Determinants of Corporate Value and Corporate Social Responsibility

This study was designed using the profitability variable as the independent variable, and firm value as the dependent variable, while corporate social responsibility (CSR) as the moderating variable. Basically, the value of the company is one aspect that is quite important in the world of investment. Because company value can be interpreted as the price a prospective buyer is willing to pay if the company is sold . The significant effect of profitability on firm value is in line with the signal theory. Profitability is a signal in the form of information stating that the company is better than other companies. Profitability is also able to reduce information asymmetry, because profitability is reliable financial information and will reduce uncertainty about the company's future prospects. Corporate Social Responsibility (CSR) has a positive effect on the relationship between profitability and firm value in manufacturing companies in the period 2013-2015. This shows that the existence of CSR is still able to strengthen the relationship between the effect of profitability on firm value


INTRODUCTION
A company should adhere to a business ethic, which is not only sticking to the number of companies that have been successfully developed, but must also stick to how to empower the community in the surrounding environment as wisely as possible, because after all the company not only has an economic obligation but also has ethical obligations. On the other hand, the assessment of a company is also measured by how much profit is generated by the company, by fulfilling the amount of profit that is according to the target, the company will automatically fulfill its obligations for its funders and can also show the company's prospect image in the future. That way the company is obliged to determine the profit target that must be generated by a company. A high profit will give an indication of the company's good prospects so that it can trigger investors Susanti (2010) shows that the results of profitability have a significant positive effect on firm value. Yuniarsih and Wirakusuma's research (2007) shows that profitability has a significant positive effect on firm value. Research by Carningsih (2009) shows that the results of profitability have a significant negative effect on firm value. The research of Zuradah (2010) shows that the results of profitability have a significant negative effect on firm value. Rahayu's research (2010) shows that the results of profitability do not have a significant effect on firm value. The number of differences in the results of the research that has been done shows the inconsistency of the research that has been done.
The effect of profitability on firm value will be stronger if the company has achieved profitability as targeted, the company must have a strategy or steps that have been taken by the company with company policies through company activities in paying attention to the surrounding environment or empowering the community around the company as the company's object in increase the company's value image. Corporate social responsibility is a process of communicating the social and environmental impacts around the company on the economic actions taken by the company towards certain groups in society and on society as a whole. This expands the company's responsibility in providing financial reports to owners of capital, especially shareholders. That way, the company's responsibility is not only to seek profit for shareholders, but also to provide social responsibility reports to the community.
According to Kuntari and Sulistyani (2007), there are three approaches in social performance reporting, namely: Social Audit . The social examination measures and reports the economic, social and environmental impacts of the social-oriented programs of the company's operations. Social examination is carried out by making a list of company activities that have social consequences, then the social auditors will try to estimate and measure the impacts caused by these activities.
Various alternative report formats for presenting social reports have been proposed by academics and practitioners. The approaches that can be used by companies to report their social responsibility activities are summarized by Dilley and Weygandt 24 into four groups as follows (Henry andMurtanto, 2001 in Kuntari andSulistyani, 2007): a) Inventory Approach . The company compiles and discloses a comprehensive list of corporate social activities. This list must include all of the company's social activities, both positive and negative. b) Cost Approach . The company makes a list of the company's social activities and discloses the amount spent on each of these activities. c) Program Management Approach . The company not only discloses social responsibility activities but also the objectives of these activities as well as the results that have been achieved by the company in accordance with the stated objectives. d) Cost Benefit Approach . The company discloses activities that have social impacts and the costs and benefits of these activities. The difficulty in using this approach is that it is difficult to measure the social costs and benefits that companies impose on society.
Social Disclosure in Annual Report Social disclosure is the disclosure of information about 25 companies' activities related to the company's social environment. Social disclosure can be made through various media, including annual reports, interim reports, prospectuses, announcements to stock exchanges or through mass media. Companies tend to disclose information relating to their activities and the impact caused by the company. Gray, et al., in Florence, et al., (2004) mentioned three studies, namely: a) Decision Usefulness Studies. Belkaoui (1989) in Anggraini (2006) suggests that companies that carry out social activities will disclose them in financial statements. Some of the studies conducted by researchers expressing this opinion find evidence that social information is needed by users of financial reports. Analysts, bankers and other parties involved in the research are asked to rank accounting information. The accounting information is not limited to traditional accounting information that has been valued so far, but also other information that is relatively new in the accounting discourse. They place information on the company's social activities in a moderately important position. b) Economic Theory Studies. This study uses agency theory which analogizes management as an agent of a principal. Typically, principals are defined as shareholders or other traditional users. However, the definition of principal 26 extends to the entire interest group of the company concerned. As an agent, management will try to operate the company in accordance with the wishes of the public. c) Social and Political Theory Studies. Studies in this field use stakeholder theory, organizational legitimacy theory and political economy theory. Stakeholder theory assumes that the company's existence is determined by the stakeholders. Social disclosures made by companies are generally voluntary, unaudited and unregulated (not influenced by certain regulations). Darwin (2004) in Anggraini (2006) states that Corporate Social Responsibility is divided into 3 categories, namely economic performance, environmental performance and social performance.
Through a CSR disclosure strategy, a company will gain social legitimacy and maximize its financial strength in the long term (Kiroyan, 2005 in Yuniasih andWirakusuma, 2007). Because lately many companies have realized the importance of implementing CSR as a business strategy in achieving a positive image among investors and the public. On july 20, 2007 the government also passed Law No. 40 of 2007 on Limited Liability Companies which regulates the company's obligation to implement Corporate Social Responsibility. With the enactment of the PT Law, it is hoped that it can increase the extent of CSR disclosure by companies because CSR was originally Voluntary to become Mandatory for companies.
Given that when financial strength increases, the level of profitability will be even higher. If the level of profitability is higher, the legitimacy obtained by the company will have an impact on the increase in firm value. This can be done if the company implements CSR and discloses it, so that stakeholders know that the company is able to carry out its ethical obligations that have implemented CSR. This CSR disclosure is very useful for both parties, both the company and the surrounding environment. Where the company will get a good image among the public and investors, the public can get something useful both material and non-material, so that the company's shares will be more attractive to investors. If many company shares are interested indirectly, the demand for shares by investors will be higher, and when the demand for shares increases, the company value will increase.
Based on the description above, the researcher designed the problem formulation, with the following questions: 1. Does profitability have a significant effect on firm value? and is CSR able to moderate the effect of profitability on firm value?. This study also aims is to determine whether the effect of profitability on the value of the company and to examine whether disclosure of CSR able to moderate influence on the value of the company's profitability.

METHOD
This study was designed using the profitability variable as the independent variable, and firm value as the dependent variable, while corporate social responsibility as the moderating variable. This research is included in the category of quantitative research, where this study emphasizes theory testing through measuring the research variables by numbers and analyzing data using statistical procedures. This study uses a deductive approach that aims to test the hypothesis (Paramita, 2015;6). The object of this study is the profitability of the company against firm value. By using the corporate social responsibility variable as a moderating variable. The research was conducted on manufacturing companies listed on the Stock Exchange Indonesia (BEI) a tender period 2016-2018. The data used in this study are secondary data. Research data is taken from annual reports that have met the criteria and are published. The data sources include: Indonesia Stock Exchange, www.idx.co.id. The data used in this study are annual reports, which include income statements, cash flows, CSR disclosure reports, and notes on financial statements.
The population of this research is manufacturing companies listed on the stock exchange and disclose social responsibility reports both within the framework of the annual report or separately within a period of 2016 -2018. This study used purposive sampling/judgment sampling technique, which means that sampling is based on certain criteria. The criteria used can be based on a balance (judgment) or based on a certain quota. Shile the purpose of sampling the sample is determined by using the following criteria: 1.Manufacturing companies that have been listed on the IDX in 2016, which publish annual reports in a row. 2.The sample companies carried out their CSR disclosures in their annual reports during 2016-2018. 3.The company has published its annual report ending on 31 December of the period 2016-2018. 4.The company has the data or ratio required in the research. 5.The sample company has all the required data in full. The technique used in this study uses documentation techniques, and the researcher studies all the records needed in the annual report of the companies that are the research samples such as CSR, ROA, mentioning the price of outstanding shares, and other necessary data. The variables used in this study were divided into three parts, namely the dependent variable and the independent variable and the moderating variable. Profitability. According to Irawati (2006), which states that: profitability ratios is a ratio used to measure the efficiency of the use of company assets or is the ability of a company to generate profits for a certain period (usually semester, quarterly, etc.) to see the ability companies in operating efficiently.
Company Value. According to Sujoko and Soebaiantoro (2007) in Hermuningsih (2009), company value is an investor's perception of the level of cleanliness of a company which is closely related to its share price. A high stock price makes the company value high, and increases market confidence not only in the company's current performance but also in the company's future prospects. The share price used generally refers to the closing price, and is the price that occurs when the stock is traded on the market. Corporate Social Responsibility . The initial concept of CSR originated from Howard R. Bowen in 1953 with the definition that CSR is an obligation or social responsibility of the company based on alignment with the objective objectives and values of a society. The value of the company q=((EMV+D))/((EBV+D)) x 100%. Profitability ROA= ((Laba brsih)/(total aktiva)) x 100 %. Corporate social responsibility CSRDI = Xj/Nj. Descriptive statistics are used to determine the level of disclosure of Corporate Social Responsibility (CSR), firm value, profitability in manufacturing companies listed on the IDX. The disclosures used in this study are minimum, maximum value, mean and standard deviation. This classical assumption test aims to determine and test the feasibility of the regression model used in this study. This test is also intended to ensure that the regression model used does not contain multicollinearity and heteroscedasticity and to ensure that the resulting data contributes normally (Ghozali, 2006).
The heteroscedasticity test aims to test whether in the regression model there is an inequality of variability from one observation to another. If the residual variance from one observation to another observation is still called homoscedasticity, and if it is different it is called heteroscedasticity. A good regression model is one that has homoscedasticity or heteroscedasticity does not occur (Ghozali, 2009). The way to detect the presence or absence of heteroscedasticity is by looking at the plot graph between the predicted values of the related (dependent) variable, namely ZPRED and the residual SRESID. Detection of the presence or absence of a certain pattern on the scatterplot graph between SRESID and ZPRED where the Y axis is the predicted Y, and the X axis is the residual (Y prediction -real Y) that has been studentized. Basic analysis: a. If there is a certain pattern, such as the dots that form a certain pattern (wavy, widened then narrowed). Then it indicates that there has been heterocdasticity. b. If there is no clear pattern, as well as dots that spread above and below the number 0 on the Y axis, there is no heteroscedasticity.
Test normali ta s aims to test whether the regression model, or residual confounding variables have a normal distribution. As is well known, the 1 and F tests imply that the residual value follows a normal distribution. If this assumption is violated then the statistical test is invalid for a small sample size. In principle, normalists can be detected by looking at the distribution of data (points) on the diagonal axis of the graph or by looking at the histogram of the residuals. Basic decision making: 1) If the data spreads around the diagonal line and follows the diagonal line or the histogram graph shows a normal distribution pattern, the regression model fulfills the normality assumption. 2) If the data spreads far from the diagonal and / does not follow the direction of the diagonal line or the histogram graph does not show a normal distribution pattern. So the regression model does not meet the normalist assumptions. Normalized test with graphs can be misleading if you are not careful visually it looks normal, even though statistically it can be reversed. Therefore it is recommended that in addition to the graphical test it is equipped with statistical tests. Another statistical test that can be used to test for residual normality is the Kolmogorov-Smirnov (KS) nonparametic statistical test. The KS test is done by making a hypothesis H 0 : Residual data are normally distributed, H a : Residual data are not normally distributed Multicollinearity test cannot be used in simple linear regression analysis. Test Autokolerasi test does not need to be applied to cross sectional. The autocorrelation test aims to find out whether there is a correlation between members of a series of observational data sorted according to time or space. Autocorrelation is a correlation between one observation disorder variable and another observation disorder variable.
The data that had been collected were analyzed using statistical analysis tools, namely: Simple linear regression analysis ( Simple Regression analysis ) Y = a + B 1 X 2 + e, Multiple linear regression analysis ( multiple regression analysis) Y = a + B 1 X 1 + B 2 X 2 + B 3 X 1 X 2 + e. The extraction test or often referred to as Moderated Regression Analysis (MRA) is a special application of linear multiple regression where the regression equation contains an element of interaction (multiplication of two or more independent variables (Ghozali, 2006). The multiplication variable between CSR (X 2 ) and Profitability (X 1 ) is a moderating variable because it illustrates the moderating effect of the profitability variable (X 1 ) on the relationship between CSR (X 2 ) and Company Value (Y) Regression analysis is basically a study of the dependence of the dependent variable with one or more independent variables, with the aim of estimating and / or predicting the population average or the average value of the dependent variable based on the value of the known variable (Guharati, 2003in Ghozali 2006. According to Ghozali (2006), the accuracy of the sample regression function in estimating the actual value can be measured from its goodness of fit. Statistically, at least this can be measured from the coefficient of determination, the value of the F statistic and the value of the t statistic. Statistical calculations are called statistically significant if the value of the static test is in a critical area (areas where H 0 is rejected). Conversely, it is called insignificant if the statistical test value is in the area where H 0 is accepted.
The coefficient of determination (R 2 ) essentially measures how far the model's ability to explain variations in the dependent variable. The coefficient of determination is between zero and one. The value of R 2 small means the ability of independent variables in explaining the variation is very limited dependent variables. A value close to one means that the independent variables provide almost all the information needed to predict the variation in the dependent variable.
The fundamental weakness of using the coefficient of determination is the bias towards the number of independent variables included in the model. Each additional one independent variable, then R 2 is definitely increasing no matter whether these variables significantly influence the dependent variable. Therefore many penliti recommend using niali Adjusted R 2 , adjusted R 2 can go up or down when the independent variable is added to the model. The F test is carried out to test whether the regression model used is fit. The basis for making decisions is: If F-count <F-table, then the regression model is not fit (hypothesis is rejected), If F-count> F-table, then the regression model is fit (hypothesis is accepted). The F test can also be done by looking at the significance value of F at the output of the regression results using SPSS with a significance level of 0.05 ( a = 5%). If the significance value is greater than a , the hypothesis is rejected, which means that the regression model is not fit. If the significance value is smaller than a , the hypothesis is accepted, which means the regression model is fit. The t statistical test is carried out to show how far the influence of one independent variable individually is in explaining the variation in the dependent variable. The basis for making decisions is: If tcount <t-table, then the independent variable individually has no effect on the dependent variable (hypothesis is rejected), If t-count> t-table, then the independent variable individually affects the dependent variable (hypothesis is accepted). The t test can also be done by looking at the significance of t of each variable in the regression output using SPSS with a significance level of 0.05 ( a = 5%). If the significant value is greater than a , the hypothesis is rejected (the regression coefficient is not significant), which means that individually the independent variable does not have a significant effect on the dependent variable. If the significance value is less than a , the hypothesis is accepted.

RESULTS AND DISCUSSION
Of the companies engaged in the manufacturing sector, the research focus is on annual financial reports. From these annual financial reports, the objects of this study are profitability, firm value, and corporate social responsibility (CSR) which can be found in company reports. The following table will present the manufacturing company data sampled in this study.  The classical assumption test is a statistical requirement that must be met in multiple linear regression analysis based on ordinary least square (OLS). Thus regression analysis that is not based on OLS does not require the requirements of classical assumptions, such as logistic regression or ordinal regression. In linear regression analysis, it is necessary to test classical assumptions which aim to avoid the possibility of deviations from classical assumptions. There are four classical assumption tests that will be carried out. The results of the normality test can be seen from the normal probability plot graphic image below. It should be reminded that the normality assumption referred to in the classic assumption of the OL approach is (data) residuals formed by linear regression models that are normally distributed, not the independent variables or the dependent variable . The criteria for a (data) residual that is normally distributed or not with the Normal PP Plot approach can be done by looking at the distribution of points in the figure. If the distribution of the points is close or dense in a straight line (diagonal) it is said that the residual (data) is normally distributed, but if the distribution of the points is away from the line then it is not normally distributed. From the figure, the SPPS results show that the distribution of data points is close to the line, which means that the residual (data) is normally distributed.
Heteroscedasticity is done by making a Scatterlot (distribution flow) between the residual and the predicted value of the dependent variable which has been standardized. The results of the heteroscedasticity test can be seen in the Scatterplot image . From the figure, it can be seen that the point distribution does not form a certain pattern, so it can be concluded that there is no heteroscedasticity or in other words there is homoscedasticity. The classical assumption about heteroscedasticity in this model is fulfilled, which is free from heteroscedasticity. This test (scatterplot) is prone to errors in drawing conclusions. This is because the determination of whether there is a pattern / plot on the drawn tag points is very subjective. It could be that some people say there is no pattern, but some others say there is a pattern. There is no definite measure when a scatterplot forms a pattern or not. Decisions rely solely on the observation or sight of the researcher.
The autocorrelation test aims to find out whether there is a correlation between members of a series of observational data sorted according to time or space. Autocorrelation is a correlation between one observation disorder variable and another observation disorder variable. Autocorrelation often appears in time series data . Autocorrelation arises because lucky observations over time are related to one another. Autocorrelation can be detected through the Durbin-Waston (DW) method on the basis of the decision making of the Watson Durbin test by looking at the following table:  Multicollinearity test aims to test whether in the regression model that is formed there is a high or perfect correlation between the independent variables. Multicollinearity is a linear relationship between independent variables in multiple regression. A good regression model should not have a correlation between the independent variables. Several methods can be used to perform the multicollinearity test . One of them is the multicollinearity test by analyzing the calculation of the tolerance value and variance inflation factor (VIF). A low tolerance value is the same as a high VIF value (because VIF = 1 / tolerance ). Value cutoff commonly used to indicate the presence of multicollinearity is value toleranc e < 0.10 or equal to 10. The results of the VIF VIF is less than 10 so that it can be concluded not happen multicoloniarity . The multicollinearity test results can be seen in the Coefficients table for the last two columns. Based on the classical assumption test that has been carried out, it can be seen that the data in this study are normally distributed and there is no heteroscedasticity. Therefore, the available data fulfills the requirements to use a simple regression model and moderated regression analysis (MRA). Simple regression analysis and moderates regression analysis (MRA) are used to determine the extent to which the relationship between the independent variables and the dependent variable can be found in the following table: Hypothesis 1 testing states that : profitability has a significant effect on firm value. Based on table 4.4, the regression coefficient test above explains that the profitability variable shows the sig value. of 0.001, this value is less than the value = 0.05 or p-value 0.001> 0.05. So it can be concluded that the profitability variable has a positive effect on firm value. Thus H 1 is accepted. Testing the hypothesis 2 states that : Based on the results Moderated Regression Analysis (MRA), table 4.7 test regression coefficient above me n explain that the variable CSR (moderation) shows the value of R Squere first test at 0,071 and the value of R squre both show the value of 0.111, which means in testing the first and second regressions the value of R Square increases. It can be concluded that the CSR variable has a positive effect on the relationship between profitability and firm value. This shows that CSR can increase firm value when company profitability is high. Thus H 2 is accepted.
The linear test results show that the profitability as measured by using Return On Assets (ROA) has a significant effect on firm value. Thus this study supports the first hypothesis (H1) which states that a significant effect of profitability on firm value is in line with signal theory (signaly theory). Profitability is a signal in the form of information stating that the company is better than other companies. Profitability is also able to reduce information asymmetry, because profitability is reliable financial information and will reduce uncertainty about the company's future prospects, and if the company is able to provide a solid expectation of the value (results) in the future it will cause the company to be assessed. community level (Indriyo, 1984).
The results of this study support Utaminingsih's (2014) study entitled the effect of profitability on firm value with the extent of CSR disclosure as a moderating variable. This study uses objects in all companies included in LQ45 during the 2009-2011 period. The sampling method was carried out by purposive sampling with the criteria included in the LQ45 during 2009-2011 consecutively and using the rupiah as the unit in the financial reports so that 24 companies were used as research samples. The data analysis technique used moderated regression analiysis . The results obtained are that profitability has a significant effect on firm value and the extent of CSR disclosure is not a moderating variable in the effect of profitability on firm value.
The regression test results show that the Corporate Social Responsibility (CSR), which is measured using RSDI, has a significant effect on firm value. In other words , CSR can increase firm value when company profitability is high. The impact of CSR on the relationship between profitability and Company Value is caused by the development of manufacturing companies starting in 2014 in implementing CSR activities. And this shows that the company is stingy / economical, which means the company has high profitability but low CSR activities (Suharto , 2007).
Thus this study accepts the second hypothesis (H2) which states that the company's ability to provide high expectations of future value (results) causes the company to be highly valued by the community. However, a research gap found in several previous studies indicates that there are other factors that influence investors in assessing companies. CSR is expected to influence the relationship between profitability and firm value. In order to increase corporate value in a sustainable manner, companies must pay attention to the economic dimension (Kusumadilaga, 2010). A high level of profitability does not always guarantee an increase in the value of a company. According to Bowman & Haire (1976) and Preston (1978) in Agustine (2014), the higher the level of company profitability, the greater the disclosure of social information by the company. This is because people currently tend to choose companies that are responsible and care for the surrounding environment because by supporting these companies, the community indirectly participates in maintaining the surrounding environment (Susanti and Santoso, 2011). In addition, companies that care about the environment are considered to pay more attention to the prospects for the company's future performance so that investors will assess them positively. Therefore, companies with a high level of profitability will always try to increase the disclosure of social activities carried out by the company in an effort to convince investors that the company not only pays attention to short-term goals (profit), but also long-term goals, namely increasing company value (Yuniasih and Wirakusuma, 2007). So it can be concluded that CSR increases firm value when company profitability increases. Based on this description, Corporate Social Responsibility can moderate the effect of profitability on firm value.
This study supports the research of Hermawan (2014) in his research entitled the effect of financial performance on firm value with the disclosure of corporate social responsibility as a moderating variable, in this study researchers used objects in food and beverage companies listed on the IDX. The data analysis of this study used multiple regression methods to determine the variables involved in the study. The result of the research is that partially, the financial performance variable (return on assets) has no significant effect on the company value. Furthermore, partially the corporate social responsibility variable is able to moderate the performance relationship

CONCLUSION
From the results of data research and discussion, the following conclusions are obtained ; Profitability has a significant effect on firm value in manufacturing companies in the 2013-2015 period. This shows that profitability is an important element of the company's value element. With stable profitability, investors will see the positive side of financial performance that affects firm value. The value of the company whose measurement is measured from the share price divided by the book value per share is also inseparable from the size of the company's profitability. Corporate Social Responsibility (CSR) has a positive effect on the relationship between profitability and firm value in manufacturing companies in the period 2013-2015. This shows that the existence of CSR is still able to strengthen the relationship between the effect of profitability on firm value.