Corporate social responsibility and financial performance: Fact or fiction? A look at Ghanaian banks

Northumbria University has developed Northumbria Research Link (NRL) to enable users to access the University’s research output. Copyright © and moral rights for items on NRL are retained by the individual author(s) and/or other copyright owners. Single copies of full items can be reproduced, displayed or performed, and given to third parties in any format or medium for personal research or study, educational, or not-for-profit purposes without prior permission or charge, provided the authors, title and full bibliographic details are given, as well as a hyperlink and/or URL to the original metadata page. The content must not be changed in any way. Full items must not be sold commercially in any format or medium without formal permission of the copyright holder. The full policy is available online: http://nrl.northumbria.ac.uk/pol  i cies.html


Introduction
In the last few years, Corporate Social Responsibility (CSR) as a concept has captured the attention of the majority of management scholars, with studies of Corporate Citizenship, Ethics and Social Responsibility appearing with greater frequency.CSR as a term is often used interchangeably in these studies with such concepts as corporate responsibility, corporate citizenship, social enterprise, sustainability, sustainable development, triple-bottom line, corporate ethics and, in some cases, corporate governance (Bassen, Hölz & Schlange 2006).These interchangeable terminologies have influenced the various ways in which actors understand and define CSR.Consequently, varied definitions of the concept have been put forward by various observers (governments, companies, business associations, business consultants, non-governmental organisations, shareholders, employees, consumers and communities) in an effort to endorse, encourage or criticise its practical implications.These definitions are based on the different values and expectations that each of these stakeholders bring to their relationship with corporations (Moon 2007).The abundance of different definitions of CSR has increased confusion regarding the concept (Margolis & Walsh 2003).In the final report to the International Standards Organization's Committee on Consumer Policy (ISO/COPOLCO) on Desirability and Feasibility of ISO CSR Standards, it was noted that 'corporate social responsibility' or 'social accountability' are approximately equivalent terms, however, 'corporate responsibility' is the most inclusive concept for reflecting the focus on a firm's triple bottom line as well as a firm's social engagement and interaction with stakeholders in society for economic, social and environmental purposes (ISO Working Group 2002).
Notwithstanding the foregoing, the proliferation of definitions reveals the widespread attention being paid to CSR, not only by organisations, but also governments, international institutions and other stakeholders.Some of these actors have further developed structured principles, standards and frameworks by which CSR activities can be governed (Branco & Rodrigues 2006;Kashyap, Rajan & Stein 2008).Moreover, several indices have been developed by CSR rating institutions, such as the Dow Jones Sustainability Index (DJSI), the Financial Times Stock Exchange's FTSE4Good, the Fortune 500 and the Johannesburg Securities Exchange (JSE) index.These indices rate firms based on various criteria including, but not limited to, human rights, environmental protection, worker health and safety, labour standards, marketing, accountability and supply-chain management.
The applicability of these standards, principles and indices in different geographical, cultural, political, economic and environmental conditions has been questioned (Halme, Roome & Dobers 2009).This is down to the fact that much of the research and extant literature on CSR have been focused on the developed world, where institutions are effective and efficient, whilst limited research attention on the subject has been observed in developing countries and regions (Arli & Lasmono 2010).Blowfield and Frynas (2005) therefore call for, as a necessity, a critical CSR research agenda focusing on developing countries.The African region in particular is the most under-researched and the region's uptake on CSR in comparison with Europe is very low.Visser (2005), for instance, has revealed that only 12 of Africa's 53 countries have had any research published in core CSR journals, with 57% of those articles focused on South Africa and 16% on Nigeria.Thus the applicability of the standards, principles and practices designed in the developed world is questionable in an African setting.In this vein, Muthuri and Gilbert (2011) advocate the development of an 'Africanised' CSR agenda.
Moreover, of the limited published papers on CSR in the region, many have thrown more light on philanthropic, ethical and corporate governance issues (Abor 2007;Kyereboah-Coleman & Biekpe 2007;Ofori 2007a), whilst others have examined wholesale CSR typology issues (Ofori 2006(Ofori , 2007b;;Ofori & Hinson 2007), leaving other important aspects completely unexplored.For instance, no existing study has explored empirically the impact of CSR practices on companies' financial performance (return on assets [ROA] and return on equity [ROE]).This is an important issue worth exploration considering the generalised argument that CSR benefits not only stakeholders but also the organisations themselves.The aim of this article is thus to cover this lacuna in the CSR research agenda in the region, by examining empirically (1) bank managers' and CSR officers' views on CSR, (2) their motives for engaging in CSR practices and (3) the impact of the CSR activities on their financial performance (ROA, ROE).This was done using the Ghanaian banking sector as the empirical setting.The banking industry was selected for the study because, as profit-orientated businesses, executives of banks have a responsibility to maximise profits.Moreover, banks are also faced with the added responsibility of fulfilling the increasing demands of diverse and complex stakeholder groups.Another reason for selecting the banking sector is that the Ghana Banking Survey (2007) report raised questions about the motives of Ghanaian banks in developing and implementing CSR practices.This study therefore explores the views of Ghanaian banks on CSR practices, the motives behind their CSR activities and the relationship between CSR practices and their financial performance.
The next section reviews the existing literature, followed by the methodology and data analysis.The results are subsequently presented and discussed, leading to a discussion of their theoretical and practical implications.The article concludes with a recommendation regarding future research agendas.

Literature review
Traditionally, scholars have argued that the one and only social responsibility of a business is to use its resources and engage in activities designed to increase its profits (Friedman 1970;Henderson 2005;Lantos 2001).Proponents of this view chronicle that CSR is characterised by external interference, over-regulation and legislation, corporate irresponsibility, misguided virtues, unethical and misappropriation or unjustified appropriation of the owners' property, which amount to theft and violation of shareholders' rights (Friedman 1970;Halfon 1998;Lantos 2001).They further rationalise that CSR activities lower economic efficiency and profit; impose unequal cost amongst competitors; impose hidden costs passed on to stakeholders; and place responsibility on business rather than individuals (Friedman 1970;Jensen 2001;Lantos 2001).In this sense, firms are advised to focus on 'making profit' -the overriding goal and responsibility of businesses.
In recent times, however, in fast-developing CSR literature, many scholars and global institutions have rejected the notion that firms should focus all their efforts and resources on maximisation of profit (Carroll 1991;Crane, Matten & Spence 2008).The interdependence of business and society has been recognised by these scholars and global institutions who argue that it is a managerial obligation to take action to protect and improve both the welfare of society and the interest of their organisations (Cannon 1992;Davis & Blomstrom 1975;Lunden 1988).Accordingly, firms have other equally-important social responsibilities aside from profit maximisation.These additional activities are contained in a number of CSR models propounded by researchers that seek to capture the various responsibilities of firms.
For instance, Carroll (1991;2004) introduced four variables (economic, legal, ethical and philanthropic) in his model for assessing CSR practices, whilst Lantos's model (Lantos 2001) considers ethical, altruistic and strategic variables.Crane et al. (2008) also introduced a contemporary model which groups CSR activities into four components: the Marketplace, Workplace, Environment and Community.Additionally, Wood's CSR model (Wood 1991) looks at principles, processes and policies, whilst Meehan, Meehan and Richard's 3C-SR model (Meehan, Meehan & Richard 2006) measures a firm's CSR using three variables: consistency, commitment and connections.Galbreath (2009) also structures his framework within the strategic scope of a firm, including a firm's longterm objectives (mission), internal and external impact on a firm's mission (strategic issues), a firm's markets (markets), products and/or services needed (customer needs), available internal resources (assets, skills, competences, relationships, finance and facilities -resources) and a firm's performance within competitive markets (competitive advantage).
These models offer various angles from which the activities and performance of corporations are judged and viewed.The models thus underscore the important insight that corporations have other responsibilities to society and their stakeholders, which go beyond the narrow economic, technical and legal requirements.To meet this end, organisations are advised to identify those with interests in their activities (stakeholders) and to engage effectively with them in order for their activities to be legitimised by these stakeholders, allowing them to ensure their long time survival and profitability (Mitchell, Agle & Wood 1997).
These models also underscore the point that organisations view CSR activities differently.Whereas some may consider it to be a moral obligation to society with no returns expected, others may view it as a strategic activity that could enable them to gain legitimacy or protect their profits.Yet others may also view it as a moral stand in order to protect or sustain the environment.These differing perspectives may arguably underscore why companies undertake different types of CSR activities, with some focusing on issues in the community, whilst others focus on the environment.Additionally, others may focus on issues at work or in the marketplace, whilst others consider compliance with standards to be the best CSR activity.
Extant literature on CSR in Africa suggests that CSR activities are mainly philanthropic in nature and centre on community development (Baughn, Bodie & McIntosh 2007;Blowfield & Frynas 2005;Chapple & Moon 2005;Ofori 2007a;Ofori & Hinson 2007).Research by Gesellschaftfür Technische Zusammenarbeit (GTZ) (2009) research on CSR in six sub-Saharan African countries reveals that companies' CSR activities in the region are predominantly philanthropic.According to the study, companies mandate individuals from their marketing, finance, human resources, communications or corporate affairs departments to be responsible for all their CSR activities.The majority of these individuals were found to work within the marketing, communications and/or corporate affairs departments.The findings in that study suggest that CSR is perceived differently by firms in the region, with many of them viewing it as a strategic marketing, branding or public relations tool.Based on the above discussion, the following hypotheses have been formulated: • H1: Ghanaian banks view CSR activities as being a strategic activity.
There are different motives behind firms' CSR initiatives and practices.Findings from existing studies reveal that firms' motives for engaging in CSR activities tend to cluster around both strategic and moral justifications (Brønn & Vidaver-Cohen 2008).In their study of 500 Norwegian firms, Brønn and Vidaver-Cohen (2008) found legitimacy, profitability and sustainability reasons as the three main motives of firms' CSR activities.Accordingly, legitimacy and sustainability motives tend to dominate, followed by strategic profit-protecting motives.There is also the moral motive for CSR which is anchored on the idea that businesses have an ethical duty to give back to society.Studies have documented instances in which 'doing the right thing' appears to be a stronger motive for social initiative than the practical benefits generated by CSR activities (Berthoin 1992;Hahn & Scheermesser 2006).
For instance, Holmes' study of executives' attitude to social responsibility for the years 1970-1975 (Holmes 1976) in the USA revealed that 97% of the executives had the strongest response with regard to solving social problems, whether or not business contributed to creating those problems.Galaskiewicz and Colman (2006) also disclosed that many managers genuinely believe that businesses have a duty to improve local communities and create a better world for the future, which forms the motivation behind their CSR activities.
Many studies, however, found strategic reasons to be the main motive behind firms' CSR activities (Kotler & Lee 2005).Knox, Maklan and French (2005) reveal that in some firms, business outcomes can be linked to a firm's participation in social initiatives.Several other studies have shown that companies develop a social portfolio because managers believe these activities can build competitive advantage, provide new business opportunities, insulate firms from costly regulation, or help them meet shareholder demands (Campbell, Moore & Metzger 2002;Fombrun & Shanley 1990;Galaskiewicz & Colman 2006;Gardberg & Fombrun 2006).
From this perspective, CSR activities are considered strategic in nature, with the understanding that firms can do well in the long run by doing good (Vogel 2005).This is evidenced in the global move from single-bottom line (economic) to the triple-bottom line (economic, environmental and social) approach (Global Reporting Initiative 2002), at times also referred to as the three 'Ps': people, planet and profit (Elkington 1997).The significance of this strategic measure is that companies are not only audited according to their economic or profit impact but are also penalised according to their non-performance with regard to environmental and social impact.Subsequently, in some countries, share prices reflect positively the ethical dimensions of a company's operations where financial markets judge companies by their wider impact on society (Elkington 1997;Frankental 2001).This triple-bottom line introduces the demand for strategic focus and the inclusion of CSR as essential to all core management functions.Consequently, CSR is now perceived to be about satisfying the needs of both firms and stakeholders, that is, maximising profit whilst still meeting wider stakeholder demands.
In some industries, firms consciously invest extra effort and resources in order to appear socially inclined in the hope of differentiating themselves from less-responsible colleagues (Barnett 2007;King, Lenox & Barnett 2002).According to Utama (2008), a company is motivated to practise CSR if the company's stakeholders (consumers, investors and other stakeholders) reward it (for instance with higher product and/or share price) for this practice.A survey by MORI (2003) points out that 70% of consumers will pay a higher price for a product they think is ethically more superior, which will lead to profit maximisation.Profit-making motives thus revolve fundamentally around managerial beliefs that engaging in social initiatives can have a direct impact on profitability.Tudway and Pascal (2006) disclose that shareholder value can be maximised when companies pursue a visible social agenda.A study of Dutch managers showed that whilst respondents believe CSR can enhance reputation, strengthen employee commitment to the firm and improve overall profitability, they also express an equally-strong desire to make the world a more moral and better place (Graafland & Van de Ven 2006).This leads to the following hypothesis: • H2: The main motives of Ghanaian banks' CSR activities are profit maximisation and legitimacy.
CSR as a factor of profitability, according to Albinger and Freeman (2000), has the ability to motivate, attract and retain the desired workforce and improve financial performance.Dolan (1997) underscores this point by referring to Forbes reports which found that more than half of 2100 MBA (Master of Business Administration) students revealed that they would accept a job with a lower salary in a sociallyresponsible company.CSR thus enhances corporate image and reputation, which are also important drivers of cost reduction and overall financial performance of companies.
According to Roberts and Dowling (2002) there are emerging empirical findings of a strong and positive link between reputations from CSR and financial performance.Others such as Fombrun (1996), Peloza (2006) and Gardberg and Fombrun (2006) found that reputation is a critical intangible asset which provides a crucial link between social responsibilities and profitability by reducing transactional costs and increasing product demand.Ofori and Hinson (2007) inferred that the strategic innovation of CSR has the propensity for producing outcomes such as financial performance and organisational effectiveness.Hahn and Scheermesser (2006) also revealed that the practice of CSR can yield positive financial results, either by generating new revenue or by protecting existing profit levels.Coelho, McClure and Spry (2003:18) note that a firm's interactions with its stakeholders are recognised as having a direct influence on profitability and that ethical executives should consider this as being part of their fiduciary duties to shareholders.Pearce and Robinson (2009:28) explain further that overlooking the enduring concerns of society (customers, suppliers, creditors, ecologists and regulatory agents) may produce profit in the short term but, overtime, the financial consequences are likely to be detrimental.
Empirical studies examining the relationship between CSR activities and firms' financial performance have presented mixed findings.A greater number of these, however, show a positive correlation between CSR and firms' financial performance (Frooman 1997;Griffin & Mahon 1997;Roman, Hayibor & Agle 1999;Waddock & Graves 1997).Margolis and Walsh (2003) also examined studies published between 1971 and 2001 and found no relationship in 21 of these studies.However, 18 studies reported a positive relationship and seven found a negative relationship.Again, in a metaanalysis of 52 studies, Orlitzky, Schmidt and Rynes (2003) found an overall positive link between CSR and financial performance.Even though associations have been found, the strength of CSR contribution to firms' profitability has not been discussed.Thus we examine both the relationship and the strength of CSR contribution to banks' overall financial performance.Financial performance measures the results of a firm's CSR activities in monetary terms.According to Griffin and Mahon (1997), researchers tend to use principally accounting-based measures in examining the financial performance of firms.Some of these measures include ROE, ROA, profitability and earnings per share, amongst others.Following on from the findings of Gil-Estallo, Giner-de-la-Fuente and Griful-Miquela ( 2008), we also employ ROE and ROA in measuring firms' financial performance in this study and in assessing the association between CSR activities and financial performance.Our next two hypotheses are thus: • H3: There is a positive relationship between Ghanaian banks' CSR activities and their financial performance (ROA and ROE).• H4: Ghanaian banks' CSR activities significantly contribute to their profitability (ROA and ROE).

Research design
The design of this research and data collection were influenced by the need to obtain rich data in order to examine banks' views on their CSR activities in a meaningful and empirical manner, their motives for engaging in CSR, the association between CSR and their financial performance and, finally, the level of contribution of CSR to banks' profitability.A quantitative methodology was thus considered appropriate for exploration of these issues in this study.This involved collecting primary data using survey questionnaires from respondents.

Sample and sampling technique
The sample frame for the study was all banks licensed by the Bank of Ghana to operate commercial banking services in the country.As at May 2009, the Ghana Banking Survey reported that there were 25 such banks in Ghana.These banks make available their financial statements to the public and fulfil obligations to regulatory agencies such as the Environmental Protection Agency, Labour Commission, Ghana Stock Exchange and Bank of Ghana.Twenty-two out of 25 banks participated in the study, representing 89.3% participation.
Since CSR is normally handled at senior levels (Crane et al. 2008) and due to the strategic nature of the information sought, people in senior management positions, heads of departments or officials knowledgeable and responsible directly for issues relating to CSR were therefore targeted.A purposive sampling technique was employed to select respondents since the respondents were those officials directly involved in the formulation, implementation and evaluation of CSR activities in their firms.A background check was undertaken by the researchers on the banks to identify and get in touch with these respondents who were directly in charge of CSR practices.This was made possible because most of these banks were listed on the Ghana Stock Exchange.A total of 176 questionnaires (eight to each of the 22 banks) were finally distributed and at the end of the survey, 133 of them were returned, filled in properly and useable.This represents a very good response rate of 76%.

Research instrument
Data-collection technique employed was the questionnaire survey method.In collecting information on the independent variable CSR, the primary data source employed was a questionnaire consisting of closed-ended questions.Research questions were designed as concisely as possible in order to obtain maximum information on management views and motives for CSR practices.The items in the questionnaire were thus used to measure their views on CSR practices, as well as their motivation for CSR practices.Questions covering views influencing the practice of CSR sought to understand the views behind this practice.The rest of the questions sought to examine the banks' motivation for practising CSR.For these questions on banks' views on CSR and their motivation for CSR activities, we adopted a 5-point Likert scale adapted from Abdul and Ibrahim (2002) and Brønn and Vidaver-Cohen (2008) respectively.The scale ranged from '1 = strongly agree' to '5 = strongly disagree'.The respondents were thus asked to indicate the extent to which they agreed with statements on the two main issues (see Appendices 1 and 2 for the questions).
These scales were subjected to reliability analysis.The most widely-used reliability measure in research is the Cronbach's alpha which assesses the consistency of the entire scale and indicates how well the items correlate positively to one another.Cronbach's alpha ranges from 0 to 1, with 0 standing for a completely unreliable test, higher values close to 1 indicating higher internal reliability and 1 standing for a completely reliable test.Different positions on what constitutes an acceptable value of Cronbach's alpha have been observed.For example, Nunnally (1967) originally advocated a threshold value of 0.50 as acceptable whilst Pallant (2007: 95) later pointed out that a generally-acceptable Cronbach's alpha value is 0.70 and above.Hair et al. (2010:125) recently suggested that a threshold value of 0.60 was acceptable.
In this study, the result of the 7-item scale on the view of CSR has a Cronbach's alpha (α) of 0.702, whilst the 9-item scale on motivations has a Cronbach's alpha (α) of 0.760 (see Tables 1 and 2).Both results show acceptable levels of internal consistency and reliability (Hair et al. 2010;Nunnally 1967;Pallant 2007).The results are also consistent with that of Abdul and Ibrahim (2002) and Brønn and Vidaver-Cohen's (2008) studies on CSR, which yielded results of 0.6052 and 0.980 respectively.Additionally, prior to the main survey, the questionnaire was pretested with nine managers responsible for CSR activities in nine banks.Two academic scholars knowledgeable in CSR were also given the questionnaire in order to review and provide feedback.Feedback from the pilot study allowed further refinement of questions to enhance the clarity and comprehensibility of the instrument.The data on financial performance were obtained from the Ghana Banking Survey, Ghana Stock Exchange, Annual Reports and accounting records of the banks.

Panel study
As in previous studies, the performance indicators ROA and ROE were employed as the dependent variable (financial performance) of the study.Panel data were obtained and used in exploring CSR impact on financial performance.
Using panel technique is one of the ways to avoid dispersing information on cross-sectional units observed over time.The use of panel estimation techniques have the advantage of capturing a larger portion of the variability in the data and also reduce the importance of a key econometric problem that often arises in empirical studies, namely, the omitted variables that are correlated with explanatory variables.
According to Akhter (1993), pooling of the data increases the sample size and the efficiency of the parameters because it adjusts for any heteroskedasticity of error terms and autocorrelation in time series data.Furthermore, the pooled data on panel technique reduce the problems generated by multicollinearity amongst the explanatory variables in the model (Baltagi 1995:4;Kennedy 2003:302).Pooling data on a panel gives more information, more variability, more degrees of freedom and more efficiency in the parameter estimates produced.Therefore, a pooled time series and cross-section is recommended over individual cross-section and time series analysis.

Company profile
Further examination of the 133 fully-completed questionnaires returned at the end of the survey revealed that 76% of them were from managers and 24% from non-managers who were knowledgeable about their CSR activities.Of the 76% respondent managers, 8% were top managers, 24% senior managers and 44% middle managers (Figure 1).Job titles  and positions on an organisational chart provide information regarding relationships with other departments as well as core business functions (Pearce & Robinson 2009).It is essential that senior-to top management officials are placed in charge of CSR, since they influence directly decisions that can promote considerations for the broader interest and holistic view of CSR practices.

Data analysis
The obtained data from the questionnaire were entered into two analytical software packages: the Statistical Package for Social Sciences (SPSS, version 17) and STATA.To explore the banks' view on CSR activities as well as their motives, both descriptive and inferential statistics, including mean scores, standard deviation, and analysis of variance (ANOVA) analyses, were used.For a meaningful interpretation of the mean scores, the following intervals adopted from Gravetter and Wallnau (2009:278-302)  Again, for the examination of the relationship between CSR and banks' financial performance, both descriptive and inferential statistics, including Pearson correlation and multiple regressions (in addition to the methods mentioned above), were employed.In this regard, the mean scores of banks' CSR practices were aggregated into one independent CSR variable and, based on the scores, dummy variables of 1 or 0 were assigned for Strategic CSR or Non-strategic CSR respectively.The dependent variables ROA and ROE measured the financial performance of the banks.To observe effectively the association of CSR activities and banks' performance, other control variables were included in the regression model.These control variables include debt ratio (DR), origin (ORIG), size (SIZE) and growth (GROW), as shown below.
Regression equation: Where: The results of the analyses are presented and discussed below.

Ethical considerations
For ethical reasons, respondents were not coerced into filling the questionnaires; they did so on a free-will basis.Again, the data were not collected undercover, as permission was sought from management of the banks to enable the authors to administer questionnaires to their employees.The individual identity of respondents was not mentioned anywhere in the study.

Trustworthiness
The analyses and discussion of findings of the study are all based on the data collected, as well as other secondary information obtained from the banks.Authors were as objective as possible in reporting the findings.

Managers' view of corporate social responsibilityin Ghana
We used the mean score to examine banks' views on their CSR activities.Thus, in order of the strength of agreement, the view that CSR 'improves business image and reputation' ranks top, followed by 'promotes socioeconomic development', 'longterm profits for business', 'serve as a source of competitive advantage', 'enhance the implementation of core business activities', 'attract investors' and 'stimulates and develops customer demands respectively'.The standard deviation figures further suggest that respondents and, for that matter, banks have the most closely-related view regarding the fact that CSR improves business image, as it records the least standard deviation (0.46).On the other hand, the highest standard deviation, obtained for 'sustains and develops customer demands' (1.11), suggests that respondents have the most diverse views with regard to that issue.
However, the general agreement with all these statements reveals the potential benefits of CSR to firms and underscores the strategic nature of CSR activities.Additionally, these perspectives are captured by the contemporary view on CSR (Branco & Rodrigues 2006;Brønn & Vidaver-Cohen 2008;Crane et al. 2008;Quazi & O'Brien 2000).It is also consistent with the finding of Holmes (1976) in his study of executives' perception of CSR in the USA, that more than half of the managers strongly agreed that corporate social activities enhance corporate reputation and goodwill.Similarly, Abdul and Ibrahim's (2002) study reveals that about 70% of their respondents disagree with the view that 'business already has too much social power and should not engage in social activities that might make it more'.
Finally, the Least Significant Difference (LSD) test was used to find out whether the difference between the various views on CSR was significant statistically.The results indicate that there was a significant difference F (24.149) = 2.303, p = 0.001; < 0.01, at a 99% confidence level, between banks' views on their CSR practices.This implies that the responses to the contemporary view of CSR practices are not due to errors but are rather a true measure of differences in responses.Again, all of these reflect the strategic nature of CSR activities in contemporary times (Branco & Rodrigues 2006;Carroll 1979;1991;2004;Galbreath 2009;Kashyap et al. 2008;Meehan et al. 2006;Wood 1991).Thus, H1 is supported.

Motivation for practising corporate social responsibility
In Table 2, the main motives for banks to engage in CSR activities are scored according to the means.The results reveal that Legitimacy is the dominant motive behind banks' CSR activities as it has the highest mean (4.24) and the lowest standard deviation (0.47).This is followed by the Profit protectability motive (3.63) and, finally, the Sustainability motive (3.37).
Consequently, ANOVA analysis was used to examine whether significant differences exist in the mean scores of the legitimacy-, profit protectability-and sustainability motives.At a 95% confidence interval, there was no significant difference F (2.6) = 1.318, p = 0.335; < 0.05 observed for the legitimacy-, profit protectability-and sustainability motives.Thus, the results revealed that there is no significant difference amongst the motives governing CSR practices within the Ghanaian banking sector.This implies that banks in Ghana are motivated to practise CSR for legitimacy reasons as much as they are motivated by profit protecting and sustainability reasons.H2 is therefore supported.

Corporate social responsibility and financial performance
To examine the relationship between CSR and banks' profitability, we employed correlations and regression analyses.In our analysis, we first examined the presence of multicollinearity in our dataset, which is considered to exist when the variables correlate highly.Even though the rule of thumb is that Variance Inflation Factor (VIF) figures of more than 10 depict the presence of multicollinearity, Freund and Littell (2000) suggest that VIF figures below 5 indicate the absence of multicollinearity.TheVIF figures in our regression output (Table 3) confirm the absence of multicollinearity between variables in this study.
Table 3 presents the summary descriptive and correlation statistics for the dependent, independent and controlled variables used for the regression analysis.ROA registers a mean of 0.037 with a standard deviation of 0.240.This suggests that ROA for the banks over the period of the study averages about 3.7% with individual yearly figures differing around 24.0%.The high variations for ROA as indicated by the standard deviation are confirmed by the minimum and maximum values of -1.160 and 2.190 respectively.The mean score of ROE (0.252) is, however, much higher than that of the ROA (0.037) and the variation around this value is also  very high, as indicated by its standard deviation (0.712).The ROE mean value of 0.252 suggests that banks are able to give their shareholders about 25.2% return on their investments.The debt ratios of the banks over the period are very high (4.766) which suggest that an average bank has about 476.6% of its assets as total liabilities.This confirms Saunders and Cornett's (2008) assertion that banks are generally highlyleveraged (risky) firms.
Table 3 further reveals an important insight that growth in the banks during the period of the study was about 44.4% as indicated by the mean of 0.444.The aggregated transformed CSR mean (0.812) reflects our earlier finding that banks strongly regard their CSR activities as Strategic.
The correlation coefficients figures in Table 3 also reveal the relationship between CSR and banks' profitability.
The correlation coefficients indicate a direct and positive relationship between CSR and the two profitability measures (ROA and ROE), with both being significant at a 5% significance level.The positive relationship between CSR and the profitability measures (ROA and ROE) supports the general idea that the more a firm carries out CSR activities, the more profitable the firm will eventually become.In this case, the more the bank initiates, develops and implements CSR activities, the more the returns on its assets.A similar interpretation can be given to the positive correlation between ROE and the CSR.Thus, the more the firm is strategic in CSR practices, the more it increases its shareholders' value.This confirms H3 in that there is a positive relationship between CSR activities firms' profitability.
Apart from CSR, firm size too has a significantly positive relation with ROA and ROE.This suggests that the more banks expand their assets, the more profitable they become.Growth is also found to be related positively to the two profitability measures but is not significant at a 5% significance level.The results on size and growth are consistent with that of Abor (2007), who also discovered a positive relationship between growth and size and these indicators in his analysis of working-capital management and profitability, where these variables were used as control variables.However, the debt ratio registers mixed results.It is found to be related negatively to ROA but positively to ROE.Origin also registered a positive correlation for both ROA and ROE, but was not significant at a 5% significant level.
Finally, to support or reject H4 in full, we examined the unique contribution of CSR and each of the other four factors to firms' ROA and ROE.The resulting models as depicted in However, CSR activities do not seem to contribute significantly to ROA and ROE when compared with growth, size and, perhaps, debt ratio.Thus, even though the correlation results reveal a positive relationship between CSR and firms' profitability, the regression result depicts that the predictive ability of CSR activities with regard to banks' profitability is not very strong.Notwithstanding the foregoing, there may be reasons for this outcome.To start with, it is possible that the magnitude of CSR activities undertaken by banks in Ghana is very small compared with their size, growth and the amount of debt they have.With the positive association, however, it is debatable to suggest that the more CSR activities banks undertake, the greater the possibility of it contributing in the long run to their ROA and ROE.The results, however, reject H4, which posits that banks' CSR activities contribute significantly to their ROE and ROA (profitability).

Conclusion and implications
Corporate Social Responsibility practice within a strategic scope is integral to corporate strategy.An organisation's concern for profit does not exclude taking into account the interests of all who have a stake in the firm (stakeholders).This study has explored banks' views and motives for CSR practices as well as its association and contribution to financial performance of banks in Ghana.Results from the study show that banks in Ghana generally view CSR activities as being a strategic move toward enhancing the reputation and overall legitimacy of their operations.Additionally, the study found a positive relationship between banks' CSR activities and their profitability (ROA and ROE).However, the regression results depict that banks' CSR activities in their current form are not a dominant predictor of their profitability as compared with their size, growth, debt ratio and origin.Hence, firms engaging in CSR practices should do so in synchrony with other factors that have a significant impact on financial performance.These findings have important theoretical and practical implications.
Theoretically, this study has contributed to the literature in a number of ways.Firstly, it validates empirically the association between CSR activities and financial performance, in that firms perceive CSR activities as a way of gaining and retaining legitimacy which is strategic.Secondly, it adds the Ghanaian perspective of CSR to the growing sub-Saharan African literature on CSR, thereby enhancing our present understanding of how banks in Ghana view the impact of CSR activities.Practically, the study findings imply that managers should spend more effort and resources on their CSR activities as their organisations will be rewarded with legitimacy by stakeholders and, in the long run, perhaps financial performance.Additionally, the initiation, development and implementation of CSR activities should be strategic, not haphazard, in order to gain the associated positive benefits.

Limitations and recommendations for future research
This study, like any other, is subject to certain limitations.One such limitation is that the study is focused on a single industry (the banking sector) which means the results are biased and only applicable to that sector.The second limitation was the small sample size of the study, arising from the small population of banks in the country.Thus, the extent to which the results of this study can be generalised is questionable.Also, a structured and closed-ended questionnaire was employed in the survey, limiting additional contextual issues that could have been generated from interviews.
In this regard, future studies covering many other sectors or, better still, having wider geographical coverage, may provide more comprehensive evidence.Such studies could also include interviews so as to narrate an in-depth qualitative research.Also, a comparative study of the impact of CSR and financial performance can be done for various industries such as the oil and gas, food and manufacturing and sectors in order to provide more insight on the differences amongst them.To elicit further insight, however, future studies on strategic CSR should also capture the intensity and content of strategic planning as well as the quality of expertise employed, as these factors are known to influence financial performance.
Lastly, further research could examine the influence of CSR on firms' human capital development, knowledge and talent management, organisational development strategies and organisational settings (organisational culture and climate).
ROE = earnings after interest and taxes/total equity for firm ROA = earnings before interest and taxes/total assets for firm SIZE = size of the firm (log of total assets) for firm GROW = growth in sales for firm DR (debt ratio) = total liability over total assets ε (error) = the error term β0 = the average performance of the bank in the absence of the control factors

TABLE 1 :
Summary statistic for banks' view of corporate social responsibility practices.

TABLE 2 :
Summary statistic for banks' motives for corporate social responsibility practices.

TABLE 3 :
Summary descriptive and correlation statistics for both dependent and independent variables.
ROA, return on assets; ROE, return on equity; DR, debt ratio; CSR, corporate social responsibility.

TABLE 4 :
Standard multiple regression analyses.