The Impact of CSR on Tax Avoidance: The Moderating Role of Political Connections

Abstract

This paper investigates the association between corporate social responsibility (CSR) and tax avoidance (TA), with a particular focus on how political connections influence this relationship. The study examines non-financial companies listed on the Egyptian Exchange (EGX) over the period from 2017 to 2022, encompassing a final sample of 70 firms and 420 firm-year observations. Pooled Ordinary Least Squares (OLS) and fixed-effects regression methods are utilized for statistical analysis. The findings reveal a significant positive correlation between CSR activities and a higher degree of TA, suggesting that companies involved in CSR are more likely to engage in TA. Moreover, political connections are shown to have a moderating effect, further strengthening this relationship. To the authors’ knowledge, this research is one of the first attempts to explore the moderating influence of political connections on the CSR-TA relationship in an emerging market context. By doing so, it extends the debate in the literature regarding the negative role played by political connections in increasing TA in developing markets. Previous studies primarily focused on the direct link between CSR and TA, but this study sheds light on the nuanced interaction between these factors when political ties are considered.

1. Introduction

Corporate social responsibility (CSR) has become a significant area of academic focus worldwide, especially among businesses that prioritize transparent reporting of their CSR initiatives. The central aim in analyzing CSR practices often lies in evaluating how effectively these companies uphold their commitment to open communication about both financial and non-financial elements of their activities [1]. Past research in less-developed countries (LDCs) has linked CSR activities to a range of factors, including corporate legitimacy, earnings quality, cash reserves, earnings management, firm value, and overall performance [2,3,4,5,6,7,8,9]. Other studies highlight the role of CSR in fostering employee creativity [10], addressing social and political challenges [11], and influencing operational cost strategies. Specifically, CSR disclosure can sometimes serve as a rhetorical tool to obscure cost-cutting measures [12], which, alongside its ties to firm opacity [13,14,15,16], raises questions about CSR as a strategic choice for securing tax incentives or avoiding tax responsibilities [17,18,19].

In essence, the study of CSR involves exploring how these initiatives can reinforce ethical corporate behavior and sustainable practices aimed at community development [20,21]. However, CSR disclosure can be complex; managers might sometimes leverage CSR practices to mask a lack of financial transparency, possibly using it to manage stakeholder perceptions and divert attention from less favorable financial reports. This dual perspective emphasizes CSR’s multifaceted role, balancing corporate ethics with practical organizational strategies [8,22]. For example, Li et al. [23] reveal that charitable activities contribute to the long-term value of firms in China by improving their public image and building political connections; this social capital gives firms insurance against nugatory procedures, alleviates social pressure for the executives, and provides a moral screen permitting executives to engage in questionable behaviors or cover up illegal acts, such as tax avoidance practices.

Subsequently, this study seeks to deepen the understanding through extending the literature of the links between CSR, political connections, and TA [7,16,17,18,20,24,25,26,27]. We intend to do this by addressing an overlooked aspect: the moderating role of political connections in the CSR-TA relationship. Past research has predominantly used stakeholder theory and legitimacy theory to examine the CSR-TA relationship [20,26,28,29,30], as well as to explore the connection between political influence and TA [30,31,32,33,34,35]. However, few studies have specifically analyzed how political connections might alter the relationship between CSR activities and TA strategies. Despite extensive reviews of the existing literature, only one prior study has examined the moderating influence of political connections on CSR and TA [30]. This research, set within Bangladesh’s banking sector, reported a negative moderating effect of political ties on CSR’s relationship with TA. The general lack of research on this topic highlights a critical gap, suggesting the need for further investigation into how political connection may affect corporate tax strategies in firms that emphasize CSR, especially in emerging markets.

The Egyptian market offers a compelling context for studying the connections between CSR, TA, and the role of political ties due to its unique economic and regulatory characteristics. Since the 1990s, Egypt has shifted its economic system from a state-controlled model to one favoring open-market policies, privatization, and economic reforms [36]. This transition has introduced notable regulatory and structural changes, which can significantly influence how CSR practices interact with tax strategies and how political connections might impact this relationship [37,38,39,40,41].

During this period, Egypt also took steps to improve financial reporting standards, enhancing transparency in the stock exchange by adopting improved accounting practices [37,39]. Despite these advancements, the Egyptian market still faces challenges compared to more developed markets like the U.S. and the U.K. Ongoing government intervention and certain regulatory limitations have hindered the full alignment with international standards, creating a complex environment for CSR and tax practices to evolve. These conditions underscore Egypt as a valuable area for examining CSR, political connections, and TA under transitional regulatory pressures [11,42,43].

Egypt’s regulatory environment is often portrayed in research as having relatively low enforcement strength, with accounting and disclosure practices shaped substantially by political influences [39,43,44]. This environment can contribute to limited transparency in corporate disclosures, governance, and the widespread use of TA strategies, a trend observed across several emerging economies [22,39,45,46]. In light of this, Egypt provides a valuable context for examining CSR and TA within a regulatory framework distinct from mature markets. This study explores whether recent developments in Egypt’s economic, governance, and regulatory systems have advanced tax compliance behaviors.

To address this, two research questions are posed: (1) What is the relationship between CSR practices and TA in the Egyptian market? (2) Do political connections moderate this relationship? The findings indicate a significant positive relationship between CSR engagement and higher levels of TA, suggesting that companies involved in CSR are also likely to engage in tax-saving strategies. Additionally, political connections appear to reinforce this link, as they intensify the effect of CSR activities on TA. Therefore, the study highlights the impact of political affiliations on corporate tax strategies within Egypt’s evolving regulatory landscape, offering insights into one of the factors that shape tax compliance.

This research provides multiple contributions to the existing literature. First, previous studies on CSR and TA are predominantly based in developed economies, where business environments differ significantly from those of developing countries, such as Egypt. As one of Africa’s leading emerging markets and a key financial center in the MENA region [37,39], Egypt’s distinctive political and economic backdrop, especially post-2011 and 2013 revolutions, makes it an important case study with broad relevance across the MENA region [43].

Second, this research uniquely investigates how political connections influence the relationship between CSR practices and TA, filling a critical gap in studies of emerging markets. Third, the study offers practical recommendations for investors, shareholders, and policymakers in emerging economies that share structural and socio-economic similarities with Egypt, helping to guide CSR policies and tax regulation. Finally, the findings of this study challenge prior research, which largely emphasized stewardship and stakeholder theories without considering the complex moderating role of political ties.

The rest of this paper is organized as follows: Section 2 presents the theoretical framework; Section 3 reviews the relevant literature and outlines hypotheses; Section 4 discusses the methodology; Section 5 details the empirical results and analysis; and Section 6 concludes with limitations and future research directions.

2. Theoretical Framework

Research examining the links between CSR, TA, and corporate governance which reduce political connections highlights several theoretical frameworks, including stakeholder theory, resource dependency theory, agency theory, legitimacy theory, and risk management theory, each offering distinct perspectives based on varied contexts [30]. Among these, stakeholder theory holds particular significance; it posits that a company’s success is contingent upon addressing the interests and welfare of various stakeholders not only for legal compliance and profitability but also for ethical responsibility [20]. This approach underscores that aligning business operations with the needs and expectations of these groups fosters legitimacy and strengthens long-term organizational resilience [30].

A company’s CSR practices aim to create value for its stakeholders, reflecting the principles of stakeholder theory, which encourages firms to prioritize the needs of all parties involved or impacted by their actions [47,48]. This perspective implies that firms committed to CSR would prioritize transparent and ethical business practices, potentially reducing their engagement in TA activities. Studies have suggested that firms with robust CSR initiatives are less likely to partake in aggressive tax strategies compared to those with lower CSR involvement, as these actions align more closely with stakeholder expectations for responsibility and ethical conduct [24,49,50,51,52]. Furthermore, stakeholder influence—including groups like employees, communities, and suppliers—can shape the relationship between CSR and tax practices. For instance, having women in top management, as part of the stakeholder framework, may impact both CSR efforts and approaches to taxation [20].

In the same context, from the perspective of agency theory, managers (agents) are expected to give priority to their own interests at the expense of shareholders (principals). This situation is often more apparent in developing countries that suffer from weak governance mechanisms and ineffective regulatory bodies. Therefore, the presence of members with political connections on the board of directors enhance opportunistic behavior of management, as managers tend to increase their CSR disclosure to cover up the adoption of tax avoidance strategies or to obtain the expected benefits of CSR reporting [44].

Another theory relevant to this research is resource dependency theory, particularly concerning the influence of women in corporate decision-making related to tax strategies. Research by ref. [53] suggests that female directors tend to engage less in tax planning activities. Gender has been shown to significantly impact corporate decision-making processes [54]. Resource dependency theory posits that firms rely on external resources provided by stakeholders, and that maintaining legitimacy is crucial to sustaining their support [55]. This theory proposes that firms structure board membership to ensure access to valuable external resources, and that increased diversity among board members enables the company to better understand and respond to environmental demands [56,57]. A diverse board composition, therefore, has the potential to strengthen corporate effectiveness through a broader resource pool, ultimately impacting corporate performance, including CSR practices. Stakeholder theory, resource dependency theory, and agency theory all indicate that a diverse board, with varied perspectives and backgrounds, can influence the relationship between CSR initiatives and corporate TA strategies [48,55].

3. Literature Review and Hypotheses Development

This section explains the relationship between political connections, CSR, and TA. Three hypotheses were explored in this investigation. The first discusses how CSR and TA are related. The nexus between political connections and TA is discussed in the second. The third discusses the moderating effect of political connections on the CSR-TA relationship.

3.1. CSR and TA

In order to carry out their activities, companies must fulfill their commitments to a variety of stakeholders with differing interests [58]. All of them concur, therefore, that businesses must enhance societal welfare [58]. Companies can support social welfare by increasing their tax payments to the government [59]. In turn, the governments employ taxes to raise the standard of living in society. Consequently, businesses pay taxes if they believe that the government utilizes them to help the entire community rather than just a select number of citizens [60]. Conversely, companies may decide not to pay taxes if they believe the money they provide to the government will not be used appropriately for community welfare. In this case, they might opt not to pay taxes and use the money they save for CSR initiatives [59].

Therefore, the literature has given a lot of attention to the topics of CSR and TA. Past studies on the relationship between CSR and TA have produced mixed results. CSR is the duty of corporations to take society’s needs into account. Businesses should have a procedure for integrating social, environmental, ethical, human rights, and consumer concerns in order to meet their CSR requirements [25]. The term “tax avoidance” describes a business’s attempt to minimize taxes paid to taxing authorities using a range of strategies, some of which are entirely lawful because they transfer money from the government to corporate owners through value-maximizing operations [61].

The concordance of tax payments with CSR contributions has been shown by several research. Transparency, honesty, and a solid reputation are thought to be prerequisites for more socially conscious businesses [50]. The stakeholder theory suggests that businesses should not engage in this kind of behavior since it might be detrimental to both the stakeholders and society as a whole. In a similar vein, companies should act appropriately, according to corporate culture theory. These theories anticipate increased taxation for companies that engage in greater CSR. According to the corporate culture theory, a number of research showed a negative correlation between CSR performance and TA [62]. For instance, ref. [51] discovered that companies that prioritize corporate social responsibility substantially reduced TA rates. Similarly, companies that report more CSR information have a lower likelihood of engaging in tax-aggressive activities [50].

Furthermore, ref. [52] found that tax-aggressive behavior, such as higher book-tax disparities, low effective tax rate (ETRs), and significant tax-sheltering activities, is more prevalent among companies with low CSR ratings in the U.S. Similarly, ref. [63] showed that enterprises’ profitability and CSR expenditure are heavily dependent on their payment of taxes, which is their contribution to public goods, suggesting that the involvement of businesses in CSR initiatives is essential to motivating them to lessen tax evasion. A evaluation of 47 quantitative empirical studies on the relationship between CSR and tax evasion was done by ref. [28]. They stated that the majority of research revealed a negative correlation between CSR and TA. Moreover, ref. [24] examines the connection between irresponsible CSR practices and aggressive TA. Firms that engage in excessively irresponsible CSR exhibit a greater inclination towards TA. Ref. [52] demonstrates similar results, which suggests that companies that exhibit social irresponsibility also tend to be more aggressive with regard to taxes and have larger unrecognized tax benefits than other firms.

However, several research found a positive relationship between CSR and TA [25,26,62,64], which according to the risk management perspective of CSR, companies engage in CSR to cover up their aggressive tax avoiding and other socially irresponsible behavior [64]. This implies that CSR disclosure may increase managerial risk-taking incentives, which leads to more tax avoidance practices. Similar findings were made by ref. [26] using data from French non-financial companies, who found that companies who participate in CSR are more likely to shield themselves from the damaging effects of active TA methods. Additionally, ref. [49] examined the CSR practices of S&P 500 companies between 2000 and 2008 and discovered a positive correlation between tax aggressiveness and social responsibility. Based on a sample of Canadian firms, ref. [65] demonstrate that a company’s tax conduct is not always consistent with its CSR. Furthermore, according to ref. [66], the attitudes of firms about CSR have a significant bearing on how much and how ready they are to lower their tax liabilities.

Similarly, ref. [67] demonstrates that CSR companies have smaller ETRs and larger book-tax discrepancies. This suggests that TA by CSR companies is more active. According to these results, companies take part in CSR initiatives as a risk-reduction strategy. Furthermore, ref. [25] discovered a positive correlation between TA and CSR. Companies with headquarters in jurisdictions with low financial-tax reporting conformity are more likely to participate in CSR as a protective measure against the possible detrimental effects of aggressive TA practices than companies with headquarters in jurisdictions with high financial-tax reporting conformity. TA methods are expected to result in a rise in CSR disclosures [68]. The legitimacy theory holds that businesses should disclose more information about the environment, society, and governance in order to allay community fears about reduced tax revenues and establish legitimacy. The following hypothesis is put out in light of the theories, empirical literature, and research environment:

H1. 

There is an association between CSR and TA.

3.2. Political Connections and TA

There are a number of reasons why managers would want to get into politics, including the desire to be free from regulations and to be under less observation and oversight [69]. The shielding corporate political ties provide against legal scrutiny and potential lawsuit threats is one of the key benefits of doing so. Ref. [70], in their investigation of the relationship between TA and risk, pointed out that although lowering taxes might boost a business’s profitability, managers strive to do so without subjecting the enterprise to the dangers associated with TA strategies. Ref. [71] states that in the event that tax manipulation is discovered, firms may be subject to heavy fines from tax authorities as well as securities regulators, which might negatively impact the company’s total value. Consequently, companies which can lower their tax liabilities without incurring penalties stand to gain the most from TA, especially those headed by CEOs with political connections [71]. Additionally, the market frequently puts less pressure on companies with political links to be transparent [72]. According to ref. [70], TA may lead to greater opacity, which makes it more difficult for regulators to completely comprehend the financial operations of a company. Therefore, political connections can help minimize the market and political repercussions associated with financial opacity or aggressive tax tactics [73].

The second benefit of having political connections for your firm is the capacity to learn about impending changes in tax laws and enforcement, which makes it possible for these firms to employ complex tax strategies to benefit from time-series fluctuations in tax rules or enforcement [74]. Ref. [73] makes the case that politically connected firms frequently have access to preferred information, which helps them to predict when aggressive tax planning would likely be accepted.

In line with the argument that firms may leverage political connections to secure or maintain lower tax rates, a number of empirical research have examined the influence of political connections on TA. Researchers like refs. [33,72,74,75] have found a positive relationship between company political ties and TA. Furthermore, previous research showed that political connections enhance firms TA [76,77]. However, several studies have discovered a negative relationship between TA and political connections [32,78]. On the other hand, research by refs. [79,80] indicates that political connections do not significantly affect firms TA. These conflicting findings draw attention to contradictions in the literature and point to the need for more research. As a result, we offer the following hypotheses:

H2. 

There is a relationship between political connections and TA.

3.3. Political Connection Moderating Role on the CSR–TA Relationship

Ref. [72] offers proof that businesses with political ties are more likely to avoid taxes. There are several explanations for this behavior, such as fewer overt risks, improved awareness of impending tax law changes, reduced political costs, and closer relationships with legislators and policymakers.

Also, it is often acknowledged that board member political affiliations pose a substantial risk factor for TA activities [31,33,81]. Politically linked boards of companies can get loan financing without having to provide creditors access to financial information, which alleviates tax managers’ worries about the informational repercussions from aggressive tax methods [72,82]. Ref. [83] state that this political relationship creates an environment that is favorable to hazardous and aggressive tax planning by acting as a type of insurance during catastrophic occurrences. Additionally, political connections make it more likely that future tax regulations will be loosened, which promotes even more aggressive tax planning [84]. In their investigation of whether political links on business boards in China limit the power of tax authorities to stop TA, ref. [31] discovered that in economies that are subject to political control, these linkages can seriously impede the effective implementation of tax compliance laws. Similarly, ref. [35] looked at how political ties affected local companies’ tax planning in China and found that companies only stopped using TA strategies once new political leaders were appointed.

Furthermore, political governance frameworks are frequently the cause of high levels of corruption, poor accountability, and limited transparency in emerging economies [74,85,86]. Ref. [87] states that firms that have political links may deliberately use CSR programs to curry favor with the government and take use of their proximity for financial advantage. Thus, CSR incentives change from being normative to being strategic. The conflict of interest that results between the board and shareholders when politically connected directors prioritize personal gains by influencing board decisions about CSR and tax payments exacerbates agency problems [88]. Moreover, firms with political ties may make use of their standing to increase societal approval, which would lessen the belief that CSR is a legitimate technique. Therefore, these companies are more likely to use TA strategies. In light of this, the following hypothesis is put forth:

H3. 

Political connections moderate the CSR–TA association.

4. Research Design

4.1. Data Description

The initial sample selected for the current research was based on the top companies listed on the Egyptian Exchange, representing EGX 100. The insurance and financial companies were excluded because of their special characteristics in accounting and financial reporting [42,43,89]. The final sample consists of 70 companies, with 420 observations representing 12 industries over 2017–2022, as described in detail in

Table 1. Sample structure.

Industry Sector	No. of Firms	No. of Observations	%
Paper and Packaging	4	24	5.8
Basic Resources	5	30	7
Building Materials	6	36	8.6
Travel and Leisure	6	36	8.6
Real Estate	9	54	13
Contracting and Construction Engineering	8	48	11.5
Textile and Durables	5	30	7
Communication Services, Media, and IT	4	24	5.8
Trade and Distributors	4	24	5.8
Healthcare and Pharmaceuticals	4	24	5.8
Industrial Goods, Services, and Automobiles	5	30	7
Food, Beverages, and Tobacco	10	60	14.1
Total	70	420	100

. Data in this research were collected from annual reports, company websites, Bloomberg Asharq, and Mubasher Egypt.

4.2. Variables’ Measurement

4.2.1. Corporate Social Responsibility (CSR)

To measure CSR, we depended on the Egyptian Social, Environmental, and Governance Index (S&P/EGX ESG index methodology) issued by the Egyptian institute of Directors in cooperation with Standard and Poor’s and the Egyptian Stock Exchange (ESE) to measure the level of disclosure of social responsibility. It has been used by many studies conducted in the Egyptian environment as it is more stable and related to the needs of society [25].

The Egyptian index consists of two parts: first, governance mechanisms, which include ownership structure and shareholders’ rights, financial and operational information, information about the board of directors and senior management, and business ethics and corporate responsibility, and second, social and environmental responsibility, which include environmental protection, information about employees, community development, customer satisfaction, and product quality. This study relies on the binary score applied by Standard and Poor’s to prepare the index, where the element takes the value (1) if the company discloses it, and (0) if it is not disclosed. Then, the percentage of the company’s disclosure level is calculated, which is the number of disclosure points for each company divided by the total number of index elements (the maximum disclosure limit) [90].

$$\mathbf{CSR} \mathbf{reporting} \mathbf{Index}={\sum }_{\mathit{t}=1}^{\mathit{N}\mathit{j}}{\displaystyle \frac{\mathit{X}\mathit{i}\mathit{j}}{\mathit{N}\mathit{j}}}$$

Whereas:

Nj: Total number of social responsibility index items (32) item.

Xij: Number of social responsibility index items disclosed by the firm j during the period t.

4.2.2. Tax Avoidance (TA)

Given of the clandestine nature of tax returns and other revenues submitted to the internal revenue service, measuring quantified TA has become a major challenge in accounting studies [91]. In accounting research, TA is commonly quantified using the current effective tax rate (CUR_ETR), calculated by dividing the current income tax expense by pre-tax earnings. This measure excludes deferred income tax expenses since deferred taxes arise from timing discrepancies between tax and financial reporting. Using CUR_ETR allows for a clearer picture of the immediate tax liabilities associated with a company’s operational earnings, avoiding distortions from deferrals. [20]. In order to simplify interpreting results we multiplied this proxy with a negative one and excluded all sample firms with negative returns on equity.

4.2.3. The Moderating Variable: Political Connections (POLCONs)

POLCONs are used as moderating variables in our study. To measure this variable we assessed the number of political persons on the board or their kins, or any formal characters that have served in a government position such as prime ministers, ministers, members of the parliament, or ambassadors, or work at political party, or have political connections through their family ties [31,86].

4.2.4. Control Variables

Our research controls for a number of firm characteristics that may have an impact on TA practices proportionate with previous researches [50,88,92,93,94]. Firm size (SIZE) accounts for the influence of a firm’s scale on corporate tax avoidance. Following the argument by Rego [95] that larger firms may achieve economies of scale through tax planning, leveraging resources and incentives to lower group tax, we included firm size as a control variable in our regression model [96,97]. Thus, firm size was represented by the logarithm of total assets. Leverage was factored in, calculated as the ratio of total debt to total assets. Return on assets (ROA) was included, defined by dividing pre-tax income by total assets, and cash holdings were measured as the ratio of cash and cash equivalents to total assets. Additionally, industry and year effects were incorporated into the primary models to control for sector-specific and temporal variations.

4.3. The Study Models

To test the proposed relationships in line with the research hypotheses, we conducted analyses using Pooled Ordinary Least Squares (OLS) and fixed-effect regression models, applying robust standard errors to address potential heteroscedasticity and autocorrelation issues. The Hausman test results indicated that the fixed effects model provided a better fit than the random effects estimator, justifying its use in this study. Therefore, three regression models were generated to formulate the three hypotheses as follows:

Model 1:

TAit = β0 + β1 CSRit + β2 SIZEit +β3 LEVit + β4 ROAit + β5 Cashit + βt + βind + εit

(1)

Model 2:

TAit = β0 + β1 POLCONsit + β2 SIZEit +β3 LEVit + β4 ROAit + β5 Cashit + βt + βind + εit

(2)

Model 3:

TAit = β0 + β1 CSRit + β2 POLCONsit +β3 CSR*POLCONsit + β4 SIZEit + β5 LEVit + β6 ROAit + β7 Cashit+ βt + βind + εit

(3)

where CSR is corporate social responsibility; TA is tax avoidance; POLCONs is political connections; SIZE is firm size; LEV is firm leverage; ROA is return on assets; Cash is cash holdings and cash equivalents; βt represents the time fixed effects; and βind represents the industry-fixed effects.

Table 2. Variable definitions.

Variable	Measurement
Dependent variable
Tax Avoidance (TA)	(Total Income Tax Expenses—Deferred Tax Expenses)/Pretax Income. This proxy is multiplied with a negative one.
Independent variable
Corporate Social Responsibility (CSR)	ESG score based on the ESG ratings generated by Standard and Poor’s (S&P) in collaboration with the Egyptian Stock Exchange (ESE), as described above.
Control variables
Return on Assets (ROA)	Measured as pretax income divided by total assets
Firm Leverage (LEV)	The ratio of total debt to total assets
Cash Holding (Cash)	Cash and cash equivalents/total assets
Firm Size (SIZE)	The natural logarithm of total assets
Moderating variables
Political Connections (POLCONs)	We use a dummy variable that coded as 1 if the firm has a number of politician directors or their kin on the board, or have formally served in a government position (i.e., prime ministers, ministers, members of the parliament, ambassadors), and 0 otherwise.

provides definitions for all the variables.

5. Main Results

5.1. Descriptive Statistics and Pairwise Correlation

Table 3. Descriptive statistics.

Variables	N	Minimum	Maximum	Mean	Median	Std. Deviation
TA	420	−1.728	−0.028	−0.201	−0.204	0.236
CSR	420	0.435	0.768	0.568	0.537	0.115
POLCONs	420	0.000	1.000	0.084	0.000	0.243
SIZE	420	7.492	14.943	9.385	9.014	1.596
ROA	420	−0.293	0.408	0.075	0.056	0.102
LEV	420	0.000	0.874	0.417	0.419	0.213
Cash	420	0.000	0.896	0.098	0.049	0.134

Notes: Definitions for all variables can be found in Table 2.

offers the descriptive statistics for the full sample that involves 70 Egyptian firms. Looking at the outcomes of the table. We can notice that the average (median) of CSR disclosure is 0.568 (0.537) and ranges between 0.435 and 0.768. It is noteworthy that this result is close to the result of the study of ref. [25], who found that the mean of CSR disclosure in Egyptian companies is 0.539. In addition, the TA mean is 20.1%, which is lower than the legal tax ratio in Egypt of (22.5%). This means that the firms, on average, have been taxed aggressively by paying or accruing lesser tax than the juristic requirement. Regarding the POLCONs, the average of POLCONs is 0.084 within the range from 0% to 1%. This implies that 8% of the sample firms are politically connected through politicians or their relatives who work at their boards of directors. The log transformation for the firm size (FSIZE) averages 9.385 and ranges between 7.492 and 14.943. The leverage (LEV) averages 0.417 and registers a maximum value of 0.874. On average, the return on assets (ROA) is 7.5%. The average (median) for cash is 0.098 (0.049).

The results of a pairwise correlation matrix among the study variables are exhibited in

Table 4. Correlation.

Variables	(1)	(2)	(3)	(4)	(5)	(6)	(7)	Variance Inflation Factor (VIF)
(1) TA	1.000							----
(2) CSR	0.246 ***	1.000						1.086
(3) POLCONs	0.361 ***	0.179 ***	1.000					1.473
(4) SIZE	0.417 ***	0.172 ***	0.522 ***	1.000				1.468
(5) ROA	−0.029	0.035	0.162 ***	0.002	1.000			1.082
(6) LEV	−0.045	0.060	−0.040	0.156 ***	−0.083 *	1.000		1.067
(7) CASH	0.108 **	0.167 ***	0.023	0.027	0.197 ***	−0.092 *	1.000	1.083
Mean VIF								1.212

Note: asterisks: ***, ** and * correspond to 0.01, 0.05 and 0.10 significant.

. As displayed in the table, TA correlated significantly and positively with both CSR at 0.246, POLCONS at 0.361, firm size at 0.417, and cash holding at 0.108. Moreover, the study found that TA has an insignificant negative correlation with both ROA at −0.029 and LEV at −0.045. In addition, the correlation matrix reveals that none of the coefficients is higher than 0.522 (i.e., between POLCONS and SIZE); this refers to the fact that the research has no multicollinearity problem [98]. Furthermore, the study calculates the VIF to estimate the multicollinearity problem. Table 4 indicates that the highest noticed VIF value among the variables of the study is 1.473 for POLCONs, which is less than conventional cutoff of 10.0.

5.2. Multivariate Outcomes

5.2.1. CSR and TA (Model 1)

Table 5. Association between CSR and TA—Model 1.

Variables	DV: TA—Model 1
	Pooled OLS Regression Model		Fixed Effects Regression Model
	Coef.	Z	Coef.	Z
Constant	−0.628	−8.08 ***	−1.416	−10.91 ***
CSR	0.378	3.91 ***	0.145	2.05 **
SIZE	0.062	9.09 ***	0.162	11.76 ***
ROA	−0.15	−1.33	−0.234	−1.20
LEV	−0.138	−2.64 ***	−0.196	−2.13 **
CASH	0.125	1.52	0.309	2.77 ***
Year (FE)	Included
industry (FE)	Included
Adj. R2	0.225		0.357
F-statistic	24.093 ***		38.343 ***
Hausman test results	Fixed effects
No. observations	420		420

Notes: ***, **, and * denote correlation is significant at 1, 5, and 10 percent. Variable definitions are presented in Table 2.

shows the regression results of Model (1) to examine the influence of CSR on TA. As reported in the table, we can notice that there is a significant and positive correlation between CSR and TA, as the regression coefficient was 0.378 at the level of significant p < 0.01. This outcome demonstrates that firms involved in TA strategies tend to increase its level CSR contribution, (e.g., employee relations, corporate governance, community relations, human rights, environment, and products). Further, in conformity with legitimacy theory, our findings indicate that TA firms attempt to create a positive public and media perception of being socially responsible entities and cover up their illegal practices of paying less tax than is due to the state. Moreover, in the Egyptian context, this finding aligns with the study of ref. [25], who show that corporate TA is positively associated with CSR disclosure. In addition, this finding corresponds to prior research conducted in developed markets such as in France [99], China [71], and the U.S. [100], and are supported by both the agency theory and legitimacy theory. This finding supports the view that engaging in tax avoidance practices is one of the provenances of shareholder wealth maximization, and corporate social responsibility is no longer viewed as a firm’s obligation to society, but rather as a constraint. In contrast, this finding is inconsistent with the outcome of some studies conducted in developed markets, like Australia [50], Spain [101], and Indonesia [102], as well as one study conducted in a developing market [30], who argue that the higher the firm’s CSR performance, the lower the firm’s tax avoidance. In light of the previous discussion, it can be said that the lack of agreement between the results of prior researches regarding this relationship confirms the controversy of this relationship so far, which calls for conducting more research on this relationship. Finally, we have to emphasize and accept the first hypothesis (H1), which suggests that the level of CSR is significantly related to corporate tax avoidance.

5.2.2. Political Connections and TA (Model 2)

Table 6. Association between POLCONs and TA—Model 2.

Variables	DV:TA—Model 2
	Pooled OLS Regression Model		Fixed Effects Regression Model
	Coef.	Z	Coef.	Z
Constant	−0.328	−4.41 ***	−0.851	−4.87 ***
POLCONs	0.179	3.87 ***	0.409	4.33 ***
SIZE	0.05	6.18 ***	0.103	5.32 ***
ROA	−0.229	−1.99 **	−0.169	−0.89
LEV	−0.1	−1.90 *	−0.132	−1.46
CASH	0.19	2.35 **	0.338	3.12 ***
Year (FE)	Included
industry (FE)	Included
Adj. R2	0.225		0.388
F-statistic	24.017 ***		43.812 ***
Hausman test results	Fixed effects
No. observations	420		420

Notes: ***, **, and * denote correlation is significant at 1, 5, and 10 percent. Variable definitions are presented in Table 2.

displays the results of the Pooled OLS regression and fixed effects regression that we conducted to test the correlation between political connections and TA. In both models, POLCONs is significantly positively associated with TA at the 1% level, which is consistent with H2. This finding is in agreement with the results of prior studies conducted in developed markets like ref. [31,74,75], and the study of [30] performed in a developing market, suggesting that increasing political connections through board directors increases the likelihood of firms engaging in more tax avoidance practices. Moreover, this outcome boosts the view that political connections result in interfering policies between the personal and the public dimension of political connections; thus, increasing commission of tax avoidance practices.

Regarding the control variable, TA associated significantly and positively with firm size (Size) at 1% level, in both models. This outcome is consistent with study of [50], who found that small firms would be less aggressive towards their tax policy than big firms. In addition, TA has a significant and positive relationship with cash at 5% and 1% levels, respectively, in both models. In contrast, the study found an insignificant and negative influence of LEV and ROA on TA.

5.2.3. Political Connections Moderate the Relationship Between CSR and TA (Model 3)

Table 7. POLCONS’s moderating role in association between CSR and TA—Model 3.

Variables	DV: TA—Model 3
	Pooled OLS Regression Model		Fixed Effects Regression Model
	Coef.	Z	Coef.	Z
Constant	−0.284	−3.15 ***	−0.86	−4.71 ***
CSR	0.173	1.79 *	0.151	1.87 *
POLCONS	0.083	1.81 *	0.359	2.83 ***
CSR*POLCONS	1.957	5.83 ***	0.793	2.52 **
SIZE	0.034	4.24 ***	0.102	5.18 ***
ROA	−0.236	−2.16 **	−0.173	−0.90
LEV	−0.086	−1.72 *	−0.138	−1.52
CASH	0.175	2.24 **	0.334	3.06 ***
Year (FE)	Included
industry (FE)	Included
Adj. R2	0.305		0.389
F-statistic	25.818 ***		31.232 ***
Hausman test results	Fixed effects
No. observations	420		420

Notes: ***, **, and * denote correlation is significant at 1, 5, and 10 percent. Variable definitions are presented in Table 2.

summaries the results of the Pooled OLS regression and fixed effects regression that we used to test the moderating effect of political connection on the association between CSR and TA. As reported in the table, the interaction term’s (CSR*POLCONS) coefficient is positive and significant at 1% and 5% levels, respectively, in both models, suggesting that the presence of politicians as board directors increases the contribution of firms in CSR activities, and hence, TA practices would be more aggressive due to the existence of political connections on the board. This finding disagrees with the results of study [30], which was conducted in a developing market (Bangladesh), which found that existing political connections as board directors weakens the association between CSR and tax avoidance. In general, our result supports the argument that politically connected firms, through their connections to officials in government agencies, try to create an image of being socially responsible firms to alleviate public concerns regarding their illegal practices such as TA, which they consider a risky investment opportunity in order to increase their cash flows [51].

Robustness Analysis: Alternative TA Measures

To assess the robustness of the findings of our fundamental analysis, we re-ran our three models using book-tax differences (BTDs) as another proxy of TA, measured by this equation: (net Income before tax − Taxable income)/Total assets). Previous studies point out that BTDs are correlated significantly with TA [91,100,103]. The results shown in

Table 8. Results of alternate-measurement TA (BTDs).

	Model 1 DV: TA Fixed Effects		Model 2 DV: TA Fixed Effects		Model 3 DV: TA Fixed Effects
	Coef.	Z	Coef.	Z	Coef.	Z
Constant	−1.202	−6.67 ***	−0.19	−0.80	−0.572	−2.57 **
CSR	0.212	2.11 **	----	----	0.153	2.02 **
POLCONS	----	----	0.738	5.74 ***	0.593	2.90 ***
CSR*POLCONS	----	----	----	----	1.751	2.63 ***
Controls	Included
Year (FE)	Included
industry (FE)	Included
Adj. R2	0.408		0.458		0.457
F-statistic	47.587 ***		58.253 ***		41.229 ***
No. observations	420		420		420

Notes: ***, **, and * denote correlation is significant at 1, 5, and 10 percent. Variable definitions are presented in Table 2.

propose that our evidence is durable to the use of this alternative TA measure, and the results of Model 1, Model 2, and Model 3 were aligned with the previous findings, as displayed in Table 5, Table 6 and Table 7.

Robustness Analysis: Tobit Model

To assess the robustness of the results of our hypotheses, we applied Tobit regression as our dependent variable (TA) is non-positive (restricted to the negative side). We utilized the Tobit model because it accounts for the censoring of values at 0. The results on

Table 9. Results of Tobit regression model.

	Model 1 DV: (TA)		Model 2 DV: (TA)		Model 3 DV: (TA)
	Coef.	Z	Coef.	Z	Coef.	Z
Constant	−0.529	−8.14 ***	−0.235	−4.21 ***	−0.269	−3.26 ***
CSR	0.236	3.94 ***	----	----	0.169	1.89 *
POLCONS	----	----	0.181	3.85 ***	0.092	1.87 *
CSR*POLCONS	----	----	----	----	1.869	5.74 ***
SIZE	0.073	9.16 ***	0.036	6.45 ***	0.044	4.11 ***
ROA	−0.138	−1.36	−0.236	−2.30 **	−0.225	−2.11 **
LEV	−0.126	−2.76 **	−0.105	−1.97 *	−0.079	−1.79 *
CASH	0.115	1.43	0.186	2.43 **	0.183	2.32 **
obs.	420		420		420

Notes: ***, **, and * denote correlation is significant at 1, 5, and 10 percent. Variable definitions are presented in Table 2.

support our hypotheses.

Test of Endogeneity: Generalized Method of Moments (GMM)

Our main regression results, as presented in Table 5, Table 6 and Table 7, may be affected by endogeneity and omitted variable bias, as highlighted by Schultz et al. [104] and Wooldridge [105]. To mitigate these concerns, we employed fixed effects model estimators, which help address endogeneity and omitted variable bias to some extent. However, fixed effects models do not account for endogeneity arising from unobserved heterogeneity, such as interactions between unobserved firm characteristics and variables like corporate governance or audit factors [106].

To address this issue, we utilized the Generalized Method of Moments (GMM) estimator proposed by Arellano and Bond [107], which allows us to examine tax avoidance in Egyptian firms while controlling for endogeneity. The system GMM estimator evaluates the equations in levels, using the first difference of each variable as an instrumental variable (IV).

Table 10. Results of two-step system GMM approach.

	Model 1 DV: (TA)		Model 2 DV: (TA)		Model 3 DV: (TA)
	Coef.	Z	Coef.	Z	Coef.	Z
Constant	0.139	0.32	−0.035	−0.11	−0.260	−0.56
TAit-1	0.149	2.35 **	0.163	2.84 **	0.192	2.91 **
CSR	0.110	2.43 **	----	----	0.197	2.88 **
POLCONS	----	----	0.271	2.97 ***	0.245	2.78 **
CSR*POLCONS	----	----	----	----	2.069	2.99 ***
Controls	Included
Year (FE)	Included
industry (FE)	Included
F-statistic	39.63 ***
obs.	420		27.51 ***		21.58 ***
AR1 (p-value)	0.005		420		420
AR2 (p-value)	0.314		0.002		0.007
Sargant test (p-value)	0.455		0.324		0.358

Notes: ***, **, and * denote correlation is significant at 1, 5, and 10 percent. Variable definitions are presented in Table 2.

presents the GMM regression results, analyzing the relationship between CSR and tax avoidance (TA). The coefficients for CSR are positive and significant. Similarly, the interaction between CSR and POLCONS are positively and significantly related to TA.

The diagnostic statistics for the GMM model in Table 10 indicate first-order autocorrelation in the error term, as suggested by the AR1 statistic (p < 0.10). However, the absence of second-order autocorrelation (p > 0.10) confirms the model’s validity [108,109]. Additionally, the Sargan test’s p-value (p > 0.10) supports the exogeneity and validity of the IVs used [110].

Overall, the results consistently reveal a positive relationship between CSR and tax avoidance, providing further support for our hypotheses. Thus, we conclude that our main regression findings are robust to endogeneity concerns, including potential reverse causality.

6. Conclusions, Limitations and Future Research

This study utilizes a sample of 70 non-financial firms listed on the Egyptian Stock Exchange (EGX), encompassing 420 observations over the period 2017 to 2022, to explore the relationship between CSR and TA. In addition, the study examines the moderating role of political connections in the association between a firm’s CSR engagement and its level of TA. Findings reveal a positive, significant relationship between CSR and TA, suggesting that firms engaged in CSR practices tend to exhibit higher levels of TA. This outcome aligns with earlier studies, which observed that CSR activities can serve as a means for firms to obscure their true financial standing from shareholders [25,26,62,64].

Notably, this research appears to be the first to assess the moderating effect of political connections on the CSR-TA relationship in the Egyptian market. Unlike findings from Bangladesh [30], where political connections were found to have a different moderating impact, our study identifies a positive moderating effect, suggesting that the unique institutional and economic context of Egypt influences the interplay between CSR and TA practices. This divergence underscores the importance of considering contextual factors and their impact on corporate behavior regarding tax strategies.

This study contributes meaningful insights, but also has certain limitations. Primarily, it is focused solely on non-financial firms listed on the Egyptian Stock Exchange (EGX) between 2017 and 2022. Future research can address this limitation by replicating the study across different African markets, particularly in the MENA region, where studies on TA remain limited. Such replications are especially valuable given the region’s recent economic and political shifts, influenced by the Arab Spring, which may impact CSR and tax behaviors.

Another limitation arises from the exclusive use of data from non-financial companies, restricting the generalizability of results, as CSR practices may vary widely across industries. In particular, financial firms and non-listed companies in Egypt might exhibit distinct CSR and TA behaviors, warranting a broader scope for future research.

Practical implications of this study suggest that policymakers and regulators in Egyptian market should consider the dual-edged nature of CSR. While such activities can be seen as socially beneficial, their association with tax avoidance highlights the need for monitoring mechanisms and stricter regulations to ensure that CSR initiatives do not serve as a façade for aggressive tax activities. This is especially critical for firms with political connection, which may exploit their influence to extract tax benefits.

Additionally, tax authorities in Egypt should prioritize audits and evaluations of politically connected companies with extensive CSR activities. This will help ensure compliance with tax regulations and prevent undue utilization of tax benefits under the guise of CSR. Egyptian companies, investors, and regulators could benefit from enhanced shareholder awareness of management’s potential CSR strategies that could be used for non-transparent objectives. Establishing effective internal controls is recommended to curtail TA practices, and CSR initiatives should shift towards sustainable practices rather than serving as a façade to obscure unethical activities. Furthermore, both domestic and international investors might consider looking beyond CSR disclosures when assessing investment opportunities, particularly with companies displaying high CSR engagement. The study’s findings also underscore the value of regulatory improvements to better safeguard shareholder rights, potentially making the Egyptian market more appealing to foreign investors.

Looking ahead, future studies could broaden the sample to include financial firms listed on the EGX and explore sector-specific comparisons within the Egyptian market. Future studies could explore the impact of managerial risk-taking incentives, which may lead to more TA through CSR. Also, corporate governance may be explored to mitigate this problem. Additional research might also incorporate non-Egyptian firms to validate the findings further. Exploring other moderating factors, such as audit firm characteristics (size, tenure, specialization), could add layers of insight into CSR and tax practices. Moreover, investigating board governance characteristics (e.g., board size, gender diversity, and meeting frequency) could reveal further nuances in the CSR and TA relationship, particularly in how these elements might moderate managerial opportunism.