THE EFFECT OF EARNING PER SHARE, INVESTMENT OPPORTUNITY SET, TOTAL ASSET TURNOVER, AND COLLATERALIZABLE ASSET ON DIVIDEND POLICY (Research on Consumer Goods Industrial Sector Companies Listed on the Indonesian Stock Exchange for the 2015-2019 period)

Dividend policy is a company's decision to determine whether the profits earned by the company will be distributed to investors in the form of dividends or by increasing the company's retained earnings. The purpose of this research was to determine the effect of earnings per share, investment opportunities, total asset turnover, and collateralizable assets on dividend policy in consumer goods industry entities listed on the Indonesia Stock Exchange for the 2015-2019 period. The population in this research was the consumer goods industry listed on the Indonesia Stock Exchange of 52 entities in 2015-2019. The technique of determining the sample used is purposive sampling. The samples obtained are 13 companies incorporated in the consumer goods industry sectors in the Indonesia Stock Exchange for 2015-2019. Based on the analysis, the result shows that the variable total asset turnover partially has a significant positive effect on dividend policy. Collateralizable asset variable has a significant negative impact on dividend policy. In contrast, the investment opportunity variables and earnings per share don't affect dividend policy. This research can be used as a reference for stakeholders in the company to consider the total asset turnover that affects dividend policy in a company.


INTRODUCTION
Financial management consists of three major decisions that are very important for the company. First, planning activities to implement company goals and objectives. Second, funding activities, namely the decision to seek funds to run the business. Third, investment activity refers to company acquisition and investment management to sell products, provide services, and invest excess cash (Subramanyam, 2017:16). One of the three decisions that usually become a problem in every company is the funding decision.
Dividends play a critical role in the company because they can describe its stability for the long term, explain its ability to produce profits from its operational activities, and manage its funds by paying dividends. The dividend is considered that it can affect investment opportunities for company capital formation. The dividends can be in the form of cash/cash, property, liquidation, shares, and certificates. The number of dividends paid will reduce retained earnings and money available to the company, and the amount depends on the policies of each entity. The proposal and distribution of dividends will be approved at the General Meeting of Shareholders (GMS).
Dividend policy is one of the three critical decisions of financial management. The central aspect of dividend policy is to decide whether the profits earned by the entity will be allocated to investors in the form of dividends or by increasing the company's retained earnings to finance investment in the future. If the company chooses to hand out the profits earned as dividends, it will reduce internal funding sources resulting in dependence on its external funding sources. Conversely, if the company chooses not to allocate dividends, the company's internal THE EFFECT OF EARNING PER SHARE, INVESTMENT OPPORTUNITY SET, TOTAL ASSET TURNOVER, AND COLLATERALIZABLE ASSET ON DIVIDEND POLICY Annisa Rizal and Dedik Nur Triyanto 210 sources of funds will be even greater. Another aspect that is considered for paying dividends is to reduce agency problems by meeting the interests of shareholders and maintaining the company's reputation for the management.
Dividend policy involves two interested parties: the shareholders with their dividends and the management with their retained earnings. Dividend policy often creates a conflict of interest between the management and the shareholders because they expect income in the form of dividends from the return on their investment. In contrast, the management less favours this because it can reduce the company's internal funds, which will impact the company's poor performance.
Conflicts of interest can cause the company to bear agency costs. Agency theory states that conflicts of interest can be reduced by implementing appropriate supervisory mechanisms to harmonize the interests of many parties in the company. Company management must determine the optimal dividend policy, a fair policy for shareholders with dividends and company management with company growth (Dewi & Sedana, 2018).
One of the entities that pay cash dividends is Gudang Garam Tbk. (GGRM) which distributed cash dividends of IDR 5 trillion in profit to shareholders in 2018. The allocation of dividends by the company can only be done if the company has positive retained earnings. However, not always entities that have positive retained earnings pay dividends. This happened to the company PT Gudang Garam Tbk. In 2019, Gudang Garam also managed to gain an increase in profit of Rp10.88 trillion. With such high net income, the company does not carry out a dividend policy that is profitable for shareholders. Based on the results of the AGMS, the shareholders agreed to determine the use of the company's profits for the 2019 financial year entirely to be included in the retained earnings account, which will be used to increase working capital. Investors were disappointed by the decision not to distribute dividends even though Gudang Garam had a brilliant performance for the 2019 financial year (market.bisnis.com).
Several factors influence dividend policy, one of which is earnings per share. Earning per share is a financial ratio used to quantify the company's ability to generate profits from invested shares. High earnings per share reflect good company performance, and the company can manage its finances. So, entities with high EPS are considered to distribute profits in the form of dividends and increase investor interest in investing. This step is based on the research of Kurniawati & Isroah (2017) and Diantini & Badjra (2016) that earnings per share positively and significantly affect the dividend policy. Unlike Kurniawati & Isroah (2017), according to Pamungkas et al. (2017) and Astuti (2020) and earnings per share doesn't affect the dividend policy.
Investment opportunities are opportunities owned by the company to invest with assets and investment options held by the company with a positive net present value. From this explanation, it can be interpreted that entities with high investment opportunities will require high funds for their investment needs so that dividend payments will decrease or may not be distributed at all. There is an inversely proportional relationship between investment opportunities and dividend payments. Putri & Susetyo (2020) and Sudarmono & Khairunnisa (2020) say that investment opportunities don't affect the dividend policy. Meanwhile, according to Chayati & Asyik (2017) and Putri & Andayani (2017), investment opportunities positively affect the dividend policy.
Total asset turnover (TATO) is a ratio used to measure the company's effectiveness using its assets. High asset turnover will reflect that the management can manage its assets properly to maximize its achievement. Therefore, the higher the total asset turnover rate, the higher the company's ability to pay dividends. According to Triono & Artati (2019) and Simanjuntak (2016), total asset turnover positively affects the dividend policy. Meanwhile, Jurnal Akuntansi ISSN 2303-0356 Vol. 11, No.3, October 2021 Angeline (2019) and Lindi et al. (2019) said that total asset turnover doesn't affect the dividend policy. According to Muslih & Husin (2019), collateralizable assets are guaranteed by the company when making loans to creditors. Creditors often ask for collateral for the debt they have lent in the form of assets. High collateralizable assets make creditors more secure to provide loans and don't need to impose restrictions on the company's dividend policy so that entities can pay dividends in high amounts. Entities that have high fixed assets are also considered to be able to owe high quantities. The high collateralizable assets can be used as a guarantee that will reduce agency conflicts in the company so that the company can pay dividends in more significant amounts. The assessments of Mangasih & Asandimitra (2017) and Darmayanti & Mustanda (2016) said that Collateralizable assets positively affect the dividend policy. However, according to Helmina & Hidayah (2017) and Setiawati & Yesisca 2016), Collateralizable assets don't affect the dividend policy.
This study will examine the effect of earnings per share, investment opportunities, total asset turnover, and collateralizable assets on dividend policy in consumer goods industry entities listed on the Indonesia Stock Exchange for the 2015-2019 period. This study's data analysis methods are descriptive statistics, classical assumption tests, panel data regression analysis, and hypothesis testing.
Earning per share is the ratio used to measure the net profit that the company can obtain from every rupiah of shares outstanding in one period. Earnings per share data can be seen from the company's income statement. In general, entities compare earnings per share each period with the same period in the previous year to describe the company's profit growth rate, which can later be used as a reference in making dividend policy decisions. The success of management in managing the company is reflected in the increase in earnings per share so that entities that have high EPS are considered to distribute profits in the form of dividends and increase investor interest in investing. Kurniawati & Isroah (2017) and Diantini & Badjra (2016) state that EPS positively affects dividend policy.

H1: Earning per share has a positive effect on dividend policy
An investment opportunity is an investment decision in the form of a combination of assets in place with future investment choices that will affect the company's value. Investment opportunities are opportunities for entities to make investments that are expected to obtain high returns in the future. Entities experiencing growth or with good financial conditions tend to choose new investments rather than pay dividends. The profit earned by the company is considered more profitable to invest in further investments than to share profits. It means that there is an inverse relationship between investment opportunities and dividend payouts. Therefore, it is in line with Purnami, Kadek Diah Arie & Artini (2016) research that investment opportunities negatively affect dividend policy.

H2: Investment opportunity has a negative effect on dividend policy
Total asset turnover is a measure of the activity ratio used to measure the company's level of efficiency in using its assets as a whole to generate sales. To get high sales, a company must work hard in turning its assets. Entities with high asset turnover illustrate that the company can obtain high sales, which has an impact on high company profits. The higher the total asset turnover ratio of a company, the greater the company's profit, so the amount of dividends to be distributed is also more significant. The research results support the statement's hypotheses by Firdaus and Handayani (2019) and Simanjuntak (2016), which states that total asset turnover positively and significantly affects dividend policy. Collateralizable assets are the number of assets guaranteed by the company to creditors when applying for loans. Collateralizable assets are considered as collateral for creditors to provide loans. Entities with more collateralizable assets have more minor agency problems, and this happens because these assets can be used as collateral for loans to creditors. Entities can use significant collateralizable assets to obtain large loans because high collateral can be used for large loans. It means that entities with large collateralizable assets are considered to increase profits. So, they can distribute large amounts of dividends compared to entities with small amounts of collateralizable assets. This research is in line with Mustanda (2016)

RESEARCH METHODS
This study's data analysis methods are descriptive statistics, classical assumption tests, panel data regression analysis, and hypothesis testing. The tests in the classical assumption test include linearity, autocorrelation, heteroscedasticity, multicollinearity, and normality tests. However, not all tests need to be performed. If the model has been assumed to be linear, then the linearity test is almost unnecessary. The autocorrelation test would be usable if the time series data format were set. A regression model uses more than one independent variable, it is necessary to do a multicollinearity test. (4) To provide certainty that the regression model has estimation accuracy, is unbiased, and is consistent, a heteroscedasticity test is needed. (5) In the Best Linear Unbias Estimator (BLUE) requirement, the normality test is not included and is not required to be fulfilled. The classical assumption test has been explained with the multicollinearity and heteroscedasticity test.
This research uses consumer goods in industry sector entities listed on Indonesia Stock Exchange (IDX) as a population. The sample of this research is consumer goods industry entities listed in Indonesia Stock Exchange (IDX) period 2015-2019. Sampling was carried out using a purposive sampling technique. This research takes a sample of consumer goods industrial sector entities listed on the Indonesia Stock Exchange (IDX) for 2015-2019. This research is classified into quantitative research and determines variables of theory testing based on data measurement and statistical analysis. The criteria for selection can be seen in table 1: Jurnal Akuntansi ISSN 2303-0356 Vol. 11, No.3, October 2021 Hal. 209 -222 The companies selected to be the sample 13 Total research sample 2015-2019(13 x 5) 65 Test analysis was conducted using secondary data in the company's annual report on the stock exchange website and the company's official website, and relevant supporting journals. This research used the dependent variable dividend policy, and the independent variables are earning per share, investment opportunity set, total asset turnover, and collateralizable assets.

Variable Operational Definition Dividend Policy
The dividend policy ratio is used because it's considered more able to describe managerial opportunities' behaviour by looking at the amount of profit distributed to shareholders as dividends and how much is kept in the company to compare the ratio in previous years with the current year.

= ℎ ℎ
Earning Per Share According to Kasmir (2018), earnings per share is a ratio used to measure the success of management performance in achieving profits for shareholders, calculated by reducing the amount of profits with taxes, dividends, and other rights for priority shareholders. ℎ = ℎ

Investment Opportunity
According to Myers on Putri & Susetyo (2020), the investment opportunity is a component of company value whose expenditure is determined by management based on future investment options expected to generate greater returns. The investment opportunity ratio in this research uses a price-based proxy for Capital Expenditure to Book Value of Assets (CAPBVA) because it describes its growth prospects.

Total Asset Turnover
According to Gitman & Zutter (2015: 123), total asset turnover is one measure of the activity ratio used to the company's level of efficiency in using overall assets to generate sales.

Collateralizable Asset
Collateralizable assets are the ratio of fixed assets to total assets used as collateral for agency costs caused due to conflicts between shareholders and creditors (Suci, 2016.  Based on that table, the dividend policy projected by the dividend payout ratio (DPR) has a mean value of 0.52783 and a standard deviation of 0.28746, so it can be interpreted that the data from the dividend policy (DPR) does not vary or are grouped. The maximum value of the dividend policy (DPR) is 1.22683, namely from SIDO in 2017. The minimum value of the dividend policy (DPR) of 0.15055 comes from SKLT in 2016.
Earning per share (EPS) has a mean value of 336.02544 and a standard deviation of 491.01184, so it can be interpreted that the EPS variable data varies or is not grouped. The maximum value of the EPS variable is 2,707.92910, namely from TCID in 2015. The minimum value of the EPS variable is 7.22107, namely from CINT in 2019.
The investment opportunity projected by capital expenditure (CAPBVA) has a mean value of 0.03199 and a standard deviation of 0.04138. It can be interpreted that the investment opportunity variable data is varied or not grouped. The maximum value of the CAPBVA variable is 0.26594, namely from SKLT in 2016. The minimum value of the CAPBVA variable is -0.02368, namely from TCID in 2019. ISSN 2303-0356 Vol. 11, No.3, October 2021 Total asset turnover (TATO) has a mean value of 1.24813 with a standard deviation of 0.54634, which means that the TATO variable data is grouped or does not vary. The maximum value of the TATO variable was 2.39188 UNVR in 2016. The minimum value of the TATO variable is 0.54634, namely from ROTI in 2017.

Jurnal Akuntansi
Collateralizable assets (COLLAS) have a mean value of 0.36658 and a standard deviation of 0.14461, so it can be interpreted that the data is grouped or does not vary. The maximum value of the COLLAS variable is 0.67301 from ROTI in 2015. The minimum value of the COLLAS variable is 0.05920, namely from DLTA in 2018.

Classic Assumption Test Multicollinearity Test
The following are the result of multicollinearity testing on each independent variable of the research. From the multicollinearity table, it can be known that each independent variable has a correlation value of < 0.800. This value indicates that the independent variable in this research does not have a correlation or is free from multicollinearity

Heteroscedasticity Test
The following are the result of heteroscedasticity testing on each independent variable of the research.  (2021) From table 4, it can be known that the probability value for each independent variable is > 0.050. It means that the independent variable in this research does not occur heteroscedasticity.

Uji Chow
This test is conducted to determine which model is most appropriate for estimating panel data between the common or fixed-effect models. Based on table 5 of the test results using the Chow test, it can be seen that the value of the cross-section is 0.0000, which means that the value is less than 0.05, then H0 is rejected, and H1 is accepted. Based on these data, it can be concluded that the fixed effect model is better to use than the standard effect model. After it was concluded that the fixed-effect model was selected, it was necessary to re-test, namely by conducting the Hausman test.

Uji Hausman
This test is conducted to determine which model is the most appropriate for estimating panel data between Fixed Effect or Random Effect Models.

Uji Lagrange Multiplier
This test is conducted to determine whether the random effect model is better than the common effect method using the Lagrange multiplier test.  Table 7 shows the results of the Lagrange Multiplier test with the probability value (pvalue) of Breusch-Pagan being 0.0000 less than the standard significance of 0.05. It means that H0 is rejected or the panel data regression model uses the random-effects model because it is considered better than the common effect model.

Panel Data Regression
Based on the model testing that has been done, the model that should be used in this study is the random effect model. The data results (table 8) show that the Panel data regression model can formulate the random effect model testing. It explains the effect of earnings per Jurnal Akuntansi ISSN 2303-0356 Vol. 11, No.3, October 2021 share, investment opportunities, total asset turnover, and collateralizable assets on the dividend policy of consumer goods industry companies listed on the Indonesia Stock Exchange in 2015 -2019. The interpretation of the regression equation shows the constant value is 0.353486. It means that if the value of the independent variable earnings per share (EPS), Investment Opportunity (IOS), total asset turnover (TATO), and collateralizable assets (COLLAS) is 0, then the value of the dependent variable dividend policy (DPR) ) of 0.353486 units. The regression coefficient value of earnings per share (EPS) is -0.650E05. It shows that every time there is an increase of one EPS unit, assuming other variables are constant, the dividend policy (DPR) will decrease by -0.650E05 units. The value of the investment opportunity regression coefficient (IOS) is 0.726776. Showing that every time there is an increase of one IOS unit, assuming other variables are constant, the dividend policy (DPR) will increase 0.726776 units. The regression coefficient value of total asset turnover (TATO) is 0.340184. It indicates that every time there is an increase in one unit of TATO with the assumption that other variables are constant, the dividend policy (DPR) will increase by 0.340184 units. The value of the regression coefficient for collateralizable assets (COLLAS) is -0.707503. It indicates that for every increase in one unit of COLLAS with the assumption that other variables are constant, the dividend policy (DPR) will decrease by -0.707503 units.

Coefficient of Determination Test
The following are the result of coefficient of determination testing on each independent variable of the research.   Table 9 shows the value of the coefficient of determination (R 2 ) is 0.252091 or 25.2091%. It means that independent variables can describe the dependent variable (dividend policy) of 25.2091%, and other variables of 74.7909% describe by the other variable.

Simultaneous Test (F Test)
The following are the result of simultaneously testing the research. From table 10, it can be known that the F test shows the value is 0.000236, which is smaller than the 0.05 significance level, then H1 is accepted. It means that earnings per share, investment opportunity set, total asset turnover, and collateralizable assets simultaneously affect the dividend policy in consumer goods industrial entities listed on the Indonesia Stock Exchange in 2015-2019.

Partial Test (t)
The following are the result of partial testing of the research. From table 11, it can be known that the coefficient values for EPS have a probability value of 0.1604, IOS has a probability value of 0.2087, TATO has a probability value of 0.0002, and COLLAS has a probability value of 0.0271.

Effect of Earning Per Share on Dividend Policy
Earnings per share have a probability value of 0.1604, which means that earnings per share do not affect dividend policy. These results support the results of research conducted by Jurnal Akuntansi ISSN 2303-0356 Vol. 11, No.3, October 2021 Hal. 209 -222 Pamungkas et al. (2017) and Astuti (2020), which state that earnings per share do not affect dividend policy. The higher earnings per share will not affect the dividend policy because it is not always a company that earns a profit that will decide to distribute dividends to shareholders so that the increase in company profits will be used for profitable investment purposes.

Effect of Investment Opportunity on Dividend Policy
The probability value of the investment opportunity variable is 0.1087 > 0.05, which means that investment opportunities do not affect dividend policy. These results support the results of research conducted by Putri dan Susetyo (2020) and Sudarmono & Khairunnisa (2020), which state that investment opportunities do not affect dividend policy. It means that size of the investment opportunity value will not affect the high or low value of the dividend policy. This research proves that a company does not pay attention to investment factors for decisions in paying dividends. The company is more focused on efforts to generate profits in order to have sufficient profit reserves to be invested without having to reduce the distribution of dividends to shareholders.

Effect of Total Asset Turnover on Dividend Policy
The value of total asset turnover is 0.0002 < 0.05, which means that total asset turnover positively affects the dividend policy. This research is the same with research by Firdaus dan Handayani (2019) and Triono & Artati (2019) that total asset turnover positively affects the dividend policy. This research proves that the faster the turnover of assets, the net profit generated will increase because the company can utilize assets to increase sales, which affects revenue. Thus the more effective the turnover of company assets results in high company performance to increase company profits and impact the increase in the rate of return obtained by investors. It means that there is an increase in Total Asset Turnover (TATO). If the value of dividend policy increases and vice versa, it can be said that dividend policy and total asset turnover are directly proportional.the coefficient of total asset turnover is 0.340184, meaning that each increase of one unit of total asset turnover will increase dividend policy regression coefficient of 0.340184, assuming that other independent variable of total asset turnover are considered constant.

Effect of Collateralizable Asset on Dividend Policy
The probability value of the collateralizable asset is 0.0271 <0.05. It means that the collateralizable asset has a negative effect on the dividend policy. The negative coefficient on the t-test results indicates that entities with a high level of asset security will have a low dividend policy and vice versa. The high level of collateralizable assets in the company illustrates that the greater the number of fixed assets in the company's total assets, the greater the assets that can be used as collateral in taking the company's debt. Entities with a high level of collateralizable assets are more likely to use the profits generated to pay for asset collateral rather than being distributed to shareholders to pay off existing obligations. So, that the number of dividends to be distributed will be reduced because the company prioritizes paying its debts. The result of this research is the same as research by Muslih & Husin (2019) and Wahjudi (2019) that collateralizable assets negatively affect the dividend policy. If the value of collateralizable assets is high, then the value of the dividend policy will be low. Entities with a high level of collateralizable assets are more likely to use the profits generated to pay for asset collateral rather than being distributed to shareholders to pay off existing obligations. So that the amount of dividends to be distributed will be reduced because the company prioritizes paying its debts.

CONCLUSIONS AND SUGGESTION
The research shows that earnings per share, investment decisions, total asset turnover, and assets that can be pledged simultaneously affect dividend policy. While partially earnings per share and investment opportunity set don't affect dividend policy. Meanwhile, total asset turnover positively and significantly affects dividend policy, and collateralizable assets negatively and significantly affect dividend policy. This research suggests improving the company's performance and managing total asset turnover effectively to achieve high sales. As well as managing assets properly and effectively to increase sales which can later achieve the desired profit or high profits so that they can consistently distribute dividend policies to shareholders so that they are interested in investing. Investors are recommended to analyse the Total Asset Turnover (TATO) to be used as a source of information in providing an overview of developments to invest in or not in a company. According to the results of this research, it is stated that high total asset turnover affects dividend policy. The current study has many limitations, only using consumer goods industrial sector entities, and only takes a sample of 13 entities and only has four independent variables.