The Effect of Probability and Good Corporate Governance (GCG) on Tax Avoidance in Listed Industrial Sector Companies in Indonesia Stock Exchange (BEI)

The purpose of this research are to find empirical evidence of Good Corporate Responsibility (GCG) variabel, that consists of a board of commissioners and institutional ownership, also profitability variable can effect the tax evasion variable, which is the object of research are industrial corporate sector that listed on the Indonesia Stock Exchange (IDX). To determine the sample for this research, purposive sampling methode is used for. The kind of data from this study are secondary data which is collected from prospectuses, financial reports


INTRODUCTION
Tax is a mandatory contribution charged to taxpayers to be deposited into the state treasury owed by individuals or coercive entities with no direct compensation.In fact, taxes are used by the government for state purposes and national development as well as equitable distribution of public welfare.However, during the Covid-19 pandemic, several sectors have changed due to limited activities that have an impact on the country's economic system, without exception in the tax sector.Currently, the phenomenon of government revenue in the tax sector can be said to have not been maximized.And this can be seen from the realization data of the 2020 State Budget, where the realization of tax revenue was recorded at Rp. 1,072 trillion or 19.6% compared to the realization in 2019.This realization is 89.4% of the state budget target from Presidential Decree 72 or there is a shortfall of around Rp126.7 trillion (Ministry of Finance, 2020).This shortfall factor ultimately has an impact on swelling the realization of financing in the State Budget of Rp. 945.8 trillion or in other words there is an increase in the budget deficit to 6.1% of Gross Domestic Product (GDP).
Reported through the IMF, in 2016 Indonesia was declared included in the country with the largest row of tax avoidance cases, it proves that there are still companies that carry out massive tax avoidances practices by utilizing loopholes in the Law, and there are still companies that have not fully implemented GCG.In the Tax Justice Network report entitled The State of Justice 2020: Tax Justice in the time of it was revealed that the findings of tax avoidance of US $ 4.78 billion, which if rupiah is equivalent to Rp. 6, 8.7 trillion, including in Indonesia.And US $ 78.83 million or around Rp1.1 trillion of which came from individual taxpayers.(Sukmana, 2020) The phenomenon of tax avoidance that occurs in Indonesia INI among others occurs in companies engaged in the manufacturing industry.And companies identified as tax evasion include PT.Cola-Cola Indonesia (CCI) committed tax avoidance with a lack of tax payments amounting to Rp. 49.24 billion.(Rahmadian & Wijaya, 2023) The factor that influences tax avoidance is none other than a poor corporate governance system (Calista & Susanty, 2022).
Corporate governance or commonly known as Good Corporate Governance (GCG) is a process and structure used by companies to increase business success potential and also corporate accountability in order to realize shareholder value in the long term while still paying attention to the interests of other stakeholders referring to the Law (Zahrowati & Saputra, 2020).
Historically, the term Good Corporate Governance in Indonesia began to be known in early 1997, when the economic crisis hit Indonesia.There are adverse effects of the monetary crisis, including many companies that went bankrupt because they were unable to survive the turmoil of the monetary crisis that hit almost all countries in the world.Poor governance is one of the factors causing the economic and political crisis in Indonesia that began in 1997 The government also gave a strong impetus to the implementation of GCG in Indonesia, evidence of this concern can be seen with the establishment of regulations that regulate GCG practices in Indonesia.Starting from the establishment of the National Committee on (Wirawan & Diyanty, 2014) (Rizkia, 2023) Companies that carry out tax avoidance practices certainly do tax planning with the aim of minimizing the tax burden incurred but this will bring a bad image to a company, this is because tax avoidance has elements of confidentiality that can reduce the transparency of a company so that elements of good company management are needed, The number of companies that do tax avoidance proves that Corporate Governance in Indonesia has not been fully carried out (Minarta, 2021) .Governance structures can be implemented by various mechanisms, such as institutional ownership and audit committees (Maharani &;Suardana, 2014) .
In addition to good corporate governance, there are also other factors that can affect tax avoidance practices, namely profitability.In his research, it is stated that profitability is one of the benchmarks of company performance.Profitability shows the ability of a company to generate profits in a certain period at the level of assets, sales, and share capital.Meanwhile, according to Profitability is one of the financial ratios that can be used to assess a company and to find out how well the company's ability to produce profits (Maharani &;Suardana, 2014)(Kashmir, 2014) (Pricilia & Susanto, 2017) .
Profitability can be calculated by return on investment (ROI), gross profit margin (GPM), return on equity (ROE), return on assets (ROA), operating profit margin (OPM), net profit margin (NPM), return on common stock equity (RCSE), basic earning power (BEP), and earnings per share (EPS).In this study, researchers used indicators that are indicators of return on equity (ROE) as a measure of the ratio of profitability to tax avoidance.The results showed that Profitability measured using the (Ariawan & Saad, 2017)Return on Assets (ROA) ratio had a positive effect on Tax Avoidance.However, the results of research by I Gusti Ayu and Suardana (2014) show that profitability measured throughReturn on Assets (ROA) negatively affects Tax Avoidance.
The phenomena and differences in the results of previous research conducted by and become a reference for researchers to conduct research to examine the effect of institutional ownership, independent board of commissioners, and profitability on (Ariawan & Saad, 2017) (Maharani &;Suardana, 2014)tax avoidance using different sectors, namely the Industrial sector and differences in measuring profitability variables from previous researchers.The reason researchers use the industrial sector is because this sector is a sector that contributes greatly to Indonesian taxation.Researchers used panel data to measure samples, starting from the period 2018-2022.The title of this study is, "The Effect of Profitability and Good Corporate Governance (GCG) on Tax Avoidance in Industrial Sector Companies Listed on the Indonesia Stock Exchange (IDX)".

METHODE
This study uses tax avoidance variables as dependent variables.The operational definition of tax avoidance variables for this study is the company's effort to minimize the tax burden paid by the company without violating applicable tax laws in Indonesia.The facts show that, almost all or the majority of companies in Indonesia still do tax avoidance.However, there are also some companies that do not do tax avoidance.In this study, an effective tax rate (ETR) measure was used to measure tax avoidance.The ETR measurement method is to compare the company's tax burden with profit before tax, where the data is obtained through the company's income statement.
Then, for the independent variables used in this study are as follows:

Institutional Ownershipl
Operational definition of variables Institutional ownership for this study is stock ownership by institutions such as pension funds, financial institutions, insurance companies, and others.The higher the level of institutional ownership in the company can have a positive impact on supervision in the company in question.In this study, institutional ownership variables are measured using the proportion of share ownership by institutions divided by the number of shares outstanding.These data are obtained from the company's annual financial statements and calculated using the following formula:

Independent Board of Commissioners
The elements of the board of commissioners referred to in this study are members of the board of commissioners who are not affiliated with management and other members of the board of commissioners who can influence the company's performance.In this study, the independent board of commissioners is measured by the following formula:

Profibilitas
The operational definition of profitability variables for this study is a measure of the ability of an enterprise to earn profit in a certain period.In this study, profitability is measured by the return on equity (ROE) ratio.According to measuring (Kashmir, 2014)return on equity , it is to compare the company's net profit after tax with the company's capital.These data are obtained through the company's financial statements.The ROE formula is as follows: The type of data used by researchers is quantitative data with data sources in the form of secondary data.In this study, researchers used secondary data in the form of annual reports and financial statements owned by industrial sector companies listed on the IDX in 2018-2022.This research data is obtained directly from the official website of the Indonesia Stock Exchange (IDX), namely www.idx.co.id and from the company's website.
The population of this study is industrial sector companies listed on the Indonesia Stock Exchange (IDX) during the 2018-2022 period, totaling 62 companies.The sample is part of the industrial sector listed on the IDX in 2018-2022, totaling 19 companies.
The sample selection in this study uses the purposive sampling method which means the determination of the sample based on certain criteria, the sample selection criteria set in this study are as follows: 1. Industrial companies listed on the IDX during the research period, namely the period 2018-2022 2. Companies that present financial statements in rupiah (Rp) in full during the period 2018-2022 3. Companies that have a complete GCG structure, namely the 2018-2022 period Based on the population information and samples described above, the total number of research samples is 105 research samples with the following details: The data collected in this study is in the form of secondary data where the data collection method used is the documentation method, namely by tracing and elaborating financial statements and annual reports published by companies in the IDX industrial sector from 2018 to 2022.Researchers also conduct library research by studying various references / literature related to this research.This study used multiple liner regression data analysis method that examined the effect of institutional ownership, independent board of commissioners, and profitability.The data used were as many as 105 samples (one hundred and five) samples of companies.This model was chosen because in this study it was designed for independent variables that affect the dependent variable.Furthermore, in this study the data was processed using statistical methods using statistical package for social sience 25 (SPSS 25) software.
In statistical testing, later this study will provide a descriptive statistical picture and test classical assumptions before conducting regression testing.Testing classical assumptions to avoid deviations from classical assumptions so that problems do not arise in the use of the analysis.
Classical assumption tests in this study include normality tests, multicolonicity tests, heteroskesdacity tests, and autokeleration tests.
The normality test is performed to test whether in the regression model the independent variable and the bound variable have a normal data distribution or not.To find out whether the data is normally distributed or not, the Kolmogorov-Smirnov test will be carried out by looking at the level of significance.Normal detection when Kolmogorov-Smirnov significance value > 0.05.
The multicollinearity test aims to test whether the regression model found a correlation between independent variables.Methods that can be used to test multicollinearity can be seen from the correlation matrix of independent variables.
The heteroscedasticity test aims to test whether in linear regression models there is an inequality of variance from residuals or observations to other observations.If variance from one observation to another is fixed then it is called homoscedasticity (Ghozali, 2018).The autocorrelation test aims to test whether in the linear regression model there is a correlation between errors in the t-1 period.If there is a correlation, there will be autocorrelation.Autocorrelation occurs because there are sequential observations throughout the year related to each other, a good model is autocorrelation-free (Ghozali, 2018).
Regression analysis is basically the study of the dependence of the dependent variable (bound) with one or more independent variables (explanatory or independent variables), with the aim of estimating and predicting the population mean or the average value of the dependent variable based on the known value of the independent variable.
The regression equation model is formulated: Y = a + b1X1 + b2 X2 + b3 X3 + e Information: Y: Tax Avoidance X1: Institutional Ownership X2: Independent Commissioner X3: Profitability a: Konstanta β1 -β3: Regression coefficient of independent variable The coefficient of determination R 2 measures how far the model is able to explain the variation of the dependent variable, with values between zero and one.With each additional independent variable R2 increased regardless of whether it had a significant effect on the dependent variable, many researchers recommend using the Adjusted R2 value when evaluating which regression model is best.If it approaches 1 (the greater the value of R 2 ), it indicates that the independent https://doi.org/10.24256/kharaj.v5i3.4007 variable provides almost all the information needed to predict the variation of the dependent variable (Ghozali, 2018) .
The F test is used to test the joint hypothesis that b1, b2, and b3 are simultaneously equal to zero.The test criteria are: 1.If the significant value of the F test > 0.05, then there is no simultaneous influence of the independent variable on the dependent variable.2. If the significant value of the F test < 0.05, then there is a simultaneous influence of the independent variable on the dependent variable.The purpose of hypothesis testing is to determine whether the theoretical answers contained in the hypothesis statement are supported by the facts collected and analyzed in the process of testing data.The t-test basically shows how far one explanatory or independent variable has influence individually in explaining the variation of the dependent variable.The test criteria are partial with a level of significant α = 5% and a confidence level of 95%, that is, if the significance value t < 0.05, then the hypothesis is accepted, meaning that there is a significant influence between one independent variable and the dependent variable.And if the significance value t > 0.05, then the hypothesis is rejected, meaning that there is no significant influence between one independent variable and the dependent variable (Ghozali, 2018) .

RESULTS AND DISCUSSION Descriptive Research Object
The population in this study is industrial sector companies listed on the Indonesia Stock Exchange (IDX) during the 2018-2022 period.Based on the results of the selection that has been carried out, as many as 62 companies were registered and selected in the criteria, namely as many as 105 data samples based on the criteria that have been carried out by researchers using purposive sampling as follows:

Data Analysis Descriptive Statistical Analysis
The following data have been collected as-is without the intention of making generalized or generalized conclusions.Based on Table 2. above the results of descriptive statistics from 70 industrial sector company data from 2018 to 2022 for the dependent variable Tax Avoidance (Y) and independent variables including Institutional Ownership (X 1 ), Independent Board of Commissioners (X 2 ), and Profitability (X 3 ), this study shows the following: 1. Tax Avoidance as a dependent variable is proxied with ETR with a minimum value of 0.17 and a maximum value of 0.88 while the average value (mean) is 0.2059 and a standard deviation value of 0.81189 which means that the tax avoidance variable fluctuates because there is a large enough difference between the minimum and maximum values.In addition, this also indicates that the tax avoidance variable is normally distributed when viewed from a standard deviation value that is smaller than the average value (mean).2. Institutional ownership as an independent variable is proxied by share ownership by institutional parties divided by the number of shares outstanding.The minimum value of the institutional ownership variable is 0.13 and the maximum value of 1 at the mean value of 0.5858.As well as the standard deviation value of 0.24324 when viewed from the standard deviation and average value, the institutional ownership variable fluctuates greatly because the difference between the minimum and maximum values is quite large.This indicates that the variable of normally distributed institutional ownership in terms of standard deviation is smaller than the mean value.3. Independent commissioners as an independent variable are proxied by the number of independent commissioners divided by the number of company commissioners multiplied by 100%.The minimum value is 0.33 and the maximum value is 0.60 while the mean is 0.4311 and the standard deviation is 0.09176.When viewed from the relatively large difference between the minimum and maximum values, the independent commissioner variable fluctuates greatly.It also indicates that the data is normally distributed because the standard deviation value is smaller than the mean value.Profitability as an independent variable measured by net profit after tax divided by the company's total equity multiplied by 100%.It has a minimum value of 0.01 and a maximum value of 0.25.The mean is 0.1034 while the standard deviation is 0.08189.The variable profitability fluctuates quite a lot because it is seen from the difference between the minimum and maximum values which are quite large.This indicates that the variable profitability is normally distributed in terms of its standard deviation value which is smaller than the mean value .

Normality Test
In this study, the normality test against residuals used the Kolmogrov-Smirnov test using 105 samples of initial companies before outliers were carried out on several samples of companies that were not normally distributed.The regression model can be said to satisfy the normality test if the significant value > 0.05.Based on the results of the normality test that has been performed in table 3, the value of Asymp.Sig shows a value of 0.0016.When viewed the value has a value smaller than 0.05 which means that the data is not normally distributed so that outliers are carried out on some samples with the aim of making the data normally distributed.
An outlier is a case or data that looks very different from other observations and appears in the form of extreme values for either a single variable or a combination (Ghozali, 2018)  After outliers were carried out on 33 company samples, the n used in this study was 72 samples.From table 4. above it can be seen that Asym.Sig which shows a value of 0.200 is greater than the significance level of 0.05 which means that the data has been normally distributed and can be done for further testing of multiple regression models.

Multicholinerity Test
Methods that can be used to test multicollinearity can be seen from the correlation matrix of independent variables.In the correlation matrix if between independent variables there is a high correlation (generally, above 0.90) then this is an indication of multicollinearity by showing the value of Tolerance and Variance Inflation Factor (VIF) (Ghozali, 2016) If the VIF value ≥ 10 and the tolerance value ≤ 0.1 illustrates that the regression model experiences multicollinearity and vice versa, if the VIF value ≤ 10 and the tolerance value ≥ 0.1 then the regression model is free from multicollinearity.Based on table 5.above which shows that the Tolerance value is greater than 0.1 and the VIF value is smaller than 10, so that overall the variables above show that multicollinearity does not occur.
From this value, it can be explained in detail that the institutional ownership variable has a tolerance value of 0.943 which is greater than 0.1 with a VIF value of 1.060 which is smaller than 10, the independent commissioner variable has a tolerance value of 0.968 which is greater than 0.1 and a VIF value of 1.033 which is smaller than 10, then the profitability variable shows a tolerance value of 0.969 which is greater than 0.1 and a VIF value of 1.032 more small out of 10.Heteroskedaticity Test Based on the scatterplot above which shows that the data points spread above and below or around the number 0, and the points do not collect below or above only, the spread of the points above does not form a certain pattern, so it can be concluded that there is no heteroschedality problem.

Autocorrelation Test
Autocorrelation test in this study to test whether in linear regression model there is a correlation between errors in period t-1.In this study, the Durbin-Waston test was used in the summary model where the data was said to be autocorrelation-free if the indigo du<dw< 4-du.The results of the autocorrelation test can be seen in table 6. as follows.Based on the results of the SPSS output above, it is known that the value of Durbin-Watson of 1,691 values can be compared with the values in table 6. Durbin-Watson at a significance of 0.05.It is known that n is the number of research samples totaling 72 and k is the number of independent variables in the study, which is as many as 3 variables.Furthermore, the DW value must be greater than du, the number can be seen in the table value 6. durbin-watson and found a d U value of 1.7054.From these results, the Durbin-Watson value between the limits (d U) and (3-d U ) is 3 -1.7054= 1.2946, the result of d U 1.2946 is smaller than 1.691, which can be concluded that there is no autocorrelation, thus the results of multiple linear regression analysis for research hypothesis testing can be continued.

Multiple Linear Regression Analysis
Regression analysis is basically the study of the dependence of the dependent variable (bound) with one or more independent variables (explanatory or independent variables), with the aim of estimating and predicting the population mean or the average value of the dependent variable based on the known value of the independent variable.Based on table 7. Multiple linear regression above then obtained the following equation: Y = 0, 258 + 0, 018X 1 + 0, 053X 2 -0.251X 3 + e From the equation above can be explained as follows: 1.The value of the Constant (α) is 0.258 which means if all independent variables are equal to 0, then the value of Tax avoidance is 0.258.2. The value of the regression coefficient of institutional ownership is 0.018 which is a positive value of this value it can be stated that the variable of institutional ownership has a positive effect on tax avoidance.It also indicates that if the regression coefficient value of other variables remains, the change in institutional ownership (X 1) of 1 percent will increase tax avoidance practices by 0.018 or 1.8 percent.3. The value of the coefficient of the independent commissioner variable of 0.053 is positive, from this value it can be defined that the independent commissioner variable has a positive effect on tax avoidance and it also has meaning if the value of the regression coefficient of other variables remains, then a change in the independent commissioner variable (X 2 ) of 1 percent will increase tax suppression by 0.053 or 5.3 percent.The value of the coefficient of the profitability variable of -0.251 is negative, from this value it can be defined that the profitability variable has a negative effect on tax avoidance, and it has a meaning that if the value of the regression coefficient of other variables remains then a change in the profitability variable (X 4 ) of 1 percent will reduce tax avoidance by 0.251 or by 25.1 percent.

Coefficient of Determinity (R 2 )
The coefficient of determination R 2 measures how far the model is able to explain the variation of the dependent variable, with values between zero and one.With each additional independent variable R2 increased regardless of whether it had a significant effect on the dependent variable, many researchers recommend using the Adjusted R2 value when evaluating which regression model is best.If it approaches 1 (the greater the value of R 2 ), it indicates that the independent variable provides almost all the information needed to predict the variation of the dependent variable .(Ghozali, 2018) Table 8 When viewed from table 8.The results of the Adjusted R Square coefficient of determination test are 0.146 based on this value, it can be interpreted that the variables of institutional ownership, independent board of commissioners, and profitability can explain the variation in tax avoidance by 4.6% and the remaining 95.4% which can be explained by other factors.

Simultaneous Test (Test F)
Simultaneous tests in this study were used to determine the influence of simultaneously bound variables.The acceptance criteria in this test can be seen at a significance value smaller than 0.05 which means that the independent variable simultaneously affects the dependent variable.And vice versa if the significance value is greater than 0.05 then simultaneously the independent variable has no effect on the dependent variable.followingthe results of the simultaneous test (F) can be seen in table 9. below.Based on the results of the output above, it can be stated that in table 9. ANOVA significant values show a value of 0.009 smaller than 0.05 which means that the regression model used is feasible and simultaneously the independent variable has a significant effect.

Simultaneous Test (T Test)
In simultaneous testing (T test) is used to see how far the influence of individual independent variables that explain the variation of the dependent variable.The test results of the t test are shown in table 10. coefficients below.Based on the results of the output above, you can see the following detailed explanation below: 1.The institutional ownership variable has a beta (β) value of 0.057 and a significant value of 0.015 which is smaller than 0.05 then it can be concluded that H1 is acceptable.2. The independent board of commissioners variable has a beta value (β) of 0.144 and a significant value of 0.012 which is smaller than 0.05 from the result it can be concluded that H2 is accepted.3. The profitability variable measured by return on equity has a beta value (β) of -0.218 with a significant value of 0.097 which is greater than 0.05 it can be concluded that H3 is rejected.
From the explanation above, it can be seen in table 10.A summary of the acceptance and rejection of the hypothesis is below.

Pemabahasan
The Effect of Institutional Ownership on Tax Avoidance Based on the results of the previous t test, it can be seen that the variable of institutional ownership of tax avoidance is significant at a confidence level of 0.05 so that it can be concluded that institutional ownership can affect tax avoidance practices.Furthermore, the value of the coefficient (β) can be seen in table 10.Coefficients on the results of the t test that institutional ownership has a positive direction with a value of 0.015 which shows that the greater institutional ownership in a company, the more able to minimize tax avoidance practices.
Jensen &; Meckling, (1976) in agency theory states that institutional ownership has an important role in minimizing agency conflicts that occur between agents and principals.The existence of institutional ownership in one company can maximize the monitoring process in decision making carried out by agents (managers) so as to minimize the actions of deviant agents.(Hamdani, 2016) The results of this study are in line with previous research conducted by which stated that institutional ownership has a positive and significant effect on tax avoidance practices.(Mulyani et al., 2018) The Influence of the Independent Board of Commissioners on Tax Avoidance Based on the results of the existing hypothesis test, it shows that the independent board of commissioners on tax avoidance has a significant value with a confidence level of 0.012 so that it can be concluded that the independent board of commissioners influences tax avoidance practices.When viewed from the regression coefficient, the existence of an independent board of commissioners in the company can maximize the monitoring system for managers in decision making and this can certainly reduce tax avoidance practices in the company.
According to Jensen & Meckling, (1976) in agency theory which states that the existence of an independent board of commissioners that has no affiliation with the authorities and acts neutrally can minimize tax avoidance practices in a company and can maintain company value in a good governance system.Companies generally have an independent board of commissioners amounting to 30% of the total members of the company's board of commissioners (Hamdani, 2016) (Pohan, 2013).
This research is in line with agency theory according to Eisenhardt, (1989) in (Aisha & Susilowati, 2021) stating that agency theory is based on several assumptions.These assumptions include assumptions of basic human nature, organizational assumptions and assumptions of information.Assumptions of human nature that emphasize that humans have limits and do not like to take risks, organizational assumptions between members and the assumption of information that makes information as commodity goods that can be traded.While the independent party functions as a mediator between the agent and the principal and other functions as a monitor of the agent's behavior in carrying out the duties of authority with the objectives of the principals.
However, the results of this study are not in line with previous research conducted by the results that the independent board of commissioners has no influence on tax avoidance practices and research conducted by that the independent board of commissioners negatively affects tax avoidance practices.(Praditasari & Saad, 2017) (Maharani &;Suardana, 2014) The Effect of Profitability on Tax Avoidance https://doi.org/10.24256/kharaj.v5i3.4007 Based on the results of the output results of table 10.which shows that the profitability variable has no effect on tax avoidance practices, judging from the significance value of 0.097 which is greater than 0.05 and profitability has a coefficient value of -131 which gives an insignificant negative direction.
In agency theory, it explains that there is an information asymmetry between shareholders and managers used by potential investors to assess the state of a company whether it has a high value, besides that in agency theory it also explains that there can be differences between individuals and agents on the interests of company profits and tax collectors.The tax collector expects the maximum possible income from tax collection while the management wants the maximum possible profit and minimizes the tax burden (Rahayu, 2019).This research is in line with agency theory which states that high-value companies certainly have good management, including profitability.In this study, the measurement used is return on equity where profitability is measured using net profit after tax compared to total capital.The research is in line with previous research conducted by which showed that profitability negatively affects tax avoidance.The higher the company's ROE value, the better.(Maharani &; Suardana, 2014)

CONCLUSION
Based on the results of testing and discussion, it can be concluded that: 1.Institutional ownership has a positive and significant effect on tax enforcement practices with a sig value of 0.015 < 0.05 and a value (β) of 0.57 which can be seen in table 10.From these results, it can be concluded that the higher the level of institutional ownership in the company, the better the corporate governance system in minimizing tax avoidance practices.2. The independent board of commissioners has a positive and significant effect on tax avoidance practices with a GIS value of 0.012 < 0.05 and a value of (β) 0.144 which can be seen in table 10.From these results, it can be concluded that the higher the proportion of independent board of commissioners, the stronger the supervisory system in the company in minimizing tax avoidance practices.3. Profitability measured using ROE has no effect on tax avoidance practices with a sig value of 0.97 > 0.05 and a value of (β) -0.131 which can be seen in t10.From these results, it can be concluded that profitability as measured by ROE gives a negative and insignificant direction, which is a contributing factor is the lack of management ability to manage capital deposited by investors in increasing company profits, so that it can create loopholes in tax avoidance practices.
For subsequent research and development, researchers suggest adding research samples to corroborate research generalizations.Researchers also suggest reviewing and adding other probability variables that can affect tax avoidance activities.(Pricilia & Susanto, 2017)
Corporate Governance Policy (KNKCG) in KEP/31/M.EKUIN/08/1999 concerning the Establishment of KNKCG which then issued Good Coporate Indonesia guidelines, and then continued with the establishment of the National Committee for Governance Policy (KNKG) as a replacement for KNKCG through decree Number KEP/49/M.EKON/11/2004 which consisted of Public Subcommittees and Corporate Subcommittees.

Table 7 .
Multiple Linear Regression Analysis Test ResultsInfluence Profibilitas and Good Corporate Governance (GCG) Towards Tax Avoidance...