Communicating philanthropic CSR versus ethical and legal CSR to employees: empirical evidence in Turkey

Abstract

Purpose

This study investigates the mechanism through which banks employ corporate social responsibility (CSR) commitment to engage in employees. The values of different types of CSR engagement (i.e. philanthropic CSR vs ethical and legal CSR) are distinguished and their influences on employee identification are analyzed. The moderation effect of CSR communication through corporate social media is examined in this context.

Design/methodology/approach

A sample of 254 respondents was collected through surveying the employees of one of the largest banks in Turkey.

Findings

Findings suggest that ethical and legal CSR is perceived more importantly than philanthropic CSR by employees in the banking industry. In addition, the level of transparency and frequency of CSR communication through corporate social media moderates the CSR types–employee identification relationship distinctively.

Practical implications

Special attention should be paid to the conditions under which CSR communication takes place effectively, as skeptics toward certain types of CSR initiatives may occur along with the disclosure of information about them.

Social implications

If organizations use social media communication in a way that would bring the CSR interests of their employees to light, it is likely that CSR initiatives will become more meaningful and have a greater societal impact.

Originality/value

This study contributes to the CSR literature through identifying the value of different types of CSR initiative and confirming the importance of transparent and proactive CSR communication on employee identification in the banking sector.

Citation

Wang, Y. and Pala, B. (2021), "Communicating philanthropic CSR versus ethical and legal CSR to employees: empirical evidence in Turkey", Corporate Communications: An International Journal, Vol. 26 No. 1, pp. 155-175. https://doi.org/10.1108/CCIJ-01-2020-0014

1. Introduction

Corporate social responsibility (CSR), defined as “behaviors [aligned] with the norms and demands embraced by their main stakeholders” (Maignan and Ferrell, 2004, p. 6), serves to meet the expectations of various stakeholders (Agarwal and Berens, 2009; Clarkson, 1995; Schoeneborn et al., 2020). It has its roots in stakeholder theory that the sustainable growth of a company rests on its relationship with a variety of important stakeholders (Carroll, 1991; Wang and Holznagel, 2020). In recent decades, the digitalization-driven CSR challenge is even more pressing for companies because they need to meet a variety of stakeholder expectations for their operation and survival (Chen et al., 2019; Cheng et al., 2019). While some types of CSR may be perceived positively by primary stakeholders (i.e. who can affect or is affected by the achievement of the organization's objectives), they can be perceived negatively by others (Maignan and Ferrell, 2004; Wang and Berens, 2015) or vice versa. For example, Hemingway and Maclagan (2004) and Wang et al. (2008) suggested that philanthropic CSR which is highly desirable and purely voluntary should be distinguished from other CSR initiatives, which are either required or expected by the stakeholders. Primary stakeholders, such as employees, may associate the financial success of an organization with their own benefits (Donia and Sirsly, 2016; Ellemers et al., 2011). As a consequence, they may challenge the purpose of investing in philanthropic CSR, which can diversify the corporate resources away from business operations. On the contrary, ethical and legal CSR, which are distinguished from philanthropic CSR in Carroll (1979, 1991), may be recognized by employees as valuable resources, given that they are either required or expected to meet social norms and thus should be accounted in corporate strategy.

Recently, in the literature, a growing interest in the impact of CSR on employees focused on how individuals’ attitudes are affected by an organization's CSR engagement (Donia and Sirsly, 2016). A few positive links have been established, such as the impact of CSR initiatives on organizational identification (Carmeli et al., 2007) and on group commitment to an organization (Brammer et al., 2007). Despite extensive research on how the global trend of CSR engagement influences employees' attitudes and behaviors (see, e.g. Lee et al., 2013a, b; Bauman and Skitka, 2012; Kim et al., 2010; Bhattacharya et al., 2008), studies on the impact of distinct types of CSR on employees is underrepresented in the literature. Rangan et al. (2012) urged researchers to pay more attention to “…what the best way forward is” (p. 4), rather than “whether” to embrace CSR or not. Thus, this paper intends to meet the call by examining how employees assess and react to philanthropic CSR versus ethical and legal CSR. Organizational theorists view managing employees' attitudes as a deterministic factor for successfully implementing CSR initiatives (Martinko, 2004). Widely recognized as an important stakeholder group, employees' perceptions toward CSR may affect the level of their job satisfaction, productivity and retention and consequently influence the growth potential and profitability of an organization in the long run (Bhattacharya et al., 2008; Carmeli et al., 2007). Along with it, positively perceived CSR engagement also helps to attract prospective employees in the job market (Greening and Turban, 2000; Turker, 2009) due to their belief in good opportunities for personal growth in these organizations (Bhattacharya et al., 2008).

An effective CSR communication contributes to constituting positive outcomes of social responsibility engagements (Du et al., 2010). Many studies in the literature have focused on CSR communication toward external stakeholders such as customers and the local community (see, e.g. Kim and Ferguson, 2014; Lee et al., 2013a, b). Our paper differs from existing literature through clarifying the impact of internal CSR communication on employees. In particular, with the increased use of social media in corporate communication (Kesavan et al., 2013), we are interested in knowing whether such a new management tool enhances employees' attitudes toward the engagement in CSR initiatives.

Our empirical research is conducted in the banking sector in Turkey. Ararat and Göcenoğlu (2006) argued that Turkey is often described as a country of conflicts. Its economic and historical factors coupled with its unique geography cause duality and diversity in economic, social and cultural dimensions. The development in the banking sector is crucial for the Turkish society, as it makes up approximately 80% of the whole financial system (Taşkın, 2015). In the banking sector in Turkey, employees are recognized as one of the most important stakeholder groups. Such a recognition is influenced by the nature of the market the banks operate (Ararat, 2008). As banks are strongly regulated, to gain the license to operate, they need to comply with a variety of disciplines and responsibilities, such as conforming to ethical decision-making and the social norms (Smeltzer and Jennings, 1998; Salmones et al., 2009; Bennett and Kottasz, 2012; Krasodomska, 2015). Hence, the role of ethical and legal CSR in fulfilling the expectation of employees might be more prominent compared to that of philanthropic CSR, despite the fact that more and more banks started to proactively integrate philanthropic CSR into corporate practices after the financial crisis in 2008 (Fatma et al., 2015). On the other hand, the understanding of CSR in Turkey originated from the Ottoman Empire, when philanthropy was a common form of CSR in the Islamic tradition (Gokcenoglu and Onan, 2008). Despite a rich philanthropic tradition, legal frameworks which support corporate philanthropy are relatively weak in Turkey. Recently, the engagement of activist from international nongovernmental organizations (NGOs) further elevated the importance of philanthropic CSR in Turkey (Arguden, 2002; Gokcenoglu and Onan, 2008). Ararat (2008) argued that as Turkish business environment becomes more competitive, we are more likely to observe a wider adaptation of philanthropic CSR. For example, national banks in Turkey now started to engage in CSR activities that are related to supporting education and arts (Taşkın, 2015). Given a good fit of the ethical and legal CSR with the banking sector and the growing interests in philanthropic CSR, how employees in this sector in Turkey perceive the value of philanthropic CSR over ethical and legal CSR is not straightforward. Thus, our study aims to clarify the mechanism through which different types of CSR affect corporate employees distinctively and to provide the banks with guidelines on developing effective communication strategies for aligning employees through an effective engagement in CSR.

2. The theoretical framework

2.1 Philanthropic corporate social responsibility versus ethical and legal corporate social responsibility: the distinction

Stakeholders are broadly defined as “any group or individual who can affect or is affected by the achievement of the organization's objectives” (Freeman, 1984, p. 46). Clarkson (1995) distinguished stakeholders into two groups with respect to their impact on an organization: primary and secondary stakeholders. The support of primary stakeholders is indispensable for an organization's survival and support (Hillman and Keim, 2001). Employees who possess power, legitimacy and urgency, their right belongs to this category (Mitchell et al., 1997). Since they take part in the core operation of an organization, their commitment is critical for the organization's survival and growth (Clarkson, 1995; Mitchell et al., 1997). Thus, how employees respond to the organization's CSR initiatives will not only affect their attitudes and behavioral intentions toward their employer but will also further determine the outcomes of the CSR strategy (Santhosh and Baral, 2015).

The conceptualization of CSR varies in different disciplines, from the obligation of a company to maximize shareholders' value in early economic domain (Zenisek, 1979) to the responsibility of businesses for their impacts on society (European Commission, 2011). In this study, we adopted the stakeholder's perspective of CSR (Maignan and Ferrell, 2004). According to it, CSR serves to meet the expectations of various stakeholders, as opposed to improving social welfare as a whole (Pirsch et al., 2007; Agarwal and Berens, 2009). Since different CSR initiatives may affect specific stakeholder groups distinctively (Maignan and Ferrell, 2004; Wang and Berens, 2015), some types of CSR may be perceived positively by primary stakeholders, such as employees, but not by secondary stakeholders (e.g. local community, government and NGOs). Considering the resources that an organization can allocate to CSR are often limited, it might be difficult to satisfy all stakeholders. Thus, it would be of the organization's interest to identify which CSR initiatives may help engage in employees, in case that they are regarded as the key stakeholder group with which the organization needs to satisfy.

Carroll (1979, 1991) clarified four distinctive aspects of CSR, namely, economic, legal, ethical and philanthropic CSR. This classification has been widely recognized and adopted in the literature (Wang and Berens, 2015; Fatma et al., 2015). Carroll's framework is based on stakeholders' expectations, from the most required CSR to the least required but desired CSR. Economic responsibility, which refers to producing goods and services that the society wants and selling them at a profit, is the primary and required responsibility of a company. Legal and ethical responsibilities stand in the middle, which are both expected and desired by the stakeholders. Philanthropic responsibility, accounting for donations to social and charitable causes, such as support for education, culture and minorities (Godfrey, 2005), is highly desirable and purely voluntary (Hemingway and Maclagan, 2004; Wang et al., 2008).

An organization's decision concerning corporate philanthropy is often based on the management team's discretion (Carroll, 1991). Some scholars argued that corporate philanthropy contributes to enhancing corporate image and reputation (Lii and Lee, 2012) and reducing governmental burdens (Wang and Qian, 2011). In contrast, others suggested that corporate philanthropy may represent a pure corporate expenditure which diverts valuable resources of a company to unrelated operations (Friedman, 1970). As claimed by Godfrey (2005), corporate philanthropy stands between shareholders' value maximization and business citizenship.

As employees may associate the financial success of an organization with their own benefits, they may challenge the altruistic purpose of investing in a purely voluntary CSR initiative, which can diversify corporate resources away from business operations. Thus, they are expected more likely to value the endorsement of corporate philanthropy as an additional expenditure, as opposed to an indispensable instrument to achieve competitive advantage. On the contrary, ethical and legal CSR may be recognized as valuable resources by employees, given that they are either required or expected in the banking sector and thus should be accounted in corporate strategy. Therefore, the role of ethical and legal CSR on conforming to the expectation of employees might be more prominent compared to that of philanthropic CSR in the banking sector.

2.2 Corporate social responsibility and employee identification

Social identity theorists argued that individuals classify themselves and others into social groups and understand their place in the world by their personal identities and a number of social identities (Turner et al., 1979). As a consequence, employee identification is a perception of oneness with or belongingness to an organization (Riketta, 2005; Mael and Ashforth, 1992). Employees may define themselves by the similar attributes that define their organization (Dutton et al., 1994; Tourky et al., 2020). Bauman and Skitka (2012) distinguished identification from distinctiveness: the former depends on the value congruence between stakeholders and an organization, while the latter refers to the relative position of one organization compared to others. In other words, a high identification can be achieved if employees' value proposition that aligns with that of the organization. Employees who identify strongly with their organization view its successes as their own, which further determines their self-image (Bhattacharya et al., 2008).

In case that an organization engages in CSR proactively, employees may develop a strong identification with it due to the desire to make a social contribution and the feeling of pride to work for a good citizen (Rodrigo and Arenas, 2008). Thus, CSR initiatives, in general, may contribute to determine the level of employee identification with an organization (Kim et al., 2010; Greening and Turban, 2000; Dutton et al., 1994). Given the industry standard for ethical decision-making and conforming to the social norms (Bennett and Kottasz, 2012; Krasodomska, 2015), employees in the banking sector are expected more likely to value the endorsement of corporate philanthropy as an additional expenditure while considering ethical and legal CSR conforming to their expectations. The prediction is summarized in Hypothesis 1.

H1.

The positive impact of ethical and legal CSR on employee identification is stronger than that of philanthropic CSR.

2.3 Personal fit with corporate social responsibility

The person–situation fit draws on the congruence between individuals and the attributes of the situation (O'Reilly et al., 1991). Accordingly, employees' personal fit with respect to CSR refers to the match between employees' value propositions toward CSR and the types of CSR the organization is engaging in. Employees' responses to CSR are highly dependent on the personal-CSR fit of the individuals (Bhattacharya et al., 2008). For example, if a business supports dental health and decides to engage in corporate philanthropy (e.g. donating money to a dental association), cause marketing (e.g. giving a percentage of toothpaste purchases to the dental association) and corporate social marketing programs for promoting behavior change (e.g. supporting clinics that train children about dental care), employees may view the importance of the three approaches differently depending on their own value propositions (Bhattacharya et al., 2008). The more the employees agree with the rationale underlying a type of CSR initiative (i.e. a higher personal-CSR fit), the higher the level of positive responses will be achieved among employees (Rupp et al., 2006).

Despite the importance of personal fit in determining the perceptions among employees, only a few scholars have assessed this factor in the context of CSR engagement. For instance, Rodrigo and Arenas (2008) found that not only employees' opinions about a company but also their worldviews play an important role in developing attitudes toward the social and environmental responsibility. The study by McShane and Cunningham (2012) confirmed that to form a judgment employees first assess their own emotional engagement with a CSR initiative. It implies that employees need to possess a personal connection with the CSR initiative and feel passionate about it, in order to develop their beliefs in the authenticity of the CSR and in turn identify themselves with the organization. In line with the findings in the literature, we conjecture that if employees' personal value propositions align with the organization's CSR strategy, the CSR initiatives will form a more positive perception among employees. That is, a personal-CSR fit affects employee identification with the organization positively.

H2.

Personal fit with CSR influences employee identification with an organization positively.

2.4 The moderation effect of corporate social responsibility communication through corporate social media

One challenge for organizations committing to CSR is to communicate with stakeholders about their CSR engagement effectively (Amaladoss and Manohar, 2013). Employees, in particular, rely on proactive internal CSR communication to acquire a comprehensive understanding of their employer's engagement. Despite so, empirical findings in the literature pointed out that stakeholder's low awareness of CSR initiatives remains as a major issue in carrying out CSR-related strategies (Du et al., 2010; Morsing et al., 2008). Many employees, for instance, are neither familiar with nor fully aware of their organization's commitment (Hoeven and Verhoeven, 2013). Therefore, to evoke employee identification with CSR, it is crucial for an organization to develop effective communication strategies toward their employees.

Recently, social media has become a popular tool in corporate communication (Etter et al., 2019; Kesavan et al., 2013). While traditional media is based on monologue and one-way communication to disseminate messages, social media enables dialogical and two-way communication with no centralized control entities (Colleoni, 2013). The nature of social media allows the employees to initiate a dialogue with an organization through user-generated content, which may contribute to align the organization's agenda with the expectations of employees. In addition, as user-generated content is recognized as more credible and trustworthy, it may result in a stronger impact on shaping the perceptions among stakeholders compared to corporate-generated content (Christodoulides et al., 2012).

When communicating CSR on social media, two factors may determine its effectiveness on generating employee identification: the frequency of the disclosure and the transparency of the messages framed (see, e.g. Fieseler and Fleck, 2013; Coombs and Holladay, 2011). A regularly updated CSR platform online keeps the dialogue between an organization and its stakeholders alive and enhances the awareness of certain issues that the organization aims to focus on (Fieseler and Fleck, 2013). As a consequence, employees might be motivated to engage in online discussions about the ongoing CSR activities. Organizations, on the other hand, may receive immediate feedback from the employees enabling adjustment of the strategy toward the interests and expectations of employees (Lee et al., 2013a, b). Meanwhile, as an active communication on social media indicates the organization's endeavor to engage in CSR, employees may feel less skeptical about the corporate expense on CSR, in particular, on philanthropic CSR. Therefore, a high frequency of CSR disclosure on social media contributes to aligning employees' interests. In other words, it moderates the impact of CSR endorsement on employee identification with the organization.

H3.

The impact of CSR on employee identification is strengthened by the frequent CSR communication on social media.

Parallel to frequency, transparency of the message framed is also recognized as a deterministic factor in CSR communication (Coombs and Holladay, 2011). It requires an organization to provide truthful, substantial and useful information that enhances understanding of stakeholders, participation of stakeholders when identifying and distributing the information and being accurate in actions, policies and practices (Rawlins, 2009). Stakeholders' sensitivity of the information transparency is associated with their skepticism toward certain types of CSR initiatives (Morsing et al., 2008). Transparency, thus, is used as a cue by stakeholders to evaluate the authenticity and trustworthiness of an organization's CSR engagement (McShane and Cunningham, 2012). Therefore, an organization's CSR communication should take the three aspects into account as suggested by Balkin (1999) that not only provide the stakeholders with clear and timely information but also invite them to participate in the discussion. Correspondingly, social media platforms serve as a relevant tool to achieve these goals due to its highly interactive and timeless feature. Moreover, although companies often focus on communicating about their achievements in CSR, it is also important to be transparent with respect to negative news or outcomes (Collier and Esteban, 2007). For example, responding to negative comments proactively while sharing good details on CSR strategy may mitigate negative perceptions among stakeholders (Ali et al., 2015). In summary, a high transparency in CSR communication on social media contributes to forming positive perceptions among employees. In other words, it moderates the impact of CSR endorsement on employee identification with the organization.

H4.

The impact of CSR on employee identification is strengthened by the transparent CSR communication on social media.

The conceptual model is presented in Figure 1.

3. Method

To test the hypotheses, we conducted a survey among employees working at one of the largest banks in Turkey. A survey method was employed as it enabled us to investigate the perceptions with respect to CSR engagement among employees directly. The questionnaire was conducted using Qualtrics and generated in Turkish through applying existing Turkish translations of the measurements as well as the back-translation method. A convenient sampling method was used to recruit the participants. The data were collected in April 2016. A total of 328 employees participated in the survey, among which, 262 completed the questionnaire. Also, eight respondents indicated that they were not aware of the bank's CSR initiatives at all and thus were deleted from the sample. The final data set consisted of 254 valid participants among whom approximately 55.9% (N = 142) were female and 44.1% (N = 112) were male. The demographic information of the sample is presented in Table 1. The majority of the participants (56.7%) fall in the age range of 25–34. In the sample, 38.6% aged between 35–44 years, 3.9% aged between 45–54 years and a further 8% were in the age group of 18–24 years.

Previously validated measures were used and adapted to the context of our study. Both CSR types (i.e. philanthropic and ethical and legal CSR) were measured by three items adapted from previous study by Salmones et al. (2005), exploring the influence of CSR on loyalty and valuation services. As the bank under investigation engages in three themes of philanthropic CSR, namely, education, culture and environment, the items measuring philanthropic CSR were modified to fit into these themes. The measure of employee identification was based on the validated scales by Mael and Ashforth (1992). The construct of transparency of CSR communication on social media was assessed by four items developed by Rawlins (2009). The items were based on Balkin’s (1999) dimensions of transparency: informational, participatory and accountability. Then, these items were adapted to measure the frequency of CSR communication. In addition to the latent variables, personal awareness of and personal fit with CSR type were measured with one item, respectively. The former was measured through the question “are you aware of the social responsibility initiatives of the bank,” after the introduction of the CSR concept. The latter refers to the question “I believe that a bank should play a role in society that goes beyond the mere generation of profits.” All items aforementioned were assessed using the seven-point Likert scale. Further, we followed previous literature to control the demographic factors which were revealed to determine employee identification in other contexts, such as tenure (Organ and Ryan, 1995), age, gender, position and the level of education (Hall et al., 1970; Organ and Ryan, 1995; Stinglhamber et al., 2015).

A confirmatory factor analysis (CFA) in AMOS (version 22) was used to test the validity of the latent variables. All variables in the model were standardized as a preparation for generating the interaction terms (Aiken and West, 1991). The results are presented in Table 2. It can be observed that all constructs have Cronbach's alpha values above 0.7 (Nunnally, 1978) suggesting a good lower bound for reliability. The factor loadings of the same constructs are all higher than 0.5 except for Phil3 and Trans3. In addition, the average variances extracted (AVE) are all above the 0.5 benchmark (Fornell and Larcker, 1981) and the composite reliability all above 0.7, confirming a satisfied convergent validity. Overall model statistics shows that the comparative fit index (CFI), goodness-of-fit index (GFI) and root mean square error of approximation (RMSEA) (0.953, 0.899 and 0.074, respectively) are above or nevertheless close to the threshold, suggesting an acceptable model fit. However, the chi-square for the model is 225.57 (df = 94, p > 0.05). The relatively high χ²/df value might be attributed to the number of items involved (Hair et al., 2010; Theo et al., 2013). Similar issues are found in other published articles too (e.g. Liao et al., 2009; Conway et al., 2015).

Table 3 presents the correlation of all variables. As shown in the table, the independent variables, philanthropic CSR, ethical and legal CSR, personal fit, frequency and transparency are all correlated with identification but modestly. Values of the latent variables were computed as the average of all indicators associated with one construct. Then, the interaction terms between CSR-related variables (i.e. philanthropic and nonphilanthropic) and communication-related variables (i.e. frequency and transparency) were calculated to obtain the moderators. In addition to this approach, we also imputed the factor scores of the latent variables through the CFA results. Compared to the equal-weighted method, this approach takes the relative importance of each indicator into account.

4. Results

Ordinary least squares (OLS) regression was employed to test the direct and moderation effects predicted in the hypotheses. The regression results are shown in Table 4. Model 1 presents the regression results with the direct effects only. Model 2 controls the direct effect of two moderators – frequency and transparency, while model 3 also includes the interaction terms. As shown in model 1, both philanthropic (b = 0.211, p < 0.001) and ethical and legal CSR (b = 0.339, p < 0.001) influence employee identification significantly. The impact of ethical and legal CSR is only slightly higher than that of philanthropic CSR, weakly confirming the prediction in Hypothesis 1. In addition, a significant impact of personal fit on identification is observed (b = 0.176, p < 0.001), suggesting that the higher the personal-CSR fit, the more the employees identify with the organization engaged in CSR. This result confirms our prediction in Hypothesis 2.

As for the results presented in model 3, we found opposite moderation effects on philanthropic and ethical and legal CSR. Since both types of CSR are observed affecting employee identification positively, a positive moderation effect will strengthen the direct effect and vice versa. Frequency, for instance, strengthens the impact of philanthropic CSR on identification, though the interaction term is only significant at 0.1% level. It partially confirms our prediction in H3 that a proactive communication on corporate social media implies the financial institution's endeavor to engage in philanthropic CSR, thus, leading to higher employee identification. On the other hand, frequency weakens the impact of ethical and legal CSR, suggesting that over communicating expected CSR engagement (i.e. ethical and legal CSR) on social media may not be regarded as a positive signal by employees. In contrast, a strong and positive moderation effect is observed from transparency to ethical and legal CSR, implying that a high transparency of CSR communication on social media strengthens the impact of nonphilanthropic CSR on identification, whereas transparency negatively moderates the impact of philanthropic CSR. The moderation effects are illustrated in Figure 2.

5. Discussion

This study centralized on clarifying the mechanism through which different types of CSR affect corporate employee identification with an organization distinctively. We also examined whether corporate social media as a new management tool enhances employees' attitudes toward CSR commitment. A few theoretical, managerial and societal implications are summarized in this section.

5.1 Theoretical implications

The findings confirmed the importance of distinguishing different types of CSR when aiming at aligning specific stakeholder groups, such as employees, as suggested by Maignan and Ferrell (2004) and Wang and Berens (2015). We found that ethical and legal CSR is perceived more importantly than philanthropic CSR by employees in the banking sector in Turkey. While philanthropic CSR often turns out to be appealing to corporate constituents in developed countries (Dobers and Halme, 2009; Baughn et al., 2007), we confirm that more attention is assigned to abiding rules and regulations in emerging economies. In this context, employees tend to hold higher expectations toward the bank's commitment to ethical and legal CSR, though they stay positive toward the bank's engagement in philanthropic CSR as well. Ararat (2008) argued that despite a rich philanthropic history, legal and fiscal frameworks which support corporate philanthropy are relatively weak in Turkey. Hence, charity foundations are often set up with altruistic motives, even though they aim at supporting educational institutions, hospitals and arts and cultural centers. Given such a national context, strengthening legal and ethical CSR becomes a precondition for efficient labor markets and corporate competition in Turkey and thus is prioritized by the employees (Ararat and Göcenoğlu, 2006). Also, Ararat (2008) found that the importance of legal and ethical responsibilities is not clearly differentiated in Turkey implying that both CSR dimensions are perceived equally important among employees. The fact that philanthropic CSR is also perceived positively may be attributed to several factors, such as the influence of the social norms set in the global market (Baughn et al., 2007), the increasing negotiation power of international NGOs and the regulations imposed upon Turkey as a precondition for it to become a member of the European Union.

In comparison, other studies investigating CSR in the banking sector identified a positive relationship between philanthropic CSR and another key primary stakeholder group – the customers. For instance, McDonald and Lai (2011) while examining CSR in Taiwanese banking industry found that consumers are in favor of customer-centric philanthropic initiatives. Compared to ethical CSR, Salmones et al. (2009) revealed that customers from the Spanish retail banking industry are more identified with the banks engaging in philanthropic CSR. It is in line with our conjecture that internal (e.g. employees) and external stakeholders (i.e. customers) value the importance of discretionary CSR differently. While external stakeholders desire more philanthropic CSR, as suggested by Salmones et al.'s work, internal stakeholders may challenge the altruistic purpose of investing in a purely voluntary CSR initiative which can diversify the corporate resources away from business operations.

Second, we observed distinctive moderation effects of transparency and frequency on philanthropic and nonphilanthropic CSR, respectively. In particular, a high transparency in communicating philanthropic CSR on corporate social media affects the CSR–employees identification relationship negatively. It implies the challenge in CSR communication: as the stakeholders may become skeptical toward certain types of CSR initiatives, a voluntary disclosure of CSR information may not always lead to the support among corporate constituents (Morsing et al., 2008). Instead, it takes the risk to be perceived as hypocritical and monolithic with the intention to dominate and control the communication platform (Lee et al., 2013a, b). For instance, since the value of philanthropic CSR is not as clear as that of ethical and legal CSR for employees, communicating about the former proactively may cause confusions among the employees and make them suspect the true intention of the company to support philanthropy-related CSR causes (Rifon et al., 2004). Therefore, it will affect employee identification with the organization negatively. This finding encourages scholars to further investigate the value of social media in CSR communication with respect to different types of CSR initiatives. As suggested by Kim and Ferguson (2014), what makes CSR communication effective from the stakeholders' perspective is particularly important up-to-date. Without distinguishing “what to communicate” and “how to communicate” about CSR from the CSR activity itself, it is challenging to assess the success of the communication process itself.

Next, our results confirmed the positive link between the personal-CSR fit and employee identification. It is consistent with the findings in the literature that employees' personal preference with respect to CSR commitment affects their perceptions toward an organization (Rodrigo and Arenas, 2008; Bauman and Skitka, 2012; McShane and Cunningham, 2012; Story and Neves, 2015). In a broader context, Kim and Ferguson (2014) stated that the public expects CSR initiatives to be personally relevant too. The importance of personal-CSR fit implies the direction for future research to study CSR at the microlevel (i.e. individual level of analysis), in order to identify the underlying mechanisms and microfoundations through which firms can best align with the values of individuals, as suggested in Aguinis and Glavas (2012).

5.2 Managerial and societal implications

While many organizations fail to align employees with their CSR policy (Santhosh and Baral, 2015), we confirm the importance of addressing the interests of employees toward CSR. The influence of CSR on employees' perception not only depends on the CSR types (i.e. philanthropic vs ethical and legal CSR) and CSR communication but also on employees' personal-CSR fit. Thus, despite CSR initiatives as a useful management tool to engage in employees, carrying out a clear plan for internal CSR communication is a critical success factor. For example, Nan and Heo (2007) found that promoting CSR which has a high fit with a company's expertise is more likely to influence stakeholders' acceptance of the sincerity of the company's CSR motives. As ethical and legal CSR tends to fit a financial institution's expertise better than philanthropic CSR, communicating about the former toward the employees may result in higher employee identification with the organization.

The challenge, however, remains in “how to communicate” companies' CSR activities. Previous studies have suggested the importance of communication channels (Morsing and Schultz, 2006), credible communication sources (Maignan and Ferrell, 2001) and information consistency (Coombs and Holladay, 2011). The moderation effects of transparency and frequency identified in our study further support the value of informativeness in CSR communication (Morsing and Schultz, 2006). However, special attention should be given to the conditions under which transparency leads to a positive outcome, as skeptics toward certain types of CSR initiatives may occur along with the disclosure of information about them. Likewise, the use of corporate social media for internal CSR communication should be tailored toward employees' needs and expectations as opposed to promoting a top-down CSR strategy online. Organizations must consider employees as viable and impactful stakeholders in CSR communication and cultivate them as boundary spanners through fully capitalizing the value of corporate social media as a two-way communication tool. For example, organizations may even call on employee representatives as corporate ambassadors to communicate the corporate CSR policy with the external world through the online portal.

If organizations use social media communication in a way that would bring the CSR interests of their employees to light, it is likely that CSR initiatives will become more meaningful and have a greater societal impact. Ararat and Göcenoğlu (2006) argued that ethical corporate behavior, if managed to align stakeholders' interests, will accelerate the convergence of business culture toward CSR in Turkey and “may partially neutralize the societal cultural characteristics that may be unsupportive of CSR” (p. 32). In addition, if the employees' views of CSR converge with that of their organization, they will identify with their employer and perceive a strong oneness (Riketta, 2005; Mael and Ashforth, 1992). Hence, work is no longer a chore, and both parties – the organization and the employees – will benefit from such a win–win situation. However, one downside is that CSR will become a corporate tool for brainwashing the newcomers, who are more vulnerable and prone to identification for the purpose of social recognition.

6. Conclusion

Employee identification is crucial for an organization's sustainable growth. The CSR strategy of an organization, as a consequence, can only maximize its value with the support of employees. In this context, we studied how philanthropic CSR, in comparison to ethical and legal CSR, enhances employees' attitudes toward CSR commitment. The findings suggested that employees tend to hold higher expectations toward the financial institutions' commitment to ethical and legal CSR due to the general norms of abiding rules and regulations in the banking sector. Moreover, we identified the role of voluntary disclosure of CSR information as a double-edged sword. Despite its value on enhancing the positive relationship of ethical and legal CSR and employee identification, overexposing CSR information which is not necessarily appealing to employees may lead to their skeptics.

Despite our contributions to the literature, it is important to acknowledge several limitations of this study. First of all, Carroll's framework (1979, 1991) was adopted in this study to distinguish the CSR types due to it being recognized as the most well-known model in CSR literature. However, scholars interested in the specific value of different CSR types may also explore other classifications which placed CSR into a broader context, such as strategic versus responsive CSR (Kramer and Porter, 2006) and congruent versus incongruent CSR (Hietbrink et al., 2010). Second, CSR communication was investigated in this study from the corporate side only. Although it can be argued that this approach is highly relevant to managing and communicating CSR at the corporate level, the value of corporate social media as a two-way communication tool was not thoroughly discussed. Scholars focusing on user-centric communication (i.e. users, not organizations controlling the creation and distribution of information, see, e.g. Christodoulides et al., 2012) could explore effects of employees' CSR communication through corporate social media as a new line of research. Next, we admit the fact that our study only focused on one bank. Given such a case study approach, the results might not be generalizable. Despite so, our study provided an in-depth analysis that could not have been accomplished by looking at multiple banks. The case study approach was necessary, given that little had been known about the impact of internal CSR communication on employees in Turkey. An analysis of multiple banks would have provided shallow information, which is beyond our scope when we research phenomena that we have not studied before (Yin, 2017). Further, while examining internal CSR communication, our study did not control the role of the internal culture, the corporate values and the leadership style in the employees' CSR perceptions and corporate identification. For example, quality relationships between leadership and employees can lead to the latter's higher embracement of CSR (Mallory and Rupp, 2014). The bank we studied could have had a predominantly transformational leadership style and a positive internal culture that may have influenced the employees' views of CSR and their identification. Future reach can consider examining the impact of CSR communication on employee identification while controlling these factors (e.g. internal culture, corporate values and leadership style).

Figures

Figure 1

The conceptual model

Figure 2

Moderation effects

Table 1

Demographics

Gender	Frequency	Percentage
Male	112	44.1
Female	142	55.9
Age
18–24	2	0.8
25–34	144	56.7
35–44	98	38.6
45–54	10	3.9
55 and above	0	0.0
Level of education
High school graduate	15	5.9
Bachelor's degree	164	64.6
Master's degree	73	28.7
PhD	1	0.4
Other	1	0.4
Tenure
Less than a year	6	2.4
1–2 years	8	3.1
3–5 years	33	13.0
6–10 years	107	42.1
more than 10 years	100	39.4
Position
Junior auditor/auditor	10	3.9
Lawyer/architect/engineer	5	2.0
Manager	66	26.0
Officer	37	14.6
Junior specialist/specialist	115	45.3
Other	21	8.3

Table 2

Results of the confirmatory factor analysis

Construct	Items	Factor loadings	AVE	Composite reliability	Cronbach's alpha
Philanthropic CSR (Phil)			0.553	0.776	0.734
	Phil → Phil1	0.895***
	Phil → Phil2	0.819***
	Phil → Phil3	0.432***
Ethical and legal CSR (EL)			0.666	0.857	0.855
	NonP → NonP1	0.723***
	NonP → NonP2	0.890***
	NonP → NonP3	0.826***
Frequency (Freq)			0.787	0.940	0.942
	Freq → Freq1	0.853***
	Freq → Freq2	0.860***
	Freq → Freq3	0.945***
Transparency (Trans)			0.592	0.800	0.760
	Trans → Trans1	0.891***
	Trans → Trans2	0.904***
	Trans → Trans3	0.405***
Identification (Ident)			0.682	0.895	0.892
	Ident → Ident1	0.697***
	Ident → Ident2	0.761***
	Ident → Ident3	0.921***
	Ident → Ident4	0.903***

Note(s): ***p < 0.001, **p < 0.01, *p < 0.05, †p < 0.1

χ² = 225.570, df = 94, GFI = 0.899, CFI = 0.953, RMSEA = 0.074

Table 3

Correlation

	Phil	EL	Ident	Freq	Trans	Fit	Awareness	Gender	Age	Education	Position	Tenure
Phil	1
EL	0.444**	1
Ident	0.426**	0.491**	1
Freq	0.233**	0.161*	0.284**	1
Trans	0.123	0.163**	0.136*	0.741**	1
Fit	0.158*	0.109	0.287**	0.339**	0.190**	1
Aware	0.363**	0.174**	0.230**	0.107	−0.003	0.034	1
Gender	−0.017	0.011	0.102	−0.220**	−0.240**	−0.019	0.038	1
Age	−0.123*	−0.170**	−0.228**	−0.122	−0.074	−0.073	−0.037	0.029	1
Education	0.077	−0.048	0.057	0.035	−0.084	−0.011	0.016	0.065	−0.102	1
Position	−0.090	−0.137*	−0.038	−0.035	−0.043	0.076	−0.081	0.053	0.278**	0.188**	1
Tenure	−0.120	−0.053	−0.082	−0.087	−0.045	−0.025	−0.122	0.151*	0.658**	−0.090	0.259**	1

Note(s): **p < 0.01, *p < 0.05

All variables are standardized

Table 4

Regression results

	Model 1	Model 2	Model 3
	Standardized estimates (Std. error)	Standardized estimates (Std. error)	Standardized estimates (Std. error)
Constant	−0.091 (0.310)	−0.125 (0.311)	−0.144 (0.308)
Philanthropic CSR	0.211*** (0.065)	0.187** (0.065)	0.168* (0.065)
Ethical and legal CSR	0.339*** (0.057)	0.342*** (0.057)	0.364*** (0.058)
Personal fit	0.176*** (0.045)	0.133** (0.047)	0.131** (0.047)
Awareness	0.081† (0.048)	0.068 (0.047)	0.076 (0.048)
Gender	0.159† (0.090)	0.198* (0.092)	0.163† (0.091)
Position	0.023 (0.031)	0.025 (0.031)	0.016 (0.031)
Age	−0.281** (0.104)	−0.263* (0.103)	−0.242* (0.101)
Tenure	0.075 (0.066)	0.070 (0.065)	0.069 (0.064)
Education	0.050 (0.079)	0.023 (0.079)	0.054 (0.079)
Frequency		0.206** (0.074)	0.178* (0.074)
Transparency		−0.123 (0.081)	−0.091 (0.081)
Phil * frequency			0.174† (0.094)
EL * frequency			−0.196* (0.083)
Phil * transparency			–0.297** (0.104)
EL * transparency			0.188* (0.092)
R-squared	0.378	0.399	0.428
F-value	16.505***	14.590***	11.877***

Note(s): ***p < 0.001, **p < 0.01, *p < 0.05, †p < 0.1