Methodology: In this study, Islamic Sharia-based banking companies listed on Dhaka Stock Exchange in Bangladesh is selected. Four listed Islamic Sharia-based banks (66.67%) among six listed banks are taken as sample randomly for this study. Content analysis is used to measure the extent of CSR disclosure of sample banks. A CSR checklist covering sixteen themes with eighty-five items is prepared and a period of 5 years ranges from 2011-2015 is taken for this study. Three accounting based financial performance indicators are used in this study such as return on assets, return on equity and earnings per share. Ordinary least square regression models and Panel data regression models (Fixed effects model or Random effects model) are used to analyze data. The Bruachpagan multiplier test and Hausman specification test have been conducted to employ appropriate model for data analysis. Findings: The findings of the study indicate that there is a statistically significant positive association between CSR disclosure and financial performance of Islamic sharia-based commercial banks in Bangladesh. The regression results show that CSR disclosure of sample banks is positively related to return on equity(ROE) and earnings per share (EPS) of Islamic Sharia-based commercial banks in Bangladesh at 10% (p=0.085) and 5% (p=0.042) level respectively. Listing age is positively associated with financial performance measures (return on assets at 1% level, return on equity at 1% level and earnings per share at 5% level). Despite showing the positive relationship between CSR disclosure and financial performance, the findings show no statistically significant linkage between CSR disclosure and return on assets. The ratio of independent director on board is negatively linked with earnings per share but not statistically significantly associated with return on assets and return on equity of Islamic shariabased commercial banks. The coefficient of size (log of total assets) is negatively associated with return on assets on hand and no significant association is found with return on equity and earnings per share on the other hand. # Introduction he development of CSR concept has a long history. Parker (2011) found four industrial pioneers as leading actors in CSR thought of their time. They are Robert Owen (1771-1858), Titus Salt (1803-1876), George Cadbury and William Hesketh Lever . Their charitable activities includes co-operative movement; child schooling; trade unionism; modernization of well-organized mill production technology; improvements in employees' quality of life by pursuing social welfare strategies for the workforce; implementing cooperative working practices; increasing employee benefits and sharing corporate prosperity with its workers; donating major gifts or property to a wide variety of organizations and charities; developing community facilities such as the building of hospitals, construction of a convalescent home for children, and many others (Parker, 2011). These four 'legends' are examples of such sophisticated business entrepreneurs who simultaneously conducted corporate operation to peruse business and social goals. It is astonishing that these people pursued these kinds of humanitarian activities willingly not from facing pressure from stakeholders or any legal bodies. Actually, they desired to integrate social and business goal in a single frame. In other words, these institutions seemed to To measure the level of CSR disclosure of Islamic Sharia-based commercial banks in Bangladesh, the annual reports of the sampled banks of the study were reviewed employing content analysis. Annual reports have been chosen as the source of information since these reports are frequently to communicate financial and non-financial information to external stakeholders. These reports are prepared under the II. Literature Review and Theoretical Perspectives a) Literature review Islam as complete code of life gives guidelines to its followers in all sphere of life (Hamidullah, 2005). In religious perspective, life has two parts -life before death and life after death. Whether it is human being or other creature of almighty, none can stay alive for unlimited time in this earth. It is also very easily understandable that one who creates something once; he or she can surely built it twice or more. In this context, the Almighty Lord says in the Holy Quran: ""Allah -there is no deity except Him. He will surely assemble you for [account on] the Day of Resurrection, about which there is no doubt. And who is more truthful than Allah in statement." [Quran,4: # 87]" The above mentioned Ayah indicates that the each human being of this earth will be reborn and will be evaluated for his beliefs and activities in this worldly life. A person will also be asked about fulfilling rights of fellow humans. Basically, Human being involves in business to earn profit. Business itself has a separate entity. Banking companies play key role in the money market. To conduct financial transaction, Islamic banking companies provide fruitful services to the customers. Majority of the people in Bangladesh is the follower of Islam, so, there is a satisfactory demand on Sharia-based financial institutions. In responding the emerging demand of Sharia-based banks in Bangladesh, a number of banks are operating. The main purpose of Islamic banking is to ensure financial services with justice and fairness to the society. It guarantees every customer to be served equally without any prejudice. Basically, research in social, environmental or non-financial disclosure in developing counties has appeared in late 1990s and early 2000s (with a few exceptions). As the basis of this research is supervision of accountants and verified by auditors and The banking industry is now a unique sector in an economy and plays a key role to ensure financial stability of a country. But now banking sector goes far beyond ensuring financial stability and involves in forming new strategies to provide desired services to various stakeholders. The banking industry is at the heart of society and it is expected to be more socially responsible (Chamber & Day, 2009). As a result, scholars all over the world have confessed and analyzed the significance of CSR programs of banking industry. In the present economic environment, it is crucial to integrate social, ethical, moral and environmental issues in their business activities (Evangelinos et al., 2009, p. 167). As a part of service industry, it is acknowledged that CSR is deep-rooted idea in the banking sector (Scholtens 2009, p. 159). Islamic banking, Islamic finance or Sharia complaint finance is getting popularity day by day. Islamic sharia-based financial institutions perform financial transactions by complying with Sharia or Islamic law. The general aim of Islamic financial institutions is to contribute to the growth of an economy under the umbrella of Islam. The principle of Islam is to exercise equal rights among individual and ensure adequate benefits to live in the society. So, Islamic financial institutions utmost duty and responsibility is to engage in social, ethical and environmental activities. In other words, the "Islamicity" of Islamic banks requires CSR as a part of its proposed empirical model (Platnova et el., 2018). Islamic financial Institutions (IMIs) operate social programs but may not disclose those activities in its annual reports or websites. To motivate the management of Islamic banks to increase the CSR disclosure in their reporting documents, International regularity authority such as Accounting and Auditing Organization for Islamic Financial Institutions (AAOIFI) have formulated standards for Islamic institutions. AAOIFI has issued standard No. 7 on regulation standard for Islamic Sharia-based banking companies in relation to CSR activities and disclosure in 2010. In this standard, CSR is described as "all activities carried out by an IFI to fulfill its religious, economic, legal, ethical and discretionary responsibilities as financial intermediaries for individuals and institutions." So, to act in accordance with the standard, Islamic banking companies necessitates to conduct various aspects of business functions differently in comparison to those of conventional banks. provide a consistent measure (Tilt, 1994). In doing the disclosure analysis, this study has formulated a comprehensive CSR index focusing on CSR-related expectations of the stakeholders. Earlier studies on CSR disclosure of Islamic financial companies especially Islamic banks are used as a basis or guide for this study such as (Hossain & Neogy, (2019) Platnova et al., (2018), Haniffa & Hudaib (2007), Aribi and Gao (2012) and Aribi & Arun (2015). The structure of the paper is as follows: Section 2 provides the literature review and theoretical perspectives. Objectives and methodology of the study are presented in the Section 3 and 4 respectively. Section 5 represents the results and discussion on findings. Section 6 provides the policy implications of the study and section 7 summarizes the study. on a developing country, namely Bangladesh, it is relevant to review some available literatures relating to developing country perspectives. A few earlier research works on CSR reporting and financial performance are reviewed in the following paragraphs. # Millin et al (2013) has written an article on "Corporate Social Responsibility and Financial Performance in Islamic Banks". The objective of the study is to investigate the relationship between CSR disclosure and financial performance of Islami banks. The CSR index indicates that Islamic banks involves in wide range of social activities both as individual banks and as countries. It is seen that Islamic banks are more committed to some dimensions such as vision and mission, board and top management, financial products or services and less attention is given on environment theme. The regression results show a positive association between CSR disclosure and financial performance. Siswanty (2018) has written an article on "The Influence of Financial Performance and Non-Financial Performance on Islamic Social Responsibility Disclosure: Evidence from Islamic Banks in Indonesia". The study aims to find out factors that influence the Islamic social disclosure in Islamic Banking companies and the linkage between social reporting and financial performance. The findings of the research indicate that financial performance of Islamic banks does not have any influence on CSR reporting while compliance of sharia principles, Sharia supervisory board have significant influence on Islamic social reporting. limitations of the study are: The results of the study may not be generalized as the sample size of the study is very small. As non-random sampling method was used, so the research is less reliable and relevant. Hasan (2013) has written an article on "Corporate Social Responsibility in the Banking Sector: A Comparative Study on Some Commercial Banks in Bangladesh". The objectives of the study are, (a) to ascertain the main areas where CSR programs are being accomplished by private commercial banks (PCB) and state-owned banks (SOB) and (c) to evaluate the contribution of CSR activities of PCB and SOB. This study found that the expenditure incurred by commercial banks to CSR programs is very immaterial. SOBs have not reached so ahead in participating in CSR areas. The limitations of the study are: (1) The study is exclusively descriptive in nature; (2) Only secondary sources of data are used; (3) The sample size of the work is very small and (4) The study covered only two years, so the findings cannot be generalized in the other periods. # b) Theoretical perspectives of CSR disclosure CSR disclosure has been subject to considerable academic research in the field of accounting. A number of theories have been proposed to explain why business organization report CSR information voluntarily. Islamic banks' literatures suggest two major issues that influence the Islamic banks CSR disclosure-socio-political context within which banks operate and economic opportunities available to Islamic Platnova et al., (2018) have written an article on "The Impact of Corporate Social Responsibility Disclosure on Financial Performance: Evidence from the GCC Islamic Banking Sector". The purpose of the study is to examine the relationship between corporate social responsibility (CSR) and financial performance of Islamic banks in Gulf Cooperation Council (GCC) region over the periods 2000-2014. The study reveals that there is a significant positive relationship between CSR and financial performance of GCC Islamic banks. The results also indicate that there is positive impact of current CSR disclosure on future financial performance of Islamic banks in the GCC countries. Furthermore, the findings of the study show that there is no statistical significant relationship between CSR dimensions and current financial performance of sampled banks except "mission and vision' and 'product and services' themes. Shafiqur, Sadia & Nicholas (2010) have written an article on "CSR by Islami Banks in healthcarestakeholders' perception". The objective of the study is to explore the stakeholders' insight of CSR support in the healthcare sector by Islami Bank Bangladesh Limited (IBBL). The finding of the study is that the stakeholders believe that this hospital (Islami Bank Central Hospital) is significantly contributing to the society through its funding in healthcare sector. The Amirul et al., (2017) have written an article on "Corporate Social Performance (CSP) Influences on Islamic Bank's Financial Performance". The objective of the research work is to assess the impact of corporate social performance (CSP) such as legal, ethical, economic and discretionary responsibilities on Islamic banks financial performance. The finding show that Islamic banks has meet up minimum level of each responsibilities and provide much attention on discretionary responsibilities which is the pinnacle stage of CSP. This study is focused on only one Islamic bank and may not reflect other Islamic banks' picture in the county or other areas. The number of observations is too small to conduct robust and rigorous analysis. El-Halaby & Hussainey (2015) assessed the corporate social responsibility conduct and disclosure (CSRCD) of 138 Islamic banks from 25 countries by employing content analysis (websites and annual reports). Their findings confirmed the early works conducted by Maali et al., (2006) which described the number of Islamic banks still low in term of disclosing their CSP and social activities. Nevertheless, there is progress from 2006 to 2015 when El-Halaby & Hussainey (2015) result shows higher percentages than Maali et al., (2006), and consistent with Haniffa and Hudaib (2007) on the gap between the findings and expectations towards the CSRCD. banks. In this regard, wood 1991 conceptualization and political economy theory are noteworthy. A brief discussion about these theories is made in the following section. # i. Wood 1991 Conceptualization A model proposed by Wood (1991) establishes a noteworthy improvement in CSR study. Wood (1991:693) defined Corporate Social Performance (CSP) as 'A business organization's configuration of principles of social responsibility, processes of social responsiveness, and policies, programs, and observable outcomes as they relate to the firm's societal relationships.' The CSP model (Wood, 1991) Principles of CSR Wood (1991: 694). In this model, CSP is hypothesized as including three aspects -Principles of CSR, Processes of CSR and Outcomes of corporate behavior. Principle of CSR 1 includes institutional principle, organizational principle and individual principle. In processing CSR programs, environmental assessment, stakeholder management and issues management are duly considered. In this model, outcomes are classified into three categoriesthe social impacts of corporate behavior, the programs companies conduct to implement social responsibility and the policies developed by the companies to handle social issues and stakeholder interests. # ii. Political Economy Theory It is assumed by many scholars that legitimacy theory and stakeholder theory have both arisen from a bigger theory termed as political economy theory. The 'political economy' itself has been defined by Gray et al. (1996, p. 47) as "the social, political and economic framework within which human life takes place". Deegan (2009) mentioned that by regarding political economy theory, an academic is capable to reflect wider (social) issues that influence on how corporation runs, and what information it takes to disclose. The political economy treats accounting statements as social, environmental, political or economic evidence and they are used as an instrument to contribute to the businesses' private interest (Guthrie and Parker, 1990). A distinction has been drawn by Buhr (1998) between the legitimacy theory and political economy theory. Buhr tells that both (legitimacy theory and political economy theory) serve to ensure legitimacy but means and motivation are seen in a different way. The difference happens in the explanation of reporting choice. For example, if 'the disclosure choice' focuses on the social constructionist perspective, it is consistent with the legitimacy theory, whereas if it focuses on the hegemonic perspective, it is more consistent with political economy theory (Buhr, 1998). # III. # Objectives of the Study Overall objectives of the study is to investigate the relationship between CSR disclosure and financial performance of Islamic Sharia-based commercial banks in Bangladesh The specific objectives of this study are as follows: i. To examine the association between the extent of CSR disclosure and return on assets (ROA) of Islamic Sharia-based commercial banks in Bangladesh. ii. To investigate the empirical relationship between CSR disclosure and return on equity (ROE) of Islamic Sharia-based commercial banks in Bangladesh. iii. To assess the impact of CSR disclosure on earnings per share (EPS) of Islamic Sharia-based commercial banks in Bangladesh. IV. # Methodology of the Study The methodology is the general research strategy that outlines the way in which research is to be undertaken. In simple terms, it is a planned and systematic way of doing research. In the following section, the methodological aspects of this study have been provided. # a) Sample Selection Simple random sampling technique is employed to select the sample banks. There are 6 Islamic Sharia-based commercial banks which are listed on Dhaka Stock Exchange (DSE) in Bangladesh. Among these 6 banks, 4 (66.67% to population) banks are selected randomly for this study (Appendix A). # b) Construction of CSR Index A comprehensive CSR disclosure index has been developed based on previous literature review. The CSR index comprise of 16 themes with 85 items of CSR reporting. Un-weighted index is employed providing each items equal weight to total score. If a banking company (bank) disclosed CSR items in its annual report it scored "1' while banks that did not disclose an item scored '0' (Gujarati, 2009). The disclosure model is additive and un-weighted index are calculated as follows ( Considering different scholars' opinions, this study adopts three accounting based measures to examine the relationship with CSR disclosure. These three ratios are Return on Asset (ROA), Return on Equity (ROE) and Earning per Share (EPS). These three indicators of financial performance are preferred because these are the most practical and significant measures to detect the association between CSR disclosure and financial performance in previous studies (Mcgruie et al, 1988). These three accounting tools are explained in the following sections. # i. Return on Asset (ROA) Basically, Return on Asset (ROA) is a fundamental profitability indicator that is employed to assess the efficiency of a business's assets. It implies # ii. Return on Equity (ROE) Return on Equity (ROE) is the amount of net income returned as percentage of shareholders' equity. Return on equity measures a company's profitability by revealing how much profit a company generates from the money shareholders have invested. ROE is especially used for comparing the performance of companies in the same industry. ROE is a measure of management's ability to generate income from the equity available to it. ROEs of 15-20% are generally considered good. # iii. Earning per share (EPS) One of the important indicators of profitability of a company is earnings per share. It is the share of a company's profit allocated to each outstanding share of common stock. how efficiently and effectively a business is receiving earning advantage based on its assets. This ratio is widely and popularly used to compare banks' financial performance from year to year and bank to bank. Many # d) Measurement of Control Variables This study considers some variables as control variables which are related to the linkage between CSR disclosure and financial performance of sample banks. Therefore, this study has used (i) firm size (ii) firm age (iii) percentage of independent directors in board as control variables for measuring the link between CSR disclosure and financial performance. # i. Firm Size # ii. Firm Age Age of a business organization is also another vital factor that may influence the volume of corporate social responsibility disclosure. The relation between CSR and financial performance could be affected by firm age (Schreck, 2011, Peloza, 2006, Suttipan, 2012). Some studies advocate that younger businesses do not have much concentration on CSR activities; rather, emphasize on maintaining profitability. After achieving healthy financial performance, businesses may invest in protecting their reputation by increasing CSR programs. Roberts (1992) has found positive relationship between firm age and CSR disclosure. Again, Roberts argue that "Sponsorship withdrawal could signal to stakeholders that the corporation expects financial or managerial In the light of above information, it can be said that firm age may influence the association between CSR disclosure and financial performance and age of business organization is included in this study as a control variable. Age is measured by number of years since the firm was listed on the Dhaka Stock Exchange (DSE) in Bangladesh. # iii. Independent director Corporate governance is must for successful operation of a business. An independent director is a part of corporate governance. Independent directors are non-executives of a company who assist a company in improving corporate credibility and governance standards. Our companies Act 1994 does not provide any provision on Independent directors. However, Indian Companies Act 2013 provides the definition of Independent director as under: An Independent director in relation to a company, means a director other than a managing director, a full time director or a nominee directora) Who, in the opinion of board a person of integrity and possesses relevant expertise and experience; b) Who or was not a promoter of the company or its holding, subsidiary or associate company c) Who has or had no precautionary relationship with the company, its holding, subsidiary or associate company or their promoters, or directors during the two immediately preceding financial years or during the current financial years. d) None of whose relatives has or had pecuniary relationship or transaction with the company, its holding, subsidiary or associate company, or their promoters, or directors, amounting to two per cent or more of its gross turnover or total income or fifty lakh rupees or such higher amount as may be prescribed, whichever is lower, during the two immediately preceding financial years or during the current financial years. e) Who, neither himself nor any of his relativesi. Hold or has held the position of a key managerial personnel or is or has been employee of the company or its holding, subsidiary or associate company in any of the three financial year immediately preceding the financial year in which he is proposed to be appointed: ii. Is or has been an employee or proprietor or a partner, in any of the three financial years immediately preceding the financial year in which he is proposed to be appointed, of-It may be noted that use of weighted average number of shares outstanding is more reliable and accurate as number of shares outstanding may alter over time. Several studies have employed an EPS to reveal the relationship with CSR disclosure and other variables (Pava & Krausz, 1996;Moore, 2001;Dragomir, 2010). Kwanbo (2011) recommends that the link between societal disclosure and earnings per share of public companies in Nigeria is not significant. So, CSR reporting is not so crucial issue in increasing profit of a firm. On the other hand, it is evident that the association between CSR and EPS (ROA and ROE) is positive and significant (Ehsan & Kaleem, 2012). So, the issue relating to direction of the relationship between CSR disclosure and earnings per share (EPS) is still unresolved. Firm size is a crucial factor that might influence the functions and performance of a firm. In earlier studies, it is seen that, firm size was significantly correlated with CSR disclosure and financial performance (Patten, 1991;Gray, 1995;Deegan & Rankin, 1996, Waddock & Graves (1997) indicate that bigger firms may have larger resources than those of smaller firms to conduct more societal and eco-friendly activities. Firm size can be defined in a number of ways and there is no overriding cause to favor one over another(s) (Cooke, 1991). A number of measures of size have been used by many investigators such as turnover, total assets, fixed assets, paid up capital, shareholders' equity, capital employed, number of employees, number of shareholders, number of branches etc. The present study includes total assets as a proxy for size. disturbances". However, Ehsan & Kaleem (2012) argue that age of firm in Pakistan was not found considerably related to CSR disclosure. Mennessa (2012) and Sukcharoensin (2012) find that there is no relationship between CSR disclosure and firm age. A. A firm of auditors or company secretaries in practice or cost auditors of the company or its holding, subsidiary or associate company ; or B. Any legal or consulting firm that has or had any transaction with the company, or its holding, subsidiary or associate company amounting to ten percent or more of the gross turnover of such firm; iii. Holds together with his relatives two percent or more of the total voting power of the company iv. Is a chief Executive or director, by whatever name called, of any non-profit organization that receive twenty five percent or more of its receipts from the company, any of its promoters, directors or its holding, subsidiary or associate company or that holds two percent or more of the total voting power of the company. Following 2012) claim that increased number of independent directors of a board try to boost managers to disclose more voluntary information. It is also seen that a higher number of independent directors were positively and significantly related with CSR (Cheng, et al, 2006). In this research, the independent directors of banking companies are measured as the percentage of independent directors on the board of banking companies in relation to total number of directors in the board. IDIRECTOR jt = A variable of the independent director ratio of firm j in period t. It is defined as number of independent director divided by total number of directors in the board of firm ? jt = A random error of firm j in period t. # Source: Author's Preparation e) Regression Models for Analysis This study tries to compare two types of regression models to analyze the relationship between CSR disclosure and financial performance of Islamic Sharia-based banking companies in Bangladesh. These two types of regression models are ordinary least square models and panel data models. They are described as follows: # i. A Pooled Ordinary Least Squares Model (POLS) or Constant Coefficient Model The pooled OLS considers constant coefficients with the usual assumption for cross-section data (Cameron and Trivedi, 2008) and controls for year. This method assumes that regressors are not correlated with the error or the intercept and the slope are equal for all data. # ii. Panel Data Model Panel data is also known as longitudinal data. In panel data model, the behavior/ information of individual or firms or units are measured across time (Baum, 2006). Gujrati (2009) emphasizes on panel data models as it can identify and measure or analyze such type of information or data that cannon be detect in crosssection analysis. Depending upon the nature of panel data, random effect models (RE) or fixed effect models (FE) are used in regression analysis. The random effect model (RE) assumes that the unobservable individual effects are random variables variables in the model. Furthermore, the fixed effect model (FE) explores the link between the predictor and outcome variables within a unit. This model assumes that unobservable individual effects are potentially correlated or impacted with the observed (outcome) regressors. It removes the influence of time-invariant characteristics of data (Torres-Reyna, 2013). To better comprehend the association between CSR disclosure and financial performance in the Bangladeshi context, this study has used the above described regression models to examine the link between CSR disclosure and financial performance, as well as other firm characteristics, such as size of entity, age of company and percentage of independent directors. The dependent variables in this study are ROA, ROE and EPS as the measures of financial performance. The independent variables are CSR disclosure index and the control variables. # f) Specimen Test The following tests have been adopted to specify the appropriate model among pooled OLS, random effects model and fixed effects model. # i. The Breusch and Pagan Lagrange Multiplier Test: The Bruesch-pagan LM test is employed to identify appropriate model between pooled OLS and random effects model. The null hypothesis is that the variance across entities is zero or no random effects exist in the model (Ho: ? 2 jt =0) (Gujrati, 2009). It implies that there exists no panel effect or difference across entities. A Chi square distribution value with 1 degree of freedom will be determined, if the null hypothesis is rejected, then random effects is appropriate over the pooled OLS model. If null hypothesis is accepted, the pooled OLS model is more suitable than other models. ii. Hausman Test: Random Effects or Fixed Effects Model The Hausman test is employed to examine whether random effects or fixed effects are appropriate. The null hypothesis is that the individual effects are not correlated with other regressors in the model. If the test shows an insignificant P value, the null hypothesis is accepted. The random effects produce unbiased estimators, indicating that random effects are appropriate. If the test indicated a significant P-value, the null hypothesis is rejected and it indicates that fixed effects model is more appropriate than random effects model. # g) Hypotheses Development A hypothesis is a definite, testable estimate about what we expect to happen in our study. To being consistent with the aims of this study, the following hypotheses are framed. Here, hypothesis defines CSR disclosure as independent variable. The alternative hypothesis consists of the following sub-hypotheses: H a1 There is a positive and significant relationship between CSR disclosure and Return on asset (ROA). # H a2 There is a positive and significant relationship between CSR disclosure and Return on equity (ROE). # H a3 There is a positive and significant relationship between CSR disclosure and Earning per share (EPS). # h) Model and Hypothesis Testing To test the relationship between CSR disclosure and financial performance in Islamic Sharia-based banking companies listed under DSE in Bangladesh, this study has used different regression models (Fixed effects or Random effects models). It has been mentioned that this study defines regression models based on financial performance measures as dependent variables. As a result, this study has estimated the following models using financial performance indicators' as dependent variable and CSR disclosure as independent variable. # CSR disclosure and Return on Asset (ROA) testing # k) Tools of Analysis To achieve the objectives of the present study and to ensure reliability of data, different statistical tools such as average, standard deviation, Minimum, Maximum, skewness, kurtosis, correlation, regression, correlation matrix, variance inflation factor (VIF) have been used for analyzing the data and testing the hypotheses. For making the analysis easy and fast, SPPS and Strata software were used. V. # Analysis and Discussion # a) Descriptive Statistics of the study Table 2 presents the descriptive statistics of the variables as part of the model examined, covering both dependent variables and independent variables across the 20 observations collected. Table 2 shows that the average CSR disclosure of the entire sample banks (Islamic Sharia-based commercial banks in Bangladesh) over the periods covered is 26.383% with a standard deviation of 13.20%. The mean CSR disclosure indicates that the level of CSR disclosure is relatively low despite the Islamic ethics at the center of Islamic banking and the standard deviation reflects a small level of dispersion from the mean. The highest CSR disclosure value is 48.24% and the lowest value is 7.06%. Thus, the CSR disclosure of Islamic Sharia-based commercial banks in Bangladesh over the period 2011-2015 is quite unstable. # Source: Author's Preparation based on sample banks annual reports Table 2 shows that the highest ROA for all sample banks over the study period is 1.75% and the lowest value is -9.97%. The mean ROA value is -0.5636 % with a standard deviation of 3.123895%. The average ROE of sample banks is 9.152% with a standard deviation of 5.135919%. The minimum ROE of sample banks is 00.63% and the highest ROE is 17.42%. The mean EPS of sample banks for the period covered (2011-2015) is 1.449 with a standard deviation of 1.91. The highest EPS is 4.84 and the lowest value of EPS of sample banks is -2.7. Considering the control variables of this study, we found that the listing age of sample banks ranges from 4 years to 30 years with a mean age of 10.88 years. The percentage of independent director on board ranged between 42 percent and 6.67 percent, having a mean value of 10.01 percent. It indicates that Islamic Sharia-based commercial banks in Bangladesh do not comply with the Bangladesh bank guideline regarding appointment of independent director on board as the average percentage of independent director is below 20%. Table 2 represents that the highest value of total assets is 8595201.6 and the lowest value is 12752.9. The mean value of total assets of the entire sample banks of the period covered is 246713 with a standard deviation of 270686. To test the normality of data, skewness and kurtosis statistics are found out. Normality in data is often a conventional assumption in the estimation process (Bai & Ng, 2005). Data distribution with either a high skewed nature or with high kurtosis is indicative of non-normality which has random effects on specification or estimation (Hall & Wang, 2005). "An outlier is a case with such an extreme value on one variable (a univariate outlier) or such a strange combination of scores on two or more variables (a multivariate outlier) that they distort statistics (Tabachnick & Fidell, 2001). At the beginning stage, descriptive statistics analysis using the mean score of components of dependent and independent variables was conducted and it was found that Kurtosis scores are(<3) for all the variables of the study (shown in Table 3). For skewness, indices have been used for acceptable limits of ±2 (Trochim & Donnelly, 2006;Field, 2000;Gravetter & Wallnau, 2014). It is found that all skewness statistics are below ±2 (Table 3). So, it can be concluded that the data of the study are normally distributed. # c) Testing of Multicollinearity Based on normally distributed data, VIF test were conducted to detect the existence of multicollinearity among the independent variables of the study. As can be seen in In order to explain the relationship between CSR disclosure and return on assets of Islamic Shari-based banking companies in Bangladesh, panel data regression models are run and the obtain results are discussed in the following sections. The obtained results are shown in Table 6 & 7. # a. The Bruech-Pagan Lagrange Multiplier Test Here it is assumed that the pooled OLS model is more appropriate than random effects model (Null hypothesis). The Breusch-Pagan Lagrange Multiplier test (Table 6) with the Chi squire value of 0.00 is significant at 100% level of significance indicates that the null hypothesis is accepted and alternative is rejected. # b. The Hausman Test In this part, the Hausman test is employed to determine whether random effect model is appropriate or fixed effect model. From the result reported in Table it is observed that the null hypothesis is rejected as the value of Chi-squire (31.22) is statistically significant 1% level of significance which implies that the individual effects are correlated with other independent variables taken in the regression model. So, fixed effects model is more appropriate than random effects model. 8) with the Chi squire value (0.00) is not statistically significant at level of significance indicates that the null hypothesis is accepted and alternative is rejected. # b. Results of the Hausman Test In this part, the Hausman test is employed to determine whether random effect model is appropriate or fixed effect model. From the result reported in Table 8 it is observed that the null hypothesis is rejected as the value of Chi-squire (5. 30) is statistically significant at 25.81 % level of significance which implies that the individual effects are not correlated with other independent variables taken in the regression model. Finally, OLS regression model is more appropriate over both fixed effect and random effect model. Table 7 represents the regression results using fixed effects model where return on assets (ROA) is used as dependent variable and CSR disclosure, listing age, percentage of independent director on board and log of total assets are used as independent variables. It is found that the model is significant at 1% level (Prob> Chi 2 =0.000) and explains 50.62% of the variability of dependent variable (ROA) in the Islamic Shari-based banking companies in Bangladesh. As can be seen from Table 7, the CSR disclosure index is statistically significant at P=0.819. On the basis of results, it is evident that there is no significant relationship between CSR disclosure and return on assets of Islamic Sharibased banking companies in Bangladesh. As the results show, two out three control variables are statistically significant: age at p=0.001 and size (total assets) at p=0.018. The coefficient of Independent director is negative but statistically insignificant. It implies that there is no significant association between return on assets and Independent director. The results relating to the relationship between CSR disclosure and return on assets (ROA) is inconsistent with the previous studies, such as those of Scholtens (2009) and Simpson & Kohers (2012). CSR information in annual reports of Islamic Shariabased commercial banks in Bangladesh is a crucial factor in increasing earnings per share. The coefficient of age is positively associated with earnings per share in the fixed effects model at 5% level (P=0.038). The independent directors on board are negatively and significantly linked with earnings per share (EPS) at 5% level of significance in the regression model. But the size proxied by log of total assets of Islamic Shariabased banking companies in Bangladesh is not significantly related to earnings per share in this fixed effect model. The overall R-squared value is 0.14 indicates that the independent variables of the model can explain 14% variability of dependent variable 'earning per share'. # VI. # Policy Implications The above mentioned results might have significant theoretical and practical implications. This study contributed to the existing literature in a number of ways. Firstly, the positive influence of CSR disclosure on financial performance (FP) has supported the political economy theory. It implies that Islamic banking companies should involve in social responsibility as much as possible (Freeman & Evan, 1990). Secondly, this study fills the gap in existing literature on CSR disclosure adding Bangladesh perspective on the phenomena, thereby increasing our current realization about the association between CSR reporting and FP in Islamic banking industry. Thirdly, this research has conducted in Bangladesh that is thought at now as a very emerging economy in the globe especially in the south-east sub-continent. So, the findings could have serious implications on other emerging countries in the globe. The findings of the research have exposed also some practical implications and suggestions. Firstly, the results imply that managers should increase the level of expenditure and efforts on CSR programs and they might desire a positive result on financial performance in the long-run. Secondly, Stakeholders of the company can be better aware about the policy and actions taken by the management, if they were engaged in social activities. Thus, manager should ensure accountability and fairness in managerial process and report full and relevant financial as well as non-financial information to all stakeholders. # VII. # Conclusion In recent years, both CSR and Islamic banking are seen in an increasing trend. Due to the significant demand of Islamic banking in this modern time, some commercial banks operate windows or sections that give Islamic banking services to clients. In this study researcher has investigated 20 annual reports of 4 Islamic Sharia-based commercial banks in Bangladesh over the period from 2011 to 2015. This research work has assessed the extent CSR reporting but mainly focused on the impact of CSR disclosure on financial performance indicators. The empirical results of Table 7 show that there is no significant association between CSR disclosure and return on assets (ROA) of Islamic Sharia-based commercial banks in Bangladesh. Regression results of Table 9 represents that there is a statistically significant association between CSR disclosure and financial performance proxied by return on equity (ROA) of Islamic Sharia-based commercial banks in Bangladesh. Hence, it can be said that higher the extent of CSR disclosure, the better the Islamicsharia based banks financial performance in Bangladesh. The results also indicate that there is a statistically positive and significant linkage between CSR disclosure and earnings per share (EPS) of Islamic Sharia-based commercial banks in Bangladesh. So, it is an inspirational finding to them who made investment decision on the basis of earnings per share (EPS) in Islamic banking industry as higher EPS leads to the declaration of higher percentage of dividend to the investors. The findings of the study also suggest that it is advantageous to integrate CSR dimensions in Islamic Sharia-based commercial banks as business strategy. Formulation and implementation of a standard CSR policy could increase the reputation or image and assist to cope with external negative news or imagedamaging incidents. Thus, it protects banks from incurring losses in the future. 1Corporate Social Disclosure Index VariableCSRI EPS jt =A variable of earning per share of firm j in the period of t. It is defined as the earnings after tax and preferred dividend scaled by total outstanding share.Control VariablesSIZE-TAjt = A variable of corporate social disclosure index of j firm in the period of t. It is defined as number of CSR items which firm disclosed divided by total CSR disclosure items (85 items).Financial Performance Measures Variables ROA jt =A variable of return on assets of firm j during period t. It is defined as Profit after tax (PAT) divided by total assets ROE jt = A variable of return on equity of firm j during period t. It is defined as Profit after tax (PAT) divided by shareholders' equity. jt = A variable of firm size of firm j in the period of t. It is defined as the Total Assets.AGE jt = A variable of firm age of firm j in the period of t. It is defined as number of years since firms was listed on DSE. 2VariablesObservationMeanStd. Dev.MinMaxCSR disclosure2026.38313.199657.0648.24ROA20-0.56353.123895-9.971.75ROE209.1525.1359190.6317.42EPS201.4491.905084-2.74.84Age201710.88217430IDDirector2018.739510.014416.6742Total Assets20246713270686.712752.98595201.6 3SkewnessKurtosisDescriptionStatisticStatisticCSRD.235-1.004ROA-1.1462.039ROE-.218-.985EPS-.368.232OP.146-1.515Age.000-2.133IDirector1.168.570SIZE (TA)1.247.208Source: Author's Preparation based on sample banks annual reports 4 4Independent variablesCollinearity StatisticsToleranceVIFCSRD.1715.864Age.5181.929IDirector.3432.928Log-asset.3942.541Source: Author's preparation based on annual reportsd) Extent of CSR disclosure of Islamic Sharia-based banking companies (2011-2015) 5BankMean CSR disclosureStandard deviationIBBL44.24%4.68%FSIBL19.53%8.98%SJIBL29.18%4.95%ICBIBL10.59%3.33%Source: Author's preparation from annual reports of sample banks 5e) Examination of Empirical Relationship between CSRDisclosure and Financial Performancei. The relationship between CSR disclosure andReturn on assets (ROA) (ROA as dependentvariable) 6Test nameChi-SquareProb.CommentHausman Test31.220.000Alternative acceptedBreusch-Pagan Lagrange Multiplier0.001.000Alternative rejectedSource: Author's preparation based on panel data of the study 7Explanatory variablesCoefficientsStd. errort-valuep-valueCSRD0.017450.0745280.230.819AGE1.1405040.25674974.440.001IDIRECTOR-.06454780.0611253-1.060.312SIZE (TA) (log)-5.9098342.166947-2.730.018Constant49.8269223.040662.160.051Group:4; Total observations; 20; R-Squared: 0.5062Source: Author's preparation based on panel data of the studyii. The relationship between CSR disclosure andReturn on equity (ROE) (ROE as dependentvariable)a. The examination of random effect using theLagrange multiplier testHere it is assumed that the pooled OLS modelis more appropriate than random effects model (Nullhypothesis). The Breusch-Pagan Lagrange MultiplierTest (Table 13Source: Author's preparation based on regression results HypothesisRelationship between two variablesExpected relationshipResultsOutcomeH a 1 H a 2 H a 3CSR on ROA CSR on ROE CSR on EPSPositive Positive PositivePositive Positive NegativeP =0.819 P =0.085 P=0.042Rejected Accepted AcceptedYear 201925Volume XIX Issue II Version I( )Global Journal of Management and Business ResearchG * Corporate Social Performance (CSP) Influences on Islamic Bank's Financial Performance Amirul Journal of International Business, Economics and Entrepreneurship 2 1 2017 * Management Perceptions from IFIs in Bahrain Journal of Business Ethics 129 4 * Narrative Disclosure of Corporate Social Responsibility on Islamic Financial Institutions ZAAribi SSGao Managerial Auditing Journal 27 2 2012 * The Banking Sector and CSR: An Unholy Alliance CChamber RDay Financial Regulation International 12 9 2009 * CDeegan Financial Accounting Theory Sydney McGraw Hill Australia Pty Ltd 2009 3rd eds. * The Determinants of Social Accountability Disclosure: Evidence from Islamic Banks around the World SEl-Halaby KHussainey International Journal of Business 20 202 2015 * An analysis of corporate social responsibility (CSR) and sustainability reporting assessment in the Greek banking sector KIEvangelinos ASkouloudis IENikolaou WLFilho Professionals' perspectives of corporate social responsibility SO LIdowu & W Filho London; New York Springer 2009 * Accounting and accountability: Social and environmental accounting in a changing world RGray DOwen CAdams 1996 Prentice Hall Hemel Hempstead * Basic Econometrics. McGraw Hill DGujarati 2003 1002 * Corporate Social Disclosure Practice: A Comparative International Analysis JGuthrie LParker Advances in Public Interest Accounting 3 1990 * Corporate Governance Structure and Performance of Malaysian listed Companies RHaniffa MHudaib Journal of Business Finance & Accounting 33 7 & 8 2006 * Corporate Social Responsibility in the Banking Sector: A Comparative Study on Some Commercial Banks in Bangladesh Hasan DIU Journal of Business and Economics 7 1 2013 * The Current Status of Corporate Social Responsibility Disclosure of Islamic Shari-based Commercial Banks in Bangladesh &Hossain Neogy American Journal of Trade and Policy 19 1 2019 * Corporate Social Responsibility and Financial Performance: An Empirical Analysis on Greek Companies TKaragiorgos European Research Studies 13 4 2010 * An Effectiveness of Social Disclosure on Earnings per Share in Nigerian Public Comoanies MLKwanbo World Journal of Social Science 1 1 2011 * Social reporting by Islamic banks BMaali PCasson CNapier Abacus 42 2006 * Corporate Social Responsibility and Firm Financial Performance Mcguire The Academy of Management Journal 4 1988 * Corporate Social Responsibility: An Exploratory Study of the Quality and Extent of Social disclosures by Lebanese Commercial banks FMenassa Journal of applied accounting research 11 1 2010 * Value Divergence from the Business Case for Corporate Social Responsibility: Contemporary Insights from British Industrial Pioneers', paper presented in a seminar in the University of Stirling LParker 2011. September 19. Autumn 2011 * Exposure, Legitimacy and Social Disclosure DMPatten Journal of Accounting and Public Policy 10 4 1991 * The Association between Corporate Social Responsibility and Financial Performance: The Paradox of Social Cost Pava JKrauasz Journal of Business Ethics 15 3 1996 * Using Corporate Social Responsibility as Insurance for financial Performance JPeloza California Management Review 2006 * The Impact of Corporate Social Responsibility Disclosure on Financial Performance: Evidence from the GCC Islamic Banking Sector Platnova J Bus Ethics 151 2018. 2018 * Reviewing the business Case for Corporate Social Responsibility: New Evidence and Analysis PSchreck Journal of Business Ethics 103 2 2011 * The Influence of Financial Performance and Non Financial Performance on Islamic Social Responsibility Disclosure: Evidence from Islamic Banks in Indonesia ISiswanti The International Journal of Accounting and Business Society 26 2 2018 * M-GSoana 2011 * An Empirical Investigation of the Relationship between Corporate Financial Responsibility and Financial Performance: Evidence from the Manufacturing Sector of Pakistan S &Ehsan Kaleem Da Journal of Basic and Applied Scientific Research 2 3 2012 * Corporate Social Responsibility and Financial Performance: Correlation or Misspecifications A &Mcwilliams DSiegel Strategic Management Jouranl 21 3 2000 * Corporate Social Responsibility and Financial Performance in Islamic Banks Millin The Banker 2013 * CSR by Islami Banks in healthcare-stakeholders' perception Sadia &Shafiqur Nicholas 2010 * Bangladesh Journal of Medical Science 09 4 * from the Financial Sector Corporate ownership and Control 8 * The Relationship between Corporate Social Performance and Corporate Financial Performance in the Banking Sector M-GSoana Journal of Business Ethics 104 1 2011b * The Determinants of Voluntary CSR Disclosure of Thai listed firms SSukcharoensin International Proceedings of Economic Development and Research 2102 46 * The Influence of External Pressure Groups on Corporate Social Disclosure Some Empirical Evidence CATilt Accounting, Auditing and Accountability Journal 7 4 1994 * The Corporate Social Performance-financial Performance link Sa &Waddock Graves Sb Strategic Management Journal 18 4 1997 * The relation between board characteristics and environmental disclosure: Evidence from Taiwan Wang Middle Eastern Finance and Economics 2012 * Corporate Social Performance Revisited DWood The Academy of Management Review 16 4 1991