The role of independent commissioner in the effect of family ownership on capital structure in family companies in Indonesia

* Corresponding author, email address: rahmatsetiawan@feb.unair.ac.id ABSTRACT The purpose of this research is to analyze the effect of family ownership on capital structure with independent commissioner as a moderating variable in family companies in Indonesia. The study used the population consisting of family businesses listed on Indonesia Stock Exchange. The data were taken from the Indonesian Capital Market Directory (ICMD), financial statements, and annual report and the observation period was done from 2012 to 2018. The sampe was taken by a purposive sampling based the specified criteria. Based on the analysis result, it can be concluded that family ownership has a significant negative effect on capital structure. Independent commissionerweakens significantly the negative effect of family ownership on capital structure. Furthermore,firm sizehas a significant positive effect on capital structure, but profitability has a significant negative effect on capital structure.


INTRODUCTION
One of the businesses which is very common in many countries is family business. Historically and sociologically, companies in Indonesia are family owned and controlledcompany. Although family companies have become public companies, their control is still held by the family. In Indonesia, the number of majority shares owned by the founder or founding family is varied, ranging from 4.48% to 96.64% (Wijayanti, 2014). Besides that, these companies, in general, are active family companies because the family is not only the majority shareholder but also serves as the company's board of directors. About 67% of companies registered in Indonesia are controlled by families (Claessens, Djankov, and Lang, 2000). In addition, based on a survey conducted by PricewaterhouseCoopers (PwC) in 2014 that 87% of family companies in Indonesia are active family companies where the role of the family is not only as a shareholder but also in management.
Family companies in Indonesia contribute to around 40% of market capitalization and have a considerable effect in key industries such as consumer goods, property, and agriculture. Research by the Boston Consulting Group shows that Indonesian family businesses that can survive in the first generation are only around 30% and around 9% can be passed on to the third generation. This can indicate that many family businesses in Indonesia are still in a growth phase. Besides that the prospects in family businesses have significant uncertainties so that making decisions taken by family companies in Indonesia so that companies can continue to survive is interesting to study.
Constraints in achieving company goals are triggered by differences between majority and minority shareholder'sinterests which will eventually lead to agency problems and have an impact on the company's performance as explained by agency theory type II. The problem that arises in family businesses is agency problem type II because there is a conflict between the majority shareholders (family) and minority shareholders (nonfamily). Family shareholders will try to maximize the company's value and prioritize their personal interests rather than the minority shareholders' interests. They also minimize the risks faced by the company so that the company can be passed on to the next generation. Yet, the non-family shareholders need to obtain dividends or capital gains so that they can lead to prosperity and wealth of minority shareholders. Also, previous research shows when cash flow rights is lower than their family control, family shareholders have a strong incentive to expropriate the wealth of minority shareholders (Faccio, Lang, and Young, 2001).
Strategic decisions like external financing can be influenced by sound corporate principles as such decisions are taken at the board level (Wasim, 2016). Capital structue is one of the most important decisions that must be taken by company management. It shows how much debt is used to fund the company's assets. In the family business, this decision is affected by family control in the business.
The difference between the majority (family) and minority shareholder's interests will be more prominent in the family company. The founders and their families have an incentive to influence financing policy as they wish. In this context, two motivations for influencing capital structure decisions are excessive risk avoidance and control considerations (Schmid, 2013).
The literature shows that strong and non-diversified ownership such as family ownership can create more cash available in family firms. This can result in companies less dependent on the use of debt for financing (Jensen, 1986). Shleifer and Vishny (1986) also found that family firms will look for sources of funds with a low probability of default because the owners and managers do not want to use funding with high risk. This imples there is a greater dependence on their own capital than on the use of debt in the capital structure they. Faccio et al., (2001) also provided evidence that family firms use lower debt to reduce fixed commitments to their cash flows.
The independent commissioner in a family company plays an important role in reducing agency problems between family and non-family shareholders. This is because, according to UUPT, the independent commissioner has the aim to create a more objective climate and to place equality (justice) among various interests including corporate interests and the interests of stakeholders. The presence of independent commissioners has an important effect on board effectiveness because independence is assumed to be closely related to the strength of the board of commissioners. In this case, the independent commissioners have greater objectivity and independence in their analysis of management and company behavior (Uribe-bohorquez, Martínez-ferrero and García-sánchez, 2018). In governance in family companies, independent commissioner can influence the board to choose a higher dividend payment policy and/or to adopt a higher level of debt to have better corporate governance (Atmaja, 2010).
Based on the results of the above phenomena and problem formulation, the purpose of this research is to develop the previous research and to test as well as to examine the role of the independent commissioner in the effect of family ownership on capital structure in the family business in Indonesia.

Definition of Capital Structure
The capital structure reflects the use of longterm debt to fund its assets. Sudana (2011: 143), states that capital structure is a comparison between long-term debt and equity. Besides that the capital structure is related to long-term funding decisions to be taken by the company. This can be referred to Margaretha (2014: 305), who states that the capital structure is a permanent financing company that consists of long-term debt and equity capital. According to Sjahrial (2014: 250), capital structure is a balance between the use of loan capital consisting of short-term debt that is permanent, long-term debt, and own capital consisting of preferred shares and common shares.

Definition of Family Ownership
Family ownership is total shares held by family compared to total outstanding shares. Anderson and Reeb(2003), Faccio and Lang(2002), La Porta, Lopez-de-Silanes, and Shleifer(1999) define family ownership as part of the shares owned by the family in a company. Pukthuanthong, Walker, Thiengtham, and Du (2013) also define family ownership as a portion of shares owned by a founder or founding family in a company.
From some of the notions above, it can be concluded that family ownership is the proportion of shares held by family members in the company both individually and through family institutions of the total outstanding shares. Family relationships are identified by tracking the founder's family share ownership directly or indirectly by being identified with the same clan name or last name as the company founder.

Definition of Independent Commissioner
Independent commissioners are members of the board of commissioners. In addition, independent commissioners are free from business relationships or other relationships that can affect their ability to act independently. According to Financial Services Authority Regulation No. 57 of 2017-concerning the Implementation of Governance of Securities Companies Conducting Business Activities as Underwriters and Brokers-independent commissioners are members of the board of commissioners originating from outside the company and fulfill the requirements as independent commissioners as referred to in the Authority Regulations Financial Services. When the board of commissioners consists of more than two people, the percentage of the number of independent commissioners must be at least 30% (thirty percent) of the total members of the board of commissioners.

Effect of Family Ownership on Capital Structure
According to pecking order theory, companies tend to adopt hierarchical financing sequences: first, they use internal resources and then external financing. If external financing is needed, they prefer debt over equity. This happens because there is information asymmetry between managers and potential investors, which limits access to external financing.The same reason is also found in family companies, which tend to use more internal resources, because of the greater potential for expropriation, and external financial costs are more sensitive to unclear information (Chen, Dasgupta, and Yu, 2014;Ma and Tian, 2017).
Family companies have large and concentrated shareholder characteristics, Jensen (1986) shows that strong and nondiversified ownership can result in more free cash flow available in the family company, which make the company more dependent on the company's internal funding. The characteristics of a family company where managers are often the owners also, make them have greater ability to modify the company's portfolio of assets to use benefits and provide funds for themselves, and the family (Ramalho and Da Silva, 2009;Roger and Schatt, 2016). Therefore, in order to continue to benefit itself, funding decisions taken by family companies are driven by a desire to avoid monitoring inherent in the use of external funds (Koropp, Kellermanns, Grichnik and Stanley, 2014). Faccio et al.(2001)also show that family firms have lower debt levels that are used to reduce fixed commitments in the form of interest in their cash flows.
Family shareholders are usually less diversified investors (Anderson and Reeb, 2003). Therefore, they face a high risk of one single asset, that is, a family company. Therefore, families have incentives to reduce risk at the company level (Schmid, 2013). The use of large debt will increase the risk of bankruptcy faced by the company because debt has a fixed cost of interest, and debt has a relatively long bond, so it has a high enough risk. Therefore, families will use more internal funding sources that have low risks such as retained earnings to fund company assets. Therefore, the hypothesis in this study isstated as follows: H 1 : Family ownership has a negative effect on capital structure.

The role of the independent commissioneron the effect of family ownership on capital structure
In resolving agency problems (agency conflicts between principals and agents), among other things, they can do it together with supervisors from outside the company. These advisors have absolutely nothing to do with the company, for example with the presence of independent commissioners (Jensen and Meckling, 1976). Improving the independent commissioner helps to resolve the effective monitoring function of the board towards the management of the company (Purag and Bujang, 2016). Independent commissioners play an important role in monitoring the executive. Due to the effective monitoring, the board tends to align their interests with the interests of shareholders and, as such, can use more debt in their capital structure (Tarus and Ayabei, 2015). This is consistent with agency theory, a board with a large number of independent commissioners can limit the implementation of managerial policies. This can exploit the wealth of minority shareholders. Independent commissioners help to resolve disagreements between managers and residual claimants. A board consisting of independent commissioners will provide a balance so insiders do not take advantage of their position and shareholder wealth (Badu and Appiah, 2017).
Similarly, referring to the resource dependency theory, companies with large independent commissioners are more likely use higher leverage because they have access and ability to connect companies with debt capital providers. Thus, the hypothesis can be stated as the following.

H2: Independent commissioner weakens the negative effect of family ownership on capital structure
Framework Based on the relationship between variables that have been stated, a framework can be drawn in relation to the effect of family ownership on capital structure with moderating variable toward the in-dependent commissionerin family business in In-donesia. The framework is shown in Figure 1.

RESEARCH METHOD
The population consists of family businesses listed on Indonesia Stock Exchange. The data is obtained from the Indonesian Capital Market Directory (ICMD), financial statements, and annual report. The observation period is from 2012 to 2018. The method used for sampling technique is a purposive sampling that is the sample selection technique by using the specified criteria. A seven-year study period is chosen because it describes the relatively new condition in the family business in Indonesia, so, it is expected that the result of this research will be more relevant to understand the actual conditions in Indonesia. The companies that become the samples are companies chosen based on purposive sampling The method to attain representative samples is in accordance with the specified criteria. The sample criteria that will be used are as follows:

Companies listed on the Indonesia Stock Exchange
The company published annual financial statements that end on December 31 in 2012-

Figure 1
Research Framework 2018. The financial statements are presented in rupiah The company is a non-financial company. The company is a family company.
Moderation variables commonly can be classified into 2 types, namely pure moderation, and quasi moderation, In brief, the 2 types of classification of moderation variables can be explained as follows:

Pure moderation
Pure moderation is a variable that moderates the relationship between independent variables and the dependent variable in which pure moderation variables interact with independent variables without becoming an independent variable.Pure moderation occurs when the moderation coefficient is declared insignificant but the interaction coefficient is statistically significant.

Quasi moderation
Quasi moderation is a variable that moderates the relationship between the independent variable and the dependent variable in which the quasi moderation variable interacts with the independent variable as well as being an independent variable.Quasi moderation occurs when the moderation coefficient and interaction coefficient are statistically significant.

Research Data
The data used were secondary data,taken from the annual and financial report for the period of 2012-2018. This study used a quantitative approach by conducting a hypothesis test. The sources of data were collected from Indonesian Capital Market Directory (ICMD), financial statements, and annual reports obtained from the official website of Indonesia Stock Exchange, www.idx.co.id. The data are collected, selected, and taken as the samples and processed in the study.

Research Variables
The dependent variable in this study is the capital structure measured by the book value debt ratio and the market value debt ratio, and the independent variable in this study is family ownership. This study usesa moderating variable namely,independent commissioner.

Operational Definition and Variable Measurement Measurement of capital structure
The capital structure was measured by two measurements, namely the book value of debt ratio (BOOKDEBT) and the market value debt ratio (MARKETDEBT) based on Baek, Cho, and Fazio(2016)  The book value debt ratio measures the proportion of funds sourced from long-term debt to finance company assets. This ratio is past-oriented because the data used to measure is historical data derived from the company's financial statements. The greater this ratio will show the portion of the use of long-term debt in financing investment in assets is greater, while the market value of the debt ratio measures the value of debt to the market value of the company. This ratio is future-oriented because it uses a market value in its measurement. The greater this ratio shows that the higher the value of debt to the market value of the company.

Measurement of family ownership
Family ownership (FAMOWN) in this study was measured bytotal shares held by family (FAMSHARES) compared to total outstanding shares (TSHARES), based on Baek, Cho and Fazio(2016) research with the following formula:

FAMOWN = FAMSHARES / TSHARES
The greater this ratio reflects the greater the family control in the company and also the higher the family participation in company management.

Measurement of independent commissioner
The independent commissioner (INDEPCOM) was measured through the proportion of independent commissioners in management to the number of commissioners based on Atmaja(2010), as measured by:

INDEPCOM= Number of Independent commissioner / Total Commissioner
The greater this ratio reflects that the greater the oversight of independent commissionerin the company.

Control Variables
Several factors expected to influence capital structure decisions.According toFrank and Goyal(2009), factors that influence capital structure arefirm size, tangibility, and profitability.
Firm size (SIZE) is measured by the natural logarithm of assets, firms that are large (in terms of assets) tend to have higher leverage because large, more diversified, firms face lower default risk. In addition, older firms with better reputations in debt markets face lower debt-related agency costs. Thus,more mature firms to have relatively more debt.
Tangibility (TANG) was measured by the ratio of plant, property, and equipment to total asset, firms that have more tangible assets tend to have higher leverage.The higher tangible assets ratio provides a greater level of security for creditors, due to collateral for loans provided by creditors so that in the event of bankruptcy the creditor can liquidate the assets guaranteed by the company to meet its obligations.The higher collateralwas given to creditors, the greater the opportunity the company has to get a bigger debt because the company is considered capable of paying off its loans.
Profitability (ROA) is measured by the ratio of net income to total assets, firms that have more profits tend to have lower leverage.Profitable firms face lower expected costs of financial distress and find interest tax shieldsmore valuable. Thus, the tax and the bankruptcy costs perspective predicts that profitable firms use more debt.

Descriptive Statistics
The descriptive statistic of this study illustrates the minimum, maximum, and mean value of the variables used in family companies listed on the Indonesia Stock Exchange in 2012 to 2018 are shown in Table 1.
The dependent variable is the capital structure while the independent variable in this study is family ownership. And the control variables are the size of the company, profitability, and tangible assets, and the moderating variable in this study is the independent commissioner.
Table 1, shows that the capital structure as measured by the book value debt ratio and the market value debt ratio of family companies in Indonesia in the period 2012 to 2018 showed an average of 0.1903 and 0.2448. The lowest values for these variables are 0.0022 and 0.0030, while the highest values are 0.5864 and 0.8330. This figure shows that companies use more funding than debt to fund their assets.
The family ownership variable shows an average of 0.5774. The lowest value for the family ownership variable is 0.0439, while the maximum value for the family ownership variable is 0.9720. This figure shows that the majority shareholders of non-financial companies in Indonesia during the observation  The independent commissionervariable shows an average of 0.3996. The lowest value for the variable is 0.0151, while the maximum value for the variable family ownership is 0.8000. This figure shows that almost all companies have independent commissioners because they are under regulations set by the Financial Services Authority in Indonesia.
The following classical test assumptions are performed to determine whether there are problems with the regression model.

Normality Test
Based on the Kolmogorov-Smirnov normality test in table 2 shows that Assymp. Sig (2-tailed) of 0.052 and 0.056. This shows that the residual value in the regression equation is normally distributed because of the Assymp value. Sig.

Multicollinearity Test
Multicollinearity test was doen to analyze whether in themodel of regression there is a relationship between independent variables. The tool used to conduct a multicollinearity test is the Variance Inflation Factor (VIF). If the VIF value is < 10 or 0.1, this means that multicollinearity does not occur. However, if the VIF value is > 10, the data variable experiences multicollinearity (Ghozali, 2016b: 103). Multicollinearity test results can be seen in Table 2 Based on Table 3, the results of testing in both models of this study indicate that tolerance values > 0.10 and VIF <10 except the interaction variables between family ownership and independent commissioner,. It can be seen that there is no multicollinearity between the independent variables in the regression model.

Autocorrelation Test
Autocorrelation test serves to test whether there is a relationship between confounding errors in the period of this study and those in previous studies. The tool used in this test is durbin watson. The results are presented in Table 4.
Based on the spss output in the table above, the d-w value is obtained when there is no autocorrelation. According to gujarati and porter (2009:436), the regression test did not experience autocorrelation if the d-w value was between d u and 4-d u . Based on the results of the autocorrelation test it can be concluded that in this research model there is no autocorrelation.

Heteroskedasticity test
Heteroscedasticity test to describe the case where the variance of errors or the model is not the same for all observations, while often one of the basic assumptions in modeling is that the variances are homogeneous and that the errors of the model are identically distributed. heteroscedasticity testing in this study uses the glejser test.the glejser test proposes to regress a residual absolute value as a dependent variable with an independent variable. If the independent variable significantly influences the absolute residual, then there is an indication of heteroscedasticity, on the contrary, if the   Table 5, shows that all independent variables, both in the book debt and market debt models are not significant to their absolute residuals (sig. Greater than 0.05). Thus, it can be concluded that there are no symptoms of heteroscedasticity.

Results of Analysis and Discussion
Based on Table 6, the results of the regression model analysis show that family ownership has a significant negative effect on the two measurements of capital structure, namely the book and market value debt ratio. This means that the higher the family ownership, the lower the use of debt in funding company assets. This is because family ownership reflects the level of family control over the company, and family shareholders are generally less diversified shareholders, so the risks faced by family shareholders are quite large. Therefore, to reduce the risk faced by the company, family ownership tends to reduce funding by using debt because it has a high risk and increases the  The sign *, **, *** shows significant at α of 10%, 5%, 1% The numbers in parenthesis is the t-statistic value Source: spss output data that has been processed risk of bankruptcy that the family shareholders want to avoid.
The interaction of family ownership and the independent commissioner has a significant positive effect on capital structure. This shows that independent commissioners the negative influence between family ownership, which means that the greater proportion of independent commissionersin the company. This is because the independent commissioner can reduce agency problems faced by the family company because the existence of independent commissioners can make the supervision and control of the board of managers more effective. These results are consistent with atmaja (2010) which shows that independent commissioner weakens the negative influence between family ownership and capital structure.
The size of the company has a significant positive effect on the ratio of capital structure. This is because the risk of bankruptcy faced by large companies is lower when compared to smaller companies. Large companies are more able to diversify their businesses than small companies. In addition, large companies can also increase the level of creditor confidence, because large companies are considered as an indicator of a company that has the good financial capability so that the company is able to repay its loans. Roa has a significant negative effect on the book value debt ratio and the market value debt ratio. This means that the higher the roa, the lower the use of debt in funding company assets.these results are consistent with the research of frank and goyal (2009) which shows that company size has a positive effect on capital structure.meanwhile, tangibility does not significantly influence both measurements of capital structure.
The coefficient of determination (r 2 ) for each book value debt ratio variable and the market value debt ratio variable are 0.443 and 0.504. This shows that 44.3% of the book value debt ratio and 50.4% of the market value of the debt ratio can be explained by the variables studied, while the remaining 55.7% and 49.6% are explained by other variables not contained in the model.

CONCLUSION, IMPLICATION, SUG-GESTION AND LIMITATION
This study was conducted to analyze the role of the independent commissioner in the effect of family ownership on capital structure in the family business in Indonesia. This research was conducted by analyzing family firms in Indonesia from 2012-2018. Based on the results of the analysis and discussion, the researchers rpovide several conclusions. First,family ownership has a significant negative effect on the capital structure both on the measurement of book debt ratio and market debt ratio. Second,independent commissioner weakens significantly the negative effect of family ownership on capital structure.It means that the negative effect of family ownership on the capital structure will be weaker when the proportion of independent commissioners is higher.Third, firm size has a significant positive effect on capital structure, but profitability has a significant negative effect on capital structure, meanwhile, tangibility has no effect on capital structure.
Based on these research results, there are several implications. First, theoretically, the result showing that family ownership has a negative effect on capital structure supports the theoretical idea that the higher the family ownership, the greater the risk faced by the family therefore the company will reduce the use of debt to reduce risk. Second, the more independent commissioners can increase the bravery of family firms to add more debt to finance their profitable investment, so the family firm can utilize more debt to finance their profitable projects with monitoring by more independent commissioners.
This study still has some limitations in terms of measurement and variable. This includes: (1) proxy corporate governance mechanism to reduce agency problems using only one variable, independent commissioner, (2) measurement of family ownership that uses only one measurement. Therefore, it is suggested that further research: (1) add other corporate governance variables, such as board size, audit committee and remuneration committee, (2) add other measurements in the measurement of family ownership, such as the use of dummy variables.