Anugerah.
Saraswati and Andayani: Quality of Disclosure and Corporate Social.
Quality Of Disclosure And Corporate Social Responsibility Reporting Practices In Indonesia Eza Gusti Anugerah.
Erwin Saraswati.
Wuryan Andayani Faculty of Economic and Business.
University of Brawijaya.
Malang Email: ezanugerah@gmail.
Abstract: This research aims to analyze the influence of corporate social responsibility (CSR) reporting practices on CSR disclosure quality in Indonesia.
This research used a sample of 103 companies across industries .
xcept for natural resource companie.
listed on Indonesian Stock Exchange from 2014 to 2016.
This research found thatthe voluntary practice of stand-alone report, assurance and reporting guideline does not enhance the quality of disclosure.
This practices tend to be usedas symbolic approach to fulfillcompanies This symbolic approach has the meaning that the companieswhich voluntarily disclose theirCSR information, merely aiming a positive impression from their Companies tend to disclose CSR information by diluting the relevant CSR information with unnecessary information to build their desired images.
Keyword: quality of disclosure, legitimacy theory, stand-alone report, assurance.
GRI.
BACKGROUND
The development of corporate social responsibility (CSR) disclosure was followed by the emergence of CSR reporting practices.
As an effort to enhancethe quality of disclosure, companies may apply three CSR reporting practices, stand-alone report, assurance, and reporting guideline.
First CSR reporting practice is a stand-alone report.
The annual report provides information related tocompanies' performance over a period to business and investment communities, while the stand-alone report provides additional information to stakeholders (GRI, 2.
A stand-alone report in general, disclose the companyAos social and environmental information.
Second CSR reporting practice is the verificationfrom external parties .
According to Wong and Millington .
, assurance can strengthen stakeholdersAo assessment for the usefulness of CSR report.
Other researchers also stated that assurance can increase the credibility of CSR report (Casey and Grenier, 2015.
Cohen and Simnett, 2015.
Habek and Wolniak, 2.
Third CSR reporting practice is the use of a reporting guideline, which in this research is the Global Reporting Initiative (GRI) reporting framework.
GRI reporting framework has been known as the leading international standardization of sustainability reporting (Bebbington et al.
, 2012.
Gray, 2010.
Mahoney et al.
, 2.
In 2013.
GRI issued a framework for preparing a sustainability report under the name GRI G4.
However.
Michelon et al.
explained that there isa possibility for companies to misuse CSR reporting practices in a voluntary context.
CompaniesAo effort to fulfill their legitimacy encourage this possibility.
There are two approaches that organizations commonly use to seek legitimacy,substantive approach and symbolicapproach.
Through a substantive approach, companies have a strategy to fulfill their legitimacy which in line with social norms.
Companies commit tomaintaintheir social and environmental Jurnal Akuntansi/Volume XXII.
No.
September 2018: 337-353 Anugerah.
Saraswati and Andayani: Quality of Disclosure and Corporate Social.
aspect from the negative impact of their activities.
High CSR activities and disclosures indicate that the company has a strong commitment to its employees, social and environment (Hong and Andersen, 2.
As for the symbolic approach, companies tend to express CSR as a strategic effort to gain legitimacy from their stakeholders without regard to social norms.
CSR is used as a tool to achieve certain goals such as corporate image (Cho et al.
, 2012.
Dobbs and Staden, 2016.
Hopwood, 2009.
Merkl-Davies and Brennan, 2007.
Michelon et al.
Based on the explanation above, the quality of information disclosedin voluntary practice is something that is not yet clear.
The voluntary CSR reporting practices need to be analyzed for their possible misuse.
This is important to note because stakeholders must obtain credible information from companiesAo CSR report even though its a voluntary disclosure.
Previous research found that CSR reporting practices did not improve the quality of disclosure (Michelon et al.
, 2.
Authors interested in conducting a research similar to previous research anduse developing countries as the sample of this article.
Aside from being a research that confirms previous research, this research also simultaneously meets the limitation of the previous research in terms of empirical evidence from other countries.
The diverseneeds and expectations of stakeholdersfrom one country to another (Freundlieb and Teuteberg, 2013.
Habek and Wolniak, 2.
are encouraging authors to do this Authors use the framework created by Michelon et al.
as the main reference for measuring the quality of disclosures in this research.
Hopefully, this research can provide an overview of CSR reporting practices in Indonesia.
THEORETICAL REVIEW
Companies disclose information related to social and environment in a form called CSR report (Wolniak & Habek, 2.
CSR report is important for stakeholders and internal corporate decision making (Vurro & Perrini, 2.
The importance of CSR reporttakes the attention of previous researchers to examine its quality (Habek & Wolniak, 2015.
Michelon et al.
, 2015.
Yusoff et al.
In general.
CSR reports are available in aform of sustainability report, triple bottom line report, and corporate social responsibility report.
Indonesia, this type of reportcommonly is a voluntary report.
Voluntary reporting has a tendency to be adjusted according to companies needs.
It makes the credibility of the information issued by companies became questionable (Gray, 2010.
Husillos et al.
, 2.
an efforts to improve the quality of information disclosed,companies adopt three CSR reporting practices, stand-alone report, assurance, and reporting guidelines (Michelon et al.
Stand-alone report provides additional information to stakeholders (GRI, 2.
Companies whichdisclose social and environmental information on a stand-alone report can improve their images (Mahoney et al.
, 2013.
Michelon et al.
, 2.
Stand-alone report has a role in increasing companiesAo accountability.
It will help investors to get additional information needed to assess the company's overall CSR performance(Dhaliwal et al.
, 2012.
Dhaliwal et al.
, 2.
In addition, companies also issued a stand-alone report as a strong commitment to social and environmental issues (Hong & Andersen, 2.
Assurance is a service which guarantees the credibility of CSR information carried out by external parties.
Wong & Millington .
found that assurance can strengthen stakeholdersAoassessment in terms of CSRreport usefulness.
This finding alsosupported by Habek & Wolniak .
which found verification from independent party enhance the Jurnal Akuntansi/Volume XXII.
No.
September 2018: 337-353 Anugerah.
Saraswati and Andayani: Quality of Disclosure and Corporate Social.
credibility of CSR report.
Other researchers also have a similar finding that assurance can increase the credibility of CSR reporting (Casey & Grenier, 2015.
Cohen & Simnett, 2.
Reporting guideline is a framework used as a guideline to prepare CSR report.
The reporting guideline analyzed in this researchis the framework issued by GRI.
GRI issued GRI G4 framework in 2013 which provides guidelines for the preparation of CSR reports with broad dimensions.
Unfortunately, the ability of three CSR reporting practiceswere questioned when it comes tovoluntary use.
There is a possibility for companies toabuse the voluntary CSR reporting practice (Michelon et al.
, 2.
It also has the potential to be used as a tool to achieve certain goals such as corporate image, positive perception, and impressionfrom stakeholders (Cho et al.
, 2012.
Deegan, 2002.
Hopwood, 2009.
Merkl-Davies & Brennan.
The possibility of CSR report abuse can be explained by understanding two approaches that companiescommonly use to seek legitimacy (Michelon et al.
, 2.
There are the substantive approach and the symbolicapproach.
Substantive approach represents companies movement to fulfill their legitimacy through authentic actions and changes in accordance to social norms.
Companies respond thechanges of external conditionswith some strategies followed by concrete actions.
Companies need to inform the stakeholders that they have changed and moving forward in accordance to their expectations or social norms.
Therefore.
CSR disclosure becomes an important medium to inform the stakeholders about the changes in companies strategies and actions (Hopwood, 2.
Companies that have high CSR activities and high disclosures, indicating that companies have strong responsibility towards employees, social and environment (Hong & Andersen, 2.
Symbolic approach tends to positively influence stakeholdersAo perceptions of the company, and likely doing apparent actions to convince the key stakeholders that the company has committed to social norms.
companies take action and disclose it by using CSR CSR report used as a strategic tool to get a positive impression and manage relationships with external parties of the company (Gao & Bansal, 2.
CSR disclosure is also used as a tool to achieve certain goals such as corporate image to gain legitimacy for the company (Cho et al.
, 2012.
Deegan, 2002.
Hopwood, 2009.
Michelon, et al.
, 2.
From this perspective it can be understood that CSR disclosures can be misused for the benefit of the company, and information disclosed has the possibility of not fully describing the actual state of the company to stakeholders.
Based on the theoritical review aboves, the credibility of information disclosed in voluntary CSR report becomes unclear.
These two opposing approaches to CSR reporting pose an important research question.
Companies may use the three CSR reporting practices to improve their reporting practices.
However,they may provide more but not enhanced social and environmental information.
Stand-alone report.
Stand-alone report used to disclose financial, social and environmental information to stakeholders (Cho et al.
, 2015.
Wolniak & Habek, 2.
Furthermore, it has an important role in improving corporate accountability.
Investors can get additional information from stand-alone reportto help them assess the company's overall CSR performance (Dhaliwal et al.
, 2012.
Dhaliwal et al.
, 2.
However, if the stand-alone report is issued voluntarily, it is possible to be misused.
There are two perspectives in understanding the meaning behind voluntary stand-alone report, substantive approach and symbolic Jurnal Akuntansi/Volume XXII.
No.
September 2018: 337-353 Anugerah.
Saraswati and Andayani: Quality of Disclosure and Corporate Social.
Based on the substantive approach, issuing stand-alone report can provide additional information, as well as improving companies accountability.
Companies may use stand-alone report as a medium to inform their authentic actions and changes in accordance to social Issuingstand-alone reports are also a form of companyAosimprovement, thus demonstrating a strong commitment to social and environmental issues (Hong & Andersen.
Based on the symbolic approach, companies used stand alone report as a tool to achieve certain goals such as positive impression and corporate image (Cho et al.
, 2012.
Deegan.
Gao & Bansal, 2013.
Hopwood, 2009.
Merkl-Davies & Brennan, 2007.
Michelon, et , 2.
In this perspective, stand-alone report is only used as opportunistic actions that benefit the company.
Based on the conceptual framework created, researchers formulate the following hypotheses:
H1: Issuing stand-alone report affects the quality of CSR disclosures.
Assurance.
Credibilityis a value that must be considered by the company in preparing CSR Companies guarantee the credibility of CSR reporting by using verification from external parties .
However, if assurance is used voluntarily, it is possible to be There are two perspectives in understanding the meaning behind the voluntary use of assurance services, substantive approach and symbolic approach.
Based on the substantive approach, using assurance for CSR report will improve its Wong & Millington's .
found that the need for assurance service is positively related to stakeholdersAo assessment in terms ofCSR report usefulness.
Habek & Wolniak .
also found thatindependent verification is needed to get a credible CSR Furthermore.
Edgley et al.
, .
found that assurance increase stakeholdersAo inclusivity in the reporting process because it changes managersAoattitudes towards them.
From this perspective, assurance servicecould provide real changes in a companyAos processes and substantively increases credibility of companies CSR disclosure.
On the other hand, there are doubts about assurance capabilities in improving transparency of CSR reporting (O'Dwyer & Owen, 2.
There is also a tendency for companies to use assurance models that are in line with the company's desires (Michelon et , 2.
Since voluntary reporting in general did not regulated, companies have flexibility in using assurance services.
Furthermore.
Cho et al.
, .
found that assurance does not have a relationship with a higher market value for companies which issued CSR report, but the company gets the impression of being more committed to social and environmental.
The findings support the argument that the benefits of assurance are limited only to perceptions of the companiesAosocial and environmental image, and does not enhance the social and environmental information.
Based on the symbolic approach, it can be understood that the assurance benefits are only to get a positive impression from stakeholders but does not enhance the information provided.
Based on the conceptual framework created, researchers formulate the following hypotheses:
H2 : The use of assurance services affects the quality of CSR disclosures.
Reporting Guideline.
The GRI reporting frameworkis widely acknowledged for leading the international standardization of sustainability reporting (Bebbington et al.
, 2012.
Gray, 2010.
Mahoney et al.
, 2.
GRI reporting framework are also one of the innovations in the field of CSR in terms of regulatory development (Voegtlin & Scherer, 2.
GRI reporting framework provides broad dimensions guideline, including economic, social and Jurnal Akuntansi/Volume XXII.
No.
September 2018: 337-353 Anugerah.
Saraswati and Andayani: Quality of Disclosure and Corporate Social.
However, the voluntary use of reporting guideline is possible to be misused.
There are two perspectives in understanding the meaning behind the voluntary use of reporting guideline, substantive approach and symbolic approach Based on the substantive approach, companies which use reporting guideline in preparing CSR reports are known to have a higher level of commitment to CSR (Michelon et , 2.
Furthermore, the use of the GRI reporting framework provides broad dimension of information in terms of companiesAo CSR and also a form of companiesAo improvement and innovation in the field of CSR disclosure (Voegtlin & Scherer, 2.
In symbolic approach, there is a tendency for companies to use reporting guidelines in biased way (Freedman, 1.
It is because in voluntary disclosure, companies have the flexibilities to choose a reporting guideline that suits them.
In addition, companies tend to use GRI reporting framework for a positive impression from stakeholders by giving an idea that companies follow the development of CSR reporting practices.
For GRI guidelineAos case.
Michelon et al.
, .
states that companies may misuse GRI guideline by trying to meet all GRI indicators, therefore increasing their reporting assessments.
From this perspective, companies uses the GRI reporting frameworkonly to impress the stakeholders by meeting all the indicators, so companiesmay achieve high scores in their performance without regard to the quality of information.
Based on the conceptual framework created, researchers formulate the following hypotheses:
H3 : The use of GRI guidelines affects the quality of CSR disclosures.
METHOD
This research uses a positive approach to analyze the effect of CSR reporting practices on the quality of CSR disclosures.
The type of data used in this study is secondary data.
Data sources in this study are annual reports and sustainability reports from the companiesAo This research uses time-series and cross-sectional based data.
Researchers use content analysis to analyze CSR disclosures.
The population are all companies listed on the IDX except natural resources companies.
This selection is based on the different obligations in reporting CSR.
Natural resources companies are required .
to carry out CSR activities in accordance to UUNo.
tahun 2007 pasal 74, while the other companies are still voluntary.
This research focuses on CSR reporting that is voluntary, so natural resources companies are not included in this researchAos observation.
This researchAos year of observationare 2014 to 2016.
This decision based on consideration of GRI G4 usage period in Indonesia.
The sampling method used in this research is purposive sampling.
The samplesAo criteria of this study are: .
Companies observed are continuously listed on the IDX from 2014 to 2016.
Companies issues a stand-alone report or annual report that can be accessed either from companies' website or IDXAos website.
Companies have CSR or sustainability information in their annual report.
Independent variables.
Stand-alone report is used to disclose financial, social and environmental information to stakeholders (Cho et al.
, 2015.
Wolniak & Habek, 2.
The Stand-alone CSR report is a dummy variable.
The value is 1 if the company provides a standalone CSR report, and the value is 0 if the company's CSR information is contained in the company's annual report.
Assurance is a third party verification service.
Assurance is a dummy variable.
The value is 1 if there is an assurance statement .
hird party verification statemen.
in the Jurnal Akuntansi/Volume XXII.
No.
September 2018: 337-353 Anugerah.
Saraswati and Andayani: Quality of Disclosure and Corporate Social.
companyAos CSR report, and the value is 0 if there is no assurance statement in its CSR report.
A reporting guideline is a guideline used by companies in preparing their CSR reports.
The reporting guideline in this study is GRI G4 issued by GRI (Michelon et al.
, 2.
The GRI guideline is a dummy variable.
The value is 1 if there is a company statement of adherence to the GRI guideline, and the value is 0 if there is no company statement of adherence to the GRI guideline.
Dependent variables.
The dependent variable in this research is the quality of disclosure.
Authors use the framework created by Michelon et al.
to measure the quality ofCSR There are four indexes that can be used as a measurement of the quality of disclosure, relative quantity index (RQT), density (DEN), accuracy (ACC), and managerial orientation (MAN).
Relative quantity index measures a company level of disclosure compared to others companies level of disclosure in the same industry .
ee Michelon et al.
, 2.
Relative quantity index can be seen in the following formula:
Definition:
RQTit = Relative quantity indexfor companyiin yeart Discit = Observed level of disclosure for companyiinyeart it = Estimated disclosure level for companyiin yeart.
Density index measures the number of sentences which contain relevant information to GRI G4 compared to the total sentences expressed in the document.
Density can be seen in the following formula:
Definition:
DENit = Density indexfor companyiin yeart = Number of sentences in the document analyzed for companyi in yeart CSRijt= 1 if the sentencejin the document analyzed for companyiin yeartcontains CSR information, otherwise 0.
Accuracy index measures how companies disclose information in their CSR reports.
Accuracy can be seen in the following formula:
Definition:
ACCit = Accuracy indexfor companyiin yeart = The number of sentences containing CSR information in the document analyzed for companyiin yeart CSRijt = 1 if the sentence j in the document analyzed for company i in yeartcontains CSR information, otherwise 0 = 1 if the sentence j in the document analyzed for company i in year t is Jurnal Akuntansi/Volume XXII.
No.
September 2018: 337-353 Anugerah.
Saraswati and Andayani: Quality of Disclosure and Corporate Social.
qualitative, w = 2 if the sentence j in the document analyzed for company i in year t is quantitative, w = 3 if the sentencej in the document analyzed for company i in year t is monetary.
Managerial orientation index measures how the company disclose their CSR Companies may use whether boilerplate approach or committed approach to express their CSR information.
In boilerplate approach, companiestend to express general expectations concerning the future and provides rules, initiatives and strategies in expressing the results.
In committed approach, the company provides stakeholders with future objectives and goals with an account of the results and outputs from the actions taken to meet stakeholders expectation.
Managerial orientation can be seen in the following formula:
Definition:
MANit = Managerial orientation indexfor companyiin yeart = The number of sentences containing CSRinformation in the document analyzed forcompany i in year t OBJit = 1 if the sentence j in the document analyzed forcompany i in year tcontains CSR information on goals and objectives, otherwise 0 RESijt = 1 if the sentence j inthe document analyzed for company i in year tcontains CSR information on results and outcomes and 0 otherwise.
The four indexes are then synthesized with the following formula:
Definition:
Qualityit = Quality of disclosureobtained from the standardized value of fourindexes RQTsit = Standardized relative quantity index for company i in year t DENsit = Standardized density index for companyiin yeart ACCsit = Standardized accuracy index for companyiin yeart MANsit = Standardized managerial orientation indexfor companyiin yeart.
Control variables.
Authors usefourcontrol variables to control the influence between the independent variable and the dependent variable.
Size, previous research found that size significantly has a positive effect on the level of disclosure.
The greater the size of the company, the higher the level of CSR disclosures issued by the company (Sun et al.
, 2.
Based on these findings, the size of the company canbe used as a control variable.
The proxy of the company size is measured by referring to Michelon et al .
using natural logs of total sales.
Total sales are suitable for this research, because there is a tendency for the number of sales to affect the company's CSR budget.
Environmentally or Socially Sensitive Industries (ESSI) is a companiesAo classification according to their sensitivity to social and environment.
ESSI is used as a control variable, because there is a different social exposure from one industry to another industries (Brammer & Millington, 2.
Following the study of Michelon et al.
, .
, companies which engaged in the fields of chemistry, mining, iron, oil, paper, and utility industries are Jurnal Akuntansi/Volume XXII.
No.
September 2018: 337-353 Anugerah.
Saraswati and Andayani: Quality of Disclosure and Corporate Social.
companies that are socially and environmentally sensitive.
Based on the explanation above.
ESSI is a dummy variable whose value is 1 if the company is one of ESSI category, and 0 if the company is not in ESSI category.
Corporate Social Performance (CSP) is an indicator to know the company's performance in carrying out its corporate social responsibility.
Following Michelon et al.
, authors use CSR as a control variable to get more accurate results.
CSP is measured using the GRI G4 assessment.
CSP can be seen in the following formula:
CSPit = Keterangan:
CSPit = CSP index for companyi in yeart = Total items disclosed by the companyiin yeart = Total itemsin GRI G4 The Company's age has been generally accepted as one of the characteristics of the company (Suttipun & Stanton, 2.
A company that has been operating for a long time, have a tendency to be better prepared in facing stakeholdersAo expectations.
Based on the explanation above, authors uses companyAos age as a control variable in this research.
The companyAos age is the length of time that the company has been through until 2016.
Data analysis method.
At first, authors with their assistants carried out the process of content analysis to get the raw data.
Then, authors process the raw data using the STATA After that, authors determine the best model using two model estimation techniques.
the first test is the Chow test, used to choose between common effect models or fixed effects the second test is Hausman test, used to choose between fixed effect models or random effects model.
Lastly, authors analyze the regression results from the best models RESULT AND DISCUSSION Descriptive results.
This research has 103 companies with three years of observation.
Total samples are 309 observations.
The characteristics of the sample can be seen in table 1.
Table 1.
Descriptive statistics (Panel A) CSR Report
Year No Yes
Total
Total 254 55
(Panel B) CSR Report
Assurance No Yes Total
Yes
Assurance Year No Yes
Total
GRI
Year Yes Total Total 288 21 Total 254 55 CSR Report GRI No Yes Total Assurance GRI No Yes Yes Jurnal Akuntansi/Volume XXII.
No.
September 2018: 337-353 Yes Total Anugerah.
Saraswati and Andayani: Quality of Disclosure and Corporate Social.
Total Total 254 55 Total 288 21 Based on Table 1, panel A shows that only 55 companies .
8%) issued stand-alone reports, 21 companies .
8%) used assurance services, and 55 companies .
8%) used GRI Panel B provides an overview about the use of two CSR reporting practices Panel B shows 21 companies issued stand-alone reports and used assurance services, 54 samples issued stand-alone reports and used GRI guideline, and 21 samples used assurance services and GRI guideline.
Lastly,samples which use all three CSR reporting practices are 36 companies.
7%).
From the table 1 it can be concluded that GRI guideline are the most popular guideline used for preparing stand-alone report.
Of the 55 companies that published stand-alone report, there were 54 companies that also used GRI guidelines.
It is worthy mentioning that 0.
6% of the samples use the GRI guideline but did not issue a stand-alone report.
These samples provide a different picture from other samples where companies generally issue stand-alone report using GRI G4 as the Furthermore, 0.
6% of the samples issued a stand-alone report but did not use the GRI guideline when most of the samples used the GRI guidelines for their stand-alone Overall, these results show that the practice of CSR reporting in Indonesia is still in the development stage.
CompaniesAo lack of interest in using CSR reporting practices are likely due to their voluntary nature.
There is a possibility companies did not consider them Regression results.
Before performing regression, authors test the model first.
This is important to get the right model for the observed samples.
The tests were the Chow test and the Hausman test.
Chow test is used to determine which is the best among common effect model or fixed effect (FE) model.
If the p-value is smaller than then FE model is suitable for this research.
Otherwise, common effect model is suitable for this research.
Authors used STATA program to do the Chow test.
The results of the Chow test are in table 2.
Table 2 shows that P-value = 0.
0000 is smaller than the value of = 5%, so it can be concluded FE model are the best modelfor this study.
Table 2.
Chow test Chow Test F test that all u_i=0 : F.
= 8,36 Prob > F = 0,0000 The next test is Hausman test.
Hausman test is used to determine which is the best among FE model or random effect (RE) model.
If the p-value is smaller than then FE model is suitable for this research.
Otherwise.
RE model is suitable for this research.
Authors used STATA program to do the Hausman test.
The results of the Hausman test are in table 3.
Table 3 shows that P-value = 0.
7170 is higher than the value of = 5%, so it can be concluded RE model are the best modelfor this study.
Table3.
Hausman test Hausman Test chi2 .
= .
-B) ' [(V_b-V_B)^(-.
-B) = 3,70 Jurnal Akuntansi/Volume XXII.
No.
September 2018: 337-353 Prob > chi2 = 0,7170 Anugerah.
Saraswati and Andayani: Quality of Disclosure and Corporate Social.
After finding RE model is the suitable model, authors used pearson correlation to determine the relationship between variables without being influenced by other variables.
Table 4 provides an overview of the relationships between variables and their significance It was found that three variables (CSR Report.
Assurance and GRI) have a negative and significant correlation .
= 1%) to disclosure quality (Qualit.
This results suggesting that the adaption of three CSR reporting practice decreases the disclosure quality.
There is a possibility for companies to dilute their CSR information with unnecessary information.
The relationship strength between the three reporting practices and Quality is 0.
3602, 0.
2255, and It can be concluded that 13% variation in Quality is explained by CSR report, 5% variation in Quality is explained by Assurance, and 13% variation in Quality is explained by GRI.
Table 4.
Correlation matrix
Quality CSR
Report
Assurance GRI
Size
ESSI
CSP
Quality 1,0000 CSR Report
3602***
1,0000
Assurance 2255*** 5803*** 1,0000 GRI
3654***
9779***
5803***
1,0000
Size
-0,4413***
4646***
2486***
4646***
1,0000
ESSI
0,1563***
-0,1042*
-0,0732
-0,1042
0,0257
1,0000
CSP
-0,3432***
0,6876***
0,3962***
0,7038***
0,5021***
0,0089
1,0000
AGE
-0,2213***
0,4174***
0,3323***
0,4184***
0,3288***
-0,0643
0,3389***
AGE
1,0000
***p<0,01.
**p<0,05.
*p<0,1.
The correlation between the quality of disclosure with four control variables have a significant correlation at 1% level.
The only positive correlation was ESSI, while others have negative correlation.
The result indicates companies which are in ESSI categorieshave a tendency to increase their disclosure quality.
Itmay occur due to the pressures of their stakeholders andstrict regulations in Indonesia for companies that are sensitive to the social and environment.
Hypotheses testing uses RE model with the following formula:
Qualityit = 1 1CSR Reportit 2Assuranceit 3GRIit 4Sizeit 5ESSIit 6CSPit 7Ageit e Keterangan: Qualityit = Quality of disclosurecompanyiin yeart.
= Constant.
Regression coefficient.
CSR Reportit = Stand-alone reportcompanyiin yeart.
Assuranceit = Assurance of companyiin yeart.
GRIit = GRI guideline of companyiin yeart.
Sizeit = Size of companyiin yeart.
ESSIit = Industrial categories of companyiin yeart.
CSPit = CSP of companyiin yeart.
Ageit = Age of companyiin yeart.
e = Error Hypotheses testing is carried out in two steps.
F-test and t-test.
F-test can be done by comparing the prob> chi2 value with significance level ().
If the p-value is more than the significance level, then the model cannot be used to explain the dependent variable.
Otherwise.
If the p-value is less than the significance level, then the model can be used to explain the dependent variable.
T-test can be done by comparing the significance value of each independent variable with the significance level.
Jurnal Akuntansi/Volume XXII.
No.
September 2018: 337-353 Anugerah.
Saraswati and Andayani: Quality of Disclosure and Corporate Social.
Table 5.
RE model regression result Wald chi2 .
= Prob > chi2 = R-squared 43,06 0,0000 0,2507 Quality Koefisien
Standar error
Signifikansi
CSR Report
0,0179731
0,0562503
0,749
Assurance -0,0102518 0,0335743 0,760 GRI
-0,0709632
0,0588499
0,228
SIZE
-0,0260961
0,0065151
0,000***
ESSI
0,0549805
0,0292435
0,060*
CSP
-0,0502883
0,1099639
0,647
AGE
-0,0003259
0,0008386
0,698
***p<0,01.
**p<0,05.
*p<0,1.
Based on Table 5, prob> chi2 is less than the value, it can be concluded that the regression model fits the data and can explain the variations in dependent variable.
addition, it is known that the R-squared value is 0,2507, which means that this model can explain 25% variation in the quality of disclosure.
Based on table 5.
The stand-alone report (CSR Repor.
is not significantly.
associated withquality of disclosure (Qualit.
This result shows that stand-alone reportdid not enhance the quality of CSR disclosure.
This result is consistent with Michelon et al.
as the main reference of this research.
Publishing a stand-alone report does not enhance the quality of CSR disclosure and its just only a symbolic actions.
In line with Hopwood .
and Merkl-Davies & Brennan .
research, voluntary information disclosure is only a tool to give an impression.
CSR disclosure is used as a strategic tool to get a positive impression and manage relationships with external parties of the company (Gao & Bansal, 2.
CSR disclosure is also used as a tool to achieve certain goals such as the company's image to gain legitimacy for the company (Cho et al.
, 2.
These results also indicate the possibility of greenwashing (Mahoney et al.
, 2.
, that is the actions of a companywhichissue stand-alone report only to provide an image that the company follows social and environment values.
It can be concluded that companieswhich voluntarily issued stand-alone reportare using symbolic approach, and aim for positive impression from their stakeholders regardless of enhancement of the information disclosed.
Assurance is not significantly .
associated withquality of disclosure (Qualit.
This result shows that assurance servicedid not enhance the quality of CSR disclosure.
There is a tendency for companies use a symbolic approach in using assurance services.
Companies which use symbolic approach tend to aim positive images from stakeholders through assurance services bycreating an image that the company follows the development of international reporting practices.
This result is consistent with Michelon et al.
that assurance services do not improve the quality of CSR disclosures and tend to use a symbolic In line with Cho et al.
that assurance services are used only to get the company's image of its commitment to social and environmental.
In summary, assurance Jurnal Akuntansi/Volume XXII.
No.
September 2018: 337-353 Anugerah.
Saraswati and Andayani: Quality of Disclosure and Corporate Social.
servicesare limited to increase the stakeholders' perception rather than enhance the quality of information disclosed.
GRI guideline (GRI)is not significantly .
associated withquality of disclosure (Qualit.
This result shows that the use of GRI guideline is only a symbolic act anddid not enhance the quality of CSR disclosure.
Companies which use symbolic approach in achieving their legitimacy tend to engage in apparent act as well as using GRI guideline for their CSR Authors found several companies which tried to meet the GRI G4 Core indicator without a clear or concrete information.
They usegeneral expectations sentencesto meet the GRI G4 indicators, resulting ina decrease inquality of their CSR report.
This results are consistent with Michelon et al.
that the use of GRI guidelines does not improve the quality of CSR disclosures.
Companies tend to misuse the GRI guidelines by trying to meet all GRI indicators, with the intention of increasing reporting assessments.
It can be concluded that GRI guideline did not enhance the quality of disclosure and have a tendency to be misused by the companies.
The quality of disclosure measurement framework which created by Michelon et al.
define the recording unit as single sentences.
If the quality of disclosure scoreis low, then one or all indexes such as RQT.
DEN,ACC and MAN have a low scorefromthe CSR This can possibly occur if companies diluteCSR relevant information with other information which irrelevant to CSR, or companies only reports their intention or expectation without any actions or subsequent performance.
One example of a sentence that reduces the quality of disclosure was found in the annual report issued by PT.
Acset Indonusa Tbk.
AuCSR program are designed to provide a space for the Companyto develop not only economically, but also socially and forthe community on the surroundings of operational area.
Thisinitiative serves as an indicator for a successful sustainableAy.
These sentences has no score from MAN index, because they have no information regarding the companyAosCSR objectives or the companyAos CSR results.
Based on that finding, these sentencesare considered irrelevant information.
If companies want to get a better quality of CSR reports, then they should pay more attention onhow they deliver their CSR report should contain an information which is required by stakeholders but avoid excessive and unnecessary detail (Michelon et al.
, 2.
The summary of hypotheses testing can be seen in table 6.
Table 6.
Hypotheses testing results Independent & Dependent Coefficient P>.
Results CSR Report & Quality 0,0179731 0,749 Insignificant Assurance & Quality GRI & Quality -0,0102518 -0,0709632 0,760 0,228 Insignificant Insignificant For the control variables results, companies size (SIZE) have a negative and significant influence on the quality of disclosure (Qualit.
This negative influence can meanthe big companies may have a low quality of disclosure and can also mean smaller companies may have a better disclosure quality than the larger companies.
Its possibly occurs if the company use symbolic approach for their voluntary CSR disclosure.
If a company uses symbolic Jurnal Akuntansi/Volume XXII.
No.
September 2018: 337-353 Anugerah.
Saraswati and Andayani: Quality of Disclosure and Corporate Social.
approach for its voluntary disclosure, then there is a tendency for the company to only focus in aiming stakeholdersAo perception regardless the quality of disclosure.
Furthermore,ESSI has a positive and significant influence on the quality of disclosure (Qualit.
Its shows that socially and environmentally sensitive companies have a tendency to enhancetheir disclosures qualities.
They have better disclosure quality than the others This may occur due to the strict regulationsfrom Indonesia government for companies that are socially and environmentally sensitive.
CSP and AGE are not significantly associated withquality of disclosure (Qualit.
These results indicate that CSR performance and age of the company have no influence on the quality of disclosure.
It can be interpreted that the longer a company operates or the better the CSR performance is, does not guarantee an increase in the quality of CSR disclosure.
Additional analysis.
Researchers conducted an additional analysis to provide more complete information regarding the research on CSR disclosure quality in Indonesia as well as to increase the robustness of this study.
Based on the existing samples, authors re-analyzed each industry by grouping the observations into 3 large groups, .
Manufacture industry.
Nonfinancial services industry.
Financial services industry Table 7.
Regression result Ae additional analysis Coefficient and Significance Quality Manufacture Non-financial services industry Financial services 0,0118007 -0,1314393 0,0069692 .
* .
Assurance -0,0459507 -0,0381014 -0,0214114 GRI
-0,0717103
CSR Report
SIZE
-0,0455838
-0,0099965
-0,0230507
***
0,012379
0,1110536
0,1546532
0,1989867
-0,1131444
-0,0000552
-0,0015993
0,0005030
Wald chi2 .
= 13,80 20,10 17,13 Prob > chi2
0,0320
0,0027
0,0088
R-squared 0,2604 0,2972 0,3792 ESSI
CSP
AGE
***p<0.
**p<0.
*p<0.
Jurnal Akuntansi/Volume XXII.
No.
September 2018: 337-353 Anugerah.
Saraswati and Andayani: Quality of Disclosure and Corporate Social.
Based on Table 7, stand-alone report and assurance in manufacturing industry are not significantly associated withquality of disclosure (Qualit.
These results are consistent with the main results.
GRI is removed from this analysis due to high collinearity with CSR report, because most of the companies from the data collectedare issuingstand-alone report together with GRI guidelines.
Its makes the relationship of these two variables very strong.
Company size are significant and has a negative coefficient, which means the increase in company size may decrease the quality of disclosures.
Its possibly occur if the company uses a symbolic In non-financial services industry.
CSR Report are significantly associated withquality of disclosure (Qualit.
The negative coefficient shows that the company which issued a stand-alone report had a lower quality of disclosure than a company that did not issue a standalone report.
In other words, companies issuestand-alone report but does not enhancethe quality of disclosure.
There is a possibility for companies to dilute the relevant CSR informations with unnecessary informations which decrease the quality of disclosures.
Assurance is not significantly associated withquality of disclosure (Qualit.
GRI is removed from this analysis due to high collinearity with CSR report, because most of the companies from the data collectedare issuingstand-alone report together with GRI guidelines.
In this industry.
ESSI are significant and has a positive coefficient.
This result indicates that companies which are sensitive to social and environmental aspects have a tendency to improve their quality of disclosure.
Some samples in non-financial services are engaged in construction and utilities which have higher CSR activities than others non-financial This result is likely due to strict regulations on environmentally sensitive companies, thus motivating companies to improve theirdisclosure quality.
In financial services industry, the CSR Report.
Assurance and GRI variables are not significantly associated withquality of disclosure (Qualit.
These results are consistent with the main results.
Similar to the results of the manufacture industry that company size are significant and has a negative coefficient.
it means that the increase in company size may decrease the quality of disclosures.
ESSI removed from analysis because all sample in the financial services industry are not ESSI.
Overall, the three CSR reporting practices in all three industry groups in Indonesia do not enhance the quality of disclosure.
ENDING
Conclusion.
This study aims to analyze the effect of three CSR reporting practices on the quality of disclosure in Indonesia.
This research found thatvoluntary practices of stand-alone report, assurance, and reporting guideline in Indonesia tend to be used as symbolic approach to fulfill companiesAo legitimacy.
CompaniesAomain target are stakeholdersAo perceptions, engaging in apparent actions to make stakeholders believe that the company is committed to societal requirements.
Companies use CSR reporting practices to get a positive impression, improvingtheir image and also gaining legitimacy from their stakeholders at the same Furthermore,the results from additional analysis are also supporting the main results.
Three CSR reporting practices in all three industry groups in Indonesia do not enhance the quality of disclosure.
This finding is consistent with Michelon et al.
as the main reference in this research whichshows, on average, three CSR reporting practices donot enhance the quality of disclosure, and only usedas tools to enhance perceived accountability.
It is worthy to mention that company size andESSI affect the quality of disclosure.
For company size, it has a negative influence on the quality of disclosure which means Jurnal Akuntansi/Volume XXII.
No.
September 2018: 337-353 Anugerah.
Saraswati and Andayani: Quality of Disclosure and Corporate Social.
largercompanies may have a lower quality of disclosure than the smaller companies.
Its possibly occur if larger companies dilute their relevant information with unnecessaryinformation to lead their stakeholders to their desired image.
For ESSI, it has a positive influence on the quality of disclosure which means socially and environmentally sensitive companies have a tendency to enhancetheir disclosures qualities.
Its possibly occur because of the strict regulation from the government and pressures from their stakeholders which encourage them to do a substantive approach of CSR.
This research implies that companies should pay more attention onhow they deliver their information.
Companies should avoid excessive and unnecessary detail in their CSR report and focus in enhancing their social and environmentalinformation.
Furthermore, companies should pay more attention on aligning their strategies and processes to social norms, so that the fulfillment of legitimacy can be done through substantive approach.
This research contributes in providing empirical evidence of CSR reporting practices in Indonesia, and also fills the limitations of Michelon et al.
related to empirical evidence from other countries to support their research.
Its important to note because CSR practices in one country with others countries may have different results.
This research has several limitations.
This research focus only on three years of data .
, because GRI G4 is widely used by many companies in Indonesia at that period of time.
In order to get spesific results and overview related to the use of GRI G4, authors decided to pick that period of time without considering other periods.
In terms of measuring the relevance of an information, this research used GRI G4.
For future research, authors recommend specific and deeper research which uses stakeholdersAo expectations as a measure for information relevance.
Authors hope that this research will motivate further research about CSR reporting practicein the future, so a better quality of CSR report can be achieved.
REFERENCES