Increased Tax Avoidance By Leverage Through Corporate Social Responsibility Disclosure

This study aims at examining and analysing the effect of leverage on tax avoidance. The moderating variable is disclosing corporate social responsibility. The population in the current study is manufacturing companies in the consumer goods industry sector. The companies were listed on the Indonesia Stock Exchange (ISE) from 2013 to 2017. To examine the hypotheses, Moderation Regression Analysis (MRA) was applied. The results prove that leverage has a positive and significant effect on tax avoidance. Moreover, the corporate social responsibility disclosure is proven to strengthen the effect of leverage on tax avoidance.


INTRODUCTION
Tax is a mandatory contribution to a country. According to Rodriguez & Ar ias. , (2012) Leverage (LEV) is a factor that affects companies in paying taxes. Leverage is the level of debt used by companies in financing. A percentage of total debt measures financial leverage to the company's equity in a period, also called Debt to Equity Ratio (DER). Darmawan & Sukartha (2014) states that t he incr ease in the amount of debt will result in interest expenses that must be paid by the company. The interest expense component will reduce the company's pre-tax profit. This statement is in line with research conducted by Zahirah et al. , (2017) which states that leverage affects tax avoidance, the research results are supported by research by (Jasmine et al., 2017) and (Nursari et al., 2017).
Based on the empires results, it can be seen that a clear pattern of relationship between leverage and tax avoidance is that companies t hat have a high degree of leverage tend to avoid taxes. Nevertheless, on the other hand, t he company is also required to keep maintaining the level of investor confidence so that this pressure encourages companies to disclose CSR as a form of cor por ate social responsibility. This is in line with the signal theory that companies with high leverage levels are expected to make more CSR disclosures to reduce information asymmetry that might result in stronger creditors and investor pressure on the company. By providing information disclosure such as CSR, it is expected that parties such as creditors and investors can see this as a guar antee for the company's business so that their rights as creditors and investors are guaranteed and do not put greater pressure on the company (Christiawan & Putri, 2014). However, despite the apparent trend patterns, some researchers revealed different results. Faizah & Adhivinna, (2017) and Hidayat, (2018) who succeeded in proving that leverage does not affect tax avoidance. B ased on t he description, the hypothesis in this study is as follows. H1: Leverage affects tax avoidance (tax avoidance).
According to Watson, (2011), if the company does not disclose CSR, it indicates that the company does not have a social responsibility. Therefore, it is often the parties concerned to make a tax strategy. The more forms of accountability a company does to its environment, the company's image will increase, and investor interest will increase. Social responsibility is one of the main factors in the survival of a company.
This study selects the object of research in the manufacturing companies in the consumer goods industry sector. The reason is that t he consumer good s industry sector has good sales each year and the consumer goods industry sector is a large company that promises high profits, therefore the possibility of greater tax avoidance. Then, related to CSR disclosure, according to Ramadhani, (2012) companies in the consumer goods industry sector have more influence/ impact on the surrounding environment as a result of the activities carried out by the company and meet aspects on the theme of CSR disclosure.
Based on the background described and the conflict between previous researchers (research gap), the authors are interested in re-analysing the effect of leverage on tax avoidance with CSR disclosure as moderating. It is hoped t hat through the findings of this research, at least it can reduce the resear ch gap and be input for future research.
Dealing with the current research, there are some theories, concepts and methods described. They are as follows. Legitimacy is a psychological st ate of partiality of people and groups of people who are very sensitive to the symptoms of the surrounding environment, both physical and non-physical (Hadi, 2011). According to Dowling & Pfeffer, (1975), legitimacy is the most impor t ant t hing for an organisation, social norms and values emphasise the limits, and the reaction to these limits encourages the importance of organisational behaviour concerning the environment. Legitimacy Theory focuses on the interaction between the company and the community. It is based on the view that a company should create harmony between the social values and the norms of behaviour that exist in the social system. As long as the two value systems are in harmony, this can be seen as the company's legitimacy. However, when t here is actual disharmony between the two value systems, there will be a t hr eat t o t he legitimacy of the company (Dowling & Pfeffer, 1975).
Stakeholders are interested parties in the company, which include employees, consumers, suppliers, the community, the government as a regulator, shareholders, creditors, and others. According to Anggraeni, (2011), Stakeholders Theory states that companies are not entities that only operate for their interests, but provide benefits for stakeholders, government, and society. In the perspective of stakeholder theory, society and the environment are the company's core stakeholders that must be considered. Stakeholder theory relat es to the concept of corporate social responsibility where corporate responsibility is not only limited to maximising the profits and interests of shareholders but must also consider the community, customers and suppliers as part of the company's operations. The assumptions of stakeholder theory are built on the statement that the company develops to be very large and causes the community to be very related and pay attention to the company, so the company needs to show accountability and responsibility more broadly and not limited to shareholders.
Signal theory is a theory that explains why companies have t he d r ive t o provide financial statement information to external parties (Sari & Zuhrohtun, 2006). One of the information that must be disclosed by the company is information about corporate social responsibility. The company made CSR disclosures in the hope of increasing the company's reputation and value (Rustiarini, 2010). Information about CSR disclosure is a signal to the company to communicate the company's performance in the long run, because CSR is related to acceptability and sustainability, which means that the company is accepted and sustainable to run somewhere long-term, who do not and disclose CSR activities (Adisusilo & Sudarno, 2011).
Tax avoidance is part of tax planning that is done to minimise tax payments. Pohan, (2011) states that tax avoidance is a tax avoidance effort that is carried out legally and safely for taxpayers without conflicting wit h applicable taxation provisions because the methods and techniques used are by utilising the weaknesses contained in Tax Laws and Regulations. Tax avoidance is an effort made by every company manager to reduce the tax burden, and even t axes ar e not paid to the government to take advantage of certain individuals or groups.
According to Kasmir, (2014), leverage is a ratio used to measure the extent to which a company's assets are financed by debt, meaning how much debt burden the company bears compared to its assets, or this ratio is to measur e the company's ability to pay all of its obligations both short and long term (total debt / total assets). The problem of leverage arises because companies use debt which causes the company to bear a fixed burden. This ratio is the debt t o t otal asset s ratio, Debt to Equity Ratio, fixed charge coverage ratio and debt service coverage. A percentage of total debt measures financial leverage to the company's equity in a period, also called Debt to Equity Ratio (DER).
Hadi, (2011) argues that Corporate Social Responsibility (CSR) is a for m of action that departs from ethical considerations of companies that are dir ect ed to improve the economy, which is accompanied by improving the quality of life for employees and their families, as well as improving the quality of life of surrounding communities and communities as a whole. Corporate social responsibility (CSR) is a form of social responsibility given by a company to increase the commercial value of a company without leaving ethical values for the quality of the environment and society. Corporate Social Responsibility (CSR) is a phenomenon of corporate strategy that accommodates the needs and interests of its stakeholders. Based on the description, the hypothesis in this study is as follows: H2: Leverage influences Tax avoidance which is moderated by Corpor at e Social Responsibility (CSR).
The model of the current research is shown in the following figure. Companies that do not have outlier data. Based on these criteria, seven companies were observed for five years. The analytical method used in this study is a moderated regression analysis with the formula: ETR = α + β1DER + ε ………………………………………………………………….(1) ETR = α + β2DER + β3DER*CSRDI + ε…………….……………………………….. (2) The independent variable is leverage. A percentage of total debt measures financial leverage to the company's equity in a period, also called Debt to Equity Ratio (DER). DER is a measure used in analysing the amount of collateral fo r creditors. The greater the DER reflects the relatively high risk of t he company. The dependent variable, namely tax avoidance, is an attempt to reduce, or even eliminate the tax debt that must be paid by the company by not violating existing laws. Low ETR indicates a lower income tax burden than pre-tax income. The formula for calculating the effective tax rate is as follows:

RESULT AND DISCUSSION
The normality test aims to test whether, in the regression model, confounding or residual variables have a normal distribution. A good regression model is having normal or near-normal data distribution (Ghozali, 2016). In this study, the normality test was detected by analysing the normal probability plot gr aph and the non-parametric Kolmogorov-Smirnov Z analysis (1-Sample K-S). This statistical test is normally distributed if the significant residual value in the Kolmogorov-Smirnov test is more than 0.05. The results of the normality test with the Kolmogorov-Smirnov (K-S) test, can be seen in the table as follows: Multicollinearity test aims to test whether the regression model found a correlation between independent variables. To detect the existence of multicollinearity problems can be done by looking at the tolerance value and  Based on the Table 2, the results of the data processing show that ther e is no single variable that has a tolerance value below 0.10 and the value of Variance Inflation Factor (VIF) above 10. Then it can be concluded that in this study multicollinearity did not occur.
The autocorrelation test aims to assess whether a linear regression mod el correlates with the fault error in the t period and the error in the previous per iod (t-1): autocorrelation test or residual independent assumptions using the Durbin-Watson method. The Durbin-Watson test is only used for first-order autocorrelation (first-order autocorrelation) and only requir es t he existence of intercepts (constants), in the regression model and no more between independent variables (Ghozali, 2016). The results of the autocorrelation test can be seen in the Table 3. The data shows that the Durbin-Watson (D-W) test obtained was 1, 275. Based on the theory in the research model, there is no auto-correlation, because the D-W value of 1.275 is between -2 to 2, which means that it is free from the auto-correlation problem.
Heterokedastisitas test aims to test whether the regression mod el occur s variance invariance from residuals of one observation to another. Heteroscedasticity symptoms are detected using scatterplot charts. The results of the heteroscedasticity test can be seen in the figure below.

Source: Research Data, 2020
Based on the classic assumption test that has been done, it can be seen that the data in this study do not have the problem of classical assumptions. Therefore, the available data meet the requirements to use a simple and moderation regression model. In this study, two linear regression equat ions ar e consisting of simple linear regression and moderation linear regression contained in the interaction element. Next, in the first regression, the result of the moderation linear regression in which there are elements of int eraction for t he second equation are as follows: ETR = 0,240 + 0,039DER while in the second regression, it is: ETR= 0,241+0,436DER+1,103DER*CSRDI Based on the results, it is stated that the First Hypothesis Testing Result s, H1 Leverage affects Tax avoidance. It is seen in the following Based on the Table 4, it can be concluded that the results of the t-test between leverage against tax avoidance are obtained count 2.672> greater than ttable that is 2.037 with a significance value of 0.012 <0.05 which means H 1 is accepted, so it shows that leverage affects tax avoidance. Based on the calculations performed in the table shows that the coefficient of determination (R2) proposed from the Adjusted R-Square value of 15.3% variable Y (tax avoidance) can be explained significantly by DER. Other variables outside the model explain the remaining 84.7% generated from (100% -15.3%).
The results obtained from testing the first hypothesis indicate that the higher the amount of funding from third party debt used by the company and the higher the interest costs arising from the debt. The higher interest costs will have the effect of reducing the company's tax burden. Thus the higher the value of leverage, the level of tax avoidance by companies will be higher. The results of this study support the research (Zahirah et al., 2017). This positive r elationship occurs because companies that have high debt will get tax incentives in the for m of discounts on loan interest. The results of this study are also in line with the view of the trade-off theory. It states that the use of debt by companies can be used for tax savings. The tax savings are in the form of interest costs that can be deducted from taxable income.
Thus, companies that have high tax rates can increase the portion of d ebt more than companies that have low tax rates. However, this study is not in line with the results of Hidayat, (2018), who said that leverage does not affect tax avoidance. Second Hypothesis Testing Results, H2: Leverage affects Tax avoidance which is moderated by Corporate Social Responsibility (CSR).
The results of this study support the theory of legitimacy and stakeholders, where companies must consider the satisfaction of all parties in implementing company activities and decision making. The higher the level of CSR disclosure, the higher the tax avoidance carried out by the company. Fulfilment of CSR obligations is carried out by companies to cover up the company's image so that it merely looks good, has the support of the community and the environment.
Companies in the consumer goods industry sector on average have high debt and interest expenses, of course, giving the effect of reducing the company's tax burden. With a high level of CSR disclosure, as if the company has fulfilled its obligations, the company's image looks good and has the support of the community and the environment, then the company that conducts tax evasion in the presence of CSR will get higher levels of tax avoidance.

CONCLUSION
Based on the results of data analysis regarding the DER proxied leverage against tax avoidance with CSR as a moderating variable, it can be concluded as follows, Leverage affects tax avoidance. It means that the higher the leverage is, the higher the level of tax avoidance. Corporate Social Responsibility (CSR) strengthens the relationship of Leverage Against Tax avoidance, meaning t hat the more increased CSR disclosure, the more strengthened the relationship between leverage and tax avoidance.
Next, undeniable, the current research has its limitation. The research limitations are presented as follows. The variables in this study only use two variables, namely Leverage (DER), and corporate social responsibility (CSR). But, likely, there are still many other variables that can affect tax avoidance . This study only uses manufacturing companies in the consumer goods industry sector as a sample, so the results cannot be used as a reference for other companies on the Indonesia Stock Exchange (IDX).
Finally, based on the discussion and conclusion, it is suggested as follows. First, future studies are suggested to be able to add independent variables and use measurements other than those used in this study. Second, in the next research, it is expected to use other types of companies other than manufacturing companies in the consumer goods industry to get more developed results.