and tax avoidance is mainly revealed among firms with a
substantial degree of industry competition than their peers with weak industry competition.

4.4.3. Heterogeneity of local government tax effort

Much of the literature on the determinants of corporate tax avoidance suggests that the extent of tax avoidance is affected by the tax
efforts of tax authorities [53,54]. To examine the impact of tax efforts on corporate tax avoidance, we divide the sample companies into
two groups using the median of the annual local government tax efforts. Columns (1) and (2) of Table 7 show the regression results
grouped by high and low tax effort, respectively. The results show that the significant impact of EID on corporate tax avoidance is
mainly found in cities with low tax efforts, in line with our predictions.

4.4.4. Heterogeneity of local government competition

Local governments in highly competitive environments may relax tax enforcement for companies in their jurisdiction, increasing
corporate tax avoidance activities [57,58]. To explore the impact of local government competition, we divide the sample into two
subsamples based on the median. Columns (3) and (4) of Table 7 show the regression results grouped by weak and strong local
government competition, respectively. The findings indicate that firms tax avoidance increases as the interregional competition faced
by local governments in a firm’s jurisdiction increases.

5. Conclusion

5.1. Empirical results

Using data on Chinese A-share listed companies from 2004 to 2012, we empirically investigate for the first time the impact of
corporate EID on tax avoidance and its related mechanisms. The main findings of this study can be summarized as follows. First, in this
study, we provide compelling evidence that EID can promote corporate tax avoidance. Through a comprehensive and rigorous
analysis, the study’s findings remain robust, demonstrating the reliability and validity of the observed relationship between EID and
tax avoidance. Second, we identify two distinct channels through which EID influences corporate tax avoidance: green innovation and
capital investments. Companies that actively pursue green innovation initiatives and make substantial investments in capital assets
exhibit a greater propensity to employ strategies that enable them to reduce their tax burdens. Third, the heterogeneity analysis
conducted in the study focuses on exploring the specific environments in which the impact of EID on tax avoidance is most pronounced.
The analysis reveals that the effects of EID on tax avoidance are particularly heightened in highly polluting industries, highly
competitive industries, and regions in China characterized by low tax effort and intense government competition.
Our research findings support the existing conclusions that environmental regulations contribute to increased corporate tax
avoidance. Additionally, we have substantiated the specific mechanisms through which EID increases corporate tax avoidance.
Moreover, by considering EID as a component of CSR, we have provided additional evidence for the relationship between CSR and tax
avoidance, demonstrating that CSR activities also contribute to increased corporate tax avoidance.

5.2. Policy recommendations

The disclosure of environmental information assumes a pivotal role in the ongoing battle against climate change and the
enhancement of environmental quality. Nonetheless, it is imperative for governments to consider the economic costs imposed on firms
when formulating environmental regulations and to adopt measures to mitigate any adverse consequences stemming from such
regulations. We make the following policy recommendations.
First, tax authorities must strengthen their efforts to collect and manage revenues from firms affected by environmental regulations.
Our results suggest that firms are more likely to engage in tax avoidance in highly competitive and highly polluting industries, as well
as in areas characterized by weak tax efforts and competitive local governments. Consequently, it is strongly recommended that tax
authorities intensify enforcement measures in these specific areas to alleviate the inequity arising from regional and industry dis­
parities in firm tax burdens. Second, the government should extend policy support to bolster technological innovation and advance the
sustainable development of firms. For instance, feasible measures can be taken, such as providing green subsidies to companies that

6
According to the “EID guide for Chinese listed companies” issued by the Ministry of Environmental Protection of China in 2010, we define the
following 16 industries as high-polluting: mining, textile, coal, tanning, pharmaceutical, brewing, fermentation, petrochemical, chemical, steel,
cement, metallurgy, paper making, building materials, electrolytic aluminum, and thermal power [55].

10
Y. Gu and S. Wang Heliyon 9 (2023) e21492

Table 6
Heterogeneity of industry type and industry competition level.
Dependent variable DDBTD

(1) (2) (3) (4)

EID 0.0077** 0.0027 0.0045* 0.0006

(0.0036) (0.0031) (0.0026) (0.0058)
Constant − 0.0094 − 0.0126 0.1587*** − 0.0514
(0.0678) (0.0343) (0.0409) (0.0508)
Observations 2263 5558 4374 2976
R-squared 0.2274 0.1355 0.1836 0.1779
Control Variable Y Y Y Y
Firm FE Y Y Y Y
Province–Year FE Y Y Y Y

Note: *** denotes significance at 1 %, ** at 5 %, and * at 10 %. All regressions control for firm and province-year fixed effects. The control variables
include firm size, return on assets, asset-liability ratio, capital intensity ratio, inventory intensity ratio, and cash holdings. Robust standard errors are
reported in parentheses, clustered at the city-year level.

Table 7
Heterogeneity of local government tax effort and competition.
Dependent variable DDBTD

(1) (2) (3) (4)

EID 0.0043 0.0080** 0.0010 0.0125***

(0.0072) (0.0038) (0.0045) (0.0041)
Constant − 0.0140 − 0.0589 − 0.0669 − 0.0710
(0.0601) (0.0517) (0.0483) (0.0526)
Observations 3116 3761 3975 3417
R-squared 0.1853 0.1851 0.1554 0.1595
Control Variable Y Y Y Y
Firm FE Y Y Y Y
Province–Year FE Y Y Y Y

Note: *** denotes significance at 1 %, ** at 5 %, and * at 10 %. All regressions control for firm and province-year fixed effects. The control variables
include firm size, return on assets, asset-liability ratio, capital intensity ratio, inventory intensity ratio, and cash holdings. Robust standard errors are
reported in parentheses, clustered at the city-year level.

exhibit low emission levels, augmenting support for green credit accessible to businesses, and intensifying tax incentives targeted at
green technologies and pollution control equipment. Third, it is advisable for the government to cooperate with firms, cultivating a
close partnership that encompasses sharing information on environmental regulations. Moreover, the government can play a crucial
role by offering valuable technical guidance and advisory support, effectively mitigating the costs associated with policy imple­
mentation for firms. This multifaceted approach supports sustainable practices while ensuring fair and effective tax compliance.

5.3. Limitations and future research

First, listed companies have a distinct advantage in terms of greater financing opportunities and lower financing costs than do other
small and medium-sized enterprises. This advantage enables them to better navigate the cost pressures associated with EID policy.
Considering the comprehensive and extensive nature of the research, we suggest expanding the research scope in subsequent steps to
include small and medium-sized enterprises. Doing so will facilitate a more comprehensive understanding of the challenges and op­
portunities they encounter with respect to EID.
Second, we concentrated exclusively on the identification of two mechanisms: green innovation and corporate investments. Due to
data limitations, we did not explore the existence of other mechanisms. As postulated in hypothesis 1, the implementation of EID
policies may contribute to tax avoidance through the “insurance effect” arising from the fulfillment of CSR or the proactive relaxation
of tax enforcement by local governments. This issue warrants further exploration and can be effectively addressed in future research.
Finally, while the research findings of this study indicate that, on the whole, EID may contribute to an increase in tax avoidance, it is
crucial to acknowledge that theoretically, EID still presents possibilities that could reduce tax avoidance. Moreover, within the scope of
this study, it is challenging to separate the net effects that contribute to an increase or decrease in tax avoidance. This aspect warrants
further exploration in future research endeavors.

Author contributions

Yu Gu: Conceived and designed the experiments; Analyzed and interpreted the data; Contributed reagents, materials, analysis tools
or data; Wrote the paper.
Shanhui Wang: Performed the experiments; Contributed reagents, materials, analysis tools or data; Wrote the paper.

11
Y. Gu and S. Wang Heliyon 9 (2023) e21492

Funding

This study has been funded by the National Social Science Fund of China (Grant number 3CJY004) and the Zhejiang Provincial
Philosophy and Social Sciences Planning Project.

Availability of data and materials

The original contributions presented in the study are included in the article/supplementary material, and further inquiries can be
directed to the corresponding author.

Declaration of competing interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to
influence the work reported in this paper.

Appendix A. Supplementary data

Supplementary data to this article can be found online at https://doi.org/10.1016/j.heliyon.2023.e21492.

References

[1] S. Konar, M.A. Cohen, Information as regulation: the effect of community right to know laws on toxic emissions, J. Environ. Econ. Manag. 32 (1997) 109–124,
https://doi.org/10.1006/jeem.1996.0955.
[2] L. Zhang, A.P.J. Mol, G. He, Transparency and information disclosure in China’s environmental governance, Curr. Opin. Environ. Sustain. 18 (2016) 17–24,
https://doi.org/10.1016/j.cosust.2015.03.009.
[3] L. Du, X. Wang, J. Peng, G. Jiang, S. Deng, The impact of environmental information disclosure quality on green innovation of high-polluting enterprises, Front.
Psychol. 13 (2022), 1069354, https://doi.org/10.3389/fpsyg.2022.1069354.
[4] M. Du, S. Chai, W. Wei, S. Wang, Z. Li, Will environmental information disclosure affect bank credit decisions and corporate debt financing costs? Evidence from
China’s heavily polluting industries, Environ. Sci. Pollut. Res. 29 (2022) 47661–47672, https://doi.org/10.1007/s11356-022-19229-4.
[5] Z. Zhang, Z. Su, K. Wang, Y. Zhang, Corporate environmental information disclosure and stock price crash risk: evidence from Chinese listed heavily polluting
companies, Energy Econ. 112 (2022), 106116, https://doi.org/10.1016/j.eneco.2022.106116.
[6] S. Wang, H. Wang, J. Wang, F. Yang, Does environmental information disclosure contribute to improve firm financial performance? An examination of the
underlying mechanism, Sci. Total Environ. 714 (2020), 136855, https://doi.org/10.1016/j.scitotenv.2020.136855.
[7] S. Ren, W. Wei, H. Sun, Q. Xu, Y. Hu, X. Chen, Can mandatory environmental information disclosure achieve a win-win for a firm’s environmental and economic
performance? J. Clean. Prod. 250 (2020), 119530 https://doi.org/10.1016/j.jclepro.2019.119530.
[8] J. Meng, Z.X. Zhang, Corporate environmental information disclosure and investor response: evidence from China’s capital market, Energy Econ. 108 (2022),
105886, https://doi.org/10.1016/j.eneco.2022.105886.
[9] W. Luo, X. Guo, S. Zhong, J. Wang, Environmental information disclosure quality, media attention and debt financing costs: evidence from Chinese heavy
polluting listed companies, J. Clean. Prod. 231 (2019) 268–277, https://doi.org/10.1016/j.jbusres.2022.03.089.
[10] Y. Luo, G. Xiong, A. Mardani, Environmental information disclosure and corporate innovation: the "inverted U-shaped" regulating effect of media attention,
J. Bus. Res. 146 (2022) 453–463, https://doi.org/10.1016/j.jbusres.2022.03.089.
[11] Q. Zhang, W. Chen, Y. Feng, The effectiveness of China’s environmental information disclosure at the corporate level: empirical evidence from a quasi-natural
experiment, Resour. Conserv. Recycl. 164 (2021), 105158, https://doi.org/10.1016/j.resconrec.2020.105158.
[12] L. Huang, Z. Lei, How environmental regulation affect corporate green investment: evidence from China, J. Clean. Prod. 279 (10) (2021), 123560. https://
doiorg/10.1016jjclepro.2020.1235607.
[13] M. Fonseka, T. Rajapakse, G. Richardson, The effect of environmental information disclosure and energy product type on the cost of debt: evidence from energy
firms in China, Pac. Basin Finance J. 54 (2018) 159–182, https://doi.org/10.1016/j.pacfin.2018.05.001.
[14] R. Bird, K. Davis-Nozemack, Tax avoidance as a sustainability problem, J. Bus. Ethics 151 (2018) 1009–1025, https://doi.org/10.1007/s10551-016-3162-2.
[15] S.D. Dyreng, M. Hanlon, E.L. Maydew, Long-run corporate tax avoidance, Account. Rev. 83 (1) (2008) 61–82, https://doi.org/10.2308/accr.2008.83.1.61.
[16] M. Hanlon, S. Heitzman, A review of tax research, J. Account. Econ. 50 (2–3) (2010) 127–178, https://doi.org/10.1016/j.jacceco.2010.09.002.
[17] S.D. Dyreng, M. Hanlon, E.L. Maydew, The effects of executives on corporate tax avoidance, Account. Rev. 85 (4) (2010) 1163–1189, https://doi.org/10.2308/
accr.2010.85.4.1163.
[18] S. Chen, X. Chen, Q. Cheng, T. Shevlin, Are family firms more tax aggressive than non-family firms? J. Financ. Econ. 95 (1) (2010) 41–61, https://doi.org/
10.1016/j.jfineco.2009.02.003.
[19] L. Oats, P. Tuck, Corporate tax avoidance: is tax transparency the solution? Account, Bus. Res 49 (2019) 565–583, https://doi.org/10.1080/
00014788.2019.1611726.
[20] J.L. Hoopes, D. Mescall, J.A. Pittman, Do IRS audits deter corporate tax avoidance? Account, Rev 87 (5) (2012) 1603–1639, https://doi.org/10.2308/accr-
50187.
[21] H. Cai, Q. Liu, Competition and corporate tax avoidance: evidence from Chinese industrial firms, Econ. J. 119 (2009) 764–795, https://doi.org/10.1111/j.1468-
0297.2009.02217.x.
[22] G. Richardson, G. Taylor, R. Lanis, The impact of financial distress on corporate tax avoidance spanning the global financial crisis: evidence from Australia,
Econ. Modell. 44 (2015) 44–53, https://doi.org/10.1016/j.econmod.2014.09.015.
[23] Y. Geng, W. Liu, K. Li, H. Chen, Environmental regulation and corporate tax avoidance: a quasi-natural experiment based on the eleventh five-year plan in
China, Energy Econ. 99 (2021), 105312, https://doi.org/10.1016/j.eneco.2021.105312.
[24] H. Yu, L. Liao, S. Qu, D. Fang, L. Luo, G. Xiong, Environmental regulation and corporate tax avoidance: a quasi-natural experiments study based on China’s new
environmental protection law, J. Environ. Manage. 296 (2021), 113160, https://doi.org/10.1016/j.jenvman.2021.113160.
[25] C. Feng, X. Zhu, Y. Gu, Y. Liu, Does the carbon emissions trading policy increase corporate tax avoidance? evidence from China, Front. Energy Res. 9 (2021),
821219, https://doi.org/10.3389/fenrg.2021.821219.
[26] C.K. Hoi, Q. Wu, H. Zhang, Is corporate social responsibility (CSR) associated with tax avoidance? Evidence from irresponsible CSR activities, Account. Rev. 88
(6) (2013) 2025–2059, https://doi.org/10.2308/accr-50544.

12
Y. Gu and S. Wang Heliyon 9 (2023) e21492

[27] W. Jiang, C. Zhang, C. Si, The real effect of mandatory CSR disclosure: evidence of corporate tax avoidance, Technol. Forecast. Soc. Change 179 (2022), 121646,
https://doi.org/10.1016/j.techfore.2022.121646.
[28] M. Nguyen, J.H. Nguyen, Economic policy uncertainty and firm tax avoidance. Account, Finance 60 (2020) 3935–3978, https://doi.org/10.1111/acfi.12538.
[29] Y.T. Lin, N.C. Liu, J.W. Lin, Firms’ adoption of CSR initiatives and employees’ organizational commitment: organizational CSR climate and employees’ CSR-
induced attributions as mediators, J. Bus. Res. 140 (2022) 626–637, https://doi.org/10.1016/j.jbusres.2021.11.028.
[30] D.B. Minor, J. Morgan, CSR as reputation insurance: primum non nocere, Calif. Manag. Rev. 53 (2011) 40–59, https://doi.org/10.1525/cmr.2011.53.3.40.
[31] N. Ahmad, H.Z. Li, X.L. Tian, Increased firm profitability under a nationwide environmental information disclosure program? Evidence from China, J. Clean.
Prod. 230 (2019) 1176–1187, https://doi.org/10.1016/j.jclepro.2019.05.161.
[32] M. Amidu, W. Coffie, P. Acquah, Transfer pricing, earnings management and tax avoidance of firms in Ghana, J. Financ. Crime 26 (2019) 235–259, https://doi.
org/10.1108/JFC-10-2017-0091.
[33] C.R. Austin, R.J. Wilson, An examination of reputational costs and tax avoidance: evidence from firms with valuable consumer brands, J. Am. Tax. Assoc 39
(2017) 67–93, https://doi.org/10.2308/atax-51634.
[34] H. Karl, C. Orwat, Environmental labelling in Europe: European and national tasks, Eur. Environ. 9 (1999) 212–220, https://doi.org/10.1002/(sici)1099-0976
(199909/10)9:5<212::aid-eet203>3.0.co;2-i.
[35] US EPA, Toxics Release Inventory (TRI) 2016 National Analysis, Environmental Protection Agency, Washington, DC, 2016.
[36] P. Zhang, Target interactions and target aspiration level adaptation: how do government leaders tackle the “environment-economy” nexus? Public Adm. Rev 81
(2020) 220–230, https://doi.org/10.1111/puar.13184.
[37] R. Ran, Perverse incentive structure and policy implementation gap in China’s local environmental politics, J. Environ. Policy Plan. 15 (2013) 17–39, https://
doi.org/10.1080/1523908X.2012.752186.
[38] L.S. Jacobson, R.J. Lalonde, D.G. Sullivan, Earnings losses of displaced workers, Am. Econ. Rev. 83 (1993) 685–709, https://doi.org/10.17848/wp92-11.
[39] L. Wu, Y. Wang, W. Luo, P. Gillis, State ownership, tax status and size effect of effective tax rate in China, Account. Bus. Res. 42 (2012) 97–114, https://doi.org/
10.1080/00014788.2012.628208.
[40] G.B. Manzon, G.A. Plesko, The relation between financial and tax reporting measures of income, Tax L. Rev. 55 (2001) 175–195, https://doi.org/10.2139/
ssrn.264112.
[41] M.A. Desai, D. Dharmapala, Corporate tax avoidance and high-powered incentives, J. Financ. Econ. 79 (2006) 145–179, https://doi.org/10.1016/j.
jfineco.2005.02.002.
[42] D.M. Christensen, D.S. Dhaliwal, S. Boivie, S.D. Graffin, Top management conservatism and corporate risk strategies: evidence from managers’ personal political
orientation and corporate tax avoidance, Strateg. Manag. J. 36 (2015) 1918–1938, https://doi.org/10.1002/smj.2313.36, 12 1918-1938.
[43] G. Richardson, B. Wang, X. Zhang, Ownership structure and corporate tax avoidance: evidence from publicly listed private firms in China, J. Contemp. Account.
Econ. 12 (2) (2016) 141–158, https://doi.org/10.1016/j.jcae.2016.06.003.
[44] S. Richardson, Over-investment of free cash flow, Rev. Acc. Stud. 11 (2006) 159–189, https://doi.org/10.1007/s11142-006-9012-1.
[45] S.A. Rhoades, The Herfindahl-Hirschman index, Fed. Res. Bull. 79 (1993) 188. https://EconPapers.repec.org/RePEc:fip:fedgrb:y:1993:i:mar:p:188-189:n:v.
79no.3.
[46] R. Bahl, A regression approach to tax effort and tax ratio analysis, IMF Econ. Rev. 18 (1971) 570–612, https://doi.org/10.2307/3866315.
[47] R.M. Bird, J. Martinez-Vazquez, B. Torgler, Tax effort in developing countries and high income countries: the impact of corruption, voice and accountability,
Econ. Anal. Policy. 38 (2008) 55–71, https://doi.org/10.1016/S0313-5926(08)50006-3.
[48] S. Chen, Y. Ye, K. Jebran, Tax enforcement efforts and stock price crash risk: evidence from China, J. Int. Financ. Manag. Account. 33 (2022) 193–218, https://
doi.org/10.1111/jifm.12145.
[49] X. Li, C. Liu, X. Weng, L.A. Zhou, Target setting in tournaments: theory and evidence from China, Econ. J. 129 (2019) 2888–2915, https://doi.org/10.1093/ej/
uez018.
[50] M. Bertrand, E. Duflo, S. Mullainathan, How much should we trust differences-in-differences estimates? Q. J. Econ. 119 (1) (2004) 249–275, https://doi.org/
10.1162/003355304772839588.
[51] R. Chetty, A. Looney, K. Kroft, Salience and taxation: theory and evidence, Am. Econ. Rev. 99 (2009) 1145–1177, https://doi.org/10.1257/aer.99.4.1145.
[52] J. Ding, Z. Lu, C.H. Yu, Environmental information disclosure and firms’ green innovation: evidence from China, Int. Rev. Econ. Finance 81 (2022) 147–159,
https://doi.org/10.1016/j.iref.2022.05.007.
[53] M.A. Desai, A. Dyck, L. Zingales, Theft and taxes, J. Financ. Econ. 84 (2007) 591–623, https://doi.org/10.1016/j.jfineco.2006.05.005.
[54] J. Kovermann, P. Velte, The impact of corporate governance on corporate tax avoidance – a literature review, J. Int. Account. Audit. Tax 36 (2019), 100270,
https://doi.org/10.1016/j.intaccaudtax.2019.100270.
[55] X. Chen, W. Qian, Effect of marine environmental regulation on the industrial structure adjustment of manufacturing industry: an empirical analysis of China’s
eleven coastal provinces, Mar. Policy 113 (2020), 103797, https://doi.org/10.1016/j.marpol.2019.103797.

13