The CEO’s foreign experience and the CEO’s share ownership: Does tax aggressiveness matter?

How to cite this paper: Pebriyanti, D. M. D. S., Firmansyah, A., Wijaya, S., & Irawan, F. (2022). The CEO’s foreign experience and the CEO’s share ownership: Does tax aggressiveness matter? Journal of Governance & Regulation,


INTRODUCTION
Taxes are the largest source of income for the state employed to finance state expenditures (Ayem & Setyadi, 2019). Therefore, the Government expects that business actors such as companies pay more taxes so that state revenues will also be higher. However, in accounting, taxes are an expense for the company, which will reduce net income. It causes companies to intend to pay taxes to a minimum, so companies need to carry out specific strategies to reduce their tax expenses (Pratiwi & Ardiyanto, 2019). The company's strategy to reduce taxes is tax aggressiveness (Luke & Zulaikha, 2016). This strategy is usually carried out by taking advantage of loopholes in tax regulations, commonly known as the gray area (Charisma & Dwimulyani, 2019). Tax redu by planning tax aggressiveness is cing the amount of tax paid legally and illegally by taxpayers, both personal and corporate taxpayers (Nugraheni & Murtin, 2019). Companies commonly employ tax planning to increase debt because the debt they have will cause interest expenses, and this interest expense can reduce company profits so that tax payable will decrease (Nugraheni & Murtin, 2019). In addition to increasing debt, companies can carry out tax aggressiveness by taking advantage of Indonesia Income Tax Regulations concerning non-tax objects by generating non-taxable income; for example, a company can invest in companies in Indonesia with ownership of at least 25%, so that dividends received by the company will be categorized as income that is excluded from the tax object (Luke & Zulaikha, 2016). Indonesia is an emerging country with various economic sectors. In addition, as a G-20 member-country, Indonesia adopted several provisions enacted by G-20, including tax policy. Nowadays, the Government attempts to promote its tax ratio to catch up with the other countries. The Indonesia Tax Authority cooperates with the OECD member countries to cope with similar problems in tax administration, such as international tax avoidance and tax aggressiveness.
There have been many companies that practice aggressive actions against taxes. One example of the latest case occurred in the United States (US). On Monday, October 19, 2020, the Chief Executive Officer (CEO) of software company Reynolds & Reynolds, Robert T. Brockman, was indicted for the US's most significant tax evasion case, which was worth US$ 2 billion. He has been doing this tax evasion for 20 years (Damayanti, 2020). In the US, cases of tax evasion also occur in Indonesia as an emerging country. A company operating in the health services sector, an affiliated company in Singapore, PT RNI, is suspected of carrying out tax evasion efforts. PT RNI has been registered as a limited liability company. However, in terms of capital, the company relies on affiliate debt for a living. Owners in Singapore provide loans to RNI in Indonesia. The owner does not invest but pays debt because the capital is included (reducing taxes). This company can practically avoid tax obligations (www.kompas.com). The practice of tax avoidance by PT RNI attempts to reduce profits by increasing loans because the interest from paying debts can reduce taxes.
Tax aggressiveness is an action that may only be carried out unilaterally by management without shareholders' approval. It can benefit management because the company can earn profits shortly, not long-term profits according to shareholders' wishes (Yuwono, 2019). For shareholders, it is an action that can harm them because the company will face the tax authorities, considering that such action is considered detrimental to state revenues from the taxation sector. If the company is caught doing tax evasion, then this action can harm the company's sustainability, which impacts shareholders as owners of the company. The condition in which the management carries out tax aggressiveness for personal gain occurs because of agency theory, namely a contract between one or several company owners (principals) that authorizes management (agents) to make decisions in running the company (Jensen & Meckling, 1976). Efficient contracts occur when the management and owners of the company have symmetrical information. However, symmetrical information never occurs (Hestanto, 2020).
The factors influencing tax aggressiveness are crucial because tax aggressiveness can cause considerable losses to society (Lanis & Richardson, 2012). Tax aggressiveness carried out by company management can also harm shareholders because of the potential costs that must be incurred, namely in the form of tax penalties and reputation costs (Kurniawan & Syafruddin, 2017). If management's tax aggressiveness does not obtain attention, then the losses incurred for the state and shareholders will be even worse. Based on previous studies, tax aggressiveness is mainly associated with activities or policies carried out by management. Therefore, examining the characteristics or involvement of management in tax aggressiveness is interesting to do. Management has characteristics that will influence decisionmaking (Sofiati & Zulaikha, 2018). One example of the characteristics of management can be seen from the mindset. The mindset of management can be formed from the experiences they have experienced. Therefore, it can be concluded that management experience will affect the mindset and, ultimately, affect decision-making, including tax aggressiveness. In addition to management characteristics, the company's characteristics also affect tax aggressiveness, and one example is the company's share ownership. Company shareholders can come from institutional, managerial, or public sources (Nugraheni & Murtin, 2019). Managerial ownership is when a company manager owns company shares (Hadi & Mangoting, 2014). Managerial ownership reduces tax aggressiveness because the goals are aligned between managers and shareholders, so-called increasing the company's stock price (Winata, 2014).
The CEO's experience can shape the mindset that influences the decisions (Kwalomine, 2017). It will be different depending on where the experience is obtained (Kwalomine, 2017). The experience gained by the CEO can derive from educational or career experiences obtained from within the country or abroad (Hambrick & Mason, 1984). It influences the director's decision on corporate tax aggressiveness (Wen, Cui, & Ke, 2020). Research on the effect of a director's foreign experience on tax aggressiveness has been conducted in China by Wen et al. (2020). The study concluded that director with overseas experience performs no tax aggressiveness actions. Research examining the effect of a director's foreign experience on tax aggressiveness has never been conducted in Indonesia. Therefore, it is essential further to investigate the effect of the director's foreign experience.
The shares ownership by directors causes them to have the same goal as shareholders, namely to obtain profits from investments (Lubis et al., 2018). Therefore, the director will be more prudent in managing the company not to cause losses to the company and maintain the company's sustainability. One of the profits from the investment obtained by shareholders comes from the increase in the company's share price. If a company is proven to have carried out tax aggressiveness that violates the rules, the company's reputation will be damaged and risk a decline in the company's stock price. Therefore, share ownership by directors tends to reduce tax aggressiveness. However, in academia, managerial ownership on tax aggressiveness is still a matter of debate. Meanwhile, in this study, managerial ownership is proxied by CEO share ownership divided by the total outstanding shares. The CEO is the highest position in the executive ranks who is responsible for all operational activities of the company (Sudana & Aristina, 2017) so that the CEO also has full authority or power in making significant decisions within the company is the reason this study examines the effect of characteristics and the CEO ownership on tax aggressiveness. The CEO's ownership of company shares is a measure that is not affected by changes in the characteristics of the board of directors or top management (Firmansyah et al., 2021). Suppose there is a relationship between the level of managerial ownership and the manager's obligation to the company. In that case, this relationship should be most visible when using CEO ownership as a managerial measure because the CEO is the most influential manager in a company.
This study also employs control variables, namely company size, and cash flow operation. The company's size is associated with tax aggressiveness conducted by large companies. Many companies with significant assets indicate high company operating activities, so profits and taxes also increase (Nurdiana et al., 2020). While operating cash flow is related to tax payments included in the company's operating cash flow. According to Hery (2016), a high operating cash flow ratio indicates high corporate profits, encouraging companies to do tax aggressiveness.
This research is expected to contribute to completing the literature related to testing the characteristics and involvement of management on tax aggressiveness which is still being debated and adding to the literature related to testing the relationship between the experience of CEO's foreign experience and CEO's ownership on tax aggressiveness. In addition, this research is expected to contribute to the government in improving policies or supervision to increase tax revenue from corporate income tax. The Indonesian Tax Authority may provide a more sophisticated tax administration to promote tax revenue and tackle tax avoidance. In a more detailed manner, it can be elaborated as follows. The Tax Authority classifies the taxpayers based on their business characteristics and ownership, especially foreign and listed companies. Next, identify the CEO's experiences and analyze the firms' policy in providing remuneration. In other words, a deep understanding of qualitative information from financial reporting is needed.
We elaborate this research into six sections. In Section 1, we describe the phenomena that occur. Then, we formulate the research question and objective of the research. We also compare the previous studies in order to decide variables that will be deployed. Section 2 contains the references that we used to build the research framework and establish hypotheses. Section 3 is the research methods, including the data observation and research model. Section 4 is the description of all results, both descriptive statistics, and inferential statistics. To wrap up the discussion, we put all of the research findings in Section 5. The last is Section 6, which summarizes all parts of this paper, including the limitation and implications. In addition, we recommend some perspectives for future research. Jensen and Meckling (1976) explained that an agency relationship is a contract between one or more people (the principal) and another person (the agent) which contains an order by the principal to the agent to carry out a job, and the agent is given the authority to make the best decisions in the interests of the principal. In the financial management framework, agency theory describes the relationship between shareholders and management. Management is an agent authorized by the shareholders (principals) to manage the company. Both parties to the contract have the same goal of maximizing each other's utility with their information. One of the factors that must be met to create an efficient contract is that the agent and principal have symmetrical or in the same quality and quantity (Hestanto, 2020). Managers have more information because it is an internal part of the company. When managers have more information, it will open up opportunities to take actions that can benefit them. This statement is supported by Eisenhardt (1989), who expressed three assumptions of human nature, namely: 1) self-interest, humans are generally selfish, 2) bounded rationality, humans have limited thinking power regarding future perceptions, 3) risk-averse humans always try to avoid risk. The first assumption of human nature supports that it will benefit when managers have more information. In addition, shareholders find it difficult to monitor the activities carried out by managers. This strategic situation increasingly triggers managers to take policies that can maximize their utility.

Agency theory
This asymmetric information is what causes agency problems. Scott (2015) stated that asymmetric information generates adverse selection and moral hazard. Adverse selection is a type of asymmetric information wherein, in a business transaction, one of the parties conducting the transaction has more information than the other party does. Adverse selection occurs when the manager, as an internal party of the company, knows more about the condition of the company and the company's prospects in the future than the shareholders. Likely, managers do not convey facts that could influence decisions to shareholders.
Meanwhile, moral hazard is a condition in which one of the parties takes advantage of the ignorance of the transaction partner to take actions that violate the contract, which may not be ethically feasible. Zhou and Xu (2019) argued that asymmetric information is something inevitable in business operations. Moreover, asymmetric information can also be considered two perspectives where control can be separated, but not for management. On the other hand, Ranaldo and Somogyi (2021) explained that if several people own important information concerning asset value, then there is the possibility to sell the information to the market. Guo and Hung (2020) emphasize that asymmetric information can also affect the credit markets.
Consequently, it can also affect the possibility of tax evasion conducted by companies. Desai and Dharmapala (2006) explained that tax aggressiveness is a form of agency problem where management manipulates company profits so that the tax expenses borne by the company are reduced to maximize its benefits. This manipulation can occur because of asymmetric information between management, which gives them authority to prepare financial statements and shareholders as users of financial statements.

Hypotheses development
Based on the agency theory, conflicts of interest between managers and shareholders occur due to information asymmetry in which managers who are an internal part of the company have more information than shareholders do. This situation opens opportunities for managers to take policies that can maximize their benefits. One form of conflict of interest between managers and shareholders is reflected in the aggressiveness of corporate taxes (Desai & Dharmapala, 2006). Tax aggressiveness is a tax planning activity that utilizes the gray area of tax regulations to minimize tax expenses and increase company profits. When the management responsible for preparing the financial statements is faced with disclosing the actual amount of tax expense and company profits but does not get any reward or arranges a strategy so that the tax expense is lower and company profits are high. Management obtains bonuses, and then management will choose a decision that can maximize profits for his gain. This opinion assumes agency theory, namely that each individual is more concerned with himself (Jensen & Meckling, 1976). Therefore, management will always choose to do tax planning. Nevertheless, some management chooses to reduce corporate tax aggressiveness and even chooses to pay additional taxes. This statement is supported by the research of Wen et al. (2020), who found that the UK companies agreed to pay millions of dollars in additional corporate taxes to avoid embarrassment and reputational damage.
Thus, the policies taken by managers also depend on the characteristics of the managers themselves. The experience of managers, especially the CEO, will significantly determine the decisions or strategies chosen to run the company's business and affect organizational outcomes (Hambrick & Mason, 1984). The CEO's foreign experience, work experience, and study experience influence the CEO's decisions, including tax planning decisions. CEOs who return from abroad may not want to get involved with the mass media regarding tax issues that can damage the company's reputation, so they choose to reduce the level of tax aggressiveness. Desai and Dharmapala (2006) supported this statement, which stated that reputation costs limit the extent to which companies will undertake tax aggressiveness because reputation costs will impact the CEO personally and the company. In addition, Bankman (2004) suggested that companies that are very aggressive towards taxes will be labeled -poor corporate citizens‖ to affect company performance negatively.
Furthermore, a survey by Graham, Hanlon, Shevlin, and Shroff (2014) found that 69% of companies penalized for being tax aggressive could damage the reputation of returnee directors. Wen et al. (2020) also concluded that the director's foreign experience has a negative effect on tax aggressiveness. It means that returnee directors choose not to be too aggressive to maintain their reputation and reputation. Companies that employ returnee directors may have low or no tax aggressiveness. Therefore, the first hypothesis in this study is as follows: H1: The CEO's foreign experience is negatively associated with tax aggressiveness.
According to the agency theory, agency problems occur due to a conflict of interest between the agent and the principal. The separation of authority between management and shareholders will allow management to maximize profits to maximize its interests due to asymmetric information between management and shareholders who cannot monitor management actions (Lubis et al., 2018). To maximize profits, management needs to do tax planning or what is called tax aggressiveness. Therefore, tax aggressiveness is a form of the agency problem. According to Desai and Dharmapala (2009), tax aggressiveness minimizes the tax expenses and covers their activities that can harm shareholders. As long as management and shareholders still have different functions, management will take advantage of tax aggressiveness to obtain maximum personal gain without thinking about the risks that will be faced for the company and shareholders as owners of the company.
Agency theory states that agency problems can be reduced by managerial ownership. Herdianti and Husaini (2018) stated that managerial ownership is the percentage of shares owned by the board of directors and commissioners as top management. Managerial ownership causes managers also to act as shareholders, so managers will be more careful in making decisions. If there is no managerial ownership, managers will make decisions that can maximize their profits without thinking about the impact on shareholders.
Thus, managerial ownership is related to tax aggressiveness. Pramudito and Ratna Sari (2015) stated that managerial ownership could reduce the level of tax aggressiveness. Management is now the manager of the company and the company owner so that management will be more careful in making decisions related to tax planning for the sake of the company's survival. This opinion is also in line with Oktaviana and Wahidahwati (2017), examining the effect of share ownership on tax aggressiveness. This study concludes that managerial ownership has a negative effect on tax aggressiveness, which means that the more managerial ownership, the lower the level of tax aggressiveness. The results of Wijaya and Saebani's (2019) research also conclude that managerial ownership can reduce tax aggressiveness. This research is also in line with Boussaidi and Hamed-Sidhom (2015) and Rahmawati, Endang, and Agusti (2016). Thus, managerial ownership is considered to align the interests between management and shareholders. If in a company there is management who is also included in the shareholder component, the level of tax aggressiveness of the company will be low. Therefore, the second hypothesis in this study is as follows: H2: The ownership of the CEO is negatively associated with tax aggressiveness.

RESEARCH METHODOLOGY
This study employs secondary data obtained from the company's financial statements and annual reports from 2016 to 2019. The financial statements and annual reports are attained from the official website of the Indonesia Stock Exchange (IDX) (www.idx.co.id) and the company's official website. The sample of this research is non-financial sector companies. We use the observation period starts in 2016 because during the year, the government of Indonesia enacted the tax amnesty law. Since the previous fiscal year, 2015, the policy showed a low percentage of tax revenue. Thus, the starting period used in this research, 2016, can give companies' financial statements a clear condition. The number of observations in this study should be 116, but after re-examining, there are 20 data whose total ownership of the main director is 0% so that 20 data must be removed. Furthermore, this study uses an effective tax rate (ETR) with a range of values from 0-1 (Lanis & Richardson, 2012), so that an ETR with a negative value is excluded from this study. The number of ETR with negative values is 8 data so that the final number of observations in this study is 88 observations. Table 1 is a summary of the sample criteria based on purposive sampling. The dependent variable used in this study is tax aggressiveness, while the independent variable used is the CEO's foreign experience and CEO's ownership, and the control variables are company size and operating cash flow. This study measures tax aggressiveness with two proxies. The main proxy employed is the effective tax rate (ETR), which refers to the calculations made by Hanlon and Heitzman (2010). The value of ETR with the level of tax aggressiveness has an inverse relationship, which means that a high ETR value indicates a low level of tax aggressiveness. On the contrary, if the ETR value is low, the tax aggressiveness is high. The opinion of Lanis and Richardson (2012), which states that ETR is the most widely used proxy in previous studies to describe tax aggressiveness, is the reason for using ETR as the main proxy in this study. In addition, compared to other proxies, ETR can capture tax aggressiveness through transactions such as overseas sales and diversion of income from high-tax jurisdictions to low-tax jurisdictions, where these actions can reduce a company's ETR (Lanis & Richardson, 2012). Dyreng, Hanlon, Maydew, and Thornock (2017) also stated that ETR could capture all acts of tax aggressiveness that take advantage of loopholes in tax regulations, either legally or illegally. Therefore, this study employs the ETR proxy, which has also been used by several previous researchers, namely Boussaidi and Hamed-Sidhom (2020), Lanis, Richardson, and Taylor (2017), Deslandes, Fortin, and Landry (2019) with the following formula: (1) Meanwhile, this study uses additional proxies to test the sensitivity of the research results. The additional proxy used in this study is the abnormal book tax difference (BTD) found by Desai and Dharmapala (2006). It refers to previous research by Nurhandono & Firmansyah (2017), which measures tax aggressiveness using abnormal BTD with the following formula: (2) where, is the difference between accounting profit and corporate tax profit i in year t; is total accruals of the company i in year t; , are residual of the regression equation.
The residual regression equation is obtained by finding the value first with the following formula: Meanwhile, the company's total accruals were obtained from: (4) where, is net income of the company i in year t; is the operating cash flow of the company i in year t.
All of the above variables are divided by the total assets of the previous period. The value of tax aggressiveness is the residual value of the crosssectional regression from 2016 to 2019.
The CEO's foreign experience is measured using a dummy variable, which refers to research by Giannetti, Liao, and Xu (2015), Yuan and Wen (2018), and Wen et al. (2020). The dummy variable used to measure the CEO's foreign experience in this study is 1 to certain companies with CEO's experience working or studying abroad and 0 to companies whose CEO has no overseas experience. Data related to the experience of the CEO is obtained from the company's annual report.
The proxy of CEO's ownership in this study refers to Noval (2015), La Masinonda (2017), and Sudanda and Aristina (2017) that is adjusted by company share owned by CEO's ownership so that the formula used is as follows: The company's total assets can measure company size by calculating the logarithm of total assets (Hartono, 2015). Therefore, the formula used to calculate firm size is an indicator of total assets according to Hartono (2015) and has also been used by Lanis  ( ) (6) This study employs operating cash flow because tax payments are included in the company's operating cash flow. According to Hery (2016), a high operating cash flow ratio indicates high company profits, encouraging companies to do tax aggressiveness. According to Gazali, Karamoy, and Gamaliel (2020), the operating cash flow can illustrate whether the amount of cash generated from the company's operations can generate sufficient cash flow to pay taxes. The calculation of CFO in this study refers to research by Purwanti (2014), Utama, Kirana, and Sitanggang (2019), and Sadjiarto, Mustofa, Putra, and Winston (2019), with the following formula: Hypotheses testing is conducted by multiple linear regression with ordinary least squares (OLS). This study employs two models. These models test the association of the CEO's foreign experience and the CEO's share ownership on tax aggressiveness. The main model in this study uses ETR to measure tax aggressiveness, while the additional model measures tax aggressiveness with abnormal BTD. The following are the main regression models in this study: (8) Meanwhile, the additional regression model in this study is as follows: (9) where, ETR is effective tax rate of the company i in year t; TRA is residual value of the abnormal BTD of the company i in year t; CFE is the CEO's foreign experience of the company i in year t; MOW is CEO ownership in the company i in year t; SIZE is natural logarithm of total assets of the company i in year t; OCF is operating cash flow of company i in year t. Table 2 shows the descriptive statistics for each variable in the main model and the second model employed in this study. Descriptive analysis in this study employs a description of the size of the data concentration and the size of the data spreadmeasures of data concentration in the form of mean and median. The mean is another terminology for the average, the ratio between the amount of data and data. Meanwhile, the median or the middle value is the data value that lies in the middle of the data set is sorted from smallest to largest. Furthermore, the size of the data spread consists of maximum, minimum, and standard deviation.   Table 3 shows the hypothesis test conducted through OLS. The test is carried out with the goodness of fit test, F-test, and T-test. The goodness of fit test describes the independent and control variables' ability to explain the variation of changes in the dependent variable. The F-test is used to determine the effect of all independent variables simultaneously on tax aggressiveness. Meanwhile, the T-test was used to determine the effect of the independent variable individually or partially on the dependent variable.  The result of hypothesis testing with the primary model in this study indicates that the CEO's foreign experience is positively associated with ETR. This result suggests that the corporate tax aggressiveness led by the returnee CEO will be lower because the ETR value with tax aggressiveness has an inverse relationship. Thus, the CEO's foreign experience has a negative effect on tax aggressiveness. The test result of this study is in line with Wen et al. (2020). The result of the second model suggests that the CEO's foreign experience is not associated with tax aggressiveness. However, the primary model will then be employed to analyze the result of hypothesis testing. Based on the descriptive statistical test, the number of CEOs with overseas experience in this study is 53 or 60.2% of the observations, which means CEOs with overseas experience had the dominant number. The data can be associated with the descriptive statistics of ETR with an annual average of more than the prevailing tax rate in Indonesia (25%). Only in 2016, the average ETR was below 25%. Thus, the companies in this study tend not to do tax aggressiveness.

RESULTS
In managing the company, the director is responsible for its shareholders and third parties with legal relations (Zaki, Ginting, Devi, & Bariah, 2019). In Indonesian Law No. 40 of 2007 concerning Limited Liability Companies, Article 97, paragraph (2), states that each director runs the company's management for the company's benefit and represents the company both inside and outside the court. Therefore, the director is personally responsible if there is an error in making decisions that do not prioritize the company's interests (Firdaus & Bachtiar, 2018). Article 97, paragraph (2) of the Company Law also states that the board of directors does not conflict interest in carrying out their duties. If the directors are negligent in carrying out their duties, causing losses to the company, they are personally responsible. It is regulated in Article 97 paragraph (3) of the Company Law. Therefore, the directors are not allowed to take advantage of themselves over the company.
Furthermore, the decision of the returnee CEO will also affect the tax regulations that apply in Indonesia. Currently, the tax collection system in force in Indonesia is a self-assessment system. In the self-assessment system, the taxpayer's amount of tax payable is determined (Mardiasmo, 2016) so that the taxpayer himself carries out activities from calculating, depositing, to reporting the tax payable. The implementation of the selfassessment system is one form of tax reform. Tax reform is a change in the tax system as a whole, including reforming tax administration, improving regulations, and increasing the tax base to reduce tax evasion and manipulation (Direktorat Jenderal Pajak, 2018). The government's implementation of the self-assessment system is exemplary and positively influences taxpayer compliance (Lasmaya & Fitriani, 2017). In addition to implementing the self-assessment system, one form of tax reform implemented is the modernization of the tax administration system. Modernization of the tax administration system can make taxpayers deposit and report the actual tax payable (Fajarwati, Kertahadi, & Kurniawan, 2014). Efforts to modernize the administrative system and technology make it easier to detect tax evasion coupled with the increased probability of becoming the object of tax audits because better technology encourages taxpayers to report the correct amount of tax (Broto, 2018). This condition is in line with the findings by Wen et al. (2020), which stated that the decision of returnee directors related to tax aggressiveness affects the regulations in force in China.
In addition, returnee directors are more competent in observing the costs and benefits of tax aggressiveness (Wen et al., 2020). Costs related to tax aggressiveness include reputation costs and acceptable sanctions (Kurniawan & Syafruddin, 2017). Reputation costs are related to the company's reputation and the reputation of the CEO (Desai & Dharmapala, 2006). The reputation of returnee directors was damaged because the company was caught in a legal case related to tax evasion. Therefore, returnee directors prefer not to do tax aggressiveness (Graham et al., 2014).
Further analysis, decisions related to tax aggressiveness also depend on the pattern of filling the position of the CEO. In Indonesia, filling the position of CEO can be through political connections. Political connection is the level of closeness of the company to the government (Pranoto & Widagdo, 2015). In practice, political connections are common in Indonesia, namely by filling the positions of commissioners or directors by people close to the government (Pranoto & Widagdo, 2015). However, the selection of directors and commissioners in Indonesia also pays attention to the competence of the candidates who will occupy positions by conducting competency tests. This test is a concern for the government so that the positions of directors and commissioners are occupied by people who can make a positive contribution to the state, one of which is through the contribution of tax payments to the state (Pranoto & Widagdo, 2015). In addition, the Indonesian Tax Authority provides awards to companies that pay the most considerable taxes, which provides a positive image to the company (Wibisono, 2018). Of course, it will also provide a positive image for company directors to reduce aggressive actions. The tax so that the tax paid is more outstanding.
Furthermore, the decision of returnee directors related to tax aggressiveness is also related to how investors supervise management activities in their home country (Wen et al., 2020). Supervision of management activities by investors related to corporate governance. Good corporate governance improves communication between management and shareholders, making it easier for shareholders to monitor management activities (Hapsari, 2018) to limit management from taking opportunistic actions. The implementation of corporate governance in Indonesia shows a good trend because almost all companies that have been listed on the IDX have implemented it (Pranoto & Widagdo, 2015). Therefore, CEO directors choose not to be tax aggressive. The result of this test does not confirm agency theory but tends to be in line with stewardship theory. In stewardship theory, management is assumed to be motivated to act in the principal's best interests and prioritize the company's state as a whole (Davis, Schoorman, & Donaldson, 1997; Zahra, Gedajlovic, Neubaum, & Shulman, 2009). Long-term interests prioritize management attitudes in this theory and even above personal interests (Hernandez, 2012). Therefore, returnee directors can carry out the mandate of principals who focus on company goals rather than their own goals (Donaldson & Davis, 1991;Hernandez, 2012 Of the total sample of 29 companies, only four companies experienced a significant increase in the CEO's ownership percentage. The low ownership of the CEO indicates that the CEO in this sample is not a shareholder who has significant ownership so that the CEO cannot influence decisions related to tax planning. The majority shareholder usually influences decision-making, including decisions related to tax planning that tend to take aggressive actions (Pratiwi & Ardiyanto, 2019).
The agency theory states that conflicts of interest occur due to the separation of authority between the principal and the agent (Hestanto, 2020). Furthermore, this conflict of interest can be overcome by managerial ownership because managerial ownership can align the position between management and shareholders (Firmansyah et al., 2021). Therefore, when making decisions, management will put themselves in the position of shareholders, so management chooses to reduce tax aggressiveness for the sake of the company's survival. However, the test results of this study cannot confirm this view. The compensation they get from the company is greater than the benefits they obtain by doing tax aggressiveness. Management compensation is an award in material or nonmaterial given to managers to align the interests between managers and shareholders (Budiadnyani, 2020). One of the factors that can affect management compensation is the company's ability, where large companies can provide more significant compensation (Sutrisno, 2016). When viewed from the results of descriptive statistical tests, the average company size is 31,201 with total assets of Rp. 10.7 T, so that the average company in the sample of this study is a large company that can provide more significant compensation. In addition, management who also owns company shares does not make the decrease in tax aggressiveness a matter that must be considered because they consider that tax aggressiveness has become a concern by other management who only acts as an agent.
Humans always attempt to avoid the risk associated with its fundamental nature (Eisenhardt, 1989). Likewise, CEOs who share ownership incentives will tend to use their discretion to avoid tax aggressiveness through tax planning because they can deal with tax authorities that can harm them. However, the CEO's discretion may be used for other corporate strategies such as performance that impacts bonuses, stock market value, and dividends. Thus, this study does not confirm that the CEO can reduce the agency problem by share ownership.

CONCLUSION
The returnee CEO's decision regarding tax aggressiveness also depends on the regulations in their country of origin. The returnee CEO is more competent in observing costs and benefits related to tax aggressiveness, one of which is reputation costs. Therefore, the returnee CEO will be more careful in making decisions regarding tax aggressiveness because the costs incurred can be greater than the benefits, such as damage to the company's reputation and affect the company's continuity. In addition, the ownership of the CEO has not reduced the agency problem, tax aggressiveness activity. In addition, the CEO ownership in Indonesia is still low, so it has not influenced his decision on tax aggressiveness.
There are several limitations to this study. First, this study only examines non-financial sector companies listed on the IDX, so the results of this study cannot generalize the character of all companies in Indonesia. Future research can employ samples from the financial sector. Second, the period in this study is too short, which is only from 2016 to 2019. Future research can use a more extended research period in order to provide more comprehensive results and discussion.
The results of this study can be used as input for the Indonesian Tax Authority to pay more attention to factors that can affect tax aggressiveness, such as the characteristics of the company's CEO and the company's ownership structure. Consequently, the tax auditors and the account representative should analyze the quantitative aspect (i.e., financial statement) and the qualitative information (i.e., returnee CEO). Interview with the taxpayers should go in a profound way to understand the companies' tax planning. This study indicates that the CEO's foreign experience is good because the returnee CEO has high morale. However, it needs to be monitored further by the Indonesia Financial Services Authority (OJK) as the institution that oversees all activities in the financial services sector to monitor how the company behaves in running its business, whether the returnee directors carry out all business ethics only or adequately those related to tax aggressiveness. Therefore, the tax authorities also need to coordinate with OJK.
Furthermore, in conducting audits, the tax authorities also need to pay attention to the experience of the chief director. The tax authorities are also advised to close the gap for taxpayers to carry out tax aggressiveness by reducing ambiguity in tax regulations and implementing tax law enforcement evenly. Companies think that the costs related to company reputation are too high when they carry out tax aggressiveness. Investors can view the company profile based on the CEO experience when making decisions regarding investment in the capital market to determine how the CEO will run their company to choose an investment place that suits each investor's risk profile.