The Impact of Ownership Characteristics and Gender on Earnings Management: Indonesian Companies

Abstract

Earnings management is a behavior performed by management to show good performance to principals. This effort creates information bias in the study of agency theory, which in turn increases information asymmetry. In Indonesia, the average company has a family ownership structure. Therefore, this study aims to examine the effect of family ownership characteristics and gender on earnings management. This study includes gender diversity in the board of commissioners and board of directors. This research uses the non-financial companies’ data in Indonesian Capital Market. Furthermore, the data were analyzed using multiple regression based on ordinary least squares. Research results show that the proportion of females in both board of commissioners and board of directors, as well as company size contribute significantly to earnings management, whereas, family ownership, ROA, and leverage do not have a significant impact. This research provides a practical contribution to the study of the composition of the board of commissioners and directors regarding earnings management actions for owners, investors and other stakeholders.

1. Introduction

Corporate governance (CG) is a mechanism that is allegedly effective in reducing information asymmetry between agents and principals. The existence of information asymmetry encourages management to manipulate in showing better earnings information (earnings management) (Agustia 2013). Earnings management is conducted by management as an agent to gain more attention from the principal. Good profits are then their main concern in order to attract more attention from investors. Managers may also actively engage in EM because the mentioned income affects their compensation. However, powerful CG mechanisms in phrases of board characteristics may additionally help lessen competitive EM, as a result lowering company costs (Lara et al. 2017).

Investors often focus on the information on company profit without paying a close attention to the procedure applied to compose it (Beattie et al. 1994). Procedure intended in this context refers to a process to make or to modify the financial statements in order to appear well-managed and attract the investors to make investment, which is usually termed as earnings management. With regard to the earnings management practice, many studies have been conducted; nevertheless, those focusing on family enterprise are still limited in number (Salvato and Moores 2010). This condition underlies the observation conducted by the researchers because family enterprises dominate the economy of developed and developing countries. PricewaterhouseCoopers Company (PWC) survey results on family business in Indonesia mention that more than 95% of Indonesian businesses are owned by families. Additionally, the Indonesian Institute Corporate and Directorship (IICD) reveals that families own and control more than 95% of Indonesian businesses. (Sukmadilaga et al. 2015).

Many studies have been conducted previously related to corporate governance to reduce earnings management behavior that may be carried out by management. Marzuki (2022) show that Companies meet analysts’ forecasts of earnings, which is important for companies which dominated Bumiputera audit committees in Malaysian companies. In addition, relevant works (Barua et al. 2010; Emadi and Garkaz 2015; Gavious et al. 2012; Gunawan and Situmorang 2019; Harakeh et al. 2019; Novilia and Nugroho 2016) show that the more women occupy positions on the board of directors in a company, the more possible it will be to reduce earnings management actions through its policies. Abbott et al. (2012) suggest a significant relation between the presence of at least one female director in the Board and the low probability of restatement.

Due to the gradual but steady increase the number of female directors serving on corporate boards, board gender diversity is receiving a lot of attention from policymakers, regulators, investors, corporations, scholars and the general public. (Ammer and Ahmad-Zaluki 2017; Pathan and Faff 2013). Several previous studies have explained how gender diversity can improve financial information to be more accurate and transparent. For example, Damak (2018) and Zalata et al. (2018) argue that the independence, efficiency and monitoring activities of the board are enhanced by the appointment of female directors. Ammer and Ahmad-Zaluki (2017) and Ginesti et al. (2018) suggest that the practice of CG mechanisms, in improving company reporting practices, is very possible with the presence of female directors. The existence of a negative influence between the presence of female directors on boards and EM practices has also been recognized by other studies. (Gavious et al. 2012; Harakeh et al. 2019; Lara et al. 2017). However, Ye et al., in contrast to these findings, found no correlation between gender diversity and earnings quality.

In Indonesia, more attention is focused in studies of good corporate governance and earnings management to factors of institutional ownership, managerial ownership, board size, audit committee, and independent board of directors. (Jao and Pagalung 2011; Praditia and Marsono 2010; Saffudin and Prasetiono 2011; Ujiyantho and Pramuka 2007; Veronica 2013; Widiatmaja and Syaffrudin 2010). On the other hand, previous researchers have never paid attention to the phenomenon of large family ownership on average companies in Indonesia.

The novelty of this research is to examine specifically on family ownership (FO); Female proportion in a Board of Commissioners (FemKom); Female proportion on Board of Directors (FemDir). In addition, this study presents control variables: Company size (SIZE); Return On Assets (ROA); Leverage (LEV). The results of this study will serve as a theoretical contribution in giving description of Agency Theory, especially regarding the relation between shareholders and company management, and also between majority shareholders and minority ones. In practice, this study also contributes in terms of supervisory practices of public company management by owners, investors, and other stakeholders.

2. Literature Review and Hypothesis Development

2.1. Agency Theory

Earnings management studies cannot be separated from agency theory. Agency theory is the basis to understanding corporate governance in which each individual is motivated by their own interests so that conflicts between principals and agents can arise. The principal is motivated to enter into a contract to prosper themselves with ever-increasing profitability. Meanwhile, Agents are motivated to meet as many of their financial and psychological needs as possible. Agency issues arise because shareholders, as owners, and management, as agents, do not share the same interests. (Surya and Yustiavandana 2006).

Jensen and Meckling (1976) state that the agency relationship is a contract in which the owner hires a manager to provide a service and then delegates decision-making authority. The principal authorizes the agent to conduct good corporate governance. Good corporate governance is a system, structure, mechanism and culture that will protect the interests of shareholders and stakeholders. Therefore when the agent is able to manage the company properly, the agent will get compensation from the principal (Lestari 2022). If the owner and manager have different interests, it will cause an agency conflict, namely, the separation of functions between ownership on the part of investors and control in management. Managers may also actively engage in EM because of the mentioned income affect their compensation. However, powerful CG mechanisms in phrases of board characteristics may additionally help lessen competitive EM, as a result lowering company costs (Lara et al. 2017).

Organization issues happen in light of the division of proprietorship and control in the company which results in potential conflicts between owners and managers. The goals of management differ from those of the company’s shareholders (Van Horne and Wachowicz 2001). Jensen and Meckling (1976) potential problems that arise in the perspective of agency theory, namely, the existence of information asymmetry so that it can cause two problems to arise, namely, moral hazard and adverse selection. The emergence of these problems can make the company bear agency costs in the form of monitoring costs, bonding costs and residual costs.

The emergence of agency costs above can conceptually be reduced by implementing corporate governance. The structural components of GCG consist of: (1) managerial ownership, (2) institutional ownership, (3) independent board of commissioners and (4) audit committee. This study focuses on the mechanism of share ownership that exist in the average company in Indonesia and the diversity of the composition of the board of commissioners and the board of directors. The study of share ownership and the composition of the board of directors is expected to contribute to the development of agency theory in the future.

2.2. Gender Diversity

Caplan (1987) emphasized that gender is a difference in the behavior of men and women apart from biological context, most of it is actually formed through social and cultural processes. Humans consist of men and women who live together and work together. The existence of both of them involves playing themselves in family activities/activities and in the midst of society. The roles played by each are inseparable from the skills they have also according to their character. As a result, the environment rewards anyone who benefits the environment, regardless of what and who based on gender.

Caliper interviewed 177 of the best female leaders (CEOs, managing directors, senior directors, vice-presidents, general managers, and others) in various companies in Asia Pacific and found five characteristics that many female leaders have (Fitriani 2015): (1) Ability to persuade: In most cases, women leaders are more persuasive than men. They have a tendency to be more ambitious than men, and the satisfaction they derive from convincing others to say “yes” boosts their ego. However, their social, feminine, and empathetic nature will not vanish when forced to do so. (2) Proving criticism wrong: Women as leaders have a lower ego strength than men, so they can still experience rejection and criticism because they are “not cheeky yet”. However, they are able to quickly recover, learn from mistakes, and move forward with a positive “I’ll prove it” attitude thanks to their high levels of courage, flexibility and friendliness. (3) Team spirit: When it comes to problem solving and decision making, great female leaders typically employ a comprehensive leadership style. Additionally, they more adaptable, considerate and helpful to their employees. However, women have a lot to learn from men when it comes to problem-solving and decision-making accuracy. (4) The leader: Good female leaders usually have good charisma and so do men. They are persuasively confident ambitious and energetic to get the job done. (5) Daring to take risks: Women leaders are now daring to break the rules and take risks, just like men, while paying the same attention to the smallest of details. They speculate outside the company and reject all structural rules, such as policies and regulations.

2.3. Earnings Management

Kim and Sohn (2013) stated that current period earnings can be managed in two ways, namely, through freedom to choose accruals in accordance with Generally Accepted Accounting Principles (GAAP) and distorting earnings activity significantly. Accruals seem to be a way for financial managers to beautify the appearance of profits. Under certain conditions, the company lowers profits so that the feedback given to investors is low.

According to Perwira and Darsono (2015), earnings management is the result of intervention from the management in the process of compiling financial reporting. Accountability through financial reports must be standardized properly. However, the nature of accounting which contains a lot of estimates (estimates), judgments (judgments) and the nature of accruals opens up opportunities to manage earnings (Alwiyah and Sholihin 2015). Earnings arrangements can also occur when an independent party (auditor) does not provide comments when conducting an audit. The role of the auditor is also important in assessing company profits which are part of the financial statements.

Kim and Sohn (2013) state that accrual earnings management based on GAAP occurs when the accounting period ends. This determines whether the company’s profits have been as expected or not. Management will see a gap to be able to manage profits in such a way as to generate the expected profit. Of course, this is still within the limits of applicable regulations. Managers may have several considerations when managing earnings (Merchant and Van Der Stede 2014). They may use personal information about the company’s prospects to address minor short-term disruptions in earnings measures to provide users with more informative performance indicators. This is considered to provide an overview of the condition of the company going forward.

Sulistyanto (2008) states that earnings management is a strategy used by business leaders to manipulate financial statements in order to the deceive stakeholders interested in the company’s performance and condition. This is what makes many people think that earnings management is an act of fraud. Others still view earnings management as a positive effort by reason of the suitability of the actual conditions. Rules are still maintained in presenting information in financial statements. This is the reason why earnings management is still maintained to this day.

Companies that manipulate earnings from accruals to improve their appearance are still considered reasonable. Wild et al. (2005) refer to these as “cosmetics” which have no cash flow consequences. In this way, earnings management is an action taken by agents to manipulate the recognition of accounting transactions related to profit/loss which is a form of good signal of agent performance. This action is not a real action taken by the agent to eliminate transactions that affect the real value of the company’s profit/loss which results in the loss of company assets. However, this condition can reflect business decisions that often reduce shareholder wealth (Wild et al. 2005).

Earnings management is a practice usually associated with Agency Theory, which explains how an agent and a principal interact when an investor wants to get the most out of his return on investment, and the management wants to gain the investor’s trust. The discrepancy in the interests between manager and investor stimulate the manager to manipulate the information which is different from the real condition. This shows that the manager has full control over the company information which is asymmetrical between the company manager and the investor. The information asymmetry may lead the manager to develop opportunistic behavior that promotes earnings management. The negative effects of opportunistic earnings management are low quality of reported profit (Velury and Jenkins 2006), decrease in accounting information relevance (Habib and Azim 2008) and share liquidity decline (Fathi et al. 2011).

Khlif et al. (2015) used the interaction between disclosure, accounting policies and the cost of equity capital. Rosita (2014) showed that information asymmetry, voluntary disclosure and the management of earnings simultaneously have a significant impact on COE. Albeit partially, only information asymmetry has a positive effect on COE and the other two variables have no significant effect. According to Rinobel and Laksito (2015), firm size and earnings management have a positive effect on the cost of equity. Kim and Sohn (2013) suggest that the level of earnings management (real activities manipulation controlled from the effect of accruals) has a positive impact on equity capital costs.

Information manipulation results in irrelevant information being used as the foundation to make decisions. The useless information might have an effect on the investor. Many cases related to profit information manipulation took place in many countries, such as: Satyam Computer Service scandal in India which was exposed in 2009, the company also manipulated its profit record (Bhasin et al. 2013); and Toshiba scandal in Japan which was disclosed in 2015 related to the overstatement of its profit amounting USD 1.2 billion (Bisnis.com 2015). Financial information manipulation cases also happened in Indonesia involving: (1) the Kimia Farma scandal in 2001 which was related to inflated profit in its financial statements (Tempo.com 2003); (2) the Garuda Indonesia scandal which was exposed in 2019 and related to the overstatement of its profit in 2018 as much as IDR 11.33 billion (www.finance.detik.com, 26 July 2019); and (3) the Hanson scandal which was revealed in 2019 and related to the earnings manipulation since 2016 which caused and overstatement of as much as IDR 613 billion (Wicaksono 2019).

2.4. Family Ownership

A survey by PWC on family business in Indonesia revealed that 95% of Indonesian businesses were owned by families. In addition, the data of Indonesian Institute for Corporate and Directorship (IICD) show that more than 95% of businesses in Indonesia are owned and controlled by families (Sukmadilaga et al. 2015). Other research also argues that in developing countries, most of the companies are still controlled by family ownership (Bhaumik and Gregoriou 2010). This indicates that family business has a strong implication in accounting practices including earnings management. Yudha and Singapurwoko (2017) assume that there is a difference in leverage levels due to different leverage management among founder, sibling partnership and cousin consortium, which is caused by different styles of leadership. The difference might also result from the various races of Indonesia which might affect such different styles of leadership.

Gibb Dyer (2006) explores the effect of family on the performance of a company in reference to Agency Theory and from a company resource-based perspective. Miller and Le Breton-Miller (2006) employ the propositions of Agency Theory and management to suggest an assumption on the factors which make a family-owned company a strong contender. The rapid development in the study of family business eventually shows that family involvement can bring benefits and costs to the company.

The potential costs arising from the family involvement in a business include: (1) Conflict of interests between minority shareholders and family shareholders; (2) More opportunities for the family members who belong to top management to realize their family interests as controlling instead of non-controlling shareholders; (3) More attention given by family shareholders to the company growth and business survival instead of maximizing the shareholders’ profits; (4) High-paying jobs created by family business for its descendants; (5) More dividends given to family shareholders; (6) Tendency to gain more personal profits by sacrificing minority shareholders; (7) Limiting external recruitment to encourage family members to sit in the management; (8) Maintaining the management position for the controlling shareholders regardless of their inadequate competence; (9) The generosity of parents afforded to their offspring in managing the family business despite rising agency costs; (10) Subconscious action of limiting the nature of human resources to recruit family members; and (11) People’s distrust arising from family economy-oriented business.

The Family Firm Institute (Sukamdani 2012) portrays the family business proportions in various countries. In the United States, the family business proportion is 64% which contributes to the US Gross Domestic Products and absorbs 62% of the US workforce. In Brazil, 70% of businesses belong to families. Chile and Australia report that 75–90% and 67%, respectively, of the businesses in both countries are owned by families. The reasons why family business can reserve a big role in the economy are compiled from various resources, including: (1) swift decision making, (2) strong commitment and motivation of family members, (3) long-term orientation, and (4) strong family connection to understand and trust each other.

One of the strengths of a family business is the awareness that the business has a vital contribution to the national economy, including Indonesia. In addition, other strengths of family business are ownership, supervision, and management, while the weaknesses include inability to play a maximum role due to lack of good corporate governance, and the problems threatening its existence because the business is still run by the first generation. Family business usually has a short lifespan because of the lack of separation between family matters and business concerns, unfair human resources management in terms of competence and performance evaluation, the lack of clear organizational roles, rules and structure, “spoiled kid syndrome”, where the company recruits incompetent family members, and high reliance on the founder.

There are four stages in the family business, including: (1) Development stage (family members are the main engine); (2) Management stage: a stage where several issues matters, such as value conflict, succession, organizational structure, compensation, competence, earnings distribution and harmony; (3) Transformation stage, where non-family professional manager must be recruited to create fresh ideas; and (4) Sustainability stage, when the family business already completes the development and profitability stages, and all organizational systems, procedures and policies of the organization are well-established. Good family management practices can: (1) stimulate the thoughts and understanding of new business strategies, (2) recruit and maintain capable good non-family managers, (3) create a flexible and innovative organization, (4) produce and protect the capital, and (5) prepare the leadership succession.

Family business also faces structural obstacles including: (1) high likelihood of conflict of interests among family members resulting in unfocused strategy development, decision making and resource allocation, (2) poorly managed company sustainability due to unsmooth succession agenda, and (3) difficulty to change where the transformation takes place due to the dominance of the founders’ strategies, systems, culture and leadership styles.

2.5. Hypothesis Development

Dang and Nguyen (2021) examined the significance of external audit and internal management in preventing future price decline and discovered that the risk of future stock price decline is significantly linked to corporate governance. Particularly, the risk of a future stock price crash was found to be positively correlated with a strong board of directors. However, there is a negative correlation between accident risk and audit committee effectiveness. Based on these findings, good internal governance can lessen the likelihood of stock prices plummeting. This study also demonstrates that audit committees are more effective at preventing acute risks when their external audits are of higher quality. The findings hold a number of significant repercussions for risk management and portfolio investment.

Nguyen and Dang (2022) examined the connection between bank risk and risk management effectiveness at various institutional quality levels and found that risk management efficiency has a negative effect on bank risk. This relationship is further enhanced in countries with higher institutional quality. This suggests that risk management can prevent bank risk more effectively in countries with better institutional quality. Nguyen (2022) examines the determinants of bank risk management structure and finds that risk management structure (including the size of the audit committee, the independence of the audit committee, the presence of professionals in finance and accounting on the audit committee, the frequency of the meetings of the audit committee, the graduation of the risk committee, and external quality assurance) is positively related to the operational scale of the bank and the benefits of supervision, but negatively related to the cost of supervision and the bargaining power of the CEO. However, the relationship between high-risk and low-risk banks is different and between banks in countries with different institutional characteristics.

Bhuiyan et al. (2020) tested the impact of an independent risk committee on risk appetite and corporate goodwill. The quality of corporate governance is improved by having an independent risk committee. This improves investor protection by lowering a company’s risk appetite and increasing shareholder value. This study found that multiple risk measures were significantly reduced in organizations with independent risk committees compared to organizations with joint audits and risk committees. The existence of an independent risk committee is positively related to firm value.

Nguyen (2022) analyzed the effect of audit committee effectiveness on bank stability. Focusing on the structure of the Audit Committee, this study shows that a smaller Audit Committee with more independent members can increase the stability of a bank. This shows that the effectiveness of the audit committee is positively related to the stability of the bank. The effectiveness of the Audit Committee is primarily in helping to improve bank stability by providing incentives to balance results and maintain a higher capital adequacy ratio. In addition, the relationship between bank stability and audit committee effectiveness is highly dependent on the stability of each bank and the institutional quality of each country.

Nguyen (2020) tested of the influence ownership structure on bank risk taking behavior ASEAN nations and demonstrated that the characteristics of each bank are correlated with the relationship amongst ownership structure and bank risk taking behavior according to the quantile regression. First, government and foreign ownership has a positive effect on bank risk at high-risk banks, while at low-risk banks, it has a negative effect. Second, across all bank risk allocations, the relationship between ownership concentration and risk appetite is negative. Based on the risk level of each bank, these findings suggest that the appropriate ownership structure can limit a bank’s risk appetite.

Some of the studies above have yielded many results suggesting that good corporate governance, including the existence of commissions in it, has an influence on the value of the company in general. However, we found several gaps, including that the above research has not examined in detail the existence of women’s gender in commissions/council. The influence observed in the studies above still allows it to be devoted to earnings management, which is currently more common. The above-average studies use research data from developed countries, so that there are still opportunities for research to be carried out in developing countries such as Indonesia. The gap that we found then leads us to develop the following hypothesis:

Hypothesis 1 (H1).

Family ownership is related to earnings management.

Hypothesis 2 (H2).

Females as board commissioners are related to earnings management.

Hypothesis 3 (H3).

Females in board of directors are related to earnings management.

3. Sample and Methodology

The sample is the non-finance companies listed at Indonesia Stock Exchange from 2016 to 2019. The number of units of analysis was 365 studies which were determined using the purposive sampling method. The criteria used in this research purposive sampling are:

• Companies listed in 2016–2019;
• Manufacturing companies that publish annual report data and audited financial report data consistently for 2016–2019;
• Manufacturing companies which use rupiah currency;
• Having a financial year ending on 31 December;
• Companies that have complete data needed for research for 2016–2019.

Furthermore, the variables of this study consist of the dependent variable and the independent variable, as well as the control variable. The dependent variable used is the Earnings Management variable. The independent variables consist of: Family ownership (FO); Female proportion in the Board of Commissioners (FemKom); Female proportion on the Board of Directors (FemDir). In addition, this study presents control variables: Company size (SIZE); Return On Assets (ROA); Leverage (LEV). Furthermore, the measurement of each research variable is measured as shown below (see

Table 1. Variable operational measurement.

No.	Variable	Measurement
1.	Earnings Management	The earnings management level in this research was measured using Discretionary Accruals referring to (Friedlan 1994) $Discretionary Accruals=\left(\frac{Total accrual{s}_{test period}}{Sale{s}_{test period}}\right)-\left(\frac{Total accrual{s}_{benchmark period}}{Sale{s}_{test period}}\right)$
2.	Family ownership (FO)	The ownership structure of a company by a family amounting more than 5% (Klein et al. 2005)
3.	Female proportion in Board of Commissioners (FemKom)	The number of women on the Board of Commissioners
4.	Female proportion in Board of Directors (FemDir)	The number of women on the Board of Directors
5.	Company size (SIZE)	Company total assets
6.	Return On Assets (ROA)	Company profitability
7.	Leverage (LEV)	Ratio of company ability to fulfill its responsibilities

).

In the next step, after the research data were collected, the data were tested for outliers, normality, and classical assumption fulfilment. Moreover, the additional tests to assess the hypotheses with multiple regression were also conducted to analyze the model significance, including adjusted R-Squared test, F-test, and t-test. The model equation is as follows:

DAC = a + β₁FO + β₂FemKom + β₃FemDir + β₄Size + β₅Lev + β₆ROA + ε

where:

• DAC = Discretionary accrual;
• FO = Family ownership;
• FemKom = The number of women on the Board of Commissioners;
• FemDir = The number of women on the Board of Directors;
• Size = Company size;
• Lev = Leverage;
• ROA = Return On Assets.

4. Results

4.1. Descriptive Statistic

Descriptive statistics describe the data, involving the minimum, maximum, mean, median, and standard deviation for each variable. The descriptive statistics of the data used in this study are presented in

Table 2. Descriptive Statistics.

	DAC	FO	FemKom	FemDir	SIZE	ROA	LEV
Maximum	1.713	0.982	0.667	0.750	18.344	0.416	0.940
Minimum	−7.679	0.112	0.000	0.000	9.320	0.000	0.000
Mean	−0.067	0.578	0.143	0.165	14.555	0.070	0.173
Median	−0.026	0.580	0.000	0.143	14.591	0.050	0.148
Range	9.393	0.869	0.667	0.750	9.025	0.416	0.940
Std. Dev	0.762	0.213	0.180	0.188	1.391	0.071	0.142

Source: Data processed by the researchers, 2022. Notes: DAC: Discretionary accrual; FO: Family ownership; FemKom: Female proportion in BOC; FemDir: Female proportion in BOD; SIZE: Company size; ROA: Return On Assets; LEV: Leverage.

.

Table 2 shows the value of Discretionary Accrual between −7.679 and 1.713 with mean −0.067 and standard deviation 0762 which is quite far and tends to be negative. This might be caused by the low average profitability level of the companies.

Family ownership has the maximum value of 0.982 and the minimum 0.112 with mean 0.578 and standard deviation 0.213. This indicates that the sample companies of this research have a very high level of family ownership. The mean of this variable dominates with a value of over 0.5.

The data on the proportion of females in Board of Commissioners show the maximum value of 0.667 and the minimum 0.000 with mean 0.143 and standard deviation of 0.180. This implies that there are still many companies which have not yet appointed a woman as a member of Board of Commissioners. From the mean, it can be concluded that the proportion of female commissioners is still very small.

Similarly, the data on female proportion in Board of Directors show the maximum 0.750 and the minimum 0.000 with mean 0.165 and standard deviation 0.188. This shows that there are still many companies which count more on male directors.

4.2. Correlations

Table 3. Correlation Matrix.

	DAC	FO	FemKom	FemDir	SIZE	ROA	LEV
DAC	1.000
FO	−0.009 ***	1.000
FemKom	−0.101 ***	−0.031 ***	1.000
FemDir	−0.109 ***	0.130 ***	−0.012 ***	1.000
SIZE	−0.079 ***	0.068 ***	−0.033 ***	−0.048 ***	1.000
ROA	0.063 ***	0.038 ***	−0.223 ***	−0.060 ***	−0.009 ***	1.000
LEV	−0.033 ***	0.060 ***	−0.113 ***	0.074 ***	−0.193 ***	0.200 ***	1.000

Source: Data processed by the researchers, 2022. Notes: The table presents the correlation between variables. *** denotes a significance level of 0.05.

below shows the results of the correlation test between research variables. Table 3 below shows that all research variables have a correlation value close to 0.8. This shows that the relationship between research variables is weak. Therefore, it is concluded that all variables do not indicate multicollinearity.

4.3. Result of Multiple Regression Analysis

Table 4. Result of multiple regression test.

Variable	B	Std. Error	t	p-Value	VIF
Intercept	0.776	0.429	1.807	0.072
FO	0.029	0.188	0.153	0.878	1.029
FemKom	−0.437	0.227	−1.929	0.054	1.061
FemDir	−0.452	0.215	−2.102	0.036	1.031
Size	−0.048	0.029	−1.667	0.096	1.051
ROA	0.439	0.581	0.755	0.451	1.096
Lev	−0.297	0.284	−1.048	0.296	1.100
Multiple R: 0.189 R Squared: 0.036 Adjusted R Squared: 0.020 Standard Error: 0.754 F: 2.201 Sig.: 0.042

Source: Data processed by the researchers, 2022.

presents the results of the multiple regression test conducted in this research.

Table 4 shows that the value of the VIF is less than 10. The regression model can be concluded to be free from multicollinearity symptoms. These results are consistent with the correlation matrix Table 3 which shows the weak relationship between the study variables.

Table 4 gives a general description suggesting that the independent variables basically influence the dependent variables of this research with R Squared value less than 0.05. This also means that there are still other variables outside the model which influence earnings management performed by the companies.

Table 4 indicates that the model developed in this research is fit and eligible to use with the p-value 0.042 which is lower than the significance level or α of 0.05. However, it still opens the opportunities to develop further study by exploring and inserting other variables than those used in this research; thus, a better formula can be composed. Next, Table 4 demonstrates the results of the hypothesis test as follows.

Table 4 denotes that family ownership does not influence earnings management. This implies that the first hypothesis of this research is not supported. In conclusion, a higher or lower rate of family ownership does not encourage or discourage earnings management.

4.4. Robustness Test

The strength of the research model was tested to review the consistency between the results with the ordinary least square method and the fixed effect model. Test results were obtained by changing the method of analysis. The test results using the fixed model yielded the following results:

Table 5. Robustness Test Fixed Effect Model.

Var	Coefficient	t-Stat	p-Value
Constanta	0.783637	0.957354	0.3393
FO	0.000856	0.002324	0.9981
FemKom	−1.684739	−4.303615	0.0000
FemDir	−1.175394	−3.082765	0.0023
Size	−0.027527	−0.493548	0.6220
ROA	−0.654304	−0.785518	0.4328
Lev	0.137337	0.627433	0.5309

Source: Data processed by the researchers, 2022.

shows the results of the research model robustness test. The robustness test was performed using panel data based on the fixed method, where there was no significant difference from the ordinary least square test conducted on the research hypothesis. The robustness test shows that female commissioners and female directors have a significant negative impact on the management of earnings. Meanwhile, family ownership, size, ROA, leverage have no impact on the management of earnings. This result is similar to the result of the ordinary least squares, suggesting that there is no difference between the ordinary least square test and the fixed method.

5. Discussion

The results of testing Hypothesis 1 show that family ownership has no effect on earnings management. This means that family ownership of companies in Indonesia does not direct its influence on earnings management decisions. These results are in accordance with the study by Richardson and Leung (2011), which showed that family ownership did not affect earnings management practices. With the direct control of the family, the indication of earnings management usage was reduced.

Testing Hypothesis 2, which states that the position of female commissioners has an influence on earnings management decisions, shows that companies in Indonesia with the position of female commissioners are able to influence earnings management actions. The proportion of female commissioners has a significant adverse effect on earnings management. The Board of Commissioners functions to oversee the policy of the directors in running the company and give advice to them. Commissioners are independent, which means that they are not involved in the company management and expected to be able to perform their tasks and functions objectively duly for the interests of the company, regardless of the conflict of interests among related various parties. Octaviani and Kartikaningdyah (2019) mention that the core of a company’s corporate governance is its Board of Commissioners which is assigned to safeguard the implementation of company strategies, monitor the management in running the company, and impose the company accountability realization. According to Trisanti (2017), the presence of female commissioner is influential since women are more detailed in monitoring the business operations and can give advice to the directors. The functions indirectly influence the profit quality reported by the company management to the Board of Commissioners which later can direct the policies and actions of the directors regarding financial reporting. Li (2021) claimed that it is established from the perspective of the system, female social background, and data from China’s listed companies that female managers typically enhance earnings management. This phenomenon is becoming increasingly apparent as the proportion of female managers in the Chinese market grows. Orazalin (2020) shows that companies that have more women on their boards are better at controlling earnings management.

Hypothesis 3 states that the position of a female director has an influence on earnings management action decisions. The results show that companies in Indonesia with the position of female director are able to influence earnings management actions. Female proportion in Board of Directors significantly and negatively influences earnings management. Relevant research findings (Chen et al. 2017; Faccio et al. 2016; Liu et al. 2014) reveal that more female directors will lead to higher profitability, lower profit volatility, and more opportunities for business survival. Orazalin (2020) showed that companies that have more women on their boards are better at controlling earnings management. The negative effect between the presence of female directors and EM practices has been carried out by many other studies (Damak 2018; Gavious et al. 2012; Harakeh et al. 2019; Lara et al. 2017). According to data from Chinese enterprises, Ye et al. (2010) found no correlation between gender diversity and earnings quality, contrary to these findings.

Furthermore, testing the control variable size, profitability and leverage shows that only size is able to show its effect on earnings management, while profitability and leverage are not able to be control variables on earnings management. The size of a company also significantly influences earnings management. According to the Agency Theory of (Jensen and Meckling 1976), agency relationship shows the functional separation as a principal presents authority to an agent to manage the principal’s resources. Consequently, the company owner bears the costs to supervise the company manager in running the business. When the costs become larger, tighter supervision is conducted, so the manager must be careful and accurate in providing the information to minimize earnings management.

Profitability does not present a significant influence on earnings management. Agency Theory postulates that the principal as the company owner does not completely know about the actions performed by the manager, so it is very likely that the agent or manager does something beyond the principal’s sight or so-called moral hazard which is an agency problem caused by information asymmetry. Information asymmetry occurs between principal and agent because of imbalance in gaining information. Agent is motivated to use the information collected without the principal knowing to maximize his own profit which is achieved by exhibiting good performance of the company to increase the company quality. However, the principal prefers to make a contract which guarantees their personal welfare through growing profitability since it is important for a company to gain maximum profits. Conversely, high levels of ROA will result in tax increase. To avoid that, the management will manage the company profits in such a way to attract more attention and capital from external sources.

Leverage also does not significantly influence earnings management which means that the lower or higher leverage rate will not affect the management to manipulate the company profits. A company with high leverage rates as a result of the large ratio of total debts to total assets will face high default risk where the company is very likely unable to fulfil its liabilities. In this case, earnings management cannot be used as the mechanism to prevent the default from happening. The management of the companies with averagely safe leverage rate is not interested or motivated to conduct earnings management since the companies do not require the practices which might be needed only for certain situations.

6. Conclusions

The findings of this research show that the proportions of female members in both the Board of Directors and the Board of Commissioners significantly influence earnings management. Nevertheless, family ownership, SIZE, ROA, and leverage do not show significant influences on earnings management. These results indicate that the exist of women the board of commissioners and the board of directors determines the decision-making process for taking profitability management actions. However, family ownership in Indonesian companies does not affect earnings management actions, so even though on average the shareholding companies are owned by families, it does not affect the earnings management actions taken.

Even so, this study has several weaknesses, including that this research was conducted on manufacturing companies only, so these results may be different from other corporate sectors in carrying out earnings management. We suggest for other researchers to conduct further research to explore and insert the variables other than those used in this research model to develop a better corollary of earnings management. This research presents a theoretical contribution to Agency Theory in describing earnings management. In practice, the research contributes to supervisory practices, especially ones related to earnings management.