THE ROLE OF CORPORATE VALUES TO CORPORATE SOCIAL RESPONSIBILITY AND DIVIDEND POLICIES THAT AFFECT STOCK RETURNS

The purpose of this study is to analyze whether firm value mediates the effect of Corporate Social Responsibility (CSR) and dividend yield on stock returns. The samples are 12 companies; they chosen by purposive sampling method. The data is processed by panel regression and Sobel Test using eViews 9.0. The results showed that firm value is mediating the effect of Corporate Social Responsibility (CSR) to stock return. The second result is firm value isn’t mediating the effect of Dividend Yield to stock return.


THEORETICAL REVIEW
In this research, there are four variables, CSR and Dividend Yield are independent variables, company value is a mediating variable, and stock return is the dependent variable. The following is the framework of this research: Corporate CSR is expected to provide long-term benefits, positive relationships with stakeholders must be maintained to minimize conflicts that will occur (Gitman & Zutter, 2015). According to Lins, et al (2017) with companies implementing CSR activities that are beneficial to the environment can increase investor confidence in the company, the company's image will look better in the eyes of the public and will produce something positive. At present the responsibility of the company must rest on the triple bottom lines, namely corporate responsibility on social, environmental and financial aspects so that each company is required to disclose information about corporate social responsibility or Corporate Social Responsibility (Hartoyo, 2016). To measure the Corporate Social Responsibility Index used categories derived from the GRI (Global Reporting Initiative) indicator. = Dividend Yield. Investors want a relatively stable dividend distribution because it reduces uncertainty about the expected results of investments made and can also increase investor confidence in the company so that the value of shares also increases. Dividend yield is one indicator that is often used by securities analysis to measure the performance of a company in distributing profits on income of shares traded in the capital market. The yield dividend in this study can be formulated as follows: Company Value. To measure the value of a company in this study, Tobin's Q method is used as a valuation ratio (Valuation Ratio), which is the most comprehensive measure to assess the work of a company, because the ratio reflects a combination of risk-ratio effects and return-yield ratios. If the value of Tobin's Q is more than one company, this will stimulate investment, whereas if the value of Tobin's Q is less than one, it will reduce investment. Tobins'Q calculations are as follows: Stock Return. Based on the two concepts by Tandelilin (2011), the total return of an investment can be calculated by using yield and capital gains obtained from an investment. In this study, the researcher calculates stock returns by using actual return, which is the difference between the current stock price and the stock price in the previous period or can be formulated as follows: Where: R it = Stock return in the t period; P t = Stock price in t period; P t-1 = stock price before t period.

METHODS OF RESEARCH
The population in this research includes listing companies on the Indonesia Stock Exchange (IDX) during the 2012-2016. The sampling method used by the author in this research was using purposive sampling method. Samples taken from populations that have met the criteria there are 12 companies. This research will use panel data regression because the data consists of time series and cross section.
In estimating panel data regression there are three models, namely Common Effect (CE), Fixed Effect (FE), and Random Effect (RE). To choose the best model, Chow test, Huasman test, and Lagrange test were conducted. In addition to conducting panel data regression, in this study a Sobel test was conducted to determine whether the company value variable mediates the effect of CSR variables and yield dividends on stock returns. The first step is by calculating the indirect coefficient value, after that it looks for the standard error value from the indirect effect coefficient.
If the result of t count is greater than t table with a 0.05 level of significance, it can be concluded that the mediation coefficient is significant which means that the mediating variable influences the influence of the independent variable on the dependent variable.

RESULTS AND DISCUSSION
To prove the first hypothesis is whether the company value variable mediates the effect of CSR on stock returns, regression and Sobel Test are carried out with the following results: Because the value of t count of 2.0017 is greater than the value of t table of 1.67155, then there is a mediating effect of the company value variable in the effect of CSR on stock return. CSR variables directly do not affect stock returns, but indirectly CSR influences stock returns by mediating firm value variables. CSR has a significant positive effect on company value. The greater the CSR activities carried out by the company, the better the company's reputation in the eyes of the public and investors. This causes an increase in the value of the company. The results of this study support the results of Chen & Lee (2017) research, namely that the higher the company invests in the form of CSR will increase institutional shareholders, and develop the size of the company which results in the growth of profits and increasing corporate value. The value of the company has a significant positive effect on stock returns, the higher the value of the company, the higher the investors value the company so that investors want to buy shares at a higher price. So the increasing value of the company will cause an increase in stock returns. This result is different from the results of the research from McMillan (2017), which is that firm value has a significant negative effect on stock returns and Rahman & Mustafa (2018) states that firm value does not significantly influence stock returns.
Next, to prove the second hypothesis is whether the company value variable mediates the effect of the Dividend Yield on stock returns, Sobel Test is carried out with the following results:

CONCLUSION AND RECOMMENDATIONS
The value of the company mediates the effect of CSR on stock returns. CSR variables directly do not affect stock returns, but indirectly CSR influences stock returns by mediating firm value variables. CSR has a significant positive effect on company value. Dividend yield directly affects stock returns, and firm value does not mediate the effect of dividend yield on stock returns. Dividend yield has a significant negative effect on stock returns. From the results of this study, the authors recommend to the issuers to pay attention to the activities of CSR and dividends which are costs for the company because it affects stock returns.