Essay Writing Service

1 Star2 Stars3 Stars4 Stars5 Stars (No Ratings Yet)
Loading...

Impact of Corporate Social Responsibility Disclosure on Financial Performance

 The Impact of Corporate Social Responsibility Disclosure on Financial Performance: A Comparative Study between Mozambique and South Africa in the Banking Sector

Abstract

Nowadays there is a growing interest in Corporate Social Responsibility (CSR) activities, both in the professional and academic fields. The purpose of this article is to examine the relationship between Corporate Social Responsibility (CSR) and Financial Performance (FP) by using the top 10 ranked banks respectively in Mozambique and the Republic of South Africa (RSA). Covering five years of observation (2012-2016) and applying the content analysis to assess the CSR dimensions and the general measure of FP such as Return On Asset, Return On Equity and Return On sales, we find a significant association between FP and CSR, suggesting that CSR behavior improves banks’ performance. The study further concludes that the RSA banks are disclosing more information regarding CSR than Mozambican banks. Additionally, the evidence indicates that Mozambican banks primarily disclose information about their customer & products comparatively to other categories of CSR, and the RSA banks mostly disclose information about their environmental initiatives.

KEYWORDS: Corporate Social ResponsibilityFinancial PerformanceBanking Sector, Mozambique, RSA

 

INTRODUCTION

Companies are increasingly focused on participation in Corporate Social Responsibility (CSR) activities and the contribution of the banking sector to sustainable development is being questioned, because the banking sector consume vast amounts of natural resources, such as energy, paper and others, and create wastes. CSR activities are a way of contributing directly or indirectly to the sustainability of the whole society. On the other hand, Siueia & Wang (2017) indicates that the banking sector is vital to improve the economy of a country, in this context, bank managers must act in a sustainable way as a concern for environmental protection,  rather than the maximization of the shareholder returns.

Several authors, have examined the relationship between Corporate Social Responsibility (CSR) and Financial Performance (FP), in the banking sector (El Moslemany and Etab, 2017; Tijani et al., 2017;  Mulyadi & Anwar, 2012; Bae, Kang and Wang, 2011; Mallin et al., 2014; Platonova et al., 2016; Galant & Cadez, 2017). However, the findings are mixed. For example, some surveys found a neutral relationship in the field, some found a negative relationship and others found a positive relationship between CSR and FP. Furthermore, of these studies focused in Western countries, Asian countries, North African countries and, evidence from sub-Saharan African countries remains rare, in a special case, when it comes to a comparative study among Mozambican (Moz) banks and the Republic of South Africa (RSA) banks the studies does not exist. There is a gap in the literature that must be fulfilled. Therefore, in acknowledge of us, no indexed journal has ever published any article related to the banking sector of these two countries. This research attempts to fill the existing gap in the literature, contributing to the knowledge of how banks strive to achieve better FP through CSR activities in the African context. In practical terms, we believe that the voluntary report on CSR activities could help the banking sector to improve FP.

This study aims to analyze the impact of voluntary disclosure of CSR activities on FP comparing the banking sector of Mozambique (Moz) and the banking sector of the Republic of South Africa (RSA), covering the period from 2012 to 2016. These two countries are located in sub-Saharan Africa. It is also important to emphasize the existence of similarity among foreign banking institutions in Moz and the RSA. However, they have different CSR policies and different CSR strategies. With this, perhaps the CSR activities undertaken in the two banking sectors are also different.

Therefore, the present study tries to answer the following two research questions: “is there a positive relationship between CSR and FP in the banking sector? And, are banks primarily disclosing information regarding their customer and products or their primarily disclosing information regarding their environmental protection?” In the literature, we can find several theories that explains the relationship between CSR and FP (Supply and demand theory, legitimacy theory, agency theory or neo-classical theory, “trade-off theory”, Stakeholder theory, and others). In this research, we applied stakeholder theory to solve the questions, because we believe that this theory better explains the positive relationship between these two variables.

We tested the hypothesis based on data extracted from the audited annual statement of banks available online at the central banks of Moz and RSA, as well as at individual banks’ websites. Taking into account the prior research (Freedman and Jaggi, 1986, Simpson and Kohers, 2002), we applied ROS, ROE and ROA as a measure of FP. In addition, we used the content analysis to evaluate the dimensions of CSR activities extracted from the annual reports as well as data extracted from the respective banks’ website.

The main conclusions are as follows: First, we find a positive and significant relationship between CSR and FP, suggesting that the banking industry is socially responsible. Second, we find evidence that RSA banks primarily disclose information about their environmental protection activities, comparatively to other CSR categories. However, the Mozambican banking sector mostly disclose information regarding their customer & products. Third, we find evidence that banks of Mozambique pay little attention to reporting their CSR activities than RSA banks.

Our study differs from previous studies in several ways. First, we explor the  impact of CSR activities (measured by Evironmental initiative, Human resource policies, Customer and Product and Community involvent) on banks` FP (measured by Return On Assets, Return On Equity and Return On Sales), Second, we measured CSR applying data collected from banks` annual reports and banks` websites. Third, this is the first evidence about Mozambican and RSA Banking sector (an example of sub-Saharan country). Finally, the evidence from the study can help the regulators to understand banks’ business practices in african context.

RELEVANT RELATED LITERATURE AND HYPOTHESIS

Prior study, such as Margolis and Walsh (2003); Branco and Rodrigues (2006); Platonova et al. (2016); Galant & Cadez (2017) defined Corporate Social Responsibility in several ways. For example, Galant & Cadez (2017) recognized 37 different definitions of Corporate Social Responsibility; we use the acceptable definition of Carroll (1999). For Carroll (1999, p. 289) there are four distinct areas that constitute CSR as economic, legal, philanthropic, and ethical. The lowest level of definitions requires the firms to perform with the minimum of business ethics and, on the opposite hand; the highest level includes proactivity suggesting that firms should adopt a sustainable behavior.

Several authors argue that there is a link between Corporate Social Responsibility and Financial Performance and this link may be neutral, for others this relationship may be negative or positively related. More recently, Galant & Cadez (2017), reviewed studies where CSR was associated with Financial Performance, they conclude that the measurement approach is a reason for different conclusions or influence the relationship between CSR and FP. In addition, they pointed out that the researchers are subjective and they select bias that may influence the nature relationship between CSR and FP.

Several authors support the viewpoint of a positive relationship between CSR and FP. According to the proponents, they argue that this link is theorized by the stakeholder theory (Goss and Roberts, 2011; Wu and Shen, 2013; Mallin et al., 2014; Han et el., 2016; Plato nova et al., 2016; Maqbool and Zameer, 2018). This theory support the idea that the companies that adopt socially responsible practices are building their reputation in the society, increasing their profitability and attracting more investors to invest in the company and consequently reducing the operational costs. In contrast, the companies that do not invest in CSR policies their operational costs could be higher and hence decrease their FP.

Highlighting on some authors in the banking sector, that links positively CSR and FP. More recently, Maqbool and Zameer (2018) investigated the relationship between CSR and FP, using a data set from 28 commercial banks listed in Bombay stock exchange (BSE) as a sample of the Indian banking sector, covering the period of 10 years (2007-2016). The findings provided evidence that CSR activity is significantly associated with financial performance, suggesting that the management have to integrate CSR strategic into the business to obtain a better FP. The study by Platonova et al. (2016), using a sample of 222  Islamic cross-bank data sets for the Gulf Cooperation Council (GCC), covering 15 years (2000-2014), using a regression method to test the hypothesis. They concluded that CSR is statistically associated with profitability, implying that CSR activities have a long-term effect on banks’ profitability. Similarly, Mallin et al. (2014); using a sample of Islamic banks cross-country, examined the impact of CSR disclosure on FP over two years (2010-2011), they provided a significant and positive relationship between CSR disclosed index in the annual statements and their FP.

Wu and Shen (2013), using a sample of 162 banks in 22 cross-countries and applying the database of the Ethical Investment Research Service, during the period of 7 years from 2003 to 2009, tested the impact of CSR on FP. The result provided evidence that CSR is positive associated with profitability. Showing that bank managers are engaged in CSR initiatives is vital to the growth of profitability. While in another article, the research of Goss and Roberts (2011) examined the effect of CSR on the cost of banks’ loans. Using a sample of 3996 loans to US companies, they conclude that non-CSR firms pay for loans of 7-18 basis points compared to the socially responsible firms. Additionally, firms are engaging in CSR activities to reduce their volatility and increase their attractiveness for potential borrowers.

From this viewpoint, authors have mainly focused on the positive link between CSR and FP. Therefore we hypothesis as follow:

  • H1:Thereis a positiverelationship between Corporate Social Responsibility reporting and Financial Performancein the Banking sector. (Model 1);

There are several methods to measure CSR (content analysis, surveys, ethical rating, one-dimensional indicators, reputational measures and others). Our Research uses the scores achieved by a content analysis approach. This is widely used in banking sector to obtain the CSR score. So as a reference to our approach we select four components of the drives of CSR behavior (environmental protection, customer & products, community involvement and human resource behavior). Highlighting on some recent studies in this field that uses this method to examine the degree of CSR activities.

A review of Akin and Yilmaz (2016) conducted a similar survey focusing on the impact of voluntary CSR disclosure in the banking sector, using 33 banks as a sample of Turkey over the period of five years (2005 – 2009); based on content analysis of annual reports banks were classified into five different categories of CSR activities. Applying the multivariate regression analysis, to test the hypothesis, the findings provided evidence that the degree and quality of CSR disclosure depends on the characteristics of the bank, the banks` ownership, and the stock exchange listing.

The study of Jain et al. (2015), carried out a similar research focusing on the impact of voluntary CSR disclosure in the banking sector, employing a sample of six giant banks in 4 cross-countries (China, Japan, Australia, and India) covering seven years from 2005 to 2011. The results conclude that the banks are involved in voluntary CSR activities. Notably, the findings showed that the Indian and Chinese banking sector had a fewer level of CSR activities on human resource behavior component comparative to the Australian and Japanese banking sector. In additional, Australian banks were leading the promotion of diversity and equity, while Japanese banks were at the top of the line, promoting balance between work and life. However, regarding the category of community involvement, the results showed that the Chinese and Indian banks were increasing their engagement in CSR activities. While in another article, Ershad and Rahman (2015) reached the same conclusion, using ten Commercial banks by random sampling process between Islamic and conventional banks in the banking sector of Bangladesh. Over the year 2013 and data were based on the secondary data by using the content analysis approach. Applying six category of CSR activities such as Health, Education, Environment, Disaster Relief, Sports & Culture and Social welfare. The results showed that both banks are directly involved in CSR activities. However, the result also showed a smaller impact in the category of social development, suggesting that much can still be done to improve this category. The study also found that Islamic banks have lower total expenses compared to conventional banks in CSR activities.

Taking into consideration the findings of the previous study, we selected solely four components of social responsibility behavior to know how each relatedwith FP. Thus, we hypothesis as follows: H2: All the categories of CSR activities have an individual positive and significant effect on the FP of African Banks. (Model 2). And, we offer four testable sub-hypothesis as following.

  • H2(1): The satisfaction of human resource is positively related to FP;
  • H2(2): The customer & products provided are positively related to FP;
  • H2(3): The environmental protection is positively related to FP;
  • H2(4): The local community is positively related to FP.

As regards to CSR initiative relating to human resource to achieve better financial performance, some authors support the viewpoint of a positive relationship between CSR activities and Financial Performance. Following the organizational commitment in CSR initiatives, the banks can improve employee commitment, satisfaction and better performance.

A review of Pérez and Del Bosque (2012), employing 1,124 observations, as a sample of the Spanish banking sector, in 2010. Investigated the impact of customer perceptions on the CSR activities of their banking service providers. Using a new scale based stakeholder theory, the results revealed that Spanish banks are involved in voluntary CSR disclosure and are disclosing mainly information concerning human resource relationships and community involvement. On the other hand, managers have specific CSR strategies that reveal different CSR policies. Implying that the existent CSR initiatives conceded by the banks increase employee satisfaction and performance. Similarly, Akinpelu et al. (2013), using a sample of 13 Nigerian commercial banks indexed on the Nigerian Stock Exchange over 2009, using descriptive data analysis, examined the factors that determined the degree of CSR disclosure activities in the annual reports and accounts of Nigerian commercial banks and the respective CSR component that were disclosed. The findings show that the sample was engaged in voluntary CSR activity. Specifically, they found that the banks disclosed less information about consumer relationship, environmental protection and product quality, however, the information about human resources activity and community involvement were mostly disclosed. Based on this argument, we hypothesis as follows:

H3.1: RSA Banks are primarily disclosing information regarding their human resources.

As regards to CSR initiative relating to customer and products to achieve better financial performance, some authors support the viewpoint that firms firstly discloser information about their customer and product. Following Mensah et al. (2017), the majority of researchers have frequently focused on the impact of CSR activities on customer behavior, motived to influencing customers and subsequently improving investment, and contributing to environmental protection and achieving social development.

kilic (2016) using a sample of 25 banks of Turkish and using website content analysis approach examined the impact of online CSR activity in the banking sector. They conclude that banks were disclosing primarily information about customers and products. Nevertheless, they do not disclose information about environment initiative and energy. The study of Yeung, S. (2011), carried out a research focusing on the impact of the Role of Banks in CSR. Employing a sample of 3 cross-countries (Hong Kong, USA and Scotland), using questionnaires to collect data set. The results conclude that the banks need to meet customer requirements to become a social responsible bank. Similarly, the study of Macdonald and Rundle (2008), using a sample of 720 observations and applying a multi-method approach, investigated the relationship between CSR and customer satisfaction in the banking sector. They conclude that customer satisfaction is less affected by CSR activities compared to customer-oriented events. And if the bank decides to engage in CSR activities, these activities should be appropriately chosen. Based on this argument, we hypothesis as follows:

H3.2: Mozambican Banks are primarily disclosing information regarding their customer and products.

RESEARCH DESIGN

SAMPLE AND DATA

The sample comprised of 20 top ranked banks, 10 banks for each country, and we did not make a distinction between domestic and foreign banks as well as listed and unlisted banks (Table 1). We excluded in our sample the Bank of Mozambique (the Central Bank of Mozambique) and the South African Reserve Bank (the Central Bank of RSA), because of their different characteristic compared to other banks. Thus, the sample is small we did not consider the effect of outliers in the regressions.

This study is constructed on base of secondary dataset extracted from the annual reports and other bank reports by using content analysis technique to measure CSR disclosure and the regression model to analysis data, covering five years of observation from 2012 to 2016.

ESTIMATION METHODS

MEASUREMENT OF CSR

At present, there are several methods to measure CSR disclosure (surveys, one-dimensional indicators, reputational measures, ethical rating, and content analysis). In line with previous studies, we applied content analysis approach to select and sort banks into CSR disclosure and non-CSR disclosure. (Akin and Yilmaz, 2016; Kilic, 216; Ershad and Rahman, 2015). Previous authors argues that content analysis is a better approach to process, analyze, examine,  interpret, and sort different types of content data and permit to segregate CSR activities into different groups. For each bank, we carefully collected and processed the total number of statements according to specific category of CSR activities. We have divided into four CSR areas such as Environment Protection (EP), Human Resource (HR), Customer and Product (CP) and Community Involvement (CI). More specifically, each of these categories were divided in sub-categories of twenty-two CSR activities are given those four areas, (Table 2). Van Staden & Hooks (2007), argues that companies use several media to disclosure CSR activity. Moreover, because of the improvement in information technology coupled with the globalization, banks’ websites are revealing additional information regarding CSR activities comparatively the annual reports and, the banks’ websites became the principal media for CSR disclosure. Thus, we complement these data searching websites of the banks to identify banks that have issued CSR activities in their other reports. For each categories, banks was scored one if it disclosed related information and zero otherwise. If a component appears in several locations, it is calculated only once. CSR reports via their website as a standalone document for June of 2016 to December of 2016, for each area the banks were scoring from zero to two points. The scale range using following criterion:

0 = the bank does not provide any social information on their website.

1 = the bank provides general social information on their website.

2 = the bank provides particular social information on their website.

Finally, we measured the CSR disclosure (CSR_Dis) as a sum of CSR­­_Score and CSR_Web and it received a score one if the bank were considered disclosing bank and zero where otherwise. The CSR disclosure index (CSR_Dind) was released five consecutive years. In agreement with the previous research, we estimate the CSR disclosure index using the following equation, (Platonova et al., 2016):

CSR_Dind =∑i=1nKijtN

(1)

Where CSR_Dind indicates the CSR disclosure index for dimension j and period t; Kijt is variable K (1, …n) for dimension j and time t; N is the number of statements.

MEASUREMENT OF FP

Several studies used different accounting methods to measure FP (return on assets =ROA, return on equity = ROE, return on sales = ROS, growth rate of sales = GRS) others studies used market measures such as market to book ratio (MTB) or stock returns (SR) and,  net profit margin (NPM). (E.g. Waddock and Graves, 1997; Chen and Yang, 2011; Platonova et al., 2016; El Moslemany and Etab, 2017 and others). In line with Tsoutsoura (2004) in El Moslemany and Etab (2017), we applied ROA, ROE and, ROS as the measure of FP.

 

BASIC MODEL

We formulate our basic models as follows to test the hypothesis:

We propose the equation (2) and equation (3) for model 1 and, for model 2 respectively:

FP = λ0 + λ1 CSR_Dind + λ2 LNSizei,t + λ3Cap.Ri,t + λ4Loan_Ri,t + λ5Debit_Ri,t + εi,t

(2)

FP = λ0 + λ1 CSR_Dis + λ2 LNSizei,t + λ3Cap.Ri,t + λ4Loan_Ri,t + λ5Debit_Ri,t + εi,t

(3)

VARIABLE EXPLANATION

DEPENDENT VARIABLE

Based on the objectives of our study, use widely measures of profitability namely ROA, ROE and ROS (Tsoutsoura, 2004 in El Moslemany and Etab, 2017 and others). Because we believe that, some of the banks here have not been listed on stock exchanges, we ignore the measure of market-return in our study.

INDEPENDENT VARIABLE

CSR_Dind denote CSR activities index, comes from equation (1) and CSR_Dis represents the overall individual category of CSR activities (environmental protection, customer and products, human resource and local community). To explore how each dimension interacts with financial performance (Branco and Rodrigues, 2006 and others).

CONTROL VARIABLES

Reference to previous research we include some control variables that could affect CSR activity and financial performance (bank size, capital ratio, Loan ratio and, Debt ratio) are defined as follows: First, we control Bank size (LNSize) measured by the natural logarithm of total assets. Size has widely been recognized as a proxy for profitability. The larger size facilitate bank to attract cheaper capital, therefore, the bank will have more recourse to invest in CSR activities and, consequently the larger bank makes additional contributions to the society than the small banks (El Moslemany and Etab, 2017; Platonova et al., 2016; Wu and Shen, 2013 and others). The expectation is a positive and significant association between bank size and FP. Second, we control another proxy of profitability, the capital ratio (Cap.R) measure as equity capital divided by total assets. It’s an indicator of bank invisible risk of default that designates the bank’s ability to grow under the present capital structure (Siueia & Wang, 2017). Following Patonova et al. (2016), highest capitalized banks tend to involve aggressively in CSR behavior, we predict that the relationship between CSR and FP will be positive. Third, in the same line; we included Loan ratio (Loan_R) considered as total loans divided by total assets to control the earning power of the bank (Platonona et al., 2016). This variable is widely used as an indicator of bank credit risk also; symbolize one of the main sources of the revenue of banks. In agreement, whith Siueia & Wang (2017), bank managers can increase provision when they presumed that loan will decrease and affect negatively the profit. The expectation is a positive and significant association with FP and CSR, respectively. Debt reduces the profitability at the same time; debt is also a commonly tested factor to examine the influential elements of CSR policies (Lev et al., 2010). Debt ratio (Debt_ R), indicates the long-term debt divided by total assets for bank i at moment t. It’s expected that the coefficient on debt ratio variable to be negative through the FP and CSR. Finally, λ0 indicates the intercept; λ1 … λn designates the regression coefficients; εi,t denote the error term.

RESULTS AND DISCUSSION

STATISTICS AND CORRELATION MATRIX

Table 3 presents the descriptive statistics of the variables of interest for the object of the study. In terms of the variable CSR_Dind that represent the voluntary CSR disclosure index, we can observe that the standard deviation for Mozambique banks is higher (0.1916 VS 0.0928) than the standard deviation for banks in RSA. This implies that there is a huge disparity among Mozambique banks regarding their attitude to Voluntary disclosure of their CSR activities. However, more than 16% of RSA banks have reported voluntary CSR activities in their annual report or their banking website compared to 6% of Mozambique`s banks. This indicates that banks are disclosing lowest level of CSR activities.

In terms of the performance variable, Table 3 show the highest mean value for ROA is around 56%, for ROE is 62% and for ROS is around 58% for RSA banks, and the lowest mean for Mozambican banks for ROA is 0.4561 with a standard deviation of 0.2196, respectively. However, we underscore the period understudy and indicate that banks did not recorded losses according to the usually positive outcome of the ratio of FP.

In terms our control variables, we can observe that the RSA banks had the highest mean value for Bank size around 7.0042 with a standard deviation of 0.1785. For bank capital categories, variable Cap.­R have, a mean equals 0.2255 compared to mean equals 0.1926 for Mozambican banks. Additionally, the highest Loan ratio average is around 20% and the standard deviation equals 0.0031 for RSA banks than the mean 0.0294 from banks of Mozambique. Finally, the debt ratio ranging from 0.0401 to 0.0779 and the average is around 5% for RSA banks compared to mean equals to 0.0524 for Mozambican banks.

A Pearson correlation matrix of the main variables used in our research is revealed in Table 4. We drew attention to the correlation matrix between the CSR disclosure index and each FP variables are statically positive and significant. This result suggest that the banking sector is socially responsible. Moreover, These findings are consistent with the studies of Wu and Shen, 2013; Mallin et al., 2014; Han et el., 2016; Platonova et al., 2016, and, among others.

(Insert around here Table 3 and Table 4)

The Table 5, the first 3 columns; display the estimations for equation (2) using ROA. The following 3 columns displays estimations for equation (2) using ROE. Additional the last 3 columns displays estimations for equation (2) using ROS. We used ROE and ROS to check the robustness of our model. Regarding ROA we can observe that the RSA banks are slightly over performing Mozambican banks. However, CSR indexes for both samples are relatively low (around 6%).

Regarding our four control variables, it can be observed that all are statistically significant (at least 0.1, 0.05 and 0.01 respectively) and positive as predicted, debt ratio variable is negative implying that our models are consistent. Jointly the result suggests the presence of voluntary CSR practice in FP behavior.

Our model overall is statically significant (P-value <0.05) and the highest Adj. R-squared equal 61.7% for RSA banks than 51.8% for Mozambican banks; indicating the strength and explanatory power of our model. This result, also displayed the maximum VIF – variance inflation factor as 2.439 for RSA than 2.422 for banks in Mozambique, suggesting that the multi-collinearity does not appear to be a problem. Similarly, With respect to the autocorrelation problem, we performed the Durbin-Watson test and maximum value was 2.121, so we rejected the residual autocorrelation hypothesis. On the other hand, we also solve the endogeneity problem by performing the Hausman test (Hausman and Taylor, 1981) for all models.

Similarly, regarding the robustness test (regression in model 1) using ROE and ROS. As presented in Table 5, the last six columns, is similar to the first model (ROA) and both models are significant at least at Prob. F < 0.1 and robust to expect the effect of the variable. Particularly, the CSR indexes for both samples are positive and significant (λ1 = 0.0431, P-value <0.05 and t-stat = 3.728), for the ROE of banks of Mozambique than (λ1 = 0.0533, P-value <0.05 and t-stat = 3.422) for RSA banks. For ROS variables also we observed a positive and significant coefficient λ1 on CSR_Dind variable (λ1 = 0.0604, P-value <0.05 and t-stat = 4.255, respectively) for RSA banks than λ1 = 0.0711, P-value <0.05, and t-stat = 4.127for Mozambican banks.

Regarding our control variables, it was observed that all were statistically significant (at least at 0.01, 0.05 and 0.1 respectively) and positive as predicted, however, debt ratio variable is negative suggesting that our models are consistent. Additional, the Adj. R-squared is equal to 0.631 means that our ROE model explains about 63% of the behavior of the independent variable and, the remaining 27% could be explained by unknown factors, for RSA banks. For Mozambican banks, the Adj. R-squared explains only about 52% of the behavior of the independent variable for ROE model and, the remaining 48% could be unexplained factors. In term of ROS, Adj. R-squared is high in RSA banks than Mozambican banks (around 51% vs 40%). This outcome also, exhibited the highest value of Durbin-Watson test as 2.199 for ROE model and 1.846 for ROS model, suggesting rejections of the residual autocorrelation hypothesis. We also displayed the highest VIF as 2.418 for RSA banks than the lowest VIF as 2.375 for Mozambican banks. Therefore, both models did not show the multi-collinearity problems, the autocorrelation problem even the problems of heteroscedasticity because we approved the respective statistics test to solve it, as presented in Table 5.

The conclusions confirm the expectation based on the stakeholder’s theory, because for all three models (ROA, ROE and ROS) we found a positive and statistically significant effect of CSR activities on financial performance. On this evidence, our first hypothesis is accepted. Implying that the banking sector is voluntary socially responsible also with this bank behavior increases their financial performance. These findings are in line with Platonova et al. (2016); Wu and Shen (2013) international studies and opposes the findings of El Moslemany and Etab (2017); Bae, Kang, and Wang (2011) international studies who those found insignificant or, negative relationship between Corporate Social Responsibility and financial performance in the banking sector.

(Insert around here Table 5)

Regarding the regression in model 2 (examine the effect of each category of CSR activities on the FP). We can observe in Table 6, that each individual category of voluntary CSR activity is positive linked with FP and P-value is lesser than the standard 0.05. Suggesting that the banking sector use this for elements of social responsibility behavior as a tool to increase financial performance, except for environmental protection category (CSR_D1) it means that, the banking sector invest less in the environmental activities comparatively to another category of CSR activities, for Mozambican banks present in our research.

The insignificant effect of environmental protection activities ((λ1 = 0.0001, P-value > 0.1) on FP, maybe because Mozambique is less developed compared to RSA. Following previous study, the level of development of the country is fundamental to involve the managers in CSR activities. Particular, bank managers in less developed country, are not involved in pursuing environmental initiative as opposed to developed countries where managers emphasis their CSR efforts on environmental initiative and issues that contribute to increase development and sustainability of the community. These findings, also suggest that not all individual categories of voluntary CSR activity affect PF. Therefore, this result indicates a statistically insignificant link between the financial performance and all individual categories of CSR practices, suggesting insufficient evidence to support our hypothesis H2.

Regarding the control variables, it can be seen that all are statistically significant (at least at 0.01, 0.05 and 0.1 respectively) and positive as predicted, however, debt ratio variable is negative suggesting that our models are consistent. Additional, the Adj. R-squared is equal to 0.621 means that our ROA model explains about 62% of the behavior of the independent variable and, the remaining 28% could be explained by unknown factors for RSA banks. Moreover, for Mozambican banks, the Adj. R-squared explains only 52% of our model.

Regarding the autocorrelation, multi-collinearity, endogeneity and, heteroscedasticity, as presented in Table 6, does not appear to be a problem because we carried out the respective statistic test to solve it.

 (Insert around here Table 6)

This paper draw attention to the highest CSR score for the top10 banks in both country. According to the research results presented in Table 7, it can be concluded that the hypothesis H3.1 has not been supported. In other words, RSA banks present the highest CSR score on environmental activities (18.735%), the environmental category presents t-value = 8.765, P < 0.01 showing that RSA banks are relatively stronger commitment to environmental dimension and,implying that the RSA banks are disclosing firstly information relative to their environmental initiative however, they did not ignoring other categories of voluntary CSR activities. Additional, for banks of Mozambique the maximum CSR score, it can be seen on customer and products category (16.788%),the customer and products dimension present t-value = 5.524, P < 0.01 showing that Mozambican banks are relatively stronger commitment to customer and products category and,implying that the banks of Mozambique are disclosing primarily information relative to their customer and products than other three categories of CSR activities, so we have evidence to support our hypothesis H3.2.

Environmental category, the Mozambican banks demonstrate weak commitment. Their total five-year score was 3 out of a possible 30 points.  Their only disclosed environmental information in the year of 2016. Thus, the banks of Mozambique should considerably improve the environmental protection category. On the other hand, the RSA banks demonstrate strong growth rate from a lowest 33.33% starting point adopted in 2012 to 86.67% by 2016, suggesting a progressive increase on CSR activities disclose. We find evidence that RSA banks are involved in environmental policy, green programs, recycling activities practices, sustainability, and conservations of energy since 2012. In contrast Mozambican banks failed in environmental category, only two banks reported environmental issue in 2016.

There are 6 events under the human resource element that reveal a bank`s social commitment to its workforce. Over the period of our analysis, both countries revealed growth in their human resource activities scores. Similarly, the RSA banks had the highest number of CSR category disclosed over the period, achieving an annual score of 23 out of a possible 30 points compared to 20 points out of 30 points possible in Mozambican banks sub-sample.

While, when it comes to customer and product category disclosing, the Mozambican banks are significantly ahead of RSA banks. Their total five‐year score was 69 out of a possible 125 points, compared to 65 points for RSA banks. There are 6 events under customer and product element that reveal a bank`s social commitment to sustainability. We found the Mozambican banks are largely supporting women branch and student branch (more than 50%) and RSA banks largely supporting the customer satisfaction and product quality (more than 50% for each).

Likewise, in community involvement category that reveals a bank`s s social commitment to supporting their local community. The findings reflects that Mozambican banks are largely creating job opportunities and supporting culture and sport (35% and 28%, respectively) and, the RSA banks a disclosing mostly information regarding public health care and housing policy (32% and 37%, respectively).

Over all the RSA banks had the highest number of CSR category disclosed over the period, achieving an annual score of 26 out of a possible 30 points compared to 20 points out of 30 points possible in Mozambican banks.

On the other hand, the results also conclude that, Mozambican banks pay little attention to reporting their CSR activities than RSA banks. This suggest that, there is significant difference in the CSR activities taken up by the African banking sector. These findings can be explained by different level of development of this two countries understudied (RSA GDP is high than Mozambique GDP). Following previous study the development level of a country is essential to determine the CSR behavior.

(Insert around here Table 7)

 

INTRUMENTAL VARIABLES (2SLS) REGRESSION ANALYSIS

Followingprevious studies (e.g. Mallin et al., 2014; Platonova et al., 2016), to avoid endogeneity problem we carried our 2SLS regression to validate the interpretation of the results.The findings shows that the coefficient of CSR_Dind still positive and significant at the conventional level (λ= 0.0518, P < 5%), even after accounting for endogeneity. This findings implies that endogeneity cannot account for the positive and significant link between CSR activity and financial performance. Similarly, we carried out the F-test cross the three models is 33.76 and is above the conventional level (minimum of 10). Also, we rejected the null hypothesis (p < 0.000).

CONCLUSION

Based on the hypothesis, we investigated the relationship between CSR and FP, using the RSA banking sector and the Mozambican banking sector as samples, over five years (2012-2016).  Using ROA, ROE and ROS as our FP measure and, applying content analysis of annual reports to obtain quantitative data, and complementing this data, by searching the banks’ websites reports about CSR disclosure as a measure of CSR, our key findings can be summarized in these points. Firstly, the coefficient λ1 that captured the association between FP and CSR activities was statistically significant. Consistent with the prior study of Platonova et al., 2016, and others, suggesting that bank managers are involved in CSR activities to improve FP. Additionally, we concluded that Mozambican banks pay little attention to reporting their CSR activities compared to the RSA banks.

Secondly, we examined the effect of each category of voluntary CSR activities on the financial performance, the conclusions were no significant for environmental protection category of CSR activities for Mozambican banks, however, and for the other three categories of CSR activities, and the results were significant and positive. Consequently, the finding was inconsistent with our hypothesis H2.

Furthermore, we expected but did not provide evidence that the RSA banks primarily disclose information about human resource behavior comparatively to other categories of CSR activities in our research. On the other hand, we expected and provided evidence that the Mozambican banks principally reveal information about their customer and products comparatively to other dimension of CSR activities. In general, our results confirm other previous studies in the field providing evidence that the CSR activities are associated to financial performance.

This study is subject to at least three limitations: first, the fact that the study concentrates only a small sample size (the top 10 ranked banks in each country) may be, if the sample were composed of all banks, the result could be more interesting. Second, the fact that the research covered only 5 years of data for analysis, may be if the time period were extended, we might have a different conclusion. Finally, the study was based on self-collected information from the annual reports and bank websites; perhaps if the CSR dada were collected from other source the result could be different.

Future research could consider the expansion of the period of analysis. In practical terms, we believe that the voluntary report on CSR activities could help the banking sector to improve FP. In the scientific scope, we believe that this study makes a great contribution filling the gap in the literature, particularly in the African context.

 

REFERENCE

Akin, A. and Yilmaz, I. (2016). Drivers of Corporate Social Responsibility Disclosures: Evidence from Turkish Banking Sector. Procedia Economics and Finance, 38, 2 – 7. Available online at www.sciencedirect.com

Akinpelu, Y. A.; Ogunbi, O. J.; Olaniran, Y. A. and Ogunseye, T. O. (2013). Corporate Social Responsibility activities disclosure by Commercial Banks in Nigeria. European Journal of Business and Management, 5, 173-185.

Bae, K. H.; Kang, J. K. and Wang, J. (2011). Employee treatment and firm leverage: A test of the stakeholder theory of capital structure. Journal of Financial Economics, 100, 130-153.

Branco, M. C., & Rodrigues, L. L. (2006). Communication of Corporate Social Responsibility by Portuguese banks: A legitimacy theory perspective. Corporate Communications: An International Journal, 11(3), 232-248. http://dx.doi.org/10.1108/13563280610680821

Carroll, A. B. (1999). Corporate Social Responsibility evolution of a definitional construct. Business & Society, 38(3), 268-295. http://dx.doi.org/10.1177/000765039903800303

Chen, H. and Wang, X. (2011). Corporate Social Responsibility and Corporate Financial Performance in China: an empirical research from Chinese firms. Corporate Governance: The International Journal of Business in Society, 11(4), 361-370.

El Moslemany, R. and Etab, M. (2017). The effect of corporate social responsibility disclosures on financial performance in the banking industry: empirical study on Egyptian banking sector. International Journal of Business and Economic Development, 5(1), 20 – 34.

Ershad, A. and Rahman, S.M.M. (2015). Corporate Social Responsibility disclosure: A comparative study between Islamic banks and conventional banks in Bangladesh. International Business and Management, 10(1), 9-17.

Freedman, M. & Jaggi, B. (1986). An analysis of the impact of corporate pollution disclosures included in annual financial statements on investment decisions. Advances in Public Interest Accounting, 1(2), 193-212.

Galant, A. & Cadez, S. (2017). Corporate Social Responsibility and Financial Performance relationship: a review of measurement approaches. Economic Research-Ekonomska Istraživanja, 30(1), 676-693, DOI: 10.1080/1331677X.2017.1313122 https://doi.org/10.1080/1331677X.2017.1313122

Goss, A., Gordon S., & Roberts, G. S. (2011). The Impact of Corporate Social Responsibility on the Cost of Bank Loans. Journal of Banking & Finance, 35(7), 1794-1810. http://dx.doi.org/10.1016/j.jbankfin.2010.12.002

Han, J. J.; Kim, H. J. and Yu, J. (2016). Empirical study on relationship between corporate social responsibility and financial performance in Korea. Asian Journal of Sustainability and Social Responsibility, 1, 61-76 https://doi.org/10.1186/s41180-016-0002-3

Hausman, J. A. and Taylor, W. E. (1981). Panel data and unobservable individual effects. Econometrica, 49(6), 1377-1399.

Jain, A.; Keneley, M. and Thomson, D. (2015). Voluntary CSR disclosure works: Evidence from Asia-Pacific banks. Social Responsibility Journal, 11(1), 2-18.

Kilic, M. (2016). Online corporate social responsibility (CSR) disclosure in the banking industry: Evidence from Turkey. International Journal of Bank Marketing, 34(4): 550–569.

http://dx.doi:10.1108/IJBM-04-2015-0060

Lev, B.; Petrovits, C. and Radhakrishnan, S. (2010). Is doing good for you? Yes, charitable contributions enhance revenue growth. Strategic Management Journal, 31(2): 182-200.

McDonald, L., & Rundle-Thiele, S. (2008). Corporate social responsibility a bank customer satisfaction. International Journal of Bank Marketing, 26(3), 170-182. http://dx.doi.org/10.1108/02652320810864643

Mallin, C.; Farag, H. and Ow-Yong, K. (2014). Corporate Social Responsibility and Financial Performance in Islamic banks. Journal of Economic Behavior & Organization, 103, 21–38.

Maqbool, S. and M. N. Zameer (2018). Corporate social responsibility and financial performance: An empirical analysis of Indian banks. Future Business Journal, 3(1), 84-93. https://doi.org/10.1016/j.fbj.2017.12.002

Margolis, J.D. and Walsh, J.P. (2003). Misery loves companies: Rethinking Social Initiatives by business. Administrative Science Quarterly, 48(2), 268-305.

Mensah, H. K. and Nuertey, D. (2017). The effect of corporate social responsibility on organizational commitment of employees of rural and community banks in Ghana. Cogent Business and Management, 4, 1-19.

Mulyadi, M. S. & Anwar, Y. (2012). Impact of Corporate Social Responsibility toward Firm Value and Profitability. The Business Review, Cambridge, 19(2), 316 – 323.

Pérez, A., & Del Bosque, I. R. (2012). The Role of CSR in the Corporate Identity of Banking Service Providers. Journal of Business Ethics, 108(2), 145-166. http://dx.doi.org/10.1007/s10551-011-1067-7

Platonova, E.; Asutay, M.; Dixon, R. and Mohammad, S. (2016). The Impact of Corporate Social Responsibility Disclosure on Financial Performance: Evidence from the GCC Islamic Banking Sector. Journal of Business Ethics, http://dx.doi.org/10.1007/s10551-016-3229-0

Simpson, W. G., & Kohers, T. (2002). The link between Corporate Social and Financial Performance: Evidence from the banking industry. Journal of Business Ethics, 35, 97–109.

Siueia, Tito Tomas & Wang, Jianling (2017). Loan Loss Provisions, income smooth, Signaling, Capital Management and Pro-Cyclicality: Empirical evidence from Mozambique’s Commercial Banks. International Journal of Economics and Finance, 9(11), 48-63.

https://doi.org/10.5539/ijefv9n11p48

Tijani, A. A.; Adeoye, A. O. and laka, N. S. (2017). Effects of Corporate Social Responsibility on Banks Financial Performance in Nigeria: A Study of United Bank of Africa. International Journal of Business, Economics and Management, 4(6), 136-147 https://EconPapers.repec.org/RePEc:pkp:ijobem:2017:p:136-147

Van Staden, C. J. & Hooks, J. (2007). A comprehensive comparison of corporate environmental reporting and responsiveness. British Accounting Review, 39(3): 197-210.

Waddock, S.A. and Graves, S. B. (1997). The Corporate Social Performance – Financial Performance link. Strategic Management Journal, 18(4), 303-319.

Wu, M. W. & Shen, C. H. (2013). Corporate Social Responsibility in the banking industry: Motives and Financial Performance. Journal of Banking & Finance, 37(9), 3529-3547. http://dx.doi.org/10.1016/j.jbankfin.2013.04.023

Yeung, S. (2011). The Role of Banks in Corporate Social Responsibility. Journal of Applied Economics and Business Research, 1(2), 103-115. http://www.aebrjournal.org/uploads/6/6/2/2/

6622240/4.shirley.pdf

Websites

  1. http://ww.bancomoc.mz Bank of Mozambique. Last accessed 30th December 2017.
  2. http://www.bvm.co.mz Mozambique Stock Exchange. Last accessed 28th December 2017.
  3. https://www.resbank.co.za/ South African Reserve Bank. Last accessed 30th December 2017.
  4. https://www.jse.co.za/ Johannesburg Stock Exchange. Last accessed 28th December 2017.

Tables

 

Table 1: Top 10 Banks by asset

Moz Banks RSA banks
BancABC Absa Bank,
Banco Terra African Bank
Banco Unico Barclays Africa Group
Barclays Bank Bidvest
BCI Capitec Bank
BNI First Rand Bank
Ecobank Grindrod
FNB Investec
Millennium BIM Nedbank Limited
Standard Bank Standard Bank

 

 

Table 2: Corporate Social Responsibility (CSR) disclosure category and sub-category

Environment Protection Human Resource Customer and Product Community involvement
Environmental policy Recruitment policies Product quality Creating job opportunities
Environmental management Student recruitment/Training Introduced new product Support for education
Lending and investment policy Employee evaluation/training Customer satisfaction Support for arts, culture and sport
Sustainability Employee health & safety Women branch Support for public health care
Recycling activities practices Employee recreation & sport Student branch housing policy
Conservation Resources/energy Employment of women

Table 3: Statistics

Variable Country
Moz RSA
Mean Max Min Std. Dev. Mean Max Min Std. Dev.
CSR_Dind 0.0593 0.1682 0.0067 0.1916 0.1569 0.2087 0.1338 0.0928
ROA 0.4561 0.7204 0.3065 0.2196 0.5649 0.7183 0.3942 0.1644
ROE 0.5372 0.6998 0.2978 0.1917 0.6233 0.7515 0.3106 0.1729
ROS 0.4897 0.5066 0.2605 0.2012 0.5809 0.6916 0.3831 0.6823
Size 4.0631 5.9984 2.0058 0.1846 7.0042 8.0337 5.9824 0.1785
Cap.R 0.1926 0.4207 0.1015 0.0998 0.2255 0.5238 0.1997 0.1552
Loan_R 0.0294 0.0478 0.0099 0.0991 0.1983 0.2126 0.1372 0.2011
Debit_R 0.0524 0.0608 0.0076 0.1001 0.0508 0.0779 0.0401 0.1985
Obs 50 50

 

Table 4: Correlation matrix of main variables

Variable CSR_Dind ROA ROE ROS Size Cap.R Loan_R Debit_R
CSR_Dind 1.0000
ROA 0.3855 1.0000
ROE 0.3417 0.4013 1.0000
ROS 0.3313 0.3011 0.1023 1.0000
Size 0.4131 0.2123 0.3101 0.2281 1.0000
Cap.R -0.3342 0.3272 0.3153 0.2977 -0.3231 1.0000
Loan_R -0.3341 -0.2251 0.2691 0.2981 -0.3218 0.3951 1.0000
Debit_R -0.3019 -0.2411 0.2178 0.3384 0.4005 0.0131 -0.3227 1.0000

Table 5: Result of Model 1 (FP and CSR index)

  ROA ROE ROS
Country Moz RSA Moz RSA Moz RSA
Variables Coeff. t-Stat. Coeff. t-Stat. Coeff. t-Stat. Coeff. t-Stat. Coeff. t-Stat. Coeff. t-Stat.
Intercept -0.045   -4.054** -0.0313   -3.296** -0.452   -4.071** -0.1578   -3.447** -0.0345   -2.377* -0.03   -2.350*
CSR_Dind 0.0627  4.917** 0.0749  4.215** 0.0431  3.728** 0.0533  3.442** 0.0604  4.255** 0.0711  4.127**
Size  0.0407  5.981***  0.0439  5.280***  0.3372 4.818***  0.3894 4.613*** 0.0437 5.243*** 0.0484 5.994***
Cap.R  0.0564  4.379**  0.0752  4.138** 0.0426   3.902** 0.0430   3.800** 0.0762   4.219** 0.0811   4.319**
Loan_R  2.0947  3.037*  2.2635  3.104* 1.9975 2.874* 1.8761  2.110* 3.0634  3.044** 3.1235  3.0453**
Debit_R -0.019  -1.063* -0.014  -1.031* -0.205  -1.083* -0.1826  -1.053* -0.0301  -1.076*** -0.0282  -1.053**
R-square 0.628 0.731 0.631 0.742 0.508 0.629
Adj_R square 0.518 0.617 0.524 0.623 0.403 0.517
F-Statistic 4.964 5.003 5.175 5.002 5.631 5.469
Prob. (F-Statistic) 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
Durbin-Watson Statistic 2.121 2.118 2.075 2.199 1.774 1.846
Max VIF 2.422 2.439 2.397 2.418 2.375 2.402
Obs. 50 50 50 50 50 50

*** P<0.01, ** P<0.05, * P<0.1

Table 6: Result of Model 2 (ROA and each CSR category)

CSR_D1 CSR_D2 CSR_D3 CSR_D4
Variables Moz RSA Moz RSA Moz RSA Moz RSA
Coeffi. t-stat. Coeffi. t-stat. Coeffi. t-stat. Coeffi. t-stat. Coeffi. t-stat. Coeffi. t-stat. Coeffi. t-stat. Coeffi. t-stat.
Intercept -0.0693   -4.496** -0.0557   -4.744** -0.067   -4.491** -0.0553   -4.744** -0.0577   -4.319** -0.0541   -4.028** -0.0531   -4.015** -0.055   -3.992**
CSR_D1 0.0001 1.997 0.6778  4.012**
CSR_D2 0.4628  4.201** 0.5335  4.028**
CSR_D3 0.6268 4.005** 0.5262 4.022**
CSR_D4 0.0535 3.913* 0.6766 4.011*
Size  0.0406  4.572***  0.0428  5.314***  0.0406 4.156***  0.0428 5.044***  0.0406 5.457***  0.0428 5.023*** 0.0405 4.313* 0.0431 5.151*
Cap.R  0.0558  4.169**  0.0697  4.013**  0.0558   4.168**  0.0697   4.013**  0.0558   4.168**  0.0697   4.014** 0.0556   4.169** 0.0698   4.013**
Loan_R  2.0094  3.186*  2.0679  3.192*  2.0094  3.353**  2.0679  2.123**  2.0096  3.353**  2.0678  2.137** 2.0093  3.355** 2.0599  2.125**
Debit_R -0.033  -1.073* -0.024  -1.068* -0.033  -1.073* -0.024  -1.068* -0.033  -1.074* -0.024  -1.069* -0.033  -1.073* -0.024  -1.068*
R-square 0.644 0.751 0.651 0.732 0.561 0.599 0.562 0.601
Adj_R square 0.521 0.619 0.532 0.631 0.442 0.501 0.451 0.502
F-Statistic 5.004 5.271 6.005 5.997 5.924 5.326 5.886 5.973
Prob. (F-Statistic) 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
Durbin-Watson Statistic 1.914 1.896 1.881 1.936 1.925 1.903 1.921 1.922
Max VIF 2.241 2.24 2.401 2.405 2.411 2.397 2.226 2.301
Obs. 50 50 50 50 50 50 50 50

Table 7: Total CSR categories score by Country

Variables Moz RSA
Year CSR_D1 CSR_D2 CSR_D3 CSR_D4 CSR_D1 CSR_D2 CSR_D3 CSR_D4
2012 0 6 6 1 10 10 10 5
2013 0 6 11 3 13 8 10 7
2014 0 7 14 8 13 9 14 10
2015 0 10 18 8 15 11 13 13
2016 3 20 20 12 26 23 18 20
Total (2012-2016) 3 49 69 32 77 61 65 55
t-value 1.000 3.694 5.524 3.258 8.765 4.443 5.564 4.188
Sig. (2-tailed)


Recommendation
EssayHub’s Community of Professional Tutors & Editors
Tutoring Service, EssayHub
Professional Essay Writers for Hire
Essay Writing Service, EssayPro
Professional Custom
Professional Custom Essay Writing Services
In need of qualified essay help online or professional assistance with your research paper?
Browsing the web for a reliable custom writing service to give you a hand with college assignment?
Out of time and require quick and moreover effective support with your term paper or dissertation?
Did you find someone who can help?

Fast, Quality and Secure Essay Writing Help 24/7!