The Sustainable Development Goals (SDGs) were set to solve the problem of poverty and promote inclusive and sustainable growth by 2030 (Bappenas, 2015). To achieve this, significant resources are needed, including financial resources. In this case, taxation is one of the main sources of revenue in the Indonesian APBN. Tax revenue reaches 70% of the total government revenue.

It is important to maintain the resilience of tax revenue. It is still necessary to increase Indonesia's tax-to-GDP ratio. Indonesia's tax ratio is still below the Asia-Pacific average tax ratio of 19,1% (O.E.C.D., 2022). Indonesia's tax ratio in the period 2016-2022 is still at 9% - 11% (Damara, 2023). This figure illustrates the narrowness of Indonesia's tax base and low tax compliance in Indonesia.

Tax avoidance erodes tax contributions to the country. Tax avoidance schemes can limit the government's ability to fund public facilities, thereby hindering progress towards the SDGs (Trade & Development, 2019). Tax avoidance is an effort to minimize the amount of taxes paid through a tax planning strategy (Hanlon & Heitzman, 2010). Tax avoidance can be used as an option to minimize the tax burden listed in financial reports in a way that does not violate the law (Mardiasmo, 2018). However, the government does not want this.

Foreign direct investment (FDI) is cross-border investment where an investor based in one economy has long-term interests and significant influence in an enterprise based in another economy (O.E.C.D., 2022). FDI in Indonesia was recorded at US$20,1 million in 2021 and is the second largest in Southeast Asia (dataindonesiaid, 2022). FDI can boost economic growth in developing countries.

FDI is an important channel for technology transfer between countries, promotes international trade through access to foreign markets, and can be an important means of economic development. Today's business world is interconnected between countries. The challenge is that multinational companies may be able to minimize or double non-taxation (Indonesiagoid, 2023). Base Erosion and Profit Shifting (BEPS) is a threat to tax fairness in many countries through profit shifting, especially in developing countries.

Tax avoidance attracts policymakers' and researchers' attention. (Atwood et al., 2012)), stated that firms avoid taxes due to various factors, such as firm-internal factors (firm size, leverage, operating costs, firm performance, operations of multinational firms) as well as cross-country factors that can influence corporate tax avoidance, such as tax rates, earnings volatility, and institutional factors. The presence of foreign ownership can also influence tax avoidance (Yuanita et al., 2020) (Suranta et al., 2020). In Malaysia, as a developing country, the level of tax avoidance by multinational firms is closely related to the presence of foreign direct investment (Salihu et al., 2015). Meanwhile, in other studies, foreign institutional ownership has a negative relationship with tax avoidance ((Pujiningsih & Salsabyla, 2022); (Akbar et al., 2021); (Maisaroh & Setiawan, 2021); (Idzni & Purwanto, 2017)).

Based on the above description, this research will examine the influence of foreign ownership on the tax avoidance practices of firms in Indonesia. In addition, fundamental financial factors such as profitability, leverage, capital intensity, and firm size will be considered as control variables.

Positive accounting theory explains the hypothesis that leads management to engage in earnings management. Based on the political cost hypothesis, firms lobby the government when legal or accounting standards reduce their profits (R. L. (Watts & Zimmerman, 1978)). Based on the tax hypothesis, management makes discretionary decisions to choose accounting policies that can increase or decrease tax payments (Watts & Zimmerman, 1986). Management strives to maximize the book value of the firm, but this may affect the firm's earnings. If a firm tries to manipulate tax rules to pay less tax, it can earn higher profits. This means that profits influence the company's tax policy.

This research uses quantitative methods to process and analyze data so that conclusions can be drawn in the form of relationships between observed variables. Indonesia was chosen as the population because Indonesia's tax ratio is still below the Asia-Pacific average tax ratio, while FDI into Indonesia is recorded as the second largest in Southeast Asia.

The sampling technique used is purposive sampling. The criteria used are companies that are included in the IDX80 index category, and the company does not have a negative pre-tax income in 2018-2021. The IDX 80 Index is an index that includes the 80 most liquid stocks on the exchange. The index considers liquidity factors and fundamental factors such as financial condition and growth prospects, which are naturally of interest to investors (IDXcoid, 2021).

The proxy used in this study to calculate tax avoidance is the book-tax difference (BTD). BTD is the difference between accounting profit (pre-tax income) and taxable income as reported in the financial statements. These differences arise due to different treatment in financial accounting standards and tax regulations. ETR was not used in this study because it has weaknesses when used with panel or time series data. ETR is measured only on an annual basis, which can lead to significant volatility. ETR also does not consider temporary book-tax differences (Firmansyah et al., 2021).

The company could increase its book income to show investors that its economic performance is strong. On the other hand, companies need to report taxable income to reduce their tax burden. This can be seen through the BTD proxy, which can indicate tax avoidance. Refer to (Lietz et al., 2013); (Saragih et al., 2021) the formula for calculating the book-tax difference (BTD) is as follows,

BTD_i,t = PreTaxIni, t− then divided by Total Asetit

were,

BTD_i,t = Book Tax Difference

PreTaxIni, t = Pre-tax income (accounting based)

CurrTaxi, t = Current tax expense

StatRatei, t = Corporate income tax rate in year t

The interests of foreign parties in this research represent foreign influence. This reflects the foreign direct investment made by a firm. In this research, the foreign influence variable is measured using a proxy adopted from (Salihu et al., 2015).

First, foreign influence is measured by the proportion of foreign ownership divided by the total number of shares in the firm. The second proxy is the proportion of the board of directors. Considering that Indonesia has a two-tier system as regulated by Law No. 40 of 2007 on Limited Liability Enterprises, the board of directors is divided into a board of commissioners (supervisors) and a board of directors (management). For this reason, the second proxy is measured by the proportion of the board of commissioners in the firm. The third proxy is seen from the proportion of the board of directors in the firm.

In addition to the dependent and independent variables observed as described above, fundamental financial factors such as profitability, leverage, capital intensity, and firm size are considered as control variables.

Profitability is the ability of a company to make a profit during a certain period (Khan & Nuryanah, 2023). Profitability influences tax avoidance practices because it is used as a basis for calculating corporate income tax (Fauzan et al., 2019) (Sunarto et al., 2021). In this study, profitability is measured using the return on assets (ROA) ratio (Salihu et al., 2015) (Yuanita et al., 2020).

Leverage reflects the portion of financing that comes from debt. High leverage allows companies to incur higher interest costs. Increased interest costs will reduce corporate profits that are subject to taxation. Leverage is one of the factors that affect the tax policy of companies ((Fauzan et al., 2019); (Maharani & Baroroh, 2019)). In this study, leverage is measured by the ratio of long-term debt divided by total assets (Chen et al., 2010) (Salihu et al., 2015) (Yuanita et al., 2020).

A firm's capital investment is determined by its decision to finance itself by investing in the firm's assets to conduct its business and earn a profit. Capital intensity affects the firm's depreciation expense. Large assets result in high depreciation expenses for the firm. High depreciation expense will result in lower firm profits, so the income tax payable will be lower. In this study, capital intensity is measured by the ratio of fixed assets to total assets (Chen et al., 2010) (Salihu et al., 2015) (Yuanita et al., 2020).

Company size is a metric used to indicate the size of a company. It is important to control for variations in the level of company investment in assets with tax incentives (Salihu et al., 2015) due to the possibility of differences in load recognition times (Chen et al., 2010). A larger company size typically indicates greater funding, which in turn may lead to higher return expectations. Management's desire for maximum profits can lead to the encouragement of tax avoidance, as highlighted by (Dewinta & Setiawan, 2016). This study measures company size using the natural logarithm of total assets (Salihu et al., 2015) (Yuanita et al., 2020) to avoid extreme fluctuations in variable data compared to the sizes of other variables in this study.

To understand each studied variable, we first carry out descriptive data processing. A series of panel data estimation model tests were conducted, including the Chow test, Hausman test, and Lagrange test. These tests are used to determine the most appropriate regression model for the panel data at hand. Based on the results of the selected panel data regression estimation model, classical assumptions are tested, and hypotheses are tested.

This research employs the regression model described above.

BTD_it =

Table 3 presents descriptive statistics describing the characteristics of the research sample. Tax avoidance, as measured by the book-tax difference (BTD), has a maximum value of 0,1930 and a minimum value of -0,3672. A positive value means that the pre-tax income according to the accounting books is higher than the taxable income. The median value of 0,0006 shows that there are still more companies that have a high BTD or can be said to practice tax avoidance. The standard deviation of BTD is 0,0488 while the mean is -0,0010. The standard deviation is greater than the average, which indicates that the distribution of the book-tax difference variable is heterogeneous or varied.

The foreign ownership ratio has the lowest value of 0,0013 (0,13%), which means that all public companies included in the IDX80 index have foreign investors. The lowest value is owned by PT Kimia Farma Tbk, which is a state-owned company. The highest percentage of foreign ownership was 93,10%, namely PT Unilever Indonesia Tbk in 2020.

The highest ratio of foreign board of commissioners is 0,75, which means that 75% of the board of commissioners are foreign nationals. The highest ratio is owned by Indofood Sukses Makmur Tbk but the highest number is owned by Astra International Tbk. The average composition of the Board of Commissioners is 10,68%, which indicates that the position of the Board of Commissioners is small in Indonesia.

Meanwhile, the highest ratio of foreign boards of directors is 0,67, which means 67% of the board of directors are foreign nationals. The highest ratio is owned by H.M. Sampoerna Tbk at the same time as the highest number of foreign boards of directors. Overall, the proportion of foreign Board of Directors is even lower with an average of 9,90%.

Profitability has an average value of 0,0799 with a standard deviation of 0,730. This value indicates that, on average, the companies included in the IDX80 Index have relatively competitive profitability. A positive minimum profitability value ensures that the sample in the study does not have a negative pre-tax profit. The highest profitability is led by PT Unilever Indonesia Tbk in 2018, 2019, and 2020. This also shows that the non-cyclical consumer industry sector was able to survive the impact of the Covid-19 pandemic wave.

Capital intensity has an average value of 0,3502 and a standard deviation of 0,3127. Similarly, leverage has a mean of 0,2048 and a standard deviation of 0,1698. Both show a lower standard deviation than the average, which means that capital intensity and leverage in companies included in the IDX80 index have relatively standard values.

The firm age in the descriptive statistics is explained based on the age of the company from the time of its IPO to the year the financial report was prepared. The oldest is 39 years old in 2021, namely PT Unilever Indonesia Tbk. Meanwhile, the youngest company is PT Buyung Poetra Sembada Tbk (HOKI), which has only been registered since June 2017. In testing the regression model later, the company age uses the log value so that it does not have a value that is too unequal to other ratio variables.

The firm size in this descriptive statistical table is explained based on the total value of the firm's assets in millions of rupiah. The highest value is owned by PT Astra International Tbk in 2021 with a value of IDR 367,3 trillion, while the lowest value is owned by PT Buyung Poetra Sembada Tbk with a value of IDR 758,8 billion. In testing the regression model later, company age uses the log value so that it does not have a value that is too unbalanced with other ratio variables.

The results of testing the accuracy of the selection of the panel data regression model are presented in Table 4.

The results of the Chow test show that the probability value of the chi-square cross section is 0,0000. At the confidence level (α = 95%), it is concluded that FEM is more appropriate than CEM. Next, the results of Hausman's test show that the probability value of the random cross-section is 0,0640. This value is greater than α 0,05, so it can be concluded that it is more appropriate to use REM rather than FEM. The next step in selecting the best model is to perform a Lagrange multiplier test. Based on the Lagrange multiplier test, it can be concluded that it is more appropriate to use REM than CEM. Thus, of the three best panel regression model test results above, REM is the most appropriate to use at the α level of 0,05.

Hypothesis testing in this research was carried out after the classical assumption test was satisfied. The classical multicollinearity assumption test was carried out to show that there is no correlation between the independent variables that exceed 0,8 (Ghozali, 2017).Table 5 shows the results of the multicollinearity test, and it can be seen that the highest value is only 0,62. Therefore, the data in this study did not experience multicollinearity problems.

Based on the panel regression model that was selected and passed the classical assumption test for panel data, the results of regressing the random effects model are shown in Table 6 below. The F-statistic test results show a Prob (F-statistic) value of 0,0000 or less than the α value of 0,05. This shows that the independent variables and control variables can simultaneously influence the dependent variable (tax avoidance) in the company. This research model is suitable to be tested using regression.

Based on Table 6, it is known that the adjusted R square of this model is 0,3304. This shows that the independent variable can explain 33,04% of the tax avoidance measured by BTD, while the rest is explained by other variables outside this research.

Foreign ownership has a positive effect (α<0,05) on tax avoidance with a coefficient of 0,0464. This means that the higher the composition of foreign investors, the higher the tendency of the firm to engage in tax avoidance. These results suggest that management will maintain good corporate performance by reporting high pre-tax income but lower taxable income to avoid taxes. This is to satisfy interested owners, including foreign investors. These results are in line with previous studies conducted by (Salihu et al., 2015), (Khan et al., 2017), (Shi et al., 2020), (Alkurdi & Mardini, 2020), (Nainggolan & Sari, 2020).

Meanwhile, foreign boards of directors and foreign boards of commissioners do not affect tax avoidance. These results suggest that the number of directors or commissioners does not necessarily affect a firm's tax policy. They do not aggressively exploit the differences between tax and accounting rules. Given that the companies included in the IDX80 index are companies that are of primary interest to investors, it is even more important to maintain the company's reputation. Research by (Suranta et al., 2020) also found no evidence of the influence of foreign commissioners on tax avoidance practices. The regulations on access to financial information for tax purposes (AEOI), which came into effect in August 2017, as regulated in PMK number PMK-70/PMK.03/2018, may also support the findings of this research.

The control variable in this research, namely profitability, shows a significance value of 0,0166 (α<0,05) with a coefficient of 0,1222, which proves that the profitability ratio has a positive effect on tax avoidance. This result is consistent with the research findings of Richardson et al. (2013b) which state that the profitability ratio has a positive effect on tax avoidance practices. Meanwhile, capital intensity in this study shows a negative effect on tax avoidance. These results suggest that the lower the capital intensity, the more aggressive the tax avoidance. The capital intensity variable is related to depreciation costs, which may affect the amount of taxes paid. These results are consistent with the research of (Yuanita et al., 2020). This research cannot show that the leverage ratio, firm age, and firm size affect tax avoidance.

The results of this research show that foreign ownership has an impact on tax avoidance. The higher the value of shares or investments owned by foreign parties, the higher the book-tax difference value. This is evidence that companies with foreign direct investment may be able to minimize their tax burden due to the complexity of regulatory differences.

Meanwhile, the number of foreign directors and the number of foreign commissioners does not affect tax avoidance. This shows that the tax policy of the company does not depend on the number of directors. This research is limited to dividing the number of foreign directors by the total number of directors, but it does not consider whether the position is that of a finance director or similar positions that influence the company's tax policy. To overcome this limitation, future research could consider looking at specific positions of foreign directors or using audit committee variables that are directly related to financial reporting.

With purposive sampling, it is possible that the data obtained may not be fully representative of the population of all types of firms. Therefore, further research is needed beyond the IDX80 index.

This research may contribute to positive accounting theory. Shareholders want managers to be able to increase their wealth. For this reason, managers use policy discretion to choose accounting methods that can reduce the tax burden but still show a high book value. This research also serves as an input for the DJP in anticipating the risk of tax avoidance by foreign corporations.