“Corporate social responsibility disclosure and profitability: Evidence from Islamic banks working in Yemen”

This study aims to examine the influence of corporate social responsibility (CSR) dis- closure determinants on profitability of Yemeni Islamic financial institutions. The empirical study was based on a balanced panel for twelve years from 2005 to 2016. Banks’ profitability is measured by four indicators such as return on assets (ROA), return on equity (ROE), profit after tax (PAT), and earnings per share (EPS), while corpo- rate social responsibility, financial leverage, inflation rate, asset size, and age of Islamic banks are considered as independent variables. The results of this study with regard to ROA indicated that corporate social responsibility, asset size, inflation rate, and age of Islamic banks have a significant influence on profitability (ROA). With respect to ROE, the result indicated that financial leverage, asset size, and inflation rate are the most important variables affecting bank profitability (ROE). Concerning PAT, the outcome revealed that financial leverage and age of Islamic banks have a significant effect on profitability (PAT). Finally, the result with respect to EPS indicated that financial le- verage, asset size, inflation rate, and age of Islamic banks have a significant impact on bank profitability (EPS). The result will be beneficial to scholars, investors, stakehold- ers, managers, and policymakers in the Islamic financial sector. in activities, and proper disclosure has helped to improve their accounting-based financial performance proxied return on equity (ROE) and return on (ROA). The findings indicate that the Islamic bank’s background information, corporate governance information, corporate social disclosure, bank and bank age have a negative and substantial effect on on assets. The findings reveal that board diligence, the size of the audit committee, audit committee composition, and board diligence are linked to return on assets. The outcomes revealed that (a) CSR positively affects financial performance, (b) state ownership weakens the relationship between CSR and financial performance, and (c) industry competition strengthens the relationship between CSR and financial performance for both state-owned and non-state-owned firms. The findings revealed that the relationship between a firm’s operating expenditures and its profitability is non- linear, and with certain threshold values of CSR scores, operating spend has a more significant and negative effect on profitability. The study results showed that, with the CSRD as a dependent variable, return on assets (ROA) and net interest margin (NIM) have an opposite influence, but return on equity (ROE) has no effect on CSRD, while CSRD has a different influence through ROA, ROE, and NIM on the profitability dependent variable.


INTRODUCTION
According to the growing public concern about corporate impacts on society and the environment, "corporate social responsibility" (CSR) disclosure and social reporting have become a crucial issue for many companies. Corporate social responsibility has been around for decades and is now rising at an incredible rate with no signs (Quazi et al., 2015). Avoiding unethical behavior and obtaining 'social authorization' (Gunningham et al., 2004, p. 308) have led to an increased number of firms getting involved in preparing CSR sustainability reports published.
The main objective of corporate social responsibility disclosure is to achieve an "ethical and socially responsible approach to the stakeholders of the company" (Hopkins, 2004). The principles of "corporate social responsibility" are consistent with the services of Islamic institutions in different countries following Islamic law. Each person is responsible for promoting equality and prosperity in society, as well as seeking the blessings of God for the sake of prosperity in this life and the hereafter. Despite the importance of "corporate social responsibil-

LITERATURE REVIEW
Many empirical studies examined the association between CSD in Islamic banks and banks' profitability in different countries such as Amran Bolanle et al. (2012) indicated the correlation between "corporate social responsibility" and benefit. The regression results indicated that "corporate social responsibility" spending has an effect on profitability in the Nigerian banking industry (Beta = 0.945, p.01). Rahman and Bukair (2013) studied empirically and reported the impact of the "Shariah supervisory board" (SSB) and its features on the level of CSR disclosure. The findings revealed that the volume of CSR data reported in Islamic banks' annual reports has risen. Ompusunggu (2016) investigated the effect of benefit on the disclosure of "corporate social responsibility" (CSR) transparency. The findings of this review show that CSR transparency has a major effect on profitability through "ROA, ROE, and NPM". ROA and NPM have a positive impact on CSR transparency, while ROE has a negative effect. Chiang et al. (2015) investigated the connection between "corporate social responsibility and financial reporting quality." Companies can easily minimize their level of earnings control and offer quality financial reporting by practicing CSR, according to the analytical findings of the total samples. Platonova et al. (2016) demonstrated the link between "corporate social responsibility" (CSR) and financial success in Gulf Cooperation Council Islamic banks (GCC). The banks sampled were recorded in the analysis. According to the results of this report, there is an important positive link between CSR transparency and Islamic banks' financial success in the GCC region. Amran et al. (2017), exploring two developed nations, Indonesia and Malaysia, considered the social responsibility monitoring of full-fledged Islamic banks. The study's findings revealed that Islamic banks' CSR disclosure has risen in Malaysia and Indonesia. The workplace and group aspects were discovered to be the most widely revealed fields by Islamic banks in both countries. Wuttichindanon (2017) examined the determinants of "corporate social responsibility" (CSR) disclosure of companies listed on the Thai Stock Exchange, as well as report options used for CSR disclosure (SET). The findings revealed that big or government-owned businesses are more likely to favor the sustainability study. Chen et al. (2018) examined the effect of mandatory "corporate social responsibility" (CSR) transparency on firm results and social external costs. Mandatory CSR reporting companies suffer a drop in productivity since the mandate, according to the study. Maqbool and Zamir (2019) in-vestigated Indian corporations' "corporate social responsibility" (CSR) transparency. The findings show that 'mining and mineral' companies are the most vocal about CSR disclosure, accompanied by 'power sector' firms. Maqbool and Bakr (2019) investigated the correlation between "corporate social responsibility (CSR) and financial performance" (FP), which has sparked a lot of discussion among academics. The results show that CSR and FP have a curvilinear relationship, implying that two long opposing views can be complementary. Lin et al. (2019) tested the influence of CSR factors on a company's financial performance. According to the report, allocating resources to diversity, human rights, workplace treatment, wages, insurance, recruitment, wellness, and workforce protection will help an organization to build value.
However, most empirical studies examined the association between CSR and financial performance such as Ridwan and Mayapada (2020); also, the influence of sharia governance on corporate social responsibility transparency in the Indonesian Islamic industry was examined. Szegedi et al.
(2020) tested the influence of "sharia governance on corporate social responsibility disclosure in Indonesian Islamic banks". The results show that financial institutions are involved in CSR operations and that adequate transparency has helped to boost their accounting-based financial efficiency, as measured by "return on equity (ROE) and return on assets (ROA)". The researchers looked at whether the effect of a company's CFP varies based on its degree of CSR output and annual operational expense. The results found that the correlation between a company's operating expenses and productivity is non-linear and that operating spending has a more substantial and detrimental effect on the financial performance at certain CSR score thresholds (Lee & Yang, 2021). The partnership between benefit and "corporate social responsibility" transparency was investigated. Return on assets (ROA) and net interest margin (NIM) have opposite effects on the CSRD dependent factor, but return on equity (ROE) has a little impact on CSRD, whereas CSRD has a separate impact on the productivity dependent factor than ROA, ROE, and NIM (Tran et al. 2021). Table 1 shows the summary of empirical studies that examined the link between CSD and financial performance in different countries.
Accordingly, this paper aims to study the link between corporate social responsibility disclosure factors with Islamic banks' profitability in Yemen during the period of examination from 2006 to 2016. Bank profitability is considered as a dependent variable, whereas corporate social responsibility factors are considered as independent variables. This review bridges the existing gap in the corporate social responsibility disclosure and bank profitability literature in the Republic of Yemen.

METHODS
This study applies a quantitative method. The Bank Scope Database and the websites of the sample Islamic banks are used to compile annual financial reports. The sample of the study consists of three Islamic financial institutions during the period from 2005 to 2016. The study is based on secondary data collected from the annual reports. The dependent variable of this study is bank profitability measured by four indicators such as return on assets (ROA), return on equity (ROE), profit after tax (PAT), and earnings per share (EPS). The independent variable of this study is measured by five factors such as corporate social responsibility, financial leverage, inflation rate, asset size, and age of Islamic banks.
The un-weighted content analysis approach is used to analyze the CSR disclosures through annual reports, ensuring the quality and reliability of the analysis. As a result, '1' is assigned to each CSR disclosed in the financial statement, and '0' is assigned to the annual report if no CSR disclosure items are disclosed otherwise (see Cooke, 1989) as follows: where TVDS is the cumulative score for corporate social responsibility disclosure), dj is 1 if shown by Islamic firms, or 0 otherwise, and n is the maximum score for each Islamic institution.
A total of 10 disclosure index items of CSR are developed based on previous studies such as Hossain (2008), Hossain  The results showed significant differences in the level and the extent of the disclosure between IFIs and CFIs, largely due to the disclosure made by IFIs of religions-related themes and information, including Shari'a supervisory board reports, the Zakah and charity donation, and free interest loan.

Mirfazli (2008) Indonesia 42 companies 2004
The results indicated that the main foci of social disclosure from companies registered at the Indonesia Stock Exchange are labor theme (51.60 percent), followed by customer theme (19.40  The results of this study showed that there are significant effects on profitability through ROA, ROE, and NPM on the disclosure of CSR. ROA and NPM have positive effects on the disclosure of CSR, while ROE negatively affects CSR disclosure. The results revealed that CSR disclosure of Islamic banks has generally grown both in Malaysia and Indonesia. More specifically, it was found that workplace and community dimensions were the most highly disclosed areas by the Islamic banks in both countries.

Author Country Sample Data Period Results
Wuttichindanon (2017) Thailand 451 firm

Secondary 2014
The results revealed that government-owned firms or large firms are more likely to prefer the sustainability report. The study found that mandatory CSR reporting firms experience a decrease in profitability after the mandate. In addition, the cities most impacted by the disclosure mandate experience a decrease in their industrial wastewater and SO2 emission levels. Maqbool  The results suggested an increase in overall CSR disclosure by all banks in the sample, and the involvement of commercial banks in CSR activities, and its proper disclosure has helped to improve their accounting-based financial performance proxied by the return on equity (ROE) and return on assets (ROA). Al-Homaidi et al. The findings indicate that the Islamic bank's background information, corporate governance information, corporate social disclosure, bank size, and bank age all have a negative and substantial effect on return on assets. Al-Homaidi et al. The findings reveal that board diligence, the size of the audit committee, audit committee composition, and board diligence are linked to return on assets. The outcomes revealed that (a) CSR positively affects financial performance, (b) state ownership weakens the relationship between CSR and financial performance, and (c) industry competition strengthens the relationship between CSR and financial performance for both state-owned and non-state-owned firms.
Lee and Yang (2021) Taiwan TOP 50 2013 to 2017 The findings revealed that the relationship between a firm's operating expenditures and its profitability is nonlinear, and with certain threshold values of CSR scores, operating spend has a more significant and negative effect on profitability. The study results showed that, with the CSRD as a dependent variable, return on assets (ROA) and net interest margin (NIM) have an opposite influence, but return on equity (ROE) has no effect on CSRD, while CSRD has a different influence through ROA, ROE, and NIM on the profitability dependent variable.

Multiple regression and measurement of variables
Multiple regression was applied to examine the association between Islamic banks' profitability and corporate social responsibility disclosure in Yemen.  Table 2.   Table 4 presents the results of the link between dependent (bank profitability) and independent (corporate social responsibility disclosure) variables in the Islamic industry in Yemen during the period of the study. The results of the study indicated that there is a positive link between profit after tax, earnings per share, and financial leverage with bank profitability (ROA, ROE, and PAT), while corporate social responsibility, inflation rate, and age of Islamic bank have a negative correlation with profitability defined by ROA, ROE, and PAT. However, financial leverage, inflation rate, asset size, and age of Islamic banks have a positive relationship with earnings per share (EPS), except it has a negative link with corporate social responsibility transparency in Islamic firms in Yemen. The results indicated no multicollinearity diagnosis as the p-value of variables is less than 80. Table 5 shows the outcomes of model 1 and model 2. The result shows "pooled and fixed" effect models for two estimates. The findings revealed that the adjusted R square for the ROA model is 0.37 for a pooled model and 0.46 for a fixed model, while the outcomes offered that the adjusted R square for ROE is 0.47 for pooled and 0.45 for fixed-effect models. The results for both models 1 and 2 are fit because the p-value is less than 0.05.

Determinants of ROA and ROE
With regard to an ROA pooled model 1, the results show that asset size and age of Islamic banks have a highly significant effect on profitability at the level of 1% (p-value < 0.01). The results also indicated that corporate social responsibility and age of Islamic banks have a negative correlation with firms' financial performance, whereas financial leverage, asset size, and inflation rate have a positive association with firms' financial profitability (ROA). With regard to (ROA) a fixed model 1, the outcome revealed that asset size and age of Islamic banks have a highly significant influence on firm profitability at the level of 1% (p-value < 0.01), corporate social responsibility has a significant effect on ROA at the 10% (p-value < 0.1), inflation rate has a significant effect on ROA at 5% (p-value < 0.05). The results also indicated that corporate social responsibility and age of Islamic banks have a negative link with ROA, whereas financial leverage, asset size, and inflation rate have a positive link with bank profitability (ROA).
With regard to ROE, the results of the study found that financial leverage and age of Islamic bank have a negative and significant effect on banks profitability (ROE) in pooled and fixed models, except the age of bank has a significant impact on ROE in a pooled model and insignificant effect in a fixed model. Asset size and inflation rate indicators have a significant impact on banks' financial profitability (ROE) in pooled and fixed models. With regard to (PAT) pooled model 3, the result suggested that financial leverage, asset size, and age of Islamic industry have a significant influence on financial institutional banks' profitability (PAT), while corporate social responsibility and inflation rate have an insignificant influence on profitability (PAT). The results also indicated that corporate social responsibility and asset size have a negative association with profitability (PAT),  whereas financial leverage, inflation rate, and age of Islamic banks have a positive effect on bank profitability (PAT). With respect to (PAT) fixed model 3, the outcomes of the study showed that financial leverage and age of Islamic banks have a significant impact on PAT, while corporate social responsibility, asset size, and inflation rate have an insignificant influence on PAT. The findings also indicated that corporate social responsibility has a negative effect on PAT, whereas financial leverage, asset size, inflation rate, and age of Islamic banks have a positive link with banks profitability (PAT).

PAT and EPS determinants
With regard to (EPS) pooled model 4, the outcome indicated that corporate social responsibility, financial leverage, and asset size have a strongly significant effect on firms' financial profitability (EPS), except the inflation rate has an insignificant correlation with EPS. Corporate social responsibility and asset size have a negative association with EPS, whereas financial leverage, inflation rate, and age of Islamic banks have a positive relationship with EPS. The results of a fixed model 4 suggest-ed that financial leverage, asset size, inflation rate, and age of Islamic banks have a significant impact on profitability (EPS), except corporate social responsibility has a negative and insignificant effect on EPS. The results also found that financial leverage, asset size, and inflation rate have a positive effect on financial profitability ratio (EPS), except age of Islamic banks has a negative effect on EPS.

DISCUSSION
This section discusses the current review for the period of this study. During the research, a highly important association was found between Islamic institutions' profitability and corporate social responsibility transparency.  According to the findings, the average degree of corporate social transparency is low. The low-level transparency index items are used for a number of reasons. Non-compliance in Yemen may be the most critical explanation for those planning the financial reports of Yemeni Islamic banks who do not provide this kind of detail in external annual reports. Furthermore, the high cost of preparing and releasing reports to the public is another potential explanation for why commercial bank management may not be able to include such information.

CONCLUSION
The objective of this research is to explore the effect of corporate social responsibility transparency factors on Yemeni financial institutions' profitability over the twelve years period from 2005 to 2016. The results of this study show that corporate social responsibility, asset size, inflation rate, and age of Islamic banks have a significant influence on profitability (ROA). It is also suggested that financial leverage, asset size, and inflation rate are the most important variables affecting