This essay has been submitted by a student. This is not an example of the work written by our professional essay writers.
The aim of this study is to provide comparative evidence on corporate social responsibility practices and to identify factors that influence corporate social responsibility disclosures. A unique sample of 80 sustainability reports for the year ending 2009 from 24 countries is examined using a comprehensive disclosure index. The World Renowned Global Reporting Initiative 2006 Guidelines are used as the benchmark disclosure index checklist. The empirical results indicate that on average, level of corporate social responsibility disclosure is 44 percent. In addition, the amount and type of corporate social responsibility information disclosed varies across the country. The results also indicate that firm size, industry type and company that have their sustainability reports assured are significantly more likely to disclose corporate social responsibility disclosure. Further analysis indicates that industry type and assurance influence all type of corporate social responsibility disclosure (economic, environmental, and social).
The first objective of this study is to provide comparative evidence on corporate social responsibility disclosure (CSRD) practices using data from 24 countries and to determine to what extent CSRD does vary across jurisdictions. The second is to analyse how company and institutional factors account for CSRD.
This study is motivated by the evidence that there has been an increase of CSR reporting in many world economies (Bebbington 2008). The KPMG International Survey provides data that CSR reporting has gone mainstream, with nearly 80 percent of the largest 250 companies from 22 countries issuing stand alone reports, up from about 50 percent in 2005 (KPMG 2008). Yet, CSR reporting is imbalanced globally. This survey shows that CSR in developed countries is higher than developing countries. Variations CSR reporting between countries may be influenced by the stage of economic development, different at social and economic development present different concerns and priorities (Xiao et al. 2005).
In the same way, the intensification in corporate social responsibility disclosure (CSRD) separate recording coincides with an upturn in community concern regarding social and environmental matters and in academic accounting research (Van der Laan Smith et al. 2005). The growing accounting research in CSRD area has largely focused on the extent and determinants of CSRD (see for example Spicer 1978; Trotman et al. 1981; Cowen et al. 1987; Belkaoui and Karpik 1989; Ness and Mirza 1991; Robert 1992; Hackston and Milne 1996; Haniffa and Cooke 2005; Naser et al. 2006; Nurhayati et al. 2006; Branco and Rodriguez 2008; Clarkson et al. 2008, Islam and Deegan 2008; Reverte 2009). Most of these studies have been conducted in a single country setting. However, CSRD studies with data from only a single country evokes the potential for failure to link empirical results to the political, economy or cultural characteristics of broader jurisdictional units (Williams 1999).Whereas an international comparison as conducted in this study, offers deeper insights.
There has been some international research examining institutional factors that may influence CSRD (see for example Williams 1999; Buhr and Freedman 2001; Holland and Foo 2003; Chapple and Moon 2005;Van der Laan Smith et al. 2005; Xiao et al. 2005; Baughn et al. 2007). These studies indicate that country/region, culture, the stage of nation's social and economic development and legal and regulatory context are important determinants of the level and type of CSRD. In addition, Baughn et al. (2007) reveal that relationship between CSRD and jurisdictional economic, political, and social contexts reflect the importance of a country's development of such institutional capacity to promote and support CSRD practices. However, most of these studies are generally limited only to very specific environmental or social aspects. They usually have not comprehensively analysed the broader holistic areas of CSR, ranging from economic, human rights, labour practices and decent work, product responsibility and society aspects. Therefore, this study utilises a different and more comprehensive measurement construct to close the gap and investigate all aspects of CSR using globally broad data. This is important because it is argued that CSR is more than just caring about the environment, all aspects of CSR should be included for a comprehensive analysis (Godfrey and Hatch 2007).
There are five major contributions offered from this study. First, previous studies on CSRD practices have almost solely focused on the annual report. They assume that annual reports contain the key information on social and environmental. However, as a reporting medium, the annual reports still face the limitation in the number of indicators observed and diversity of information provided (Frost et al. 2005). Recent research has shown that companies now rely more heavily on alternative media to communicate social and environmental information such as discrete reports (Frost et al. 2005; Ho and Taylor 2007). The use of other media raises questions about the assumed importance of annual report as main channel for conveying sustainability information (Frost et al. 2005). Frost et al. (2005) conclude that annual reports provide very limited insights into corporate sustainability compared to stand-alone sustainability reports. This study focuses solely on sustainability reports as a better media source to capture and analyse the sustainability information. This research is important to advance a greater a deeper comprehending the extent of companies' CSRD details in sustainability reports
Second, this study uniquely explores the role of assurance to improve the credibility of disclosed information (Simnett et al. 2009). There are a great number of companies that rely on assurance statements in order to improve credibility and transparency of corporate social responsibility information. Over 60 percent of sustainability reports issued by companies in France, Spain, Korea, and Italy include assurance statement (KPMG 2008). This raises questions about what drives company seeks assurance statement on sustainability reports. Despite these advance of demand of assurance statement, research on the effect of assurance on corporate social responsibility reporting is rare (Kolk and Perego 2010). Empirical findings of this study will add greater insights about the adoption decision of assurance services for sustainability reporting. In addition, it also contributes to recent literature in international accounting and auditing and to extend a better understanding of adoption of assurance services.
Third, this study looks at companies throughout the globe using at rich data set in 24 countries rather than merely looking at the CSRD made by companies in a single country. Most past financial disclosure studies have been conducted in a single country setting. However, CSRD studies with data from only a single country evokes the potential for failure to link empirical results to the political, economy, national institutions or cultural characteristics of broader jurisdictional units (Adam et al. 1998; Williams 1999). Further, multi-country sample allows for the introduction of country-specific effects as a significant institutional dimension (Aerts and Cormier 2006 p.302). Differences in the type of corporate governance systems (see for example Van der Laan Smith et al. 2005), the type of business systems (see for example Buhr and Freedman 2001; Chapple and Moon 2005), the type of legal system and level of enforcement (see for example Williams 1999; Buhr and Friedman 2001; Holland and Foo 2003), and the level of economic development (see for example Buhr and Friedman 2001; Xiao et al. 2005) comprise a main source of variations in jurisdictional context with potentially important ramifications on CSRD and overall corporate communication (Aerts and Cormier 2006).
Fourth, this study analyses all types of CSRD rather than just a single social or environmental disclosure item. Prior studies are often limited only to very specific and limited environmental or social aspects. They usually have not analysed the broader holistic areas of CSR. This study utilises a more comprehensive measurement construct to close the gap and better investigate corporate communication. By assessing all the types of CSRD, this study could capture the whole set of values, issues, and processes that companies must address in order to best highlight their societal interactions (Ho and Taylor 2007).
Finally, this study contributes to future research and development in social and environmental accounting disclosure by testing legitimacy theory. Legitimacy theory seeks to understand what factors may cause variability in CSRD and to what extent the variables of interest within an organization may influence organizational actions in seeking legitimacy (Haniffa and Cooke 2005). Past studies of CSRD have adopted a variety of perspectives to explain differences in CSRD, including agency theory (see for example Ness and Mirza 1991), political economy theory (see for example Williams 1999; Xiao et al. 2005), legitimacy theory (see for example Guthrie and Parker 1990; Patten 1991,1992, 2002; Lindblom 1994; Deegan and Rankin 1996; Deegan and Gordon 1996; Adam et al. 1998; Wilmshurst and Frost 2000; Campbell 2000, 2003; O'Dwyer 2002; O'Donovan 2002; Newson and Deegan 2002; Van Staden and Hooks 2007; Cho and Patten 2007; Islam and Deegan 2008; Aerts and Cormier 2009; Archel et al. 2009), institutional theory (see for example Aerts et al. 2006), stakeholder theory (see for example Robert 1992; Van der Lan Smith et al. 2005; Eljido-Ten 2007). Deegan (2002) and Islam and Deegan (2008) argue that legitimacy theory is widely used to explain social and environmental reporting. They posit that corporate annual report disclosure is best explained as a tool for maintaining legitimacy.
The research questions of this study are:
To what extent does corporate social responsibility (CSRD) vary across the sample?
What company and institutional factors explain corporate social responsibility disclosure (CSRD) communication?
Literature Review and Hypotheses
Corporate Social Responsibility
Corporate social responsibility has become a more attractive concept in the last three decades, though both corporate and academic world differ as to exactly how CSR should be defined (Dahlsrud 2008). Carroll (1979 p.500) defines CSR as the social responsibility of business encompassing the economic, legal, ethical, and discretionary (for example philanthropy) expectations that society has of organizations at a given point in time. While, Hackston and Milne (1996 p.78) define corporate social disclosure as the provision of financial and non-financial information relating to an organization's interaction with its physical and social environment as stated in annual report or separate social reports. It includes details of the physical environment, energy, human resource, products and community involve matters.
GRI (2006) characterises CSR or sustainability as the practice measuring performance on economic, environmental, and social impacts to internal and external stakeholders. This study adopts the definition used by GRI (2006), because it is broader than other definitions and it covers the six key themes that can be found in annual report, company's website or stand alone reports which are economic, environment, human rights, labour and decent work, product responsibility, and society. This broader definition is also used by Frost et al. 2005 and Clarkson at al. (2008). Frost et al. (2005) argue that GRI is employed as an indicator of the content and scope of sustainability reporting. Since, this study will use sustainability reports as a medium of CSRD; the GRI definition can provide an appropriate template to measure company's sustainability communications (Frost et al. 2005).
Prior Research of Corporate Social Responsibility Disclosure
The growing accounting research in social and environmental reporting area within a country has focused on the extent and determinants of CSRD (see for example Trotman and Bradley 1981; Cowen et al. 1987; Belkaoui and Karpik 1989; Ness and Mirza 1991; Robert 1992; Hackston and Milne 1996; Haniffa and Cooke 2005; Naser et al. 2006; Nurhayati et al. 2006; Branco and Rodriguez 2008; Clarkson et al. 2008, Islam and Deegan 2008; Reverte 2009). In international comparisons researchers have begun to examine the institutional factors to influence CSRD (see for example Williams 1999; Buhr and Friedman 2001; Holland and Foo 2003; Chapple and Moon 2005;Van der Laan Smith et al. 2005; Xiao et al. 2005; Baughn et al. 2007). These studies indicate that country of origin (Newson and Deegan 2002; Van der Laan Smith et al. 2005), culture and institutional factors (Buhr and Friedman 2001), the stage of nation's social and economic development (Chapple and Moon 2005; Xiao et al. 2005), and legal and regulatory context (Holland and Foo 2003) are potential important determinants of the level and type of CSRD. For example, Baughn et al. (2007) reveal that relationship between CSRD and country economic, political, and social contexts reflect the importance of a country's development of such institutional capacity to promote and support CSRD practices.
Legitimacy theory posits that the legitimacy of a business entity to operate in society depends on an implicit social contract between the business entity and society. Companies can lose their license to operate in society by breaching society's norms and expectations. Accordingly, legitimacy theory predicts that companies adopt environmental and social responsibility reporting to legitimise their operations when society's norms and expectations of the business entities change or the business entities perceive themselves in breach of existing norms and expectations of society (Deegan 2002; O'Donovan 2002; Deegan and Blomquist 2006). According to Deegan (2006 p.275):
"Legitimacy theory asserts that organisations continually seek to ensure that they are perceived as operating within the bounds and norms of their respective societies, that is, they attempt to ensure that their activities are perceived by outside parties as being "legitimate", these bounds and norms are not considered to be fixed, but change over time, thereby requiring organisations to be responsive to the ethical (or moral) environment in which they operate".
Legitimacy may be seen as an important benefit for resource allocation to the firm. Dowling and Pfeffer (1975), and O'Donovan (2002) argue that within legitimacy theory, "legitimacy" is considered to be a resource on which an organisation is dependent for survival. The empirical investigations of CSRD practices have produced a very diverse body of academic literature which engage many different theoretical perspectives. Many researchers have discussed CSRD practices within the theoretical framework of legitimacy theory (see for example Guthrie and Parker 1990; Patten 1992; Wilmshurst and Frost 2000; Deegan 2002; Islam and Deegan 2008). The below five hypotheses are evolved from legitimacy theory tenets.
In previous studies company size has persistently been found to significantly and positively associate with CSRD. Firm size is probably the most commonly used proxy for firm visibility. Firms with higher visibility that are more sensitive to laws and regulation will tend to disclose more intensive social and environmental information. Size has been shown an antecedent of legitimacy; larger companies undertake more activities, have a greater impact on society and have more stakeholders who might be concerned with the social activities undertaken by the company Spicer (1978), Trotman et al. (1981), Belkoui and Karpik (1989), Patten (1991, 1992), Deegan and Gordon (1996), Hackston and Milne (1996), Adam et al. (1998), Williams (1999), Haniffa and Cooke (2005), Nasser et al. (2006), Ho and Taylor (2007), Clarkson et al. (2008), Cahaya et al. (2008), Branco and Rodrigues (2008), Said et al. (2009), Aerts and Cormier (2009), and Reverte (2009) provide evidence that company size has a positive impact on corporate social disclosure. Therefore, H1 is:
H1: There is a positive relationship between firm size and the extent of CSRD in sustainability reports.
Type of industry influences political visibility and drives disclosure in order to minimize pressure and criticism from society (Patten 1991). In previous studies, industry type is another common variable employed to explain the content and extent of CSRD (Cowen et al. 1989; Ness and Mirza 1991; Roberts 1992; Hackston and Milne 1996; Adam et al. 1998; Williams1999). Overall, more sensitive industries are considered to be those with more risk of being criticized in CSR matters because their activities the perception of higher risk (Reverte 2009).
Roberts (1992) and Hackston and Milne (1996) measure industry type as a dichotomous classification industries into high profile and low profile industries. Roberts (1992 p.605) defines high profile industries as those with consumer visibility, a high level of political risk or concentrated intense competition. In this study, the resource and manufacturing industries are classified as high profile industries. Whereas, finance and service industries are included as a low profile industries. Hackston and Milne (1996) provide evidence that high profile industries disclose significantly more social and environmental information than low profile industries. Given their greater need for legitimacy. H2 hypothesis:
H2: Firms in high profile industry will provide a higher extent of CSRD in sustainability reports than firms in low profile industry.
The practice assurance statement in sustainability reports serves as a communication mechanism and it arguably enhances the clarity and reliability of the statements (Deegan et al 2006). Simnett et al (2009 p.939) argue that companies purchase of assurance is driven by their objective to increase stakeholder or user confidence in the quality of the sustainability information provided and/or to increase stakeholder trust in level of organizational commitment to sustainability agendas.
Kolk and Perego (2010) argue that the demand of assurance services is significantly influenced by the legal environment in which a firm operates. Moreover they conclude that governance mechanism, country level institutional mechanism, and level of awareness on sustainability are factors that related with adoption of assurance services. In summary, assurance plays important role to improve the credibility and transparency of the sustainability information disclosed. This stronger need for legitimacy leads to the third hypothesis:
H3: There is a positive relationship between assurance statement and the extent of CSRD in sustainability reports.
Jurisdiction trait is defined as a particular characteristic of the power or influence that a country or group of similar countries has to carry out legal decisions enforce laws and/or affect change to influence firm communication. Several studies in the accounting literature state that jurisdiction traits may make a difference to the pattern of CSRD. There are several factors that impact of disclosure at the national level such as the type of corporate governance systems (see for example Van der Laan Smith et al. 2005), the type of business systems (see for example Buhr and Freedman 2001; Chapple and Moon 2005), the type of legal system and level of enforcement (see for example Williams 1999; Buhr and Friedman 2001; Holland and Foo 2003), and the level of economic development (see for example Buhr and Friedman 2001; Xiao et al. 2005).
A key variable is whether the market is predominantly shareholder-oriented (equity-oriented) or creditor-oriented (debt-oriented). In some countries, when the firms reach a certain size they turn to the stock market as their main source of capital. These capital markets are characterized as shareholder-oriented or equity-oriented capital markets (for example US, UK, and Canada). In other countries, the companies still depend on bank financing as their primary source capital, these are characterized as creditor-oriented or debt-oriented. Example of country with creditor-oriented are Germany, Japan, and Switzerland. Under the equity-oriented jurisdiction view, the objective of manager is to maximize the value of the firms. In contrast, under the creditor-oriented, the corporation is viewed as a separate entity operating very much as a part of the social, political, and economic fabric of society.
The corporation is accorded legal status by society and in turn is expected to fulfill certain
social responsibilities, to the contractarian viewpoint, firms have social responsibilities not only towards their stockholders but to all their stakeholders(Van der Laan Smith 2005 p.130). Given this distinction, Van der Laan Smith et al (2005) argue that management in creditor-orientated societies would be more likely to perform and disclose social responsibility activities as part of strategically managing stakeholder relationships. It is hypothesized that creditor-oriented jurisdictions have higher incentives to provide a greater level of corporate social responsibility disclosures than equity-oriented.
H4: Firms in creditor-oriented jurisdiction will provide a higher extent of CSRD in sustainability reports than firms in equity-oriented.
The empirical governance literature suggests that the degree of board independence is related to composition and independence and will elevate board effectiveness (Said et al. 2009). In this study, the number of independent director is as a proxy of corporate governance. Directors who are more independent to the management may have more power to encourage disclose more social and environmental information to the stakeholders. Consequently, the dominant number of independent directors in the board arguably results in greater disclosures (Haniffa and Cooke, 2005).
The presence of independent directors in the board composition may strengthen the public perception of corporate legitimacy. The public may value an entity highly if it has a balance or many independent directors on the board because such a condition might signify a more effective board in supervising the management activities (Nurhayati et al. 2006). Independent directors also play important role in enhancing corporate image and act as a monitoring role in ensuring that companies is properly managed by its management (Said et al. 2009). The final hypothesis states:
H5: There is a positive relationship between independent directors and the extent of CSRD in sustainability reports.
The data collection focuses on the 2009 fiscal year. The data collected for this study is sourced from the 80 public companies from 24 separate countries. The types of data acquired for this study include: (1) data on a corporate social responsibility disclosure; (2) data on firm characteristics: size, industry, return on asset (ROA), auditor; and (3) data on assurance statement, (4) data on jurisdictional traits, and (5) data on the independent directors. The data of CSRD and assurance statement are derived from stand-alone sustainability reports. Most of the sustainability reports are obtained from the GRI's websites. Sustainability reports not available from GRI's websites are obtained from the company's websites. Data sources for firm characteristics; size, industry, independent director, return on assets and auditor are obtained from annual reports and ORBIS databases.
Sample and Sampling
The sample includes 80 companies from 24 countries listed on the GRI's websites at 2009 fiscal year. The companies chosen as sample in this study are all public companies. The selection of companies in each country is located by contemplating the representativeness of characteristics of each region/country and manageability of data gathering. The proportional stratified random sampling method is used to select the sample. With proportional stratified random sampling, this study controls the relative quantity of each stratum, rather than letting random processes control it. This guarantees the proportion of different strata within a sample and therefore produces a final sample that has more equal representation of each sub-group from the population than simple random methods provide (Neuman 2005).
Measuring Dependent Variable
In this study, the extent of CSRD is dependent variable. CRSD score is calculated by referring to the globally-respected GRI guidelines version 2006. The GRI reporting guidelines contain 79 items that comprehensively reflect the spirit of to the CSR and sustainability reporting. The 79 items are from six key GRI-based categories: economic (9 items), environmental (30 items), human rights (9 items), labour practices and decent work (14 items), product responsibility (9 items), and society (8 items).
Each indicator includes a corresponding set of core and additional indicators. From 79 items, 49 items are core and 30 are additional. The 49 core items have been developed by GRI to identify generally applicable indicators and are assumed to be material for most organizations. An organization should report on core indicators unless they are deemed not material on the basis of the GRI reporting principles while additional indicators represent emerging practice or address topics that may be material for some organizations, but are not material for others. This study employs all items (total) and core items to measure the extent of CSRD.
Consistent with past studies, the CSRD index is calculated as a dichotomous equally weighted index on a 0-100% scale. Marston and Shrives (1991) reviews the disclosure index literature, report that a wide variety of studies have adopted a disclosure index as a mechanism to measure the extent of disclosure. All items are equally weighted and each of the 79 possible items that are disclosed is awarded a score of 1 (and if not disclosed a score of 0 is given). Items are removed from the equation when they are clearly not applicable.
Measuring Independent Variables
Firm size, commonly used as a proxy for public visibility is measured by firm's total asset. The more visible the companies the more social responsibility activities will be considerated and disclosure will become a way to enhance corporate reputation. Williams (1999), Ho and Taylor (2007) Branco and Rodriguez (2008), Cahaya et al. (2008) have concluded that firms size appears to be a significant determinant of CSRD. Firm size is logged to reduce skewness.
This variable is measured by dummy variable. 1 if the company is high profile and 0 if low profile. Differing industry profiles could represent the extent of legitimacy and related political costs. High profile industries can be expected to exhibit greater concern in CSRD than low profile industries. Patten (1991), Hackston and Milne (1996), Adams et al. (1998) and Newson and Deegan (2002) note that industry type has a significant impact on social disclosures in annual reports.
This variable is measured by dummy variable. 1 if company has assurance statement in their sustainability reports and 0 if company does not have assurance statement in their sustainability reports. Simnett et al (2009) find that assurance statement is used by companies as a medium to enhance their credibility.
This variable is measured by dummy variable. 1 if the country is shareholder-oriented and 0 if the country is creditor-oriented. Van der Laan Smith et al (2005) find that countries with creditor oriented have a higher level and quality CSRD in their annual reports.
This variable is measured by the percentage of independence directors to total directors. The empirical governance literature suggests that and increase the level of corporate communication the independence will fosters board effectiveness and enhance perceptions of legitimacy (Haniffa and Cooke 2005; Said et al. 2009).
Measuring Control Variables
Two control variables are employed. These are Return on Assets (ROA) and auditor type. ROA is used as a term to measure profitability. Firms with high profitability tend to have higher levels of CSRD (Aerts and Cormier 2009). ROA is measured as of ratio total net profit divided by total assets. Previous research also suggests that the audit firm can influence disclosure levels. Big4 auditors may be more sophisticated with better audit quality (Gupta and Nayar 2007). A higher quality auditor may help clients prepare more sophisticated sustainability reports. Auditor is measured by dummy variable, 1 if firms audited by Big 4 and 0 otherwise.
This study employs several statistical techniques to test the hypotheses. One-way ANOVA will be used to test whether there is a significant difference in the extent of CSRD between high and low profile industries, report with and without assurance in each countries, shareholder and creditor oriented, and auditor type. This study also utilises Pearson and Spearman correlation and Ordinary Least Squares (OLS) regressions to test the associations between the dependent and predictor variables. The regression models used are:
CSRD = Î²0 + Î²1 SIZE + Î²2 INDUSTRY + Î²3 ASSURANCE + Î²4 JURISDICTION + Î²5 INDEPENDENT DIRECTORS + Î²6 ROA + Î²7 AUDITOR + Îµ
80 sustainability reports from 24 countries are analysed in order to test H1 to H5. Table 1 column (1) shows the name of the 24 countries in this study, column (2) reports level of CSRD index, column (3) the number of sample firms in each country. The highest number of firms are 7 while those the lowest number of firms are 2. Column (4), (5), (6) and (7) show the number of sustainability reports, assured, not assured, and proportion sustainability reports assured. The Spain companies tend to disclose the most (62.8%), followed by South Korea (61.4%) and Austria (61.0%). The country with the highest number of sustainability assured is Spain (5; 100%) and the lowest number is Japan and Turkey (0; 0%). Column (8) presents the jurisdiction traits of country. The sample countries that are shareholder-oriented are Australia, Canada, China, India, Singapore, South Africa, Turkey, UK, and US; the number of countries with creditor orientation are Austria, Belgium, Denmark, Finland, France, Germany, Italy, Japan, Netherlands, Norway, Portugal, South Korea, Spain, Sweden, and Switzerland.
The types of industry involved was divided into high profile and low profile industries. The nature of a company's industry has been identified as a factor potentially affecting CSRD practices (Roberts 1992; Meek et al 1995; Hackston and Milne 1996; Adam et al 1998; Newson and Deegan 2002; Haniffa and Cooke 2005; Reverte 2009). In this study, the high profile industries are agriculture, forest and paper, automobile and airlines, mining, metal, oil, utilities, chemicals, and banking and finance whilst the low profile companies are consumer goods, construction and property, service, food and retails. Table 2 presents the industry composition of sample firms. The high profile sample contains 43 firms or 53.75% with concentrations in mining, metal, oil, utilities, and chemicals (30%) while low profile sample contains 37 firms or 46.25% and majority are from consumer goods industries (21.25%).
The CSRD constitutes 79 items categorised into six categories of performance indicators, namely economic, environmental, human rights, labour practices and decent work, product responsibility, and society. Table 3 provides descriptive statistics for both dependent and independent variables. The mean (median) of disclosure index total is 44.4 (41.0) out of 79. The result also showed that the mean (median of disclosure index core is 31.6 (32.5) out of 49.The social theme has the highest number of disclosure in sustainability reports, followed by environmental and economic.
Regarding the predictor variables, Table 3 summarizes that the mean of total assets is $6.7. The mean of board independence is 8.4. The mean of return on assets (ROA) ratio is 3.90%. With regards to the descriptive statistics of categorical variables is shown in Table 3, 53.75% firm in the sample is classified as high profile while 46.25% as low profile. For 51.25% firm in the sample has assurance statement in their sustainability reports while 48.75% firm does not have. Of the 80 firms from 24 countries, 33 firms or 41.25% are from country with shareholder-oriented, 47 firms or 58.75% are creditor-oriented. Finally, of the 80 firms that had their sustainability reports, 73 or 91.25% audited by a Big 4 audit firm while 7 firms or 8.75% audited by non Big 4 audit firm.
Table 4 presents the CSRD index and the results of the One-way ANOVA tests the relationship between CSRD and type of industry, assurance, jurisdiction, and auditor. The results indicate that there are significant differences across high profile and low profile industries (F = 7.39, pvalue = 0.00) with respect to all items (total index), as well as all four index (core, economic, environmental, and social). Assurance variable also indicates the similar result that there are significant differences between assured and not assured report to all type of CSRD. However, the results also indicate that there is no significant difference between CSRD across shareholder and creditor oriented, as well as CSRD across Big4 and non Big4 auditor except with respect to environmental.
Table 5 presents the multiple regression results for the 80 sample companies. Column (1) provides the result of total disclosure index. As hypothesized, the coefficient on size is positive and significant at the 5% level. It can be observed that size, industry type and assurance do have a significant and positive relationship with total disclosure index. The result of Column (1) is consistent with Column (2), (3), (4), and (5), except for size variable in two of the GRI sub-categories. The coefficient of industry type and assurance are positive and significant at 1% level. However, size is only significantly associated with core index and social index at 10% level.
The explanatory power of the regression is the highest for total disclosure index (32.8%) and the lowest is for the economic index (26.1%). Overall, the results on regression on Table 3 report that H1 is supported for the total disclosure index; bigger companies disclose more. H2 is supported for all types of disclosure, high profile industry tends to disclose more than low profile industry. H3 is also supported for all types of disclosure, this study finds that firms that do have assurance statement tend to provide more sustainability information. However, this study fails to support H4 and H5. The regression results indicate that there is no relationship between jurisdiction and auditor to CSRD.
Implication and Conclusions
The aim of this study is to provide comparative evidence on corporate social responsibility practices and to identify factors that influence corporate social responsibility disclosures across 24 countries. The main contribution of this study are: firstly, it focuses solely on sustainability reports as a better media source to capture and analyse the sustainability information, and to advance a greater a deeper comprehending the extent of companies' CSRD details in sustainability reports, secondly; it looks at companies throughout the globe using at rich data set in 24 countries rather than merely looking at the CSRD made by companies in a single country; thirdly, it analyses all types of CSRD rather than just a single social or environmental disclosure item.
The results of this study generally support the predictions that size, type of industry, and assurance are determinants of the extent of CSRD. This results also indicate that the extent of CSRD ( index total and index core) is significantly higher for firms with high profile and for firms with assurance statement. This findings are consistent with past studies regarding the relationship between size and type of industry (see for example Hackston and Milne (1996), Adam et al. (1998), Haniffa and Cooke (2005), Ho and Taylor (2007), and Reverte (2009)). According to these results, it seems that the legitimacy theory is the most relevant theory for explaining CSRD practices. As outlined in legitimacy theory, the reason of large companies and high profile industries to make more CSRD is regarding to accountability and visibility. Large corporations and high profile industry do have a bigger effect on the community, and therefore normally have a bigger group of stakeholders that influence the corporation (Hackston and Milne 1996; Adam et al 1998; Haniffa and Cooke 2005; Reverte 2009).
This study also find that there is a positive significant relationship between assurance statements and the extent of CSRD. This results support past studies that companies purchase of assurance service to increase stakeholder and user confidence regarding the quality of sustainability information (Simnett et al 2009). This finding is consistent with Deegan et al (2006) and Kolk and Perego (2010) that argue that assurance statement may ehance the clarity and reliability of the sustainability information disclosed. From the legitimacy perspectives, this finding shows that the adoption of assurance statement may lift the reputation of companies and it strengthens to legitimate their social responsibility activities.
Jurisdiction traits and independent directors are not statistically significant. This finding is inconsistent with Van der Laan Smith et al (2005). However, this study focused largely on all type of CSRD (economic, environmental, and social) rather than only social disclosure. It is important to note that the Van der Laan smith et al's study is only focused on social disclosure. The result of this study implies that the relationship between CSRD across jurisdictions may need to be explored by future reseachers. The relationship between CSRD and independent director is also inconsistent with Haniffa and Cooke (2005), Said et al (2009), and Nurhayati et al (2006). Nevertheless,these studies have been conducted in a single country and limited only to very specific environmental or social disclosure. Finding of this study exhibit that in cross-country study, it needs to define independent director precisely.
Finally, this study also find that all the control variables (ROA and auditor) are not significant. Findings of this study show that there is no association between financial performance and the extent of CSRD. The insignificant is possibly explained by globally crisis and it is supported that descriptive statistics for ROA shows that mean of sample is very low. Lastly, the insignificant of association between auditor and CSRD may be explained by most of companies use Big Four auditor to assure their sustainability reports.