The Effect of Financial Leverage, Employee Stock Ownership Program and Firm Size on Firm Performance of Companies Listed in Indonesia Stock Exchange

________________________ Abstract ___________________________________________________________________ The purpose of this research is to examine and to obtain affected empirical evidence of financial leverage, firm size and employee stock ownership program (ESOP) to firm performance in manufacturing company in Indonesian Stock Exchange on 2013-2015. Independent variables in this research are Financial Leverage (DER), Firm Size and Employee Stock Ownership Program (ESOP). Dependent variables in this research are Return on Assets (ROA) and Return On Equity (ROE). The results Showed that the simultaneous test of three independent variables Significantly afftected to the ROA and ROE. The partial tests of Financial Leverage (proxy DER) and Firm Size Significantly affected to ROA and ROE. But, the results Showed that the Employee Stock Ownership Program (ESOP) did not Affect to ROA and


INTRODUCTION
A company can measure its financial condition both the success and failure of the company's performance using the tools of financial analysis using the Return On Asset (ROA) and Return On Equity (ROE). Return On Assets (ROA) is an important ratio that can be used to measure the ability of companies with investments made (its assets) at a profit (Hutomo, 2015). The size of the company (firm size) has a relationship with the company's financial structure. Where Trisusilowati (2006) in Mar'ati and Purnomo (2011), said on choosing financing ways, big companies whose shares are owned by a lot of people will choose additional common stock sales because these sales will not have much affect control of the company. Instead of small companies may prefer to avoid the issuance of common shares in its effort to keep control of the company entirely. stakeholders, value added is a measure that is more accurate in measuring return for stakeholders. As part of its responsibilities to stakeholders, management companies must be able to manage all the resources, both financial resources, as well as non-financial resources of the company in efforts to create added value for the company. If all the resources owned by the company can be managed and utilized properly it will create added value for the company, so as to generate greater profits and improve financial performance.

Agency Theory
Agency theory becomes a theory which has close links with the company's performance as well as the Employee Stock Ownership Program (ESOP) because, according to Anthony and Govindarajan (2007: 530-533) in Priansyah (2016), agency theory assumes that if the contract incentives such as bonuses, commissions, or stock options granted based on the size of the performance of the agent, then the agent will be more interested in improving the performance of an incentive to get more. Agency theory mentions the existence of agency cost which represent costs incurred by shareholders who entrust the company on the part of managers and employees to organize and run the company in order to maximize the return to the principals (Pugh, 2000 in Hartono andWibowo, 2014). Chanda and Shen (2009: 88) states that financial performance judged on a variety of financial ratio analysis. Financial ratio anaylsis can be used to analyze the existing financial statements and pro forma financial statements. Financial ratios can also help identify weaknesses and strengths of the company's financial perspective, as well as providing a way to do a comparison of financial data.

Firm Performance
The main objective of financial performance measurement is to determine the level of profitability. Measurement of profitability in this study is the ROA and ROE. ROA to determine the impact of firm size, employee stock ownership program and the use of financial leverage on assets (ROA), ROE to determine whether the company can increase the return on investors, since equity is one of the capital used by companies to get the company's resources. Weygandt, Kimmel and Kieso (2013: 700) states that ROA is an overall measure of profitability is return on assets. We compute this ratio by dividing net income by average assets. Weygandt, Kimmel and Kieso (2013: 700) states that ROE is an another widely used profitability ratio is return on ordinary shareholders' equity. It measures profitability from ordinary shareholders' viewpoint. This ratio shows how many euros of net income the company earned euro invested by the owners. We compute it by dividing net income available to ordinary shareholders 'by average ordinary shareholders' equity.

Financial Leverage
Financial leverage is a measure of how much the company uses capital and debt to finance its assets (Enekwe et al. 2014). Financial leverage can be used by companies to meet the funding needs of the company so that the company can operate, invest and develop their business. Financial leverage is expected to provide additional advantages greater for shareholders. It is based on a fixed amount of interest expense on the debt can reduce the amount of tax. However, financial leverage can also adversely affect the company because of the high financial leverage will cause financial difficulties because of its debt obligations. (Ningrum, 2014)

Debt-Equity Ratio (DER)
Leverage in this study proxied by the Debt Equity Ratio (DER). Selection of Debt Equity Ratio (DER) as a proxy based on the leverage ratio is used by investors to see how much debt the company when compared to equity held by the company or its shareholders.

Employee Stock Ownership Program (ESOP)
According to Bergstein & Williams (2013) in Haosana and Hatane (2015) ESOP is a unique financial tool for continuing business success by providing the employee stock ownership of the business. Employee Stock Ownership Plans is a company where the company contributes part of its own shares or cash to be used to buy shares to a trust (the trust) was established to buy part of the company's shares for employees. Stock options are granted directly to the individual employee to use as they are, if deemed fit, not into a pension trust (Dessler.G, 2007).

Hypothesis
Based on the description in the literature review above, hence writer formulate the research hypothesis as follows: H1a: Financial Leverage has a positive and significant effect on firm performance (ROA proxy).
H1b: Financial Leverage has a positive and significant effect on the firm performance (proxy ROE).
H2a: Firm Size has a positive and significant effect on firm performance (ROA proxy). H2b: Firm Size has a positive and significant effect on the firm performance (proxy ROE). H3a: Employee Stock Ownership Program (ESOP) has a positive and significant effect on firm performance (ROA proxy).
H3b: Employee Stock Ownership Program (ESOP) has a positive and significant effect on the firm performance (proxy ROE).

Research Model
This research was conducted using secondary data to observe and analyze the research object consisting of independent variables and the dependent variable. The independent variable is the Employee Stock Ownership Program, Firm Size, Financial Leverage proxied by the Debt-Equity Ratio (DER) Dependent variable was ROA (Return on Assets) and ROE (Return on Equity) memproksikan financial performance. Subjects used in this research is manufacturing companies listed in Indonesia Stock Exchange in 2013-2015. Data were analyzed with SPSS V.21. The research model can be described as follows:

Research's Object Selection
This study observed and analyzed the research object consisting of independent variables and the dependent variable. The independent variable is the Employee Stock Ownership Program, Firm Size, Financial Leverage proxied by the Debt-Equity Ratio (DER) Dependent variable was ROA (Return on Assets) and ROE (Return on Equity) representing firm performance. Subjects used in this research is manufacturing companies listed in Indonesia Stock Exchange in 2013-2015.

Sampling Method
The population in this study are all manufacturing companies listed in Indonesia Stock Exchange and publish the financial statements in the year 2013-2015. By using as many as 53 companies as research samples. Sample selection technique used in this research is purposive sampling method in which the methods used take samples drawn from the population must meet the criteria established researchers. Criteria established by the researchers is: a. Manufacturing companies listed in Indonesia Stock Exchange period 2013 -2015; b. Companies that publishes financial reports per period audited use Rupiah (IDR) as well as complete for use as research information; c. Manufacturing company which has several business segments within the company.

Variable Operationalization
The independent variables in this study are the Firm Size, Employee Stock Ownership Program and Financial Leverage, while the dependent variable is the Return on Assets and Return on Equity representing firm performance. Plan the company makes to contribute part of its own shares or cash to be used to buy shares to a trust which established to buy part of the company's shares for employees.

Firm Size
Firm size is a large-scale to small companies classified according to a variety of ways, including: total assets, log size, the stock market value, and others (Isbanah, 2015). In this study, the variable firm size measured by the natural logarithm of total assets n. Log (Total Assets)

DER
Describing how much debt compared to equity held by the company DER= ROE Demonstrates the success of management in maximizing returns to shareholders ROE= ROA Measuring the return on optimizing the use of company assets

Data Collection Technique
Data collection techniques are methods used to obtain the required data in a study. Data collection techniques in this study conducted by literature research and field research. The research literature is useful to get a literature review that will be used as a guide and a guide in conducting research and making discussion of the results will be more systematic. With a library research found many theoretical and empirical studies that have been done before to support this research. Field research is useful for gathering data related to this study, field research carried out by collecting the annual financial statements of various companies that have been audited by a public accountant and has been published during the study period. How to obtain secondary data is through www.idx.co.id and on the official web site of the company.

Data Processing Techniques
The data obtained will be processed and analyzed using the methods of electronic data processing, using SPSS Software V.21. Data processing techniques used in this research is multiple regression analysis techniques which states that allegedly independent variables (firm size, employee stock ownership program and financial leverage) effect on the dependent variable (firm performance).

Analysis Method
Statistical test equipment used to test the hypothesis is multiple linear regression, to test whether the independent variables affect the dependent variable with a significant level of 5% (α = 0.05). Models of multiple linear equations used are as follows: Hypothesis testing using multiple linear regression analysis model: In Table 3 the data processing is done using the proxy Return On Asset (ROA) & Return on Equity (ROE) representing firm performance as the dependent variable, and use the 53 companies with the 3-year period running from 2013 to 2015 year, which resulted in 159 samples as the study sample (N). Shown in the table above that the ROA has a minimum value of 0.0004, whereas the maximum value of 0.2615, the value of the average (mean) of 0.076321, indicating that the average company has a high return assets amounted to 0.076321, as well as producing standard deviation of 0.0509706. Shown in Table 3, that the ROE has a minimum value of 0.0006, whereas the maximum value of 0.3691, the value of the average (mean) of 0.125083, indicating that the average company has a return on equity of 0, 125 083, and has a standard deviation of 0.0509706.

Multicollinearity Test
Multicollinearity test can be seen from (1) the value of tolerance and the opponent (2) variance inflation factor (VIF). Cutoff value that is commonly used to indicate the multicollinearity presence. Tolerance is a value ≤ 0.10 or equal to VIF ≥10 (Ghozali, 2016: 103). With the results obtained from the tolerance test chart below multicollinearity test showed more than 0.10, it can be concluded not happen multicollinearity. VIF for the results obtained from testing multicollinearity test figures show over 10, it can be concluded that no multicollinearity. Thus all three variables used in this study as a linear regression model that is financial leverage, firm size and Employee Stock Ownership Program (ESOP) has been qualified to predict the Return on Assets (ROA) and Return on Equity (ROE) Based on the tables 4 and 5, with the tolerance results obtained from testing multicollinearity that showed more than 0.10, it can be concluded that the relationship has no multicollinearity. VIF for the results obtained from testing multicollinearity test figures show over 10, it can be concluded that the relationship has no multicollinearity. Thus all three variables used in this study as a linear regression model that is financial leverage, firm size and Employee Stock Ownership Program (ESOP) has been qualified to predict the Return on Assets (ROA) and Return On Equity (ROE).

Normality Test
Normality test is done to determine how the normal distribution of data. The significance of the independent variables on the dependent variable through t test can only be applied when the residuals have a normal distribution. Normality test is done with the Kolmogorov-Smirnov test. Kolmogorov-Smirnov test can be performed to test whether the residuals are normally distributed. Data will be said to be normally distributed if the value of the Kolmogorov-Smirnov and significantly higher than α (0.05) (Ghozali, 2016: 169-170 If the level of significance of the resulting data is smaller than 0.05 means between the dependent and independent variables were not normally distributed. Whereas, if the level of significance of the resulting data is greater than 0.05 means that the normal distribution occurs between the dependent and independent variables. Results of testing normality test presented in tables 6 and 7 as follows. ,582 a. Test distribution is Normal. b. Calculated from data. From the table above, test One-Sample Kolmogorov-Smirnov Test for Return On Asset (ROA) with sample (N) as many as 159 samples, generating significant value of 0.582 with 0.777 probability far above α = 0.05, it can be concluded that meet the assumptions of normality test and the data were normally distributed.

Autocorrelation Test
Autocorrelation test showed regression residual properties which are not free from one observation to another observation (Ariefianto, 2012b: 26-27). To determine whether there is autocorrelation can test Durbin Watson (DW). DW test using statistical test d, by lowering the lower limit of the critical value (dL) and the upper limit (dU). Statistical value of this test ranged from 0 through 4. d value of 2.00 is when there are no autocorrelation between residuals. At the time d close to 0, it indicates positive autocorrelation. At the time of d approaching 4, it shows a negative autocorrelation (Widarjono, 2010c: 99). Lower limit value (dL) and the upper limit (dU) depending on the number of variables and the number of observations used in the study were obtained from statistical tables Durbin Watson. Results of testing autokolerasi test presented in Table 8 and 9 to explain the results of the test as follows.  Table 9 above, using Return On Equity (ROE) as the dependent variable, it can be concluded that there is no autocorrelation between the study variables and have fulfilled classical assumption test.

Heteroskedastisitas test
According to Hutomo (2015) heteroskedastisitas test is a state where the variance and the confounding errors are not constant for all the independent variables. Heteroskedastisitas test used in the study is the Glejser test. Heteroskedastisitas test results influence financial leverage, firm size and Employee Stock Ownership Program (ESOP) of the firm manufacturing company performance period 2013 -2015: Based on the output of the heteroskedastisitas test results by using glejser test which can be seen from Table 10 and 11, it can be concluded that the data are free from heteroskedastisitas because the significant value generated financial leverage amounted to 0.050 on ROA and 0.409 on ROE, firm size for 0730 at the ROA and 0.246 on ROE, and employee stock ownership program for 0938 on ROA and ROE 0.621 in. From these results, there is not any one variable that is below 0.05, then it can be concluded that there is no heteroskedastisitas symptoms.

Hypothesis Testing
Researchers used three independent variables in the study, it must be done in a multiple regression analysis to test the hypothesis. Regression model was made in order to predict the changes in the value of the variable performance of the company by proxy Return On Asset (ROA) and Return On Equity (ROE), which is influenced by changes in the value of the variable leverage, firm size, and Employee Stock Ownership Program (ESOP) with a significance level of 0.05 or 5%.
The regression equation above shows that the constant value of -0.047. This means that if the value of the variable LEV, FS, and ESOP considered zero or constant, then the variable ROA has a value of -0.047. Variable LEV (leverage) has a coefficient of -0.044 which shows that the FS and the ESOP when variables held constant, then any increase in LEV (leverage) by 1 unit will lower ROA amounted to -0.045. The coefficient is negative means increased LEV will cause a decrease in ROA.
Variable FS (Firm Size) has a coefficient of 0.025 which shows that when variables LEV and ESOP considered constant, then any increase in SD (Diversification Strategy) by 1 unit will increase ROA by 0.025. The coefficient is positive means increased FS will lead to an increase in ROA.
Variable ESOP (Employee Stock Ownership Program) has a coefficient value of -0,015yang shows that when variables FS and ESOP considered constant, then any increase in ESOP by 1 unit will lower ROA amounted to -0.011. The coefficient is negative means increased ESOP will cause a decrease in ROA. According to the table above, the obtained results of regression analysis testing with the form of regression model on the dependent variable ROE as follows: ROE = -0.095 -0,023LEV + 0,037FS -0,009ESOP

Regression models for Return On Equity (ROE) used in this study
The regression equation above shows that the constant value of -0.095. This means that if the value of the variable LEV, FS, and ESOP considered zero or constant, then the ROE have a value of -0.095. Variable LEV (leverage) has a coefficient of -0.023 which shows that the FS and the ESOP when variables held constant, then any increase in LEV (leverage) by 1 unit will lower ROE of -0.023. The coefficient is negative means increased LEV will cause a decrease in ROE.
Variable FS (Firm Size) has a coefficient of 0.037 which shows that when variables LEV and ESOP considered constant, then any increase in firm size of one unit will increase the ROA amounted to 0,037. The coefficient is negative means that the increase in SD will cause an increase in ROE.
Variable ESOP (Employee Stock Ownership Program) has a coefficient of -0.0009 indicating that FS and LEV when variables held constant, then any increase in ESOP by 1 unit will lower ROE of -0.0009. The coefficient is negative means increased ESOP will cause a decrease in ROE.

Test F (Simultaneous)
Test F can be regarded as ANOVA. The statistical test F basically indicates whether all the independent variables or free inclusion in the model jointly have influence on the dependent variable / dependent (Ghozali, 2011: 98 F test is performed to detect the influence of the independent variables and the dependent variable the researchers used. ROA and the dependent variable is the independent variable LEV, FS and ESOP. In accordance with the results of table 14 it can be seen that the significant value of 0.000 that is smaller than α = 0.05, it can be concluded that simultaneously, there is a significant effect. This shows that there is significant influence between leverage (by proxy of debt to equity ratio), Firm Size and Employee Stock Ownership Program (ESOP) as independent variables on ROA as the dependent variable together with a confidence level of 95%. So we can conclude that the variable LEV, FS and ESOP significantly affect ROA. F test is performed to detect the influence of the independent variables and the dependent variable the researchers used. Independent variables and the dependent variable of ROE LEV, FS and ESOP. In accordance with the results of table 4:14 to note that the significant value of 0.000 that is smaller than α = 0.05, it can be concluded that there is significant influence between leverage (by proxy of debt to equity ratio), Firm Size and Employee Stock Ownership Program (ESOP ) as the independent variable on the dependent variable ROE together with a confidence level of 95%. So we can conclude that the variable LEV, FS and ESOP significantly affect ROE.

t test
The statistical test t basically shows how far the influence of the explanatory variables / independent individually in explaining the dependent variable (Ghozali, 2011: 98). According to Manurung (2006b: 171), the destination t test to determine whether the regression coefficient is significant or not. In this study, the test was conducted to determine how much influence the leverage (by proxy of debt to equity ratio), firm size and Employee Stock Ownership Program (ESOP) in explaining the variation of the dependent variable ROA and ROE by using the degree of error (α) 5% and a confidence level of 95%. The test results of independent variables on the dependent variable are presented in tables 16 and 17 Variable LEV (Leverage) has a significance value of 0.000, which means H1a accepted because the significant value of the variable LEV has a value less than 0.05. Thus it can be concluded that the variable LEV has significant effect on ROA firms as sample.
As for the variable FS (Firm Size) has a significance value of 0.000, which means H2a accepted because the significant value of the variable FS has a value less than 0.05. It can be concluded that FS variable has significant effect on ROA firms as sample.
Furthermore, analysis of the ESOP variable ROA has a significance value of 0.304, which means H3a rejected because of the significant value of the ESOP variable has a value greater than 0.05. It can be concluded that the ESOP variable has no significant effect on ROA firms as sample. Variable LEV (Leverage) has a significance value of 0.019, which means H1b accepted because the significant value of the variable LEV has a value less than 0.05. Thus it can be concluded that the variable LEV has significant effect on ROE firms as sample.
As for the variable FS (Firm Size) has a significance value of 0.000, which means H2b accepted because the significant value of the variable FS has a value less than 0.05. It can be concluded that FS variable has significant effect on ROE firms as sample.
Furthermore, analysis of the ESOP variable ROE has a significance value of 0.613, which means H3b rejected because of the significant value of the ESOP variable has a value greater than 0.05. It can be concluded that the ESOP variable has no significant effect on ROE firms as sample.

Coefficient of Determination Regression Test
According to Endang NP, Topowijono, and Vidyanata (2016), the coefficient of determination (R2) is useful to know how big the ability of the independent variables in explaining the dependent variable. If the coefficient of determination closer to the figure, the better the effect of inter-dependent and independent variables in the study. Conversely, if the coefficient of determination is getting close to zero, the smaller the influence of independent variables on the dependent variable. Adjusted R2 test results are presented in tables 18 and 19 as follows

CONCLUSION
The purpose of this study is to demonstrate empirically the effect of financial leverage, firm size and employee stock ownership program (ESOP) to the firm performance of companies listed on the Indonesian Stock Exchange (BEI) in the period from 2013 to 2015 either simultaneously or partial. The sample used in this study is as much as 53 or as many as 159 manufacturing companies sampled data and more data is processed with the help of SPSS Software V.21 Based on test results and discussion as has been presented in previous section, could be concluded as follows: 1. Variable LEV (Leverage) has a significance value of 0.000, which means H1a accepted because the significant value of the variable LEV has a value less than 0.05. Thus it can be concluded that the variable LEV has significant effect on ROA firms as sample. 2. Variable LEV (Leverage) has a significance value of 0.019, which means H1b accepted because the significant value of the variable LEV has a value less than 0.05. Thus it can be concluded that the variable LEV has significant effect on ROE firms as sample. 3. Variable FS (Firm Size) has a significance value of 0.000, which means H2a accepted because the significant value of the variable FS has a value less than 0.05. It can be concluded that FS variable has significant effect on ROA firms as sample. 4. As for the variable FS (Firm Size) has a significance value of 0.000, which means H2b accepted because the significant value of the variable FS has a value less than 0.05. It can be concluded that FS variable has significant effect on ROE firms as sample. 5. Furthermore, the analysis of the ESOP variable ROA has a significance value of 0.304, which means H3a rejected because of the significant value of the ESOP variable has a value greater than 0.05. It can be concluded that the ESOP variable has no significant effect on ROA firms as sample. 6. Furthermore, the analysis of the ESOP variable ROE has a significance value of 0.613, which means H3b rejected because of the significant value of the ESOP variable has a value greater than 0.05. It can be concluded that the ESOP variable has no significant effect on ROE firms as sample.