INTRODUCTION

Taxpayers in reducing the tax burden often do tax planning or Tax Planning. Planning by reducing the tax burden, some are allowed but must comply with tax provisions and some are not allowed which are not in accordance with tax provisions. Tax avoidance is a legal tax avoidance effort that does not violate tax regulations by taxpayers by trying to reduce the amount of tax by looking for regulatory weaknesses (Sinambela & Sinambela, 2019). Meanwhile, according to tax regulations, tax evasion (tax evasion) is a violation of tax law. Currently, tax avoidance is something that the Fiskus must pay attention to, because there are several possibilities that this tax avoidance will lead to tax evasion.

Food and beverage sub-sector companies in GDP growth data have a significant decline in 2019-2020, so it is possible for this sector to carry out tax avoidance to ease the burden of tax expenditures in order to maintain the company's net profit.

Several factors that can affect tax avoidance, one of which is profitability, profitability is the company's ability to earn comprehensive profits, convert sales into profits and cash flow (Sirait, 2017). Profitability can measure the welfare of shareholders or company investors in the form of dividend payments and profit returns (Agusti, 2014).

Companies that have low sales growth are more likely to have the possibility of minimizing financial burdens including corporate tax burdens compared to companies with high or increasing sales. This is evidenced by (Hidayat, 2018) proving that sales growth has a positive effect on company growth. The study also explains that the higher sales growth, the less corporate tax avoidance activities are caused because companies with high sales levels will have the opportunity to earn large profits and be able to pay taxes.

The age of the company, the longer the period of establishment of the company, the more experience the company has and the more human resources it has, the more skilled it is in managing the tax burden so that it tends to have loopholes in tax avoidance (Dewinta & Setiawan, 2016).

A comparison tool is needed to compare accuracy and clarity in a study, therefore the researcher includes several research results that can be used as a reference for further study. The results of previous studies have links with research variables and affect tax avoidance, including research:

According to (Dewinta & Setiawan, 2016) which analyzes the effect of company size, company age, profitability, leverage, and sales growth on Tax Avoidance. The data used used 176 research samples from manufacturing companies listed on the Indonesia Stock Exchange (IDX) for the 2011-2014 period. The analysis technique used is multiple linear regression analysis technique. The results of the analysis prove that company size, company age, profitability, and sales growth have a positive effect on Tax Avoidance.

According to (Hidayat, 2018) which examines the effect of profitability, leverage, and sales growth on tax avoidance. The data used are 25 research samples from manufacturing companies listed on the Indonesia Stock Exchange (IDX) for the 2011-2014 period. The analysis technique used is multiple regression analysis technique. The results of the analysis prove that profitability and sales growth have an effect on tax avoidance, while leverage has no effect on tax avoidance.

According to (Indriani & Juni, 2020) research examines the effect of firm size, firm age, sales growth, and profitability on tax avoidance. The data used used 10 research samples from pharmaceutical companies listed on the Indonesia Stock Exchange (IDX) for the 2016-2019 period. The analysis technique used is multiple regression analysis technique. The results of the analysis prove that company size, company age, sales growth, and profitability have no effect on tax avoidance.

RESEARCH METHODS

The strategy used in this research is associative strategy. Associative strategy is a strategy in research conducted to determine whether there is a relationship between the dependent variable and the independent variable. The dependent variables in this study are Profitability, Sales Growth, and Company Age. While the independent variable is tax avoidance in food and beverage sub-sector companies listed on the Indonesia Stock Exchange.

There are 32 companies that are included in the food and beverage sub-sector category, but the researchers here only take 29 companies because the other 3 companies have data that is less proportional because some of the variables needed are not in the company's financial statements. the researchers used purposive sampling technique in determining the sample to be taken in this study.

The data used in this study is secondary data, secondary data is data obtained from the source indirectly. The data in question is evidence, records, or historical reports that have been archived, published, or unpublished (Ghozali, 2016). In this study, the secondary data used is the 2020 financial report data of food and beverage sub-sector companies listed on the Indonesia Stock Exchange.

1. Variable Operation

a. Profitabilitas (ROA)

b. Pertumbuhan penjualan (SALES)

c. Umur Perusahaan (Age)

Uper = (Years of Research-Years of Company Establishment)

d. Penghindaran Pajak (ETR)

The data analysis method used in this study is the multiple linear regression analysis model. This analysis uses calculations using statistics to measure the strength of the relationship between 2 or more variables. The test application used in the study is Eviews 9.

2. Model fit (Adjusted R2)

The fit model in this test is 0.151026, which means that the variation or behavior of the independent variables (Profitability, Sales Growth, Company Age) is able to explain the variation of the dependent variable (Tax Avoidance) of 15.10% while the rest is 84.90 % is the variation of other independent variables that affect tax avoidance but is not included in the model.

3. F Test (Global Test)

Used to test whether there is at least one independent variable that affects the dependent variable.

4. T test (Individual Test)

Used to test one by one the independent variables that affect the dependent variable.

5. Normality test

The Kolmogorov-Smirnov Z method, if the significance value shows 0.05 then the processed data is normally distributed, if the significance value is <0.05 then it can be stated that the processed data is not normally distributed (Gunawan, 2016).

6. Multicollinearity Test

The multicollinearity test aims to test whether there is a correlation between independent variables in the model or to determine whether or not there is a correlation between independent variables. The presence or absence of multicollinearity can be seen through the correlation matrix between the independent variables. The test method used to detect the presence or absence of multicollinearity in the regression is to look at the tolerance value and the opposite of the variance inflation factor (VIF). If the coefficients of the tolerance value limit model are > 0.10 and the VIF value is < 10, it can be said that there is no multicollinearity. The following are the results of the multicollinearity test for the research model carried out.

The effect of multicollinearity on the model is that the coefficient of influence is small, the standard error estimate is large and the theory test is not proven.

7. Autocorrelation Test

Autocorrelation test is used to test whether there is a correlation between the residual error in linear regression in period t with errors in period t-1.

8. Heteroscedasticity Test

The heteroscedasticity test is used to determine whether or not there is a deviation from the classical assumption of heteroscedasticity, namely the existence of variance inequality from the residuals for all observations in the regression model. If the variance of the residual from one observation to another observation remains, it is called homoscedasticity and if it is different it is called heteroscedasticity. A good regression model is a homoscedasticity model or there is no heteroscedasticity. The test model uses the Glejser test which aims to determine the symptoms of heteroscedasticity shown by the regression coefficients of each independent variable to the absolute value of the residual. If the probability is > 0.05 (ɑ) then it can be ascertained that the model does not contain elements of heteroscedasticity (Suliyanto, 2011).

RESULTS AND DISCUSSION

A. Description of Research Object

The data in this study uses secondary data in the form of annual reports (annual reports of manufacturing companies in the food and beverage sub-sector listed on the Indonesia Stock Exchange (IDX). The determination of the sample is carried out using purposive sampling technique with criteria and sampling process.

Variable	Coefficient	Std. Error	t-Statistic	Prob.
C	0.144912	0.038698	3.744637	0.0010
ROA	0.001243	0.002488	0.499578	0.6217
SALES	0.142856	0.069504	2.055348	0.0504
Age	0.001693	0.001061	1.596239	0.1230
R-squared	0.241988	Mean dependent var		0.201034
Adjusted R-squared	0.151026	S.D. dependent var		0.105808
S.E. of regression	0.097491	Akaike info criterion		-1.690667
Sum squared resid	0.237613	Schwarz criterion		-1.502075
Log likelihood	28.51468	Hannan-Quinn criter.		-1.631603
F-statistic	2.660333	Durbin-Watson stat		2.150363
Prob(F-statistic)	0.070026

Source: Processed data (2022)

C. Model fit (Adjusted R2)

Conclusion: the tax avoidance model has a weak goodness of fit.

D. F test (GLOBAL TEST)

The hypothesis:

Ho : b1 = b2 = b3 = 0 means that all independent variables do not affect the dependent variable

Ha : there is at least one independent variable that affects the dependent variable

Decision:

1. If the p-value (prob) of F < 0.05 then Ho is rejected,

2. If the p-value (prob) of F > 0.05 then Ho is accepted.

So, from the processed results, the p-value of F is 0.070026 > 0.05 so Ho is accepted (Ha is rejected).

E. t Test (Individual Test)

The Effect of Profitability (ROA) on Tax Avoidance

Ho : b1 < 0 means that ROA has no effect or has a negative effect on tax avoidance

Ha : b1 > 0 means that ROA has a positive effect on tax avoidance

The theory: (+)

Equation Interpretation

C = 0.144912 + 0.001243Y

a = 0.144912 indicates if the ROA is 0 then the average tax avoidance is 0.144912

b = 0.001243 means that if ROA increases by an average of Rp. 1, then tax avoidance will increase by 0.001243 and vice versa.

Decision:

1. If the p-value t/2 0.05 then Ho is rejected

2. If p-value t/2 > 0.05 then Ho is accepted

Processed results obtained p-value of t/2 = 0.31085 > 0.05 then Ho is accepted (Ha is rejected) so that it is proven that ROA has no effect or has a negative effect on tax avoidance.

F. Effect of Sales Growth on Tax Avoidance (+)

Ho : b1 < 0 means Sales Growth has no effect or has a negative effect on tax avoidance

Ha : b1 0 means that Sales Growth has a positive effect on tax avoidance

The theory: (+)

equation interpretation

C = 0.144912 + 0.142856Y

a = 0.144912 indicates if Sales Growth is 0, then the average tax avoidance is 0.144912

b = 0.142856 means that if Sales Growth increases by an average of Rp. 1, then tax avoidance will increase by 0.142856 and vice versa.

Decision

1. If the p-value t/2 0.05 then Ho is rejected

2. If p-value t/2 > 0.05 then Ho is accepted

Processed results obtained p-value of t/2 = 0.0252 <0.05 then Ho is rejected (Ha is accepted). So it is proven that Sales Growth has a positive effect on Tax Avoidance.

G. Effect of Company Age on Tax Avoidance (+)

Ho : b1 0 means that the age of the company has no effect or has a positive effect on Tax Avoidance

Ha : b1 < 0 means that the age of the company has a negative effect on tax avoidance

The theory: (+)

equation interpretation

C = 0.144912 + 0.142856Y

a = 0.144912 indicates if the age of the company is 0, then the average tax avoidance is 0.144912

b = 0.001694 means that if the age of the company increases by an average of Rp. 1, then tax avoidance will increase by 0.001694 and vice versa.

Decision

1. If the p-value t/2 0.05 then Ho is rejected

2. If p-value t/2 > 0.05 then Ho is accepted

Processed results obtained p-value of t/2 = 0.0615 > 0.05 then Ho is accepted (Ha is rejected). So that it is proven that the age of the company has a positive effect on tax avoidance.

H. Normality test

The normality test aims to test whether in the regression model, the confounding or residual variables have a normal distribution. The normality test of the data was carried out using the Kolmogorov-Smirnov Test, by looking at the significance level of 5%. Data is said to be normally distributed if the significance of the dependent variable is more than 5% or the asymp significance value is greater than 0.05 (Suliyanto, 2011).

Based on the results of the Kolmogorov-Smirnov Test, the following results were obtained:

F-statistic	1.099314	Prob. F(2,23)		0.3500
Obs*R-squared	2.530305	Prob. Chi-Square(2)		0.2822
Test Equation:
Dependent Variable: RESID
Method: Least Squares
Date: 06/30/22 Time: 15:14
Sample: 1 29
Included observations: 29
Presample missing value lagged residuals set to zero.
Variable	Coefficient	Std. Error	t-Statistic	Prob.
C	-0.006690	0.039866	-0.167821	0.8682
ROA	-4.14E-05	0.002559	-0.016191	0.9872
SALES	0.013177	0.076245	0.172820	0.8643
AGE	0.000191	0.001129	0.169133	0.8672
RESID(-1)	-0.031027	0.226706	-0.136861	0.8923
RESID(-2)	0.299820	0.211297	1.418951	0.1693
R-squared	0.087252	Mean dependent var		-1.06E-17
Adjusted R-squared	-0.111172	S.D. dependent var		0.092121
S.E. of regression	0.097106	Akaike info criterion		-1.644032
Sum squared resid	0.216881	Schwarz criterion		-1.361143
Log likelihood	29.83846	Hannan-Quinn criter.		-1.555434
F-statistic	0.439726	Durbin-Watson stat		1.969845
Prob(F-statistic)	0.816138

Source: Processed data (2022)

F-statistic	0.909418	Prob. F(9,19)		0.5371
Obs*R-squared	8.731295	Prob. Chi-Square(9)		0.4624
Scaled explained SS	6.447691	Prob. Chi-Square(9)		0.6944
Test Equation:
Dependent Variable: RESID^2
Method: Least Squares
Date: 06/30/22 Time: 15:18
Sample: 1 29
Included observations: 29
Variable	Coefficient	Std. Error	t-Statistic	Prob.
C	0.006319	0.010614	0.595338	0.5586
ROA^2	3.95E-06	3.27E-05	0.120524	0.9053
ROA*SALES	-0.001460	0.002641	-0.552886	0.5868
ROA*AGE	3.07E-06	2.34E-05	0.131506	0.8968
ROA	-0.000542	0.001145	-0.473478	0.6413
SALES^2	0.022497	0.030833	0.729646	0.4745
SALES*AGE	0.000105	0.000808	0.130022	0.8979
SALES	0.012435	0.034292	0.362629	0.7209
AGE^2	-2.42E-06	9.65E-06	-0.250535	0.8049
AGE	0.000181	0.000628	0.288853	0.7758
R-squared	0.301079	Mean dependent var		0.008194
Adjusted R-squared	-0.029989	S.D. dependent var		0.011755
S.E. of regression	0.011930	Akaike info criterion		-5.752703
Sum squared resid	0.002704	Schwarz criterion		-5.281222
Log likelihood	93.41419	Hannan-Quinn criter.		-5.605041
F-statistic	0.909418	Durbin-Watson stat		2.039588
Prob(F-statistic)	0.537063

Source: Processed data (2022)

Heteroscedasticity Hypothesis

Ho : no heteroscedasticity

Ha : there is heteroscedasticity

Testing using a white test with a decision

1. If the p-value of Chisquare <0.05 then Ho is rejected

2. If the p-value of Chisquare> 0.05 then Ho is accepted

From the processing results, the p-value of the chi-square is 0.537063 > 0.05, then Ho is accepted so that in the resulting model there is no heteroscedasticity.

CONCLUSION

The results of the study can be concluded that Profitability has no effect on decisions in tax avoidance or Tax Avoidance in food and beverage sub-sector companies listed on the Indonesia Stock Exchange (IDX) for the period 2020. Sales growth affects decisions in tax avoidance or Tax Avoidance in companies. food and beverage sub-sector listed on the Indonesia Stock Exchange (IDX) for the period 2020. Company age does not affect decisions in tax avoidance or Tax Avoidance for food and beverage sub-sector companies listed on the Indonesia Stock Exchange (IDX) for the period 2020.

For further researchers, they can add the year of observation and the number of companies in the food and beverage sub-sector. As well as adding other independent variables that are suspected of influencing tax avoidance, such as corporate governance which includes audit quality, independent commissioners, audit committees, and other parties that can influence the company's decision to avoid tax. For companies to be able to carry out tax planning properly because tax avoidance can be indicated as a tax evision.

BIBLIOGRAFI

Agusti, W. Y. (2014). Pengaruh profitabilitas, leverage, dan corporate governance terhadap tax avoidance (studi empiris pada perusahaan manufaktur yang terdaftar di BEI tahun 2009-2012). Jurnal Akuntansi, 2(3), 1–20. Google Scholar

Dewinta, I. A. R., & Setiawan, P. E. (2016). Pengaruh ukuran perusahaan, umur perusahaan, profitabilitas, leverage, dan pertumbuhan penjualan terhadap tax avoidance. E-Jurnal Akuntansi, 14(3), 1584–1615. Google Scholar

Ghozali, I. (2016). Aplikasi Analisis Multivariete dengan Program IBM SPSS. In Semarang, Universitas Diponegoro. Badan Penerbit Universitas Diponegoro. Google Scholar

Gunawan, I. (2016). Metode penelitian kuantitatif: Teori dan Praktik. Jakarta: Bumi Aksara.

Hidayat, M. (2018). Pengaruh Manajemen Pajak, Pertumbuhan Penjualan Dan Uku-Ran Perusahaan Terhadap Nilai Perusahaan Pada Perusahaan Manufaktur Di Bei Periode 2014-2016. MEASUREMENT: Journal of the Accounting Study Program, 12(2), 206–215. https://doi.org/10.33373/mja.v12i2.1745. Google Scholar

Indriani, M. D., & Juni, J. (2020). Pengaruh Ukuran Perusahaan, Umur Perusahaan, Pertumbuhan Penjualan dan Profitabilitas terhadap Penghindaran Pajak (Tax Avoidance) (Studi Empiris pada Perusahaan Sub Sektor Farmasi yang Terdaftar Di Bursa Efek Indonesia Periode 2016-2019). STIE Indonesia Jakarta. Google Scholar

Sinambela, L. P., & Sinambela, S. (2019). Manajemen Kinerja; Pengelolaan, Pengukuran, dan Implikasi Kinerja. Rajawali Press. Google Scholar

Sirait, P. (2017). Definisi Profitabilitas. Yogyakarta.

Suliyanto, S. (2011). Perbedaan Pandangan Skala likert sebagai Skala Ordinal atau skala Interval. Prosiding Seminar Nasional Statistika Universitas Diponegoro 2011, 51–60. Google Scholar

Stages	Sample Determination Description	Number of Companies
1	Manufacturing companies in the food and beverage sector listed on the Indonesia Stock Exchange (IDX) 2020	32
2	Manufacturing companies that do not publish audited financial reports and complete annual reports during 2020	(3)
3	Companies that do not publish audited annual financial statements during the 2020 period in rupiah	29
4	Food and beverage companies that provide the accounts needed to perform variable calculations in research	29
Number of Observations used		29

	Coefficient	Uncentered	Centered
Variable	Variance	VIF	VIF
C	0.001498	4.569353	NA
ROA	6.19E-06	1.421143	1.095287
SALES	0.004831	1.058982	1.041103
AGE	1.13E-06	4.804694	1.058037