Does CEO Duality Affect Corporate Reporting and Tax Aggressiveness In Indonesia?

Tax Aggressiveness

Indonesia established various tax compliance schemes from 2016 to 2020. These measures include tax amnesty, lowering tax rates for SMEs, boosting taxpayer services, and strengthening audit staff at the Directorate General of Taxation to improve law enforcement. The Financial Transaction Report and Analysis Center (PPATK) states that suspicious financial reports, including probable tax fraud, have gradually increased over the past three years. 2018 had 501 cases, 2019 had 738, and 2020 had 793. From 2016 to 2020, the Indonesian Directorate General of Taxation reported that 20% to 50% of registered companies as taxpayers paid corporate taxes, producing 18% to 19% of total tax income. According to Balakrishnan et al. (2019) and Frank et al. (2009), tax aggressiveness includes legitimate tax planning and fraudulent tax avoidance. Tax aggressiveness lowers income tax. Aggressive corporate tax planning can pass tax savings to shareholders, increasing their wealth (Arora & Gill, 2022; Campbell et al., 2020).

CSR Disclosures, Aggressive Financial Reporting, and Tax Aggressiveness

Tax aggressiveness is often regarded as socially irresponsible and unethical in business and management research. Despite companies' CSR efforts, aggressive tax companies could damage their reputation. Companies are urged to make their tax payments complete through CSR initiatives. According to Chouaibi et al. (2022), Issah and Rodrigues (2021), and Mgbame et al. (2017), taxes from companies fund public services like education, national security, law enforcement, and health care. CSR disclosures and tax aggressiveness relationship is still being debated. Some studies stated that CSR disclosures negatively relate to tax aggressiveness (Chouaibi et al., 2022; Mgbame et al., 2017; Raithatha & Shaw, 2022). They argue that companies with more CSR disclosures are more likely to adopt ethical tax policies to match their public image of social responsibility. In contrast, CSR disclosures and tax aggressiveness are positively correlated in other studies (Abid & Dammak, 2022; Mao, 2019; Rohyati & Suripto, 2021). They suggest companies can use CSR to improve their reputations and offset unfavorable public views from aggressive tax planning.

Adopting aggressive financial reporting strategies and reducing income tax can enable companies to evade tax payments. The relationship between aggressive financial reporting and tax aggressiveness has been the subject of many studies, but no firm conclusions have yet been drawn. Sánchez-Ballesta and Yagüe (2021) identified a pair of compromises between aggressive financial and tax reporting practices. Taxes rise and fall with a company's income. However, Firmansyah (2019), Nugroho et al. (2020), and Rachmawati et al. (2020) stated that no trade-off occurred between aggressive tax planning and financial reporting. Managers employ aggressive financial reporting to increase shareholder value. Management avoids adopting accounting methods that cause earnings increases since it increases income tax liabilities. Managers must aggressively tax to compensate for aggressive financial reporting.

CEO Duality's Effect on CSR Disclosures, Aggressive Financial Reporting, and Tax Aggressiveness

Voinea et al. (2022) show that CEO duality negatively relates to CSR disclosure. They indicate that companies with no CEO duality are the higher-performing organisations that share CSR information more frequently. Ananzeh et al. (2022) and Sun et al. (2022) also confirmed the result. Their findings confirm the legitimacy theory that assumes a company will engage in and publicise socially responsible activities to maintain a good reputation. In contrast, Lassoued and Khanchel (2023) argued that CEO duality streamlines decision-making processes, as the CEO has direct control over the board and can expedite the incorporation of CSR considerations into corporate strategies. CEO duality can result in more effective implementation of CSR initiatives and enhanced CSR disclosure.

Academics and practitioners argue about CEO duality and aggressive financial reporting relationships. Some researchers argue that CEO duality signals an imbalance in the internal monitoring and control system, designed to minimise the possibility of conflicts of interest and facilitate aggressive financial reporting (Gavana et al., 2022; Hasan et al., 2022). However, some researchers (Cudia et al., 2021; Paino & Iskandar, 2021) argue that CEO duality can reduce aggressive financial reporting by improving transparency, accountability, and quality. CEO duality can be held directly accountable for accurate and consistent disclosure.

Cao et al. (2021) and Ezejiofor and Ezenwafor (2021) found that CEO Duality and tax aggressiveness were positively correlated. CEO duality increases board-management interaction, increasing managers' opportunities to get involved in tax aggressiveness. Consequently, a tax aggressiveness strategy can be quickly enacted and executed. However, Chytis et al. (2020) found that businesses with dual CEOs have larger ETRs and, consequently, less tax planning. The findings assume that CEOs with a long-term interest in the company may be more tax cautious. They may prefer to follow tax laws rather than risk legal or reputational consequences.

Hypothesis development

The present study employs the theories of agency, stewardship, and legitimacy to accomplish its research objectives and address its questions. Based on agency theory (Jensen & Meckling, 1976), CEO duality can cause shareholders' and management's conflict of interest because it optimises the management's interests rather than the shareholders'. However, stewardship theory (Davis et al., 1997) assumes that CEO duality tends to be responsible for managing the company and its stakeholders. CEO duality could encourage CSR disclosure and lower tax aggressiveness to protect long-term interests and meet stakeholder expectations. Moreover, CEO duality can affect CSR disclosure and tax aggressiveness in legitimacy theory (Deegan, 2002) to obtain and sustain social legitimacy. CEO duality promotes CSR disclosures and demonstrates social responsibility. Figure 1 summarises the research framework related to CEO duality, corporate reporting, and tax aggressiveness.

Corporate governance involves two structures of decision-making authority concentration: the CEO and the board chairman held by one person (CEO duality) or by a different person. This concentration of decision-making authority may influence the firm's CSR Disclosure, aggressive financial reporting, and tax aggressiveness decisions.

There exist two prevailing theories on this matter. The agency theory promotes separating responsibilities to establish an effective control mechanism over management strategies and policies. In cases where a company exhibits CEO duality, the efficacy of the board members may be endangered due to the potential for the CEO to interfere in board affairs, leading to a potential decline in overall performance. Cooray et al. (2020), Harun et al. (2020), and Malik et al. (2020) believed that separating responsibilities indicates professional supervision, better regulation, and reduced potential for partial revelation, thereby enhancing disclosure quality. The stewardship theory presents an alternative perspective to agency theory, proposing consolidating decision-making authority in a single individual. Advocates of this perspective maintain that improved administration, proactive oversight, reduced interference, and advancement toward organisational goals are possible. According to this theoretical perspective, corporate managers are perceived as safeguarding the organisation's interests and acting in its best interests. Numerous research supports the theory that this perspective is a good signal of decision-making authority (Pham & Pham, 2020; Pucheta-Martínez & Gallego-Álvarez, 2020; Wijethilake & Ekanayake, 2020).

Thus, we hypothesise:

H₁:CEO Duality has a moderating effect on CSR disclosures and tax aggressiveness relationships.

H₂:CEO Duality has a moderating effect on aggressive financial reporting and tax aggressiveness relationships.

See Full Size >

Population and sample

This study analyses a sample comprising 665 firm-year observations to examine the CEO Duality's moderating effect on the relationship between CSR disclosures and aggressive financial reporting toward tax aggressiveness. The sample is drawn from the population of nonfinancial companies listed on the Indonesian stock exchange from 2016-2020. The selected period for this study spans from 2016 to 2020, as during this period, the Indonesian government actively enforced multiple policies to enhance tax compliance among taxpayers. Therefore, this period is deemed representative of the years of tax enforcement in Indonesia. The selection process for the industry groups involved identifying those exhibiting the highest degree of tax aggressiveness, as determined by the value of their permanent discretionary differences. Specifically, four out of the eight industry groups were ultimately chosen for analysis. This study has examined four industries, namely Infrastructure, Utility, & Transportation/IUTR (consisting of 19 companies), Basic Industry & Chemicals/BCHE (comprising 44 companies), Consumer Goods Industry/CONS (comprising 26 companies), and Property, Real Estate, & Building Construction/PROP (comprising 44 companies). The study's data analysis is categorised based on industry classification, as various features and business cycles are expected to influence decisions regarding tax aggressiveness.

Measurement of instruments

The current study considers tax aggressiveness as the dependent variable. The independent variable is two types of corporate reporting: corporate social responsibility (CSR) disclosures and aggressive financial reporting, and CEO duality is a moderating variable. Additionally, this study incorporated three control variables, specifically return on assets (ROA), leverage, and firm size.

Data Analysis

Panel data analysis is performed since this study uses time series and cross-section data. This study used Panel Unit Root (Stationarity) test to verify data stationarity—the Breusch Pagan Lagrange Multiplier (BP LM) test determined panel data poolability. The Hausman test then chooses the best model: random or fixed effect.

Empirical Model

This study uses two multiple regression equations. $DUALITY$ moderates $CSR$ and $TAG$ relationships (Model 1) in the first equation (3). In the second equation (4), $DUALITY$ moderates $AFR$ and $TAG$ relationships (Model 2).

Model 1:

${TAG}_{it}=\alpha +{\beta }_{\mathrm{1}}{{CSR}_{it}\mathrm{*}DUALITY}_{it}+{\beta }_{\mathrm{2}}{CSR}_{it}+{\beta }_{\mathrm{3}}{DUALITY}_{it}+{\beta }_{\mathrm{4}}{ROA}_{it}+{\beta }_{\mathrm{5}}{LEV}_{it}+{\beta }_{\mathrm{6}}{SIZE}_{it}+{\epsilon }_{it}$(3)

Model 2:

${TAG}_{it}=\alpha +{\beta }_{\mathrm{1}}{{AFR}_{it}\mathrm{*}DUALITY}_{it}+{\beta }_{\mathrm{2}}{AFR}_{it}+{\beta }_{\mathrm{3}}{DUALITY}_{it}+{\beta }_{\mathrm{4}}{ROA}_{it}+{\beta }_{\mathrm{5}}{LEV}_{it}+{\beta }_{\mathrm{6}}{SIZE}_{it}+{\epsilon }_{it}$(4)

where ${TAG}_{it}$ is tax aggressiveness; ${CSR}_{it}$ is Corporate Social Responsibility disclosures; ${AFR}_{it}$ is aggressive financial reporting; ${DUALITY}_{it}$ is CEO Duality; ${ROA}_{it}$ is Return on Assets; ${LEV}_{it}$ is Leverage Ratio, ${SIZE}_{it}$ is Firm Size; ${{CSR}_{it}*DUALITY}_{it}$ is the interaction of $CSR$ and $DUALITY$; and ${{AFR}_{it}*DUALITY}_{it}$ is the interaction of $AFR$ and $DUALITY$.

Descriptive Statistics

The descriptive statistics for each industry are presented in Tables 1,2,3, and 4. IUTR has the highest Tax Aggressiveness (TA) relative to the other three industries. Table 1 illustrates this with the mean TA value of 0.1313 and the maximal TA value of 1.0227. In contrast, the PROP industry exhibits the lowest TA value, as evidenced by a mean of 0.0705 and a minimum value of -0.9268, as presented in Table 4. The positive mean value observed in the TA's analysis indicates that most companies operating within the specified four industries employed aggressive tax strategies from 2016 to 2020.

Furthermore, it is observed that the PROP industry has disclosed the most extensive item of CSR activities. This result is supported by the mean CSR disclosure value of 0.4534 and the highest recorded value of 0.6688, as presented in Table 4. According to the GRI G4 index, the maximum value signifies that companies operating in the PROP industry disclose 66.88% of the total disclosure items.

Moreover, it is worth noting that companies operating in the CONS industry exhibited the highest value for the aggressive financial reporting (AFR) variable, specifically at 0.8963, as presented in Table 3. Nevertheless, the mean AFR values observed in the three sectors, namely IUTR, BCHE, and CONS, exhibit negative values. The negative mean values suggest that most companies in the sample were not engaged in aggressive financial reporting.

Additionally, the data on CEO duality, ranging from 0 to 1, indicates the presence or absence of CEO duality within the companies belonging to the four selected industries.

Stationarity Test Results

Since this model is based on panel data analysis, we use the panel unit root test to confirm data reliability. Three-panel unit root tests—Harris-Tzavalis, Breitung, and Hadri LM—show no unit root in any variable, and the data are stationary.

Regression Results

Tables 5 and 6 below show that the diagnostic results for Model 1 and Model 2 were similar. First, four industry multicollinearity tests demonstrate no issues (VIF < 5). Second, the Wooldridge and Modified Wald Statistics tests show autocorrelation and heteroskedasticity in CONS industry data. IUTR, BCHE, and PROP industry data only have heteroskedasticity concerns. This study estimates the robust standard error for random effect models and uses the cluster-by-code command to address heteroskedasticity and serial correlation.

As shown in Table 5, Model 1 used IUTR, CONS, and PROP data in a random-effects panel data regression model. The random effects model is preferred over the combined OLS and fixed effects models because the F test and Breusch-Pagan Lagrange Multiplier (BP-LM) test for the panel data regression model are significant (p-value <0.05), and the Hausman specification test is not (p > 0.05). BCHE used the fixed-effect model since the Hausman specification test p-value < 0.05. In Model 2, the random effect model is appropriate for the four industries (Table 6).