Keywords

1 Introduction

Corporations become is an integral part of the society and the only way to achieve sustainability and survive in the society respects the ethical values of the society (Davis 1975, p. 13). There is a widespread academic interest in the corporate social responsibility. With respect to the CSR, the type of researches which has observant growth focuses on the determinants of CSR disclosures of the firms. The relationship of CSR disclosure as determined by firm characteristics whether financial such as liquidity, leverage, and profitability or not financial such as size, ownership, and audit firm size, has been exceedingly inspected and reported in the researches which investigate the CSR (Haniffa and Cooke 2005; Mahadeo et al. 2011; Alareeni 2019).

The phenomenon of CSR is attracting increasing international attention. Today, corporations are expected not only to focus and pursue profit, but also to consider the CSR. In general, CSR means that corporations take into consideration the fears and concerns of corporate stakeholders (e.g., employees, shareholders, suppliers, government, customers, and the local community). The CSR principles include the fairness of social and environmental sustainability into the business process (Alkababji 2014; Awadh and Alareeni 2018). However, as the societies’ awareness about the importance of CSR disclosure increase, the researchers become more focusing on the topic of CSR, such as, (Razak 2015) examined the association between some firm characteristics and CSR disclosure in Saudi Arabia, also from Saudi Arabia, Macarulla and Talalweh (2012) examined the level of CSR disclosure of the 134 firms that were listed on the RSE, Saleh (2009) provided empirical evidence on CSR disclosure practices in Malaysia. Sufian (2012) investigated the association between firm’s characteristics and corporate social responsibility disclosure (CSRD) in Bangladesh, moreover, Yao et al. (2011) studied the determinants of CSR disclosure by Chinese firms.

This research attempts to review the relationship between firms’ characteristics and the level of CSR disclosure by firms listed on Bahrain Bourse. The purpose of this research is set out to examine the CSR disclosure practices in Bahrain by analyzing the annual reports of different types of firms listed on Bahrain Bourse. That is to find out whether the level of CSR disclosure is influenced by firm-specific characteristics or not?. The study, therefore, examines the association between an eight firm-specific characteristics and the level of CSR disclosure. These characteristics include: firm size, profitability, financial leverage, firm age, size of audit firm, ownership, industry type, and liquidity.

The next section discusses previous studies as well as formulating research hypotheses to answer the research question.

2 Literature Review and Hypothesis Development

This study attempts to review previous studies that have been published by well-known journals in the area of CSR and identify the theoretical framework which had led to the development of the hypotheses to be tested and analyzed.

Since the CSR disclosure is voluntary, there will be no power to requiring companies to disclose CSR, which might affect its flow, or make it hostage to the interests of companies with a blind eye to the interests of stakeholders. Moreover, there is a lack in researches that investigate the CSR reporting of firms in Bahrain.

CSR defined as the voluntary obligation of firms to contribute to social goals and developments. In last few decades, corporations become more aware of the fact that it is an integral part of the society and the only way to achieve sustainability (Davis 1975, p. 13).

2.1 Evolution of the Concept of CSR

Madrakhimova (2013, p. 36), made a study to investigate the history of the CSR’s concept, he traced that the evolution of CSR started in 1950, in the 1960s, the definitions got expanded and it began expanding in the 1970s. In 1980, some fewer new definitions appeared and the empirical research became more mature. In the 1990s, CSR continued to serve as the basic design, but inferior or turns into the alternative thematic framework.

Mohamed et al. (2014, p. 56) summarizes the timeline of social accounting in two periods; the first period when the social accounting has been discovered in the 1960s, and it was charred only the areas that concern human dimension. Then, the second in the 1980s, it was extended to the protection of the environment in the name of environmental accounting. Thus, the object of a social accounting therefore encompasses environmental and social concerns.

In a related study, Maguire (2011), p. 214) link the CSR evolution with events happened during the last four decades as given in Table 12.1.

Table 12.1 CSR evolution
Full size table

This study contributes to the corporate social responsibility literature, because it provides insight into the CSR disclosure practices of listed companies with respect to their operations.

Furthermore, such this research is useful to the stakeholders to evaluate the level of voluntary CSR disclosure and the firm’s compliance with social responsibilities and making decisions. In addition, managers may realize the importance of environmental and social disclosures and learn the determinants which lead to better disclosure practices. This will result in better provision of CSR disclosure to stakeholders.

2.2 Theoretical Framework

Corporate social responsibility (CSR) can be defined as the voluntary commitment of firms to contribute to social goals and developments. In last few decades, corporations become more aware of the fact that it is an integral part of the society and the only way to achieve sustainability and survive in the society is respect the ethical values of the society, Davis (1975).

There is a widespread academic interest in the corporate social responsibility, the empirical investigations of CSR practices has produced a very diverse academic literatures that engages different theoretical perspectives in support of corporate social reporting. With regard to the empirical research on CSR, Reverte (2009) divided the empirical studies into three types, the first one related to ‘descriptive studies,’ which report on the extent and nature of CSR with some comparisons among periods and countries, the second one is related to ‘explicative studies,’ which interested in investigating the determinants of CSR reporting. The third one is interested in the ‘impact of CSR information’ on stakeholders.

Empirical studies have shown that CSR disclosure activism varies across companies, industries, and time, Gray et al. (1995), Hackston and Milne (1996). They have also shown this behavior to be importantly and systematically determined by a variety of firm and industry characteristics that influence the relative costs and benefits of disclosing such information, Belkaoui and Karpik (1989), Cormier and Magnan (2003).

Campbell (2007) offers a comprehensive institutional theory on CSR comprising a series of propositions specifying the conditions under which corporations are likely to behave in socially responsible ways. Chih et al. (2008) focus of this section, based on (Campbell 2007), is placed on providing an explanation of the determinants of CSR and proposing appropriate measures to proxy for these determinants in our empirical study.

This study adopts second type of empirical studies which mentioned by Reverte (2009), and it attempts to examine the association between eight firm-specific characteristics and the level of corporate social responsibility disclosure of firms listed in Bahrain Bourse in order to evaluate the practice of corporate social responsibility disclosure. These characteristics include: firm size, profitability, financial leverage, firm Age, size of audit firm, ownership, industry type, and liquidity which some of them are widely used in such researches.

2.3 Study Hypothesis

Firm Size

Yao et al. (2011, p. 214) identified the determinants of CSR disclosure of corporations in China, indicated that there is an affirmative relationship between the company size and the social and environmental information disclosure.

Rettab et al. (2009, p. 41) advise that the ability of large firms to communicate their social activities to stakeholders is more than small firms. On other hand, some studies showing opposite results, like Ebiringa et al. (2013, p. 61), they found that there is a negative relationship between number of corporate social disclosed and firm size in Nigerian oil and gas sector. Thus, we hypothesize that:

H1: There is a relationship between firm size and the level of CSR disclosure.

Profitability

Most of prior researches end up with similar results related to the variable of profitability. For example, Alareeni (2018), Hussainey et al. (2011, p. 41), Ismail and Chandler (2005, p. 32), and Roberts (1992, p. 124) found that profitability has positive relation with social reporting. Thus, it is hypothesized that:

H2: There is a relationship between firm’s profitability and the level of CSR disclosure.

2.3.1 Financial Leverage

Chek et al. (2013, p. 61) found that there is no correlation between firms’ leverage and the level of CSR disclosure, and their finding was agreed with the research made by Mustaffa and Tamoi (2006, p. 314). They found that leverage is not appeared as a determinant for the company to disclose their social activities. To examine the relation between leverage and the level of compliance with CSR disclosure among firms listed in the Bahrain Bourse, it is hypothesized that:

H3: There is a relationship between firm’s financial leverage and the level of CSR disclosure.

Firm Age

Firm age is defined by the time span between the sample’s year and when a firm was listed on stock exchange. Yao et al. (2011, p. 31) found that firm age is negatively correlated with the level of CSR disclosure in China, and justify it by the nature of China’s capital market, and therefore, it is hypothesized that:

H4: There is a relationship between firm’s age and the level of CSR disclosure.

Size of Audit Firm

The firm’s external auditor impacts the quality and quantity of information disclosure in annual report. Big and international audit firms require more information and adopt high quality of internal procedures in which it contributes the firm’s disclosure, Uwuigbe and Egbide (2012, p. 32). This leads to the fourth suggested hypothesis:

H5: Bahraini firms audited by large auditing firms disclose more social information than those audited by small auditing firms.

Ownership

Eng and Mak (2003, p. 32) argued that the government ownership will lead to high conflict between firms financial goals and society’s expectations. High portion of shares held by government will lead to high expectation of society in the aspect of social activities and programs such as training programs, donations, and pension plans (Naser 2006, p. 51). The following hypothesis is suggested:

H6: Firms with Bahraini ownership disclose more social information than firms owned by foreign.

Industry Type

Industry type is another variable used to explain the level of CSR disclosure; companies belong to different sectors and operating in different activities. Hence, companies that have manufacturing operations involved in more activities than companies with services operations, which may require them to report more social information especially in the environment and product aspect.

H7: Manufacturing industries disclose more social information than non-manufacturing firms.

Liquidity

Ezat and El-Masry (2008, p. 112) found a positive relationship between levels of company Internet reporting and liquidity. In order to test this relationship for companies listed in the Bahrain Bourse, it is hypothesized that:

H8: There is a relationship between firm’s liquidity and the level of CSR disclosure.

3 Research Methodology

The study sample divided into six sectors: (commercial banks, investment banks, insurance, services, industrial, and hotels & tourism) as given in Table 12.2. The subsequent multiple linear regression model was fitted to the data:

$$ \begin{aligned} {\text{CSR Dis }} & = \, \beta_{0} \, + \, \beta_{1} \, TA \, + \, \beta_{2} {\text{ Prof }} + \, \beta_{3} {\text{ Fin Lev }} + \, \beta_{4} {\text{ age }} \\ & \quad + \, \beta_{5} {\text{ Audit F siz }} + \, \beta_{6} {\text{ Own }} + \, \beta_{7} {\text{ Ind type }} + \, \beta_{8} {\text{ Liq }} + \, e \\ \end{aligned} $$
Table 12.2 Classification of sampled companies by sector
Full size table

where:

CSR Dis = Corporate social responsibility disclosure

TA = Total assets (firm size)

Prof = Profitabiltiy

Fin Lev = Financial leverage

Age = Firm age

Audit F Siz = Audit firm size

Own = Percentage of Bahrain ownership

Ind Type = Industry type

Liq = Liquidity

e = Error term (Fig. 12.1).

Fig. 12.1
figure 1

Research model

Full size image

The data for measuring the dependent and independent variables investigated in this study were collected manually from the sampled companies’ annual report downloaded from their official Web sites as well as the Web site of the BSE. The 35 firms’ annual reports were fully covered in the study because of its small sample size and secondly, the researcher sought to determine the level of corporate social responsibility disclosure of companies listed in Bahrain Bourse. The reports of the year 2017 were selected because they were relatively more recent at the time the study was conducted and they were easier to obtain.

Independent Variables:

Eight company characteristics (e.g., firm size, leverage, firm age, audit firm size, profitability, industry type, ownership, and liquidity) were examined for their association with the level CSR disclosure. The annual reports of the firms were used to obtain the data. Table 12.3 below includes independent variables of the study, their measurements, and their expected signs. It also shows prior studies that used similar measurement methods as follows.

Table 12.3 Summary of the independent variables
Full size table

4 Data Analysis and Discussion

4.1 Descriptive Statistics

This section discusses the descriptive statistics for the dependent and independent variables used in the study. Tables 12.4. and 12.5 report the minimum, maximum, mean, and standard deviation for the sample companies. Furthermore, it states the normality tests of the variables used in the research.

Table 12.4 Descriptive statistics for the dependent variables
Full size table
Table 12.5 Descriptive statistics for the independent variables
Full size table

Disclosure checklist which representing the dependent variables is divided into four categories:

  1. i

    Community involvement (CI): According to the descriptive results, the extent of first category, which is community involvement (CI) disclosure on average, is 54.54%, with a minimum of zero percent and a maximum of 100%, indicating variations in the level of (CI) disclosure in Bahrain.

  2. ii

    Employee information (EI): According to the descriptive results, the extent of second category, which is employee information (EI) disclosure on average, is 61.90%, with a minimum of 17% and a maximum of 100%, indicating variations in the level of (EI) disclosure in Bahrain.

  3. iii

    Product/service information (PSI): According to the descriptive results, the extent of third category, which is product/service information (PSI) disclosure on average, is 60.95%, with a minimum of zero percent and a maximum of 100%, indicating variations in the level of (PSI) disclosure in Bahrain.

  4. iv

    Environmental disclosure (ED): According to the descriptive results, the extent of Forth category, which is environmental disclosure (ED) on average, is 16.19%, with a minimum of zero percent and a maximum of 100%, indicating variations in the level of (ED) in Bahrain.

Summarizing the results of descriptive statistics for dependent variables, the amount of (EI) and (PSI) were approximately same mean, which indicating that around 61% of the tested samples disclose for them, followed by the (CI) which is disclosed by 54.55% of the tested companies, and the last category which is the (ED), got the lowest mean of 16.19%, this can be justified that the operations of most of the listed companies in Bahrain Bourse are not having a significant impact on the environment because of their nature which belong to the service field.

Eight firm-specific characteristics are representing the independent variables:

As indicated by the minimum and maximum values, there is a vast domain of variation within the independent variables.

  1. i

    The mean of liquidity was 2.46, with a minimum of 0.29 and a maximum 12.58.

  2. ii

    The mean of size was 141.6 BD millions, with a minimum of 0.6 BD millions, and a maximum 123 BD millions.

    The normality classifications of both liquidity and total assets (Size) were deviated. Subsequently, natural logarithm was used in the regression analysis to moderate skewness and to bring the distribution of the variables closer to normality.

  3. iii

    The mean leverage for the firms was 43.80 with a minimum 4.15, indicating companies with somewhat high debts and a maximum of 89.69, indicating firms with very high debts.

  4. iv

    The age ranges for the firm is from 8 to 57 indicating variations with a mean of 29.77.

  5. v

    With respect to auditor type, the investigator documents a mean of 0.83 and minimum (maximum) values of 0.00(1.00) proposing that around 83% of the listed firms in Bahrain are audited by a large four auditing companies.

  6. vi

    Profitability ranges from –16.23 to 20.90 with a mean of 8.27, exhibitionist that 19.21 times the amount of equity of the company was consumed due to operations.

  7. vii

    Bahraini ownership ranges from 0.00 to 1.00 suggesting 40% of tested samples owned locally.

  8. viii

    Last but not least, industry type with and minimum (maximum) values of 0.00 (1.00) suggesting only 6% of the tested samples belong to the manufacturing field.

Table 12.6 is summarizing the Pearson correlation matrices. This will help to examine the statistical relationship among the dependent and the independent variables, and whether multicollinearity exists between the data before estimating the model. As clarified in the results, it shows that there are some moderately high correlations between variables, more specifically between firm size (Size) and financial leverage (Lev), and between social and CSR disclosures (CSR dis) and Industry (Ind). Moreover, industry (Ind) also has significant correlation with liquidity (Liq). The correlation appeared between in low degree between leverage (Lev) and two others variables which are ownership (Own) and liquidity (Liq).

Table 12.6 Correlations
Full size table

4.2 Regression Results

As illustrated in the findings (Table 12.7), F-value is 2.183 (P < 0. 01). This outcome statistically supports the importance of the regression model. Moreover, the findings appear that R2 is 0.402, which proposes that independent variables included in the model explain 40.20% of the variation in disclosure index (Table 12.8).

Table 12.7 Model summaryb
Full size table
Table 12.8 ANOVAb
Full size table

Generally, the model of CSR disclosure is accepted; two out of the eight assumptions are agreeable. Particularly, industry and audit firm size are significantly linked to CSR disclosure. The remaining six independent variables (size, profitability, leverage, age, ownership, and liquidity), however, show statistically insignificant relationships to CSR disclosure.

Hypothesis 4 predicts a favorable relationship among company age and CSR disclosure. The findings suggest that firm age is not significant in explaining the variation in the extent of CSR disclosure. The results reported a positive association between firm size, profitability, leverage, ownership, and liquidity. However, the results are statistically insignificant.

This could be due to the actuality that older companies find it complicated to sit new procedures to take into account the social responsibility, and it takes time to acclimate with the growing awareness of societies as fast as younger firms (Table 12.9).

Table 12.9 Coefficientsa
Full size table

5 Conclusion, Implications, and Recommendations

A disclosure checklist was used to measure the level of CSR disclosure. Each of the 35 sampled firms’ annual reports was tested. A regression analysis was used to test the relationship between the level of disclosure and (firm size, age, leverage, the size of the audit firm, profitability, industry type, ownership, and liquidity). The regression analysis indicates that the disclosure level also varies by audit firm size and industry type. Audit company size is affirmatively related to the level of CSR disclosure. This suggests that big audit firms such as the big 4 encourage their clients to have a higher level of CSR disclosure. Industry type is also positively associated with the level of CSR disclosure, mainly in the fourth category of disclosure checklist which is the environmental field, due to the nature of their operations that have a high influence on the environment, they disclose more about the environmental information. Other company merits such as size, profitability, leverage, age, liquidity, and ownership are not significant in explaining the level of CSR disclosure.

The findings of the analyses provided in this research should be particularly relevant to CSR codes. The study, therefore, recommends the regulatory bodies in Bahrain, to work on set a code of social responsibility to be guidance for the corporations in their social responsibility.

From a theoretical point of view, this study contributes to the existing literature on the association of firm characteristics and CSR disclosure, by investigating firms listed in Bahrain Bourse. Therefore, it expands on CSR compliance studies in the Gulf region, particularly in Bahrain.