Effect of Corporate Governance on Corporate Social Responsibility and Firm Values: Empirical Evidence From Manufacturing Companies in Indonesia

. Corporate governance (CG) and corporate social responsibility (CSR) are no longer voluntary. However, it has become a mandatory for every company, especially in increasing the value of the company, although there are still many pros and cons. This study aims to investigate the effect of CG practices on CSR as well as the influence of CG and CSR on firm value. The object of this research is manufacturing companies (Basic Industry and Chemical and Consumer Goods Industry Industry) which are listed on the Indonesia Stock Exchange in 2010-2018. Based on the process of sample selection (purposive sampling) 55 companies were obtained so there were 495 observations of research data. There are two research models tested in this study. First, it uses CG and CSR practices as independent variables and company value as the dependent variable. The second uses CG practice as an independent variable and CSR as the dependent variable. In the first model only CSR has a significant effect on the t-sig value of 0,000 (0,000 <0.05). The second model shows that CG has no significant influence on CSR.


INTRODUCTION
Corporate social responsibility has existed since the 1950s, and it remains unclear whether companies should be involved in CSR activities and whether they should be asked to report on these activities, or sustainability reports (Carroll and Shabana, 2010). In Indonesia, in 2007Law No.40 of 2007 concerning Limited Liability Companies (UU PT) was passed, which was passed on July 20, 2007. Article 74 of the Limited Liability Company Law states: (1) a company that carries out business activities in the field of and/or relating to natural resources are required to carry out Social and Environmental Responsibility (TJSL). (2) TJSL is a company obligation that is budgeted and calculated as a company expense whose implementation is carried out with due regard to propriety and fairness. (3) Companies that do not carry out the obligations referred to in paragraph 1 are subject to sanctions in accordance with the provisions of the legislation.
The Lapindo mudflow case caused by the activities of PT Lapindo Brantas, is an example of a case where Corporate Social Responsibility has not been properly implemented. Although PT Lapindo Brantas has tried to compensate the losses caused to the community, the losses borne by the community have not been in accordance with those provided by PT Lapindo (id.wikipedia.org, 2018). While at the international level several cases also occurred such as Mega Corporation, Enron, Worldcom, HIH Insurance, Parmalat Kolap in the early 2000s that received attention related to corporate behavior, including also issues of good governance, ethics, accountability, and trust (Marsiglia and Falautano, 2005). In 2006 the National Committee on Governance Policy issued a general guideline for Good Corporate Governance (GCG) of Indonesia which became the foundation for companies wishing to maintain long-term sustainability of their business within the corridors of applicable business ethics. In addition, Marsiglia and Falautano (2005) corporate governance and CSR initiatives look identical, when corporate governance means being held accountable while CSR means consideration; companies have gradually increased from the philanthropic variants of corporate capitalism to strategies and approaches designed to regain the trust of clients and society in general. Furthermore, companies that have good environmental and social performance (in the sense that will be reflected in CSR disclosures) will certainly be responded positively by investors and will be seen in an increase in share prices. Vice versa, if the company has a bad environmental performance will certainly be responded to negatively by investors who will be seen from a decline in share prices. This was shown by Wijayanti (2012) who found that CSR had a positive and significant effect on firm value.
Good corporate governance practices will also improve market performance better. States that one of the causes of the long process of improvement in Indonesia is caused by the very weak CG that is applied in companies in Indonesia. Furthermore, Suhartati et al. (2011) shows that GCG significantly influences company value.
CNN Indonesia (2016) released the news that the latest research stated that companies in Indonesia have a lower quality of corporate social responsibility (CSR) compared to companies from Thailand. The Research Center for Governance, Institutions, and Organizations of the National University of Singapore (NUS) Business School explained the company's low understanding of CSR practices, resulting in the low quality of the operation of the agenda. The research conducted a study of 100 companies in four countries namely Indonesia, Malaysia, Singapore and Thailand. The research explained that Thailand became the country with the highest quality of CSR implementation with a value of 56.8 out of a total of 100, while Singapore received 48.8. Indonesia and Malaysia each received 48.4 and 47.7 scores (Suastha, 2016).
Furthermore, Sari (2014) found that Bukit Asam Coal (Persero) Tbk and Timah (Persero) Tbk have revealed each performance indicator and CSR disclosure on average has exceeded 75%. However, the indicators and their respective aspects are not yet disclosed in detail. This information shows that it is necessary to identify corporate governance practices for CSR reporting or disclosure as well as their impact on company value. Based on the above problems, researchers are interested in examining the effect of corporate governance practices and corporate social responsibility on corporate value. For this reason, the specific purpose of this study are: To examine the effect of corporate governance practices on company value in Indonesia with manufacturing companies (Basic Industry and Chemical and Consumer Goods Industry) listed on the Indonesia Stock Exchange as case studies. To investigate the effect of corporate social responsibility (CSR) reporting on the value of companies in Indonesia with manufacturing companies (Basic Industry and Chemical and Consumer Goods Industry) listed on the Indonesia Stock Exchange as case studies. To investigate the effect of corporate governance practices on reporting corporate social responsibility (CSR) in Indonesia with manufacturing companies (Basic Industry and Chemical and Consumer Goods Industry) listed on the Indonesia Stock Exchange as case studies

Research Urgency
Rashid (2018) investigated the effect of corporate governance practices on corporate social responsibility reporting in the least developed countries, namely companies registered in Bangladesh. Rashid (2018) concluded that the practice of corporate governance has no influence on corporate social responsibility reporting. These findings, in particular, indicate that CSR disclosure by companies is not responsive to new corporate governance regulations. In previous studies, investigating corporate governance mechanisms influences corporate CSR practices, such as ownership structures (Ghazali, 2007;Dam and Scholtens, 2012), board characteristics (De Villiers et al, 2011;Chelon and Parbonetti, 2012) and corporate governance practices as a whole. general (Haniffa and Cooke, 2005;Rashid and Lodh, 2008;Jo, 2011, 2012;Rao et al., 2012;Khan et al, 2013;Bhaduri and Selarka, 2016). Likewise Harjoto and Jo (2011) find that CSR is positively related to governance characteristics. Ruangviset, Jiraporn, & Kim (2014) find evidence of statistically significant negative impact of CG on CSR.
Furthermore, some research in Indonesia related to corporate governance and CSR practices, Sari (2014) examines the disclosure of CSR in the mining company's sustainability report. The study also analyzes the disclosure of CSR in the company's sustainability report with the standards of the Global Reporting Initiatives (GRI) 3.1. The study was conducted on Bukit Asam Coal (Persero) Tbk. and Timah (Persero) Tbk. in the company's sustainability report for 2012. The results obtained are that both companies have revealed CSR in accordance with GRI3.1. on average it exceeds 75%. However, the disclosure of indicators and their respective aspects is not yet detailed. Ratnasari (2011) examines the effect of corporate governance characteristics of the extent of disclosure of social responsibility within sustainability report. The results obtained by Ratnasari (2011) show that corporate governance does not significantly influence the extent of sustainability report disclosure. Furthermore, this research shows that the leverage variable has a negative and significant effect on the extent of sustainability report disclosure. Wijiayanti (2012) examined the effect of CSR on firm value and the effect of CSR on trading volume. Furthermore Wijayanti (2012) concluded that CSR has a positive and significant effect on firm value with a significance level of 0.007. However, CSR for the trading volume of Wijayanti (2012) did not find a significant effect with a significance level of 0.382. Suhartati et al. (2011) examined the effect of CSR information and GCG practices on firm value. The test shows the results that CSR has no significant effect on firm value. Furthermore, GCG variables, namely the number of independent boards (positive) and board sizes (negative), the control size and sales growth variables significantly influence the value of the company.
From the studies above show that abroad and within the country has begun to do a lot of research conducted on corporate governance practices towards CSR as well as CG and CSR practices on corporate value. Furthermore, previous studies, have not given consistent results regarding GCG and CSR practices on company value.

METHOD
The object of this research is manufacturing companies (Basic Industry and Chemical and Consumer Goods Industry) which were listed on the Indonesia Stock Exchange (BEI) from 2010-2018 and also became a research population. The sampling method uses a purposive sampling method because it uses disclosure data. The criteria used are: Basic Industry and Chemical and Consumer Goods Industry companies listed on the Indonesia Stock Exchange from 2010 to 2018. Providing complete annual reports from 2010 to 2018. Having complete data related to the variables used in research. The type of data used is secondary data that is quantitative data obtained on the company's website, the Indonesia Stock Exchange website and other sources. The data used are annual reports, financial reports, a summary of company performance and also the Indonesian Capital Market Directory (ICMD), and data share prices of manufacturing companies listed on the IDX for the period 2010 to 2018. The research variables used consisted of several types, namely dependent, independent, and control variables. The independent variables are corporate governance and corporate social responsibility (CSR). The dependent variable are firm value and CSR in other models. Control variables are company age (size), company size (size), and company sales growth (sales growth).
This research model uses multiple linear regression which consists of two models. The first model looks at whether CG and CSR affect company value. The second model regresses CG with CSR which is used to answer whether CG has an impact on CSR. The research model can be seen below.

Firm Valueijt CSRIt
The research examines GCG practices towards CSR. In addition, this study examines whether CG and CSR affect company value. Based on the research objectives, the analysis method used is multiple linear regression. However, because the data is arranged in the form of panels, before testing the classical assumptions the selection of the right model is done using the Chow Test (is the fixed effect model better than the common effect) and the Hausman Test (is the fixed effect model from the random effect model) significance of 5%. Next, data cleansing from outlier problems and classical assumptions is carried out consisting of four assumptions namely normality, multicollinearity, autocorrelation, linearity, and heterokedastisitas. To answer the hypothesis, multiple linear analysis is used with regard to F statistics, R squared and significance values.

RESULT
Here are the results of descriptive statistical tests in which there is a description of the acquisition of research data. The number of observational data is 495, the minimum value, maximum value, mean and standard deviation of each are described in the following table 1: Because of data processing uses panel data, the selection of the best model is done first. Here are the results of the chow and hausman test. For the research model with the dependent variable the company value obtained by the chow test results with a value of p = 0,000 or p <0.05, which means the fixed effect model is better than the common effect model. This is also supported by the Hausman test with a value of p = 0,000 or p <0.05, so that the fixed effect model is still selected compared to the random effect model. However, the research model with dependent variable corporate social responsibility obtained chow test results with a value of p = 0,000 or p <0.05. However, in the Hausman test the value of p = 0.746 or p> 0.05 was obtained so that the random effect model was chosen for the model.
The classic assumption tests that need to be fulfilled are normality, multicollinity, heterocedasticity, and autocorrelation.  Normality Test (DV: CSR) Source: processed data As shown in the picture above the histogram normality test of the regression equation shows the shape of the bell and the probability value is 0.1574 (0.1574 > 0.05) for DV: Firm Value and 0.2368 (0.2368 > 0.05) for DV: CSR which shows normally distributed data. Furthermore, in the multicollinearity test there is no Correlation matrix value that exceeds 0.8, so it is free from multicollinearity problems. Next, the heterokedasticity test was performed using the Glejser test. Based on the test results the coefficients of all the independent variables for models one and two are not significant, indicating there is no heterokedastisitas. For the autocorrelation test the Durbin Watson value is 1,063 for model 1 (DV: Company Value) and 0.727 for model 2 (DV: CSR). According to Santoso (2005), the value is between the cut-off value of -2 and +2, so that both models are free from the autocorrelation problem. Based on the table above for model 1 with the dependent variable firm value, shows that the proportion of the value of the independent variable influences the dependent variable based on the adjusted R2 value (Ghozali, 2016). Adjusted R2 obtained by 74.01%, meaning that the independent variable is able to explain the dependent variable by 74.01%. While the rest is influenced by other variables outside the variables studied. Furthermore, the F-sig value was obtained at 0.000, which means the value of p <0.05, indicating that the independent variables namely corporate governance (CG) and corporate social responsibility (CSR) jointly affect the value of the company. Then by using three control variables (age, size, and growth) the hypothesis is proved by the t-sig value of corporate governance (CG) of 0.964 (0.964> 0.05). This shows that corporate governance partially does not significantly influence the value of the company. The indication of the cause is based on descriptive statistics the average disclosure of board governance criteria in the sample company is only 12,608. This value is far from the maximum total disclosure of 20. Meanwhile, the t-sig value of corporate social responsibility (CSR) is obtained at 0,000 (0,000 <0.05), meaning that corporate social responsibility partially has a significant effect on firm value. If seen the coefficient regression constanta value is 10,867 and t-sig value is 0,000, meaning that if the independent variable is zero then the value of the company is 10,867. Furthermore, the CSR regression coefficient value is 2,480, meaning that if the company discloses the CSR it does then the value of the company will rise by 2,480. Furthermore, for model 2 with the dependent variable corporate social responsibility (CSR), the independent variable is able to explain the dependent variable by 3.8% and the rest is influenced by other variables outside the studied variable. Based on the t-sig value. the results obtained by 0.094 (0.094> 0.05), thereby indicating that corporate governance (CG) does not significantly influence corporate social responsibility (CSR). This finding is consistent with Rasyid (2018) and Ratnasari (2011).

CONCLUSSION
Based on the results of hypothesis testing it was found that only corporate social responsibility has a significant influence on firm value. This also shows that the company's value will increase with the disclosure of corporate social responsibility reports. Meanwhile, corporate governance does not have a significant effect on corporate value and corporate social responsibility. This study examines two subsectors from the manufacturing sector, namely Basic Industry and Chemical and Consumer Goods Industry. For further research development, it can consider adding to the Miscellaneous Industry subsector and other sectors, while also being able to include other variables such as financial performance, market capitalization, etc.