Research Article: 2018 Vol: 22 Issue: 4
Gehan A. Mousa, University of Bahrain and Benha University
Abdelmohsen M. Desoky, University of Bahrain and South Valley University
Ghias U. Khan, University of Bahrain
This study investigates the possible association between corporate governance (CG) and disclosure of corporate social responsibility (CSR) by listed companies in the Gulf Cooperation Council (GCC) countries. An un-weighted CSR disclosure index including 41 information items has been developed to gather suitable information from corporate annual reports of 246 sampled listed companies from the six GCC countries covering the 2016 financial period. Five independent variables represent CG factors (board size, non-executive directors, role duality, female directors and size of the audit committee) in addition to three company characteristics (firm size, profitability and type of industry) are control variables in six regression models. Findings of regression models reveal that only board size and non-executive directors are determinants of CSR disclosure; conversely, no significant association is found between CSR disclosure and the other three CG variables.
Corporate Social Responsibility Disclosure, Corporate Governance, The Gulf Cooperation Council (GCC).
concept CSR was introduced for the first time in business contexts by Sheldon (1924) to reflect the global awareness of the negative impacts of firms’ operations in both environmental and social matters. Firms face strong pressure from different stakeholders, society, and governments to behave responsibly (Mallin et al., 2013). They are increasing their disclosures on CSR to present how they transact with the environmental, social and economic consequences of their different activities. Consequently, several studies reveal a steady improvement in CSR and increase the need for such disclosures in different countries (Dias et al., 2016). Hill et al. (2007) define CSR as “the economic, legal, moral, and philanthropic actions of firms that influence the quality of life of relevant stakeholders”.
Coincided with increased demand for CSR disclosure by a variety of stakeholders, Corporate Governance (GC) provides rules and procedures for greater transparency and credible disclosure (Alfraih and Almutawa, 2017; Albassam, 2014). Good practice of CG mechanisms is a key feature of CSR (Welford, 2007). The integration between these mechanisms and CSR may enhance corporate practice and its social role. The more the firms engage in social activities, the more CSR disclosure. Izzo (2014) stated that “CSR is an issue of governance and a strategic tool more than a mere communication activity, with a potential impact on both organizations and their economic and financial performance”. Bhimani and Soonawalla (2005) state “the concepts of CG and CSR are two sides of the same coin”. In this stream, earlier studies have documented a positive association between CG and CSR disclosure as an indication of the increase of social responsible behavior (Alfraih and Almutawa 2017).
Because of the unique role of CSR disclosure for the efficiency of capital markets and its importance for investors, managers and others who rely on it (Sletten, 2012), the literature on the association between CG factors and CSR disclosure has generated much attention. Habbash (2016) argues that CG factors are responsible for monitoring firms’ decisions and activities which affect all their stakeholders including the community. This supports an association between the effectiveness of CG system and CSR disclosure. Such a system is associated with disclosure in general and with disclosure of significant activities that affect the environment and community in particular. CG factors are main determinants of corporate disclosure practice (Albassam, 2014). In USA, Giannarakis et al. (2014) and Mallin et al. (2013) report that CG factors are positively associated with CSR disclosure.
The current study aims to empirically investigate:
1. The extent of CSR disclosure by sampled companies in the six GCC countries.
2. The association between a number of CG variables (i.e. role duality, non-executive directors, board size, female directors and size of audit committee) from one side and CSR disclosure by listed companies in GCC countries from the other. Five CG variables are selected as determinants of CSR disclosure.
This study has a number of contributions. The focus of the study, GCC countries as emerging capital markets relatively have few CSR research studies. Our study may help to increase the understanding the impact of CG factors on CSR disclosure in GCC business environment. To the best of our knowledge, this study is one of the first studies in the GCC area to investigate female directors as a GC variable and its impact on CSR disclosure. It may be useful for regulators in the GCC who play a unique role to improve the efficiency of capital markets, protect investors and enhance confidence in these markets. Finally, the findings of the current study may be of great interest for listed companies to improve their social responsibility and for different stakeholders to improve their decisions.
This study is organized as follows: Section 2 includes background on CG in the GCC. Section 3 includes the theoretical framework, literature review and development of research hypotheses. Section 4 includes research methods and section 5 shows Data analysis and findings. Finally, Section 6 provides summary and conclusions including limitations and future research.
The GCC six member countries are characterized by a combination of commonalities such as language, customs, traditions and religion, as well as similar social, economic and political conditions. The economic activities of these countries depend mainly on revenues of oil industries because they have significant oil and gas reserves in addition to other activities. Governments of the GCC seek to achieve the prosperity and economic progress of their people through the recovery and development of the financial markets and attract more foreign investments through various investors. Therefore, they are interested in CG practices. CG has emerged as a reaction to a series of scandals and financial crises in many countries. It is a system that includes practices and rules, which control a firm to achieve balance among different stakeholder interests. Consequently, CG codes have been established in most countries to improve CG practices (Hassan et al., 2017).
Oman is the first country in the GCC to issue a code of CG in 2002 that adopted UK-style (based on the 1992 UK Cadbury Report) voluntary compliance and enforcement regime. In the UAE, the 2007 Securities and Commodities Authority (SCA) established the CG Code, which expands government requirements and rules that should be adopted by firms listed on Abu Dhabi and Dubai stock markets. In 2009, the Ministry of Economy in UAE issued Ministerial Resolution No. 518 amending SCA’s code, which provided comprehensive CG rules that were mainly adopted from international standards. In Kuwait, the Capital Markets Authority (CMA) issued the executive rules for the governance system in 2013 but time after time, the implementation of governance was delayed until mid-2016 to allow companies to arrange their internal affairs. In Qatar, the CG Code for companies and legal entities listed on the stock market of Qatar has been developed by the Qatar Financial Markets Authority (QFMA) and based on the best international and regional codes of CG.
In the Kingdom of Bahrain, the CG Code was enacted on January 1st 2011 and companies should apply the code by the end of 2011. The code consists of 9 principles that adhere to international best practices to enhance effective board, more transparency and disclosure and equal rights of all shareholders. Finally, in KSA, CG has a number of axes such as shareholders’ rights, disclosure and transparency, the audit committee and the Board of Directors (BoD). CG is seen as a new mechanism, which enhances the integrity of financial transactions, the performance of the BoD and stakeholders’ rights. In conclusion, it can be noted that CG codes in GCC member countries have nearly similar characteristics and promote confidence and stability in the financial markets through greater transparency and disclosure.
Agency Theory and CSR Disclosure
A number of theoretical perspectives have offered different explanations and motivations for CSR disclosure. For example, stakeholder theory explains corporate disclosure as a tool for meeting stakeholders’ needs of information (Freeman, 1984); while, legitimacy theory argues that CSR disclosure is a means for companies to legitimize their various activities (Lindblom, 1994) and agency theory argues that directors or agents are likely to voluntarily present information if they have incentives (Haniffa and Cooke, 2002). Since agency theory is one of the most widely used theories to explain the association between CG factors and voluntary disclosure in general and the disclosure of CSR in particular, the current study has adopted it (Barako et al., 2006; Haat et al., 2008; Li et al., 2008). Chi and Wang (2009) described two agency problems. The first arises from the separation of management and ownership; and the second because of the varied interests of managers, owners and other different stakeholders.
Agency theory (Chi and Wang, 2009; Barako et al., 2006) proposes that agents or directors will solve or mitigate the information asymmetry problem by two mechanisms. First better monitoring which depends on board characteristics, second providing more disclosures. With regard to the first mechanism, the main responsibility of the BoD is to combat both agency problems and the tendency to introduce their personal interest above all corporate stakeholders (Charles and Fombrun, 2006). Agency theorists argue that the effectiveness of the board in firms depends on some characteristics such as size, the ratio of non-executive directors and the CEO duality. For example, agency theory suggests that a larger board will include diversified knowledge and expertise, which may influence positively corporate disclosure and performance in addition, more effective board’s monitoring role in contrast to a small board (Bassett et al., 2007).
Several studies including Eng and Mak (2003) and Barako et al. (2006) have suggested that non-executives are expected to be more accountable and transparent which reflected in a higher level of disclosure. Fama and Jensen (1983) argue that the non-executive directors in the board can play a critical role to minimize agency conflicts within the firm. They help to maintain balances of the management performance and enhance the effectiveness of the board. Moreover, the separation of the two roles (CEO and the chairman of the board) improves monitoring process by avoiding the concentration of power and authority in one person (Haat et al., 2008; Li et al., 2008). Jensen (1993) argued that internal control systems fail and the board will not be effective in monitoring activities when the CEO duality exists. Li et al. (2008) pointed out that role duality has a negative effect on CSR disclosure because it reduces the effectiveness of the board regarding disclosure policies.
Based on agency theory, we can argue that managers disclose more voluntary information such as CSR information to achieve a number of advantages. Such disclosure helps to mitigate agency conflicts; reduce information asymmetry; and disseminate information to stakeholders about the firm’s activities (Shehata, 2014); reduce the costs of agency (Barako et al., 2006); and convince different users that they are acting in an optimal way (Watson et al., 2002) and help firms to get lower capital or debt cost (Lan et al., 2013).
Several disclosure studies investigated CSR disclosure practices and identified its various determinants, especially those allied to CG factors. In general, a number of scholars (Bushman and Smith, 2001; Hassan, 2012) argued that firms with best practice of CG are intended to disclose more information to stakeholders for different reasons. CG can improve corporate reputation; avoid litigation costs; reduce agency costs; enhance firm performance and finally provide understanding on how different decisions are made within firms (Li et al., 2008).
In Malaysia, a number of studies have been conducted. Haniffa and Cooke (2002) investigated the CSR disclosure determinants for a sample of 167 Malaysian listed companies in 1995. The main results of the study showed that some CG factors such as non-executive directors and the proportion of the family members on the boards have a significant negative association with CSR disclosure. In contrast, four firm characteristics namely, size, profitability, multiple listing and industry type have a significant positive association with the disclosure of CSR. Haniffa and Cooke (2005) used the content analysis approach to examine culture and CG characteristics on CSR disclosure for 139 non-financial firms. The main findings of their study showed that firm size, profitability, multiple listing and industry are associated positively to CSR disclosures. Furthermore, Esa et al. (2012) investigated a change in the level of CSR disclosure to determine the effect of CG attributes through a sample of annual reports of Malaysian government companies. They revealed that board size is the main factor for CSR disclosure. Further, Yusoff et al. (2015) investigated the possible association between CG mechanisms and corporate environmental disclosure practices. Using three years financial reports of the 100 leading listed companies in Malaysia, they reported a positive significant association between board size and environmental disclosure, while no association was found on board independence, female directors and ownership concentration.
In the USA, Giannarakis et al. (2014) investigated the association between a number of CG factors and the extent of CSR disclosure. Some of these factors including firm size, the board commitment to CSR and firm profitability were positively correlated with the extent of CSR disclosure in contrast to financial leverage which has was shown a negative association with it. In Bangladesh, Khan (2010) investigated the effect of a number of CG factors on CSR disclosure in a sample of commercial banks for 2007 and 2008 and revealed that a significant positive association between the extent of CSR disclosure and the percentage of non-executive directors, foreign ownership, profitability and firm size. In another study on banks in the same country, Das et al. (2015) reported a significant positive correlation between firm size, board size, ownership structure and independent non-executive directors on the board and CSR disclosure. Using a sample of 116 listed Bangladeshi non-financial firms, Muttakin et al. (2015) found a negative association between CSR disclosure and women directors while, a positive association has been found between CSR disclosure and firm size, profitability and foreign directorship.
In Pakistan, Lone et al. (2016) examined the association between some CG factors and CSR disclosure by 50 firms from 8 different sectors (from 2010 to 2014) and reported a positive association between CG factors namely women directors, independent directors, board size, and the extent of CSR disclosure. Furthermore, Dias et al. (2016) examined the impact of the global financial crisis on the extent of CSR in two various periods, before the financial crisis (2005- 2008) and during the financial crisis (2008-2011) in a small sample of 36 Portuguese listed companies. They reported that there is an increase in CSR before the crisis while, there is a slight decrease after it.
In Egypt, using a disclosure index of 34 information items of environmental, customer, human resources, energy, and community involvement issues, Rizk et al. (2008) analyzed 60 annual reports of the financial year 2002 and reported that CSR disclosure significantly varied according to the type of industry. In Kuwait, Alfraih and Almutawa (2017) used an un-weighted disclosure index with a total of 49 information items to investigate corporate social and environmental activities of 52 listed firms in KSE from 2005-2008. They revealed that the voluntary disclosure in Kuwaiti firms is low.
In KSA, Macarulla and Talalweh (2012) analyzed annual reports of 132 listed firms in 2008 to assess the level of CSR disclosure. The results showed that level of CSR disclosure is low and three firm attributes (industry type, firm profitability and firm size) are the main CSR determinants. Al-Janadi et al. (2013) found a similar result of the low level of CSR disclosure and reported a significant positive association between a number of variables such as nonexecutive directors, audit quality, board size, role duality, government ownership and CSR disclosure. Using a sample of 267 non-financial listed firms from 2007-2011, Habbash (2016) investigated CSR disclosure in KSA and the potential impact of CG, ownership structure and firm features in KSA employing a CSR disclosure index of 17 items. Habbash (2016) found that CSR disclosure increased after the application of Saudi CG code in 2007. In Qatar, Naser et al. (2006) examined determinants of CRS disclosure and found three variables (firm size, business risk, and corporate growth) to be associated with the level of CSR disclosure. In addition, Al- Khater and Naser (2003) supported the existence of the CSR disclosure in Qatar. Another study was conducted in the same country, Qatar, by Al-Naimi et al. (2012) who investigated CSR disclosure for both the financial and manufacturing listed firms. They concluded that the highest disclosed information was related to human resources and product development, followed by community involvement.
Using a sample of 53 GCC Islamic banks in 2008, Bukair and Abdul-Rahman (2015) reported an insignificant association between some CG variables (board size, role duality and board composition) and CSR disclosure level. Furthermore, Khasharmeh and Desoky (2013) evaluated the level and variation of on-line CSR disclosure by 163 companies listed in the six GCC stock markets. The authors found industry type, firm size, firm profitability and firm risk as key determinants of the online CSR disclosure. The current study investigates the association between CG and CSR disclosure using a wider sample than its previous studies such as Khasharmeh and Desoky (2013) and Bukair and Abdul-Rahman (2015) that used 163 listed companies 53 islamic banks respectively. Furthermore, this study is one of the first studies in the GCC area to examine the association between female directors and CSR disclosure.
Based on the above literature (Haniffa and Cooke, 2002; Haniffa and Cooke, 2005; Al- Janadi et al., 2013; Khasharmeh and Desoky, 2013; Bukair and Abdul-Rahman, 2015; Yusoff et al., 2015), the current study uses five CG factors as independent variables (board size, nonexecutive directors, role duality, female directors, and size of the audit committee) to examine the effect of these variables on CSR disclosure after controlling three company characteristics (firm size, profitability and type of industry) as follows.
Prior studies (Haat et al., 2008; Li et al., 2008) reported that the BoD is an important internal CG mechanism, which help to reduce agency problems. Board Size may affect the way in which directors carry out their main responsibilities (Fama and Jensen, 1983). Similarly, Zahra et al. (2000) argued that the board ability to monitor and assess management and executives is based on its size. There are two opposing points of view regarding the size of the board. The first view argues that large boards are more effective to monitor firm activities, reduce conflicts and may include collective experience (Monks and Minow, 2004). “Larger boards provide greater monitoring of the financial accounting process” (Anderson et al., 2004: p. 340). In contrast, the second view pointed out that a small board size is better than large in a different manner such as communication and coordination problems between members (Jensen, 1993; Cheng, 2008). Aljifri and Moustafa (2007) found that firms with a large board size might face increases in agency costs. However, the empirical results on the association between board size and CSR are mixed. Das et al. (2015), Esa et al. (2012) and Cheng and Courtenay (2006) showed a significant positive association between board size and CSR disclosure. In contrast, Said et al. (2009) reported a weak association between board size and the extent of CSR disclosure. Based on the previous discussion, the first hypothesis is formulated as follows:
H1: There is a significant positive association between board size and the extent of CSR disclosure.
Larger the size of non-executive directors, the more independence of the board, which may reduce both the divergence between the interests among stakeholders and costs of agency (Ienciu, 2012; Andres et al., 2005). A higher percentage of non-executive directors on the board may provide a positive effect on firm performance (Chau and Gray, 2010). Achieving high levels of transparency and disclosure quality depends on the independence of the board members (Habbash, 2016). The effect of the non-executive directors on CSR disclosure is arguable. For instance, Said et al. (2009) in Malaysia and Habbash (2016) in KSA could not demonstrate the significant impact of independent directors on CSR disclosure. However, Haniffa and Cooke (2005) concluded a negative association between the proportion of non-executive directors and the extent of CSR disclosure in Malaysia. Similar findings were reported by Eng and Mak (2003). Previous arguments suggest the following hypothesis:
H2: There is a significant association between non-executive directors and the extent of CSR disclosure.
Role duality refers to same person as having two responsibilities (the chief executive officer and the authority of the chairman of the board) (Giannarakis et al., 2014). In CG literature, role duality is considered one of the most important CG factors. From an agency perspective, the capability of the board to accomplish its main functions such as monitoring, discipline, compensation of senior managers may reduce when the role duality exists in a firm (Molz, 1988). Habib and Hossain (2012) argued that role duality has a negative effect on monitoring actions of the CEO and thus may reduce value and performance of the firm. Consequently, separation of the functions between CEO and board chairman is seen as an effective monitoring tool and recommended by agency theory. Empirical evidence on the association between CEO duality and CSR disclosure is mixed. A number of empirical research concluded that role duality is more likely to enhance the quality of voluntary disclosure including the CSR disclosure (Forker, 1992; Haniffa and Cooke, 2002; Said et al., 2009). On the other hand, Habbash (2016) found no significant association between role duality and CSR disclosure. Similar results were revealed by Giannarakis et al. (2014) and Bukair and Abdul-Rahman (2015). Based on the above arguments, the following hypothesis is formulated:
H3: There is a significant association between role duality and the extent of CSR disclosure.
The BoD, which comprises different members in culture, customs, traditions, gender, practical and scientific experiences is expected to play larger social and economic roles in the community. There is association between the existence of female members on boards and firm philanthropy in the arts and other community services (Williams, 2003). In the same line, Bear et al. (2010) argued that the higher the ratio of female members in the BoD, the more the social and charitable activities. Muttakin et al. (2015) concluded that female directors are more sensitive to charitable and community matters than men. They are favorable for several reasons, such as work experience background and communicate in a more participatory manner, which often encourages a broader perspective to include different needs for all stakeholders and more CSR disclosure (Srinidhi et al., 2011). Empirical research examining the association between female directors and CSR disclosure provided inconclusive results. For example, Bear et al. (2010) reported a positive association between female directors and CSR disclosure. In contrast, Khan (2010) reported no significant association between female directors and CSR disclosure in a sample of Bangladeshi banks. This result is consistent with Giannarakis et al. (2014) who found that the presence of female members on the board does not affect the CSR disclosure. In the light of the above discussion, the following hypothesis is formulated as follow:
H4: There is a significant positive association between female directors and the extent of CSR disclosure.
Size of the Audit Committee
The audit committee is a cornerstone in CG practice that enhances financial reporting. Since one of the main responsibilities of the audit committee is to oversee financial reporting as a dynamic internal control system, the existence of an audit committee can improve the quality of corporate disclosure (Forker, 1992). Prior studies provided a number of characteristics of an audit committee to play an effective monitoring role over financial reporting (DeZoort and Salterio, 2001). These characteristics include number of meetings, independence of audit committee, size of the audit committee and sufficient financial expertise of the committee members (Hassan et al., 2017). Said et al. (2009) found a significant positive association between an audit committee and the CSR disclosure level. Because all companies in the sample have an audit committee, the present study relied on the size of the audit committee to measure the effect of an audit committee on CSR disclosure. A number of studies argued that the audit committee size is an important character that helps the committee to act more effective. An audit committee should have a small number of members with three and maximum six (Krishnan, 2005). The literature on the effectiveness of an audit committee (Pearce and Zahra, 1992; Bedard et al., 2004) concluded that an appropriate audit committee size helps members to apply their experience to improve the quality of financial reporting and serve different stakeholders. In contrast, Anderson and Reeb (2003) pointed out that a large audit committee helps to strengthen the effectiveness of the audit committee and increase its monitoring functions. Based on previous discussion, the following hypothesis is formulated as follow:
H5: There is a significant positive association between size of the audit committee and the extent of CSR disclosure.
Listed companies on the stock markets of the six GCC countries are the population of this study. Table 1 shows details on the sampled companies and their distribution within six countries. They were selected randomly from listed companies in each country and were divided in two categories (industrial and non-industrial). The selection was restricted to only companies, which provide at least one item of the CSR items of disclosure. Table 1 indicates that 252 companies were selected for this survey; however, 6 companies were excluded for a number of reasons (e.g.: none of them provided a single item of the CSR information). The rest was 246 companies representing 36.2% of all listed companies in the six GCC countries. Of the sampled companies, 105 (42.7%) are industrial and 141 (57.3%) are non-industrial companies.
Sample companies selected for the current study1
|- Listed companies||43||179||161||112||45||139||679|
|- Companies selected for the study||18||88||46||32||26||42||252|
|- Excluded companies||1||2||1||2||0||0||006|
|- Companies included in the study||17||86||45||30||26||42||246|
|- Industrial companies2||3||41||25||15||8||13||105|
|- Non-industrial companies||14||45||20||15||18||29||141|
|- Percentage of selected companies3||6.9%||35.0%||18.3%||12.2%||10.6%||17.0%||100%|
Note: 1:A full list of companies included in the study is available upon request from the corresponding author.
2:The total number of industrial companies is low in some countries (e.g., Bahrain and Qatar).
3:Percentage of companies from each country to the total sample (246 companies).
Independent and Control Variables
Independent variables include a set of five CG factors, namely, board size, non-executive directors, role duality, female directors and size of the audit committee. A number of studies have revealed various associations between company characteristics (such as firm size, firm profitability, type of industry, firm leverage, firm liquidity and audit firm) and the extent of CSR disclosure. This study considers three company characteristics namely, firm industry, firm size and firm profitability as control variables. Based on the relevant literature, the current study used “Firm net sales revenue” as a proxy for firm size and “Firm net income” as a proxy for firm profitability (Desoky and Mousa, 2013; Khasharmeh and Dessoky, 2013, Bukair and Abdul- Rahman, 2015 and Habbash, 2016). Definitions of all independent and control variables used in this analysis are presented in Table 2.
INDEPENDENT AND CONTROL VARIABLES
| Panel A: Independent variables:
1. Board size
2. Non-executive directors
3. Role duality
4. Female directors
5. Size of audit committee
Panel A: Control variables:
1- Firm industry
2- Firm size
3- Firm profitability
+ or –
+ or –
|Number of the BoD.
% of Non-executive directors to total board members.
A dummy variable; (0) if the CEO is the same as the board chair; and (1) otherwise.
% of female directors to total board members.
Number of audit committee members. A dummy variable; (1) if the firm is industrial and (0) otherwise.
Firm net sales revenue.
Firm net income.
Notes:1-Information on the above variables was collected at the end of 2016 financial period; 2-Predicted signs of independent And control variable were based on their expected effects on the dependent variable; 3-Firm size and profitability are in US$.
Construction of the CSR Disclosure Index (The Dependent Variable)
An un-weighted CSR disclosure index, which treats all included items equally with a dichotomous procedure in which an item scores: (1) if it is disclosed and (0) otherwise, was implemented in this study, indicating that all items are equal in significance. To build this index in the current study, a cautious review of a number of CSR disclosure studies, accomplished in various parts of the world, was carried out (Haniffa and Cooke, 2005; Khan, 2010; Khasharmeh and Suwaidan, 2010; Esa et al., 2012; Khasharmeh and Desoky, 2013). All possible efforts were made in this stage of the research to construct a reliable CSR disclosure index. The independent variable (CSR Disclosure), which includes 41 items, was divided into four groups (sub-variables) of CSR disclosure items as shown in Table 3 below.
THE DEPENDENT VARIABLE AND ITS SUB-GROUPS
|S. No.||Group||Items||No of Items||Symbol|
|1||Environmental and related information||01-09||09||TOTG1|
|2||Employee welfare information||10-22||13||TOTG2|
|3||Community involvement and social information||23-35||13||TOTG3|
|4||Products quality and safety information||36-41||06||TOTG4|
|5||Total CSR Disclosure||01-41||41||TOTCSR|
Descriptive statistics and statistical analysis (Pearson correlation and regression analysis) were carried out. Six linear regression (Ordinary Least Square-OLS) models with enter method were performed for continuous dependent variables, five independent variables (BSIZE, BCOMPO, ROLDUA, FEMDIR and SAUDCO), and three control variables (FINDUS, FSIZE and FPROFI). The regression equations used are as follows:Model 1
Y(TOTCSR)=β0+β1 BSIZE+β2 NEXDIR+β3 ROLDUA+β4 FEMDIR+β5 SAUDCO+ε
Y(TOTCSR, TOTG1, TOTG2, TOTG3 and TOTG4)=β0+β1 BSIZE+β2 NEXDIR+β3 ROLDUA+β4 FEMDIR+β5 SAUDCO+β6 FINDUS+β7 FSIZE+β8 FPROFI+ε
Where Y=the CSR disclosure (the dependent variable: TOTCSR, TOTG1, TOTG2, TOTG3 and TOTG4); β0=a constant; βi, i=1, …, 8 is parameters; and ε is error term. In the OLS six regression models, Model 1 involves TOTCSR (the main dependent variable of 41 CSR items) with all of the five independent variables; Models 2-6 involve TOTCSR, TOTG1, TOTG2, TOTG3 and TOTG4 (as dependent variables) with five independent variables besides three control variables (FINDUS, FSIZE and FPROFI). Furthermore, regression diagnostics were applied to assess the possibility for multicollinearity among two or more independent variables. As variance inflation factors (VIF) were less than 2 and tolerance levels were more than 0.60 for all independent variables, inter-correlation among independent variables was not seem to be problematic and multicollinearity would not be a serious concern in this research.
Table 4 presents the findings of the descriptive analysis for all variables. It shows that across the 246 GCC companies included in the sample, the maximum percentage of female directors to total board members (FEMDIR) is only 25% with a mean of 2.14% indicating that the majority of board members in GCC listed companies are male directors. This result shows a small participation rate of female directors in listed companies in GCC. The minimum board size (BSIZE) was 5, while 12 board members was the maximum with a mean score of 7.93. The minimum percentage of non-executive directors to total board members (NEXDIR) is 11% and 100% is the maximum with a mean of 69.3%. This refers that the majority of board members in GCC listed companies are executive directors.
DESCRIPTIVE STATISTICS OF THE DEPENDENT, INDEPENDENT AND CONTROL VARIABLES
|- Environmental and related information (TOTG1)||0||9||3.51 (39%)||2.802|
|- Employee welfare information (TOTG2)||2||13||5.98 (46%)||3.584|
|- Community involvement and social information (TOTG3)||0||12||4.03 (31%)||3.138|
|- Products quality and safety information (TOTG4)||1||6||2.70 (45%)||2.048|
|- Total CSR Disclosure (TOTCSR)||3||36||15.99 (39%)||9.234|
|Independent and control variables:|
|- Board size (BSIZE)||5||12||7.93||1.717|
|- Non-executive directors (NEXDIR)||11%||100%||69.3%||26.3%|
|- Female directors (FEMDIR)||0%||25%||2.14%||5.16%|
|- Size of audit committee (SAUDCO)||2||6||3.42||6.88|
|- Firm size (FSIZE) ($000)||9,029||35,199,050||873,986||2,841,387|
|- Firm profitability (FPROFI) ($ 000)||-495,937||4,728,293||80,646||507,157|
|- Role duality (ROLDUA)||242 (98.4%)||4 (1.6%)|
|- Firm industry (FINDUS)||105 (42.7%)||141 (57.3%)|
Note: 1- The above results are based on 246 companies from the 6 GCC countries; 2- For more details on each variable, refer to Table 2 above.
The table shows that the size of members of an audit committee (SAUDCO) is ranging from 2 as the minimum and 6 as the maximum number with a mean of 2.14. This result indicates that all sampled companies in the six GCC countries are having an audit committee. Additionally, in most firms (242 of 246 sampled companies representing 98.4%) the CEO and the chairman of the board are not the same persons meaning that most listed companies in GCC have a separate chairman of the board and CEO. In relation to firm size (FSIZE), it is clear that $35,199 million was the maximum and $9.029 million was the minimum firm net sales revenue. The average firm profitability (FPROFI) for the total sample was $80,646, with a standard deviation of $507,157. Of the 246 sampled companies, 105 (42.7%) are industrial, while others are non-industrial firms.
As regards CSR disclosure by the sampled companies, Table 4 shows levels of CSR disclosure for the total of 41 items and for each of the four groups of CSR items. The mean score for the total CSR disclosure (TOTCSR) is 15.99 representing 39% of 41 information items included in the CSR disclosure index with a standard deviation of 9.234, while 36 (87.8%) items is the highest total score and the lowest score is 3 (7.3%). This study reports higher level of CSR disclosure than previous research accomplished in GCC. For instance, previous studies concluded that, on average, sampled companies provided only 26% of 45 items included in the CSR disclosure index (Khasharmeh and Suwaidan, 2010); and 33.2% of 47 items (Khasharmeh and Desoky, 2013). In addition, Macarulla and Talalweh (2012) and Al-Janadi et al. (2013) reported a low level of CSR disclosure in KSA at only 14.61% (Al-Janadi et al. 2013). The above result may suggest that listed companies in GCC are now more involved in the CSR disclosure which may suggest that they are now more socially responsible in their actions. Furthermore, Table 4 shows that the second group “Employee welfare information–TOTG2” has a mean score 5.98 representing 46% of 13 items included in the group, which is the highest group in disclosure. The fourth group “Products quality and safety information–TOTG4” has a mean score of 2.70 representing 45% of 6 items as the second highest group in disclosure. The first and the third groups were found as the third and fourth in disclosure respectively.
Table 5 reveals detailed results, which are varied from one item to another not only in the overall level (41 items) but also in each group of CSR disclosure. It is clear that 83% of the sampled companies, the highest level of disclosure, provided information on item no. 20 “Staff qualifications” which is the most disclosed item and ranked first. Items nos. 21 “Pension schemes and related information” and 10 “Training programs for staff” are provided by 76% and 70% of the sampled companies and are ranked as second and third most disclosed items; followed by items nos. 24 “The company relation with the local community” and 37 “Product quality and related information” as fourth and fifth. It is clear that three of the highest five items of CSR disclosure are included in the second group of information. The lowest items disclosed are nos. 27, 29, 15, 16 and 26, which have the lowest mean of 17%, 16%, 13%, 11% and 8% respectively.
CSR ITEMS DISCLOSED BY THE SAMPLE COMPANIES
|Items||Mean||SD||Group Rank||Overall Rank|
| A. Environmental and related information:
1. A policy concerning the environment.
2. Involvement in the environment protection programs.
3. Conservation of natural resources.
4. Recycling plant of waste products.
5. Participation on pollution control plans.
6. Energy saving.
7. Carefulness on disposal of waste.
8. R&D on protecting the environment.
9. Obedience with environmental and related regulations.
B. Employee welfare and related information:
10. Training programs for staff.
11. Various educational facilities.
12. Health care inside or outside the company.
13. Safety and security arrangements.
14. Support staff entertainment and holidays.
15. Staff cultural activities.
16. Providing loans to staff with special interest rates.
17. Housing facilities.
18. Founding of training and professional centers.
19. Clear policies for remuneration package/scheme.
20. Staff qualifications.
21. Pension schemes and related information.
22. Permanence of employees’ job.
C. Community involvement and social information:
23. The company various endowments.
24. The company relation with the local community.
25. Supporting educational and cultural activities.
26. Transport facilities for staff families.
27. Participating in medical and health centers.
28. The company’s gifts and cash rewards.
29. Founding public Halls.
30. Funding various educational scholarships for students.
31. Participation on reducing the community unemployment rate.
32. Contribution toward community serving programs.
33. Establishing new projects in poor areas.
34. Providing the community with financial supports.
35. Participating and financing community celebration.
D. Products quality and safety information:
36. Efforts to develop the company’s products including its packaging.
37. Product quality and related information.
38. Alertness to customer criticisms or complaints.
39. The company’s role in controlling prices and optimizing profits.
40. Customer service programs, market, product, warranty, etc.
41. Obedience with customer protection and related regulations.
The univariate analysis presents information on the association between CSR disclosures (the dependent variable), CG factors (dependent variables) and firm factors (control variables). Table 6 shows, as predicted, significant association between CSR disclosure and three independent variables namely BSIZE, NEXDIR and FEMDIR, however, this association is not strong as the coefficient value is 0.196, -.289 and -.171 respectively. At the same time, no significant association is found for other independent variables (ROLDUA and SAUDCO). In contrast to the above findings, very weak non-significant positive associations are reported between the dependent variable (CSR disclosure) and FINDUS, FSIZE and FPROFI (control variables) with coefficient values of 0.002, 0.106 and 0.052 respectively.
CORRELATION BETWEEN CSR DISCLOSURE (THE DEPENDENT VARIABLE) AND OTHER VARIABLES
*=Correlation is significant at the 0.05 level (2–tailed); **=Correlation is significant at the 0.01 level (2–tailed)
Notes: 1-Significant correlations are in bold; 2-Dependent, independent and control variables are defined in Tables 2 and 3;
3-Pearson correlation was performed for all variables; 4-All coefficients are based on 246 observations (2016 financial period).
A significant note on the univariate analysis is although the findings verify some significant association amongst independent variables (e.g. BSIZE vs. SAUDCO and NEXDIR vs. FEMDIR) and control variables (e.g. FSIZE vs. FPROFI), these associations, which is 0.185, 0.223 and 0.592 respectively, do not exceed 0.7 and do not show a serious problem of multicollinearity in the current study. In conclusion, among the five independent variables, three are significantly associated with CSR disclosure. This finding partially supports what was argued earlier by Bushman and Smith (2001) and Hassan (2012) suggesting that firms with best practice of CG are intended to disclose more information to stakeholders for different reasons; and (Albassam, 2014) who argued that some CG factors are main determinants of corporate disclosure practice. Finally, the above finding is consistent with what was reported in the USA by Mallin et al. (2013) and Giannarakis et al. (2014) as CG factors are significantly associated with CSR disclosure.
The Multivariate Results
Regression findings are presented in Table 7 below. Two general notes about regression results: First, the first five models are significant (p ≤ 0.05) while only the last one is insignificant (p=0.07); Second, results of the two main models (Models 1 and 2) are mostly similar. Models 1 and 2, which involve the main dependent variable “TOTCSR” and are significant at p ≤ 0.05, have adjusted R<sup>2</sup> (the explanatory power of the model) of 0.179 and 0.191 and F value of 5.072 and 3.945 respectively. In both Models 1 and 2, F value is higher than 1 indicating that both are good regression models. Field (2005) pointed out that a good model should have F value bigger than 1. These results suggest that about 17.9% of the variation in the total CSR disclosure among sampled companies can be explained by the five CG variables (the first model) and about 19.1% of the variation in the total CSR disclosure scores among sampled companies can be explained the five CG variables and three control variables (the second model).
Furthermore, the above findings show a significant positive association between board size (BSIZE) and CSR disclosure suggesting that companies with higher board size are more likely to disclose more CSR information. The above result supports the idea that a large board size improves the position of a wide range of points of view between the board members, which help in sharing the diverse members’ experiences and knowledge. This may increase the opportunity to disclose more CSR information. In the light of the above, it is possible to accept H1 suggested earlier in this research. This finding is in line with what was revealed in the literature by Das et al. (2015), Esa et al. (2012) and Cheng and Courtenay (2006) who reported a significant positive association between board size and CSR disclosure. Also, the above result is consistent with what was reported in KSA by Al-Janadi et al. (2013) who reported significant positive association between board size and CSR disclosure; in Malaysia by Yusoff et al. (2015) who reported a positive significant association between board size and environmental disclosure.
A significant negative association between non-executive directors (NEXDIR) and CSR disclosure is found suggesting that companies with fewer ratios of non-executive directors are providing more CSR information. This leads to accept H2. The above result supports what was revealed by some of previous studies in this area. For instance, it supports results revealed in Malaysia by Haniffa and Cooke (2005) and Eng and Mak (2003); in KSA by Al-Janadi et al. (2013) who found significant association between the existence of non-executive directors and CSR disclosure; in Pakistan by Lone et al. (2016) who showed an association between independent directors and the extent of CSR disclosure; finally in Bangladesh by Khan (2010) who revealed a significant association between the extent of CSR disclosure and the percentage of non-executive directors. However, the above result does not support findings reported in Malaysia by Said et al. (2009) and Yusoff et al. (2015) and in KSA by Habbash (2016) who reported no association between board independence from one side and CSR disclosure from the other.
Additionally, the regression results (all models) highlight female directors variable (FEMDIR) is not significantly associated to CSR disclosure (p ≥ 0.05 in all regression models). This means that being a male or female is not correlated with CSR disclosure. In the light of the above result, it is possible to reject H4 suggested earlier in this research. This result is compatible with what was revealed in other areas of the world by Khan (2010) in Bangladesh; Giannarakis et al. (2014) in the USA and Yusoff et al. (2015) in Malaysia who found that the presence of female members on the board does not affect the CSR disclosure. However, this result is not in line with Bear et al. (2010) who argued that the higher the ratio of female directors, the more the social and charitable activities. Further, this result is consistent with Lone et al. (2016) who found a positive association between women directors and the extent of CSR disclosure in Pakistan. A possible reason may be attributed to the insignificant existence of female directors in GCC listed companies, resulting in them being incapable of affecting the company’s process of decision-making.
Concerning other independent variables, role duality (ROLDUA) and size of an audit committee (SAUDCO), all regression models revealed that both variables are not significantly correlated to CSR disclosure (p ≥ 0.05 in all regression models) leading to reject H3 and H5. This result refers that CEO/Chairman separation and size of audit committee may not be determinant of CSR disclosure by listed companies in GCC. The above result is consistent with the findings reported by a number of previous studies including Habbash (2016); Giannarakis et al. (2014) and Bukair and Abdul-Rahman (2015) who did not find significant association between role duality and CSR disclosure. However, this finding is not in line with what was reported by Forker (1992); Haniffa and Cooke (2002) and Said et al. (2009).
Based on the above regression results, it can be concluded that of the five CG independent variables used in the current study, only two (board size and non-executive directors) are found as determinant of CSR disclosure in the GCC countries.
This study examines the association between CG and CSR disclosure by listed companies in the GCC using a CSR disclosure index of 41 information items. The results indicate that the mean score for the total CSR disclosure is 39% of 41 items included in the CSR disclosure index indicating an improvement in the CSR disclosure in 2016 by listed companies in the GCC countries. Regression results showed that only two CG independent variables (board size and non-executive directors) are found as determinant of CSR disclosure in the GCC countries.
Regarding theoretical implications, the findings of the current study represent one step beyond previous research on the association between CG variables and CSR disclosure in GCC area, which is considered an important subject for the community in general and the companies’ stakeholders in particular. The current study may contribute to the extant literature on this area of research. The findings show that a larger board is associated with more CSR disclosures. However, the percentage of non-executive directors to total members on the board has proven to be negatively associated with CSR disclosures. Further, the findings provide a theoretical support that female directors variable is not significantly associated to CSR disclosure which means that being a male or female is not correlated with CSR disclosure. In addition, our findings provide support that CEO/Chairman separation and size of audit committee may not be determinant of CSR disclosure by listed companies in GCC.
For practical implications, providing empirical evidence on this area of research within listed companies in GCC countries may add a new dimension to the financial reporting in general and CSR disclosure in particular. It is expected eventually that the findings from our study may offer relevant knowledge to a number of stakeholders including regulatory bodies, government and other interested parties concerning CG factors. The findings of this study could support company management, government and regulatory bodies to be well equipped to build future strategies and CSR disclosure policies in the GCC business environment especially, to the best of the researchers knowledge, the current study is one of the first in the area to investigate female directors as a GC variable and its impact on CSR disclosure in the GCC area. The current study has put forward recommendations to companies, practitioners, the government and other policymakers regarding the potential of board size and non-executive directors with respect to better CSR disclosure. For instance, the results confirm that the CG practices by listed companies in GCC countries, at least some factors of CG, are associated with their CSR disclosure and may be one of the reasons for the improvement in CSR disclosure. This may help CG regulators in GCC countries to improve CG mechanisms. For stakeholders, our findings suggest that they should exert more pressure on the management of listed companies to provide further CSR information. Most essentially, the study provides stakeholders with better understanding of the association between female directors variable and CSR disclosure.
The current study suffers from a number of limitations. First, the sample used is somewhat small. Second, the current study evaluates financial reports of one financial period, 2016, neglects the earlier periods. Third, this study concentrates on the quantity and the extent of CSR disclosure neglecting their qualitative features. Fourth, our CSR disclosure index includes a small number of CSR disclosure items, 41 items that needs to be extended. Fifth, other CG variable such as board commitment to CSR and board nationality were out of the focus of the current study. Future research could overcome the above limitations and enrich previous conclusions by expanding the sample size; increasing number of items included in the CSR disclosure index; and undertaking a comparative study between GCC countries and other countries in the area and between the period before and after the application of the CG code in some countries. Besides, it would be stimulating to study other CG variables and company characteristics variables (i.e. issue of new shares, firm age and foreign listing) and ownership variables not included in the current study. Finally, it will be interesting to investigate the effect of country characteristics such as regulations, educational system, political regime and cultures as determinant of CSR disclosure.
1. The Gulf Cooperation Council includes six countries: Kingdom of Bahrain (Bahrain), Kingdom of Saudi Arabia (KSA), Kuwait, Oman, Qatar and United Arab Emirates (UAE).
2. The main source for information in this section is the Websites of GCC six capital markets.
Albassam, W. (2014). Corporate governance, voluntary disclosure and financial performance: an empirical analysis of Saudi listed firms using a mixed-methods research design. PhD diss., University of Glasgow.
Andres, P., Azofra, V., & Lopez, F. (2005). Corporate boards in OECD countries: size composition, compensation, functioning and effectiveness. Corporate Governance: An International Review, 13, 197-210.
Bassett, M., Koh, P., & Tutticci, I. (2007). The association between employee stock option disclosures and corporate governance: Evidence from an enhanced disclosure regime. The British Accounting Review, 39(4), 303-322.
Bedard, J., Chtourou, S., & Courteau, L. (2004). The effect of audit committee expertise, independence, and activity on aggressive earnings management. Auditing: A Journal of Practice& Theory, 23(2), 13-35.
Desoky, A., & Mousa, G. (2013). The impact of firm characteristics and corporate governance attributes on internet investor relations - evidence from Bahrain. International Journal of Business and Emerging Markets, 5(2), 119-147.
DeZoort, F., & Salterio, S. (2001). The effects of corporate governance experience and financial reporting and audit knowledge on audit committee members’ judgments. Auditing: A Journal of Practice & Theory, 20(2), 31-47.
Esa, E., Anum, N., & Ghazali, M. (2012). Corporate social responsibility and corporate governance in Malaysian government linked companies. Corporate Governance: The international journal of business in society, 12(3), 292-305,
Hassan, Y., Hijazi, R., & Naser, K. (2017). Does audit committee substitute or complement other corporate governance mechanisms: Evidence from an emerging economy. Managerial Auditing Journal, 32(7), 658-681
Khan, M. (2010). The effect of corporate governance elements on corporate social responsibility (CSR) reporting: empirical evidence from private commercial banks of Bangladesh. International Journal of Law and Management, 52(2), 82-109.
Khasharmeh, H., & Desoky, A. (2013). Online corporate social responsibility disclosures: The case of the gulf cooperation council (GCC) countries. Global Review of Accounting and Finance, 4(2), 39-64.
Khasharmeh, H., & Suwaidan, M. (2010). Social responsibility disclosure in corporate annual reports: Evidence from the gulf cooperation countries. International Journal of Accounting, Auditing and Performance Evaluation, 6(4), 327-345.
Ienciu, I. (2012). The relationship between environmental reporting and corporate governance characteristics of Romanian listed companies. Accounting and Management Information Systems, 11(2), 267-294.
Lone, E., Ali, A., & Khan, I. (2016). Corporate governance and corporate social responsibility disclosure: Evidence from Pakistan. Corporate Governance: The International Journal of Business in Society, 16(5), 785-797.
Mallin, C., Michelon, G., & Raggi, D. (2013). Monitoring intensity and stakeholders’ orientation: how does governance affect social and environmental disclosure? Journal of Business Ethics, 114(1), 29-43.
Naser, K., Al-Hussaini, A., Al-Kwari, D., & Nuseibeh, R. (2006). Determinants of corporate social disclosure in developing countries: the case of Qatar. Advances in International Accounting, 19, 1-23.
Said, R., Zainuddin, Y., & Haron, H. (2009). The relationship between corporate social responsibility disclosure and corporate governance characteristics in Malaysian public listed companies. Social Responsibility Journal, 5(2), 212-226.