Understanding Defensive Stocks with Company Fundamentals and Dividend Policy Variables as Moderation

The purpose of this study was to determine the effect of Return on Equity, Firm Size, Debt Equity Ratio and Price Earning Ratio on return defensive stocks with Dividend Policy as Variable Moderation in Manufacturing Companies. The study population was manufacturing sector companies listed on the Indonesia Stock Exchange from 2015 to 2018. There were 61 companies sampled using purposive sampling techniques. The analytical method uses multiple linear analysis and moderated Regression analysis. The results showed the Firm Size variable had a significant positive effect and the DER variable had no significant positive effect while the ROE and PER variables had no significant negative effect on return defensive stocks. The MRA test states that the Dividend Payout Ratio variable strengthens the effect of Firm Size on return defensive stocks while the ROE, DER and PER variables are weakened by the effect on return defensive stocks by dividend policy. Future studies are expected to use other variables that might increase or decrease stock returns. Article Information Article History: Received February 2020 Approved March 2020 Published September 2020


INTRODUCTION
Investment is a commitment to several funds or other resources with the hope to obtain profits in the future (Bodie et al., 2014). For investors, the capital market is a vehicle that provides various investment alternatives with various returns and levels of risk, while for the capital market issuers is a means to raise longterm funds to finance business continuity.
In investing, investors expect profits in the future. A large return will motivate investors to continue investing and be prepared to bear the risk of the investment made (Pardosi & Wijayanto, 2015). For individual investors or institutional investors, selected stock investments are expected to provide a rate of return following the risks that must be borne by investors. For investors, this rate of return becomes a major factor because returns are the results obtained from an investment (Tandelilin, 2001).
Return is the level of profit enjoyed by investors for the investment actions that have been taken (Nurmalasari & Yulianto, 2015). The calculation of stock returns consists of capital gains and capital losses Mathematically the stock return formula is calculated at the current price minus the period of last year and divided by last year (Jogiyanto, 2003).
Investments in selected securities are certainly expected to provide a level of return following the risks that must be borne by investors (Safitri & Yulianto, 2015). When investors consider the level of return and risk contained in shares, investors can choose the type of stock that is classified as defensive. This type of stock price is generally not affected by economic fluctuations, so it has a relatively low market or beta risk (Novy-Marx, 2014).
Defensive stocks tend to have low market risk, but that does not mean these types of stocks are risk-free. Therefore, investors need to analyze the company's fundamental performance. The company's fundamental performance can be analyzed through financial statements (Fitriati & Prasetiono, 2010). Some ratios that need to be considered before investing are Profitability, Firm Size, Leverage, Price Earning Ratio and Dividend Policy.
ROE shows the company's ability to generate profits from the equity used. The higher the value of ROE shows the increasing ability of companies to generate net profits using their capital (Hanafi, 2004). The company's strength in generating high profits can be interpreted that the company has a high return. This is supported by research conducted by Sakina and Prasetiono (2017) and Hunjra et al. (2014) that ROE has a significant positive effect on return defensive stocks. However, several previous studies have found that ROE has a significant negative effect examined by Novy-Marx (2014).
Firm size can be interpreted as a company size which can be found by multiplying the stock price by the number of shares outstanding. Prihanantyo (2015) states that the greater the size of the company which is proxied as market capitalization, the greater the return that will be obtained. Large market capitalization is generally one of the attractions of investors in buying shares (Taslim & Wijayanto, 2016). This is supported by research conducted by Acheampong et al. (2014) and Putri (2016) that firm size has a significant positive effect on return defensive stocks. However, some researchers have previously found a negative relationship between firm size and return defensive stocks studied by Fama and French (1992) and Fitriati and Prasetiono (2010).
A high company DER indicates that the company has a high obligation to repay loans to creditors. Fulfilment of sources of funds through debt will affect the level of corporate leverage because leverage is the ratio used to measure how far the company uses debt (Wibowo & Wartini, 2013). A high DER reduces investors in investing their funds due to companies that depend on external parties. The impact that occurs is a decrease in the company's stock price, so stock returns tend to be lower (Ang, 1997). This is supported by research from Acheampong et al. (2014) and Puspitadewi (2016) that leverage has a significant negative effect on return defensive stocks. But researchers Abdullah et al. (2015) and Sakina and Prasetiono (2017) found DER variable has a positive effect on return defensive stocks.
PER is one indicator of investor or market confidence in a company as well as an indicator of market confidence in the company's growth prospects (Haryanto, 2014).The higher the PER, the higher the price per share in the market. Investors will be interested in low-value company PER because shares can be bought cheaply in the hope that the stock returns obtained will be higher in the future that will bring (Fatahurrazak et al., 2014). Researchers Darusman and Prasetiono (2012) and Antara (2012) agree that the results of his research the relationship between price earning ratio to return defensive stocks is significantly negative. However, previous researchers found a significant positive relationship between price earning ratio and return defensive stocks examined by Arslan and Zaman (2014) and Puspitadewi (2016).
DPR is a corporate profit given to shareholders according to the size of share ownership in the company (Harlina & Khoiruddin, 2018). If the company distributes dividends to shareholders, it will reduce retained earnings. Dividend policy is often regarded as a signal for investors in assessing the merits of a company, this is because dividend policy can affect the company's stock price. Thus how many portions of the profit is distributed in the form of dividends and how many portions of retained earnings to be reinvested, is a serious problem for management (Kurniawan, 2017).
This study tries to develop previous research conducted by Sakina and Prasetiono (2017) regarding the analysis of the effect of profitability, firm size, leverage and price earning ratio on return defensive stocks with dividend policy as a moderating variable.
The purpose of this study was to determine the effect of profitability, firm size, leverage and price earning ratio on return defensive stocks and determine the ability to moderate dividend policy variables.

Hypothesis Development
ROE describes a company's ability to generate a return on investment and is used in comparing two or more companies in the same industry (Halim, 2003). Companies that generate high and stable profits will get a positive response from investors. Stock returns will rise along with rising stock prices (Sudiyatno & Suharmanto, 2011). This is considered positive by investors because the company can utilize resources and have an impact on company stock returns. H1: Profitability has a significant positive effect on Return Defensive Stocks.
Firm Size describes the resources owned by the company, but it also describes the financial condition of the company in a period. In general,

H5
stocks with large company size have good company development potential with relatively low risk, so they can attract investors to trade (Pomfret & Saphiro, 2007). H2: Firm Size has a significant positive effect on Return Defensive Stocks.
DER illustrates the attachment between the amount of long-term debt provided by creditors compared with the amount of own capital owned by the company (Syamsuddin, 2011). High leverage indicates financial risk or the risk of company failure to repay loans will be higher, and vice versa (Wijayanto, 2010). The impact that occurs is a decrease in the company's stock price, so stock returns tend to be lower. H3: Leverage has a significant negative effect on Return Defensive Stocks.
PER information indicates the amount of rupiah that must be paid by investors to obtain one rupiah of corporate earnings (Tandelilin, 2010). The higher the PER shows the prospect of the share price being valued the higher by the investor on earnings per share, so the higher PER also shows the more expensive the stock is on the earnings per share. H4: Price Earning Ratio has a significant negative effect on Return Defensive Stocks.
Companies with high profitability will also distribute high dividends to ensure the company can meet the needs and expectations of all stakeholders (Hunjra et al., 2014). Dividend policy is often regarded as a signal for investors in assessing the merits of a company, this is because dividend policy can affect the company's stock price.
H5: Dividend Policy moderates the effect of Profitability on Return Defensive Stocks.
Large firm size is one indicator of a stable company that maintains financial performance, which makes the company much sought after by investors (Darusman & Prasetiono, 2012). Large companies tend to distribute regular dividends and cause greater stock returns. Firm size influences investors decisions because of the business risk factors of small companies that tend to use their profits to expand. H6: Dividend policy moderates the effect of Firm Size on Return Defensive Stocks.
Companies with high leverage require full attention in dividend distribution because some of the company's profits are used to pay off obligations first. Companies that have low equity, the debt to equity ratio is also high so that the interest paid is also higher, therefore the company may divide profits by a small amount, thus causing a company's dividend policy is low (Larasati, 2015). H7: Dividend policy moderates the effect of leverage on Return Defensive Stocks.
According to Abdullah and Agaki (2016) the price earning ratio contains information about net income (earnings per share) obtained by the company. The higher the EPS, the smaller PER of a company. Theoretically, net income is used to predict the value of dividends to be paid by companies. The increase in the number of dividends paid indicates that the company's net income (earnings per share) has also increased and the company's PER has decreased. H8: Dividend policy moderates the effect of Price Earning Ratio on Return Defensive Stocks.

METHODS
This type of research uses quantitative research methods. Quantitative data used are secondary data in the form of summary financial statements and annual reports of companies included in the manufacturing sector from 2015 to 2018 obtained from the official website (Idx. co.id, 2019). Quantitative research emphasizes its analysis on numerical data that are processed by statistical methods (Sugiyono, 2016).
The sampling technique used in this study was purposive sampling. The research sample was 61 companies and the observation year was carried out for 4 years starting from 2015 to 2018, so the data used in the study. The variables used in this study include stock returns, return on equality, firm size, leverage, price earning ratio and dividend policy. Stock returns are changes in stock prices from the previous period. The number of stock returns in this research is determined using the following formula (Jogiyanto, 2003): Stock Returns = Profitability is proxied by Return on Equity which is the ratio between net income after tax and total own capital. The amount of ROE in this research is determined using the following formula (Sihombing, 2019): ROE = Firm size is proxied by market capitalization which is a calculation of the number of shares outstanding at the price per share. The size of the firm size in this study was determined using the following formula (Putri, 2016):

Market Capitalization = Outstanding Share x Price per Share
Leverage is proxied by Debt to Equity Ratio which is the ratio between debt and equity. The amount of leverage in this research is determined using the following formula (Sakina & Prasetiono, 2017): Practically Price Earning Ratio is obtained by dividing the stock price by earnings per share (Khajar, 2013). The amount of price earning ratio in this research is determined by the following formula: PER = Dividend policy is proxied by the Dividend Payout Ratio which is a comparison between dividend per share and earnings per share. The amount of DPR in this research is determined using the following formula (Ang, 2007):

DPR =
This study uses panel data, which consists of several categories and collected at a certain period (Rosadi, 2012). The selection of panel data is because in this study uses several years and also many companies.
The model I uses multiple linear regression analysis, this model is used to determine the effect of independent variables on the dependent variable (Ghozali & Ratmono, 2017). This test uses a t-test statistic that shows how far the influence of one independent variable individually in explaining the dependent variable. The error rate in this study was 5%.
Model I regression equation: The model II uses Moderated Analysis Regression, MRA uses an analytical approach that maintains sample integration and provides a basis for controlling the influence of moderator variables (Ghozali, 2016). The purpose of this analysis is to determine whether the moderating variable will strengthen or weaken the relationship between the independent variable and the dependent variable. If the moderated variable constant is positive, then the moderating variable strengthens the influence of the independent variable on the dependent variable and vice versa (Astakoni et al., 2019). In this study using dividend policy as a moderating variable, to determine the ability of dividend policy in strengthening or weakening the influence of independent variables on the dependent variable.
Model II regression equation:

Descriptive Statistical Analysis
Descriptive analysis is used to describe and describe the variables in this study. Descriptive statistical analysis in this study includes the mean, maximum, minimum and standard deviation (Yanto et al., 2017). The results of the descriptive statistical analysis can be seen in Table.1 below: The results showed that the average Return on Equity value of 0.03 in the 2015-218 time span. The maximum value of ROE is 5.09 and the minimum value of ROE is -11.04. ROE standard deviation is 1.07 for the period 2015-2018.
The results showed that the firm size variable obtained an average value of 27.92 in the 2015-2018 time span. The maximum firm size value is 550,184,503,737,000 or 33.94 after the results of data processing using natural logarithmic calculations and the minimum value of variable size is 33,120,000,000 or 24.22. Firm size standard deviation value is 2.20 during the 2015-2018 period.
The results showed that the Debt to Equity The results showed that the Dividend Payout Ratio obtained an average of 0.25 in the 2015-2018 time span. The maximum value of the DPR is 5.35 and the minimum value of the DPR is 0.00, this is because the company does not distribute dividends to shareholders. The standard deviation value of the dividend policy is 0.50 for the period 2015-2018.

Selecting Panel Data Estimation Model
Estimation aims to choose which method is best for determining the regression model.
Determination termination of the panel regression model estimation method is done using three approaches, namely Common Effect Model, Fixed Effect Model and Random Effect Model. The selection of the most appropriate panel regression estimation method can be done by conducting several tests. Based on the regression model I after conducting the Chow test with the value of prob. 0.000, then the Fixed Effect Model model is selected, the next process is to perform the desired test results obtained by prob. 0,000 then the Fixed Effect Model model is selected and ignores the Lagrange Multipllier test.
Cross-Section Random 61.379938 0.000 Based on the regression model II after conducting the Chow test with the value of prob. 0.000, then the Fixed Effect Model model is selected, the next process is to perform the desired test results obtained by prob. 0,000 then the Fixed Effect Model model is selected and ignores the Lagrange Multipllier test.

Hypothesis Test
Test (t) is carried out to find out individually (partial) independent variables that influence the dependent variable significantly or not. If the significance level (Sig t) is smaller than α = 0.05 and t arithmetic ˃ t table then Ha1 is accepted. Can be seen in the following table: Following is the equation in Table 6. Model I as follows: The level of confidence of 95% or (α) = 0.05, profitability has a positive coefficient and obtained t-count of -0.926185 smaller than t table that is equal to 1.97669 with a significance level of 0.3558 > 0.05, this means ROE has a statistically negative effect and no significant effect on RDS so that Ha1 is rejected.
Firm size variable has a positive coefficient value and obtained t-count of 2.469036 is greater than t-table that is 1.97669 with a significance level of 0.0147 < 0.05, this means that firm size is statistically positive and significant effect on RDS so that Ha1 is accepted.
The DER variable has a positive coefficient value and the obtained t-test of 0.695594 is smaller than the t-table of 1.97669 with a significance level of 0.4878 > 0.05, this means that the statistically positive DER has no effect and is not significant to RDS so that Ha1 is rejected.
PER variable has a positive coefficient value and obtained t-count of -0.819515 smaller than t-table that is equal to 1.97669 with a significance level of 0.4138 > 0.05, this means that statistically negative PER has no effect and is not significant to RDS so that Ha1 is rejected. Multiple regression equations with moderation variables. If the moderated variable constant is positive, then moderating variables strengthen the influence of the independent variables on the dependent variable and vice versa (Astakoni et al., 2019). Therefore, the above equation has the following meanings: The ROE * DPR variable constant is -0.467. So that the DPR variable can weaken the effect of the profitability variable on RDS. Therefore Ha5 was rejected, where the Parliament could moderate, namely by strengthening the effect of profitability on return defensive stocks.
The SIZE * DPR variable constant is 0.034. So that the DPR variable can strengthen the effect of firm size variable on RDS. Therefore Ha6 is accepted, where the Parliament can moderate, namely by strengthening the effect of market size on return defensive stocks.
The DPR * DER variable constant is -0,353. So that the DPR variable can weaken the influence of the leverage variable on RDS. Therefore Ha7 is accepted, where the Parliament can moderate, that is by weakening the influence of leverage on return defensive stocks.
The PER * DPR variable constant is -0,324. So that the DPR variable can weaken the effect of the price earning ratio variable on RDS. Therefore Ha8 is accepted, where the DPR can moderate, namely by weakening the effect of the price earning ratio on return defensive stocks.

Effect of Profitability on Return Defensive Stocks
Based on the results of the t statistical test, it has been found that the profitability variable has a negative and not significant effect on return defensive stocks. This shows that there is no influence on the amount of profitability of a company on the value of return defensive stocks. The results of this study are following research conducted by Novy-Marx (2014) and Sihombing (2019) which states that profitability has a negative and not significant effect on return defensive stocks.
Companies that have high equity do not always produce high profits, but if the profits are large or small, they will not necessarily be allocated to stock returns. Low profit might be allocated to cover the company's debt interest (RM Musallam, 2018). Such conditions can be caused by external factors, namely declining economic conditions and internal factors of the company.

Effect of Firm Size on Return Defensive Stocks
Based on the results of the statistical t-test, it has been found that the firm size variable has apositive and significant effect on return defensive stocks. This shows that the higher firm size owned by a company can have an impact on increasing return defensive stocks. The results of this study are following research conducted by Acheampong et al. (2017) and Arslan and Zaman (2014) which states that firm size has a positive and significant effect on return defensive stocks.
The size of the firm size describes the growth rate of a company. The greater firm size of a company illustrates the company's better per-formance so that it will produce high earnings and be able to provide higher returns (Putri, 2016).

Effect of Leverage on Return Defensive Stocks
Based on the results of the t statistical test, it has been found that the leverage variable has a positive and not significant effect on return defensive stocks. This shows that there is no influence on the amount of leverage owned by a company on the value of return defensive stocks. The results of this study are consistent with research conducted by Putri (2016) and Sakina and Prasetiono (2017) which states that leverage has a positive and not significant effect on return defensive stocks.
To increase company stock returns, we can refer to the optimal level of debt of the company, the optimal level of debt will move from time to time . The company has access to external funding (debt) that can gain high benefits such as maximizing company wealth, growth opportunities and possible investment returns, generally increasing the value of the company which impact on stock returns. However, this can lead to dangerous situations such as financial risk (bankruptcy) (Khémiri & Noubbigh, 2018). Thus it can be said if DER has a positive effect on return defensive stocks, but this will not necessarily attract investors to invest their capital. So it can be concluded that leverage does not affect the size of the return generated.

Effect of Price Earning Ratio on Return Defensive Stocks
Based on the results of the t statistical test, it has been found that the variable price earning ratio has a negative and not significant effect on return defensive stocks. This shows that there is no effect of the magnitude of the price-earnings ratio owned by a company on the value of return defensive stocks. The results of this study are consistent with research conducted by Darusman and Prasetiono (2012) and Antara (2012) which states that the price-earnings ratio has a negative and not significant effect on return defensive stocks.
Companies that have a high PER are usually avoided by investors because their shares are considered expensive, causing a decline in share prices and subsequently will have an impact on increasing stock returns (Fama & French, 1992). The results of this study can be interpreted that the price earning ratio is not the main factor seen by investors in investing, but also looks at other factors such as investor psychology and macroeconomic factors.

Dividend Policy Weakens the Effect of Profitability on Return Defensive Stocks
Based on the results of the second regression equation analysis test obtained a negative value for the constant variable interaction between profitability and dividend policy. This shows that the dividend policy variable weakens the effect of profitability on return defensive stocks. The results of this study are consistent with the research Astakoni et al. (2019) which states that dividend policy weakens the effect of profitability on return defensive stocks.
Companies with small profitability can distribute dividends, but it is difficult to maintain even cannot distribute dividends for the next period (Fajaria & Isnalita, 2018). However, stock returns will remain high even though it is not influenced by high profits and dividend distribution. This is because the company's policy that forces to continue to distribute dividends in the condition of the company and the market is not good. So investors are reluctant to invest their funds in the capital market.

Dividend Policy Strengthens the Effect of Firm Size on Return Defensive Stocks
Based on the results of the second regression equation analysis test obtained a positive value for the constant variable firm size interaction and dividend policy. This shows that the dividend policy variable can strengthen the effect of firm size on return defensive stocks. The results of this study are consistent with the research Situmorang (2017) which states that dividend policy moderates and strengthens the effect of firm size on return defensive stocks.
Companies with high dividend distribution intensity with large company size have high stock returns. So this was responded by investors as a positive signal to buy shares. Hunjra et al., (2014) clarify the relationship and conclude that dividend policy has a significant positive impact on return defensive stocks. Larger companies are considered to get greater capital and profits, including stock returns that will be obtained.

Dividend Policy Weakens the Effect of Leverage on Return Defensive Stocks
Based on the results of the second regression equation analysis test obtained a negative value for the constant variable interaction between leverage and dividend policy. This shows that the dividend policy variable weakens the effect of leverage on return defensive stocks. The results of this study are consistent with the research Fajaria and Isnalita (2018) which states that dividend policy weakens the effect of leverage on return defensive stocks.
One of the funding that can be obtained is through debt and retained earnings. This makes an increase in the amount of corporate leverage, but it is still responded positively by investors because it is considered the company is increasing debt for company expansion. Even though the company distributes dividends to shareholders and leverage increases, stock returns can continue to rise because of the influence of other factors and the formation of market prices into the added value of the company caused by psychological factors (Tahu & Susilo, 2017).

Dividend Policy Weakens the Effect of Price Earning Ratio on Return Defensive Stocks
Based on the results of the second regression equation analysis test obtained a negative value for the constantly variable interaction between the price earning ratio and dividend policy. This shows that the dividend policy variable weakens the effect of the price earning ratio on return defensive stocks. The results of this study are consistent with the research Arslan and Zaman (2014) which states that dividend policy weakens the effect of price earning ratio on return defensive stocks.
Companies with an increase in dividend policy and price earning ratio tend to decrease the impact of decreasing stock returns because investors believe that rising stock prices are followed by rising company values (Husnan, 2009). However, this does not make stock returns upward because other factors such as investor psychology and macroeconomic factors also need to be considered.

CONCLUSIONS AND RECOMMENDATIONS
This study was conducted to examine the effect of profitability, firm size, leverage, price earning ratio on return defensive stocks with dividend policy as a moderating variable. The test results found that the firm size variable has a significant positive effect while the leverage variable has a significant positive effect. Variable profitability and price earning ratio are not significant negative effect. The MRA test proves that the dividend policy variable strengthens the effect of firm size and return defensive stocks. Whereas dividend policy weakens the effect of profitability, leverage and price earning ratio on return defensive stocks.
Suggestions for further research are expected to add macroeconomic variables such as interest rates and investor psychology. The limi-tation of this study is that several companies do not issue financial reports regularly during the research period, so the results obtained are less than optimal.