THE INFLUENCE OF CAPITAL STRUCTURE, COMPANY SIZE, CORPORATE GOVERNANCE ON COMPANY PERFORMANCE WITH AGENCY COST AS INTERVENING VARIABLES

This study examines the relationship between capital structure (DER), company size, and corporate governance (GCG) on company performance with a mediating variable in the form of agency costs. The relationship between variables tested using the Partial Least Square (PLS) method where the sample consists of 57 manufacturing companies in Indonesia that are listed on the Indonesia Stock Exchange and publish financial reports in 2018. Based on the results of tests conducted, capital structure has a positive and significant effect on agency theory and also agency theory has a positive and significant result on firm performance. Meanwhile, agency theory has a partial mediating effect in the relationship between capital structure and firm performance. Therefore, recommended that manufacturing companies pay attention to the composition of the capital structure and agency cost to get the best performance.

In this era of globalization, the development of the world economy has generally increased. Although on the one hand there is an economic downturn for certain countries, in general in the future world economic growth will increase. Economic growth is thus global in nature; moreover technological developments have made the world even more limitless in terms of distance and time. The reduction in barriers related to distance and time has resulted in the emergence of new market shares which result in the presence of even greater business opportunities for business people. The increasing economic growth also resulted in the emergence of the era of free trade. Economic policies to increase the world economy always face the world economic system. When communism collapsed, the world tended to refer more to a free economy where there were no more conflicts between the market and the state or capitalism and socialism at an extreme point (Martadisastra, 2011). Free trade is recognized as a way to increase economic growth, because conceptually a country may not be able to meet its individual needs. In the free trade system or on the free market, new business opportunities can be created because, with the free trade system, regulations that previously limited the space for business actors can be minimized. However, apart from that, free trade also creates a new form of competition that is more open, therefore business actors are required to be better prepared in facing this new competition. Business actors, in this case, the company, are required to continue to improve their performance to maintain their existence in existing and emerging competition. Good company performance can not only play a role in winning the competition, but also increase the welfare of the owner.
In connection with efforts to improve company performance, it is necessary to understand the factors that can directly or indirectly affect company performance. An understanding of these factors will be useful later in the consideration of corporate strategy formulation. The company's performance, which can be said to be a benchmark for evaluating the success of management in running the company, is influenced by various factors. Various previous studies have examined a lot about these factors, but most of them only examined only one determinant. Discussions about company performance and the factors that influence it cannot be separated from the operational process of managing a company. In carrying out its business practices, companies are often faced with conflicts of interest between management as company managers and the owner. According to agency theory, this conflict of interest occurs because management, namely the agent often runs the company, not only based on the interests of the principal. Conflict and tug of war between principals and agents can cause problems which in agency theory are known as Asymmetric Information (AI) (Pudja, 2011). The management who acts as the manager of the company has greater information about the company than the owner. Management often uses the imbalance in the distribution of information to accommodate their interests rather than those of the owners. These problems will eventually lead to the emergence of agency costs. Agency costs in general can be said to be the costs incurred to keep the company's goals in line with the owner's goals. Besides, agency costs can also be said to be the number of costs incurred by the principal (owner) to supervise the agent (management) (Jensen and Meckling, 1976). Reality the possibility of eliminating agency costs is impossible, but good understanding can lead to the ability to reduce agency costs to the maximum. The inability to manage agency costs makes the business process inefficient. Agency costs that are too high can have an impact on the company's performance. On the one hand, agency costs can be used to maintain the main purpose of establishing a company, namely to improve the welfare of the owner, but on the other hand, it can also harm the company if it is not properly controlled.
Many factors also can affect agency costs. Understanding these factors with their influence and how they work on agency costs is important in managing agency costs. The purpose of this paper is that it can be used to reduce agency costs to make it as effective and efficient as possible. Previous research has found many factors that influence agency costs. Previous research has found that factors such as capital structure, firm size, and corporate governance have been shown to influence agency costs, but previous research has mostly been conducted separately so that it is difficult to determine differences in the magnitude of the effects. On the other hand, these factors also influence company performance. Therefore, this study tries to examine the factors that influence agency costs and link the effects either directly or through agency costs on company performance. This research is interested to study because no previous research has examined the factors that affect company performance either directly or through agency costs.
Research on the factors that affect agency costs and company performance has been

LITERATURE REVIEW
Agency theory was developed by Jensen and Meckling (1976). Agency theory is a theory related to the principal-agent relationship. This agency theory creates a model regarding a contractual relationship between the manager (agent) and the owner (principal). The principal delegates a decision-making responsibility to the manager (agent) by following under the work contract. The duties, powers, rights, and responsibilities of the agent and principal are regulated in a work contract mutually agreed upon. There are several possible conflicts in the relationship between the principal and the agent (agency conflict), conflicts that arise as a result of the desire of the management (agent) to take actions by following under their interests which can sacrifice the interests of shareholders (the principal) to obtain returns and long-term value of the company.. The existence of agency conflicts that occur will cause costs to be used to control conflict. These costs are known as agency costs or agency fees. According to Jensen and Meckling (1986), agency costs are costs borne by shareholders to prevent or minimize agency problems and maximize shareholder returns. This profit is the company's profit distributed in the form of dividends. According to Jensen and Meckling (1986) there are three categories of agency costs, namely (1) Expenditures to monitor manager's activities (2) Expenditures for an organizational structure which will limit undesirable manager behaviors (the bonding cost) (3) Residual Cost is the opportunity cost arising from conditions. Performance is a pattern of actions carried out to achieve measured goals based on a comparison with various standards. Performance is the achievement of a goal of a particular activity or job to achieve company goals as measured by standards. The company performance appraisal aims to determine the company's operational effectiveness. Measuring company performance can be done using a method or approach. Measurement of company performance grouped into two, namely non-financial performance measurement (non-financial performance measurement and financial performance measurement). In this study, ROA is used as a proxy for company performance. Return On Assets (ROA) is used to measure the financial performance of multinational companies, especially from the point of view of profitability and investment opportunities.ROA shows the effectiveness of the company in generating profits by optimizing its assets.
The capital structure model within the scope of Balancing Theories (Myers, 1984 and Bayles and Diltz, 1994) refers to as a balance theory, which is to balance the composition of debt and equity. In essence, this theory balances the benefits and sacrifices that arise as a result of the use of debt. As long as the value is still largely, debt added. But if the sacrifice due to using debt is greater than the debt is no longer be added. The tradeoffs for using the debt can be in the form of bankruptcy fees and agency costs. Capital structure can be measure from the ratio of total debt to equity, which is usually measured by the debt to equity ratio (DER). DER can show the level of risk of a company where the higher the DER ratio, the higher the risk of the company because funding from debt is more than its capital (equity).
The total assets owned by the company represent the capital, as well as the rights and obligations it has. The larger the company size, it can be ascertained that the larger the funds it will manage and the more complex the management will be. Large companies tend to get more attention from the wider community. Thus usually large companies have a tendency to always maintain the stability and condition of the company. To maintain this stability and condition, the company will of course try to maintain and continue to improve its performance.
Implementing GCG requires an accountable form of corporate governance mechanism. Corporate governance mechanisms are clear rules, procedures, and relationships between parties who make decisions and those who will control (supervise) these decisions will guarantee and be under the running of a governance system in an organization (Syakhroza, 2005). The indicator of corporate governance mechanisms used by the author in conducting this study is the size of the board of directors. The board of directors is fully responsible for all forms of operations and management of the company to carry out interests in achieving company goals. The board of directors is also responsible for the company's dealings with external parties such as suppliers, consumers, regulators, and legal parties. With such a large role in the management of this company, the directors have significant control rights in the management of company resources and funds from investors.

METHODS OF RESEARCH
This research is causal explanatory, that is, research that wants to find an explanation in the form of a cause-effect relationship between several variables developed in the study (Ferdinand, 2006). This research is an ex post facto research because the data used comes from the issuer's annual report that has been published and used without changing it. The sample of this research is manufacturing companies whose annual reports are published through the official website of the Indonesian stock exchange in 2018.
In analyzing the data from the study, two methods were used, namely descriptive statistics and path analysis using PLS (Partial Least Square). Descriptive statistics used to describe the real condition of the object through various parameters. Meanwhile, PLS is used to analyze the relationship between each independent, intervening, and dependent variable. Testing the relationship between variables is carried out in two stages, namely direct relationship testing as many as seven hypotheses and indirect testing (mediation) which tests three hypotheses.

RESULTS AND DISCUSSION
In analyzing the data in this study, two approaches were used, namely descriptive statistics and path analysis with Partial Least Square software. The data obtained came from the Indonesia Stock Exchange (www.idx.co.id) in 2018 for manufacturing companies that met the criteria for the research sample as many as 57 companies. Descriptive statistical details for the research sample are as follows: Hypothesis testing in this study uses path analysis with the PLS (Partial Least Square) tool. In this study, a significance level of 95% was used so that the probability of error in this study was 5%. The results of hypothesis testing were divided into 2, namely the direct effect and the indirect effect as shown in the figure and table below:  Hypothesis one testing shows the results that capital structure affects agency cost. The results of this study are in line with research conducted by Makhdalena (2015). Reducing agency cost in the company is done by increasing the use of debt in company funding, because debt requires the company to pay its obligations back, the free cash flow available to managers to take unnecessary actions is limited. This is because the more debt the company has, the greater the cash that must be used as a reserve to pay loan principal and interest so that debt in the capital structure of a company can be used as a mechanism for disciplining company managers who tend to overuse free cash flow. Thus debt can reduce agency costs on free cash flow by reducing the cash flow available for expenditure at the manager's discretion.
The second hypothesis testing shows the results that firm size does not affect agency cost. The results of this study are in line with research conducted by Makhdalena (2015). This is because the size of the companies in the sample of this study has large company sizes. The increasing size of the company in the number of company assets is not a determinant of the effectiveness of asset utilization. With a large company size and not in a condition of financial difficulty, companies will prefer not to be careful in making expenses. Besides expenses, which are a proxy for agency cost measurement, the bigger the company, the greater the agency cost that the company must incur. Also, the bigger the company, the agency problems it faces is bigger.
The results of testing the third hypothesis show that corporate governance does not affect agency cost. The results of this study are in line with research conducted by Kamyabi, Majbouri, and Ashae (2014) and research by Moez (2018) but not in line with research conducted by Nguyen, et., al (2020). The results of this study indicate that corporate governance which is proxied by the board of directors does not affect agency cost because the number of boards of directors in the sample companies of this study has a large enough number so that it increases agency cost because the board of directors cannot minimize agency problems in the company due to the lack of efficiency in a board of directors size.
The fourth hypothesis testing shows the results that capital structure affects company performance. The results of this study are in line with research conducted by Javed, et al (2014). Good or bad capital structure will have a direct effect on the company's financial position. With a good capital structure in the company, it can indirectly attract investors to invest in the company. Conversely, shareholders should know what capital structure can increase value for investors and the capital structure policies implemented by company management and how the capital structure impacts investors. The policy on the use of debt in the capital structure can be used to control the excessive use of free cash flow by management, in order to avoid wasted investment which in turn can increase the performance and value of the company. The results of this study are in line with the theory of Jones, et al (2009) which states that capital structure can affect company performance, this happens because debt financing causes fixed interest expenses to be paid.
The results of testing the fifth hypothesis state that there is no influence between company sizes on company performance. From the results of descriptive statistical analysis, it can be seen that the average value of the company size for the sample of this study is 28,466, which means that the companies that are the research samples are classified as large companies and the size of the company does not affect company performance. This shows that the size of the company is not a guarantee that the company will have a good performance. The results of this study are in line with the research conducted by Talebria et al. (2010).
The results of testing the sixth hypothesis indicate that good corporate governance (GCG) has no effect on company performance. This can be seen from the calculated T value of 1.827 which is smaller than the T table value of ± 1.96. Eisenberg et al (1998) stated that a small number of boards of directors are more effective in implementing company policies and increasing company sales which results in an increase in company profits which in turn can improve company performance. In this study, the number of boards of directors in the company as the sample of the study is quite large, which means that the number of people who control the operations of the company is also quite large and the information circulating in the company is also getting bigger. The eighth hypothesis testing shows that the capital structure affects firm performance through agency cost. The results of this study indicate that the use of debt in the company's capital structure reduces agency cost so that it has an impact on improving company performance. The results of this study are not in line with research conducted by Fakhrudin (2011). According to Jensen and Meckling (1976) reducing agency cost can be done by increasing debt because increasing debt will reduce waste that may be done by management. Crutchley and Jensen (1999) also state that to reduce agency costs, it can be done by increasing the use of debt in the financing, because debt requires companies to pay their obligations back, so the free cash flow available to managers to take unnecessary actions is limited. The use of debt in the capital structure can prevent unnecessary corporate expenditure and encourage managers to operate the company more efficiently. This causes agency cost, to decrease and company performance will increase (Cao, 2006). The results of this study indicate that debt funding is well managed to increase company profits.
The testing of the ninth hypothesis shows that the existence of agency cost does not mediate the relationship between firm size and performance. The direct test also shows that firm size does not affect firm performance and firm size does not affect agency cost. The results of this study indicate that the larger the company, the greater the agency cost incurred by the company. The results of this study are in line with research conducted by Fakhrudin (2011) which states that small companies can improve their performance through reducing agency costs. Conversely, if the size of a large company causes an increase in agency costs, the company's performance will decrease. Companies with large sizes are faced with a bigger agency problem so that they cannot minimize agency costs which can increase company profitability which will have an impact on improving company performance.
The tenth hypothesis testing shows that there is no indirect effect of corporate governance on company performance through agency cost. The direct test also shows that corporate governance has no effect on company performance and corporate governance also has no effect on agency cost. The results of this study indicate that the corporate governance mechanism proxied by the board of directors is unable to determine the effectiveness of monitoring activities in supervising managerial performance and cannot minimize agency problems that exist in the company so that it is unable to reduce agency costs in the company so that there is no increase in company performance. This is because the number of boards of directors in the companies sampled in this study is quite large. The effectiveness of the board as a mechanism for corporate governance depends on its number and composition. The larger the board size, the greater the number of people controlling the company's operations, which means that the information circulating in the company will be even greater. The results of this study are in line with Florackis and Ozkan (2009) who found that a high board size will lead to high agency costs as well. However, the results of this study are not in line with research conducted by Lin and Lin (2018).

CONCLUSION
Based on the results of the tests that have been carried out, it can be concluded that the capital structure has a positive and significant effect on agency cost and performance. This relationship can be seen from the results of direct (partial) or indirect (mediation) tests. However, the nature of the mediation relationship is partial mediation, meaning that agency cost does not fully mediate the relationship between capital structure and firm performance. Therefore, it can be said that from several factors that are thought to have an influence on company performance, only two variables have an influence, namely capital structure and agency cost. This means that if the company wants its performance to increase from time to time, the company must pay attention to the composition of its capital structure. In addition, the existence of agency cost must be considered by the company in order to improve company performance.