Sage Journals: Discover world-class research

Abstract

This study investigates how each environmental, social, and governance (ESG) dimension relates to tax avoidance in well- and poorly governed firms. The analysis uses data from 2016 to 2020 on A-share companies listed on the Shanghai and Shenzhen stock exchanges and three panel methodologies: pooled ordinary least squares, random effects, and fixed effects models. The results show that not all ESG dimensions are equally associated with tax avoidance, and the relationship between ESG and tax avoidance differs between well- and poorly governed firms. For well-governed firms, the social and governance dimensions positively correlate with tax avoidance, whereas the environmental dimension does not. For poorly governed firms, none of the three dimensions correlate with tax avoidance. These findings suggest that ESG and tax avoidance are more compatible than conflicting. Moreover, they underscore the importance of good corporate governance in enabling tax avoidance as an ESG-compatible strategy. This study addresses the debatable relationship between ESG and tax avoidance, which has significant implications for stakeholders, including managers, investors, and policymakers.

Plain Language Summary

Exploring the relationship between ESG and tax strategies

This study explores the relationship between environmental, social, and governance (ESG) practices and tax avoidance among A-share companies listed on the Shanghai and Shenzhen stock exchanges from 2016 to 2020. Specifically, it examines whether this relationship differs between well- and poorly governed firms. The findings reveal that the social and governance dimensions of ESG are linked to higher tax avoidance in well-governed firms, whereas the environmental dimension is not. In poorly governed firms, none of the ESG dimensions is significantly related to tax avoidance. This suggests that ESG practices can align with tax avoidance strategies, and good governance is crucial for this alignment. This study provides valuable insights for managers, investors, and policymakers on the importance of good governance in harmonizing tax strategies with ESG.

Keywords

ESG tax governance sustainability firm value

Introduction

Environmental, social, and governance (ESG) is a framework used by investors and stakeholders to evaluate corporate sustainability. The concept of ESG gained prominence after its introduction in the 2004 “Who Cares Wins” Report published by 20 financial institutions in response to a call by United Nations Secretary-General Kofi Annan (Gillan et al., 2021). The report emphasizes that firms that effectively manage ESG issues can generate long-term value for shareholders because these issues are expected to significantly impact firms’ future competitiveness and financial performance. Therefore, rather than the traditional corporate social responsibility (CSR), which centers on social responsibility and ethics, ESG focuses on creating corporate value through sustainable practices.

As environmental risk increases and business competition intensifies, ESG has become crucial for companies. Recently, taxes have been actively discussed in the context of ESG with legislative proposals mandating the disclosure of tax havens and offshore tax evasion. This legislation can be cited as the “Corporate Governance Improvement and Investor Protection Act,” which includes Title V Disclosure of tax havens and offshoring (Sturgis, 2022). The increasing emphasis on tax in ESG discussions stems from the recognition that a company’s approach to taxes is closely related to the extent to which it engages in ESG issues and builds trust with its stakeholders (Morris & Visser, 2022). Understanding the relationship between ESG and tax avoidance is crucial, as it demonstrates how companies strive to enhance their corporate value and fulfill stakeholders’ expectations through sustainable practices. Despite its importance, the link between these two concepts remains ambiguous according to existing theories and empirical research.

First, tax avoidance is considered socially irresponsible and a failure to fulfill social responsibilities because it undermines social welfare by transferring value from the state to shareholders (Avi-Yonah, 2008; Freedman, 2003; Friese et al., 2008; Landolf, 2006; Slemrod, 2004). From this perspective, tax avoidance contradicts the purpose of ESG activities; thus, the two have a negative relationship (Yoon et al., 2021). However, tax avoidance can be perceived as a legitimate strategy for enhancing corporate value and social welfare because reducing tax payments leads to higher after-tax profits, which positively impacts investment and economic development (Chamley, 1986; Desai & Dharmapala, 2006; Judd, 1985; Ohrn, 2018). Thus, ESG and tax avoidance can be compatible strategies for increasing firm value and may have a positive relationship. To address this ongoing debate, this study investigates whether tax avoidance is a conflicting or compatible strategy with ESG activities.

This study also examines whether the relationship between ESG and tax avoidance strategies varies according to corporate governance. Effective governance mechanisms in well-governed firms can mitigate the risks of tax avoidance, such as managerial rent diversion, and ensure that benefits accrue to the firm and its shareholders (Bayar et al., 2018; Desai & Dharmapala, 2009). Therefore, the goal of tax avoidance to enhance corporate value is more likely to align with ESG activities in well-governed firms than in poorly governed ones. Thus, if ESG and tax avoidance are compatible strategies for enhancing firm value, the positive association between them becomes more apparent in well-governed firms than in poorly governed ones. Conversely, if ESG and tax avoidance are conflicting strategies, effective governance mechanisms may reinforce the negative relationships. Strong governance implies greater transparency and accountability, which can reduce risky tax-avoidance practices and positively affect ESG activities (Christensen & Murphy, 2004; Lanis & Richardson, 2011). In this case, the negative relationship between ESG and tax avoidance becomes more evident in well-governed firms than in poorly governed ones. Therefore, an analysis that considers corporate governance is necessary to clarify the relationship between ESG and tax avoidance. In particular, the organizational structure of Chinese companies is mostly pyramidal, complex, and prone to agency problems, making them suitable for exploring the role of governance in the relationship between ESG and tax avoidance (Hai et al., 2022).

This study conducts three panel analyses: pooled ordinary least squares (OLS), random effects (RE), and fixed effects (FE) models on 1,841 firm-year observations of listed Chinese firms from 2016 to 2020. In China, the government has implemented strict environmental regulations to address environmental issues. These regulations are often considered the primary drivers influencing corporations’ decisions and actions (Feng et al., 2020; Zhang et al., 2022). For instance, beginning with the 2016 guidelines mandating that listed companies disclose environmental information, the government has continually refined these regulations to enhance the standardization of ESG reporting practices among Chinese firms. The guidelines announced in 2024 require listed firms to commence mandatory disclosures covering a broad range of ESG topics starting in 2026. Driven by these regulations, the implementation of ESG practices in Chinese companies has evolved rapidly. The National Plan to Control Total Emissions of Major Pollutants during the Eleventh Five-Year Plan Period (11th Five-Year Plan), initiated in 2006, is a notable environmental policy aimed at mitigating pollution (Fan et al., 2019). These regulations often lead to increased operational risks and financial constraints for companies, prompting them to avoid taxes (Geng et al., 2021). Given this regulatory landscape, this study presents an invaluable opportunity to gain deeper insight into how companies navigate their tax strategies within the ESG framework in a government-led regulatory environment.

This study contributes to existing literature in several ways. First, it is the first to provide empirical evidence on how ESG and tax avoidance are related by analyzing a substantial sample of Chinese-listed firms. The results of the three panel regressions show that the social dimension of ESG is positively associated with tax avoidance, which is attributable to unobserved firm-specific characteristics. The findings suggest that ESG and tax avoidance can be compatible strategies driven by unobserved firm-specific factors, such as the shared goal of enhancing firm value, which challenges the traditional view that they are conflicting approaches. Second, this study enriches ESG literature by exploring the role of corporate governance in the relationship between ESG and tax avoidance. It finds a positive association between the social and governance dimensions and tax avoidance in well-governed firms but not in poorly governed ones. This suggests that ESG and tax avoidance strategies can coexist and enhance corporate value in well-governed firms because good governance mitigates the risks associated with tax avoidance and ensures that tax benefits are channeled toward the firm. Third, this study decomposes ESG into three dimensions and finds that each dimension has a different correlation with tax avoidance. This supports the need for researchers to decompose the sub-dimensions of ESG and analyze them separately. Finally, this study has significant implications for various stakeholders, including managers, investors, and policymakers. This can help managers understand how to reconcile ESG and tax avoidance strategies to maximize firm value. Firms engaging in ESG and tax avoidance strategies under good governance can be viewed as having attractive investment opportunities. For policymakers, it suggests that corporate governance should be considered when designing regulations requiring the disclosure of tax strategies for ESG reporting. The remainder of this paper is organized as follows. Section “Literature Review and Hypotheses” describes the relevant literature and proposes the hypotheses. Section “Research Design and Empirical Methodology” introduces the empirical methodology used to test the hypotheses. Sections “Results” and “Robustness Test” present the empirical results and robustness tests, respectively. Sections “Discussion” and “Implications and Limitations” present the discussion, implications, and limitations. Finally, Section “Conclusion” concludes the study.

Literature Review and Hypotheses

Stakeholder theory is commonly adopted in studies on the relationship between ESG and tax avoidance strategies. According to the stakeholder theory, firms are responsible not only for investors and creditors but also for other non-financial stakeholder groups that can influence firm outcomes (Freeman, 1984). Consequently, they should prioritize cultivating positive relationships with stakeholders through socially responsible businesses (Alessa et al., 2024). However, the relationship between ESG and tax avoidance strategies under the stakeholder theory can be either positive or negative, contingent on conflicting viewpoints on tax avoidance.

First, tax avoidance practices that shift value from the state to shareholders may fail to fulfill social responsibilities by not paying a fair share of taxes. As corporate tax payments play a crucial role in financing public goods, tax avoidance policies that prevent corporations from paying a fair share of taxes undermine social welfare (Freedman, 2003; Friese et al., 2008; Landolf, 2006; Slemrod, 2004). Thus, tax avoidance and aggressive tax planning are considered socially irresponsible and threaten corporate legitimacy (Avi-Yonah, 2008; Erle, 2008; Huseynov & Klamm, 2012; Schön, 2008). Additionally, tax avoidance practices can generate reputational and legal risks as regulators and tax authorities may act against firms that engage in aggressive tax planning (Hanlon & Slemrod, 2009). Consequently, investors and other stakeholders may view tax avoidance negatively, potentially leading to stock market evaluation. Tax avoidance that is socially irresponsible and damages corporate value contradicts the purpose of ESG activities. Thus, a negative relationship is expected between ESG and tax avoidance because companies that actively implement ESG are less likely to engage in tax avoidance practices. Consistent with this view, Yoon et al. (2021) and Jiang et al. (2024) suggested a negative relationship between ESG performance and tax avoidance.

However, tax avoidance can be perceived as a legitimate strategy for enhancing corporate value (Desai & Dharmapala, 2006). Because taxes constitute a significant expense for firms, reducing tax payments can result in higher after-tax profits, which can be reinvested in new projects or distributed to shareholders in the form of dividends. This perspective is consistent with previous studies demonstrating that corporate taxes can negatively affect investment and economic development, thereby detracting from social welfare. Corporate taxes reduce after-tax returns, resulting in efficiency losses in resource allocation and hampering capital accumulation and productivity growth (Chamley, 1986; Judd, 1985). They also have detrimental effects on investment and entrepreneurship (Djankov et al., 2010). Higher tax rates discourage R&D investments, ultimately hindering economic growth (Ohrn, 2018). Therefore, reducing tax payments can be considered a fair business practice that improves community welfare. Increased after-tax profits can lead to investments and job creation, thereby benefiting the income level of the community and other tax revenue sources. Thus, tax avoidance does not hinder social welfare because increased after-tax profits contribute to community income levels and other sources of tax revenue (Sikka, 2012). If managers view tax avoidance as a strategy to enhance corporate value rather than conflicting with social responsibility, ESG and tax avoidance are compatible strategies for maximizing firm value. Thus, firms that actively commit to ESG activities are likely to engage in tax avoidance strategies, resulting in a positive relationship. In support of this argument, Davis et al. (2016), Gulzar et al. (2018), and Lanis and Richardson (2012) suggested that firms with high CSR performance are more likely to engage in tax avoidance. Similarly, Abdelfattah and Aboud (2020) revealed a positive relationship between ESG ratings and the degree of tax avoidance.

As discussed above, the relationship between ESG and tax avoidance remains debatable, and an empirical investigation is needed to verify whether these two strategies are compatible or conflicting. Therefore, the nondirectional Hypothesis 1 is proposed.

Hypothesis 1: There is no relationship between ESG and tax avoidance strategy.

From the perspective of the agency theory, corporate governance plays a crucial role in determining the impact of tax avoidance on firm value (Desai & Dharmapala, 2009). Tax avoidance can increase firm value. However, for poorly governed firms, managerial rent diversion can offset or even outweigh this benefit, thereby damaging firm value (Desai & Dharmapala, 2006, 2009). In poorly governed firms plagued by agency problems and opaque information environments (Bayar et al., 2018; Kim et al., 2011), managers can divert the benefits or rents generated by tax avoidance for personal gain. This can lead to a misallocation of resources, reduced transparency, and decreased firm value. Therefore, effective governance mechanisms are crucial to mitigate the risks associated with tax avoidance and ensure that benefits accrue to the firm and its shareholders (Bayar et al., 2018). Wilson (2009) supported this argument by showing that tax-sheltering firms with strong governance exhibit positive abnormal returns.

Because the impact of tax avoidance on firm value is mediated by firm governance, the relationship between ESG activities and tax avoidance strategies in Hypothesis 1 may vary depending on corporate governance. As strong governance prevents managerial rent diversion in tax avoidance, a tax avoidance strategy is likely to align with ESG activities to enhance corporate value in well-governed firms. Thus, the positive association between ESG and tax avoidance is expected to be more apparent in well-governed firms than in poorly governed ones.

By contrast, if tax avoidance contracts ESG because it negatively impacts society and corporate value (Landolf, 2006; Slemrod, 2004), well-governed firms reduce their tax avoidance activities to legitimize their existence (Christensen & Murphy, 2004; Lanis & Richardson, 2011). Strong governance implies that a company operates with greater transparency and accountability, which can reduce risky tax-avoidance practices and positively affect ESG activities. Thus, if ESG and tax avoidance are conflicting strategies, effective governance mechanisms may reinforce their negative relationship. This negative relationship is more pronounced in well-governed firms than in poorly governed ones.

Therefore, corporate governance is crucial to better understand the relationship between ESG and tax avoidance strategies. Based on this, the second hypothesis is proposed.

Hypothesis 2: The relationship between ESG and tax avoidance strategies varies, depending on corporate governance.

Research Design and Empirical Methodology

Data

The data included A-share companies listed on the Shanghai and Shenzhen stock exchanges from 2016 to 2020. The sample consisted of (1) manufacturing firms with a December fiscal year-end, (2) firms with financial statements obtained from the China Stock Market and Accounting Research Database, and (3) firms with ESG scores obtained from the Bloomberg Database (Liu & Lee, 2023). Companies with missing data or capital impairment were excluded, as were companies with taxable income less than or equal to zero, because they had no incentive to avoid taxes. The final dataset comprised 1,841 firm-year observations. Table 1 summarizes the data selection process used in this study.

Table 1.

The Data Selection Process.

Firm-year observations of manufacturing firms with December fiscal year during the 2016–2020 period	2,555
Less
Missing data for variables	616
Firms with impaired capital	20
Firms with taxable income below zero	78
Final number of firm-year observations	1,841

Measurement of Variables

Tax Avoidance

Tax avoidance was measured using the model developed by Desai and Dharmapala (2006). This assumes that the total difference between book income and taxable income arises from earnings management and tax avoidance activities. Thus, the book-tax difference not explained by earnings management accounts for tax avoidance (Frank et al., 2009). In this model, the proxy for earnings management is total accruals (AC), which allows for managerial discretion in financial reporting decisions. Accordingly, the residual calculated by estimating Model (1) represents the level of tax avoidance (TA).

BT D_{i, t} = A C_{i, t} + ε_{i, t}

(1)

where BTD is the total difference between book and taxable income for firm i in year t divided by total assets in year t−1; AC = total accruals (=net income − operating cash flows) for firm i in year t divided by total assets in year t−1.

ESG

Third-party rating agencies evaluate and rate corporate ESG activities. For example, Bloomberg collects public information disclosed by firms and evaluates their activities annually using a score ranging from 0 to 100, where a score close to 100 is the best score that a company can achieve (Huber & Comstock, 2017). In addition to the aggregate evaluation scores for ESG activities, scores for individual ESG subdimensions were also provided. As ESG data from Bloomberg are a proprietary score widely accepted as a stable and accurate measure, they are widely used in research (Minutolo et al., 2019). This study also uses Bloomberg scores as proxies for corporate ESG performance.

Corporate Governance

This study employs a governance score to comprehensively assess firms’ corporate governance structures. The Bloomberg governance scores encompass two crucial aspects: board composition and executive compensation. The former evaluates a board’s capacity to offer various perspectives, ensure proper management oversight, and identify potential structural risks. The latter assesses how effectively a company’s pay policies align the interests of top executives with those of other stakeholders in the long term. These scores rank the companies’ relative performance across four key areas: diversity (including gender and age), refreshment (entrenchment and balance of tenure), director roles (overboarding of directors, chairpersons, and CEO), and independence (encompassing board leadership and director independence) (Bloomberg, 2021).

Research Model

This study employs three panel methodologies to examine the relationship between ESG and tax avoidance strategies: pooled OLS, RE, and FE models. It builds a regression model in which the dependent variable is the level of tax avoidance (TA) estimated using the model developed by Desai and Dharmapala (2006). The main independent variable is the ESG score (ESG) obtained from the Bloomberg database. The environmental (ENV), social (SOC), and governance (GOV) scores, which are sub-dimensions of the ESG, are also used separately as independent variables to derive more comprehensive results. To control for factors affecting tax avoidance, the model includes several variables: firm size (SIZE), leverage (LEV), return on assets (ROA), operating cash flow (OCF), change in sales (GROWTH), change in investment (INV), a negative operating income dummy (LOSS), and year indicator variables (YR). The regression model is as follows:

\begin{matrix} T A_{t} = β_{0} + β_{1} ES G_{t} (EN V_{t} or SO C_{t} or GO V_{t}) + β_{2} SIZ E_{t} \\ + β_{3} LE V_{t} + β_{4} RO A_{t} + β_{5} OC F_{t} + β_{6} GROWT H_{t} \\ + β_{7} IN V_{t} + β_{8} LOS S_{t} + β_{9} \sum YR + ε_{t} \end{matrix}

(2)

where TA is the level of tax avoidance measured by Desai and Dharmapala’s (2006) model; ESG = aggregate ESG score; ENV = environmental score; SOC = social score; GOV = governance score; SIZE = the natural logarithm of total assets; LEV = total liabilities divided by total assets; ROA = net income divided by total assets; OCF = operating cash flows divided by total assets; GROWTH = the change in sales from year t−1 to year t; INV = the change in investment in tangible and intangible assets from year t−1 to year t; LOSS = 1 if a firm’s operating income is less than zero, and 0 otherwise; YR = year-fixed dummies.

For Hypothesis 2, Model (2) is estimated for the two subgroups after classifying the total sample into well- and poorly governed groups, categorized according to their governance scores. The methodology for dividing the total samples into well- and poorly governed groups follows the approach of Khanchel and Lassoued (2023). The well-governed group comprises firms with governance scores higher than the mean value of the total sample by year, whereas the poorly governed group comprises firms with governance scores lower than the mean value of the total sample by year. This study then compares the significance of the ESG (ENV, SOC, GOV) coefficients of the two subgroups to verify whether the relationship between ESG and tax-avoidance strategies varies depending on corporate governance. In addition, subgroups are derived by dividing the sample into quartiles and defining the fourth and first quartiles as those with good and poor governance, respectively. The results of the analysis using these extreme quartiles are not tabulated but are consistent with those of the main analysis.

Results

Descriptive Statistics

Table 2 presents the descriptive statistics of the variables in the analysis. The mean (median) value of tax avoidance (TA) is 0.020 (0.018), with minimum and maximum values of −0.153 and 0.150, respectively. The mean (median) ESG score (ESG) is 24.284 (22.314), with minimum and maximum values of 5.785 and 64.115, respectively. The mean (median) environmental score (ENV) is 14.161 (11.628), with minimum and maximum values of 1.550 and 65.625, respectively. The mean (median) of the social score (SOC) is 26.082 (22.807), with minimum and maximum values of 5.263 and 77.193, respectively. The mean (median) governance score (GOV) is 46.205 (48.214), with minimum and maximum values of 3.571 and 64.539, respectively. These values indicate that listed Chinese companies have the highest governance scores (GOV), followed by social (SOC) and environmental scores (ENV). For the control variables, the mean (median) values of firm size (SIZE), debt ratio (LEV), and return on assets (ROA) are 23.418 (23.303), 0.456 (0.470), and 0.049 (0.040), respectively. The mean (median) values of operating cash flows (OCF), change in sales (GROWTH), and change in investment (INV) are 0.066 (0.061), 0.177 (0.105), and 0.126 (0.041), respectively. The mean (median) value of the dummy variable for firms with negative operating income (LOSS) is 0.067 (0.000), indicating that firms with negative operating income constitute approximately 6.7% of the sample.

Table 2.

Descriptive Statistics of Variables.

Variables	Number	Mean	Median	SD	Minimum	Maximum
TA	1,841	0.020	0.018	0.045	−0.153	0.150
ESG	1,841	24.284	22.314	7.959	5.785	64.115
ENV	1,841	14.161	11.628	10.352	1.550	65.625
SOC	1,841	26.082	22.807	10.048	5.263	77.193
GOV	1,841	46.205	48.214	5.278	3.571	64.539
SIZE	1,841	23.418	23.303	1.152	21.218	26.440
LEV	1,841	0.456	0.470	0.172	0.077	0.829
ROA	1,841	0.049	0.040	0.051	−0.122	0.211
OCF	1,841	0.066	0.061	0.063	−0.092	0.248
GROWTH	1,841	0.177	0.105	0.364	−0.356	2.402
INV	1,841	0.126	0.041	0.322	−0.325	2.065
LOSS	1,841	0.067	0.000	0.250	0.000	1.000

Note. Definition of variables: TA = the level of tax avoidance measured by Desai and Dharmapala’s (2006) model; ESG = aggregate ESG score; ENV = environmental score; SOC = social score; GOV = governance score; SIZE = the natural logarithm of total assets; LEV = total liabilities divided by total assets; ROA = net income divided by total assets; OCF = operating cash flows scaled by total assets; GROWTH = the change in sales from year t−1 to year t; INV = the change in investment in tangible and intangible assets from year t−1 to year t; LOSS = 1 if a firm’s operating income is less than zero, and 0 otherwise.

Correlation Analysis

Table 3 presents the results of Pearson’s correlations among all the variables included in the analysis. The level of tax avoidance (TA) is positively correlated with aggregate ESG (ESG), environmental (ENV), and social scores (SOC) but not significantly correlated with the governance score (GOV). Moreover, the level of tax avoidance (TA) is positively correlated with return on assets (ROA), operating cash flows (OCF), change in sales (GROWTH), and change in investment (INV), whereas it is negatively correlated with the debt ratio (LEV) and the dummy variable for loss-making firms (LOSS).

Table 3.

Correlation Matrix of Variables.

Variables	TA	ESG	ENV	SOC	GOV	SIZE	LEV	ROA	OCF	GROWTH	INV	LOSS
TA	1.0000	0.0832	0.0749	0.0804	0.0039	0.0213	−0.2213	0.6515	0.5204	0.1695	0.1139	−0.4378
		0.0004	0.0013	0.0006	0.8665	0.3622	<.0001	<.0001	<.0001	<.0001	<.0001	<.0001
ESG		1.0000	0.9491	0.8010	0.5706	0.4677	0.1810	0.0397	0.0907	−0.0157	0.0163	−0.0574
			<.0001	<.0001	<.0001	<.0001	<.0001	0.0887	<.0001	0.5019	0.4851	0.0137
ENV			1.0000	0.6382	0.4262	0.4658	0.1783	0.0521	0.0947	−0.0115	0.0171	−0.0689
				<.0001	<.0001	<.0001	<.0001	0.0254	<.0001	0.6232	0.4622	0.0031
SOC				1.0000	0.3665	0.3032	0.0824	0.0257	0.0561	−0.0138	0.0125	−0.0166
					<.0001	<.0001	0.0004	0.2708	0.0161	0.5555	0.5917	0.4777
GOV					1.0000	0.3339	0.1832	−0.0334	0.0060	−0.0438	−0.0036	−0.0255
						<.0001	<.0001	0.1523	0.7970	0.0605	0.8783	0.2749
SIZE						1.0000	0.5230	−0.0145	0.0298	0.0442	0.1037	−0.0632
							<.0001	0.5334	0.2012	0.0582	<.0001	0.0067
LEV							1.0000	−0.3956	−0.1836	0.0532	0.0657	0.1517
								<.0001	<.0001	0.0225	0.0048	<.0001
ROA								1.0000	0.5593	0.2001	0.0826	−0.4824
									<.0001	<.0001	0.0004	<.0001
OCF									1.0000	0.0818	−0.0262	−0.1956
										0.0004	0.2606	<.0001
GROWTH										1.0000	0.4549	−0.1389
											<.0001	<.0001
INV											1.0000	−0.0971
												<.0001
LOSS												1.0000

Note. The variables are defined in Table 2.

Regression Analysis

Table 4 presents the results of the pooled OLS, RE, and FE models applied to estimate regression model (2), verifying Hypothesis 1. In the OLS model, the coefficient of ESG is positive and significant at the 5% level for tax avoidance (TA). However, in the RE and FE models, the coefficients of ESG are not statistically significant. This finding suggests that higher ESG scores are associated with increased tax avoidance in the OLS model but not in the RE and FE models. Therefore, when analyzed using the aggregate ESG scores, there is no significant association or causality between ESG and tax avoidance. As the aggregate ESG scores cannot distinguish between activities in the three sub-dimensions, we decomposed ESG into environmental, social, and governance to more specifically analyze the relationship between activities in each dimension and tax avoidance.

Table 4.

OLS, RE, and FE Regression Results on the Relationship Between ESG and Tax Avoidance.

Variables	OLS model		RE model		FE model
Variables	Coef.	t-stat	Coef.	z-stat	Coef.	t-stat
Intercept	0.0065	0.36	0.0013	0.04	−0.0002	0.00
ESG	0.0003	2.46**	0.0002	1.26	−0.0001	−0.26
SIZE	−0.0011	−1.24	−0.0010	−0.67	−0.0007	−0.11
LEV	0.0056	0.97	0.0082	0.75	0.0008	0.04
ROA	0.3687	17.03***	0.4216	10.09***	0.4849	10.09***
CFO	0.1771	12.00***	0.1996	10.28***	0.2170	11.16***
GROWTH	0.0018	0.77	0.0003	0.07	−−0.0015	−0.33
INV	0.0082	3.14***	0.0048	1.17	0.0030	0.75
LOSS	−0.0325	−9.43***	−0.0259	−5.82***	−0.0204	−4.59***
YR	Included		Included		Included
R ²	.4916		.4915		.4816
Hausman test (FEM vs. REM)	χ² = 50.13 (Prob > χ² = 0.0000)
Number of observations	1,841

Note. These variables are listed in Table 2.

and *** indicate significance levels at 5% and 1%, respectively, based on two-tailed tests.

Table 5 presents the regression results for the estimation of Model (2), in which the three ESG sub-dimensions were used exclusively as the main independent variables. Panels A, B, and C of Table 5 report the results of using the environmental, social, and governance scores as independent variables. In Panels A and C, the coefficients of ENV and GOV are statistically insignificant in all three models (OLS, RE, and FE). In Panel B, the coefficients on SOC exhibit positive and statistically significant associations with TA at the 1% and 5% significance levels in the OLS and RE models, respectively. This suggests that higher social scores are associated with increased tax avoidance. However, in the FE, which controls for unobserved firm-specific characteristics, the coefficient of SOC is not statistically significant. These findings show that a positive relationship exists between social activities and tax avoidance practices, which may be attributed to unobserved heterogeneity across firms rather than a direct causal relationship (Love & Smith, 2010). This interpretation was supported by the results of the Hausman test, in which the p-value is below the critical value of .05. This rejects the null hypothesis that there is no correlation between the error term and independent variables in the model, meaning that firm-specific factors are correlated with the regressors as omitted variables in the model (Greene, 2008).

Table 5.

OLS, RE, and FE Regression Results on the Relationship Between Three Sub-Dimensions of ESG and Tax Avoidance.

	OLS model		RE model		FE model
Variables	Coef.	t-stat	Coef.	z-stat	Coef.	t-stat
Panel A. ENV
ENV	0.0001	1.35	0.0001	0.43	−0.0002	−0.78
Control	Included		Included		Included
R ²	.4904		.4903		.4831
Hausman test (FEM vs. REM)	χ² = 58.16 (Prob > χ² = 0.0000)
Panel B. SOC
SOC	0.0003	3.46***	0.0002	1.97**	0.0001	0.26
Control	Included		Included		Included
R ²	.4932		.4931		.4833
Hausman test (FEM vs. REM)	χ² = 49.12 (Prob > χ² = 0.0000)
Panel C. GOV
GOV	0.0001	0.82	0.0001	0.47	−0.0001	−0.27
Control	Included		Included		Included
R ²	.4901		.4901		.4820
Hausman test (FEM vs. REM)	χ² = 46.22 (Prob > χ² = 0.0000)
Number of observations	1,841

Note. These variables are listed in Table 2.

and *** indicate significance levels at 5% and 1%, respectively, based on two−tailed tests.

In summary, among the three ESG sub-dimensions, a positive relationship exists between social activities and tax avoidance practices, which is attributed to unobserved firm-specific characteristics. If these firm-specific characteristics include a firm’s goal or motivation, the positive relationship between social activities and tax avoidance can be interpreted as the result of firms viewing both as strategies to increase their value. In other words, firms committed to social activities engage in more tax avoidance practices, indicating that they are more likely to perceive ESG and tax avoidance as compatible rather than as conflicting strategies.

Hypothesis 2 was examined by dividing the sample into two subgroups categorized according to governance scores: well-governed and poorly governed. Table 6 presents the results of applying the OLS, RE, and FE models to estimate the regression model (2) for both groups. The OLS model shows that the coefficient of ESG is positive and significant for TA at the 1% level in Panel A for the well-governed group, but in Panel B for the poorly governed group, the coefficient of ESG is not statistically significant for TA. Thus, in the OLS model, ESG and tax avoidance strategies are positively related in firms with good governance but not in firms with weak governance. However, in both the RE and FE models, the coefficients of ESG are insignificant for both the well- and poorly governed groups. These results suggest no significant correlation or causality between aggregate ESG scores and tax avoidance levels in either group. For a more specific analysis, we decomposed ESG into environmental, social, and governance and analyzed whether the relationship between activities in each dimension and tax avoidance varied for the well- and poorly governed groups.

Table 6.

OLS, RE, and FE Regression Results of the Relationship Between ESG and Tax Avoidance for Well- and Poorly Governed Groups.

Variables	OLS model		RE model		FE model
Variables	Coef.	t-stat	Coef.	z-stat	Coef.	t-stat
Panel A. Well-governed group
Intercept	0.0012	0.05	−0.0169	0.39	0.0714	0.44
ESG	0.0003	2.60***	0.0003	1.48	0.0001	0.21
SIZE	−0.0009	−0.78	−0.0001	0.04	−0.0034	0.48
LEV	0.0020	0.26	−0.0066	0.41	−0.0225	1.00
ROA	0.2922	10.05***	0.3292	6.27***	0.3893	6.45***
CFO	0.1888	9.43***	0.2174	8.88***	0.2357	9.31***
GROWTH	0.0063	1.83	0.0084	1.62	0.0072	1.34
INV	0.0116	3.31***	0.0588	1.85*	0.0474	1.52
LOSS	−0.0329	−7.25***	−0.0238	4.93***	−0.0175	3.85***
YR	Included		Included		Included
R ²	.4599		.4645		.4337
Hausman test (FEM vs. REM)	χ² = 43.06 (Prob > χ² = 0.0000)
Number of observations	1,030
Panel B. Poorly governed group
Intercept	0.0110	0.36	0.0087	0.21	−0.3652	1.53
ESG	0.0003	1.31	0.0002	0.55	−0.0003	0.40
SIZE	−0.0015	−1.03	−0.0015	0.78	0.0148	1.41
LEV	0.0114	1.28	0.0188	1.55	0.0189	0.58
ROA	0.4554	13.86***	0.5027	8.09***	0.5361	6.84***
CFO	0.1671	7.65***	0.1801	5.93***	0.1993	6.39***
GROWTH	−0.0027	−0.82	−0.0058	0.94	−0.0081	1.10
INV	0.0029	0.72	0.0023	0.36	−0.0006	0.09
LOSS	−0.0304	−5.77***	−0.0248	3.19***	−0.0225	2.56***
YR	Included		Included		Included
R ²	.5363		.5411		.4551
Hausman test (FEM vs. REM)	χ² = 19.46 (Prob > χ² = 0.0781)
Number of observations	811

Note. These variables are listed in Table 2.

and *** indicate significance levels at 10% and 1%, respectively, based on two-tailed tests.

Table 7 presents the results of applying the OLS, RE, and FE models to estimate regression model (2), with the environmental score as the independent variable for the well-governed and poorly governed groups. In both groups, the coefficients of ENV are statistically insignificant for TA in all three models (OLS, RE, and FE). These results indicate no correlation or causality between environmental activities and tax avoidance, regardless of corporate governance.

Table 7.

OLS, RE, and FE Regression Results of the Relationship Between Environmental Scores and Tax Avoidance for Well- and Poorly Governed Groups.

Variables	OLS model		RE model		FE model
Variables	Coef.	t-stat	Coef.	z-stat	Coef.	t-stat
Panel A. Well-governed group
ENV	0.0001	1.36	0.0001	0.73	0.0000	0.13
Control	Included		Included		Included
R ²	.4573		.4619		.4317
Hausman test (FEM vs. REM)	χ² = 42.41 (Prob > χ² = 0.0000)
Number of observations	1,030
Panel B. Poorly governed group
ENV	0.0001	0.62	0.0000	0.22	−0.0002	0.36
Control	Included		Included		Included
R ²	.5356		.5405		.4293
Hausman test (FEM vs. REM)	χ² = 18.48 (Prob > χ² = 0.1019)
Number of observations	811

Note. These variables are listed in Table 2.

Table 8 presents the results of applying the OLS, RE, and FE models to estimate regression model (2) with social scores as the independent variables for the well- and poorly governed groups. In Panel A, for well-governed firms, both the OLS and RE models show that the coefficients on SOC are positive and statistically significant for TA. This finding suggests that well-governed firms engage in higher tax avoidance when their social scores are high. However, when the FE model is applied, the relationship between social scores and tax avoidance becomes statistically insignificant. This change in significance when moving from the OLS and RE models to the FE model indicates that the observed positive correlation between social scores and tax avoidance in well-governed firms does not indicate a causal link. The results of the Hausman test, with a p-value of less than .05, show unobserved firm-specific characteristics that affect both social scores and tax avoidance. This finding suggests that the positive correlation between the two can be attributed to unobserved heterogeneity across firms. In Panel B, for poorly governed firms, the coefficients of SOC are statistically insignificant across all three models (OLS, RE, and FE), indicating no correlation or causality between social scores and tax avoidance. Thus, well-governed firms that are more committed to social activities tend to engage in more tax-avoidance practices, whereas this phenomenon does not occur in poorly governed firms. These results underscore the role of corporate governance in the relationship between ESG and tax avoidance. Effective governance mitigates the risks associated with tax avoidance and ensures that benefits accrue to the firm, enabling tax avoidance to be used as a corporate value-enhancing strategy, along with ESG.

Table 8.

OLS, RE, and FE Regression Results of the Relationship Between Social Scores and Tax Avoidance for Well- and Poorly Governed Groups.

Variables	OLS model		RE model		FE model
Variables	Coef.	t-stat	Coef.	z-stat	Coef.	t-stat
Panel A. Well-governed group
SOC	0.0003	3.62***	0.0003	1.94*	0.0001	0.31
Control	Included		Included		Included
R ²	.4633		.4674		.4343
Hausman test (FEM vs. REM)	χ² = 44.28 (Prob > χ² = 0.0000)
Number of observations	1,030
Panel B. Poorly governed group
SOC	0.0002	1.24	0.0001	0.53	−0.0002	0.40
Control	Included		Included		Included
R ²	.5362		.5411		.4496
Hausman test (FEM vs. REM)	χ² = 18.43 (Prob > χ² = 0.1032)
Number of observations	811

Note. These variables are listed in Table 2.

and *** indicate significance levels at 10% and 1%, respectively, based on two-tailed tests.

Table 9 presents the results of applying the OLS, RE, and FE models to estimate regression model (2), with the governance score as the independent variable for the well- and poorly governed groups. In Panel A for the well-governed group, the OLS and RE models reveal that the coefficients of GOV are positive and statistically significant for TA. However, in the FE model, the coefficient of GOV is not statistically significant. These results suggest that, for well-governed firms, there is no causal relationship between governance activities and tax avoidance; however, there is a positive correlation, which is attributable to unobserved firm-specific factors. This is supported by the Hausman test results, in which the p-value is lower than the critical value of .05. This implies that firm-specific factors are correlated with the regressors as omitted variables in the model, and these factors may affect both governance activities and tax avoidance. By contrast, in Panel B, for the poorly governed group, the coefficients of GOV are statistically insignificant for TA across all three models (OLS, RE, and FE). Thus, no association or causality exists between governance scores and tax avoidance in poorly governed firms. These results are consistent with those of the analysis of social scores in Table 8.

Table 9.

OLS, RE, and FE Regression Results of the Relationship Between Governance Scores and Tax Avoidance for Well- and Poorly Governed Groups.

	OLS model		RE model		FE model
Variables	Coef.	t-stat	Coef.	z-stat	Coef.	t-stat
Panel A. Well-governed group
GOV	0.0008	2.70***	0.0008	2.14**	0.0003	0.45
Control	Included		Included		Included
R ²	.4602		.4651		.4344
Hausman test (FEM vs. REM)	χ² = 40.43 (Prob > χ² = 0.0001)
Number of observations	1,030
Panel B. Poorly governed group
GOV	0.0004	1.03	0.0002	0.45	0.0000	0.04
Control	Included		Included		Included
R ²	.5360		.5407		.4577
Hausman test (FEM vs. REM)	χ² = 17.71 (Prob > χ² = 0.1248)
Number of observations	811

Note. These variables are listed in Table 2.

and *** indicate significance levels at 5% and 1%, respectively, based on two-tailed tests.

In summary, when ESG is divided into three detailed dimensions, a positive relationship between ESG and tax avoidance appears for the social and governance dimensions but not for the environmental dimension. This is only observed in well-governed firms. Therefore, Hypothesis 2 is partially supported by the social and governance dimensions. Overall, the findings suggest that for well-governed firms, tax avoidance can be a strategy compatible with ESG activities driven by the shared goal of enhancing firm value, as strong corporate governance acts as a deterrent to managerial rent diversion in tax avoidance.

Robustness Test

To validate the robustness of the findings, this study conducts additional analyses using an alternative measure of tax avoidance: the cash effective tax rate (cash ETR). Cash ETR is calculated by dividing a firm’s cash taxes paid by its pre-tax income, with lower values indicating higher tax avoidance (Dyreng et al., 2008). Thus, in the robustness analysis, the dependent variable in the regression model (2) is substituted with this alternative measure, specifically using cash ETR multiplied by −1 to represent tax avoidance.

Table 10 presents the results of the pooled OLS, RE, and FE models applied to estimate regression model (2) using cash ETR as an alternative proxy for tax avoidance. In both the OLS and RE models, the coefficients of ESG are positive and significant for tax avoidance (TA). However, in the FE model, the coefficient of ESG is not statistically significant. These results indicate a positive correlation between ESG scores and tax avoidance but no causality between the two variables.

Table 10.

OLS, RE, and FE Regression Results on the Relationship Between ESG and Tax Avoidance Measured by Cash ETR.

Variables	OLS model		RE model		FE model
Variables	Coef.	t-stat	Coef.	z-stat	Coef.	t-stat
ESG	0.0028	3.72***	0.0025	3.60***	0.0002	0.15
Control	Included		Included		Included
R ²	.0761		.0811		.0679
Hausman test (FEM vs. REM)	χ² = 71.25 (Prob > χ² = 0.0000)
Number of observations	1,841

Note. These variables are listed in Table 2.

***

indicates the significance level at 1%, based on the two-tailed test.

Table 11 presents the regression results for estimating Model (2) using cash ETR as an alternative proxy for tax avoidance, with the three ESG sub-dimensions exclusively as the main independent variables. In Panels A and B, the coefficients of ENV and SOC have positive and statistically significant associations with TA in the OLS and RE models, respectively. However, in the FE model, which controls for unobserved firm-specific characteristics, the statistical significance of the coefficients for ENV and SOC diminishes. In Panel C, the coefficients of GOV are statistically insignificant across the three models (OLS, RE, and FE). Thus, additional analyses confirm the main finding of a significant correlation between social activities and tax avoidance; environmental activities also exhibited a significant correlation.

Table 11.

OLS, RE, and FE Regression Results on the Relationship Between Three Sub-Dimensions of ESG and Tax Avoidance Measured by Cash ETR.

Variables	OLS model		RE model		FE model
Variables	Coef.	t-stat	Coef.	z-stat	Coef.	t-stat
Panel A. ENV
ENV	0.0023	3.97***	0.0021	3.87***	0.0004	0.45
Control	Included		Included		Included
R ²	.0771		.0821		.0671
Hausman test (FEM vs. REM)	χ² = 31.83 (Prob > χ² = 0.0008)
Panel B. SOC
SOC	0.0013	2.34**	0.0010	1.86*	−0.0012	−0.86
Control	Included		Included		Included
R ²	.0719		.0769		.0638
Hausman test (FEM vs. REM)	χ² = 72.95 (Prob > χ² = 0.0000)
Panel C. GOV
GOV	0.0010	0.97	0.0010	0.89	−0.0004	−0.19
Control	Included		Included		Included
R ²	.0696		.0748		.0670
Hausman test (FEM vs. REM)	χ² = 68.15 (Prob > χ² = 0.0000)
Number of observations	1,841

Note. These variables are listed in Table 2.

, **, and *** indicate significance levels at 10%, 5%, and 1%, respectively, based on two-tailed tests.

Table 12 presents the regression results for estimating Model (2) using cash ETR as an alternative proxy for tax avoidance for the well- and poorly governed groups. In Panel A for the well-governed group, both the OLS and RE models show that the coefficients of ESG are significantly positive for TA. However, in Panel B for the poorly governed group, the coefficients of ESG are insignificant for TA. Thus, the positive correlation between ESG and tax avoidance is observed only for well-governed firms. In the FE model, the coefficients of ESG are insignificant for both groups.

Table 12.

OLS, RE, and FE Regression Results of the Relationship Between ESG and Tax Avoidance Measured by Cash ETR for Well- and Poorly Governed Groups.

	OLS model		RE model		FE model
Variables	Coef.	t-stat	Coef.	z-stat	Coef.	t-stat
Panel A. Well-governed group
ESG	0.0037	4.06***	0.0036	4.24***	0.0009	0.70
Control	Included		Included		Included
R ²	.0747		.0872		.0608
Hausman test (FEM vs. REM)	χ² = 78.09 (Prob > χ² = 0.0000)
Number of observations	1,030
Panel B. Poorly governed group
ESG	0.0013	0.78	0.0003	0.14	−0.0038	−0.56
YR	Included		Included		Included
R ²	.0772		.0900		.0824
Hausman test (FEM vs. REM)	χ² = 18.59 (Prob > χ² = 0.0990)
Number of observations	811

Note. These variables are listed in Table 2.

***

indicates the significance level at 1%, based on the two-tailed test.

Tables 13 to 15 present the analysis results for each of the three ESG dimensions: environmental, social, and governance, respectively. As shown in Panel A of three tables for the well-governed group, both the OLS and RE models reveal positive and statistically significant coefficients for ENV, SOC, and GOV in relation to TA. However, in Panel B, for the poorly governed group, the coefficients of ENV, SOC, and GOV are insignificant for TA. Thus, a positive correlation between ESG and tax avoidance exists only for well-governed firms across all three dimensions. In the FE model, the coefficients of ENV, SOC, and GOV are insignificant for both groups, which is consistent with the main analysis results. Overall, additional analyses using cash ETR affirm the robustness of the findings.

Table 13.

OLS, RE, and FE Regression Results of the Relationship Between Environmental Scores and Tax Avoidance Measured by Cash ETR for Well- and Poorly Governed Groups.

	OLS model		RE model		FE model
Variables	Coef.	t-stat	Coef.	z-stat	Coef.	t-stat
Panel A. Well-governed group
ENV	0.0026	3.90***	0.0025	4.16***	0.0004	0.41
Control	Included		Included		Included
R ²	.0735		.0859		.0593
Hausman test (FEM vs. REM)	χ² = 78.74 (Prob > χ² = 0.0000)
Number of observations	1,030
Panel B. Poorly governed group
ENV	0.0017	1.36	0.0014	0.96	−0.0007	−0.21
Control	Included		Included		Included
R ²	.0786		.0917		.0901
Hausman test (FEM vs. REM)	χ² = 18.39 (Prob > χ² = 0.1043)
Number of observations	811

Note. These variables are listed in Table 2.

***

indicates the significance level at 1%, based on the two-tailed test.

Table 14.

OLS, RE, and FE Regression Results of the Relationship Between Social Scores and Tax Avoidance Measured by Cash ETR for Well- and Poorly Governed Groups.

	OLS model		RE model		FE model
Variables	Coef.	t-stat	Coef.	z-stat	Coef.	t-stat
Panel A. Well-governed group
SOC	0.0017	2.65***	0.0017	2.73***	0.0006	0.46
Control	Included		Included		Included
R ²	.0661		.0787		.0592
Hausman test (FEM vs. REM)	χ² = 77.59 (Prob > χ² = 0.0000)
Number of observations	1,030
Panel B. Poorly governed group
SOC	0.0003	0.27	−0.0011	−0.76	−0.0052	−1.09
Control	Included		Included		Included
R ²	.0765		.0883		.0748
Hausman test (FEM vs. REM)	χ² = 18.94 (Prob > χ² = 0.0899)
Number of observations	811

Note. These variables are listed in Table 2.

***

indicates the significance level at 1%, based on the two-tailed test.

Table 15.

OLS, RE, and FE Regression Results of the Relationship Between Governance Scores and Tax Avoidance Measured by Cash ETR for Well- and Poorly Governed Groups.

	OLS model		RE model		FE model
Variables	Coef.	t-stat	Coef.	z-stat	Coef.	t-stat
Panel A. Well-governed group
GOV	0.0054	2.44**	0.0052	2.54**	−0.0007	−0.24
Control	Included		Included		Included
R ²	.0652		.0777		.0566
Hausman test (FEM vs. REM)	χ² = 76.92 (Prob > χ² = 0.0000)
Number of observations	1,030
Panel B. Poorly governed group
GOV	−0.0029	−1.03	−0.0030	−0.99	−0.0019	−0.52
Control	Included		Included		Included
R ²	.0777		.0910		.0888
Hausman test (FEM vs. REM)	χ² = 14.81 (Prob > χ² = 0.2518)
Number of observations	811

Note. These variables are listed in Table 2.

indicates the significance level at 5%, based on the two-tailed test.

Discussion

This study analyzes the relationship between ESG and tax avoidance in a manner that differs from previous studies. First, it disaggregates the three dimensions of ESG and independently analyzes the relationship between each dimension and tax avoidance. Second, it runs all three panel regressions using the OLS, RE, and FE models for validating the relationship between ESG and tax avoidance. Third, the relationship between the two structures is identified more clearly by considering corporate governance.

The findings of this study are summarized as follows: First, when analyzing the entire sample, the relationship between ESG and tax avoidance is positive for the social dimension but insignificant for the environmental and governance dimensions. This provides empirical evidence that ESG dimensions are differentially related to tax avoidance. Prior research has also revealed diverse relationships between the three dimensions of ESG and tax avoidance (Salah et al., 2023; Yoon et al., 2021). However, the specific relationship between each dimension and tax avoidance varies across studies. For instance, Salah et al. (2023) revealed in their analysis of French firms that the social dimension of ESG is correlated with a lower level of tax avoidance, whereas the environmental and governance dimensions are associated with a higher level of tax avoidance. These discrepancies may be attributed to variations in the ESG landscape or tax strategies influenced by the distinct policy frameworks, regulatory environments, and cultural contexts of different countries.

Second, regressions using the three panel models show a positive relationship between the social dimension and tax avoidance in the OLS and RE models; however, this relationship becomes insignificant in the FE model. This suggests that social activities and tax avoidance practices are not causally related but rather correlated with unobserved firm-specific characteristics. If these firm-specific characteristics include a firm’s motivation or goal, the positive relationship between social activities and tax avoidance can be interpreted as the result of firms viewing both as strategies to increase their value. Thus, ESG and tax avoidance appear to be more compatible than conflicting. This is consistent with previous studies that found a positive relationship between ESG and tax avoidance (Abdelfattah & Aboud, 2020; Davis et al., 2016; Lanis & Richardson, 2012).

Third, when dividing the total sample into well- and poorly governed groups, social and governance activities are positively associated with tax avoidance in well-governed firms but not in poorly governed firms. A regression analysis using three-panel models for the well-governed group shows a positive relationship between social and governance activities and tax avoidance in the OLS and RE models; however, this relationship is insignificant in the FE model. Thus, social and governance activities are positively associated with tax avoidance practices only in well-governed firms, implying that such firms are more likely to adopt tax avoidance strategies that contribute to corporate value creation while being mindful of their social and governance-related responsibilities. This highlights that good governance can mitigate the risk of managerial rent diversion, and tax avoidance practices can be accompanied by value-enhancing strategies such as ESG.

Implications and Limitations

The theoretical implication of this study is that it confirms the existence of unobserved firm-specific factors in the relationship between ESG and tax avoidance through three panel regressions using the OLS, RE, and FE models. This underscores the importance of methodological robustness when investigating the complex relationship between ESG and tax avoidance. This study also enriches the literature by drawing on agency theory in the relationship between ESG and tax avoidance, providing evidence on how governance mechanisms shape corporate behavior with respect to sustainable business practices.

The findings of this study have practical implications for various stakeholders, including managers, investors, policymakers, and society. First, managers can consider tax avoidance as a legitimate tool to increase corporate value that aligns with ESG goals. This can guide managers to harmoniously formulate tax and ESG strategies to maximize corporate value. Second, these findings may be useful to investors, especially those interested in ESG, when making investment decisions. Firms with strong corporate governance that engage in ESG and tax avoidance strategies may be considered attractive investment opportunities because they are committed to sustainable practices to enhance corporate value. Third, policymakers can draw on the findings of this study when proposing tax-related regulations in the ESG context. Policies that require firms to disclose tax-related information while encouraging them to establish good governance can reduce the risk of tax avoidance and promote the alignment of firms’ tax stratification with their ESG objectives. Finally, when tax benefits are diverted for personal gain, it can lead to the misallocation of resources and a lack of transparency, negatively impacting society and public goods. This study suggests that governance is crucial for mitigating these risks and ensuring that tax avoidance is a responsible business practice that broadly benefits society.

This study has limitations that should be addressed in future research. For corporate governance, specific factors such as board independence, executive compensation, and ownership structure could not be considered as proxies owing to difficulties in accessing relevant data. Future research should explore specific governance mechanisms that promote responsible tax strategies and sustainable business practices. For example, future studies could investigate the mediating effects of board composition, executive compensation, and ownership structure on the alignment of ESG goals with tax avoidance practices. Additionally, the findings are based on an analysis of listed Chinese companies, and caution should be exercised when applying them to companies in other countries. Given the regional differences in laws, regulations, business practices, cultures, and economic factors, future research should conduct comparative analyses of firms in different countries. By conducting cross-country comparisons, researchers can identify the contextual factors that shape firms’ approaches to ESG and tax avoidance, thereby providing more nuanced and contextually relevant insights.

Conclusion

This study investigated the relationship between ESG and tax avoidance by conducting three panel regressions using the OLS, RE, and FE models. By analyzing a sample of 1,841 firm-year observations of Chinese listed firms from 2016 to 2020, this study found partial evidence that ESG and tax avoidance are positively correlated, and this correlation is observed only in well-governed firms. Specifically, for such firms, the social and governance dimensions correlate positively with tax avoidance, whereas the environmental dimension does not. For firms with weak governance, none of the three components correlates with tax avoidance.

This study contributes to existing literature in several ways. First, given the growing discussion on corporate tax strategies in the context of ESG, this study sheds light on the relationship between ESG and tax avoidance. It provides empirical evidence that ESG and tax avoidance are not conflicting but rather compatible approaches that can be influenced by unobserved firm-specific factors, such as the goal of increasing firm value. Second, it clarifies the nuanced relationship between ESG and tax avoidance by considering the role of corporate governance. The finding that tax avoidance is positively related to ESG only in well-governed firms indicates that it is an ESG-compatible strategy when effective governance mechanisms can reduce the risk of tax avoidance. Third, this study finds that the three ESG dimensions have different correlations with tax avoidance. Social and governance activities are linked to tax avoidance activities. In contrast, environmental activities are not significantly associated with tax avoidance. These differences support the need for researchers to distinguish and analyze the three dimensions of ESG research. Finally, the evidence from this study can help various stakeholders, including managers, investors, and standard setters, better understand corporate tax avoidance practices and ESG activities in the context of responsible business behavior and comprehend the role of corporate governance in the relationship between the two strategies.

Footnotes

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This work was supported by the Gachon University Research Fund of 2023 [GCU-202303930001].

ORCID iD

Hyun-Ah Lee

Data Availability Statement

The datasets generated during and/or analyzed during the current study are available from the corresponding author on reasonable request.

References

Abdelfattah

Aboud

(2020). Tax avoidance, corporate governance, and corporate social responsibility: The case of the Egyptian capital market. Journal of International Accounting, Auditing and Taxation, 38, 100304. https://doi.org/10.1016/j.intaccaudtax.2020.100304

Alessa

Akparep

J. Y.

Sulemana

Agyemang

A. O.

(2024). Does stakeholder pressure influence firms environmental, social and governance (ESG) disclosure? Evidence from Ghana. Cogent Business & Management, 11(1), 2303790. https://doi.org/10.1080/23311975.2024.2303790

Avi-Yonah

R. S.

(2008). Corporate social responsibility and strategic tax behavior. In Schön

(Ed.), Tax and corporate governance (pp. 183–198). Springer. http://doi.org/10.1007/978-3-540-77276-7_13

Bayar

Huseynov

Sardarli

(2018). Corporate governance, Tax avoidance, and financial constraints. Financial Management, 47(3), 651–677. http://doi.org/10.1111/fima.12208

Bloomberg. (2021). Governance Scores: analyze governance risk and opportunity. A Bloomberg professional services offering. Bloomberg.com/esg. https://data.bloomberglp.com/professional/sites/10/Governance-Scores-Fact-sheet.pdf

Chamley

(1986). Optimal taxation of capital income in general equilibrium with infinite lives. Econometrica: Journal of the Econometric Society, 54(3), 607–622. https://doi.org/10.1016/0047-2727(85)90020-9

Christensen

Murphy

(2004). The social irresponsibility of corporate tax avoidance: Taking CSR to the bottom line. Development, 47(3), 37–44. http://doi.org/10.1057/palgrave.development.1100066

Davis

A. K.

Guenther

D. A.

Krull

L. K.

Williams

B. M.

(2016). Do socially responsible firms pay more taxes? Accounting Review, 91(1), 47–68. https://doi.org/10.2308/accr-51224

Desai

M. A.

Dharmapala

(2006). Corporate tax avoidance and high-powered incentives. Journal of Financial Economics, 79(1), 145–179. https://doi.org/10.1016/j.jfineco.2005.02.002

10.

Desai

M. A.

Dharmapala

(2009). Corporate tax avoidance and firm value. Review of Economics and Statistics, 91(3), 537–546. https://doi.org/10.1162/rest.91.3.537

11.

Djankov

Ganser

McLiesh

Ramalho

Shleifer

(2010). The effect of corporate taxes on investment and entrepreneurship. American Economic Journal: Macroeconomics, 2(3), 31–64. https://doi.org/10.1257/mac.2.3.31

12.

Dyreng

S. D.

Hanlon

Maydew

E. L.

(2008). Long-run corporate tax avoidance. The Accounting Review, 83(1), 61–82. https://doi.org/10.2308/accr.2008.83.1.61

13.

Erle

(2008). Tax risk management and board responsibility. In Schön

(Ed.), Tax and corporate governance (pp. 205–220). Springer. http://doi.org/10.1007/978-3-540-77276-7_15

14.

Fan

Zivin

J. S. G.

Kou

Liu

Wang

(2019). Going green in China: Firms’ responses to stricter environmental regulations (Working Paper No. 26540). National Bureau of Economic Research. https://doi.org/10.3386/w26540

15.

Feng

Zhou

(2020). Convergence analysis of environmental efficiency from the perspective of environmental regulation: Evidence from China. Technological and Economic Development of Economy, 26(5), 1074–1097. https://doi.org/10.3846/tede.2020.13233

16.

Frank

M. M.

Lynch

L. J.

Rego

S. O.

(2009). Tax reporting aggressiveness and its relation to aggressive financial reporting. Accounting Review, 84(2), 467–496. https://doi.org/10.2308/accr.2009.84.2.467

17.

Freeman

R. E.

(1984). Strategic management: A stakeholder approach. Pitman.

18.

Freedman

(2003). Small business taxation: Policy issues and the UK. In Freedman

(Ed.), Taxing small business: Developing good tax policies (pp. 13–43). Springer.

19.

Friese

Link

Mayer

(2008). Taxation and corporate governance—The state of the art In Schön

(Ed.), Tax and corporate governance (pp. 357–425). Springer. http://doi.org/10.1007/978-3-540-77276-7_25

20.

Geng

Liu

Chen

(2021). Environmental regulation and corporate tax avoidance: A quasi-natural experiment based on the eleventh Five-Year Plan in China. Energy Economics, 99, 105312. https://doi.org/10.1016/j.eneco.2021.105312

21.

Gillan

S. L.

Koch

Starks

L. T.

(2021). Firms and social responsibility: A review of ESG and CSR research in corporate finance. Journal of Corporate Finance, 66, 101889. https://doi.org/10.1016/j.jcorpfin.2021.101889

22.

Greene

W. H.

(2008). Econometric analysis (6th ed.). Pearson Prentice Hall.

23.

Gulzar

M. A.

Cherian

Sial

M. S.

Badulescu

Thu

P. A.

Badulescu

Khuong

N. V.

(2018). Does corporate social responsibility influence corporate tax avoidance of Chinese listed companies? Sustainability, 10(12), 4549. https://doi.org/10.3390/su10124549

24.

Hai

Fang

(2022). Does business group’s conscious of social responsibility enhance its investment efficiency? Evidence from ESG disclosure of China’s listed companies. Sustainability, 14(8), 4817. https://doi.org/10.3390/su14084817

25.

Hanlon

Slemrod

(2009). What does tax aggressiveness signal? Evidence from stock price reactions to news about tax shelter involvement. Journal of Public Economics, 93(1–2), 126–141. https://doi.org/10.1016/j.jpubeco.2008.09.004

26.

Huber

B. M.

Comstock

(2017). ESG reports and rating: What they are, why they matter. Harvard Law School Forum on Corporate Governance. https://corpgov.law.harvard.edu/2017/07/27/esg-reports-and-ratings-what-they-are-why-they-matter/

27.

Huseynov

Klamm

B. K.

(2012). Tax avoidance, tax management and corporate social responsibility. Journal of Corporate Finance, 18(4), 804–827. https://doi.org/10.1016/j.jcorpfin.2012.06.00

28.

Jiang

(2024). Does ESG performance affect corporate tax avoidance? Evidence from China. Finance Research Letters, 61, 105056. https://doi.org/10.1016/j.frl.2024.105056

29.

Judd

K. L.

(1985). Redistributive taxation in a simple perfect foresight model. Journal of Public Economics, 28(1), 59–83. https://doi.org/10.1016/0047-2727(85)90020-9

30.

Khanchel

Lassoued

(2023). Are small waves fondle and big waves overturn? Market reaction and corporate governance during four COVID-19 waves. Review of Behavioral Finance, 16(2), 300–321. https://doi.org/10.1108/RBF-02-2023-0054

31.

Kim

J. B.

Zhang

(2011). Corporate tax avoidance and stock price crash risk: Firm-level analysis. Journal of Financial Economics, 100(3), 639–662. https://doi.org/10.1016/j.jfineco.2010.07.007

32.

Landolf

(2006). Tax and corporate responsibility. International Tax Review, 29, 6–9.

33.

Lanis

Richardson

(2011). The effect of board of director composition on corporate tax aggressiveness. Journal of Accounting and Public Policy, 30(1), 50–70. https://doi.org/10.1016/j.jaccpubpol.2010.09.003

34.

Lanis

Richardson

(2012). Corporate social responsibility and tax aggressiveness: An empirical analysis. Journal of Accounting and Public Policy, 31(1), 86–108. https://doi.org/10.1016/j.jaccpubpol.2011.10.006

35.

Liu

Lee

H. A.

(2023). A study on the relationship between ESG activities and earnings management of Chinese listed companies. Global Business Administration Review, 20(2), 65–90. https://doi.org/10.38115/asgba.2023.20.2.65

36.

Love

D. A.

Smith

P. A.

(2010). Does health affect portfolio choice? Health Economics, 19(12), 1441–1460. https://doi.org/10.1002/hec.1562

37.

Minutolo

M. C.

Kristjanpoller

W. D.

Stakeley

(2019). Exploring environmental, social, and governance disclosure effects on the S&P 500 financial performance. Business Strategy and the Environment, 28(6), 1083–1095. https://doi.org/10.1002/bse.2303

38.

Morris

Visser

(2022). Tax is a crucial part of the ESG conversation. Tax research & insights. PWC. https://www.pwc.com/gx/en/services/tax/publications/tax-is-a-crucial-part-of-esg-reporting.html

39.

Ohrn

(2018). The effect of corporate taxation on investment and financial policy: Evidence from the DPAD. American Economic Journal: Economic Policy, 10(2), 272–301. http://doi.org/10.1257/pol.20150378

40.

Salah

Achour

Boussaada

(2023). Impact of corporate social responsibility on tax avoidance: Evidence from French companies. American Journal of Finance and Accounting, 7(3–4), 235–257. https://doi.org/10.1504/AJFA.2023.136698

41.

Schön

(Ed.). (2008). Tax and corporate governance (Vol. 3). Springer Science & Business Media. http://doi.org/10.1007/978-3-540-77276-7

42.

Sikka

(2012). Smoke and mirrors: Corporate social responsibility and tax avoidance. In Haynes

Murray

Dillard

(Eds.), Corporate social responsibility (pp. 53–84). Routledge. https://repository.essex.ac.uk/8103/1/WP2010-5%20PS%20CSR%20and%20Tax%20Avoidance%20Revised%20April%202010.pdf

43.

Slemrod

(2004). The economics of corporate tax selfishness. National Tax Journal, 57(4), 877–899. https://doi.org/10.17310/ntj.2004.4.06

44.

Sturgis

(2022). ESG and taxation: A necessary part of a company’s strategic objectives. Tax Insider Newsletter. The Tax Advisor. https://www.thetaxadviser.com/newsletters/2022/oct/esg-and-taxation-necessary-part-of-companys-strategic-objectives.html

45.

Wilson

R. J.

(2009). An examination of corporate tax shelter participants. Accounting Review, 84(3), 969–999. https://doi.org/10.2308/accr.2009.84.3.969

46.

Yoon

Lee

J. H.

Cho

J. H.

(2021). The effect of ESG performance on tax avoidance—Evidence from Korea. Sustainability, 13(12), 6729. https://doi.org/10.3390/su13126729

47.

Zhang

Luo

Liu

(2022). Is government regulation a push for corporate environmental performance? Evidence from China. Economic Analysis and Policy, 74, 105–121. https://doi.org/10.1016/j.eap.2022.01.018

Exploring the Relationship Between Environmental,Social,and Governance and Tax Avoidance Strategies

Abstract

Plain Language Summary

Keywords

Introduction

Literature Review and Hypotheses

Research Design and Empirical Methodology

Data

Measurement of Variables

Tax Avoidance

ESG

Corporate Governance

Research Model

Results

Descriptive Statistics

Correlation Analysis

Regression Analysis

Robustness Test

Discussion

Implications and Limitations

Conclusion

Footnotes

Declaration of Conflicting Interests

Funding

ORCID iD

Data Availability Statement

References