Corporate Governance, Family Ownership, and Earnings Management: A Case Study in Indonesia

Abstract

This study aims to verify family ownership's effect on earnings management by using corporate governance as the moderation variable. This study uses data panel regression with the period of 2011-2017. Corporate governance consisted of three dimensions, namely the board of commissioners, share ownership and transparency, and disclosure and auditing. Discretionary accruals measure earnings management with a model that controls company performance. Samples are manufacturing companies listed on Indonesia Stock Exchange. Observations were conducted on 198 firms throughout the year. The results indicated that corporate governance significantly affected earnings management. However, it declined the significance of family ownership toward earnings management. Hence, corporate governance can reduce earnings management. Furthermore, of the three components of corporate governance: the board of commissioners, shareholding, and transparency, the term shareholding precisely encouraged managers to conduct earnings management. Besides, the three core bodies of corporate governance lowered the significance of shareholding toward earnings management. This study's findings suggest that in family firms in Indonesia, earnings management is becoming more intensive than in non-family firms. Additional tests show that there is an entrenchment effect on family firms in Indonesia. Furthermore, corporate governance leads to earnings management.

1. Introduction

Earnings in a financial report are vital for both the internal and external parties of a firm. According to Nichols and Wahlen (2004), earnings represent an accounting degree of the alternate within the company’s value to common equity shareholders throughout a period. Company performance appraisal based on earnings shows the importance of earnings for various parties. Investors are often only focused on earnings information without paying attention to the procedures employed to bring out earnings information (Beattie et al., 1994). This condition becomes the cause that the earnings presentation frequently does not describe the actual condition of the company’s profits. This indicates the existence of manipulation in the earnings presented, which is known as earnings management.

Manipulating information can result in irrelevant information when used as a basis for decision making, so the information becomes useless, and investors are at risk of loss. A survey conducted in America obtained evidence that 78 percent of managers used earnings management policies (Graham et al., 2005). The manipulation case of financial information also occurred in Indonesia. An earnings statement manipulation took place in 2016. The actor behind the manipulation was PT. Hanson International Tbk (MYRX). The manipulation came into light due to the investigation of another case, a violation of Financial Accounting Standards no. 44 on Accounting for Real Estate Activities by the same company. PT Bank Bukopin Tbk has revised its 2015, 2016, and 2017 financial statements, led the Financial Services Authority (OJK) to commence an investigation based on an allegation of financial statement manipulations. PT Garuda Indonesia (Persero) Tbk was also found to manipulate its 2018 earnings statement. This case was revealed in 2019 following an examination of its financial statements by The Audit Board of Indonesia (BPK). PT Asuransi Jiwasraya was also proven to have manipulated its financial statements through window dressing in 2006. Furthermore, in 2017 the company received an adverse opinion, after that, in 2018, Jiwasraya had negative equity. Only in early 2020, this case was exposed. PT Garuda Indonesia (Persero) Tbk managed to record a slick performance by the end of 2018, despite struggling until the third quarter of 2018. Garuda Indonesia, for instance, managed to book a net profit of US $ 809 thousand in 2018 or equivalent to Rp 11.56 billion.

Claessens, Djankov, Fan, and Lang (2002) have revealed that companies in East Asia, including Indonesia, were dominated by families. The results of other studies indicated that family control is central in most countries in the world (Prencipe et al., 2014), so the family business has a significant impact on economic growth. Family ownership has a significant effect on the type of earnings management chosen. Companies with a high proportion of family ownership and are included in the non-business group are more likely to choose efficient earnings management than other types of companies (Siregar & Utama, 2008). Bhaumik and Gregoriou (2010) brought together several aspects of ubiquitous family firms in many institutional contexts, often as part of larger business groups. They paid particular attention to the mechanisms by which families retain control over firms and the families’ incentives in control to expropriate other stakeholders by way of tunneling. They examined the role of earnings management in facilitating tunneling and evidence about earnings management incidence in family firms. Their review suggested that while the literature on these aspects of family control is rich, the contexts in which the empirical exercises are undertaken are relatively few. Hence, there is considerable opportunity to expand it to other contexts, particularly in the form of cross-country comparisons of the relative impact of agency conflicts and institutions on these issues (Salvato & Moores, 2010). This condition becomes why the researchers conduct studies because family companies dominate companies in developed and developing countries.

Paiva et al. (2016) had conducted an earnings management review in a family company based on the agency theory. They concluded that the Type II agency problem provides a better explanation for earnings management in a family company than Type I. The Type II agency problem in a family company has a greater motive to take over the minority interests than non-family companies. Meanwhile, Type I agency problems are more likely to arise in non-family companies due to the differences between management and the owner’s interests. The literature indicates that boards and owners use the bundle to limit the manager’s opportunism (Al-Baidhani, 2014). Paiva et al. (2016) also stated that Type II agency problems tend to be more aggressive in developing countries because of the adoption of fragile corporate governance. This allows the founding families to get personal benefit from the minority shareholders.

Various research associated with type II agency issues was conducted, wherein the study results indicated that this hassle often takes place in family firms. Stockmans et al. (2010) examined the preservation of socioemotional wealth as a motive for earnings management in specific types of private family firms by looking at the generational stage, the management team, and the CEO position. The authors’ results suggested that socioemotional wealth may play a role as a motive for upward earnings management when firm performance is poor. Under this condition, first-generation and founder-led private family firms seem to have greater incentive to engage in upward earnings management because of the preservation of their socioemotional wealth. Chi, Hung, Cheng, and Tien Lieu (2015) have proved that family firms tend to do earnings management due to weak governance systems and ineffective governance practices.

One of the key traits of a professional is adherence to a rigorous set of ethical guidelines. When someone veers too far from ethical standards, their trustworthiness and judgment come into question. The ethics of professional accountants requires them to observe the laws and regulations arranging their jurisdiction and the bodies of work. Avoiding actions that could negatively affect the profession’s reputation is a reasonable commitment that business partners and others should expect. Corporate governance is used to allocate organizational sources well to maximize advantages for all stakeholders and society. Dang et al. (2020) recommended that strong corporate governance maintains investors’ confidence, whose support can help to finance further growth.

Companies who implement good corporate governance principles into working environment life will ensure corporate success and economic growth. They are the basis on which companies can grow. Good corporate governance is a key factor in underpinning the integrity and efficiency of a company. Poor corporate governance can weaken a company’s potential, can lead to financial difficulties, and in some cases can cause long-term damage to a company’s reputation. Companies that practice the main principles of good corporate governance, especially fairness, accountability, responsibility, and transparency, will generally outperform other companies and are in a position to attract investors to finance corporate growth. Concentrated ownership and the associated pyramidal and cross-holding structures create agency conflicts between controlling owners and outside investors. Consequently, controlling owners are perceived to report accounting information for self-interested purposes, causing the reported earnings to lose credibility to outside investors. Second, concentrated ownership is associated with low earnings informativeness as ownership concentration prevents leakage of proprietary information about the firms’ rent-seeking activities (Fan & Wong, 2002; La Porta et al., 1999; Sabir et al., 2019). Family ownership has a significant influence on the type of earnings management selected. Firms with a high proportion of family ownership and non-business groups are more inclined to choose efficient earnings management than other types of firms. The |studies in Indonesia bring out exceptional possibilities to offer besides insights into company governance’s effectiveness in proscribing earnings management.

2. Literature Review

2.1. The Effects of Family Ownership on Earnings Management

Different ownership structures inspire firms to act differently. Specifically, family ownership will have an effect on the demand and delivery of financial reporting quality due to the impact of entrenchment and the impact of alignment effect (Wang, 2006). The traditional conventional view holds that family companies are much less efficient due to the fact focused ownership creates incentives for controlling shareholders to take wealth from minority shareholders (Shleifer & Vishny, 1986). Wang, (2006) investigated the relationship between founding family ownership and earnings quality using data from the Standard & Poor’s 500 companies. Existing literature has documented that financial reporting is of higher quality when firms have stronger corporate governance mechanisms and greater demand for quality financial reporting.

Wang (2006) has presented two competing theories on the effect of founding family ownership on earnings quality, namely the entrenchment effect and the alignment effect. The empirical results showed that, on average, founding family ownership is associated with higher earnings quality. In particular, Wang (2006) found consistent evidence that founding family ownership is associated with lower abnormal accruals, greater earnings informativeness, and less persistence of transitory loss components in earnings. Besides, the results suggested a nonlinear relation between family ownership and earnings quality (Demsetz & Lehn, 1985; Shleifer & Vishny, 1986). The alignment effect proposes that the interests between principals and agents be more aligned in familyowned companies so that agency problems are minimized (Bhaumik & Gregoriou, 2010).

Family-controlled companies should be more efficient than publicly owned companies because monitoring costs are lower in family-owned companies (Fama & Jensen, 1983). Berrone et al. (2020) found that family-controlled companies prevail, comply with precise strategies, and outperform non- family-controlled companies in countries with excessive business legitimacy index (FBLI) scores. Anderson et al. (2003) suggested that although founding family ownership is quite prevalent and significant in U.S. industrial companies, it did not prove that ownership of the founding family in a public company continues to lead to the takeover of the wealth of minority shareholders. Anderson et al. (2003) found that bondholders view the establishment of family ownership as an organizational structure that better protects their interests since family ownership reduced the cost of debt financing relative to non-family firms.

In developing countries, controlling ownership is often in families’ hands (Claessens et al., 2002; La Porta et al., 1999). As the manifestation of a bigger ideology of shareholder value maximization, the idea of good governance is designed for and supported via way of means of large, manager-managed listed corporations. The increasing adoption and growing legitimacy of good governance have led to the formation of dominant institutional logic, which family firms experience pressure to adopt. Particularly strong is the pressure to increase the independence of the board of directors. While the process of change towards more independent boards may not necessarily contribute to increased economic efficiency or fully fulfill the governance needs of family firms, these firms continue to adopt such practices (Ponomareva & Ahlberg, 2016). They have a really strong commitment that they will do almost anything to ensure the company’s survival, including turning a blind eye to deviant management behaviors such as earnings management (Hasnan et al., 2013).

On the opposite hand, family management can increase the chance of expropriation of non-family minority shareholders (Cueto, 2013). Family owners can take over the company resources and appoint families who do not qualify members to key positions (Claessens et al., 2002). Brotherly rivalry, generational jealousy, unfounded compensation, and ‘irrational’ strategic choices can break a corporation’s value in a family business (Gomez-Mejia et al., 2010). However, managers of publicly indexed firms are closely influenced through powerful block holders, which include institutional investors or founders and their families(Devers et al., 2013).

Efficient structures within family-controlled companies are less common in companies with business groups, commonly found in Indonesia. If there is a share of capital invested in a public company, then the takeover of minority shareholders and opportunistic earnings management will generally tend to arise even though the organization is managed by the family (Siregar & Utama, 2008). According to Stockmans et al. (2010), the issue of the low quality of corporate disclosures, in this case, earnings management, is because of the high stage of concentration of share ownership and the dearth of marketplace monitoring which causes the high opportunity of controlling shareholders to expropriate or take over minority shareholders. So, the hypothesis that we propose is as follows:

H1: Family ownership has a positive effect on earnings management.

2.2. The Implementation of Corporate Governance on Earnings Management

Family companies tend to do earnings management. It is proven that when reporting earnings, for example, family companies in Taiwan announced lower profit values compared to non-family companies (Chi et al., 2015). It is usually recommended in family-managed companies that a more independent board and accurate company governance can keep away from such problems (Anderson et al., 2003). Therefore, the role of good corporate governance is vital in overseeing the running of the company (Cheng, 2014).

Corporate governance is the combination of rules, processes, or laws by which businesses are operated, regulated, or controlled. The term encompasses the internal and external factors that affect a company’s stakeholders’ interests, including shareholders, customers, suppliers, government regulators, and management (Yoshikawa et al., 2014). Earnings management problems can be minimized by monitoring the company through corporate governance (Nazir & Afza, 2018). Earnings management practices can be reduced through monitoring mechanisms to harmonize differences in owners and management interests (Jensen & Meckling, 1976).

Most of the earlier studies measured only a few items or components of corporate governance partially, for example, size of the board of commissioners, independent commissioners, board interest commissioners, information and competence of the board of commissioners (Abdul Wahab & Holland, 2012; Khaoula & Ali, 2012). One example of study results that showed the relationship between boards and earnings management is the study of Al-Absy et al. (2020) that alerts policy-makers, stakeholders, and researchers to the influence of a board chairman serving on the audit committee in curbing earnings management. Whereas Nguyen and Duong (2020) have found that two variables – the foreign members of the board of directors and audit committee – have an opposite effect on the earnings management behavior of Vietnamese listed banks.

Due to the importance of corporate governance in our business world today, especially after the frequent nonstop financial crises, and if only one corporate governance mechanism cannot fulfill its objectives, researchers have recently got here with a bundle of corporate governance mechanisms, which may supplement or replace one another. The corporate governance of a company involves establishing a set of relationships between the company’s management, its board of directors, its shareholders, and other stakeholders. Companies need governance, and this part of corporate governance, where the main participants are: owners, managers, and the board of directors where the balance between these three actors determines the success of the correct governance of the company (Al-Baidhani, 2014). This study uses corporate governance measurement comprehensively, including mechanisms and principles of corporate governance using the board of commissioners, shared ownership and transparency, and disclosure and auditing. Based on the description above, it is expected that the interaction of corporate governance with family ownership reduces earnings management. Thus, it can be concluded in a hypothesis, namely:

H2: corporate governance decreases the positive influence of family ownership on earnings management.

H2a: The board of commissioners decreases the positive influence of family ownership on earnings management.

H2b: Ownership of shares decreases the positive influence of family ownership on earnings management.

H2c: Transparency, disclosure, and auditing decrease the positive influence of family ownership on earnings management.

3. Research Methods and Materials

3.1. Population and Sample

The population in this study is non-finance companies listed on the Indonesia Stock Exchange in 2011–2017. The purposive sampling method was employed as the sampling technique. It refers to determining the sample by using certain considerations. The sample criteria used must have complete available data, including:

1. Annual Report for 2011–2017

2. Financial Report for the period 2011–2017

3.2. Operational Definition and Variable Measurement

3.2.1. Earnings Management

Watts and Zimmerman (1986) asserted that earnings management occurs when managers have discretionary behaviors related to accounting numbers with or without restrictions, and such a behavior can be adopted to maximize company value. Earnings management in this study was measured using the Kothari Model that is also known as the performance-matched model.

To estimate the discretionary accruals, this study used a cross-sectional version of the Jones model modified by Kothari, Leone, and Wasley (2005). In this model, return on assets (ROA) is added as an additional control variable. Previous research found that the Jones model was precisely incorrect for a company performing well or performing poorly (Dechow et al., 1995; Kothari et al., 2005).

\(\begin{aligned} \text { Accruals }_{t t}=& a+b\left(1 / \text { Assets }_{t-1}\right)+c \Delta \text { Sales }_{t}+\\ & d \mathrm{PPE}_{t}+e \mathrm{ROA}_{t}+\mu_{t} \end{aligned}\)       (1)

In Regression (1), total accruals (accruals); changes in sales (ΔSales); and gross property, plant, and equipment (PPE) are respectively defined by total assets at the beginning of the year. Where ACCR_it = total accruals for the company in the year t defined as earnings before extraordinary items of cash flow from operational activities; TA_it–1 = total assets for the company in the year_{t– 1} observation. By using estimated parameters to the actual value for each company-year, it yields an estimation of total accruals. The difference between the total of actual accruals and estimation is a proxy for discretionary accruals (AB_DA). AB_DA shows discretionary accruals. Values for the dichotomous variable, AEM is one for firm/years with a positive AB_DA, indicating the incidence of earnings management by increasing earnings. AEM is marked as a zero value for other conditions.

3.2.2. Family Ownership

According to La Porta et al. (1999), a family-owned enterprise can be described as any enterprise wherein more family members are worried and the majority of ownership or management lies inside a family. A family business is a commercial organization in which decision-making is influenced by multiple generations of a family, related by blood or marriage or adoption, who can influence the vision of the business and the willingness to use this ability to pursue distinctive goals. In a family business, two or more management team members are drawn from the owning family. Family businesses can have owners who are not family members. Family businesses may also be managed by individuals who are not family members (Chua et al., 1999).

We categorized family-owned companies if the percentage of the share ownership is 20% or more (Claessens et al., 2002; Singla et al., 2014; Purkayastha et al., 2019), or at least one of the family members is sitting on the board (Banalieva et al., 2015). The family ownership will be measured by using a dummy variable, which puts a value of 1 for companies with family ownership and a value of 0 for companies with non-family ownership.

3.2.3. Corporate Governance

This study uses components of the corporate governance index, according to Javed and Iqbal (2006) and Yorke, Amidu, and Agyemin-Boateng (2016) who have adapted it to the context of corporate governance in Indonesia. The components include mechanisms and principles of corporate governance. The advantage of corporate governance measurement in this study is the components scale of the corporate governance index using a ratio scale whose measurement results can be distinguished, sorted, and compared. Most of the earlier studies measured only a few items or components of corporate governance partially, for example, size of the board of commissioners, independent commissioners, board interest commissioners, information and competence of the board of commissioners (Abdul Wahab & Holland, 2012; Khaoula & Ali, 2012) and using a nominal scale that is using variables dummy in measurement.

Corporate governance is measured comprehensively, including the board of commissioners, shares ownership and transparency, and disclosure and auditing. In this research, corporate governance is measured comprehensively, which includes mechanisms and principles of corporate governance using the corporate governance checklist from the research of Yorke et al. (2016); Financial Services Authority Regulation Number 33 / POJK.04 / 2014; Financial Services Authority Regulation Number 34 / POJK.03 / 2014; Regulations Financial Services Authority Number 55 / PJOK.04 / 2015; Lin et al., 2011; (Lisowsky et al., 2010).

Based on the information above, this study’s corporate governance includes the board of commissioners items, shared ownership items and transparency, and disclosure and auditing. Furthermore, a checklist is carried out on each corporate governance item and provides a score for each item disclosed. The scoring technique uses an unweighted dichotomy scale with the criteria that if the corporate governance items are disclosed by the company; they are assigned a number and if not given a zero. The use of an unweighted dichotomy scale aims to avoid different perceptions of items in corporate governance expressed by the company. Besides, it is also to reduce the level of subjectivity in giving weight to corporate governance items. Based on the scores obtained, the next step is to calculate corporate governance percentage using the following formula.

\(\text { Discl }=\frac{\sum_{i j} D_{i t e m}}{\sum_{i j} A D_{i t e m}}\)

Where:

Discl : The percentage of corporate governance/board commisioner/shareholding/transparency disclosures, and auditing disclosure

D_item : The total disclosure, and

AD_item : The total items in disclosures

3.2.4. Control Variable

Leverage is the level of debt used by the company in financing. Leverage describes the risk level of a company that is measured by comparing the company’s total liabilities with the company’s total assets. The greater the level of debt the company has, the greater the risk borne by the company. Company size is calculated from the logarithm of total assets owned by the company.

3.2.5. Hypothesis Testing

Hypothesis testing in this study is carried out with the following equation model:

AEM = a0 + CG + FF + FF × CG + SIZE + LEV       (1)

AEM = a0 + CG + FF + DK + KS + TPA + FF × DK + FF × KS + FF × TPA + SIZE + LEV       (2)

Where

AEM : Accrual Earnings Management

CG : Corporate Governance

FF : Family Ownership

DK : Board of Commissioners

KS : Shareholding

TPA : Transparency, Disclosure and Auditing

Size : Company Size

LEV : Level of Debt

4. Results and Discussion

4.1. Descriptive Statistics

Table 1 presents the results of descriptive statistical analysis for research with 198 observations with a sample of 33 companies taken from the annual financial statements of companies listed on the Indonesia Stock Exchange for the period 2011–2017. The lowest earnings management value during the study period is –0.1819. It shows that the company’s management carries out the earnings management practices by lowering earnings. The maximum value of the earnings management is 0.2291. The average value of the earnings management of –0.0016 shows that earnings management practices by decreasing the company earnings.

Table 1: Descriptive Statistics

The maximum value of corporate governance is 0.9524, which means that the company has quite well performed the corporate governance function. The companies have carried out and implemented the principles and mechanisms of corporate governance properly and implemented all applicable regulations. Based on the year of observation, the minimum value is 0.3690, which means that only some of the companies implemented regulations. The average value of the companies that have conducted corporate governance is 0.7706, which means that 77.6% of the companies sampled in this study have done good corporate governance. Governance is expected to be an important tool for investors in controlling managers in managing investment and not embezzling investor funds into unprofitable projects (Larcker & Richardson, 2007).

Corporate Governance in this study includes the three components, namely DK (board of commissioners), KS (share ownership), and TPA (transparency, disclosure, and auditing). The average DK value is 0.8169 which means that 81.69% of the observed companies had performed the board of commissioners’ function. The average value of shareholding (KS) is 0.6479, which means that the companies have implemented a component of shareholding of 64.79%. The average value of the TPA components (transparency, disclosure, and auditing) is 0.8470, which means that the company has performed the transparency, disclosure, and auditing function of 86.70%. The TPA standard deviation value in this study is 0.125167. Leverage has a minimum value of 0.38590 and the maximum value is 0, 00031 with an average of 0.3859, and the standard deviation is 0.1840. The average value of 0.3859 indicates that the companies during the period of observation managed to use assets financed with a debt of 38.59%. Size has a minimum value of 11.4930 and a maximum value of 19, 5047, with an average of 14.8194, and the standards deviation is 1.7937.

4.2. Hypothesis Testing Results

4.2.1. Family Ownership, Corporate Governance and Earnings Management

Based on the data regression estimation model selection panel testing (the Chow test, the Hausman test, and the Lagrange multiplier test), it was concluded that the best model is common effect.. The following are the results of the common effect model data presented in Table 2. The F value is 4.2041 with a significance of 0.0012, illustrating that this regression model can be used to estimate the effect of the independent variable on the dependent variable. Furthermore, corporate governance has a positive effect on earnings management with a coefficient of 0.2175 at α : 1% with a value of t: 3.8838. This shows that the higher the corporate governance, the higher the earnings management. The family ownership variable appears to have a positive effect on earnings management with a coefficient of 0.0045 at α : 1% with a value of t: 4.4464. It means that the higher the family ownership, the higher the earnings management. Furthermore, the interaction variables of family ownership with corporate governance show significant results with a coefficient of –0.0058 at α : 1% with a value of t: –4.4723. These results prove that corporate governance is a moderating variable. Furthermore, the control variable size and leverage are both insignificant, so the two variables are not proven as controls.

Table 2: Regression Test Results

Where: significant level 0.01, **Significant level 0.05 and *significant level 0.10. C is a constant. CG is corporate governance. FF is family ownership. FF × CG is the interaction of family ownership profit with corporate governance. LEV is leverage. ROA is the return on assets, and SIZE is the size of the company.

The results in Table 2 indicate that family ownership positively affects earnings management. Hence, earnings management is higher if family ownership is high. This indicates that in Indonesian family firms, earnings management becomes more intensive than non-family firms. The condition for the emergence of more intensive earnings management could be because majority shareholders can ask management to manage earnings to hide the expropriation of the minority. Further results show that corporate governance is a full moderator by declining family ownership’s effect on earnings management. This shows that corporate governance works properly.

4.2.2. Board of Commissioners, Transparency, Shareholding, Family Ownership, and Earnings Management

Model (2) examines the CG variable components, namely the board of commissioners, shareholding, and transparency in earnings management. Based on the data regression estimation model selection panel testing through the Chow test, Hausman test, and Lagrange multiplier test, the best model has a common effect with control variables. The following are the results of the common effect model data presented in Table 3. Based on the results of data processing shown in Table 3, it is stated that the F value is 2.4043 with a significance of 0.0133, illustrating that this regression model can be used to estimate the effect of the independent variable on the dependent variable. Regression test results show that KS share ownership (KS) positively affects earnings management with a coefficient of 0.0888 and a t value of 2.1479. On the other hand, the board of commissioners (DK) and transparency, and disclosure and auditing (TPA) have no effect on earnings management with coefficients of 0.0556 and 0.07950 and t values of 1.5694 and 1.6218, respectively. The family ownership (FF) has a positive effect on earnings management with a coefficient of 0.0043 at α : 1% with a value of t: 3.9204. The FF × DK interaction variable has a negative effect on earnings management with a coefficient of –0.0014 at α: 10% with a value of t: 1.7800. The FF × KS interaction variable has a negative effect on earnings management with a coefficient of –0.0019 at α : 5% with a value of t: –2.1564. The FF × TPA interaction variable has a negative effect on earnings management with a coefficient of –0.0024 at α : 1% with a value of t: –2.6264. The control variable of size and leverage are both insignificant, so they are not proven as control variables.

Table 3: Regression Test Results

Where: Significant level 0.01, **significant level 0.05 and *significant level 0.10. C is the constant. DK is the board of commissioners. KS is the shareholding. TPA is transparency. FF is family ownership. FF × DK is the interaction of family ownership with the board of commissioners. FF × KS is the interaction of family ownership with shareholding. FF × TPA is the interaction of family ownership with transparency. LEV is leverage. ROA is the return on assets, and SIZE is the size of the company.

The results above indicate that the board of directors and transparency do not directly affect earnings management, but the shareholding has a positive effect on earnings management. Therefore, the higher involvement of owners in the management of the company in the board of commissioners’ membership encourages management to make decisions and act in the short term by carrying out earnings management. The result of family ownership in the model (2) is robust with the model (1), where family ownership positively affects earnings management. Furthermore, the interaction of family firms with shareholding and its interaction with the board of commissioners and with transparency, disclosure, and auditing decreases earnings management.

4.2.3. Additional Analysis

The additional test results shown in Table 4 show that the F value of 6.505 with a significance value of 0.002 illustrates that this regression model can be used to estimate the effect of the independent variable on the dependent variable. The regression test results show that family ownership has a positive effect on earnings management with a coefficient of 1.18 × 10–4 at α: 10% and a value of t: 2.0947. The variable corporate governance has a negative effect on earnings management with a coefficient of –0.0716 at α: 1% with a value of t: –2.6444.

Table 4: Additional Test Results: Model without Moderating Variables

Where: Significant level 0.01, **significant level 0.05 and *significant level 0.10. C is a constant. CG is corporate governance. FF is family ownership. FF × CG is the interaction of family ownership profit with corporate governance.

The results show that family ownership increases earnings management, whereas corporate governance decreases earnings management. These results indicate that family firms operate less efficiently using short-term profits by performing earnings management. Thus, this indicates that there is an entrenchment effect on family firms in Indonesia. Furthermore, the result shows that corporate governance leads to earnings management. This indicates that corporate governance works properly.

Based on the results of the additional test in Table 5, it is stated that the F value of 8,427 with a significance of 0.000 illustrates that this regression model can be used to estimate the effect of independent variables on the dependent variable. The regression test results show that family ownership has a positive effect on earnings management with a coefficient of 1.18 × 10–3 at α : 1% and a value of t: 3,435. The variable corporate governance (CG) has no effect on earnings management with a coefficient of 0.0768 with a value of t: 1.498. The interaction variable CG × FF has a negative effect on earnings management with a coefficient of 2.99 × 10–3 at α : 1% and a value of t: –3.2055.

Table 5: Additional Test Results: Model with CG as Moderating Variables

Where: significant level 0.01, **Significant level 0.05 and *significant level 0.10. C is a constant. CG is corporate governance. FF is family ownership. FF × CG is the interaction of family ownership profit with corporate governance.

The results above indicate that family ownership increases earnings management, whereas corporate governance does not affect earnings management. The interaction variable of family ownership and corporate governance has a negative effect on earnings management, indicating that corporate governance reduces the effect of family ownership on earnings management.

5. Conclusion

Earnings management is a transfer of value from shareholders to management because funds from shareholders managed by managers are used for profitable investments. Further, shareholders are paid dividends. Consequently, managers can take opportunistic actions in managing funds provided by shareholders. In developing countries, including Indonesia, most companies are still controlled by family ownership. Type II agency problems in family companies have a greater motive for taking over minority interests than non-family companies. The implementation of corporate governance is often identified with the implementation of good mechanisms in companies that are expected to reduce the level of earnings management in family companies.

The study shows that family ownership has a positive effect on earnings management. This indicates that in Indonesian family firms, earnings management becomes more intensive than non-family firms. Corporate governance reduces the effect of family ownership on earnings management. Hence, corporate governance can reduce earnings management in family firms. Furthermore, from the three components of corporate governance: the board of commissioners, shareholding, and transparency, the term shareholding precisely encouraged managers to conduct earnings management. Besides, the three core bodies of corporate governance lowered the significance of shareholding toward earnings management.