The Effect of Dividend Policy, Investment Decision, and Ownership Structure on Company Value With Profitability as a Moderation Variable

— This study aims to determine the effect of dividend policy, investment decisions, ownership structure, firm size, and profitability on firm value and to determine the effect of moderating profitability with respect to the effect of dividend policy, investment decisions, ownership structure on firm value in manufacturing companies listed on the Stock Exchange. Indonesian Securities (IDX). By using a sample of 70 manufacturing companies during the period 2016 – 2020, the results of this study indicate that: (i) dividend policy has no significant effect on firm value; (ii) investment decisions have a positive and significant effect on firm value; (iii) managerial ownership has a positive and significant effect on firm value; (iv) institutional ownership has no significant effect on firm value; (v) firm size has a negative and significant effect on firm value; (vi) profitability (ROA) has no significant effect on firm value; (vii) profitability (ROA) does not moderate the effect of dividend policy, investment decisions, and managerial ownership on firm value; (viii) profitability (ROA) moderates institutional ownership on firm value.


INTRODUCTION
In this study, the authors wanted to examine the factors that influence firm value, one of which is dividend policy, because firm value can be seen from the company's ability to pay dividends to investors.The greater the dividend given by the company to investors, the higher the level of investor confidence in the company.According to Nurvianda et al., (2018) a dividend policy determined by company management is a signal to shareholders so they can assess the condition of the company.Many shareholders want to get excess profits.However, the company's management is trying to be able to manage funds properly.Managers are still expected to be able to increase the value of the company by increasing profitability and giving dividends to shareholders in order to prosper the company's shareholders.Dividend policy is a decision whether the profits earned by the company will be distributed to shareholders as dividends or will be retained in the form of retained earnings to finance investment in the future (Rusdiana, 2017: 129).Research conducted by Putra and Lestari (2016) states that dividend policy has a positive effect on firm value.However, research from Abidin, et al. (2016) obtained different results which stated that dividend policy had a negative and significant effect on firm value.
Factors that affect the value of the company is the investment decision.Company activities that have the goal of increasing profits are how the company determines investment decisions.When companies expect large profits in the future by managing and circulating funds, companies need to make appropriate and mature policies so that these investment decisions will not harm the company.This is because the investment decisions made by the company can affect investors' perceptions of the company, whether the investment decisions that have been made can attract investors to invest their capital or not.If the investment decision is in line with the company's goal of increasing the value of the company, then it will be a distinct advantage for the company to obtain a large rate of return.[1] state that investment decisions have no effect on firm value.On the other hand, [2] stated that there is a significant positive effect on investment decisions on firm value.
Then the factors that affect the value of the company is the ownership structure.The ownership structure in question is share ownership by managers, and by institutions that monitor the company's operational activities, so that company performance and company value can increase.Ownership of shares by managers can reduce agency problems between management and shareholders.The fundamental problem in agency conflict is selfinterested behavior because company managers may have personal goals that compete with the goal of maximizing shareholder wealth.With share ownership by managers, it is expected that the interests of managers can be aligned with the interests of shareholders.In this study the researchers chose two ownership structures consisting of managerial ownership and institutional ownership [3] [4].
This study uses manufacturing companies listed on the Indonesia Stock Exchange as a population.Manufacturing companies reflect the growth and development of the national economy and business, besides that manufacturing companies are a sector that has high business complexity.

Dividend Policy
The dividends represent direct receipts for shareholders for their investment in the company.Dividend policy according to [5] is a policy to distribute company profits to shareholders in the form of dividends or to hold them in the form of retained earnings which are then reused as investments in the future.Dividend policy is a corporate action that needs to be implemented by the company because this policy involves the number of shareholders' profits.

Investment Decision
Investment is delaying current consumption to be included in productive assets for a certain period of time [6].Parties who make investments are called investors.Investors can be classified into two, namely individual investors (individual/retail investors) and institutional investors (institutional investors).Individual investors consist of individuals who carry out investment activities, while institutional investors consist of insurance companies, depository institutions, pension fund institutions, and investment companies.Investors also learn how to manage investor's wealth which is monetary in nature.

Ownership Structure
In this study the ownership structure refers to share ownership by managers and institutions.Managerial ownership is the percentage of share ownership by management who actively participates in making company decisions in this case the manager owns the company and will try to reduce agency problems and increase company value [7].

Profitability
The profitability ratio, also known as profitability, describes a company's ability to earn profits through all existing capabilities and sources, such as sales activities, cash, capital, number of employees, number of branches and so on.The ratio that describes a company's ability to generate profits is also called the operating ratio [8].

Company Size
As a proxy for company size (size), generally studies examining the relationship between size and company profitability use the natural logarithm of total assets (Ln TA), this is used to reduce significant differences between company sizes that are too large and company sizes that are too small., then the total asset value is formed into a natural logarithm, the conversion to natural logarithm aims to make the total asset data normally distributed.

Company Value
The goal of company management is to maximize the value of the wealth of shareholders [9].The value of the company can be measured through the value of the stock price in the market based on the formation of the company's share price in the market, which is a reflection of the public's assessment of the company's performance in real terms.It is said in real terms because the formation of prices in the market is the confluence of points of stability of demand power and points of stability of price supply power which in real terms occur buying and selling of securities in the capital market between sellers (issuers) and investors, or often called equilibrium market.
Therefore, in capital market financial theory, stock prices on the market are referred to as the concept of firm value [10].The higher the stock price, the higher the value of a company METHOD This strategy was chosen because associative research is in accordance with the researchers' objectives to be achieved, namely to determine the effect of dividend policy, investment decisions, and ownership structure (independent variable), firm value (dependent variable), profitability as measured by net profit margin (moderating variable).and company size as measured by ln total assets (control variable) to firm value in manufacturing companies listed on the Indonesia Stock Exchange.The data taken in this study is in the form of financial reports presented in 2016-2020.
The population for this study were all manufacturing companies listed on the IDX in the 2016-2020 period and 70 companies were selected in the 2016-2020 period that met the criteria for being sampled.In this study, purposive sampling was used to collect samples, namely selecting samples based on certain criteria and systematics.
The criteria used in selecting the sample are: 1. Manufacturing companies in the consumer goods industry sector that have been and are still listed on the Indonesia Stock Exchange and published financial reports for the 2016-2020 period.
2. Manufacturing companies that distributed cash dividends during the 2016-2020 period.
3. Manufacturing companies that have profit growth during the 2016-2020 period.
4. Manufacturing companies reporting their financial statements during the 2016-2020 period.
The type of data used in this research is secondary data.The research data is taken from the company's published annual financial reports.The researchers' secondary data were obtained from the Indonesia Stock Exchange with the website www.idx.co.id.
In this study, researchers wanted to know the relationship between dividend policy, investment decisions, and company growth on firm value.The method used in this study is multiple linear regression analysis assisted by eviews 10 software.In the rules of multiple regression analysis, the first step that must be carried out is to produce valid estimator value parameters.A value is said to meet the category if the results of the classical assumption test carried out can fulfill the assumptions of autocorrelation, heteroscedasticity, and multicollinearity.

Descriptive Statistical Analysis
Based on the results of descriptive statistical analysis, table 1 below presents the minimum, maximum, median, mean, and standard deviation values.Each variable is calculated using the Eviews 10 software, including: Based on table 4.1 it shows that the value of Tobin's Q within 5 years of research shows that the dependent variable (y) has a maximum value of 11.26380 and a minimum value of 0.273227.The average value is 1.568154 with a standard deviation of 1.292303.Which means the standard deviation value is smaller than the average value, this indicates that the level of risk of deviation from Tobin's Q is low.

Dividend Policy (Dividend Yield)
Dividend Yield to measure the dividend policy taken at certain companies.The table shows a maximum value of 0.322840 and a minimum value of 0.000000.The average value is 0.017453 with a standard deviation of 0.027589.Which means that the standard deviation value is greater than the average value, this indicates that the level of risk of deviation from dividend policy is high.

Investment Decision (Capex)
Capex to measure whether the right investment decision, will make the company obtain profits that allow it to increase the growth of the company's assets.The table shows a maximum value of 1.167165 and a minimum value of -6.479508.The average value is -0.021097 with a standard deviation of 0.386932.Which means that the standard deviation value is greater than the average value, this indicates that the level of risk of deviation from investment decisions is high.

Managerial Ownership (KM)
KM is the percentage of share ownership by management who actively participates in company decision making.The table shows a maximum value of 9.079849 and a minimum value of 0.000000.The average value is 0.139648 with a standard deviation of 0.529789.Which means that the standard deviation value is greater than the average value, this indicates that the level of risk of deviation from managerial ownership is high.

Institutional Ownership (IP)
KM is the number of shares owned by the institution out of the total outstanding shares.The table shows a maximum value of 1.893121 and a minimum value of 0.000000.The average value is 0.623537 with a standard deviation of 0.289998.Which means the standard deviation value is smaller than the average value, this indicates that the level of risk of deviation from institutional ownership is low.

6.
Firm Size (TA) Firm Size as measured by (ln) total assets.The table shows a maximum value of IDR 340.423 billion and a minimum value of IDR 246.283 billion.The average value is IDR 285.166 billion with a standard deviation of IDR 17.193 billion.Which means the standard deviation value is smaller than the average value, this indicates that the level of risk of deviation from firm size is low.

Profitability (ROA)
ROA is a company's ability to earn profits through all existing capabilities and sources such as sales activities, cash, capital, number of employees, number of branches and so on.The table shows a maximum value of 2.640992 and a minimum value of -0.640141.The average value is 0.059127 with a standard deviation of 0.180576.Which means that the standard deviation value is greater than the average value, this indicates that the level of risk of deviation from profitability is high.

Multicollinearity Test
The multicollinearity test aims to test whether the regression model found a correlation between the independent (independent) variables.A good regression model should have a correlation between the independent variables: So it can be concluded that there is heteroscedasticity of the cross section factor.Based on the results of the heteroscedasticity test in table 4.3. in the cross section shows a probability value of 0.0000 <0.05 which means that the error has symptoms of heteroscedasticity caused by cross section (company) data.Whereas the period test shows a probability value of 0.8020 > 0.05, this shows that there are no symptoms of heteroscedasticity caused by the period (year).So, it can be concluded that there is a heteroscedasticity problem caused by the cross-section factor (company).

Correlation Test
According to Hsio and Pesaran (2014) the Correlation Test aims to measure errors between companies and measure errors between time whether they influence each other or not.The correlation test can be divided into two, namely cross correlation and autocorrelation.The following are the results of the coross correlation and autocorrelation tests:

Cross Correlation Test
Cross correlation test is used to test whether errors between companies are correlated (Hsio and Pesaran 2014: 328).Based on table 4. the Breusch-Pagan LM test is 0.0000 <0.0500, the LM scaled Pesaran test is 0.0000 <0.0500, and the CD Means test is 0.0001 <0.0500 which indicates that there is a cross correlation.Thus the three cross-correlation test methods say that the hypothesis that there is no cross-section dependency is rejected, so it can be concluded that there is a cross-corealation problem.

Autocorrelation Test
The autocorrelation test aims to determine whether there is a confounding variable in a period that is correlated or uncorrelated with other confounding variables.The way to detect autocorrelation problems is to use the Durbin Watson (DW) test and then compare the test results with the Durbin Watson (DW) table (Ghozali, 2018: 121).Based on the results of the autocorrelation test with Durbin-Watson (DW) in Table 5, it shows that the value of d is 1.5821.While the d_L value is 1,724.When viewed from the decision making that has been determined, the value of d is between the values of 0 and d_L, namely 0 < 1.5821 < 1.724 (0 < d < d_L).So it can be concluded that there is a negative autocorrelation problem in the regression model.

Uji Chow
This test aims to choose between the common effect model or the fixed effect model which should be used to determine the right regression model in research (Ghozali, 2018: 166): 1.If the probability value for the cross-section F > 0.05 means that H_0 is accepted, then the most appropriate common effect model (CEM) is used.If seen from the results of table 4.6, it shows that the cross-section F is 0.0000 with a significant value <0.05, then H_0 is rejected, thus the method used in this study is the fixed effect model.

Hausman test
The Hausman test is a statistical test to determine whether the fixed effect or random effect model is the most appropriate to use.
1.If the probability value > significance value (a = 0.05) then H_0 is accepted so the model used is a random effect approach.
2. If the probability value < significance value (a = 0.05) then H_0 is rejected so the model used is the fixed effect.If seen from table 7, it shows that the random cross-section obtained is equal to a significant value then H0, thus the method used is the final fixed effect model.

Panel Data Regression Analysis
Panel data regression is an analysis technique that combines time series data with cross-sectional data.Based on the test results above, the regression results obtained are using the final fixed effect model (FEM).
According to Widarjano (2018: 367), the fixed effect model is a technique for estimating panel data using a dummy variable to capture intercept differences.This model assumes a different intercept between companies (cross section) but the intercept is the same between time, but has a constant (fixed) regression slope between companies and between time (time series).
Due to the presence of panel symptoms of heteroscedasticity, cross correlation and autocorrelation, the final fixed effect model (FEM) estimation was performed using white cross-section weights and white cross-section coefficient covariance method.Based on the results of the final FEM model regression analysis in table 8, it shows that there is a constant value of 3.6353 with a probability of 0.0005.In the regression on the Adjusted R-squared value is quite good, which is equal to 0.9127 which explains that the variable firm value as measured by Tobin's Q is influenced by dividend policy, investment decisions, managerial ownership, institutional ownership, profitability as measured by ROE, firm size as measured by ( ln) total assets of 0.9127%, the remaining 0.9325% is influenced by other factors not examined in this study.And the following is a discussion of the equations from the results of the moderated regression analysis obtained from the final A constant with a value of 3.6353 This means that if the dividend policy variable is measured by dividend yield, investment decisions are measured by capex, managerial ownership, institutional ownership, profitability is measured by ROA and company size is measured by ln the total assets value is 0 (zero).), then the Tobins'Q variable has a value of 3.6353.
The dividend yield coefficient is -0.4409 which indicates that the dividend yield has no significant effect on firm value.This illustrates that if the dividend yield increases by 1 unit, assuming other variables are constant, it will increase Tobin's Q by -0.4409.The Capex coefficient is 0.0603 which indicates that Capex has no effect on firm value, this illustrates that if Capex increases by 1 unit assuming other variables are constant, it will increase Tobins'Q by 0.0603.
The KM coefficient is 0.1014 which indicates that KM has no effect on firm value.This illustrates that In addition, the company is expected to be able to increase or maximize the value of the company so that it can attract investors to invest in the company.This can be done by carrying out proper financial management functions, where one decision taken will affect other financial decisions and have an impact on the value of the company.

Table 7
Hausman test

Table 8 .
Panel Data Regression Analysis