Modern positive accounting research began to develop in the 1960s when (Ball & Brown, 1968) and other researchers introduced the use of empirical financial methods in financial accounting (Watts & Zimmerman, 1990). Positive accounting theory developed from several thoughts, one of which was the thought of (Ball & Brown, 1968) about how one can explain the predictive ability of a proposition based on assumptions without adequate verification. Accounting that does not have a sufficient theoretical framework that is comprehensive causes the development of differences in practice in the field (Ball & Brown, 1968).

(Watts & Zimmerman, 1986) mention three positive accounting theory hypotheses: the bonus plan hypothesis, the debt covenant hypothesis, and the political cost hypothesis. The bonus plan hypothesis explains the tendency of companies to choose accounting procedures that can increase current-period profits so that management gets bonuses from the company. The debt covenant hypothesis sees management's efforts to implement accounting policies that maintain the company's debt ratio at a safe level to avoid imposing penalties from creditors. The political cost hypothesis describes management's intention to carry out earnings management practices to minimize political costs.

Agency relations are one of the oldest and most common codified methods of social interaction. It can be said that an agency relationship exists between two or more parties, one party acts as an agent and the other party acts as a principal (Ross, 1973). (Eisenhardt, 1989) concludes that agency theory offers a unique view of information systems, outcome uncertainty, incentives, and risk.

(Jensen & Meckling, 1976) argue that an agency relationship is a contract the company's owner enters into involving other parties (agents) to carry out several functions on their behalf. Agency theory demonstrates that the emergence of various conflicts between the owner (principal) and management (agent) gives rise to agency costs, costs that exist to reduce friction (Morris, 1987). On the other hand, (Umobong & Ironkwe, 2017) argue that agency relationships require delegation of decision-making authority from the owner to the agent.

Also known as creative accounting, earnings management is a purposeful intervention in the financial reporting process to external parties. Management is motivated to carry out creative accounting due to pressure to meet the company's financial expectations (Mutuc et al., 2019). (Healy & Wahlen, 1999) argue that earnings management occurs when managers use financial reporting and transaction preparation judgment to modify financial statements.

(Sosiawan, 2012) states that earnings management adds to the bias in financial statements and can interfere with users of financial statements who believe that the engineered profit figure is an unengineered profit figure. Earnings management itself is a step in reporting manipulated profits for the benefit of company management (Trisnawati et al., 2015).

Based on the definition described above, earnings management is a practice used by company management to utilize the flexibility of accounting rules in preparing financial statements. This flexibility is exploited to meet the needs of management with various objectives that often conflict with the expectations of the company's owners.

(Pradnyani & Astika, 2019) state that firm risk is a condition with some possibilities that cause a company's performance to be lower than expected due to uncertain conditions in the future. Financial leverage is the use of debt as a source of funding at the company's expected rate of return (Firmansyah et al., 2021). In terms of debtholders, the use of financial leverage is one of the risks faced by the company. (Raningsih & Putra, 2015) mention that a high level of leverage ratio reflects that the company has a high risk, especially for the inability to pay off its debts so that investors judge it as not good.

(Bhattacharya et al., 2019) state that from an investor's perspective, there are two stakeholders: debtholders and equity holders. From the perspective of equity holders, firm risk can be measured using total risk and idiosyncratic risk (Chakraborty et al., 2019). Further mentioned by (Bhattacharya et al., 2019) that the company's risk from the equity holder side contains two risks, namely systematic risk and idiosyncratic risk.

Based on these definitions, firm risk can be interpreted as the possibility of circumstances that bring the company to unexpected conditions. This risk is closely related to the company's capital structure, considering that funding using debt or equity has a negative potential for company performance. Excessive use of debt will lead to default risk, while market and company conditions can affect the volatility of company value.

Tax risk is defined as all tax-related risks and uncertainties related to the company's operating, investing, and financing decisions. These uncertainties include tax reporting related to the company's original condition, audit risk by tax authorities, and financial accounting related to income tax payments (Hutchens & Rego, 2013). (Guenther et al., 2013) stated that tax risk is the uncertainty associated with future corporate tax payments. Therefore, tax risk has a different view from tax avoidance and tax aggressiveness.

(Firmansyah & Muliana, 2018) state that the tax risk faced by companies is caused by various company policies used to respond to tax regulations. This response is not always in line with applicable regulations. One example of tax risk a company faces is the tax sanctions imposed by the tax authorities if the company is found to have deviated from taxes, such as not reporting part of the sales proceeds or passively increasing deductible expenses (Wardani & Putriane, 2020).

Referring to several definitions that have been described previously, tax risk can be interpreted as all risks faced by companies in connection with fluctuations in corporate tax payments. This risk can be caused by fluctuations in the company's cash flow, changes in tax rules, or company internal policies.

Based on Islamic finance theory, companies that comply with sharia provisions are subject to several restrictions to maintain their compliance status (Alnori & Alqahtani, 2019). (Naz et al., 2017) state that the principles of Islamic finance are derived from sharia, Islamic law, a law based on the hadith/Sunnah of the Prophet Muhammad PBUH and the Quran, which directs all aspects of the individual and social life of Muslims.

Companies in issuing sharia securities must meet the provisions that have been set. Becoming a company that has sharia securities has the consequence that the company's activities will continue to be monitored by the Sharia Board (Tanno & Putri, 2019). Sharia compliance plays an essential role in ensuring that all company activities align with sharia principles (Zakiah, 2017).

Based on the explanation above, the definition of sharia compliance in this study is conformity with sharia principles that have been determined by the National Sharia Council of the Indonesian Ulema Council (DSN-MUI) concerning the principles of brotherhood (ukhuwah), justice ('adalah), balance, and universal (syumuliyah). Compliance with sharia provisions in the Indonesia Stock Exchange is reflected in the inclusion or absence of issuers in the Indonesian Sharia Stock Index (ISSI) list.

An agency relationship exists between two or more parties, one party acts as an agent and the other party acts as a principal (Ross, 1973). In running the company, management is always faced with risk. From an investors' perspective, firm risk can be divided into debt holders and equity holders (Chakraborty et al., 2019). The use of debt in the company's capital structure reflects the risk from the perspective of the debtholder. (Watts & Zimmerman, 1990) state that in the debt covenant hypothesis, companies with a high debt-to-equity ratio (DER) are predicted to be more likely to use accounting provisions that increase company profits. This is because the higher the DER of a company, the thinner the distance between the company and the limitations in the debt agreement (Kalay, 1982).

The company also encounters risk from the perspective of the equity holder, namely the overall market conditions and events involving the company's management can bring risks that the corporation must bear. These various risks can create severe fluctuations in the profitability and price of the company's shares which are generally the basis for calculating bonuses for the board of directors. The bonus program hypothesis states that managers lured by bonuses are more likely to use various accounting methods that can boost company profits in the related period (Watts & Zimmerman, 1990).

A large firm risk will result in management tending to carry out earnings management. It is in line with the results of research by (Pradnyani & Astika, 2019), (Suhartanto, 2015), and (Religiosa & Surjandari, 2021). Therefore, the first and second hypotheses proposed in this study are:

H1: Firm risk (debtholder) has a positive effect on earnings management

H2: Firm risk (equity holder) has a positive effect on earnings management.

Agency theory offers a unique view of information systems, outcome uncertainty, incentives, and risk (Eisenhardt, 1989). It can be said that an agency relationship exists between two or more parties, one party acts as an agent and the other acts as a principal (Ross, 1973). One of the hypotheses in Positive Accounting Theory proposed by (Watts & Zimmerman, 1986) is the political cost hypothesis. In this hypothesis, it is illustrated that management has the intention to carry out earnings management practices to minimize political costs, one of which is the tax burden.

Referring to the previous study, research on taxation factors on earnings management revolves around deferred tax expense, deferred tax assets, and tax planning. There is still no research on the effect of tax risk on earnings management. However, several researchers have studied the effect of tax risk on firm risk, as previously mentioned that (Pradnyani & Astika, 2019), (Religiosa & Surjandari, 2021), and (Suhartanto, 2015) found that firm risk influences management decisions in carrying out earnings management practices. In that study, (Guenther et al., 2013) and (Hutchens & Rego, 2015) agree that tax risk has a positive effect on company risk. Therefore, the third hypothesis proposed in this study is:

H3: Tax risk has a positive effect on earnings management.

An agency relationship is a contract entered into by company owners involving other parties (agents) to carry out several functions on their behalf (Jensen & Meckling, 1976). The existence of two different parties in one entity can trigger different interests. This difference in interest is in line with the opinion of (Trisnawati et al., 2015) which states that company owners are motivated to maintain an increase in living standards by increasing company profitability while managers are motivated to maximize economic and psychological needs in the form of obtaining investment, debt, job security, or compensation.

Companies on the Indonesia Stock Exchange can be divided into sharia complaints and non-sharia complaint companies. (Ahmed, 2009) mentions several specific provisions in sharia provisions, including funding, investment, operations, and risk management practices. Companies that do not follow sharia provisions are not subject to these restrictions. Sharia provisions, which are also aspects of religiosity, are the foundation of morals and ethics (Kanagaretnam et al., 2015).

Management in companies that tend to be religious and comply with sharia provisions can have better morals. Actions that prioritize morals and ethics will further reduce friction between principals and agents, decreasing earnings management practices. This opinion is in line with the research results of (Kanagaretnam et al., 2015). Although (Hamdi & Zarai, 2012) state that Islamic banks still tend to carry out earnings management, there is no statistical evidence of account manipulation that leads to the avoidance of revenue declines. Based on this, the fourth hypothesis proposed in this study is:

H4: Sharia compliance has a negative effect on earnings management.

The dependent variable in this study is earnings management. The proxy used to measure earnings management in this study is discretionary accruals from the modified-Jones model. Discretionary accrual was chosen because previous studies suggest that this model gives the best results (Kothari et al., 2005).

(Wibawa et al., 2020) describe the calculation of this proxy as follows:

Explanation:

EM : Earnings management

DTA : Discretionary accruals

TACC : Total accruals

NDA : Non-discretionary accruals

Total accruals are calculated using the following proxy:

Explanation:

TACCit: total accruals of company i in period t

NIit : net profit of company i in period t

CFOit : cash flow operation of company i period t

The accrual value is calculated based on the residuals from the simple linear regression equation using the following equation:

Explanation:

TA_i(t-1) : total asset of company i period t-1

∆REV_it : revenue of company i period t minus period t-1

PPE_it : value of fixed assets less accumulated depreciation of company i in period t

β : fitted coefficient obtained from the regression results in the calculation of total accruals

The value of non-discretionary accruals (NDA) is calculated using the following equation:

Explanation:

NDA_it : non-discretionary accruals of company i period t

∆REC_it : total receivables of company i period t minus period t-1

This study has four independent variables: firm risk (debtholder), firm risk (equity holder), tax risk, and sharia compliance. In terms of debtholders, leverage reflects the risks faced by the company. Leverage is measured by using the debt ratio. The leverage proxy in this study follows the model used by (Firmansyah et al., 2021), (Pradnyani & Astika, 2019), and (Titman et al., 2011),

In this study, firm risk (equity holder) is measured by using total risk. The measurement of total risk follows the model used by (Firmansyah & Muliana, 2018), (Guenther et al., 2013), and (Parendra et al., 2020),

TRISK_it = annual standard deviation of monthly stock return of company i period t

Which:

Explanation:

P_t : Share price of company i period t

P_t-1 : Share price of company i period t-1

The proxy used to measure tax risk in this study is the volatility of cash ETR. Cash ETR volatility is calculated based on the standard deviation of a company's cash ETR as used by (Firmansyah & Muliana, 2018), (Guenther et al., 2013), and (Hutchens & Rego, 2015),

CETR_VOL_it = five-year standard deviation of annual cash ETR of company i year t

Which:

Explanation:

Cash Tax Paidit: Total tax payment of company i from year t-4 to t

Pretax Incomeit: Profit before tax of company i from years t-4 to t

Sharia compliance in this study is measured using a dummy variable. The use of dummy variables follows the model used by (Alnori & Alqahtani, 2019) and (Tanno & Putri, 2019),

SHARIA_it = Dummy variable with a value of 1 if the company complies with sharia provisions and a value of 0 if otherwise

Explanation: The value of 1 in dummy variable is given to companies listed in the Indonesian Sharia Stock Index (ISSI)

In this study, two control variables are used, namely profitability and firm size. Profitability is one of the crucial parameters in measuring the achievement of company goals (Noor et al., 2020). The proxy used in measuring profitability in this study follows (Firmansyah et al., 2021) and (Mondayri et al., 2020),

Company size shows how big the company is regarding total assets owned (Januardi & Afrianto, 2017). The proxy used in this study refers to (Firmansyah et al., 2021), (Mondayri et al., 2020), and (Parendra et al., 2020),

Explanation:

Log (total Asset)_it : Natural logarithm of total assets of company i year t

Based on these variables, the model used in this study is as follows:

Explanation:

EM_it : Earnings management of company i period t

LEV_it : Leverage of company i period t

TRISK_it : Total risk of company i period t

CETR_VOL_it : Cash ETR volatility of company i period t

SHARIA_it : Sharia compliance of company i period t

ROA_it : Profitability of company i period t

SIZE_it : Company size of company i period t

α0 : Constant

εit : Error

The data used in this study are financial statement data of manufacturing sector companies that meet the following criteria:

The results of descriptive statistic analysis for all variables used in this study are presented in Table 2.

Hypothesis testing was carried out using the random effect model presented in Table 3.

The results of research that have been carried out show that firm risk (debtholder) has a positive effect on earnings management. This result is in line with (Pradnyani & Astika, 2019), which states that firm risk as proxied by leverage has a positive effect on earnings management.

One of the hypotheses proposed by (Watts & Zimmerman, 1986) is the debt covenant hypothesis. This hypothesis looks at management's efforts to implement accounting policies that maintain the company's debt ratio at a safe level to avoid imposing penalties by creditors. Failure to maintain this debt ratio can lead to default risk for the company. To avoid this, companies tend to take advantage of loopholes in accounting rules that can be exploited to increase company profits.

Companies that use debt in their funding structure place creditors as one of the principals in the company. Management as a representative of the principal should also act in the interests of creditors. However, management who acts as a representative of the company's owner can have different preferences. This misalignment of interests can be seen in agency theory.

Leverage in the form of the use of debt in the company's funding structure provides excellent benefits for management as a multiplier. However, if the debt borne has exceeded the optimal limit, the company will have difficulty paying it off, which can lead to bankruptcy in extreme conditions. The risk of default needs to be anticipated by creditors by using debt covenants. Violation of the debt agreement can cause the company to be categorized as default.

However, corporate steps taken by managers can still take the company beyond the determined limits. This condition makes earnings management practices a promising option. Accounting rules that allow companies to use management discretion in recording methods make company directors try only to practice the favorable conditions, not the ideal provisions based on conservatism. It is reflected in the results of research showing that companies with high leverage use more earnings management strategies in preparing financial statements.

Based on the study results, it is known that the firm risk (equity holder) has no effect on earnings management. This result is in line with (Suhartanto, 2015) research, which concludes that changes in stock prices have no significant effect on earnings management.

From equity holders' perspective, firm risk is reflected in total risk, which is the fluctuation of monthly returns received by shareholders during a period. Highly volatile returns reflect the magnitude of the risk faced by the company's owners. To avoid this, principals set up many provisions and targets that must be achieved by management. The achievement of the targets that have been formulated will make the directors get rewards. This scenario is the first hypothesis (Watts & Zimmerman, 1986) proposed in positive accounting theory, the bonus plan hypothesis.

The stock return, which illustrates investors' interest, is influenced by macroeconomic conditions and the company's internal conditions. However, the only thing that can be controlled by management is the conditions that occur internally. If macroeconomic conditions do not support operations, management can compensate for this by presenting a brilliant company performance. The easiest thing for management to do is present financial statements, mainly company profits, which are growing positively.

To ensure that the company's conditions are presented satisfactorily, management needs to make some adjustments. These adjustments are carried out at management's discretion by taking advantage of exploitable accounting loopholes. However, the results of this study show something different. Stock returns' fluctuations are not a factor that encourages companies to take advantage of massive earnings management practices. It is probably due to the low level of use of earnings management in the manufacturing sector. Previous descriptive statistics have shown that the average company performs earnings management intending to increase income in a deficient range. This condition causes the rise and fall of a company's stock price does not affect the manager's decision to use earnings management related to the preparation of financial statements.

The results of research that have been carried out show that tax risk has no effect on earnings management. This result indicates that a company's cash ETR volatilities are not a factor the board of directors considers in preparing financial statements' earnings management practices.

Tax payments are transactions involving external parties with high authority (government). This phenomenon is closely related to the third hypothesis in the positive accounting theory proposed by (Watts & Zimmerman, 1986), the political cost hypothesis. The hypothesis describes management's intention to carry out earnings management practices to minimize political costs, such as subsidies and tax payments, which the company will bear.

The volatility of cash ETR, the number of tax payments compared to profit before tax in one period, shows how the company faces much tax risk. Changes in the value of the tax deposit will invite the attention of the tax officers. Special attention by tax officers will increase the risk faced by the company. Tax risk can then have an impact on uncertainty in the company's operational activities. These uncertainties include tax reporting related to the company's original condition, audit risk by tax authorities, and financial accounting related to income tax payments (Hutchens & Rego, 2013).

Referring to descriptive statistics in Table 2, the average cash ETR volatility of manufacturing sector companies is only 14.06%. This low volatility puts the company in a less risky position to be examined by the tax authorities, considering that the company's income tax payments are not volatile. It makes tax risk not a factor considered by managers in making decisions regarding the use of earnings management. This condition is further supported by the low practice of earnings management in manufacturing sector companies which shows that only a few companies carry out massive earnings management in the preparation of financial statements.

Based on the study results, it is known that sharia compliance has no effect on earnings management. This condition indicates that whether or not a company is included in ISSI is not a factor influencing the board of directors to use earnings management.

The relationship between investors and companies can be explained using agency theory. The existence of two interrelated parties, the principal and the agent, will create a conflict of interest. This difference can be seen from the motivation of both parties. The company's owner is motivated to maintain an increase in the standard of living by increasing the company's profitability. At the same time, the manager is motivated to maximize economic and psychological needs by obtaining investment, debt, job security, or compensation (Trisnawati et al., 2015).

The relationship between sharia compliance and earnings management can be estimated from the selection criteria for sharia shares compiled by the Financial Services Authority (OJK). One of these criteria is the provision of financial ratios that issuers must meet, which is that total interest-based debt compared to total assets does not exceed 45%. Suppose this criterion is compared with the composition of company funding that has been presented in Table 2. In that case, it can be seen that the average company in the manufacturing sector only has 40% of DAR. With this debt-to-asset ratio, the average company in the manufacturing sector does not need to adjust financial data using earnings management to remain included in the ISSI, which means the company can continue to have more investor coverage. This condition causes sharia compliance does not affect the manager's decision in conducting earnings management.