The Effect of Collaterallizable Assets, Growth in Net Assets, Liquidity, Leverage and Profitability on Dividend Policy (Case Studies on Non-Financial Services Sector Companies Listed on the Indonesia Stock Exchange for the 2016-2019 Period)

In determining which company to invest in, investors tend to choose companies that have good corporate value which is reflected in the company's dividends distributed to shareholders. Dividends are part of the company's profits that will be distributed to shareholders (Sutrisno, 2017: 5)[21]. Dividend policy is one of the company's financial decisions that has influence between the company and investors, creditors, and various other company stakeholders. According to Hanafi & Halim (2016: 253)[10], dividend Abstract

policy is a decision related to the profits obtained by the company at the end of the year to be distributed to shareholders in the form of dividends or to be stored as retained earnings to be reused to finance company expansion and investment in the future.
In practice, companies have difficulty determining and deciding dividend policies. According to Mardiyati et al., (2014) [15] , companies tend to choose to hold profits to be used as company operational development or to be reinvested. On the other hand, investors want to get large and stable dividends. The existence of differences in interests that occur in dividend payout policies is called agency theory proposed by Jensen and Meckling (1976) [11] . Dividend policy is proxied by the Dividend Payout Ratio (DPR). DPR is a ratio that shows the ratio between dividends per share and earnings per share (Gitman & Joehnk, 2016: 330) [9] . The problem that occurs in non-financial service sector companies listed on the Indonesia Stock Exchange (IDX) is their inconsistency in paying cash dividends to shareholders. Based on the findings of non-financial service sector companies listed on the IDX for the 2016-2019 period, from 220 companies used as research objects according to the criteria. In 2016, there were 85 companies that distributed cash dividends. In 2017, there were 81 companies that distributed cash dividends. In 2018, there were 80 companies that distributed cash dividends. In 2019, only 62 companies distributed cash dividends. There are factors that are thought to influence or may influence dividend policy, such as the research results of Wahjudi (2019) [22] , Lestari et.al., (2019) [14] , and Ullah et.al., (2019) [22] explains that dividend policy can be determined through collateralizable assets, growth in net assets, liquidity, leverage and profitability.
In this study, collateralizable assets are measured using the COLLAS ratio to measure the amount of assets owned by a company that can be used as collateral for debt by the company. Growth in net assets in this study is measured using the Assets Growth ratio to determine the level of asset growth. Companies with high growth rates will also need funds in the future to finance their growth which will have an impact on dividend payments because companies tend to pay dividends in the form of retained earnings. Ratnasari & Purnawati (2019) [18] . Liquidity in this study is measured using the Current Ratio (CR) ratio. The CR ratio shows the company's ability to cover current liabilities with company assets that are expected to be converted into cash in the near future Brigham & Houston (2018: 128) [4] . Leverage is the ratio used to measure how much the company's assets are financed with loan funds (debt). In this study, leverage is measured by the ratio of Debt to Equity Ratio (DER), which is a ratio that measures the amount of debt used by the company Gitman & Joehnk (2016: 323) [9] . Profitability is a ratio to assess the company's ability to generate expected profit and measure the management effectiveness of a company which is reflected in sales activities or investment income. In this study, profitability is measured by Return On Equity (ROE) or return on equity. Companies that have good cash flow or profitability are able to pay dividends and can even increase the value of dividends distributed (Deni et.al., 2016) [6] . The purpose of this study is to determine the effect of collateralizable assets (COLLAS), growth in net assets, liquidity (CR), leverage (DER) and profitability (ROE) on dividend policy (DPR) in non-financial service sector companies listed on the Indonesia Stock Exchange. 2016-2019 period.

II. Review of Literatures
Basic Theory and Methodology 2.1 Basis of Theory a. Dividend Policy Dividend policy is a decision related to the profits obtained by the company at the end of the year to be distributed to shareholders in the form of dividends or to be retained as retained earnings to be used again as investment financing in the future Hanafi & Halim (2016: 253) [10] . The greater the dividends distributed to shareholders, reflecting that the company has good prospects for the future. So investors will increasingly trust the company and potential investors are increasingly interested in investing their capital because the company shows good financial management conditions (Monika & Sudjarni, 2018) [16] . Dividend policy is proxied by the Dividend Payout Ratio (DPR). DPR is a ratio that shows the comparison between dividends per share and earnings per share (Gitman & Joehnk, 2016: 330) [9] .

b. Collateralizable Assets
Collateralizable assets are the company's fixed assets that are used as collateral for loan applications to creditors (Ross et al., 2015) [20] . Collateralizable assets make creditors more secure because the larger the collateralizable assets shows the greater the proportion of the company's fixed assets from the total assets owned by the company that can be used for debt collateral and creditors do not need to limit the company's dividend policy so that the company can pay larger dividends to shareholders (Muslih & Husin, 2019) [17] . The larger the assets that are used as collateral, the less agency conflicts between shareholders and creditors so that the company can freely distribute large dividends. According to (Showalter, 1999) in research (Wahjudi, 2019) [23] collateralizable assets can be calculated using the following formula.

c. Growth in Net Assets
Company growth can be calculated by Asset Growth, which is proxied by the growth in net assets. Assets Growth is a ratio that shows the growth of assets, where assets are assets used for the company's operational activities (Riyanto, 2013: 267) [19] . Companies with high growth rates will also have higher funding needs in the future to finance their growth. A high growth rate will be able to influence the company in distributing dividends to shareholders because companies tend to set dividends as retained earnings and are allocated for corporate financing (Wahjudi, 2019) [23] . Based on research according to (Wahjudi, 2019) [23] , company growth can be seen by looking at the current number of assets with previous assets.

d. Liquidity
Liquidity is a ratio to measure the ability of a company, both external and internal, to pay short-term obligations or debts that are due (Kasmir, 2017: 110) [12] . The liquidity ratio shows the relationship between the company's cash and current assets and the company's current liabilities. Based on research (Wahjudi, 2019) [23] liquidity is proxied by using the current ratio. According to Brigham & Houston (2018: 128) [4] , the current ratio shows the company's ability to cover current liabilities with company assets that are expected to be converted into cash in the near future. The current ratio is calculated by dividing current assets by current liabilities.

e. Leverage
Leverage shows the extent to which a company fulfills all its obligations through debt. The higher the leveragemaka, the higher the risk faced by the company and the higher the expected rate of return (Angelia and Toni, 2020) [24] . Leverage is a ratio used to measure how much company assets are financed with loan funds (debt) (Kasmir, 2018: 151) [12] . The ratio used is the Debt to Equity Ratio (DER), which is a ratio that measures the amount of debt used by the company. The higher the use of financing (debt) used, the greater the risk of the company. Companies with high leverage (using large amounts of loan funds) will experience the risk of default on the loan (Gitman & Joehnk, 2016: 323) [9] .
High debt will make the company more likely to choose to hold its profits which will be used to pay off debt. So that companies with high levels of debt will pay dividends in small amounts. This will have an impact on the perception of investors that they will not be interested in buying company shares because the company has a high risk and has the potential for low dividend payments (Monika & Sudjarni, 2018) [16] . The formula for calculating leverage according to Gitman & Joehnk (2016: 323) [9] is as follows:

f. Profitability
Profitability according to Kasmir (2017: 114) [12] is a ratio to assess the company's ability to generate the expected profit or profit and measure the management effectiveness of a company that is reflected in sales activities or investment income. The higher the profitability of a company, the greater the company's ability to generate profits. A high profit rate will also affect the high dividend distribution value (Sutrisno, 2017: 213) [21] .
The calculation of the profitability ratio in this study is proxied by the Return on Equity (ROE) or return on equity. This ratio shows the level of efficiency in the use of its own capital, namely how the company can financially manage the resulting profit. ROE measures the return on investment of company shareholders by linking Net Profit After Tax with own capital (Brigham & Houston, 2018: 141) [4] .

Framework a. Effect of Collateralizable Assets on Dividend Policy
Collateralizable assets are collateralized assets, so creditors do not need to limit the company in distributing dividends to shareholders. The higher the collateralizable assets value, the company can use loan funds freely for company development, while maintaining the company's reputation because it continues to pay dividends consistently, even the company can provide large dividends without being limited by creditors.

b. Effect of Growth in Net Assets on Dividend Policy
Growth in net assets shows the growth in company's assets, the higher the growth in net assets, the funds needed are proportional to the growth. This means that companies wishing to expand require costs derived from company profits, thereby reducing the share of shareholder dividends. Thus it can be concluded that Growth In Net Assets has a negative effect on dividend distribution because the rate to finance company growth (growth) uses the company's retained earnings, which means the smaller the dividends paid.

c. Effect of Liquidity on Dividend Policy
The liquidity value in a company shows the company's ability to meet the obligations of internal and external parties. The greater the level of liquidity of a company, the greater the ability to distribute dividends to be distributed to shareholders. Thus it can be concluded that liquidity has a positive effect on dividend distribution because the higher the level of liquidity, the higher the dividends to be distributed. So liquidity has a positive influence on dividend policy

d. Effect of Leverage on Dividend Policy
Leverage shows the amount of company assets that are financed by debt. So that a high level of leverage means that the risk faced by the company is also higher because of the high level of debt. Companies with high levels of debt will be weak in distributing dividends to companies because companies tend to use their profits to pay debts. Thus it can be concluded that Leverage has a negative effect on dividend distribution because a company that has a high level of leverage means that the company has high debt.

e. Effect of Profitability on Dividend Policy
Profitability reflects the company's ability to manage business processes so that the company is able to generate maximum profits. Companies that have high profits will be able to distribute dividends and even dividends that will be received by shareholders are proportional to the profits generated by the company. The distribution of large dividends will increase the company's good reputation and good potential for the future. Therefore, it can be concluded that profitability has a positive effect on dividend distribution because the higher the level of profit generated means the higher the dividends to be distributed.

Research Hypothesis
Based on the theory and framework that has been described, the hypothesis in the study is as follows: 1. Collateralizable assets, growth in net assets, liquidity, leverage and profitability simultaneously have a significant effect on dividend policy. 2. Collateralizable assets partially have a significant effect on dividend policy. 3. Growth in net assets partially has a significant effect on dividend policy. 4. Partially liquidity has a significant effect on dividend policy. 5. Partially Leverage has a significant effect on dividend policy. 6. Partially profitability has a significant effect on dividend policy.

Population and Sampling
The population in this study are non-financial service companies listed on the Indonesia Stock Exchange for the period 2016-2019. The population in this study were 220 companies. This study used purposive sampling technique and produced 44 samples with a study period of 4 years, so the amount of data in the study was 176 data.

Analysis Technique
The analysis technique used in this research is descriptive statistical analysis, panel data regression analysis, and hypothesis testing. Panel data is a combination of time series data with cross sections.

Descriptive Statistical Analysis
Descriptive statistical analysis can describe and explain the data description of each variable used in the study. The description of the data obtained includes the average value (mean), amount of data, minimum value, maximum value, and standard deviation of each variable. The following are the results of the descriptive statistical test of this study:

Classic Assumption Test
In this study, the classical assumption test was carried out using the generalized least square (GLS) method approach, this method did not need to perform heteroscedasticity and autocorrelation tests because it produced an estimator to meet the best linear unbiased estimation (BLUE), which is a treatment method to overcome homoscedastic assumption violations and cross-sectional correlation (Basuki, 2015: 58) [2] .

a. Normality Test
The normality test in the regression model is used to test whether the residual value is normally distributed or not. According to Ghozali (2016: 165) [8] , testing for residual normality uses the Jarque -Bera (JB) test. The JB test is a normality test for large (asymptotic) samples.

Figure 1. The residual normality test results (Source: Data processed by the author, Eviews 11, 2021)
Based on the results of the residual normality test using the Jarque-Bera (JB) test, it can be seen that the data is normally distributed, where the probability value is 0.3998> 0.05. Based on the results of the Eviews 11 output in this study, the correlation coefficient value is smaller than 0.9 which means that there is no multicollinearity between the independent (free) variables or it can be said that there is no relationship between the independent (free) variables.

Selection of Panel Data Regression Estimation Method a. Fixed Effect Test (Chow Test) Table 4. The results of the the model specification test
Based on the results of the model specification test using the Chow test, it can be seen that the p-value of cross section F is 0.0002. This value is below 0.05, meaning that H0 is rejected and Ha is accepted. Based on these data, it can be concluded that the fixed effect model is better used than the common effect model. After the fixed effects model is selected, it is necessary to carry out the Hausman test next. Table 5. The results of the model specification

b. Random Effect Test (Hausman Test)
Based on the results of the model specification test using the Hausman test, it can be seen that the p-value of the random cross section is 0.0644. This value is above 0.05, meaning that H0 is accepted and Ha is rejected. Based on these data it can be concluded that the random effect model is better used than the fixed effect model. Table 6. The results of the Lagrange multiplier test

c. Lagrange Multiplier Test
Based on the results of the Lagrange Multiplier test, it can be seen that the Breusch Pagan value is 0.0033. This value is below 0.05, meaning that H0 is rejected and Ha is accepted. Based on these data, it can be concluded that the random effect model is better to use than the common effect model. Collateralizable Asset by one unit and the other variables are constant, the value of the dependent variable, the value of Dividend Policy (DPR), will increase by 0.369048.  The GROWTH coefficient value is -0.007802, meaning that if there is an increase in Growth in Net Assets by one unit and the other variables are constant, the dependent variable Dividend Policy (DPR) will experience a decrease of -0.007802.  The CR coefficient value is -0.004177, meaning that if there is an increase in Liquidity (CR) by one unit and the other variables are constant, the dependent variable Dividend Policy (DPR) will decrease by -0.004177  The DER coefficient value is -0.002337, meaning that if there is an increase in Leverage (DER) by one unit and the other variables are constant, the dependent variable Dividend Policy (DPR) will decrease by -0.002337.  The ROE coefficient value is 1.032629, meaning that if there is an increase in Profitability (ROE) by one unit and the other variables are constant, the dependent variable Dividend Policy (DPR) will increase by 1.032629.

Hypothesis Testing a. Simultaneous Test (Test F)
Based on the test results, it can be seen that the prob (F-statistic) value is 0.000000, meaning that the value is smaller than the 0.05 significance level. So it can be concluded that Collateralizable Assets, Growth in Net Assets, Liquidity (CR), Leverage (DER) and Profitability (ROE) simultaneously have a significant effect on dividend policy (DPR).

b. Analysis of the Coefficient of Determination (R2)
Based on the test results, it can be seen that the adjusted R-squared value of the research model shows a result of 0.297438 or 29.7438%. Therefore, the independent variables consisting of Collateralizable Assets (COLLAS), Growth in Net Assets, Liquidity (CR), Leverage (DER) and Profitability (ROE) can explain the dependent variable dividend policy (DPR) of 0.297438 or 29.7438%. , while the rest is 0.702562 or 70.2562% explained by other variables outside the research.

Discussion of Research Results a. Effect of Collateralizable Assets on Dividend Policy
Based on the test results of the Random Effect Model, it was found that the collateralizable assets (COLLAS) variable has a probability value (p-value) of 0.0000 which is smaller than the 0.05 significance level and has a positive direction with a coefficient value of 0.369048. This means that H02 is rejected, meaning that partially collateralizable assets (COLLAS) has a significant effect in a positive direction on dividend policy.
The results of this study are in accordance with the framework developed by the researcher that collateralizable assets have a positive effect on dividend distribution because the higher the collateralizable assets value of a company will reduce agency conflicts, limiting dividend distribution by creditors and the company can freely distribute high dividends to holders stock. The results of this study are in line with research according to Darmayanti & Mustanda (2016) [5] , that collateralizable assets have a positive and significant effect on dividend policy.

b. The Effect of Growth In Net Assets on Dividend Policy
Based on the test results of the Random Effect Model, it was found that the growth in net assets variable has a probability value (p-value) of 0.7850 which is greater than the significance level of 0.05 and has a negative direction with a coefficient value of -0.0078. This means that H03 is accepted, meaning that partially growth in net assets has no significant effect with a negative direction on dividend policy. The results of this study are not in accordance with the author's initial hypothesis, but are in line with the results of research conducted by Finingsih et.al., (2018) [7] who stated that growth in net assets does not have a significant effect on dividend policy.

c. The Effect of Liquidity on Dividend Policy
Based on the test results of the Random Effect Model, it was found that the Liquidity variable has a probability value (p-value) of 0.7617 which means it is greater than the significance level of 0.05 and has a negative direction with a coefficient value of -0.004177. This means that Ha4 is rejected, meaning that partially liquidity does not have a significant effect in a negative direction on dividend policy. In this study, the test results show that the size of the liquidity does not affect the company's dividend policy. The results of this study are not in accordance with the framework developed by researchers that liquidity has a positive effect on dividend policy, but it is in line with the results of research conducted by Lestari et.al., (2019) [13] which revealed that liquidity has no significant effect on policy dividends in a negative direction.

d. The Effect of Leverage on Dividend Policy
Based on the test results of the Random Effect Model, it was found that the Leverage variable has a probability value (p-value) of 0.8935 which means it is greater than the significance level of 0.05 and has a negative direction with a coefficient value of -0.002337. This means that Ha5 is rejected, meaning that partially leverage does not have a significant effect in a negative direction on dividend policy. Both companies that have leverage levels that are above average and those that have leverage levels that are below average, companies still pay dividends. Based on the signaling hypothesis theory, shareholders use dividends as a signal to predict the condition of the company in the future. Therefore, even though the company's leverage (DER) is high, the company will try to maintain cash dividend payments to shareholders with the aim that the company is considered to have good prospects in the future, so that shareholders will continue to invest. The results of this study are not in accordance with the author's initial hypothesis, but are in line with the results of research conducted by Anggara & Dwirandra (2020) [1] which revealed that leverage has no effect on dividend policy.

e. The Effect of Profitability on Dividend Policy
Based on the test results of the Random Effect Model, it was found that the Profitability variable has a probability value (p-value) of 0.0000 which is smaller than the significance level of 0.05 and has a positive direction with a coefficient value of 1.032629. This means that Ha5 is accepted, meaning that profitability partially has a significant effect in a positive direction on dividend policy. The company's profitability ratio shows the company's ability to generate profits that will be in line with the distribution of dividends to shareholders. Companies that have a high profitability value will be able to distribute dividends and even have the potential to distribute large dividends because they are in accordance with the company's ability to generate profits. The results of this study are in accordance with the framework developed by the researcher that profitability has a positive effect on dividend policy because the profitability ratio shows that the higher the profitability or ability of the company to generate profits, the company can also increase dividend payments. The results of this study are in line with research according to Monika & Sudjarni (2018) [16] stated that profitability has a positive and significant effect on dividend policy.

V. Conclusion
The results of this study concluded that: 1. Simultaneously the Collateralizable Assets, Growth in Net Assets, Liquidity, Leverage and Profitability variables have a significant effect on Dividend Policy in non-financial service companies listed on the Indonesia Stock Exchange for the period 2016-2019. 2. Partially it can be concluded as follows: a. Collateralizable assets have a significant positive effect on dividend policy b. Growth in Net Assets has no significant effect in a negative direction on Dividend Policy c. Liquidity does not have a significant negative effect on dividend policy d. Leverage has no significant effect in a negative direction on dividend policy e. Profitability has a significant effect in a positive direction on Dividend Policy

Suggestions a. For Companies
In this study, the results show that Collateralizable Assets and Profitability as proxied by Return on Equity (ROE) have a positive effect on Dividend Policy. The author suggests that companies can manage assets that will be used as collateral for debt properly in order to reduce conflicts of interest with creditors so that the company can pay large dividends. This will give a signal to investors and logically investors will choose companies that pay dividends regularly. Companies must maintain the stability of dividend payments by paying attention to the profitability ratio to give positive signals to investors that the company has good prospects for investment and has a good performance in generating profits.

b. For Investors
In deciding to invest, investors should consider the collateralizable assets ratio and the company's profitability because these two factors have an influence on the company's dividend policy. Companies that have high collateralizable assets and profitability indicate a high dividend payout ratio so that they can be taken into consideration and an overview of the company's prospects for the future.