Following existing corporate governance tax avoidance nexus literature, it is widely acknowledged that the primary objective of establishing a corporate entity is to optimize financial gains and generate sustainable profit. Shareholders who have invested their resources in the organization anticipate a commensurate financial gain on their invested capital [1]. Hence, the organization will maximize shareholder value, necessitating a sustainable competitive advantage to succeed in the marketplace. Corporate executives are strongly inclined to foster a sustainable competitive edge, employing various strategic measures, such as cost reduction through tax avoidance [2]. Practically, one may find it challenging to envision a high-ranking executive exerting an individual influence on the tax avoidance practices of a corporation.

The Chief Executive Officer (CEO) is typically inexperienced in tax matters and may have limited familiarity with various tax strategies [3]. However, the individual probably possesses a comprehensive knowledge of the fiercely competitive landscape prevalent within the industry and the inherent capacity for expansion to yield operational efficiencies through the realization of economies of scale. It is evident that the influence of a CEO extends to the formulation and implementation of the firm's operational and financial strategies, as well as potentially impacting the firm's tax avoidance endeavors [4]. The influence of a CEO on tax avoidance strategies can be discerned through the establishment of a prevailing tone at the top with regard to the tax-related operations of the organization [5]. In corporate governance, it is common for CEOs to strategically alter the relative emphasis placed on various functional areas within their organizations. These areas encompass critical domains such as marketing, operations, treasury, tax, and the allocation of resources toward engaging diverse advisors, both internal and external to the firm.

Moreover, it is imperative to consider how the corporate governance framework can influence the establishment of compensation incentives for the tax director, who plays a pivotal role in the firm's tax-related issues. In accordance with the scholarly findings of [6], it has been posited that the proclivity towards tax avoidance practices is subject to the intricate dynamics of the tax system, including the imperative of book-tax conformity and the perceived efficacy of tax law enforcement as perceived by key decision-makers. In the realm of corporate governance, it is imperative to acknowledge the intricate nature of decision-making processes that CEOs encounter. These challenges, which include information overload, ambiguous cues, and competing goals and objectives, necessitate a cognitive filtering and interpretation mechanism [7].

The inherent nature of individuals, encompassing their values, experiences, personalities, and various other human factors, significantly influences their subjective interpretations of business situations and subsequently shapes their strategic choices [8]. Corporate governance and tax policies profoundly influence various organizational outcomes, encompassing performance metrics, growth trajectories, investments in research and development (R and D), and tax obligations, among others. Furthermore, regarding the interplay between CEO attributes and tax avoidance, it is proposed that employing demographic indicators and observable characteristics as surrogates is a viable approach, given the intricate nature of cognitive foundations and values, which pose challenges in quantification Hambrick and Mason [9]. Moreover, the limited availability of senior executives, such as the CEO, for extensive psychological assessments further supports utilizing these proxies.

Nevertheless, it is imperative to acknowledge that demographic variables, despite their validity, present inherent limitations as metrics, rendering them incomplete and prone to inaccuracy. Hence, it is essential to acknowledge that the personality profile of the top executive, specifically the CEO, encompassing salient attributes such as gender and nationality, emerges as a robust construct that holds significant potential in prognosticating decision-making processes, especially decisions pertaining to the selection of tax optimization strategies [10]. Building upon the preceding discourse, this study delves into the intricate relationship between CEO attributes and the implementation of corporate tax planning strategies, specifically focusing on the tax avoidance channel.

This study employs the Upper Echelon Theory. This subsection discusses this theory and its significance to CEO characteristics concerning firms' tax avoidance. A detailed discussion of this theory is presented below.

The study provides some basic statistics information for both the explanatory and dependent variables of interest. Each variable is described based on its mean, standard deviation, maximum and minimum values. Table 3 displays the results of the study's descriptive statistics.

Table 2 shows that about 3.8% of the value of total assets consists of non-debt expenditures. It may be recalled that a non-debt tax shield refers to the tax advantage that a company gains from the tax deductibility of certain expenses, such as depreciation and amortization, unrelated to debt financing. From Table 2, it can be deduced that, on average, the usage of non-debt tax strategy as a tool for gaining tax advantage is barely low (3.8%). However, some firms within the sample frame used the tool up to about 50% during the period under study. The result also shows that some non-finance firms did not employ the strategy during the period under study. On the variable of CEO share ownership, the table shows that CEOs of some sampled listed non-finance firms in Nigeria own their company's share to as high as 85% during the period under investigation. When compared to the average, the descriptive statistics table reveals that most CEOs in the sample possessed only about 6% of their company's share, while at some point, some CEOs had no share ownership during the period under review.

From the descriptive statistic table, it can be deduced that, on average, only 6% of the sampled non-finance firms' CEOs are female. In comparison, about 64% of the observations showed that their company's CEOs remained in office over 3 years. Therefore, it invariably suggests that about 36% of the CEO observations indicate CEOs that stayed less than or about 3 years during the period under review. Concerning the independent variable of CEO nationality, the table reveals that about 27% of the CEO observations are foreigners or non-Nigerian descent. This outcome suggests that the bulk of non-finance firm CEOs (73%) in Nigeria are indigenous. Further, the data normality test is conducted using the Shapiro Wilk procedure module in Stata, and the result is presented in Table 3.

Having conducted the data normality test, the result presented in Table 4 suggests that the independent and dependent variables of interest, including CEO nationality, CEO share ownership, CEO gender, and non-debt tax, do not follow a normal distribution. This is ascertained from the probabilities of the z statistics: CEO nationality (natceo; 0.02065), CEO share ownership (sceo; 0.00000), CEO gender (ceosex; 0.00000), and non-debt tax (non-debt; 0.00000) seen to be statistically significant at 1%, respectively. The independent variable of CEO tenure (0.98291) is statistically insignificant and, therefore, is deemed to be normally distributed during the period under consideration. However, we proceeded to conduct regression analysis to test the study hypothesis but anchored the result interpretation on the probability values of the t statistics due to the non-normality of some of the variables (Greene 2003). Further pre-regression analysis involved the test for correlation among the variables of interest, as presented in Table 5.

In Table 5, the correlation analysis, which helped to determine the extent of collinearity (degree of association) among the variables of interest, suggests that no variable of interest is highly correlated. Clearly, from the table, it is evident that the degree of association was between 47% and 37% during the period under investigation. Specifically, we find that while CEO gender has a positive association with non-debt tax (10%), CEO share ownership also shows a positive association with non-debt tax (02%) during the period under review. In this study, CEO tenure is seen to have a negative association (-1%) with the non-debt tax variable during the period under review. The variable of CEO nationality showed a positive association (7%) with non-debt tax, while CEO tenure also revealed a positive correlation with the CEO share ownership variable. Clearly, no association is seen to be too high (80%) to warrant the consequences of highly correlated variables.

Specifically, to test the formulated hypotheses of this study, the pool ordinary least square regression analysis technique was first employed, and a diagnostic test was conducted. The test for heteroscedasticity and the variance inflation factor test were conducted in this study. Results from the heteroscedasticity test suggest the presence of heteroskedasticity, and the variance inflation factor test result shows no possibility of multicollinearity in the estimated model. However, a robust regression analysis technique was introduced to address the presence of heteroscedasticity in the pool least square regression estimate, as recommended by Gujarati [46]. The pool least square regression result is presented in Table 6.

Note: () and {} contains t-statistics and the respective probabilities Where: *** and ** represent 1% and 5% levels of statistical significance, respectively.

Regression Analysis Test (2023)

This study evaluates the effect of CEO attributes on tax avoidance of listed non-finance firms in Nigeria. The period scope of this study covers 10 years, ranging from 2012 to 2021. The dependent variable of tax avoidance is in terms of non-debt tax shield, while the independent variable of CEO attribute is regarding CEO tenure, gender, CEO share ownership, and CEO Nationality. The need to conduct some pre-regression analysis, descriptive statistics, correlation analysis, and normality of data analysis is of interest to the researcher to substantiate the credibility of the data set employed for analysis. Next, panel least square regression analysis was conducted, and diagnostic tests were carried out on the estimates. Diagnostic tests performed on the estimates include tests for multicollinearity and homoscedasticity. In addressing the problem of heteroscedasticity noticed in the regression estimates, robust regression was employed while the t-statistics values of the estimates were carefully interpreted in response to the issue of non-normality that the data set faced. The findings summary are as follows.

I. CEO tenure has a statistically insignificant effect on tax avoidance for listed non-finance firms in Nigeria. II. CEO share ownership has a statistically significant negative effect on tax avoidance for listed non-finance firms in Nigeria.

III. CEO gender has a statistically significant positive effect on tax avoidance for listed non-finance firms in Nigeria.

IV. CEO nationality has a statistically significant positive effect on tax avoidance for listed non-finance firms in Nigeria.

Theoretically, a firm's tax liability is proportionally related to its profitability. However, attaining a firm's wealth maximization objective through diverse means of increasing profitability poses more challenges to reducing its tax liability. Effective tax planning is strategies that maximize the firm's expected discounted after-tax cash flows. Apart from being knowledgeable about tax laws, the tax consultants of any organization should have extensive knowledge of the company, its history, and how the organization operates. It extends to coordinating parties with diverse interests and information, involving domestic and foreign operations across multiple business segments, including finance and financial reporting, management, and technology. Effective tax planning strategies should produce benefits in terms of wealth creation for the company. Hence, tax planning is a subset of a company's overall financial plan, which needs to consider the company's investment, financing, and wealth-building strategies. The majority of Nigerian firms are highly financed through equity, thus unable to enjoy the benefits of tax shield concerning CEO attributes. From the outcomes of this study, it can be concluded that CEO attributes with respect to CEO tenure, CEO gender, and CEO nationality have significant effects on tax avoidance in Nigeria.

The following recommendations are made based on the findings:

I. Based on the regression results, which indicate that increasing CEO share ownership negatively affects tax avoidance practices in Nigeria, the policy implication is that solely relying on CEO share ownership may not be an effective approach to optimizing the non-debt tax saving strategy. To address this issue and optimize the non-debt tax saving strategy, management must develop a culture that values responsible tax planning and encourages CEOs to consider the non-debt tax shield as a strategic tool. The management team may further promote transparency and open communication about tax strategies, ensuring that tax planning is integrated into overall business decision-making processes. This may also require management to encourage collaboration between CEOs, finance teams, and tax experts to identify opportunities for optimizing the non-debt tax saving strategy.

II. Based on the regression results, which indicate that female CEOs tend to deploy more of a non-debt tax shield saving strategy in Nigeria, the policy implication is that promoting gender diversity in CEO positions can be advantageous for maximizing non-debt tax shield saving. To address this outcome and leverage the potential benefits, firm management must encourage gender diversity and inclusion in CEO Hiring. In this regard, firm management must actively promote and prioritize gender diversity and inclusion in CEO hiring processes. Establish diversity goals and strategies that aim to increase the representation of women in top leadership positions, including the CEO role. Implement policies and practices that ensure a fair and unbiased selection process, giving both male and female candidates equal opportunities and creating an inclusive culture that values diverse perspectives and encourages the participation and advancement of women in leadership roles.

III. The further outcome suggests that hiring non-indigenous CEOs is a tool for maximizing the non-debt tax shield saving strategy. The policy implication is that considering non-indigenous CEOs in executive hiring can be advantageous for tax planning purposes. In leveraging the potential benefits of this outcome, management must embrace diversity in CEO hiring and foster cross-cultural expertise. Policy actions should also actively embrace diversity in CEO hiring processes by considering non-indigenous candidates and seeking out CEOs with diverse cultural backgrounds, experiences, and perspectives. This approach can help foster cross-cultural expertise within the leadership team, including a deeper understanding of tax regulations and planning opportunities in different jurisdictions, and ensure that the selection process remains fair, transparent, and based on merit, allowing for the identification of the most qualified candidates, regardless of their nationality.

Akaninyene Akpan Arthur led the conceptualization, design, and data collection for the study. Jephtar Uviefovwe Ohwoekevwo contributed to the methodology, statistical analysis, and interpretation of findings. Iniobong Udeme Asuquo participated in the literature review, drafting, and critical revisions of the manuscript. All authors collaboratively reviewed and approved the final version of the paper.

This research received no specific grant from any funding agency in the public, commercial, or not-for-profit sectors.

The data used for this study are derived from publicly available records of the selected production firms and can be made available upon reasonable request to the corresponding author.

The authors declare no conflicts of interest in relation to this research.