Can Other Comprehensive Income be Used for Tax Avoidance?

Other Comprehensive Income (OCI) is the impact of applying fair value accounting, namely the difference between the fair value of assets (liabilities) and their carrying values. Uncertainty about the time and amount of OCI which is a medium for tax avoidance. This study provides empirical evidence on whether OCI can be used for tax avoidance, data from 504 companies listed on the IDX for 2016 – 2020. The results show that companies in Indonesia do not carry out earnings management for tax avoidance through OCI, companies in Indonesia are consistent in realizing OCI according to the plan for the previous period, and there is no time delay or change in the amount of OCI realization. The novelty of this study lies in the effect of OCI on tax avoidance, in addition to previous literature on the influence of governance, political connections, foreign interests, legal systems, and CSR.


INTRODUCTION
Tax avoidance is an action by taxpayers to avoid or reduce tax obligations through the use of loopholes in the tax law. The practice of tax avoidance does not violate the tax law, but this action is contrary to the purpose of the tax law, which is to obtain tax revenue to finance the state's operations in development. Unfortunately, tax avoidance has not been explicitly regulated in the tax laws in Indonesia [26]. Some examples of tax avoidance behavior include (1) borrowing funds from banks that are held for operations so that they do not increase income, but must pay bank interest expenses which can reduce the company's tax burden, (2) attempting to increase items and nominal amounts in operating expenses, (3) taking advantage of PP No. 23 of 2018 for MSMEs by breaking up gross income of less than IDR 4.8 billion a year, and (4) delaying, reducing, or withholding purchases and activities other potential tax burdens, including through OCI (amount and timing of recognition and realization).
Tax avoidance behavior is an important topic to be studied continuously. Tax avoidance behavior is a sustainability issue that must be taken seriously, it is not only a problem for the tax authorities, but also a general social issue related to non-compliance with the law regarding the obligation to pay taxes [3]. A literature study in articles published in journals indexed by Scimago during the period 2005 -2019 concluded that research on tax avoidance is still interesting to be studied by researchers in various countries, and tax avoidance practices occur in many countries [28]. A study conducted by [33] shows that the level of tax avoidance of multinational companies in the United States and 12 countries in Europe is still high, and specifically for companies in Europe, there is a downward trend in tax avoidance from time to time. Empirical evidence that the level of tax avoidance in Indonesia is still high [27], as well as multinational companies in Indonesia [36]. The realization of state revenues from the tax sector also missed the target. The tax amnesty policy failed to reach the target [30]. The 2020 tax revenue target of IDR 1,198.82 trillion (Perpres 72 of 2020) was not achieved, the realization of tax revenue was 89% of the target, with 55% supported by corporate income tax [34]. This research was motivated by a change in the format of the presentation of the income statement, with the addition of other comprehensive income (OCI) along with net income. OCI is an unrealized gain (loss) due to an increase (decrease) in the fair value of assets and liabilities in the presentation period. OCI appears in the income statement since Indonesian SAK converges with IFRS, precisely since assets and liabilities are presented at fair value. The convergence of IFRS on Indonesian SAK has proven to be able to increase the value of the relevance of financial statements for users [10]. In this study, the value of relevance is indicated by the ability of accounting information in financial statements in the form of the book value of equity, earnings per share, and operating cash flow in influencing stock returns after the financial statements are published. The ability to influence stock returns shows the ability of accounting information to influence user decisions. In line with [10], OCI and profit attribution were able to increase the value relevance of financial statements in Indonesia [16]. OCI and earnings attribution can moderate earnings and cash flow information in influencing stock returns. OCI is a form of application of fair value accounting which is presented in the income statement since Indonesian SAK converges with IFRS. The difference between the fair value of assets and the difference between the fair value of liabilities and their book value is proven to have relevance because it can predict future earnings [23] and affect stock returns [29]. OCI consists of adjustments to the fair value of financial assets, cash flow hedges, translation of financial statements of overseas business units, revalued property and equipment, and actuarial differences in defined benefit plans. OCI is presented in the income statement after net income, and is presented separately based on the potential to be realized and will not be realized in the next period. OCI realization means that the item is sold (asset OCI) or repaid (liabilities OCI), so the OCI realization will affect net income (realized gain-loss) and cash flow. OCI recognition flexibility (when to revaluate assets) and flexibility to realize OCI (how and when to sell financial assets available for sale), can be a loophole for tax avoidance.
The tax referred to in this study is not a final tax on OCI. So tax avoidance in this study is not tax avoidance on final OCI taxes but taxes on corporate income (PPh 21 Corporate). However, in entering the OCI value as research data, it is the net OCI value after tax (final). Financial accounting standards in Indonesia require the presentation of OCI items in the income statement for period t by separating them into 2 groups, namely: (1) the "to be reclassified to net income" group, the OCI items presented in this group are OCI items which in period t+1 will be realized (assets are sold or liabilities are paid off), resulting in gains (losses) on the realization of assets (liabilities) which are presented as income so that it affects the value of net income (net income), and net income as the basis for determining corporate income tax. Herein lies the relationship between OCI and corporate income tax 21, the difference in the fair value adjustment of assets (liabilities) is presented as OCI in period t, and when the assets (liabilities) are realized in period t+1, realized gain is recognized as something that affects net income t+1 and income tax t+1. The company can delay or reduce the realization of assets (liabilities) in period t+1 to generate net income according to management's interests, including motivation for tax avoidance. (2) Group "which will not be reclassified to net income". The OCI items presented in this group are OCI items that in the t+1 period will not be realized (assets are not sold or liabilities are not repaid). The OCI items in this group do not affect net income in either period t or t+1, so they cannot be used for tax avoidance. (Revisi atas masukan dari reviewer A no 2.a).
In the period of presentation, OCI is not related to cash flows and net income, because OCI arises from the difference between the adjustment of the fair value of assets and liabilities from their book value. SAK requires OCI to be reported in the income statement along with net income, and to increase the value relevance of presenting OCI in the income statement, SAK requires companies to classify OCI items that will be reclassified and will not be reclassified based on whether there is a plan for the next period to be realized or not to be realized. Realized OCI means that assets have been sold and/or liabilities have been paid off so that the realization has an impact on cash flows and net income from gains or losses on the difference between the carrying amount and the realized fair value. However, due to the uncertainty of the amount and timing of the realization, the company can postpone, cancel, or increase (decrease) the planned amount of OCI realization, even though in the previous period it was presented in the group to be reclassified. This change in the OCI realization plan could be due to the purpose of obtaining a net profit value following the interests of management, including the motivation for tax avoidance. This study intends to examine the role of OCI in tax avoidance, through the time and amount of OCI realization. This study is important to do to provide empirical evidence of the role of OCI in tax avoidance, considering that there is no certainty in the number and timing of the realization of OCI items presented in the group to be reclassified, allowing management to cancel the realization plan for the value of net income following its interests, including tax evasion efforts.
OCI is the difference between the fair value of assets and liabilities and their historical value, which consists of adjustments to the fair value of financial assets, hedging contracts, translation of the financial statements of overseas business units, revaluation of fixed assets, and actuarial differences in post-employment benefits. Before 2012, or before accounting standards in Indonesia converged with IFRS, OCI was not reported in the income statement, but on the equity side of the balance sheet. The addition of the presentation of OCI in the income statement along with net income has proven to be able to increase the relevance value for users in Indonesia [23] and Malaysia [35]. On the other hand, the presentation of OCI can be used for earnings management and tax avoidance in Israel [2] and China [37]. Likewise, Greece [11] found evidence that tax pressures encourage companies to carry out earnings management.
Research on factors that influence tax avoidance, so far the majority are in the form of implementing corporate governance factors and company characteristics such as company size and leverage levels, and there are specific ones regarding the company's relationship with practical politics [30], foreign interest intervention, and the country's legal system [36]. However, research on the effect of OCI on tax avoidance has not been studied, even though there is no certainty about the amount and timing of OCI realization, which can be a medium for presenting the value of net income according to certain interests, including efforts to avoid avoidance.
The moderating role of tax avoidance in the effect of OCI presentation on dividend payments in 508 companies listed on the Tel Aviv Stock Exchange in Israel during the period 2001 -2012. The results show that dividend payments based on the amount of OCI are mostly carried out by companies in Israel, not only based on net income. Paying dividends based on OCI motivates company management to carry out earnings management and tax avoidance [2]. Unfortunately, this research by [2] found several findings: first, only one OCI item is used, namely the adjustment of the fair value of financial assets, even though there are five OCI items. Second, it does not involve the reclassification of OCI. Third, in measuring OCI, by using OCIbased ROA, namely OCI from financial assets divided by total assets, whereas with the obligation to present the OCI reclassification, it can be seen the potential for tax avoidance through delaying or reducing the realization of OCI, and finally, in examining tax avoidance behavior, it does not relate to the implementation of corporate governance and connections. politics, even though many studies have proven that tax avoidance is caused by the weak implementation of GCG and political connections.
The position of this research is to develop research by [2]. This research proposes three new problems. First, there are five OCI items used, using all OCI items according to Indonesian SAK. Second, involves the reclassification of OCI in measuring OCI, using the ratio of the planned OCI reclassification to actual OCI realization. Third, it involves GCG variables, political connections, and company characteristics as control variables in assessing the use of OCI for tax avoidance (Revisi atas masukan dari reviewer B no 7 → Commented This research is expected to contribute to three parties to the academic literature in the field of financial accounting and taxation and can be used as an input for developing further research related to OCI and tax avoidance. It is also expected to be of help for the Director General of Taxes in preparing the regulations that prevent tax avoidance by companies through delaying or reducing OCI recognition and the realization of group OCI items to be reclassified. Third, for DSAK IAI in preparing PSAK on standardization and certainty of the amount and timing of recognition. OCI and the realization of group OCI items to be reclassified. (Revisi atas masukan dari reviewer B no 7 → Commented [X2]).

Tax Avoidance
Agency theory [9] states that companies are prone to conflicts of interest and information asymmetry that can cause financial statements to be biased, potentially favoring one interest, including the interest of tax avoidance through financial statements. Tax avoidance is an act of avoiding or reducing the total burden of paying taxes through legal tax planning to get the value of taxable profit as desired [2]. Tax avoidance cannot be carried out sequentially over the years because it can be easily detected and is no longer a subtle means for management, for example in the use of OCI for tax avoidance. Time delays and or reductions in the realization of OCI assets (liabilities) in the current period can be detected from the presentation of group OCI items that will be reclassified in the previous period, taking into account the impact of macroeconomic and other external volatility that affects OCI. If the realization of OCI's assets (liabilities) in the current period is significantly different from the reclassification of OCI in the previous period, it can be suspected that there is an indication of tax avoidance through OCI, in particular the delay of gains from the sale of assets or the realization of OCI. (Revisi atas masukan dari reviewer B no 7 → Commented [X3]).
[25] defines tax avoidance as a strategic choice to avoid the tax burden by utilizing the gray area of tax law regulations. The government (Directorate General of Taxes) does not have access as broad as the management's regarding the information on the company's economic activities and transactions, including the financial statements that are the basis for determining the amount of corporate income tax payments. The government has an interest in maximizing state revenues from the tax sector, including corporate or corporate income taxes, which are established, operate, and earn income in a country. However, the government's goal is not supported by company management which seeks to avoid or reduce tax payments by exploiting loopholes in the law [26].
Agency theory is the grand theory in this study. Agency theory can be used to analyze the relationship between OCI and corporate income tax (PPh 21), namely the existence of information asymmetry and conflict of interest between company management and the government. The government does not have access to information on companies as broad as management. The government does not have sufficient information on the suitability of the OCI realization plan in t-1 with the actual realization in period t. The government has an interest in the company's financial statements to check the suitability of the tax burden payable from the net profit (as taxable income) with the amount that should be. Management has an interest in earnings management for tax avoidance by deferring or reducing the amount of realized assets which in the previous period the fair value adjustment of assets was reported as OCI. (Revisi atas masukan dari reviewer A no 3).
The philosophy that underlies the actions of the company's management to avoid tax is that the company is established and operates in the jurisdiction and geographical area of a country, earning income and welfare from the results of operational activities in a country. The state through the government has the right to levy taxes on income earned by companies because the state provides the market share, operating rights, labor, raw mate-rials, and various other resources. However, due to certain reasons, such as optimizing dividends, bonuses for management and employees, business expansion, and the intentional unwillingness to pay taxes, the company's management took tax avoidance actions. (Revisi atas masukan dari reviewer A no 4).
Dividend catering theory states that management will try to maximize the company's income to maximize dividend payments to owners, and one way to do that is to avoid paying taxes [1]. Management and owners have an interest in avoiding tax payments to maximize cash flows that can be transferred to owners in the form of dividends because tax payments can reduce retained earnings [31] and maximize cash flows for operating activities for the next period to increase prices, shares, and company value. Tax avoidance through financial statements is a form of conflict of interest and information asymmetry between the government, management, and owners [24].
The relationship between agency theory, dividend catering theory, and management incentives are as follows: agency theory states that in publicly listed companies there is information asymmetry and conflicts of interest between stakeholders, including company management and the government. The government does not have access to information on companies as broad as management. The government did not obtain sufficient information on the suitability of the OCI realization plan in t-1 with the actual realization in period t. The government has an interest in the company's financial statements to check the suitability of the tax burden payable from the net profit (as taxable income) with the amount that should be. The dividend catering theory states that management seeks to maximize dividend payments to owners. One of the efforts made to be able to pay dividends is tax avoidance through earnings management. Management engineered financial statements including earnings management to avoid taxes. On the other hand, the government with limited access to information used financial statements to determine the correctness of calculating the tax burden paid by the company. The government has an interest in maximizing state revenues from the tax sector (including taxes from companies), while company management seeks to avoid taxes by paying dividends to owners and bonuses for their management. (Revisi atas masukan dari reviewer B no 2).
The role of managers in tax avoidance as an example of a conflict of interest between management and the government includes 1) Misuse of managerial positions to avoid tax through the selection of tax payment methods and schemes. 2) Opportunistic managers can regulate company business transactions to reduce the tax burden and transfer it to the post. -post other resources for private interests [27]. 3) Multinational companies are more likely to do tax avoidance than national companies [36]. Managers of multinational companies can do tax avoidance, for example by utilizing transfer pricing with subsidiaries to the parent [7].
The company's motivation for tax avoidance includes maximizing dividend payments [31], maximizing bonus incentives for managers [4], maximizing stock prices and firm value [6], regulating fluctuations in cash flow, increasing the company's liquidity and growth opportunities [32], and saving company expenses or burdens [12]. Tax avoidance can save expenses by one-third of pre-tax profit, and can align with management's interests [3].
Tax sanctions are positively correlated with the level of tax compliance. The lower the fine for violating the law, the lower the level of public compliance (including companies) with legal compliance, and the lower the public demand for transparency, which results in higher aggressiveness of the company in tax evasion acts [13]. Dividend payments have increased since the IFRS convergence era in Israel, which is reflected in the presentation of OCI presented in the income statement and can encourage management to carry out earnings management and tax avoidance [2].
The influence of foreign investors' interests, the level of investor protection, and the legal system on tax avoidance in countries with Code-Law (Indonesia, Netherlands, China) and Common-Law (United States, Singapore, Malaysia) legal systems with the research period 2009 -2016. The results show that multinational companies are more likely to do tax avoidance than national companies in the same country. Companies operating in countries with a high level of investor protection and a Common-Law legal system tend to evade tax more frequently than companies operating in countries with weak investor protection levels and with a Code-Law legal system. This is because companies operating in countries with a high level of investor protection and the Common-Law legal system have stronger ties to investors, thus making more efforts to increase dividend payments, one of which is by avoiding tax payments [36].
A study on state-owned companies in China shows that the implementation of CSR is proven to be able to reduce earnings management and tax avoidance [25]. The same thing also happened in Germany, research by [7] proves that the better the implementation of CSR in oligopolistic companies, the lower the practice of tax avoidance. A study in Brazil concluded that companies with a greater concentration of share ownership by management tend to avoid tax evasion. Companies with greater management ownership do not dare to take risks from the negative impact of tax evasion practices. This study also concludes that the level of concentration of certain shareholdings affects the level of tax avoidance [1].

Other Comprehensive Income
The purpose of financial statements is to provide information to company stakeholders in making decisions. For financial reports to be relevant to the needs of stakeholders, financial reports must be of high quality. One indicator of the quality of financial reports is having a representative value, which is being able to present assets according to actual conditions. The presentation of assets based on historical value, which was previously widely used as a basic assumption, is deemed to have a weakness, because it no longer follows the current condition of assets, especially for financial assets and liabilities that are sensitive to changes in the macro-economy of the country in which the company operates. Public pressure on the application of fair value accounting grew stronger, until finally, Indonesian SAK converged for the first time with IFRS, precisely since the regulation on the use of SAK Effective as of June 1, 2012, which was adopted from IFRS on January 1, 2009. One of the impacts of the implementation of this SAK (and the use of fair value) is the emergence of other comprehensive income (OCI), which is the difference between the fair values of assets and liabilities at the date of presentation of the financial statements and their carrying values. OCI consists of five items, namely fair value adjustments: 1) available-for-sale financial assets, 2) cash flow hedging contracts, 3) translation of the financial statements of overseas business units, 4) defined benefit plans, and 5) fixed assets and revalued intangibles.
The application of all-inclusive income recognition has an impact on the location of the OCI presentation, which is presented together with net income in the income statement, although this OCI is only an adjustment, unrealized income, not related to cash flows, has high fluctuations in value because it is sensitive. with external conditions and the value cannot be controlled by company management. The emergence of OCI is caused by internal factors such as company size and asset ownership, as well as external factors such as macroeconomics in the form of interest rates, inflation, and exchange rates [17]. OCI information is used by users to make decisions, which is reflected in its effect on stock returns [16]. OCI for the current period can be used to assess cash flow prospects [20] and can predict future comprehensive income [21]. The presentation of the OCI reclassification minimizes earnings management and is proven to improve earnings quality as measured by ERC [22]. OCI and net income is called total comprehensive income, and both are attributed to owners and presented in the net income statement. This gave rise to the idea of modifying financial performance measures, such as ROA based on comprehensive income [18] and ROE with comprehensive income and attributable equity [21]. The use of the formulation of ROA and ROE based on comprehensive income has proven to be more capable of measuring asset utilization and affects firm value [19].
OCI characteristics are sensitive to external changes and have not been realized. So to increase the value of relevance, SAK requires companies to group OCI items based on the potential to and will not be realized in the future period, which are further termed the "to be reclassified" and "those that will be reclassified". will not be reclassified". Items that will be reclassified mean that the OCI item will be sold or paid off so that it has an impact on cash flow and net income. However, it is unclear how many rupiahs will be realized (whether it is the same as the plan with the certainty of realization) and when it will be realized, which can be a medium for earnings management and tax avoidance.
In period t, OCI is not related to income tax PPh 21 period t. However, OCI items that are presented in the "to be reclassified to net income" group in period t can be used as a medium for tax avoidance in period t+1 by delaying or reducing the number of assets (liabilities) realized in period t+1 to get the profit value net and desired tax payable. This is the conceptual correlation between OCI and corporate income tax (PPh 21). So the gain (loss) on adjusting the fair value of assets (liabilities) for period t OCI group "to be reclassified to net income" can affect net income for period t+1 and income tax 21 period t+1. (Revisi atas masukan dari reviewer A no 2.a). Based on this, the conceptual framework in this study is as follows:  The tax avoidance variable is measured by three indicators: 1) Effective Tax Rate, t (ETR), namely the ratio of tax expense in period t to net income before tax period t, following [5; 8], 2) Long Term Cash Effective Tax Rate, t (LTC-ETR), which is the ratio of tax expense for the period t to t+5 to net income before tax period up to t+5, following [5] and 3) Cash Flow Tax Rate i,t (CF-TR), which is the ratio of tax expense for period t to operating cash flow for period t, follows [24].

RESEARCH METHOD
The study uses secondary data extracted from financial statements of 504 companies listed on the Indonesia Stock Exchange for 5 years, namely 2016 -2020. The sample was selected using the purposive sampling method with the following criteria:  The ratio of the realization of OCI items in period t to the planned OCI items will be realized (which will be reclassified) in period t-1.
Shareholding by the government and a parent entity controlled by the government (PCON).

Number of commissioners
Company Characteristics Company size (SIZE) [15; 30] Logarithms natural of total asset Leverage (LEV) [30] The following equation model 1 is used to test hypothesis 1. OCI can be used for tax avoidance if the OCI coefficient 1 is significant at the 5% level.
The following equation 2 is used to examine the effect of each OCI item on tax avoidance: The following equation 3 is used to see how much OCI's sensitivity is planned to be realized in period t-1 in period t and the actual realization of OCI in period t, while taking into account macroeconomic factors such as inflation rate, interest rate, and the IDR exchange rate against USD in period t that affects the realized fair value of OCI:

(Eq.3)
Description = α0: constant; β1 -4: coefficient; Real OCIt : real OCI realization in period t, which in period t-1 is planned to be realized (in period t-1, it is presented in groups to be reclassified). Real realization of OCI for period t is seen from the value of gain (loss) on sale of assets and or settlement of liabilities reported in net income of period t, such as gain (loss) on disposal of financial assets available for sale in period t, and reclassification items other OCIs; Reclass OCIt-1 : OCI item for period t-1 which is planned to be realized in period t. In period t-1 these items are presented to the OCI groups to be reclassified. INFLAt : inflation rate in period t; INTERESTt : interest rate in period t; RATEt : the average exchange rate of IDR against USD in period t; ε: error.
The significance of the coefficient β1 Reclass OCIt-1 in equation 3 shows the consistency between what was planned and the reality that happened, and whether the planned OCI item will be realized, will be realized in the next period. If the coefficient β1 Rec OCIt-1 is significant, it means that it is proven that the planned OCI to be realized in period t has indeed been realized, but if it turns out that the OCIRt-1 coefficient is not significant, it means that there is an indication of the difference between the plan and the real realization, either delayed or realized by a different amount than planned. Table 3 shows the results of descriptive statistics. Corporate tax avoidance (CTA) is measured by three indicators. CTA 1 is indicated by the value of the Effective Tax Rate, t (ETR) [8] which has a mean value of 0,2198. This means that 21,98% of the net profit generated is a corporate tax liability. Avoid Tax Rate (ATR) = Rate Tax Applied -ETR [27], if the ATR value is positive, then the tax rate paid by the company is lower than the applicable tax rate. Conversely, if the ATR value is negative, the tax rate paid is greater than the applicable tax rate, where the tax rate applied is 25%. ATR = 25% -21,98% = 3,019%. These results show that the average company listed on the Indonesia Stock Exchange for the 2016-2020 period has an ATR of 3,019%, meaning that the company avoids tax at a rate of 3,019% lower than the normal tax rate applicable in Indonesia. CTA 2 which is indicated by the value of Long Term Cash Effective Tax Rate, t (LTC-ETR) which has a mean value of 0,271. This means that in the longer term of five years, an average of 27,1% of the total cost is paid to the tax authorities. CTA 3 is indicated by the value of Cash Flow Tax Rate i,t (CF-TR), which has a mean value of 0,213. This means that an average of 21.3% of the total operating cash is paid to the tax authorities.

RESULTS AND DISCUSSION
The mean of OCI is 0,874, which means that the ratio of realization of OCI items in period t to the OCI item plans to be realized (which will be reclassified) in period t-1 is 87,4%. The company consistently realizes OCI in period t as planned in period t-1. The ratio is not 100% due to the difference between the realized fair value in period t and the reclassification fair value in period t-1 caused by macroeconomic changes such as inflation, exchange rates, and interest rates. The mean of KOMIN is 0,346, meaning that on average, publicly listed companies in Indonesia during the study period had 34,6% independent commissioners and following the regulations of the capital market authority, a minimum of 30% of the total number of commissioners, and the mean of DEKOM 4, which means the average number of commissioners is 4. There are an average of 4 people who are tasked with supervising compliance with the company's guidelines. The mean of KINS is 0,602, meaning that the number of institutional ownership of publicly traded companies in Indonesia during the study period is an average of 60,2%. Table 4 shows the results of the correlation analysis. OCI is positively correlated with all indicators of tax avoidance, meaning that the higher the ownership of assets and liabilities, the greater the adjustment of historical values or carrying values to their fair values, and the greater the potential number of OCI items that can be delayed by time to be realized and or reduced by the amount realized according to the total value expected net income and tax expense. PCON is negatively correlated with all indicators of tax avoidance, meaning that the stronger the political connection of corporate management manifested in the number of government shareholdings, the lower the value of the tax avoidance proxy.
KOMIN, KINS, and DEKOM have a strong negative correlation with all indicators of tax avoidance, this indicates that the better the application of the principles of good corporate governance, the more obedient to government regulations, one of which supports efforts to optimize state revenues from the tax sector, so as not to evade taxes. Size is positively correlated with OCI, meaning that the larger the size of the company, the more economic transactions that occur in the company's activities and the amount of asset ownership, so it is necessary to adjust the carrying value to fair value, which gives rise to OCI recognition. Table 5 shows the results of the multiple linear regression analysis. Panel A is the result of regression analysis on data from all types of industries (financial and non-financial) as many as 504 companies for 5 years with a total of 2,520 observational data. Panel B is the result of regression analysis on data only from the type of financial industry, with a total of 67 companies for 5 years with a total of 335 observational data. Table 5 equation 1 shows that the OCI coefficient value on the three tax avoidance indicators in both panels A and B is not significant, meaning that OCI does not affect tax avoidance.
Companies in Indonesia do not use OCI to avoid paying taxes. Equation 1 shows that political connections and the application of good governance principles have a negative effect on tax avoidance. Equation 2 shows that the OCI item which has a coefficient of influence on tax avoidance is greater than other items, namely financial assets in the available-for-sale category, even though this item is significant at the 10% level for financial sector companies. Equation 3 shows that the reclassification of OCI in the previous period has a significant effect on the realization of OCI for the current period, meaning that the realization of OCI in the current period follows the previous plan, and companies in Indonesia do not delay or reduce the plan for the realization of OCI for tax avoidance.
Panel B was built to look specifically at the adjusted R2 value and the significance of F by only involving financial sector companies. From table 6, it can be seen that the value of adjusted R2 in panel B is greater than in panel A, this means that the feasibility and ability of the model to explain the effect of OCI on tax avoidance is better in data that only applies to financial sector companies, than all types of industries. This is because this sector has the most available-for-sale financial assets and most frequent OCI items from the fair value adjustment of available-for-sale financial assets in the group to   be reclassified. This is supported by the coefficient value of each OCI item in equation 2, that the available-for-sale category financial asset item has a greater coefficient than other OCI items, meaning that when compared to other OCI items, the available-for-sale category financial asset item most have the potential to be used for tax evasion, through delays in time or by reducing the number of sales of financial assets. This is in line with the findings in Israel [2].
OCI cannot be used for tax avoidance, this is because in the period in which it occurs is not related to cash and net income. It is only an adjustment of the historical value of assets and liabilities to their fair value at the date of presentation of the financial statements. The gain (loss) on the fair value adjustment of these assets and liabilities has not been realized, so it does not increase or decrease net income before tax. The OCI value is highly volatile and sensitive to changes in macroeconomic fundamentals beyond the company's internal control. OCI is not the result of operations management performance. The amount of OCI is smaller than net income in forming comprehensive income, as well as the nature of its persistence.
OCI cannot be used for tax avoidance, this means that companies in Indonesia are consistent in realizing OCI items in accordance with the plan for the previous period presented in the items to be reclassified. Companies in Indonesia do not perform earnings management for tax avoidance purposes. The companies are not delaying the realization time or increasing/decreasing the number of realizations of OCI items as planned in the previous period. Regarding the difference in the value of OCI items that were presented in the group to be reclassified with realizable values in the current period, it was caused by changes in market value due to macroeconomics (exchange rate, inflation, interest) which affected the fair value of the OCI item. The presentation of OCI in the income statement increases the value relevance, because it is more representative of valuing assets and liabilities at fair value, and it is proven that companies in Indonesia do not use OCI for earnings management and tax avoidance. This is contrary to the findings in China [37].
OCI cannot be used for tax avoidance, however, in companies with large amounts of available-forsale financial assets that are presented in the OCI group to be reclassified, the strength of OCI's   influence on tax avoidance is higher. This is in line with the findings in Israel [2]. The strength of the effect is shown from the value of the regression coefficient and adjusted R 2 . This is because financial assets in the available-for-sale category are the most flexible OCI items to be realized, when, and in what amount. There are no contractual restrictions that limit the realization of financial assets in the available-for-sale category, companies can delay, realize immediately, reduce or increase the amount of realization according to their interests, in contrast to other OCI items such as cash flow hedges, whose realization is in accordance with hedging contracts in terms of the amount and the time. Available-forsale financial assets are OCI items that are most often presented in the group to be reclassified, which means that this item is the most realized in the next period, as the name implies "available-for-sale financial assets" this item is held for long-term investment purposes. short, namely capital gains from the difference in the selling price (fair value at the time of sale). When market prices rise, or when companies need fresh funds, these financial assets are realized. The realization of this item can be at any time, there is no contractual limit regarding the amount and time of realization, in contrast to the settlement or receipt of hedging settlements, the realization depends on the date in the contract. The company can delay the realization of available-forsale financial assets or reduce the amount realized in period t (although in period t-1) it is planned to be realized if it is felt that realization in period t has the potential to cause an increase in tax payments. The results of this study also show that the stronger the political connection of the company, the lower the level of tax avoidance, this is in line with the findings of [36; 30], and the better the application of GCG principles, also the lower the level of tax avoidance, this is in line with the findings of [27]. The existence of an independent board of commissioners and institutional share ownership will oversee the operations and management performance, including supervision of tax avoidance practices.

CONCLUSION
The purpose of this study is to provide evidence of whether OCI can be used for tax avoidance, while still involving governance, firm size, and level of leverage as control variables, and to prove which of the five OCI items can specifically be used for tax avoidance. The results showed that OCI cannot be used for tax avoidance. Companies in Indonesia do not carry out earnings management for tax avoidance through OCI. This means that companies in Indonesia are consistent in realizing OCI according to the plan for the previous period, there is no time delay or change in the amount of OCI realization. The available-for-sale financial asset item has a higher impact on tax avoidance than the other four OCI items.
The results of this study are useful for the government, especially the Director General of Taxes (the tax authority in Indonesia) and the compilers of accounting standards in Indonesia (DSAK IAI). The tax authority can detect tax avoidance actions through OCI from delays and or reductions in the number of assets (liabilities) realized in period t, by comparing the realization in period t with OCI items in period t-1 in the group that will be reclassified to net income. The accounting standard-setting authority in Indonesia (DSAK IAI) needs to be regulated by accounting standards regarding the certainty of the number and timing of the realization of OCI items in the group to be reclassified to net income to narrow the opportunities for tax avoidance through OCI, particularly through the gain from the sale of assets in period t which in t-1 the fair value adjustment of the asset is reported as OCI. (Revisi atas masukan dari reviewer B no 7 → Commented [X7]).
The potential for tax avoidance through OCI is not only a time delay and a reduction in the number of realizations of OCI items presented in the group to be reclassified but also the recognition of OCI occurrences such as time delays and a reduction in the number of fixed assets that will be revalued. This study does not examine tax avoidance efforts through the time and amount of revaluation of tangible fixed assets, because this is related to tax avoidance on final tax from revaluation of fixed assets, which is not the scope of this study, tax avoidance in this study is corporate income tax (PPh 21 companies), not tax avoidance on final tax revaluation of fixed assets. (Revisi atas masukan dari reviewer B no 7 → Commented [X8]).Further research is recommended to examine tax avoidance through the recognition of the emergence of OCI from the postponement or reduction of fixed asset revaluation.