The Moderating Effect of Female Directors on the Relationship Between Ownership Structure and Tax Avoidance Practices

Abstract

The primary objective of this study is to investigate the intricate relationship between different ownership structures, such as family, institutional, managerial, and foreign ownership, and tax avoidance practices. It also seeks to explore the moderating influence of female board members in shaping these relationships. This study utilizes balanced panel data from 72 industrial and service firms listed on the Amman Stock Exchange during the period of 2018 to 2023. The Generalized Method of Moments (GMM) was employed to estimate the results. The results indicate that family and foreign ownership positively influence tax avoidance practices, suggesting that families may engage in tax avoidance to benefit from rent extraction, while foreign investors may pressure managers to manipulate tax liabilities or shift profits across countries to minimize taxes. In contrast, the presence of female directors as well as institutional and managerial ownership is associated with a reduction in tax avoidance. Female directors play a moderating role in the relationship between ownership structure and tax avoidance. Their presence in interaction with institutional ownership reduces tax avoidance by focusing on tax compliance strategies. However, this effect changes in family and foreign-owned firms, where control over decision-making lies with the families or foreign shareholders, limiting the impact of female directors in promoting compliance and aligning their role with the tax avoidance strategies preferred by the controlling owners.

1. Introduction

Tax is a vital and significant source of revenue for the state budget, particularly in developing countries like Jordan. These revenues are used to provide various public services and implement economic and social development plans. However, for some firms, taxes are perceived as a cost that reduces the firm’s profits, diminishes cash flow, and potentially hinders investment opportunities that could enhance long-term wealth. Consequently, firms may engage in tax avoidance, which is often considered legal and ethical to a great extent, or in tax evasion, which is illegal and morally unacceptable (Wang et al., 2020).

Tax avoidance is a major concern that affects every tax system. It is a form of tax management that a firm can do legally by exploiting the privileges and exemptions provided by the government to firms, as well as taking advantage of weaknesses in tax laws and regulations (Napitupulu et al., 2019). Through tax planning, firms can implement structured strategies to reduce their tax burden as much as possible. These strategies include increased investment in fixed assets, transfer pricing, shifting profits to tax haven countries, and intellectual property structuring (Duhoon & Singh, 2023). On the one hand, tax avoidance may benefit shareholders by lowering costs and improving cash flow, thereby enabling further investments that enhance firm value (Masripah et al., 2017). Managers may also benefit from tax avoidance as they are compensated according to the firm’s profitability associated with effective tax management (Alkurdi & Mardini, 2020). On the other hand, tax avoidance poses a major challenge for governments, which seek to maximize tax revenues to fund public services and development efforts.

The Jordanian economy has faced numerous challenges over the years, the most pressing of which has been its heavy debt burden. This burden is largely attributed to regional instability, a high influx of refugees, limited natural resources, high unemployment rates, and severe water scarcity. These challenges have placed significant pressure on the economy and have constrained investment in development initiatives. In response, the Jordanian government launched a series of economic reforms, including reductions in government spending—especially subsidies, which had placed a heavy strain on the national budget. Jordan has also collaborated with the International Monetary Fund (IMF) to implement reforms aimed at improving the investment climate. However, these reforms are politically sensitive and may take time to produce substantial results (Qawqzeh, 2023). In this context, the tax sector plays a crucial role in supporting government spending. Aggressive tax practices undermine governments’ ability to secure financial resources (Gaaya et al., 2017). Therefore, strengthening government policies aimed at preventing and combating tax evasion and transforming the tax system into a more equitable and transparent one can be an effective step in increasing government revenues, harnessing them to achieve economic growth, and enhancing financial stability.

The phenomenon of tax avoidance has recently received the attention of researchers (Dakhli, 2022; Khan et al., 2017), particularly due to the growing concern among stakeholders and governments about the risks associated with tax avoidance, and the resulting loss of revenue and waste of public benefits for society. Studies conducted in the developed markets have identified various factors that influence the extent of corporate tax avoidance, including corporate governance (Barros & Sarmento, 2020), executive management characteristics (Khelil & Khlif, 2023), firm characteristics, corporate social responsibility (Sarhan, 2024; Chouaibi et al., 2022), and management compensation (Halioui et al., 2016). In light of findings from developed countries, this study aims to examine the factors influencing tax avoidance in Jordan, with a special focus on corporate governance mechanisms—especially ownership structures and board composition. Ownership structure may play an important role in shaping strategic decisions, including tax planning, as different types of owners often pursue divergent objectives and have different investment horizons. As highlighted by Kovermann and Velte (2019), ownership structure is a key factor in a company’s propensity to engage in tax avoidance. Furthermore, there is rising recognition that gender diversity on corporate boards, particularly the inclusion of female directors, may enhance board oversight and encourage stronger compliance, including in tax matters.

Despite growing research on tax avoidance and corporate governance, significant gaps remain, particularly in emerging markets like Jordan. Existing studies have generally focused on assessing the impact of ownership structures and gender diversity on tax behavior individually, with little attention paid to how these factors interact and influence tax behavior. Jordan’s institutional context, characterized by concentrated ownership structures and evolving corporate governance practice, provides an ideal setting for studying the relationship between governance mechanisms and tax avoidance. This importance is further underscored by the ongoing challenge that tax avoidance poses to public revenues, highlighting the importance of understanding how board composition, particularly the inclusion of female directors, may enhance tax compliance. Recent regulatory reforms in Jordan further reinforce the contextual relevance of this study. In 2017, the Jordan Securities Commission (JSC) updated its corporate governance code, introducing stricter transparency requirements, mandating the separation of the roles of the CEO and chairman, and requiring the disclosure of ownership structures (Jordan Securities Commission, 2017). On the tax front, the government has undertaken several digital reforms, including the implementation of the National Electronic Billing System (NEBS), and risk-based auditing powered by artificial intelligence, to enhance compliance and reduce tax evasion (Income and Sales Tax Department, 2023). Most notably, in 2024, the JSC introduced new governance guidelines requiring at least 20% female representation on boards of directors, thereby promoting gender diversity and aligning governance practices with international standards (Jordan Securities Commission, 2024). Collectively, these institutional developments make Jordan an especially relevant context for exploring how corporate governance factors, such as ownership structure and board diversity, influence corporate tax behavior.

Accordingly, this study stems from the need to fill this empirical and contextual gap. It contributes to the literature in four main ways. First, it examines the moderating role of board gender diversity in the relationship between ownership structure and tax avoidance, an area that remains underexplored. Second, it provides novel empirical evidence from Jordan, a country with distinct institutional characteristics. Third, it uses the Generalized Method of Moments (GMM) for panel data, enabling more robust estimation by addressing potential endogeneity concerns commonly encountered in corporate governance research. Fourth, the study confirms the robustness of its findings by employing an alternative tax avoidance proxy—Book-Tax Difference (BTD)—using the PCSE approach, with consistent results supporting the reliability of the conclusions.

2. Literature Review

2.1. Tax Avoidance

Over the past few years, several researchers have proposed definitions of tax avoidance. According to Dyreng et al. (2010), tax avoidance encompasses all practices that reduce a firm’s tax payments relative to its reported pre-tax accounting income. Tax avoidance is considered a legitimate means of reducing taxes (Gokten & Kucukkocaoglu, 2018). It refers to the act of reducing tax obligations by applying the tax system in a way not intended by the taxing authority (Hoseini & Safari Gerayli, 2018). According to Suranta et al. (2020), tax avoidance is a legal effort by firms, as taxpayers, to reduce their tax burden. On the other hand, Khelil and Khlif (2023) define tax avoidance as a managerial effort to reduce or eliminate the firm’s tax burden by exploiting loopholes in tax policies and regulations. These practices can be classified into acceptable and unacceptable forms, depending on their alignment with legal standards and ethical considerations. Acceptable tax avoidance involves legal strategies that do not include fraudulent acts, whereas unacceptable tax avoidance is achieved through illegal means. While tax avoidance itself is not illegal, it is often viewed as an undesirable practice because it reduces taxes and may lead to a decline in government revenue (Sumantri et al., 2024).

2.2. Ownership Structures and Tax Avoidance

2.2.1. Family Ownership

Previous studies have identified two contrasting viewpoints regarding the relationship between family ownership and tax avoidance. According to socioemotional wealth perspective, families tend to prioritize nonfinancial goals such as family legacy, reputation, and social image, leading to a push to lower tax avoidance (Brune et al., 2019a). Family owners are less motivated to generate additional cash flows from tax savings because of potential penalties and significant reputational concerns if tax authorities uncover aggressive tax practices (S. Chen et al., 2010). As a result, family firms are less inclined to participate in aggressive tax practices than non-family firms. This is consistent with the findings of (Chalevas et al., 2024; Ibrahim et al., 2021; Brune et al., 2019b; Steijvers & Niskanen, 2014). In contrast, the opposing perspective is grounded in Type II agency theory, which highlights potential agency conflicts between controlling family shareholders and minority shareholders. Agency costs rise when the controlling family have a greater ability to expropriate private benefits at the expense of minority shareholders (Shleifer & Vishny, 1986). From this perspective, family ownership is expected to be positively associated with tax avoidance, implying that families may prioritize their own interests by extracting rents from tax-saving opportunities. Tax avoidance is anticipated to generate a larger retained cash flow for the family, allowing them to conceal losses from other parties that may arise due to expropriation (Kovermann & Wendt, 2019). Gaaya et al. (2017) found a positive and significant association between family ownership and corporate tax avoidance practices, suggesting that families particularly benefit from tax savings, while audit quality acts as a moderating factor that reduces the incentives for family firms to engage in aggressive tax behavior. Cirillo et al. (2025) concluded that family business CEOs with a strong sense of transgenerational responsibility are more inclined to engage in tax avoidance as a strategy to safeguard the firm’s financial stability for future generations. This effect becomes more pronounced when the next generation is involved in management or when the firm is experiencing financial distress. Khelil and Khlif (2023) reviewed empirical studies on the relationship between family ownership and tax avoidance and found evidence of a negative association between family firms and tax avoidance practices in the United States, Finland, and Belgium, attributing this to prevailing socio-emotional wealth perspective in these countries. Conversely, the relationship between family firms and tax avoidance practices in Germany, Italy, Brazil, India, Malaysia, and Tunisia is positive and significant, reinforcing agency theory’s argument that families exploit minority shareholders. Moreover, other studies, such as Flamini et al. (2021) and Rakayana et al. (2021) showed that family ownership has no impact on tax avoidance practices. The previous contradiction in results led to the formulation of the following hypothesis:

H1. 

There is a significant relationship between family ownership and tax avoidance practices.

2.2.2. Institutional Ownership

Institutional investors play a crucial role in promoting good corporate governance by actively monitoring their investments and protecting shareholder interests (Ho, 2010). Regarding tax avoidance practices, Jiang et al. (2021) argued that institutional ownership has two potential effects on tax avoidance behavior: it may either encourage or deter such practices. On one side, institutional shareholders with significant ownership were found to be more motivated and skilled at influencing corporate tax avoidance practices and achieving greater profitability than other minority shareholders. On the other side, such tax avoidance practices may be perceived as politically motivated, attracting negative attention from the media, government, consumers, and public interest groups—a phenomenon known as tax shaming. This negative perception may deter institutional investors from publicly supporting tax avoidance. Kovermann and Velte (2019) suggested that institutional ownership in firms works both ways: it not only has the potential to increase tax avoidance, leading to increased profitability, but it also helps control tax avoidance when the associated risks outweigh the potential benefits. Khurana and Moser (2009) classify institutional ownership into short-term and long-term horizons. Their findings indicate that firms with short-term institutional investors are more inclined to engage in tax avoidance, as these investors tend to encourage management to adopt aggressive tax strategies aimed at maximizing firm value in the short term. In contrast, institutional shareholders with a longer investment horizon tended to be more proactive in overseeing management and discourage tax aggressiveness. Susilawati and Tarmidi (2024) found that institutional ownership negatively affects tax avoidance, meaning it helps reduce tax avoidance practices. They also found that audit quality strengthens this relationship. Prior research has yielded mixed results regarding the relationship between institutional ownership and tax avoidance. For instance, Khan et al. (2017), Jiang et al. (2021), and Sunarto et al. (2021) found a positive association. Conversely, other studies such as Alkurdi and Mardini (2020) and Ghalerodkhani et al. (2018) reported a negative relationship, indicating that institutional investors may act as monitors, discouraging aggressive tax. Meanwhile, Otieno (2014) and Tanko (2020) found no significant relationship between institutional ownership and tax avoidance. In light of these conflicting findings, the following hypothesis is proposed:

H2. 

There is a significant relationship between institutional ownership and tax avoidance practices.

2.2.3. Managerial Ownership

Managerial ownership refers to the proportion of a company’s outstanding shares held by executives who are actively involved in the firm’s strategic and operational decision-making processes (Bataineh et al., 2025). Managers play a pivotal role in creating value for a firm’s shareholders and significantly influence its overall success. They implement critical strategies and make important decisions, including those pertaining to corporate tax planning (Saragih & Ali, 2023). Previous studies on the relationship between managerial ownership and tax avoidance practices have yielded conflicting results. Some studies have suggested that managers may become more entrenched in their positions as managerial ownership increases. This may lead managers to prioritize their own interests over those of minority shareholders. Zeng (2011) noted that managers’ plans and activities focus on a limited and short-term time horizon, which increases the preference for tax avoidance. R. J. Lee and Kao (2020) and Desai and Dharmapala (2009) argued that self-interested managers engage in aggressive tax avoidance to enhance their power, seek control and personal benefits, conceal the managerial transfer of corporate resources, or improve financial performance, thereby achieving higher returns. This is consistent with the findings of Cabello et al. (2019) and Qawqzeh (2023), which indicate a statistically significant positive effect of managerial ownership on tax avoidance. In contrast, others such as Huang et al. (2018) and Hanlon and Slemrod (2009) have argued that engaging in risky tax avoidance activities may generate significant risks for firms and management, including audit risks by tax authorities, penalties and litigation risks, and reputational risks that reduce shareholder value, especially in the absence of effective governance mechanisms. Therefore, CEOs may consider reducing tax avoidance. Alkurdi and Mardini (2020) and Richardson et al. (2016) found a statistically significant negative association between managerial ownership and tax avoidance practices. Accordingly, the study formulates the following hypothesis:

H3. 

There is a significant relationship between managerial ownership and tax avoidance practices.

2.2.4. Foreign Ownership

Din et al. (2022) suggest that foreign ownership enhances the monitoring function and positively influences firm performance by aligning managerial behavior with the goal of maximizing shareholder wealth. Moreover, increased foreign ownership in domestic firms has positive effects, such as job opportunities, facilitating technology transfer, and improving international competitiveness (Mardiani et al., 2024). Furthermore, foreign managers help better understand corporate strategies, such as tax avoidance, and may influence their firms’ tax avoidance decisions (Barros & Sarmento, 2020). Foreign owners target countries that offer tax exemptions or lower tax rates, complementing their investments in countries with higher taxes (Suranta et al., 2020). Foreign institutions can serve as significant shareholders with considerable voting rights and play an active role in corporate tax avoidance practices by proposing tax strategies (Najihah et al., 2024). Therefore, depending on the risk and investment horizon of foreign owners, there may be incentives to enhance or limit tax avoidance efforts. Foreign transit investors are overly focused on short-term profits (Bushee, 2001), so corporate tax avoidance and increased tax savings can increase firm value and, consequently, shareholder wealth. Putri et al. (2024) found that foreign ownership has a significant positive effect on tax avoidance in the Indonesian banking sector. This is attributed to the investment objectives of foreign investors, who often prioritize short- and medium-term returns and may therefore support strategies that reduce tax liabilities in order to enhance after-tax profitability. However, according to S. Chen et al. (2010) and Desai and Dharmapala (2009), tax avoidance activities may encourage managerial opportunism due to their intricate nature, which can negatively impact a firm’s long-term value through management rent extraction, reputational risk, and litigation risk. As a result, long-term foreign investors are less likely to engage in tax avoidance practices. Prior research has produced varying results concerning the relationship between foreign ownership and tax avoidance. For instance, Hasan et al. (2022) and Najihah et al. (2024) found a significant negative association, indicating that firms with higher levels of foreign ownership tend to engage less in tax avoidance. In contrast, Shi et al. (2020), Alkurdi and Mardini (2020), and Suranta et al. (2020) identified a significant positive association, suggesting that foreign investors may support more aggressive tax strategies. Conversely, Muji and Waluyo (2024) reported no significant association between foreign ownership and tax avoidance practices. Accordingly, this study proposes the following hypothesis:

H4. 

There is a significant relationship between foreign ownership and tax avoidance practices.

2.2.5. The Moderating Role of Female Directors

Resource dependence theory posits that the board of directors plays a critical role in providing essential resources that support an organization’s operations and strategic objectives (Pfeffer & Salancik, 1978). The most important resources provided by the board of directors to the firm were expertise, advice, building external relationships, connecting the firm with key stakeholders, and helping to formulate strategy (Hillman & Dalziel, 2003). According to Agyemang-Mintah and Schadewitz (2019), gender diversity on corporate boards is a key factor contributing to a firm’s success. They emphasized the need for women to be part of any decision-making process, as women have the same ability as men to contribute to the firm’s financial success. Moreover, women are highly involved in the effectiveness of corporate governance, as their risk-averse nature leads them to be more cautious in decision-making, as they tend to carefully evaluate executive decisions and their subsequent impacts, including tax planning (Winasis & Yuyetta, 2017). L. H. Chen et al. (2019) also found that gender-diverse boards tend to be more cautious about reputational risks linked to aggressive tax strategies. Rakia et al. (2024), Jarboui et al. (2020), and Hoseini et al. (2019) indicated that as the number of female board members increases, the likelihood of engaging in tax avoidance activities decreases. This may be because female directors are more active and effective in monitoring firms than their male counterparts and are perceived as more risk-averse and less tolerant of opportunistic and unethical behavior (Luo et al., 2017). In addition to risk aversion, female board members tend to represent a broader set of stakeholder interests, including regulators, employees, and society at large, which encourages firms to adopt ethical tax policies. According to Liu et al. (2020) and Tapver et al. (2020), women on boards are more likely than men to perform their duties in ways that reflect a commitment to stakeholder interests. This broader perspective helps reduce opportunistic behaviors such as tax avoidance and promotes tax compliance as part of corporate social responsibility and good governance. Similarly, Alkurdi et al. (2024a) indicated that the presence of females on boards can contribute to more effective resolution of agency problems and enhance accountability by ensuring that companies comply with tax requirements and pay the correct contributions. From another perspective, Suleiman (2020) examined the impact of female representation in corporate governance on tax avoidance practices and explored the moderating role of accounting conservatism in this relationship. The study found that female representation, whether as CEOs, board members, or audit committee members, had a positive effect on tax avoidance practices, while accounting conservatism reduced tax avoidance practices. Yahya et al. (2021) also concluded that CEO power increases the likelihood of tax avoidance, as powerful CEOs view aggressive tax planning as an alternative investment opportunity and a mechanism to achieve personal interests, and powerful CEOs exploit the presence of female directors to promote aggressive tax behavior. Meanwhile, studies by Pandapotan et al. (2024) and Widuri et al. (2020) found no significant impact of gender diversity on tax avoidance practices.

A review of the recent literature reveals that corporate tax avoidance studies have increasingly emphasized the roles of ownership structure and board diversity. Baatwah et al. (2025) highlighted how female ownership and gender homophily influence tax avoidance practices, revealing that greater female participation at the ownership level is associated with lower levels of tax avoidance. This suggests that gender dynamics within ownership structures may influence firms’ ethical considerations regarding tax compliance. Examining ownership effects more closely, Lin et al. (2025) found that firms with significant state ownership tend to engage less in aggressive tax planning, attributing this behavior to the political and social scrutiny faced by state-controlled enterprises. However, although these studies underscore important ownership characteristics, they tend to overlook the potential moderating role of board composition particularly the presence of female directors in shaping corporate tax behavior.

Complementing these findings, Velte (2024) provides a structured literature review on corporate ownership structures and tax avoidance. His review highlights a significant gap in the literature on the interactive effects of governance characteristics and ownership types on tax avoidance. In particular, Velte calls for more empirical research that integrates aspects of board gender diversity into the relationship between ownership and tax avoidance, a gap directly addressed by the current study. Additional research streams have explored broader contextual factors. Shim (2024) analyzes the role of financial subsidiaries within Korean business groups, identifying the mechanisms through which internal financial networks facilitate tax avoidance, but without considering governance variables such as gender diversity. Similarly, Yang et al. (2025) investigate cross-border cooperation’s influence on the tax strategies of U.S. cross-listed firms, focusing on international dimensions rather than internal corporate governance factors. On the other hand, Souguir et al. (2024) explore how corporate environmental performance, moderated by ownership structure, affects tax avoidance, questioning whether firms are genuinely adopting green policies or simply distracting regulators from tax practices. However, their focus on environmental strategies ignores moderating factors related to governance.

Building on the recent literature, this study contributes by examining how the presence of female directors moderates the relationship between ownership structure and corporate tax avoidance practices. While prior research has typically treated ownership types and gender diversity as separate determinants of tax behavior, very few studies have investigated their interaction—particularly in emerging economies. However, to the best of the researcher’s knowledge, no prior research has addressed this issue within the Jordanian context, where ownership is highly concentrated and governance frameworks are still evolving. Although some studies have examined the direct effect of board gender diversity on tax avoidance, this study offers novel insights by focusing on the moderating role of female board representation. Accordingly, the following hypotheses are formulated.

H5. 

Female directors moderate the relationship between ownership structure and tax avoidance practices.

H5a. 

Female directors moderate the relationship between family ownership and tax avoidance practices.

H5b. 

Female directors moderate the relationship between institutional ownership and tax avoidance practices.

H5c. 

Female directors moderate the relationship between managerial ownership and tax avoidance practices.

H5d. 

Female directors moderate the relationship between foreign ownership and tax avoidance practices.

3. Methodology

Sample of the Study

The study population comprises all industrial and service firms listed on the Amman Stock Exchange during the period from 2018 to 2023. Data for the independent variables (family, institutional, managerial, and foreign ownership) and the moderator variable (gender diversity) were collected manually from the annual reports of the listed firms accessible on the Jordan Securities Commission website. This study measures tax avoidance using Effective Tax Rate (ETR) and Cash Flow Effective Tax Rate (CFETR). We excluded observations for firms that did not have sufficient data to calculate the study variables. Thus, the final sample consists of 72 firms with 432 firm-year observations.

4. Measurement of Variables

This subsection outlines the definitions and measurement proxies of all variables used in the current study. As previously stated, this study investigates the moderating effect of female directors on the relationship between ownership structure and tax avoidance practices. The dependent variable (tax avoidance practices) is measured using two metrics: ETR and CFETR. ETR is used to capture tax avoidance practices that are implied in tax shelters and loopholes in tax laws (Ibrahim et al., 2021; Dyreng et al., 2017). It is measured by dividing the tax expense by pre-tax accounting income, the lower ETR indicates the higher the firm’s tendency to conduct tax avoidance. The CFETR relies on information derived from the cash flow statement, which helps to exclude the influence of earnings management (X. Chen et al., 2014). It is calculated by dividing the tax expenses by operating cash flows. The independent variables in this study include various forms of ownership, namely family ownership, institutional ownership, managerial ownership, and foreign ownership. Along with the main variables of this study, several control variables were included: firm size, firm leverage, firm profitability, and firm age.

Table 1. Variable definitions.

Variable	Abbreviation	Measurement	Previous Literature
Panel A. Dependent Variable (Tax Avoidance)
Effective Tax Rate (ETR)	ETR	Total tax expense divided by the income before taxes.	Alkurdi and Mardini (2020); Jarboui et al. (2020); Wongsinhirun et al. (2024); Hossain et al. (2025).
Cash Flow Effective Tax Rate (CFETR)	CFETR	Total tax expenses divided by the operational cash flows.	Minnick and Noga (2010); S. Chen et al. (2010); Qawqzeh (2023); Nassar et al. (2024).
Panel B. Independent Variables (Ownership Structure)
Family Ownership	FAMO	The percentage of firm ownership owned by the same family members.	Gaaya et al. (2017); Kovermann and Wendt (2019).
Institutional Ownership	INSO	The percentage of ownership held by financial institutions (insurance firms, investment funds, pension funds, and investment firms).	Abed et al. (2020); Anggraini and Widarjo (2020); Alkurdi et al. (2024b)
Managerial Ownership	MANO	The percentage of ownership owned by executives.	Wongsinhirun et al. (2024); Najihah et al. (2024).
Foreign Ownership	FORO	The percentage of shares owned by foreign investors.	Suranta et al. (2020); Bataineh (2021).
Panel C. Moderator Variable
Female Directors	FEMD	The percentage of female directors on the firm’s board.	Widuri et al. (2020); Riguen et al. (2020); Rakia et al. (2024).
Panel D. Control Variables
Firm Size	FSIZ	The logarithm of total assets at the end of the year.	Bataineh et al. (2018); Jarboui et al. (2020).
Firm Leverage	FLEV	The ratio of the total book value of liabilities to total assets.	Rakia et al. (2024); Amara et al. (2025).
Firm Profitability	FPRO	The ratio of net income to total assets.	Salhi et al. (2020); Hossain et al. (2025).
Firm Age	FAGE	The number of years since theinception of the firm.	Alkurdi and Mardini (2020); Bataineh et al. (2025)

Source: Author’s own creation.

summarizes the variables used in this study and the main sources of the proxies used.

5. Regression Model

To investigate the relationships between the dependent, independent, moderating and control variables, tax avoidance is represented by two indicators, ETR and CFETR; therefore, the present study developed the following regression models:

ETR_i,t = β₀ + β₁ ETR_i,t−1+ β₂ FAMO_i,t + β₃ INSO_i,t + β₄ MANO_i,t +
Β₅ FORO_i,t + β₆ FSIZ_i,t + β₇ FLEV_i,t + β₈ FPRO_i,t + β₉ FAGE_i,t + ε_i,t

(1)

ETR_i,t = β₀ + β₁ ETR_i,t−1 + β₂ FAMO_i,t + β₃ INSO_i,t + β₄ MANO_i,t + β₅ FORO_i,t +
β₆ FEMD_i,t + β₇ FSIZ_i,t + β₈ FLEV_i,t + β₉ FPRO_i,t + β₁₀ FAGE_i,t + ε_i,t

(2)

ETR_i,t = β₀ + β₁ ETR_i,t−1 + β₂ FAMO_i,t + β₃ (FAMO × FEMD)_i,t + β₄ INSO_i,t +
β₅ (INSO × FEMD)_i,t + β₆ MANO_i,t + β₇ (MANO × FEMD)_i,t +
β₈ FORO_i,t + β₉ (FORO × FEMD)_i,t + β₁₀ FSIZ_i,t + β₁₁ FLEV_i,t +
β₁₂ FPRO_i,t + β₁₃ FAGE_i,t + ε_i,t

(3)

CFETR_i,t = β₀ + β₁ CFETR_i,t−1 + β₂ FAMO_i,t + β₃ INSO_i,t + β₄ MANO_i,t +
β₅ FORO_i,t + β₆ FSIZ_i,t + β₇ FLEV_i,t + β₈ FPRO_i,t + β₉ FAGE_i,t + ε_i,t

(4)

CFETR_i,t β₀ + β₁ CFETR_i,t−1 + β₂ FAMO_i,t + β₃ INSO_i,t + β₄ MANO_i,t + β₅ FORO_i,t +
β₆ FEMD_i,t + β₇ FSIZ_i,t + β₈ FLEV_i,t + β₉ FPRO_i,t + β₁₀ FAGE_i,t + ε_i,t

(5)

CFETR_i,t = β₀ + β₁ CFETR_i,t−1 + β₂ FAMO_i,t + β₃ (FAMO × FEMD)_i,t + β₄ INSO_i,t +
β₅ (INSO × FEMD)_i,t + β₆ MANO_i,t + β₇ (MANO × FEMD)_i,t +
β₈ FORO_i,t + β₉ (FORO × FEMD)_i,t + β₁₀ FSIZ_i,t + β₁₁ FLEV_i,t +
β₁₂ FPRO_i,t + β₁₃ FAGE_i,t + ε_i,t

(6)

6. Empirical Results

6.1. Descriptive Statistics

Descriptive statistics of the study variables are presented in

Table 2. Descriptive statistics.

Variable	Observations	Mean	S.D.	Min.	Max.
ETR	432	0.154	0.497	−0.36	9.548
CFETR	432	0.068	0.347	−2.656	2.792
Family Ownership	432	0.13	0.242	0	0.858
Institutional Ownership	432	0.454	0.317	0	1
Managerial Ownership	432	0.030	0.099	0	0.776
Foreign Ownership	432	0.115	0.209	0	0.943
Female Directors	432	0.053	0.096	0	0.33
Firm Size	432	18.69	1.73	12.99	22.79
Firm Leverage	432	0.364	0.236	0.0028	0.998
Firm Profitability	432	0.015	0.089	−0.751	0.371
Firm Age	432	30.181	16.947	9	86

Source: Author’s own creation.

. During the period 2018–2023, the mean of ETR was 0.154 (15.4%) and the standard deviation was 0.497, while CFETR showed a lower mean of 0.068 (6.8%) with a standard deviation of 0.347. The significant difference between these two measures indicates that firms use different strategies in their tax management practices. In addition, the mean value of ETR exceeded the mean of CFETR. This result is consistent with the findings of Qawqzeh (2023), Nassar et al. (2024), and Alkurdi and Mardini (2020).

Examination of ownership structure reveals diverse patterns among the sample firms. Institutional ownership emerges as the dominant form, representing an average of 45.4% of total shares, indicating significant involvement of financial institutions in the Jordanian industrial and service sectors. Family ownership and foreign ownership show moderate levels of participation, with means of 13% and 11.5%, respectively. Managerial ownership exhibits the lowest average at 3%, suggesting limited direct ownership by executive management in the sample firms. The presence of female directors on corporate boards demonstrates relatively modest representation, with an average of 5.3% female board members. This figure ranges from complete absence in some firms to a maximum of 33% in others, indicating significant variation in gender diversity practices across the sample. The distribution suggests that gender diversity in corporate governance remains an evolving aspect in the Jordanian sampled firms. Among the control variables, firm size exhibits substantial variation, with the natural logarithm of total assets averaging 18.69, reflecting diverse organizational scales within the sector. The average firm age of 30.18 years indicates a mature sample, while the mean leverage ratio of 36.4% suggests moderate debt levels. Profitability measurements show considerable variation with a mean of 1.511%, indicating diverse financial performance across the sample.

6.2. Correlation Analysis

The correlation analysis presented in

Table 3. Pearson correlation matrix.

Variables	(1)	(2)	(3)	(4)	(5)	(6)	(7)	(8)	(9)	(10)	(11)
(1) ETR	1.000
(2) CFETR	−0.263	1.000
(3) MANO	−0.024	0.010	1.000
(4) INSO	0.085	0.109	−0.244	1.000
(5) FORO	−0.018	0.079	−0.083	0.213	1.000
(6) FAMO	−0.024	−0.032	0.347	−0.194	−0.114	1.000
(7) FEMD	0.011	0.053	0.117	0.111	0.033	−0.100
(8) LnFSIZ	0.045	0.127	−0.091	0.196	0.231	−0.150	1.000
(9) FLEV	0.065	−0.053	−0.154	0.122	−0.103	−0.021	0.068	1.000
(10) FPRO	0.078	0.115	0.009	0.279	0.188	−0.026	−0.114	0.112	1.000
(11) FAGE	1.000	1.000	1.000	1.000	1.000	1.000	0.177	0.259	−0.196	1.000

Source: Author’s own creation.

highlights key relationships among the study variables. The negative correlation (−0.263) between ETR and CFETR suggests these measures capture distinct aspects of tax avoidance behavior, providing complementary insights into corporate tax practices. The ownership structure variables demonstrate several significant interrelationships, with managerial ownership showing a positive correlation with family ownership (0.347), suggesting potential alignment between family and management interests. Institutional ownership shows positive correlation with foreign ownership (0.213), indicating possible complementary investment patterns between institutional and foreign investors.

The presence of female directors shows interesting associations with various ownership structures. Positive correlations are observed with both institutional (0.111) and managerial ownership (0.117), while a negative correlation exists with family ownership (−0.100). These relationships suggest that different ownership structures may influence board gender diversity differently. The control variables show logical interconnections, with firm size positively correlating with both foreign ownership (0.231) and firm age (0.504), while profitability shows positive associations with institutional ownership (0.279) and firm size (0.259). In short, the highest correlation coefficient between the explanatory variables is 0.504. This indicates that there is no evidence of multicollinearity problems. In addition, the researcher also conducted the Variance Inflation Factor (VIF) test as shown in

Table 4. Variance Inflation Factor (VIF) and tolerances.

Variables	VIF	Tolerance (1/VIF)
Family Ownership	1.212	0.825
Institutional Ownership	1.254	0.797
Managerial Ownership	1.276	0.784
Foreign Ownership	1.214	0.824
Female Directors	1.085	0.922
Firm Size	1.688	0.592
Firm Leverage	1.19	0.841
Firm Profitability	1.256	0.796
Firm Age	1.519	0.658
Mean VIF	1.299

, which confirms the absence of a significant multi-linear correlation between the study variables.

The mean VIF of 1.299 falls well below the conventional threshold of 10, with the highest individual VIF being 1.688 for firm size. The low VIF values across all variables and tolerance values greater than 0.1 indicate that regression estimates will not be negatively affected by multicollinearity issues, ensuring the reliability of subsequent analyses.

7. Panel Data Diagnostic Tests

The panel data diagnostic tests in

Table 5. Panel data diagnostic tests.

Test Type	ETR Model	CFETR Model
Hausman Test	χ2(8) = 5.20 (p = 0.7362)	χ2(8) = 22.31 (p = 0.0044)
Heteroskedasticity	LM Test = 3.65 × 106 (p = 0.0000)	χ2(72) = 4.4 × 107 (p = 0.0000)
Wooldridge Autocorrelation	F(1,71) = 10.425 (p = 0.0019)	F(1,71) = 0.498 (p = 0.4829)
Model Selection	System GMM	System GMM

Note: The presence of heteroskedasticity, potential endogeneity, and autocorrelation (in ETR model) supports the use of System GMM estimation for both models. GMM is particularly suitable for dynamic panel data where the dependent variable may be influenced by its past values and where explanatory variables might not be strictly exogenous.

provide crucial insights for model specification.

The Hausman test results indicate different appropriate estimation approaches for the two tax avoidance measures. For the CFETR model, the significant test statistic (χ²(8) = 22.31, p = 0.0044) suggests the fixed effects model as appropriate, while the ETR model’s results (χ²(8) = 5.20, p = 0.7362) favor the random effects approach. Heteroskedasticity testing reveals significant concerns in both models, with the Modified Wald test for the CFETR model (χ²(72) = 4.4 × 10⁷, p = 0.0000) and the LM test for the ETR model (3.65 × 10⁶, p = 0.0000) both indicating the presence of heteroskedasticity. The autocorrelation analysis presents different patterns across the models, with the CFETR model showing no significant autocorrelation (F = 0.498, p = 0.4829), while the ETR model demonstrates significant autocorrelation (F = 10.425, p = 0.0019).

According to the literature (Areneke et al., 2023; Hermalin & Weisbach, 1988; Blundell & Bond, 1998), endogeneity represents a crucial methodological concern in corporate governance research that can significantly affect the validity of empirical findings. This issue arises from three main sources: unobserved heterogeneity (where firm-specific characteristics may influence both governance structures and firm performance simultaneously), simultaneity (where governance decisions and performance outcomes may be mutually determined), and dynamic endogeneity (where current governance choices are influenced by past performance). Traditional estimation methods like OLS may yield biased and inconsistent results in the presence of these endogeneity issues. The GMM estimator emerges as a particularly suitable solution, as it can effectively address all three types of endogeneity simultaneously. Unlike other methods, GMM does not require external instruments, instead utilizing the lagged values of the endogenous variables as internal instruments, making it especially valuable when valid external instruments are difficult to identify (Abu Afifa et al., 2024; Farag et al., 2014; Orazalin & Mahmood, 2021). In the current study, we apply the GMM estimator by incorporating the lagged dependent variables ETR_i,t−1 and CFETR_i,t−1 into the regression models to control for dynamic panel bias and potential endogeneity. Furthermore, the GMM estimator is robust across various estimation techniques, as it operates without unnecessary assumptions about the precise distribution of the data generation process and error terms (Blundell & Bond, 1998).

8. GMM Estimation Results

To examine the moderating effect of female directors on the relationship between ownership structure and tax avoidance practices using ETR and CFETR as a proxy measure, we employ the System GMM estimation method.

Table 6. System GMM results.

Variables	ETR Model			CFETR Model
	Direct Impact	Direct Impact	Moderating Impact	Direct Impact	Direct Impact	Moderating Impact
	Equation (1)	Equation (2)	Equation (3)	Equation (4)	Equation (5)	Equation (6)
L.ETR	0.285 ***	0.279 ***	0.268 ***	0.238 ***	0.242 ***	0.248 ***
	−0.042	−0.04	−0.037	−0.038	−0.039	−0.041
FAMO	−0.142 **	−0.138 **	−0.148 **	0.055	0.058	0.064
	−0.061	−0.059	−0.064	−0.057	−0.058	−0.06
INSO	0.156 **	0.162 **	0.164 **	0.145 **	0.148 **	0.154 **
	−0.068	−0.07	−0.071	−0.064	−0.065	−0.067
MANO	−0.084	−0.088	−0.095	0.132 **	0.135 **	0.141 **
	−0.072	−0.074	−0.076	−0.068	−0.069	−0.071
FORO	0.068	0.072	0.079	−0.125 **	−0.128 **	−0.134 **
	−0.065	−0.067	−0.069	−0.062	−0.063	−0.065
FEMD	-	0.145 **	0.152 **	-	−0.082	−0.088
		−0.058	−0.061		−0.054	−0.056
FAMO × FEMD	-	-	−0.138 **	-	-	0.068
			−0.058			−0.053
INSO × FEMD	-	-	0.146 **	-	-	0.131 **
			−0.062			−0.057
MANO × FEMD	-	-	−0.082	-	-	0.118 **
			−0.071			−0.066
FORO × FEMD	-	-	0.069	-	-	−0.125 **
			−0.065			−0.06
LnFSIZ	0.024 **	0.026 **	0.029 **	0.022 **	0.024 **	0.028 **
	−0.01	−0.011	−0.012	−0.009	−0.01	−0.012
FLEV	−0.082 *	−0.085 *	−0.089 *	−0.095 **	−0.098 **	−0.104 **
	−0.045	−0.046	−0.048	−0.042	−0.043	−0.045
FPRO	0.055	0.058	0.064	0.118 **	0.121 **	0.127 **
	−0.038	−0.039	−0.041	−0.035	−0.036	−0.038
FAGE	0.008	0.009	0.011	0.015	0.018	0.024
	−0.009	−0.01	−0.011	−0.008	−0.009	−0.011
Constant	0.142 ***	0.138 ***	0.132 ***	0.115 ***	0.118 ***	0.124 ***
	−0.035	−0.034	−0.032	−0.032	−0.033	−0.035
Diagnostic Tests
AR(1)	−2.45 **	−2.48 **	−2.55 **	−2.38 **	−2.41 **	−2.47 **
AR(2)	0.88	0.86	0.81	0.92	0.89	0.83
Hansen J-test	42.38	43.15	45.18	40.25	41.32	43.58
Hansen (p-value)	0.285	0.278	0.254	0.295	0.288	0.275
Instruments	61	61	61	61	61	61
Observations	360	360	360	360	360	360

* p < 0.10, ** p < 0.05, *** p < 0.01.

presents the results of this analysis.

The validity of our ETR model estimation is confirmed through several diagnostic tests. The AR (1) test shows significant negative first-order autocorrelation (−2.45 to −2.55, p < 0.05), while the AR (2) test indicates no significant second-order autocorrelation (0.81 to 0.88), supporting the use of lagged variables as instruments. The Hansen J-test of over-identifying restrictions shows p-values ranging from 0.254 to 0.285, well above the conventional 0.10 threshold, indicating valid instruments. The number of instruments (61) compared to the sample size (360) maintains a reasonable ratio, avoiding potential bias from instrument proliferation. Similarly, diagnostic tests of the CFETR model confirm the validity of the model, with significant AR (1) statistics (−2.38 to −2.47, p < 0.05), non-significant AR (2) values (0.83 to 0.92), and satisfactory Hansen test results (p values: 0.275 to 0.295). These indicators show adequate model specification and instrument validity.

The results reveal that family ownership (FAMO) shows a significant negative relationship with ETR (β = −0.142, p < 0.05). Hence, hypothesis H1 is accepted. As mentioned earlier, since ETR and CFETR are reverse proxies for tax avoidance, this result suggests that a higher proportion of family ownership in a firm is associated with an increased likelihood of engaging in tax reduction practices. This may be attributed to the prevalence of Type II agency problems and the expropriation-for-private-benefit hypothesis within family firms, where family owners are expected to act as controlling owners and extract rents for their own benefit from tax savings at the expense of minority shareholders (Shleifer & Vishny, 1986). This result aligns with the findings reported in prior research (Gaaya et al., 2017; Kovermann & Wendt, 2019; and Almaharmeh et al., 2024), and contradicts the study of (C. H. Lee & Bose, 2021). On the contrary, institutional ownership (INSO) is found to have a statistically significant positive relation with ETR and CFETR through regression analysis (β = 0.156 p < 0.05) (β = 0.145 p < 0.05). Based on this finding, it can be concluded that hypothesis H2 is supported. This suggests that institutional investors may act as a restraining force on tax avoidance behaviors to minimize legal and reputational risks because such investors are oriented at long-term sustainable performance rather than short-term returns. This finding aligns with (Dakhli, 2022; Alkurdi & Mardini, 2020; Abdul Wahab et al., 2017), but contrasts with those of (Jiang et al., 2021; and Khan et al., 2017). However, the results also indicated a positive relationship between MAOWN and CFETR (β = 0.132 p < 0.05). Hence, H3 is confirmed. This means that managerial ownership significantly reduces tax avoidance activities. Higher managerial ownership incentivizes managers to adopt more conservative and transparent tax policies. An explanation for this is that when managers own a significant portion of a firm’s share, their interests are more likely to align with those of shareholders. This makes them more vigilant about the firm’s reputation and more likely to avoid the risk of intense tax audits that may negatively affect its long-term value. This supports the findings of (Huang et al., 2018; Alkurdi & Mardini, 2020), but contradicts the findings of (Cabello et al., 2019; Qawqzeh, 2023). The results indicated that FORO had an adverse effect on CFETR (β = −0.125, p < 0.05), implying that foreign ownership promotes tax avoidance practices. Hypothesis H4 is therefore accepted. This can be explained by the tendency of foreign investors to exploit opportunities for profit shifting across countries, utilizing transfer pricing mechanisms and legal loopholes to minimize tax liabilities in both host and home countries. Their focus on profit maximization, coupled with less emphasis on national tax liabilities, leads firms to adopt aggressive tax avoidance practices. This finding is consistent with (Suranta et al., 2020; Salihu et al., 2015), and contradicts (Muji & Waluyo, 2024; Wen et al., 2020).

Introducing board gender diversity (FEMD) into Equations (2) and (5) reveals a positive and statistically significant effect on the ETR model (β = 0.145, p < 0.05). This means that with more female board members, the level of tax avoidance tends to decrease. This result is consistent with the findings of several previous studies, such as Jarboui et al. (2020), Hoseini et al. (2019), and Winasis and Yuyetta (2017) who found that female presence can enhance corporate governance, increase management performance monitoring, and lead to better compliance with regulations including taxes. The moderating effects examined in both the ETR and CFETR models show that the presence of female directors has a significant negative interaction with family ownership (β = −0.138, p < 0.05) under ETR, and foreign ownership (β = −0.125, p < 0.05) under CFETR; thus, Hypotheses H5a and H5d are accepted.

As illustrated in Table 6, the presence of female directors on the board is associated with improved tax compliance. However, a negative relationship emerges when female directors interact with family ownership in the ETR model and with foreign ownership in the CFETR model. This effect may be due to the small and modest proportion of females on the board, which weakens their influence on decisions, as well as the dominance of families and foreign investors in the decision-making process related to tax planning, as these groups may prioritize tax reduction over full tax compliance. This result can also be interpreted in the context of developing countries, such as Jordan, where institutional and regulatory frameworks may not be sufficiently supportive of empowering women within boards of directors. This is particularly true in family-owned firms, which are often dominated by kinship ties and patriarchal management traditions, as well as in firms with strong foreign ownership where strategic decisions may be centralized. Such conditions may limit the effectiveness of female board members as a monitoring mechanism, reducing their ability to influence tax compliance behavior meaningfully.

Regarding the interactive role of female board presence with institutional ownership on ETR and CFETR, it was found to have a positive effect in both models (β = 0.146, p < 0.05) and (β = 0.131, p < 0.05), respectively; therefore, hypothesis H5b is accepted. This can be explained by the fact that institutional ownership has a positive effect on tax compliance as an independent variable. This effect becomes stronger when combined with female board presence, which contributes to enhancing governance and reducing opportunities for tax avoidance. The results also showed that there is a significant positive effect of the interaction variable (MANO × FEMD) on the CFETR model (β = 0.118, p < 0.05). This result supports the developed hypothesis H5c. Thus, the presence of female directors on the board moderates the relationship between managerial ownership and tax avoidance, as their presence can promote management strategies that prioritize social responsibility and adherence to legal regulations, thereby improving overall tax compliance. This aligns with several studies that highlight the moderating role of female directors in tax avoidance. For instance, Hossain et al. (2025) found that a greater number of female directors moderates the relationship between larger, more profitable firms and their tendency to avoid taxes. Similarly, Rakia et al. (2024) reported that having women on the board is a statistically significant moderating variable. Their research indicates that increasing female representation enhances the relationship between corporate social responsibility disclosure and reduced tax avoidance. The results on the control variables in both models show that firm size (FSIZ) and leverage (FLEV) show strong significance (p < 0.05) across all specifications. While firm profitability (FPRO) has a significant positive effect on CFETR, firm age remains insignificant.

9. Robustness Test Using Alternative Tax Avoidance Measure (BTD)

To validate the reliability of our primary findings on tax avoidance, we employ Book-Tax Differences (BTD) as an alternative proxy. BTD reflects the gap between book income and taxable income, offering a complementary perspective on firms’ tax avoidance behavior compared to traditional measures such as ETR and CETR (Gaaya et al., 2017; S. Chen et al., 2010). A higher BTD is generally regarded as a potential indicator of more aggressive tax avoidance practices.

Following Desai and Dharmapala (2006) and Tang and Firth (2011), we estimate our models using Panel Corrected Standard Errors (PCSE) methodology, which effectively addresses cross-sectional dependence, heteroskedasticity, and autocorrelation in panel data (Beck & Katz, 1995). The PCSE approach is particularly appropriate for the BTD measure given its distributional properties and the panel diagnostic test results reported in

Table 7. Diagnostic tests for BTD model and estimation method selection.

Test Type	BTD Model
Hausman Test	χ2(8) = 111.62 (p = 0.0000)
Heteroskedasticity	LM Test = 1.4 × 1005 (p = 0.0000)
Wooldridge Autocorrelation	F(1,71) = 7.297(p = 0.0086)
Model Selection	Robustness measure using Panel Corrected Standard Errors PCSE Model

In the BTD model we use the Panel Corrected Standard Errors PCSE Model.

.

Table 8. Panel Corrected Standard Errors results (Dependent Variable: BTD).

	BTD Model
Variables	Direct Impact	Direct Impact	Moderating Impact
FAMO	0.0066 ***	0.0066 ***	0.0064 ***
	−0.0024	−0.0024	−0.0022
INSO	−0.0065 **	−0.0065 **	−0.0069 **
	−0.0027	−0.0027	−0.0029
MANO	−0.0022	−0.0021	−0.0042 ***
	−0.0015	−0.0014	−0.0015
FORO	−0.0055 ***	−0.0054 ***	−0.0058 ***
	−0.0018	−0.0019	−0.0022
FEMD	-	0.0003	−0.0018
		−0.0023	−0.0023
FAMO × FEMD	-	-	−0.0013
			−0.0026
INSO × FEMD	-	-	−0.0036
			−0.0024
MANO × FEMD	-	-	−0.0044 **
			−0.0020
FORO × FEMD	-	-	−0.0080 **
			−0.0035
LnFSIZ	0.0124	0.0124	0.0127
	−0.0082	−0.0082	−0.0078
FLEV	−0.0003	−0.0003	−0.0003 *
	−0.0002	−0.0002	−0.0002
FPRO	0.0004	0.0004	0.0004
	−0.0007	−0.0007	−0.0007
FAGE	−0.0005 **	−0.0005 **	−0.0005 **
	−0.0002	−0.0002	−0.0002
Constant	−0.0917	−0.0918	−0.0927
	−0.0637	−0.0631	−0.0601
R-squared	0.1082	0.1083	0.1284
Wald chi2	128.40 ***	132.98 ***	315.81 ***

* p < 0.10, ** p < 0.05, *** p < 0.01.

presents the PCSE regression results with BTD as the dependent variable. The models’ statistical validity is confirmed by the significant Wald chi-square statistics (ranging from 128.40 to 315.81, p < 0.001) and R-squared values between 0.1082 and 0.1284, indicating reasonable explanatory power for tax avoidance research.

The BTD results largely corroborate our main findings while revealing additional insights. Family ownership (FAMO) demonstrates a significant positive relationship with BTD (β = 0.0064 to 0.0066, p < 0.01) across all models, consistent with the negative relationship observed in the ETR model. This confirms that firms with family ownership tend to engage in higher levels of tax avoidance. Similarly, institutional ownership (INSO) exhibits a significant negative relationship with BTD (β = −0.0065 to −0.0069, p < 0.05), which is consistent with its positive association with both CETR and ETR. This suggests that institutional investors may play a monitoring role that leads to lower levels of tax avoidance and promotes a more conservative tax strategy.

Foreign ownership (FORO) exhibits a strong negative relationship with BTD (β = −0.0054 to −0.0058, p < 0.01). This finding contradicts the negative relationship between foreign ownership and CETR. The contradiction between the negative relationship of foreign ownership (FORO) with BTD and its negative relationship with CETR can be explained by the differences in the measures used to assess tax avoidance. Firms with foreign ownership show a decrease in BTD, indicating reduced use of tax avoidance strategies based on manipulating book-tax differences. In contrast, the negative relationship with CETR suggests that foreign-owned firms pay lower cash taxes, which may result from using financial or structural strategies, such as shifting profits to low-tax jurisdictions or benefiting from tax incentives. Thus, this contradiction reflects that foreign-owned firms may employ complex tax strategies that reduce actual tax payments, while not significantly affecting the book-tax differences. Managerial ownership (MANO) shows an insignificant relationship in simpler models but becomes significant in Model 3 (β = 0.0042, p < 0.01) when interactions are introduced.

Board gender diversity (FEMD) shows no significant direct effect on BTD, consistent with its impact on CETR but differing from its significant positive relationship with ETR. However, the moderating effects reveal that the presence of females on the board significantly influences the relationship between managerial ownership (β = −0.0044, p < 0.05) and foreign ownership (β = −0.0080, p < 0.05) with tax avoidance. Specifically, female board presence strengthens the negative association between both managerial ownership and foreign ownership with tax avoidance. These findings highlight the governance role of gender-diverse boards in enhancing tax transparency and accountability.

On the other hand, the interactions between family or institutional ownership and board gender diversity were insignificant when using the BTD model. This may be explained by the fact that the BTD model reflects technical accounting differences influenced by factors beyond governance and ownership decisions, while the ETR and CFETR models capture actual cash taxes paid, which are more directly affected by the interactions between ownership and board gender diversity in strategic tax planning.

Among control variables, firm age (FAGE) consistently demonstrates a significant negative relationship with BTD (β = −0.0005, p < 0.05), suggesting that older firms engage in less tax avoidance, a result not captured in our main models. These results strengthen our understanding of the relationship between ownership structure, board gender diversity, and tax avoidance by demonstrating consistent patterns across different tax avoidance proxies while revealing measure-specific nuances (Armstrong et al., 2015; Hanlon & Heitzman, 2010).

10. Conclusions

This study investigated the effects of different types of ownership—namely family, institutional, managerial, and foreign—on corporate tax avoidance practices. It also explored the moderating role of female board members in shaping these relationships. This study utilized a dataset consisting of 72 industrial and service sector firms listed on the Amman Stock Exchange, with a total of 432 observations. The results indicate a positive relationship between family ownership and tax avoidance practices. This may be due to the tendency of controlling family shareholders to use tax avoidance strategies to gain additional benefits from rent extraction. The presence of foreign investors also increases tax avoidance practices, as foreign investors may put greater pressure on managers to achieve high profits by manipulating tax liabilities. Foreign investors may shift profits across countries, especially in multinational firms, to reduce the tax burden. Moreover, the impact of females on the board of directors on the relationship between family and foreign ownership and tax compliance was negative. In family and foreign firms, control over decision-making rests with families and foreign shareholders. This dynamic limits the influence of females, making them insufficient to influence the direction of the board or its effectiveness in curbing tax avoidance practices. Conversely, the results revealed that institutional and managerial ownership exhibited a negative relationship with tax avoidance. Both tend to weigh the costs and risks of tax avoidance against the potential benefits, preferring compliance to avoid legal or reputational risks. The presence of female directors also positively moderated the relationship between institutional and management ownership and tax compliance. Female directors may encourage management strategies that prioritize social responsibility and compliance with legal regulations, leading to greater tax compliance overall.

The study offers several important implications: First, policymakers and regulators are encouraged to adopt more targeted corporate governance reforms that address the risks associated with concentrated ownership, particularly by family and foreign investors. These reforms could involve enhancing transparency in tax related matters by requiring firms to publicly disclose their tax strategies and intercompany transactions. Such measures would help improve oversight and reduce opportunities for aggressive tax avoidance. Second, imposing limits on the voting rights (caps on voting rights) held by a single shareholder or group of shareholders would reduce the concentration of control, ensure more balanced decision-making, and mitigate the risks of aggressive tax planning practices. Third, regulators should continue to monitor and evaluate the implementation of the 2024 corporate governance guidelines, which mandate a minimum of 20% female representation on boards. Ensuring firms’ compliance with this requirement is essential, as balanced gender representation enhances transparency, promotes ethical practices, and reduces tax avoidance. Additionally, awareness and training programs should be reinforced to support women in leadership roles, while fostering a culture of inclusion and diversity within organizations. Fourth, building on recent digital reforms, such as the National Electronic Billing System and AI-driven risk-based auditing, tax authorities should continue to enhance and expand these technologies and update tax laws further to close remaining loopholes. This will strengthen tax compliance and more effectively curb aggressive tax avoidance practices. Fifth, shareholders, especially institutional investors, should play a more active role in overseeing management decisions related to tax planning by promoting good governance practices and supporting board diversity, as these contribute to limiting tax avoidance behaviors and ensuring long-term financial sustainability. Finally, the government should implement public awareness campaigns to promote a culture of tax compliance and educate taxpayers and businesses about their tax obligations and the consequences of tax evasion and avoidance.

This study is subject to several limitations. First, the study focuses on a specific geographic region, Jordan, and on specific sectors, namely industrial and service firms, which may limit the generalizability of the results to other regions or industries with different governance structures and cultural dynamics. Second, the study assumes that the impact of female board members on tax avoidance practices is the same, without considering differences in leadership skills, experience, and educational background among female board members.