Do Corporate Social Responsibility Disclosures Improve Financial Performance? A Perspective of the Islamic Banking Industry in Pakistan

Abstract

This study aims to investigate the impact of corporate social responsibility disclosures (CSRD) on the financial performance of the Islamic banking industry of Pakistan. The study employed the method of content analysis for collecting the required data from annual reports of all four full-fledged Islamic banks operating in Pakistan from 2012 to 2017. The study developed a novel comprehensive CSRD index by using the “Global Reporting Initiative” (GRI) and “Accounting and Auditing Organization of Islamic Financial Institutions” (AAOIFI). This index consists of five dimensions and 105 sub-dimensions of CSRD. The use of Ordinary Least Squares (OLS), Panel Corrected Standard Errors (PCSEs), and Generalized Least Squares (GLS) using random-effect (RE) and fixed-effect (FE) estimators revealed a significant negative relationship between CSRD and the financial performance of the sample firms. Regarding separate dimensions, the relationship of the Environmental and Economic dimensions of CSRD is significantly positive with current performance, but it is insignificant for the relationships of Legal, Philanthropic, and Ethical dimensions of CSRD with the current financial performance. In addition to contributing to the scarce literature in the Islamic banking industry of a developing country like Pakistan, the study will also help the policymakers and other stakeholders, including the AAOIFI, to develop a comprehensive CSRD policy or index and further improve the already established standards for CSRD.

1. Introduction

Adam Smith noted in his book “The Theory of Moral Sentiments” that capitalism would not sustain if not based on integrity and honesty. He suggested that the self-interests of the owners and the managers could be countered by first establishing and then following the moral and ethical values [1]. Most of the recent studies in corporate governance have focused on certain specific issues like corporate fraud, the misuse of managerial powers, and social irresponsibility in business. In response to the changing and challenging market and managerial behavior, the business firms are compelled to take externalities into account. Business ethics and corporate social responsibilities (CSR) are now crucial for being competitive; firstly, these help firms to maintain excellent and trained staff, and secondly, these assist firms in addressing and meeting the customer’s expectations [2]. CSR is a corporate move for the betterment of the whole society apart from the compulsions of law and the primary objective that firms are supposed to perform for the wellbeing of shareholders only [3]. Unlike the Western approach, the Islamic perception of CSR is different; it is considered a holy concept. Islam explains CSR in a more religious context that has its roots in both the Quran and Sunnah, which provides a better substitute or framework for the philosophy of the human’s interaction with its fellow humans and nature as a whole. The ethical and the moral principles originating from the revelations are greatly eternal, enduring, and absolute; one can perform better by exercising the social and the business responsibilities simultaneously [4].

The Holy Quran has emphasized the CSR activities and declares these for those who are the believers, such as: “Ye cannot attain righteousness unless ye donate to charity from possessions ye love. What ye donate to charity, ALLAH is aware of’’ Quran, surah 3 Ayahs 92 [3:92]. The financial contributions under CSR are emphasized by the Holy Quran and the Sunnah in the shape of Sadqat and Zakat. Zakat means purification. It is a yearly contribution that is mandatory for the rich and paid to the poor, especially to the widows, orphans, and the elderly people who are no longer able to work or earn for themselves. Islam as a religion offers a complete code of life, including instructions about faith, ethics, morality, prayers, and belief in Allah in every walk of life, such as political, business, social, economic, and religious affairs [5]. The primary purpose of Islam is holistic welfare. The ethical and moral teachings of Islam produce complete and everlasting principles that set guidelines for individuals, society, and economic and business interactions [6].

However, in modern corporations, corporate social responsibility disclosure (CSRD) refers to the improvement of financial performance by bringing significant and positive outcomes through customers’ loyalty and commitment to paying high prices and the reduction of the risk directed towards their reputation, especially in times of unfavorable economic conditions [7]. The journey towards CSR has started for the last few centuries and is still moving forward, with no signs of slowing down its pace [8,9]. CSR has much importance in the success of an organization, as customers prefer for the organizations to be responsible economically as well as socially. In Pakistan, the majority of the CSR activities are performed in the light of corporate philanthropy. Corporate financial contributions in the form of donations are considered an authentic tool to improve a corporation’s image in an extremely competitive environment [10]. For the last few years, many financial institutions have started introducing CSR practices in their operational and organizational strategies. The banking industry is relatively different, especially regarding corporate governance and other reforms and regulations, including CSR. The banking industry has a central role in society due to its higher level of social interaction; therefore, banks are expected to be socially responsible [11]. Substantial research has examined the relationship between CSRD and the financial performance of conventional financial institutions, which produced inconsistent results. The majority of these studies used the global reporting initiative (GRI) as a benchmark for developing the CSRD index. Some studies on whether CSRD has any contribution towards the financial performance of Islamic banks have also been conducted throughout the world, bearing positive, negative, and neutral results. The majority of these studies relied on the standards established by the Accounting and Auditing Organization of Islamic Financial Institutions (AAOIFI) for developing the CSRD index. Therefore, the motivation of this study is to bridge the gap between these two standards by merging them and developing a more comprehensive index. The AAOIFI standards will help cover the issues related to Shariah compliance, such as Riba, Zakat and Sadqat, Qard-E-Hassana, Shariah Supervisory Board (SSB) approvals, attestation, and other ethical issues. The GRI standards will cover the issues related to conventional aspects, such as community, employees, environment, human rights, and other legal issues.

As far as the Islamic banking industry in Pakistan is concerned, minimal work has been done on the subject matter of CSRD. The Islamic banking industry has a good market share in the overall existing banking industry of Pakistan. Due to the increasing importance of CSR and its disclosures in the banking industry in general and in the Islamic banking industry in particular, it is vital to find out whether or not the increasing market share and financial performance have any relevance to CSR practices and their disclosures in Pakistan. Hence, this study aims to answer the research questions: What is the level of exposure of CSR of the Islamic banking industry in Pakistan? What is the impact of CSR disclosure on the financial performance of the Islamic banking industry in Pakistan? What are the impacts of dimensions of CSR disclosure on the financial performance of the Islamic banking industry in Pakistan?

After answering the above research questions, the study offers several theoretical, methodological, and practical contributions. Firstly, the study contributes to the literature on Islamic banking in Pakistan, as there were limited studies in the past [9,12]. Secondly, the study has a theoretical contribution in the application of the crux of stakeholder theory in the literature of Islamic banking, which may further assist the industry in understanding the demand of multiple stakeholders and legitimizing their license to operate on the larger canvas of the society [13,14,15]. Thirdly, the study extends the existing CSRD index, which is based on the Global Reporting Initiative (GRI) framework and “Accounting and Auditing Organization of Islamic Financial Institutions” (AAOIFI); it consists of five dimensions and 105 sub-dimensions of CSRD. The study claims that the novel index is more comprehensive than those of previous studies [9,16] and has a methodological contribution in measuring CSRD in developing countries like Pakistan. Likewise, the study applied a panel data longitudinal analysis approach from 2012 to 2017, as the focus of the previous studies was limited in the period and dimensions of CSRD. The study period is important, as many environmental, social, and governance reforms (ESG), along with the concept of green banking, its regulations, and the codes of corporate governance of 2012 and 2017, were introduced during this time in Pakistan. Finally, this study provides practical implications for management and policymakers of the Islamic banking industry to understand the importance of CSRD reporting and its impact on overall firm performance. Furthermore, it would add to the systematic presentation of CSRD in developing an international attitude towards CSRD, especially in developing countries.

The rest of the paper consists of a literature review where the theoretical framework and hypothesis development followed by the conceptual framework are discussed. The data sample, measurement of financial performance and CSRD with the research model, and the estimations are discussed in the methodology section. Finally, the results and discussions with conclusions, limitations of the study, and future directions are reported.

2. Literature Review

2.1. CSRD and Financial Performance of Islamic Banks

Prior studies noted a positive and considerable association between firms’ profitability and the magnitude of their CSRD [9,17,18]. Most profitable organizations disclose their CSR information to show their social role in the wellbeing of the community and ensure their survival in the industry. A study found a significant positive relationship between the practices of CSR disclosed in the annual reports and performance of the selected Islamic banks regarding the goodwill in Malaysia [19]. In another study, the possible link between CSRD and financial performance of the Islamic banks selected from different countries from 2010 to 2011 was investigated. The analyses indicated a significant positive relationship between the CSRD and financial performance [20]. Another study investigated the information that is crucial for ethical issues in Islamic business [21]. They designed a moral benchmark that is considered vital to differentiate Islamic banks from conventional banks. Using content analyses, they collected data from annual reports and noted a significant gap between the expected and actual disclosures. They concluded that the expected gap may have originated from the indifferent behavior of the stakeholders or the Arab culture where the sample banks were located. In the end, they summarized that if the Islamic banks intend to remain in competition with conventional banks, their communication should be even more useful to improve their overall reputation in the society in which they operate. Another dimension of the study examined the degree and the nature of CSRD and its impact on the firms’ performance in a study of 90 Islamic banks from 13 countries for a period from 2010 to 2011 [20]. The study constructed a comprehensive CSRD index for the Islamic banks where the AAOIFI standard No. 7 was used to collect data. The study reported a direct link between CSRD, the financial performance of Islamic banks, and the level of CSRD for different countries. Among them, Indonesia stood first with the highest score of 54%, while Pakistan was the last with the lowest score, i.e., 30.4%. The researchers investigated CSRD in the Islamic banks of Malaysia and found that the volume and quantity of CSRD have improved with time [22]. Another study [23] noted a regular sequence in the CSRD in three phases in the Islamic banking industry of Bangladesh, i.e., the pre-1990s, 1990–2001, and post-2001. Similarly, they divided the moral and the ethical reporting of CSR into two categories, i.e., the particular style of reporting that is related to the Islamic banks, especially those practicing Shariah compliance, and the universal standards of reporting practices that mostly account for shareholders’ communities, customers, and the employees. They noticed an overall positive increase in both the particular and the universal standards of disclosures over the corresponding period, but an upward shift in the universal disclosures after 2006. However, in contrast, some studies found a negative association between CSRD dimensions and financial performance. For example, a study found a negative impact of the environmental and social dimensions of CSRD on the financial performance of Malaysian Islamic banks [9].

2.2. Theoretical Framework and Hypothesis Development

The stakeholder’s theory states that the main objective of organizations is the creation and maximization of stakeholders’ value. Whenever the firms meet the expectations of more than a single stakeholder, they come into a position to improve their performance [13,24]. Given this, the theory that explains the disclosure of CSR and its impact on the firm’s performance is the stakeholder theory. As per this theory, if an organization wants to be more efficient, it has to pay greater attention to all of the relationships, especially to those that affect or are affected by the organization’s objectives. The stakeholder management assumption of the theory is a useful tool to gain the results which are already predicted, especially the profitability. This aspect of the said theory signals a link between the stakeholders and the performance of the corporation [13]. To summarize the stakeholders’ theory, the stakeholders are assets and the managers are expected to manage them effectively. Similarly, the social responsibilities of the business organizations are to accomplish the needs of their stakeholders for which they are legally and socially responsible [25]. From a CSR perspective, society expects firms to run their operations responsibly regardless of their economic and commercial interests. The main crux of this theory is that the organization’s goals must be in line with the goals of society. The theory further supports the goals of the Islamic banks to satisfy the demand of multiple stakeholders in the normative approach and avail instrumental benefits in the shape of better financial performance [13,26]. All of these arguments and assumptions are also compatible with the teachings of Islam regarding CSR. In addition to compliance with the governmental policies and reforms, Islam also necessitates the welfare of the individuals, society, environment, and all other creatures in the universe. Most of the prior empirical studies found a positive link between the individual dimension of CSRD and financial performance [9,27]. These positive impacts are mainly attributed to the competitive advantage, which improves customers’ relations and retention and thus recognizes better financial performance. Competitive advantage is closely related to firms’ financial performance. The conventional concept of competitiveness is widely used in a business strategy, which means that a firm becomes competitive when it gains higher financial returns than the average of its industry peers [28]. Similarly, the authors further argued that the individual dimension of the CSRD is equally important, as certain information may be lost while using the aggregate CSRD against financial performance. Thus, it is necessary to find a link between the individual dimensions of CSRD and financial performance. The detailed literature review and the previous discussion hint at a positive association of the particular dimensions of CSRD, i.e., Ethical, Legal, Environmental, Economic, and Philanthropic, with financial performance. Therefore, the following hypotheses are developed.

Hypothesis 1.

The CSRD has a positive impact on the financial performance of the Islamic banking industry in Pakistan.

Hypothesis 2.

The ethical dimension of CSRD has a positive impact on the financial performance of Islamic banks.

Hypothesis 3.

The legal dimension of CSRD has a positive impact on the financial performance of Islamic banks in Pakistan.

Hypothesis 4.

The economic dimension of CSRD has a positive impact on the financial performance of Islamic banks in Pakistan.

Hypothesis 5.

The environmental dimension of CSRD has a positive impact on the financial performance of Islamic banks in Pakistan.

Hypothesis 6.

The philanthropic dimension of CSRD has a positive impact on the financial performance of Islamic banks in Pakistan.

2.3. Conceptual Framework

The conceptual framework of the study is based on Carroll’s CSR model and the two approaches, namely normative and instrumental, of the stakeholder theory. The Carroll CSR model supports the ethical, legal, economic, environmental, philanthropic, and financial performance [29]. The approaches of stakeholder theory further underpin the relationships among these variables; the Islamic banks try to satiate the demand of multiple stakeholders in the normative approach and avail instrumental benefits in the shape of better financial performance [13,26]. These relationships are depicted in Figure 1.

3. Materials and Methods

3.1. Sample for the Research

In Pakistan, the total population of Islamic Banks registered with the State Bank of Pakistan (SBP) was five banks, namely the MCB Islamic Bank Limited, Bank Al Meezan Limited, Bank Al Baraka Limited, Bank Islami Limited, and Dubai Islamic Bank Limited [30]. However, the study excluded MCB Bank Limited, as it operates in both the Islamic and conventional or commercial banking, and had no separate annual report on Islamic banking and CSRD. The study selects the data collection period from 2012 to 2017, as many reforms regarding the environment, society, and governance (ESG), along with the concept of green banking, its regulations in Pakistan, and the codes of corporate governance of 2012 and 2017, were introduced during this period.

3.2. Measurement of Financial Performance

Most of the previous studies that investigated the association between CSRD and financial performance mainly relied on the accounting aspect of firms’ performance [31]. These studies used return on assets (ROA), which is a key indicator of the firms’ financial performance showing managerial efficiency, ability, and competency to generate profits by using the assets of the banks. Some studies also used the accounting measure of Return on equity (ROE) for measuring firm performance as an additional indicator or robustness check. It is also necessary to mention that ROE is strongly linked with the ROA, and, therefore, it is important to use this ratio as an extra measure of a bank’s performance. Hence, this study uses ROA to measure the financial performance of the Islamic banks and employs ROE for further validation of the results.

3.3. Measurement of CSR Disclosure

For data collection, the study relied on the annual reports of Islamic banks of Pakistan. The procedure of content analysis was adopted for data collection from the annual reports of the sample banks. The content analysis procedure is considered less expensive and unmistakable as compared to some other methods. Hence, it is frequently used in previous studies related to CSRD [32,33]. Following prior studies, an unweighted scoring method was applied to record the data for the CSRD index by using a dichotomous procedure, where disclosure is assigned a value of 1 if an item is disclosed in the report of a bank, and 0 otherwise [26,34]. The study used five dimensions of CSRD, namely Ethical, Legal, Environmental, Economic, and Philanthropic, which constitute composite CSRD. The sub-dimensions of each dimension are marked as 1 in the case that the sub-dimension is reported in the annual report, and 0 otherwise, by using the content analysis technique for the respective bank. Finally, adding these scores contributes to the total disclosure of CSRD for each sample bank. For each dimension of CSRD, this procedure is summarized by the equation as $\frac{\sum \mathrm{d}i}{N}$, where d = 1 if the CSRD exists and 0 if not, while N is the total number of maximum possible disclosures for the given dimension. Similarly, the whole equation for the composite CSRD is given as:

$$\mathrm{CSRD}=\frac{{\sum }_{i=1}^{\mathrm{n}}{X}_{ikt}}{N}$$

where the CSRD index _kt represents the disclosure index for dimension k and time t, while X is a variable starting from 1 to n (1 … n) for the given dimension k and the time t.

ROA represents a return on assets, which is measured by dividing the net income by the total assets, while the CSRD is the total CSR disclosure score, which can be calculated as dividing the total disclosures of the sub-dimensions by the maximum number of sub-dimensions possibly reported by a bank. The CSR dimensions consist of the Ethical, Legal, Environmental, Economic, and Philanthropic dimensions, which compose the CSR aggregate, i.e., CSRD. The capital ratio, represented by Cap Ratio, is calculated as the equity capital divided by total assets for the respective year. Size represents the size of the bank, which will be calculated as the log of total assets. The debt ratio will be calculated as the ratio of long-term debt to the total assets. Similarly, the firm’s age can be calculated by taking the years since the bank was registered or started its operations. Here, έ represents the error factor, and α is the intercept in the regression model. Return on equity (ROE), used as a robustness check, is measured as the net income divided by shareholders’ equity.

3.4. Estimation

The random-effect (RE) and fixed-effect (FE) models of Generalized Least Squares (GLS) were applied to the panel data to investigate the impact of CSRD on the financial performance of Islamic banks. In addition, the study used the Ordinary Least Squares (OLS) regression and Panel-Corrected Standard Errors (PCSEs) estimators for the robustness of the GLS estimations. However, the study claims the results of GLS, as these are more robust than other estimators for panel data analyses. The GLS random-effect model assumes the variation across the entities to be random and uncorrelated with the predictor or independent variables included in the model. In contrast, the fixed-effect model of GLS controls is for all of the time-invariant differences between the cross-sections; hence, the estimated coefficients of the fixed-effect models cannot be biased because of omitted time-invariant characteristics. The author explains the difference as “… the crucial distinction between fixed and random effects is whether the unobserved individual effect embodies elements that are correlated with the regressors in the model, whether these effects are stochastic or not” [35]. In GLS, the Hausman test was employed to decide between the random-effect and fixed-effect models. If the result of the Hausman test is significant (Chi 2, Prob. > Chi 2), then the fixed-effect model is appropriate; if not, then the random-effect model is selected. Most of the previous studies related to CSR and financial performance employed the same methodology [36,37]. The results of the Hausman test in this study are reported below. The research models are shown as follows.

Model I

ROA = α + β1CSRD + β2SIZE + β3Cap Ratio + β4 Age + β5 Debt Ratio + έ

Model II

ROA = α + β1Dimensions + β2SIZE + β3Cap.ratio + β4 Age + β5 Debt Ratio + έ

4. Results and Discussion

The descriptive statistics with the Skewness and Kurtosis show that some variables were not normally distributed, which might be due to outliers. Thus, the data were normalized by using the Van der Waerden data distribution method. The Kurtosis statistics are in a range of ±3 and the Skewness in a range of ±1.96 [21], as reported in

Table 1. Descriptive statistics.

	Minimum	Maximum	Mean	Std. Deviation	Skewness		Kurtosis
	Statistic	Statistic	Statistic	Statistic	Statistic	Std. Error	Statistic	Std. Error
ROA	−0.14	0.95	0.20	0.27	1.21	0.47	0.93	0.91
ROE	0.02	0.28	0.12	0.07	0.45	0.47	0.41	0.91
Firm Size	0.10	11.81	8.46	4.93	−1.22	0.47	−0.54	0.91
Capital ratio	0.03	0.17	0.11	0.03	−0.83	0.47	0.66	0.91
Debt ratio	2.78	20.5	11.59	5.61	−0.32	0.47	−1.08	0.91
Age	2	14	8.25	3.08	−0.17	0.47	−0.42	0.91
CSRD	35	78	57.38	14.26	−0.19	0.47	−1.36	0.91
Environmental	2	10	5.21	2.58	0.25	0.47	−1.04	0.91
Legal	3	9	6.71	1.45	−0.26	0.47	0.37	0.91
Ethical	2	13	7.88	3.71	−0.37	0.47	−1.33	0.91
Economic	5	10	7.04	1.68	0.40	0.47	−0.81	0.91
Philanthropic/social	18	37	26.46	6.15	0.24	0.47	−1.14	0.91

. In addition, the mean and the standard deviation, the minimum statistics, and the maximum statistics with the mean statistics are also reported in Table 1. To check the heteroscedasticity, the Breusch–Pagan/Cook–Weisberg test and White’s test were conducted on both the composite CSRD model and the dimension model. In both the models, there was no issue of heteroscedasticity, as the probability of both the tests (Breusch–Pagan/Cook–Weisberg and White’s tests) was less than 0.5. Therefore, the data are fit for regression.

The descriptive statistics for all of the variables of the study from the given sample of 24 observations over 6 years from all four full-fledged Islamic banks across the country are reported in Table 1. The findings show that the lowest value is −0.140 and the highest value is 0.950, with a standard deviation of 0.273 for ROA. The mean for ROA is 0.202, which indicates that return on assets at the given period, i.e., 2012–2017, remained stable. However, as far as the CSRD index is concerned, the minimum statistic is 35, while the maximum value is 78, along with the mean value of 57.38. These statistics indicate that the banks have increased their disclosures during the given period, which shows their confidence and belief in CSR practices and CSRD. This also provides a clear indication of the Islamic banks’ commitment towards increasing their interaction with the customers and the investors, as well as the society at large, to reduce the communication gap and improve the overall image of the Islamic banking industry. The positive increase in the CSRD can be interpreted as: The banks consider socially responsible activities very crucial for shaping the public opinion in favor of the banks. The average size of the bank has a lowest value of 0.103 and a highest value of 11.81, with a mean value of 8.46. This shows an overall expansion in the size of the banks and their assets over the given period. Similarly, the capital ratio has a lowest value of 0.035 and a highest statistic of 0.172, with a mean value of 0.116 and a standard deviation of 0.033. The lowest statistics for debt ratio are 2.780, while the highest statistics are 20.580, with mean statistics of 11.59, which shows an increase in the level of debt. This indicates that banks increasingly dependent on debt to finance their operations. The firm’s age has a minimum statistic of 2 and a maximum statistic is 14, with a mean value of 8.25, showing that most of the sample banks have an average age of less than 10 years. The minimum statistic for the individual dimension of CSRD, i.e., the environmental dimension, is 2, and the maximum is 10, while the mean score is 5.21. The minimum number for the legal dimension is 3 and the maximum is 9, with a mean value of 6.71. The minimum statistic for the ethical dimension of CSRD is 2, while the maximum is 13 and the mean score is 7.88. The minimum value for the economic dimension of CSRD is 5, while the maximum is 10. The mean value is 7.04 with a standard deviation of 1.68, which is pretty handsome, showing a strong and growing commitment of the banks towards the disclosure of this dimension. The minimum value for the philanthropic/social dimension of CSRD is 18 and the maximum value is 37, with a mean score of 26.46.

From

Table 2. Correlation matrix.

	(1)	(2)	(3)	(4)	(5)	(6)	(7)	(8)	(9)	(10)	(11)	(12)
ROA (1)	1
Size (2)	−0.03	1
Age (3)	−0.01	0.90 *	1
Capital Ratio (4)	−0.08	0.04	−0.09	1
Debt Ratio (5)	−0.06	0.72 **	0.88 **	0.11	1
Environment (6)	0.02	0.80 *	0.79 **	−0.09	0.87 *	1
Legal (7)	−0.647 **	0.16	0.10	0.23	0.20	0.11	1
Ethical (8)	−0.70 *	0.33	0.27	0.33	0.33	0.28	0.81 **	1
Economic (9)	0.55 **	0.29	0.17	0.31	0.31	0.19	0.78 **	0.73 **	1
Philanthropic (10)	0.73 **	0.36	0.27	0.27	0.36	0.29	0.77 **	0.88 **	0.78 **	1
CSRD (11)	0.56 **	0.33	0.31	0.10	0.26	0.33	0.80 **	0.81 **	0.63 **	0.68 **	1
Lag ROA (12)	0.41	0.89 **	0.10	0.04	−0.11	0.09	0.06	0.63 **	0.69 **	0.61 **	0.78 **	1

Note: Standard errors are in parentheses. ** p < 0.05, * p < 0.1.

, as shown, the Pearson correlation matrix does not detect any multicollinearity, i.e., above 0.8 between two predictors. So, the predictors that are to be tested for regression are not highly correlated and are, therefore, fit for further analysis.

Figure 2 shows a relation between the average percentages for CSRD over the period 2012–2017. The figure indicates that the average CSRD in 2012, 2013, and 2014 is 49%, while from 2015, there is an abrupt change in percentage, reaching a score of 53%. It falls back to 51% and 50.8% in 2016 and 2017, respectively. During the corresponding period, the average percentage score for the industry remained at 51%.

The results of the Hausman test for all of the models are reported at the ends of

Table 3. CSRD and financial performance.

	(1)	(2)	(3)	(4)	(5)	(6)
	ROA OLS	ROA GLS Random	ROA GLS Fixed	ROE OLS	ROE GLS Random	ROE GLS Fixed
CSRD	−0.587 **	−0.587 ***	−0.053	0.022	0.022	0.009
	(0.201)	(0.201)	(0.286)	(0.146)	(0.146)	(0.109)
Firm size	−1.331	−1.331	0.396	0.667	0.667	1.094 *
	(1.344)	(1.344)	(1.341)	(0.740)	(0.740)	(0.524)
Capital ratio	0.028	0.028	0.725	−0.050	−0.050	0.410 **
	(0.254)	(0.254)	(0.481)	(0.136)	(0.136)	(0.179)
Debt ratio	−0.502	−0.502	−0.331	−0.365	−0.365	−0.322
	(0.492)	(0.492)	(0.514)	(0.265)	(0.265)	(0.193)
Age	2.303	2.303 *	0.876	−0.279	−0.279	−1.630
	(1.332)	(1.332)	(3.792)	(0.730)	(0.730)	(1.438)
Lag ROA	0.130	0.130	−0.067	-	-	-
	(0.250)	(0.250)	(0.261)	-	-	-
Years	−0.356 *	−0.356 *	−0.215	0.015	0.015	0.344
	(0.184)	(0.184)	(1.090)	(0.099)	(0.099)	(0.400)
Lag ROE	-	-	-	0.787 ***	0.787 ***	0.164
	-	-	-	(0.144)	(0.144)	(0.199)
_cons	716.232 *	716.232 *	433.129	−31.300	−31.300	−693.116
	(370.697)	(370.697)	(2195.810)	(199.219)	(199.219)	(806.422)
Obs.	24	24	24	24	24	24
R−squared	0.542	0.542	0.344	0.837	0.837	0.521
Hausman Test (Chi2)	-	8.3	-	-	-	34.82 ***
Prob. > Chi 2	-	0.307	-	-	-	0.000

Note: Standard errors are in parentheses. *** p < 0.01, ** p < 0.05, * p < 0.1.

and

Table 4. Dimensions of CSRD and financial performance.

	(1)	(2)	(3)	(4)	(5)	(6)
	ROA (OLS)	ROA (GLS Random)	ROA (GLS Fixed)	ROE (OLS)	ROE (GLS Random)	ROE (GLS Fixed)
Environmental	0.821 **	0.821 **	0.723	0.283	0.283	0.142
	(0.324)	(0.324)	(0.411)	(0.204)	(0.204)	(0.200)
Legal	−0.011	−0.011	0.364	−0.102	−0.102	0.364
	(0.304)	(0.304)	(0.394)	(0.176)	(0.176)	(0.205)
Ethical	−0.436	−0.436	−1.052 *	0.472*	0.472 **	−0.347
	(0.370)	(0.370)	(0.548)	(0.232)	(0.232)	(0.320)
Economic	0.645 *	0.645 *	0.384	0.583 **	0.583 ***	0.062
	(0.339)	(0.339)	(0.376)	(0.203)	(0.203)	(0.222)
Philanthropic	−0.598	−0.598	−0.675	0.009	0.009	0.056
	(0.399)	(0.399)	(0.583)	(0.230)	(0.230)	(0.289)
Firm size	−0.799	−0.799	−0.287	0.556	0.556	1.244 *
	(1.067)	(1.067)	(1.181)	(0.631)	(0.631)	(0.575)
Capital ratio	0.086	0.086	0.926	−0.238	−0.238	0.698 *
	(0.195)	(0.195)	(0.686)	(0.157)	(0.157)	(0.356)
Debt ratio	0.005	0.005	0.325	−0.326	−0.326	−0.285
	(0.371)	(0.371)	(0.487)	(0.218)	(0.218)	(0.237)
Firm age	1.821	1.821 *	3.936	0.119	0.119	−1.432
	0.821 **	0.821 **	0.723	0.283	0.283	0.142
Lag ROA	0.087	0.087	0.042	-	-	-
	(0.216)	(0.216)	(0.217)	-	-	-
Years	−0.509 **	−0.509 ***	−0.973	−0.269 *	−0.269 *	0.328
	(0.186)	(0.186)	(1.169)	(0.139)	(0.139)	(0.571)
Lag ROE	-	-	-	0.722 ***	0.722 ***	0.018
	-	-	-	(0.203)	(0.203)	(0.278)
_cons	1026.163 **	1026.163 ***	1959.252	542.467 *	542.467 *	−660.113
	(374.915)	(374.915)	(2355.501)	(279.282)	(279.282)	(1151.203)
Obs.	24	24	24	24	24	23
R-squared (Overall)	0.830	0.830	0.752	0.928	0.928	0.715
Hausman Test (Chi2)	-	5.000	-	-	11.25	-
Prob. > Chi 2	-	0.931	-	-	0.423	-

Note: Standard errors are in parentheses. *** p < 0.01, ** p < 0.05, * p < 0.1.

. In all of the models, the results indicated that the random-effect model was an appropriate choice, as it is believed that the omitted variables only vary over time or between cases. Likewise, when such a situation exists, the random-effect model can give relatively more robust results [38]. Furthermore, endorsing our results, it is also acknowledged that Islamic banks in Pakistan are limited in number; hence, they keep a close eye on each other, especially regarding CSR activities and market strategies for the satisfaction of multiple stakeholders. Table 3 explains 54.2% of the variation in ROA and 52.1% variation in ROE due to change in the predictors. The model is significant at p < 0.01. The results indicate that CSRD has a significant negative relationship with financial performance (ROA) at p < 0.01. These results, which endorse prior studies, noted a significant negative relationship between CSR and financial performance [31,39]; this may have a plausible explanation in that increases in CSR activities and their financial cost have negatively affected the financial performance of the banks by decreasing their profitability. However, these findings are contradictory to the first hypothesis, H₁, of the study; thus, it is rejected. The significantly negative coefficient (instead of positive) for CSRD with financial performance is also contrary to the suppositions of the stakeholders. Nevertheless, the findings are consistent with the postulations of the tradeoff hypothesis presented by Friedman, that the firms’ prime responsibility is to maximize the profits of the stakeholders [39]. Hence, managers and firms involved in social activities could cause extra costs to the firms that ultimately decrease their profits. In short, the theory suggests that the greater the level of CSRD, the smaller would be the financial performance of the firm [37,40].

Figure 2 indicates that the Islamic banking industry has increased CSRD, but, still, the stakeholders’ perception in this regard is not positive, which, on one hand, shows a strong commitment of the industry towards CSRD, but, on the other hand, a poor public or customer response. This might be due to the rationale that “Pakistanis are the most generous people. Pakistan’s contribution to charity is almost 1% of its GDP, which distinguishes it in the list of top-ranked countries with respect to charitable activities, like the United Kingdom contributing 1.3% of GDP and Canada contributing 1.2% of its GDP to charity; Pakistan gives about twice what India gives to charity” [41]. Despite all of these facts, the significantly negative coefficient of CSRD may also signal that the message of the Islamic banking industry has not been properly communicated to the stakeholders. As a plausible explanation for this, the annual reports published by Islamic banks are not well organized. Except for Bank Al Meezan Limited, the rest of these reports could not properly showcase the CSR-related information, which might affect the behavior of the potential customers. The average industrial age reveals that the Islamic banking industry is in the emerging stage; therefore, it would take time to acquire a stage of maturity. The country has also passed through an intense wave of terrorism and extremism starting from 2002 onward, which has affected business activities, the overall economy in general, and the banking industry in particular. However, the average CSRD shows a slight increase in the average CSRD for the last couple of years, while an average decrease has been reported in the financial performance (ROA). The findings for ROA are also consistent with those of ROE, which is used as an additional measure to gauge performance and validate the results.

The regression results reported in Table 4 explain 83% of the variation in ROA and 92.8% variation in ROE due to change in the predictors. These results show that all of the dimensions of CSRD are statistically insignificant with the firm’s current financial performance (ROA), except the environmental and economic dimensions. These results also indicate that the ethical dimension of CSRD is insignificant with the firm’s financial performance, which rejects hypothesis H₂ of the study. It is to be noted that the ethical dimension of CSRD is statistically significant with the ROE at significance level p < 0.05. The ROE, which is used as a robustness check throughout the analysis, is mostly associated with the owners of the banks. Therefore, the owners of the banks are comparatively more attracted to the ethical issues, as these are worthy of the goodwill of the banks without incurring any direct cost to them. Similarly, the legal dimension of CSRD is also statistically insignificant with the firm’s financial performance, so hypothesis H₃ is also rejected. These results endorse a previous study showing that most of the developing countries have a lower priority for the legal dimensions [42]. This does not necessarily mean that firms in the developing countries flaunt the laws, but they are under minor pressure to operate responsibly. In these countries, there are certain issues in the legal systems that do not pressure the firms to operate in full compliance with the relevant laws. These countries are also far behind the developed countries in practicing issues like human rights and others such as the CSR in their legislation [34].

However, the environmental dimension of CSRD has a positive and significant relationship with the financial performance of the sample Islamic banks at a significance level of p < 0.05. Therefore, the hypothesis H₄ is accepted. This shows that the customers and other stakeholders pay greater attention to environmental issues. The previous studies indicate inconsistent results for the relationship between the environmental dimension and financial performance. Some of these studies found a negative or neutral association [43,44]. However, the results of the current study are in line with many previous studies showing a significant positive relationship between the environmental dimension of CSRD and the financial performance [31,45,46,47,48]. The economic dimension of CSRD has a positive and statistically significant relationship with the financial performance (ROA) at a significance level p < 0.10, which confirms hypothesis H₅ of the study. Similarly, the economic dimension of CSRD is also statistically significant and positive with the ROE at a significant level p < 0.01. The findings which support H₆ are compatible with a prior study showing a positive and significant relationship between the economic dimension of CSRD and financial performance [49]. The Islamic banking industry has flourished for the last couple of years, especially in Pakistan, with a high profit margin, based on Shariah; therefore, the customers are increasingly inclined towards Islamic banking due to the economic transparency and other benefits offered. The customers are attracted to the Islamic banking industry due to its operations based on the principles of profit and loss. As indicated by Figure 2, these banks are more inclined towards publicizing the economic information because, economically, they are considered to be more transparent, committed, and ethically responsible.

The philanthropic dimension of CSR also has an insignificant relationship with ROA, which rejects hypothesis H₆ of the study. As we know, Islam has a social and legal code of conduct, which is its beauty. The philanthropic aspect of CSRD is what the customers expect from an Islamic bank. The negative and insignificant relationship of the philanthropic dimension with financial performance could be possibly explained in that Islam discourages disclosing the philanthropic efforts due to ethical considerations [50]. Islam discourages both individuals and groups from disclosing the philanthropic and social practices they carry, as this may hurt the ego, dignity, and self-esteem of those who receive Zakat, Sadqat, Qard-e-Hassana, donations, charity, etc. Islam also warns that the publicity of philanthropic activities may hurt the true sense of giving (or welfare), which affects its promised return in the eternal world. Hence, philanthropic practices are usually kept hidden and undisclosed in an Islamic society; if they are not, or are reported, the customers may perceive these negatively. In addition, the study also estimated all models through Panel-Corrected Standard Error (PCSE), which is considered an appropriate estimator for micro-panel datasets where N is small and T is large [51,52,53,54,55]. The findings reported in

Table 5. Panel-Corrected Standard Errors.

	(PCSEs)	(PCSEs)	(PCSEs)	(PCSEs)
	ROA	ROE	ROA	ROE
CSRD	−0.587 ***	0.022	-	-
	(0.198)	(0.072)	-	-
Firm size	−1.331	0.667	−0.799	0.556
	(1.017)	(0.545)	(0.869)	(0.523)
Capital ratio	0.028	−0.050	0.086	−0.238 **
	(0.239)	(0.104)	(0.202)	(0.119)
Debt ratio	−0.502	−0.365 ***	0.005	−0.326 **
	(0.481)	(0.132)	(0.243)	(0.159)
Firm age	2.303 **	−0.279	1.821 **	0.119
	(0.921)	(0.494)	(0.790)	(0.492)
Lag of ROA	0.130	-	0.087	-
	(0.241)	-	(0.198)	-
Lag of ROE	-	0.787 ***	-	0.722 ***
	-	(0.102)	-	(0.165)
Environmental	-	-	0.821 ***	0.283
	-	-	(0.262)	(0.179)
Legal	-	-	−0.011	−0.102
	-	-	(0.189)	(0.131)
Ethical	-	-	0.436	0.472 ***
	-	-	(0.339)	(0.156)
Economic	-	-	0.645 **	0.583 ***
	-	-	(0.306)	(0.181)
Philanthropic	-	-	−0.598	0.009
	-	-	(0.382)	(0.111)
Years	−0.356 ***	0.015	−0.509 ***	−0.269 **
	(0.115)	(0.062)	(0.140)	(0.106)
Constant	716.232 ***	−31.300	1026.163 ***	542.467 **
	(232.074)	(124.751)	(281.519)	(213.381)
Obs.	23	23	23	23
R-squared	0.542	0.837	0.830	0.928

Standard errors are in parentheses. *** p < 0.01, ** p < 0.05, * p < 0.1.

under the PCSEs are consistent with those of OLS and GLS, which further authenticates the accuracy and reliability of all of the estimations in the current study.

5. Conclusions and Recommendations

This study aimed to examine the relationship between CSRD and the financial performance of the Islamic banks in Pakistan. The findings indicate that the overall level of CSRD for the given sample is 51%, which is slightly different from the CSRD reported by a previous study [20]. The prior study shows that Islamic banks in Pakistan disclose around 30.4% of CSR in their annual reports. This indicates that CSRD increased from 2014 onward. The findings also reveal a significant negative relationship between the CSRD and financial performance of the Islamic banks in Pakistan. Concerning separate dimensions, the legal, philanthropic, and ethical dimensions also have no significant or positive associations with the financial performance of these banks. Though these findings are not in line with most of the established hypotheses (H1, H2, H3, and H6) of the study and postulations of stakeholder theory, still, these results are compatible with Friedman, who argued that CSRD may cause extra costs to the firms that ultimately decrease their profits. In short, the theory suggests that the greater the level of CSR, the smaller the financial performance of the firm will be. Hence, the results are contrasted with the theoretical argument of stakeholder theory, which posits that CSRD improves firms’ financial performance by increasing their goodwill. In addition, Islam supports and encourages carrying out CSR activities, but at the same time discourages their disclosure (CSRD), especially when it is related to the social and philanthropic activities for individuals and society. The discouragement has a rationale that reporting or publicizing these activities may hurt the ego, self-esteem, and respect of the individuals at the receiving end. Because of this, CSRD of the Islamic banks might have created a negative perception in the minds of customers, especially in the cultural context of Pakistan, that could not yield the desired results, i.e., improving firm performance.

Unlike the others, the economic dimension has a significant positive association with financial performance, which may be due to the customers’ confidence in the merit and transparency of financial matters of the Islamic banks. The findings might also show the rationale that customers express trust in the profit and loss sharing mechanisms of Islamic banks, as they function per the Shariah principles of Islam. The average percentage score also suggests that priority has been given to the economic dimension in the average CSRD. Similarly, the environmental dimension also shows a significant positive relationship with financial performance, while the ethical dimension shows a significant positive association with the financial performance—the ROE shows a strong commitment of the banking industry towards disclosing information regarding these dimensions, causing a positive impact on the financial performance of the Islamic banking industry. These dimension-wise results support the stakeholder theory, but contradict the Friedman theory.

These results indicate that most of the Islamic banks, as expected, disclosed comparatively high levels of composite CSR practices, but lack a balanced disclosure of the individual dimensions of these practices. It is noted that priority is given to disclosing information related to financial obligations as compared to other dimensions. Except for Bank Al Meezan Limited, the annual reports published by Islamic banks were also found to be non-systematic, as the required information was scattered throughout these reports, making it difficult for the potential customers to access it, which may be one of the reasons for the unexpected results, especially H1, H2, H3, and H6. Being based on the principles of Shariah, Islamic banks are expected to be socially, ethically, and morally responsible, and to publish their CSR at the top to justify their moral and socially friendly nature. These institutions cannot be called true Islamic banks if they do not meet the criteria set by the principles of Shariah.

The standards of the AAOIFI, with which compliance is voluntary, aim to regulate the CSRD in Islamic banks. Likewise, the GRI reporting guidelines for CSRD in conventional banks or other business firms are also voluntary. By merging both the standards of AAOIFI and GRI, this study develops a novel index to standardize and unify the reporting of CSRD according to internationally acceptable standards in the framework of Shariah for Islamic banks. This index may be mandated for increasing CSRD, as the voluntary nature of compliance with both indexes might be a reason for low CSRD. Furthermore, the Islamic banking industry may focus on channels of communication like print and electronic media, along with using simple local languages, so that the messages can be effectively communicated to the target customers/stakeholders. The industry may also follow a comprehensive CSR policy that could counter the propaganda and increase its goodwill. The AAOIFI, a regulating authority, may also revisit its standards for CSR and CSRD in light of the new index (Appendix A,

Table A1. Comprehensive Corporate Social Responsibility Disclosure Index.

Dimension of CSRD	Sub-Dimensions of CSRD
1. Environment	▪ Introduction to greenhouse products
	▪ Definition of greenhouse products
	▪ Controlling measures for the emission of greenhouse gases (CO2, CO)
	▪ Energy conservation
	▪ Measures for water consumption/availability
	▪ Biodiversity
	▪ Transportation
	▪ Supplier environmental assessment
	▪ Conferences on environment-related issues
	▪ Amount of donations to environmental protection
	▪ Investments in sustainable developmental projects
	▪ Investment in environmentally friendly projects
	▪ Focus on risk-based corrective actions
	▪ Measures for the restoration and protection of natural resources
	▪ Stakeholders’ involvement in environmental issues
	▪ Environmental grievances mechanism
	▪ Climate change policy
	▪ Any award for environmental achievement
2. Legal	▪ Follow the Shariah and accounting rules
	▪ Whether the principles of Shariah are followed
	▪ Whether the human rights labor laws are followed
	▪ Whether the bank is dealing with legal products
	▪ Whether the bank observes anti-money-laundering laws
	▪ Provides secure transaction facility to customers
	▪ Tax payment
	▪ Any internal legal advisory committee
	▪ Any legal Shariah committee
	▪ Training on anti-corruption and other legal issues
3. Ethical	▪ Any anti-money-laundering policy
	▪ Whether the transactions are free of Riba
	▪ Whether the fund’s sources are disclosed for its customers
	▪ Provides the correct information to its customers
	▪ Prevents corruptions and irregularities in the banking system
	▪ Any internal regulating body dealing with fraud and anti-corruption
	▪ Any internal regulating body dealing with sexual harassment and workplace violence
	▪ Policies regarding sexual harassment and workplace violence
	▪ Non-discriminative policies regarding sex, age, and ethnicity
	▪ Dealing with legal items only
	▪ A proper code of ethics for the accountants
	▪ A proper code of ethics for the internal auditors
	▪ Disciplinary action committee
	▪ Code of ethics for the employees
	▪ Any grievances mechanism regarding ethical issues
4. Economic	▪ Whether the revenues generated are disclosed
	▪ Employees’ wages and benefits are reported
	▪ Whether the paid taxes are reported
	▪ Whether profit and loss statements are reported
	▪ Payments to the equity/capital owners reported
	▪ Any community investment reported
	▪ Economic value distributed in the form of operating cost reported
	▪ Procurement policy
	▪ Any economic Achievement Award
	▪ Any economic award
	▪ Economic grievances mechanism
	▪ International economic appreciation award
5. Philanthropic/social	❖ Community and Social Development
	▪ Donations for health issues
	▪ Donations for sports activities
	▪ Participating in relief and disaster management issues
	▪ Donations for education
	▪ Microfinance
	▪ Funding other organizations for social activities
	▪ Establishing a comprehensive link with the public industry/society
	▪ Involvement in government-sponsored social activities
	▪ Creating job opportunities
	▪ Job opportunities for special persons
	▪ Women branches
	▪ Amount of zakat paid
	▪ Those who receive the amount of zakat/zakat beneficiaries
	▪ SSB attestation that the amount of zakat has been computed according to sharia
	▪ SSB attestation that the sources and uses of zakat are according to
	▪ Shariah
	▪ Amount of Sadaqah paid
	▪ Sadaqah beneficiaries
	▪ Qard e Hassana paid
	▪ Beneficiaries of Qard e Hassana
	▪ Policy regarding debt
	▪ Amount of debt written off
	▪ Any public policy
	▪ Credit committee
	▪ Whether local communities are taken on board in social activities
	❖ Product and Service Responsibilities
	▪ Definition or glossary for a new product
	▪ Introduction of SSB-approved new products
	▪ Whether the new product is based on the concept of Shariah
	▪ Any external or internal communication channel with stakeholders
	▪ Regarding product
	▪ Zero investment in non-permissible product or services
	▪ Market survey and feasibility report
	▪ Efforts for research and development promotion
	▪ Products with customers’ health and safety
	▪ Riba-free products
	▪ Ensuring the customers’ privacy
	▪ Whether customers are provided with access to the online banking services
	❖ Commitment towards Employees
	▪ Efforts for a diversified staff
	▪ Employees’ health and safety
	▪ Providing equal employment opportunities
	▪ Provide training on Shariah awareness
	▪ Provide training on professional skills and challenges
	▪ Providing higher education opportunities to employees
	▪ Employee appreciation
	▪ Employees’ safety and protection
	▪ Equal remuneration for men and women
	▪ Proper promotion mechanism/promotion policy
	▪ Remuneration committee
	▪ Standard labor practices policy
	▪ Employee management interaction
	▪ Secure internet facilities for the employees
	▪ Grievances mechanism for employees

Data Source: Authors.

) to meet the upcoming challenges of the Islamic banking industry. In addition, to locate CSR-related information easily, the Islamic banks are also required to publish their annual reports in Urdu and other local languages to minimize the communication gap in countries like Pakistan, where the literacy rate is not high and the general public has a low understanding of the English language.

This study investigated the impact of both the composite CSRD and its dimensions on the financial performance of the Islamic banks in Pakistan from 2012 to 2017. By using the index of this study, studies in the future may work on the data of 2017 onward in Pakistan, as well as on large sample sizes in other countries with Islamic banks. In the future, moderation and mediation of integrated strategies and corporate governance could also be explored in the model of CSRD and financial performance of the Islamic banks. Likewise, for a clearer picture, the qualitative aspect, along with newsletters and websites of the respective banks, could also be considered by future studies. Last but not least, the new green banking regulations of 2017 may also be considered from the perspective of the Islamic banking industry of Pakistan in the future.