The Influence of Corporate Social Responsibility (CSR) Disclosure and Sustainability Accounting on Earnings Response Coefficient (ERC)

The research aimed to analyze the factors which provide earning response coefficient (ERC) received by automotive manufacturing companies listed on the Indonesia Stock Exchange using indicators of corporate social responsibility (CSR) and sustainability accounting from 2013 through 2017. The purposive sampling technique was used to gather data with the criteria, according which 13 companies were obtained as samples, (1) automotive manufacturing companies in 2013 – 2017 listed on the Indonesia Stock Exchange (IDX) (2) automotive manufacturing companies that published the annual reports. Meanwhile, the data were obtained from IDX & published annual reports. The research used a panel data regression panel and data regression model as the analysis technique. The analysis consists of three methods, namely common effect method, fixed-effect and random effect, while the hypothesis testing used t-statistics to test partial regression coefficients and f statistics to test the effect simultaneously at the significance level of 5%. Eventually, the results of the analysis used E-views to show whether: (1) corporate social responsibility (CSR) influences the earnings response coefficient (ERC) and (2) sustainability accounting influences the ERC. The T-test analysis results used E-views to reveal whether CSR and sustainability accounting influence ERC in automotive manufacturing companies listed on the IDX in 2013 – 2017.

On the other hand, sustainability accounting (X2) is a sub-category of financial accounting that focuses on disclosing information on the organization's non-financial performance to stakeholders. This performance includes activities that directly impact society, the environment, and the economy. Furthermore, the earnings response coefficient (ERC) (Y) estimates changes in the company's stock price as a result of its earnings information announced to the market.

Disclosure of Corporate Social Responsibility
According to Max Caldwell (2014:79), CSR is an aspect that strengthens the level of employee engagement in an organization. This idea was put forward based on previous findings where CSR is one of the 18 triggers for employee engagement (Towers Perrin, n.d). MBR (2009). Furthermore, it lists specific benefits from involving employees in the CSR programs, such as to adhere the employees to the organization's vision and mission, increase their expertise/skills and provide them opportunities to gain new skills. This study reveals corporate social responsibility in a company's annual report as indicated with the corporate social responsibility disclosure index (CSRI). The study used the CSRI measurement instrument based on the GRI indicators (G3), classified CSR information into several categories, i.e. economy, environment, labor, human rights, social, and products. Meanwhile, the GRI (Global Reporting Initiative) is an official institution that issues sustainability reporting standards. A dichotomous approach to calculate CSRI, which implement that each CSR item of the research instrument is assigned a value of 1 whenever it is disclosed and 0 if it is not (Haniffa et al., 2014: 33). Next, each item's scores are summed up to obtain the overall score for each company. The CSRI calculation formula is as follows: Where: CSRIj is corporate social responsibility disclosure index of company j, Nj is the number of a company j items, ΣXij is dummy variable, 1 = if the item i disclosed; 0 = if the item i not disclosed. Therefore, 0 ≤ CSRIj ≤ 1.

Sustainability accounting
According to Elkington (2010), sustainability report (SR) has various definitions. SR refers to a report that contains not only financial performance but also non-financial information consisting of information on social and environmental activities that enable companies to grow sustainably (sustainable performance). On the other hand, according to Heemskerk, Pistorio and Scicluna (2002:7), a sustainability report is defined as public reports by companies to provide internal and external stakeholders with a picture of corporate position and activities on economic, environmental and social dimensions. In short, such reports attempt to describe the company's contribution to sustainable development. Sustainability accounting or also known as social accounting, social and environmental accounting, corporate social reporting, corporate social responsibility reporting, or non-financial reporting is a sub-category of financial accounting focusing on disclosure of organizational non-financial performance information to stakeholders. The performance encompasses activities that directly impact society, the environment, and the economy. Sustainability accounting contrasts to management accounting which focuses on decision making and internal policies (Dally, 2011:8).
In sustainability accounting, one of the main challenges includes a lack of understanding of the precise definition of 'sustainable development'. According to Brundland (2010:78), sustainable development refers to that which meets present needs without compromising the ability of future generations to meet their own needs. This is to raise questions about the extent of our own needs and those of future generations. As such, it is difficult to explain the relationship between time and geography between the triangles of organization, environment and social impacts. For example, how far the environmental account in a company can guarantee the benefit of the environment and how significant a portion of its role is in saving the future. More specifically, sustainability reports are proxied by the SRDI (sustainability report disclosure index) which is performed by assigning a value of 1 if the item is disclosed or 0 if not disclosed. Then all scores are summed up as a whole. After scoring each index, the value is calculated with the SRDI formula as follows: Where: SRDI is sustainability report disclosure index, N is the number of company's samples, and K refers to the number of classes.

Earning response coefficient (ERC)
Good earnings quality can be determined using ERC as a measuring tool of the earnings information content. In this study, the ERC estimate is a coefficient obtained from cross-sectional regression between cumulative abnormal return (CAR) as a proxy for stock prices and unexpected earning (UE). Furthermore, CAR is obtained using a window (time interval), in which, the CAR is calculated daily for 15 months from 1 January to 31 March. It is because that too short period used will not be able to show market reactions that may occur outside the time interval. For example, it is due to the slow reaction of investors. On the other hand, when the period used is too long, it can provide a biased measurement regarding the company's contribution of information disclosure. The calculation of abnormal return in this study used market-adjusted models which assumes that the market index return is the best measurement (Pincus in Widiastuti, 2012). Therefore, it is unnecessary to use the estimation period to form an estimation model since the estimated securities return is the same as the market index return in the period. As such, the market return index uses the return from the composite stock price index (IHSG). The calculation of the abnormal return can be performed using the following formula.
Where: ARit is abnormal return for company i on day t Rit is daily return of company i on day t Rmt: the market index return on the day t Pit: the share price of company i on day t Pit-1: the share price of company i on day t-

METHODOLOGY
It is quantitative research with a scope limited to corporate social responsibility (CSR), sustainability accounting and earnings response coefficient (ERC) in automotive manufacturing companies listed on the Indonesia Stock Exchange (IDX). The data were gathered from the corporate social responsibility report and the annual report from 2014 to 2018. The population in this study only comprised automotive manufacturing companies, the criteria of which were taken into consideration as follows: (1) listed on the Indonesia Stock Exchange (BEI) 2014 -2018, (2) issue annual report, (3) publish the corporate social responsibility report in the annual report. Based on the above criteria, the sample size was 13 companies.
Quantitative analysis methods can be tested using statistical testing, i.e. regression models with panel data as a combination of cross-section and time-series data.

RESULTS AND DISCUSSIONS
Regression analysis on panel data was carried out using the fixed effect estimation model. The test aims to estimate the fixed effect model panel data to capture the intercept differences between variables. The intercept differences might occur due to different data originating from panel data. To determine whether there is the influence of corporate social responsibility (CSR) and sustainability accounting on the earnings response coefficient (ERC), a panel data regression analysis was carried out. The results of data processing using the fixed-effect method follows the formula: Y = 0. 545908 + 0. 590392 X1 + 0.469171 X2 + e Where: Y = earning response coefficient (ERC) X1= corporate social responsibility (CSR) X2= sustainability accounting Based on the above equation, it could be drawn several conclusions as follows: 1. The constant value in this study is 0.545908, by which, if the value of corporate social responsibility (CSR) and sustainability accounting is equal to 0, the value of earning response coefficient (ERC) is 0.545908. 2. The regression coefficient for corporate social responsibility (CSR) is 0.590392, indicating that if the value of corporate social responsibility (CSR) increases by 1 point. In contrast, the other variables are constant, the earnings response coefficient (ERC) value will increase by 0.590392, and vice versa. 3. The regression coefficient of sustainability accounting is 0.469171, indicating that if the sustainability accounting increases by 1 point while the other variables are constant, the earnings response coefficient (ERC) value will increase by 0.469171, and vice versa.

Test of the coefficient of determination (
The coefficient of determination test was carried out to determine the extent to which the relationship among the variables, i.e. corporate social responsibility (CSR) and sustainability accounting on the earning response coefficient (ERC), exists. The method is to square the correlation coefficient value and then transform it to a percentage. The result of determination coefficient processed using EViews 10.0 is as follows:  The F-count value is 18.64896, which is greater than that of F-table 2.87, while the probability value (F-statistic) is 0.000000. Using the degree of error at α = 5 per cent or 0.05, the probability (F-statistic) is less than 0.05. It indicates that there is an effect of variables of corporate social responsibility (CSR) and sustainability accounting on earnings response coefficient (ERC) in automotive manufacturing companies in 2014 -2018. The fluctuation of the earnings response coefficient (ERC) variable in automotive manufacturing companies in 2014 -2018 can be explained by independent variables (corporate Social responsibility and sustainability accounting). Therefore, other factors account for 54.59%.

Partial test (t-test)
The t-test (partial regression test) was conducted to determine whether corporate social responsibility (CSR) and sustainability accounting influence the earnings response coefficient (ERC) if the t-count is at sig <0.05. Using coefficients analysis with Eviews processing, The t-test obtains the following results. Source: the output of eviews 10 a) Variable of corporate social responsibility (CSR), we obtained the results of t-count 5.984072> t-table 1.72472 in a probability value of 0.0021 greater than that of the significance level 0.05. Therefore, the hypothesis is accepted; in other words, corporate social responsibility (CSR) positively influences earnings response coefficient (ERC). The regression coefficient of corporate social responsibility (CSR) is 0.590392, indicating that if CSR's value increases by 1 point while the other variables are constant, the value of the earning response coefficient (ERC) will increase 0.590392, and vice versa. b) Variable of sustainability accounting, the t-count result is 5.798374> t-table, i.e. 1.72472 in a probability value of 0.0008, which is less than the significance level of 0.05. Therefore, the hypothesis is accepted; in other words, sustainability accounting has a significantly positive influence on the earnings response coefficient (ERC