It is common to think of corporate generosity as a success outcome; a company creates wealth and shares it. However, the high-performance service companies [studied] teach a different lesson. Generosity not only is an outcome of success, it is a critical input.

Leonard Berry, Discovering the Soul of Service (1999, p. 246)

1 Introduction

Over the course of the 20 years since the publication of Leonard Berry’s influential book Discovering the Soul of Service, corporate social responsibility (CSR) has experienced accelerated growth in importance and now occupies a prominent standing in contemporary service business practices. CSR refers to an organization’s obligations to its stakeholders and society-at-large (Maignan et al. 2005; Smith 2003). While many aspects of CSR research continue to receive considerable study and debate, there remain some aspects of CSR in need of greater theory-building and research attention. One particularly fertile opportunity for exploration is the potential impact of CSR as a preventive mechanism for frontline service employees (FSEs) when faced with customers who are deviating from normal customer behavior.

While it is well established in the literature that CSR can effectively boost job performance of FSEs (e.g., De Roeck and Maon 2016; Korschun et al. 2014; Larson et al. 2008), far less is known about the potential value of CSR in relationship to workers who routinely expend emotional labor in their roles as FSEs (Kim et al. 2018). In particular, there is a lack of research on how CSR perceptions held by FSEs may contribute to building empathy and self-efficacy which, in turn, may help FSEs respond to dysfunctional customer behavior. Expanding our understanding of CSR’s impact on empathy and self-efficacy provides important managerial revelations. In the services marketing and management literature, FSEs’ levels of empathy and self-efficacy are major determinants of quality service delivery (Nguyen and Leclerc 2011; Wilder et al. 2014). For a service performance to reach a successful fruition, it is essential for FSEs to understand their customers’ needs, while also demonstrating expertise and problem-solving abilities (Nguyen and Leclerc 2011).

The key research objective of this study is to illuminate how CSR can serve as an antecedent to customer empathy and self-efficacy and help FSEs cope with dysfunctional customer behavior. To expand extant knowledge about the value of CSR, this study investigates how the psychological mechanisms of CSR vary by stakeholder perspectives. In broad terms, CSR can be grouped into internal and external dimensions. Internal CSR initiatives, policies, and processes ensure that a company’s operations are responsible and ethical. External CSR initiatives contribute to improving the communities in which a company operates (Drumwright 2014). Through our examination of how the dual presence of external and internal CSR stakeholder-directed activities, FSEs’ perceptions of external CSR initiatives undertaken by their company are shown to strongly influence their customer orientation while their company’s internal CSR activities made a stronger impact on FSEs’ beliefs to successfully perform their jobs.

This study begins with a review of relevant services management and CSR literatures in order to gain a deeper understanding of how FSEs’ CSR perceptions may help shape their perceptions of and reactions to dysfunctional customer behavior. In the following sections, the theoretical model and hypotheses are justified. This is followed by a description of the research method used to empirically test the posited relationships, and the results are reported. The paper concludes with a discussion of the key findings and their managerial implications and directions for further research.

2 Conceptual framework and hypotheses development

The proposed conceptual model is depicted in Fig. 1. We anticipate that FSEs’ empathy and self-efficacy mediate the relationship between FSEs’ perceived CSR of their company and FSEs’ perceptions of dysfunctional customer behavior. In the posited model, the strength of CSR as a psychological mechanism is expected to be impacted by the FSEs’ perspective of internal CSR versus external CSR stakeholder activities. In order to capture potential fallout created by dysfunctional customer behavior, we anticipate that FSEs’ perceptions of dysfunctional customers directly contribute to FSEs’ intentions to retaliate against such customers. Next, we develop the constructs of our model and present supporting literature for their respective posited relationships.

Fig. 1
figure 1

Conceptual model

Full size image

2.1 FSEs’ perceptions of CSR and empathy

CSR describes the degree to which companies assume economic, legal, ethical and discretionary responsibilities toward their stakeholders (Maignan et al. 1999). Stakeholder theory asserts organizations should strive to address the needs and wants of different groups of stakeholders, including those that either affect or are affected by a company’s activities (Freeman 1984). CSR is delivered in a wide array of activities, such as employee diversity initiatives, environmental sustainability, support of community events, and monetary and in-kind donations to charitable causes (Sen and Bhattacharya 2001). Based on the principles of the stakeholder theory (Freeman 1984; Pérez et al. 2013), employees are influenced by the tendency and manner in which a company treats its employees, customers, shareholders, distributors, suppliers, and other members of the community in which it operates.

CSR initiatives can be broadly grouped by whether they are primarily intended to benefit employees themselves (i.e., internal CSR stakeholders) or other stakeholders functioning outside the organization (i.e., external CSR stakeholders) (Scheinbaum et al. 2019). FSEs’ external CSR perceptions are shaped by the overarching viewpoint that their employer is concerned with improving the general well-being of society (Berger et al. 2007). Examples of external CSR stakeholder initiatives include corporate funding and support of charitable causes, nonprofit organizations, and the community-at-large. FSEs’ internal CSR perceptions reflect how well a company treats its employees, which corresponds to fair compensation, opportunities for education and self-development, and provision of a pleasant and safe work environment. Employees perceive these internal CSR stakeholder activities as a reward for their dedication and positive contributions to the company (Elmadağ and Ellinger 2018; Gong et al. 2014).

Reciprocal employee behavior stems from the just manner in which others are treated by their company, including fellow employees, customers, and people in the local community (Whitener et al. 1998). Employees can evaluate their organizational support in accordance with their perceptions about the way their organization treats external groups (Glavas and Kelley 2014; Rupp 2011). When employees perceive their company as socially responsible, they believe their company is just and cares about the well-being of its employees (Panagopoulos et al. 2016; Rupp 2011; Wang et al. 2017).

In a service management context, FSEs are required to treat customers with empathy, even when dealing with unruly and disruptive customers (Hochschild 1983; Yagil and Medler-Liraz 2019). The nature of empathy is interpreted as a type of personal trait, behavior, or experience; it may be either cognitive or affective (Kerem et al. 2001). The present study’s view of empathy is recognizing and understanding others’ vantage points from a cognitive perspective (Bernstein and Davis 1982). A company’s mix of external and internal CSR stakeholder activities can bring positive affects toward establishing and reinforcing an ethical climate supported by individual compassion for others (Moon et al. 2014). Moreover, an ethical climate means the organization’s members establish a consensus on ethical correctness standards and how ethical issues are treated in the organization (Olmedo-Cifuentes et al. 2014).

A company’s internal CSR stakeholder activities help establish an ethical work environment (Skudiene and Auruskeviciene 2012). Employees working in a strong ethical climate are more apt to recognize the pain and hardships of those they interact with and become increasingly motivated to behave out of compassion for serving others (Kim et al. 2018; Zoghbi-Manrique-de-Lara and Guerra-Baez 2016). Employees develop greater compassion in the workplace when their personal moral values correspond with the organizational values of their employer (Moon et al. 2014). Employees’ moral value perceptions are positively influenced by how a company demonstrates socially responsible behavior toward its external stakeholder groups (Valentine and Fleischman 2008). Based on this discussion, FSEs’ perceptions of their organization’s external CSR stakeholder activities are anticipated to increase their level of customer empathy. Additionally, when a company provides its employees with fair compensation and rewards, it helps employees perceive they are part of an ethical working environment (Collier and Esteban 2007). Thus, we anticipate that FSEs’ respective perceptions of external and internal CSR activities of their employer will exhibit a direct and positive effect on their customer empathy.

H 1a

FSEs’ external CSR perceptions have a positive relationship with their empathy for customers.

H 1b

FSEs’ internal CSR perceptions have a positive relationship with their empathy for customers.

2.2 FSEs’ perceptions of CSR and self-efficacy

According to social identity theory, employees’ positive perceptions of their company contribute to intrinsic motivation to conform to that particular company’s identity (Ellemers et al. 2004). On the basis of social identity theory (Hogg and Abrams 1988), an individual’s sense of affiliation with a socially responsible company creates a positive affect and enhances their ability to maintain a positive self-definition at work, including in times of adversity. Workers identify with their employer when they believe the company’s characteristics reflect their self-concepts (Kim et al. 2010). Positive employee sentiments are amplified when they feel their company is positively impacting external stakeholder groups (Armstrong 2006; Heslin and Ochoa 2008). Intrinsically motivated employees improve their sense of self and level of confidence in their work performance (Peterson 2004; Ryan and Deci 2000). This leads to better job performance (Skudiene and Auruskeviciene 2012).

Self-efficacy reflects confidence in a person’s capability to perform certain tasks (Bandura 2001). Through positive CSR perceptions of their company, employees are more likely to be emotionally engaged and align their personal development with their company’s development (Lee et al. 2013). As a result, employees have an enhanced desire to competently perform their work-related tasks (Maertz et al. 2005). As for internal CSR stakeholder activities, developing skills and training for employees can also positively influence employees’ self-efficacy and self-esteem (Baruch and Peiperl 2000; Muafi and Gusaptono 2010). Employee self-efficacy is impacted by the manner in which a company treats its employees, as well as financially compensates and rewards its employees (Gong et al. 2014; Lin et al. 2012). Relatedly, the fulfillment of individual needs plays a key role in the process in which CSR increases employees’ intrinsic motivation (Kim and Scullion 2013). The need for achievement is a motive closely associated with individual responsibility for job performance. When FSEs take pride in the internal and external socially responsible behaviors of their company, their motivations for achievement increase, which in turn leads to FSEs’ self-efficacy. Hence, the following set of hypotheses is proposed:

H 2a

FSEs’ external CSR perceptions have a positive relationship with their self-efficacy.

H 2b

FSEs’ internal CSR perceptions have a positive relationship with their self-efficacy.

2.3 Effect of external CSR versus internal CSR on empathy and self-efficacy

Based on relevant CSR literature, we anticipate that CSR initiatives directed at internal and external stakeholders serve as different psychological mechanisms for FSEs and therefore may heterogeneously impact their respective posited relationships to FSEs’ self-efficacy and empathy. In line with the previous discussion, internal CSR activities are chiefly aimed at FSEs themselves. Meanwhile, from the FSEs’ perspective, external CSR activities are focused on external stakeholders. Extending this classification system to empathy and self-efficacy, whereas empathy means taking the customer’s perspective in the employee’s work-related attitude formation (Wieseke et al. 2012), self-efficacy pertains to job-related self-esteem and confidence in their work abilities (Lin et al. 2012). Following this logic, FSEs’ customer empathy can be viewed as holding an “other-focused” perspective whereas FSEs’ self-efficacy is regarded as “self-focused”. Hence, employees’ behavior in the workplace changes according to the psychological mechanism, depending on the types of intended internal and external beneficiaries (McNeely and Meglino 1994).

When employees perceive their company delivers social value, they may be motivated to behave in the same way and react in a manner that is generally more beneficial to their customers (Meyer et al. 2004). FSEs’ become more cognitive in their desire to both view and understand events and information from their customers’ perspective, thus increasing their level of empathy. Consequently, it is expected that the effect of FSEs’ perceptions of external CSR will be stronger in relationship to their empathy (compared to their self-efficacy to perform work-related tasks). Yet, FSEs may be motivated to produce high self-efficacy because they perceive internal CSR activities as a reward for their performance (Gong et al. 2014). In sum, the effect of FSEs’ perceptions of the company’s internal CSR is anticipated to yield a stronger impact on FSEs’ self-efficacy than on their level of customer empathy. Therefore,

H 3a

The positive effects of FSEs’ external CSR perceptions on their empathy are stronger than the effects of FSEs’ internal CSR perceptions on their empathy.

H 3b

The positive effects of FSEs’ internal CSR perceptions on their self-efficacy are stronger than the effects of FSEs’ external CSR perceptions on their self-efficacy.

2.4 FSEs’ empathy and dysfunctional customer behavior

Many FSEs suffer from unfriendly, abrasive, and sometimes even verbally abusive customers (McColl-Kennedy et al. 2009). FSEs can experience negative psychological states such as feelings of degradation and stress disorders when exposed to dysfunctional customer behavior (Yue et al. 2017). In the services marketing literature, dysfunctional customer behavior has been referred to as problem customers (Bitner et al. 1994), deviant customer behavior (Moschis and Cox 1989), aberrant consumer misbehavior (Fullerton and Punj 2004), and “customers from hell” (Harris and Reynolds 2003). Although the use of such terms creates added confusion among scholars (Fisk et al. 2010), shared among all these terms is some customers violate generally accepted behaviors and display behavioral standards that are deemed unfair by other customers, employees, and their institutions (Harris and Reynolds 2003; Jung and Yoo 2017). This study focuses on perceptions of FSEs interacting with customers in a fashion that are in violation of norms through various dysfunctional behavior such as inappropriate complaints and verbal aggression (Harris 2013; Kang and Gong 2019; Kim et al. 2018).

Based on attribution theory, the presence of FSEs’ empathy for customers can influence how they receive dysfunctional customer behavior. Attribution theory explains how individuals make causal conjectures regarding another’s motivation for a given behavior (Folkes 1988). Empathic individuals are more apt to sense a given situation correctly and pay greater attention to the entire situation rather than solely focus on the person(s) involved (Sulzer and Burglass 1968). Empathy also reduces the likelihood of negative experiences during social interactions (Hodgson and Wertheim 2007). The greater the level of empathy, the lower the likelihood the individual gets his or her feelings hurt or becomes angry about a particular situation (Konstam et al. 2001). From the customer’s perspective, customers showing higher levels of employee empathy are more forgiving of FSEs when experiencing customer dissatisfaction (Yagil and Luria 2016). From the employee’s perspective, benevolent employees are more prone to exceed their specified job requirements in order to pursue customers’ well-being (Jarvenpaa et al. 1998). FSEs’ empathy is anticipated to negatively weaken their perceptions of dysfunctional customer behavior because they are better equipped to view situations from the customer’s perspective.

H 4

FSEs’ empathy is negatively related to their perceptions of dysfunctional customer behavior.

2.5 FSEs’ self-efficacy and dysfunctional customer behavior

Self-efficacy is a motivational construct, and it affects individuals’ emotional reactions. Individuals calculate, integrate, and evaluate information on their capabilities in order to arrange choices and efforts (Gist 1987). Once individuals are strongly confident in their innate abilities, they can exert more strength in handling related services (Sok et al. 2019). Previous research reports the higher level of self-efficacy, the more efforts are put forth to implement related duties successfully (Chen and Wu 2014; O’Neill and Mone 1998). Wood and Bandura (1989) assert individuals with a high level of self-efficacy are more likely to succeed and produce good results. Hartline and Ferrell (1996) found that FSEs’ self-efficacy produces significant positive effects on customers’ perceptions of service quality. In addition, feelings of FSEs’ competence in relation to self-efficacy improve employee morale (Hartline and Ferrell 1996). Over time, employees with self-efficacy learn to cope with conflicts at work. In a study of flight attendants, researchers demonstrated how the individual beliefs pertaining to their ability to perform necessary tasks positioned them to cope with dysfunctional customer behavior (Heuven et al. 2006). Thus, FSEs’ self-efficacy is anticipated to be negatively related to their perceptions of dysfunctional customer behavior because of their higher capabilities to cope with abnormal customer conduct.

H 5

FSEs’ self-efficacy is negatively related to their perceptions of dysfunctional customer behavior.

2.6 FSEs’ perceptions of dysfunctional customer behavior and revenge intentions

Ongoing persistence of dysfunctional customer behavior can lead to FSEs’ seeking revenge tactics on customers, resulting in potentially detrimental consequences for the financial well-being to their employer (Harris 2013; Harris and Reynolds 2003). FSEs’ revenge intentions can be described as plans to punish customers for caused damages (Grégoire et al. 2009). Humans possess a basic instinct to seek punishment for social aggression (Lockwood 2006). According to Hirschman’s (1970) customer loyalty framework, customers sometimes retaliate against service providers when exiting as a form of restitution, deterrence, or punishment. In the same vein, dysfunctional customer behavior fuels FSEs’ retaliation reactions (Huang and Brown 2016). Previous studies show that FSEs exposed to negative customer interactions are likely to direct their negative emotions during subsequent contact, thus creating service sabotage behavior (Harris and Ogbonna 2006; Skarlicki et al. 2008). Emotional responses induced by dysfunctional customer behavior may lead FSEs to seek revenge as a coping mechanism (Yeh 2015). Hence,

H 6

FSEs’ perceptions of dysfunctional customer behavior have a positive relationship with their revenge intentions.

3 Research method

To test the proposed model, we conducted a survey of FSEs working at a large property and casualty insurance company in South Korea. The survey data are comprised of customer service representatives and their supervisors. Primarily serving customers with auto insurance coverage, the major duty of the surveyed customer service representatives and their supervisors was to help customers file insurance claims involving recent auto accidents. From preliminary discussions held with managers of the auto insurance division of the insurance company, we learned it was common practice for customers to insist on receiving greater compensation than allowed under the contract terms of their insurance policies.

Data collection was conducted using an online survey that was accompanied by an email from a human resource manager of the insurance company encouraging their participation. The survey was sent to all full-time customer service representatives and their supervisors who worked in the auto insurance division of the insurance company. The link to the survey was left open for 2 weeks in April 2017. A total of 268 responses were collected out of 280 individuals invited to participate. Elimination of 31 incomplete answers resulted in 237 valid responses (84.6%). More than 90% (92.8%) of the study’s respondents worked as customer service representatives. Two-thirds of the respondents were female. At 98%, almost all of the respondents were below the age of 40, with 73% between 20 and 29 years old and 25% ranging from 30 to 39 years old.

3.1 Measurement and scale items

Understanding the conceptual model’s posited relationships required measurement of FSEs’ perceptions of their employer and customers. Existing scales from the literature were used to measure the constructs of interest with some slight modifications made to fit the context of this study. All six constructs were operationalized using multi-item measures and 5-point Likert-type scales, anchored by 1 = strongly disagree and 5 = strongly agree.

A five-item scale measuring FSEs’ perceptions of the external CSR was adapted from Lichtenstein et al. (2004) and Pérez et al. (2013). Each external CSR item gauged the respondents’ perceptions of how their company served nonprofit organizations and communities as well as supported social and environmental issues. Five items of FSEs’ perceptions of the internal CSR were adapted from Pérez et al. (2013). Each internal CSR item assessed how respondents viewed whether their company provided fair salaries, just treatment, training opportunities and a good working environment. Four items of FSEs’ self-efficacy were adapted from Karatepe et al. (2006) to focus on how FSEs viewed their abilities and associated levels of confidence in relationship to their job performance. Four items of FSEs’ empathy were modified from McBane (1995) and Wilder et al. (2014) to capture the extent to which FSEs empathized with the customer by seeing the service encounter from the customer’s perspective.

Eight items of perceived dysfunctional customer behavior measure how FSEs perceive the extent to which customers were aggressively combative and overly demanding to serve. Five items of verbal aggression behaviors were adapted from Dormann and Zapf (2004) and three items measuring FSE mistreatment were modified from Wang et al. (2011) to match the context of the current study. Finally, in order to capture FSEs’ revenge intentions, four items from Harris (2013) were adopted to evaluate intentions to do something bad to customers and punish them in some way. The scale items are listed in the “Appendix”.

Because the data collection took place in South Korea, we wanted to ensure the survey questionnaire was accurately translated to avoid any potential bias. We used a double-translation procedure recommended by Brislin (1970). Specifically, two bilingual researchers translated the construct scale items into the Korean language using back-translation and group-translation methods. After the two translators reached agreement on the exact words and expressions, a third translator translated the Korean survey back into English. The translators involved in this study confirmed the equivalence of the Korean version.

It was important to ensure not only that the translation represented a faithful rendering of the original text but also that the relevant target population clearly understood its meaning. Therefore, we conducted a pretest to identify how targeted survey respondents would react to the translated questionnaire. We utilized a qualitative method which has been demonstrated to provide richer insights as a pretest approach (Douglas and Craig 2007). We selected six respondents who closely resembled the targeted sample of FSEs and these individuals examined the questionnaires in an in-depth interview manner. As a part of this process, we checked for any issues regarding the respondent’s lack of comprehension or ambiguity of meaning. Issues identified from the pretest were sent back to the translation committee, which were then used to finalize the questionnaire.

3.2 Data analysis, reliability, and validity

The hypothesized relationships were tested using a two-step structural equation modeling (SEM) procedure advocated by Anderson and Gebring (1988). This method of measurement and testing relationships allowed for rigorous testing of measurement reliability and validity of the data before subjecting the structural model to tests of fit. A covariance matrix was created and subjected to confirmatory factor analysis. Despite the p < .01 significance of the Chi square (\(\chi^{ 2}_{{\left( { 3 8 3} \right)}} \, = \, 6 3 2. 5 5 9\)), the measurement model demonstrated good fit of the data based upon a variety of other absolute and incremental fit measures, including normed fit index (NFI) = .92, comparative fit index (CFI) = .97, goodness of fit index (GFI) = .85 , and root mean square error of approximation (RMSEA) = .053.

The construct measures yielded sound reliability and validity properties (Table 1). Evidence of convergent validity was found in each construct with the parameter estimates ranging from λ = .75 to .98. Evidence of discriminant validity was displayed via average variance extracted ranging from .68 to .86, with each measure exceeding the .50 benchmark (Bagozzi and Yi 1988). To assess internal reliability, composite reliabilities ranged from .90 to .97; all were above the .70 threshold of acceptability (Nunnally and Bernstein 1994). Table 2 presents the constructs’ correlations, means, and standard deviations.

Table 1 Reliability and validity assessment of the construct measures
Full size table
Table 2 Discriminant validity of the construct measures
Full size table

To minimize common method bias, we utilized a combination of procedural and statistical remedies (Podsakoff et al. 2003). Procedural remedies were addressed by ensuring respondents’ anonymity, separating the measurement of the predictor and criterion variables and reducing the respondents’ evaluation apprehension. As statistical remedies, we issued Harman’s single-factor test in both the exploratory factor analysis and confirmatory factor analysis. As a result of factor analysis using unrotated factor solution, all measures did not converge to one factor. Moreover, one general factor did not account for the majority of the covariance among the measures (first factor eigenvalue = 43.7% of variance, second factor = 14.0%, third factor = 8.0%, and fourth factor = 7.0%). In addition, a confirmatory factor analysis was estimated in which all indicators included in the model were restricted to load on a single factor (Podsakoff et al. 2003). The fit indices obtained from this test showed a poor fit (e.g., NFI = .54; GFI = .42; RMSEA = .185), which indicated that common method bias did not pose a problem for this study (Podsakoff et al. 2003).

3.3 Mediation test results

As the next step of the data analysis, we performed a series of mediation tests. As a result of this work, we provided evidence of the mediation roles of empathy and self-efficacy between FSEs’ CSR perceptions (both internal and external) and perceived dysfunctional customer behaviors. Based on Preacher and Hayes’ (2008) approach using SPSS Process model, we conducted mediation tests for two independent variables (FSEs’ internal CSR perceptions and FSEs’ external CSR perceptions) and two mediating variables (FSEs’ empathy and FSEs’ self-efficacy), and one dependent variable (FSEs’ perceived dysfunctional customer behavior).

We conducted four separate mediation tests using a single mediator (PROCESS Model 4) with the bias-corrected bootstrapping confidence interval. The first mediation test was based on FSEs’ external CSR perceptions as an independent variable, FSEs’ empathy as a mediator, and FSEs’ perceived dysfunctional customer behavior as a dependent variable (see Table 3a). The result showed that all the coefficients were statistically significant in the expected directions. Also, the result showed that not only a direct effect of FSEs’ external CSR perceptions on FSEs’ perceived dysfunctional customer behavior, but also the indirect effect was significant. Specifically, the indirect effect of X (FSEs’ external CSR perceptions) on Y (FSEs’ perceived dysfunctional customer behavior) via M (FSEs’ empathy) was − 0.115 with the bias-corrected bootstrapping confidence interval of (− 0.203, − 0.027), which did not include the value of zero. Thus, the result provided evidence of mediation effect.

Table 3 Mediation test results
Full size table

The remainder of Table 3(b, c, and d) shows similar results. All the coefficients were statistically significant, and both direct and indirect effects were found significant, which provided evidence of a mediation effect of FSEs’ empathy and FSEs’ self-efficacy. Specifically, the indirect effect of X (FSEs’ external CSR perceptions) on Y (FSEs’ perceived dysfunctional customer behavior) via M (FSEs’ self-efficacy) was − 0.0861 with the bias-corrected bootstrapping confidence interval of (− 0.1689, − 0.012), which did not include the value of zero.

The indirect effect of X (FSEs’ internal CSR perceptions) on Y (FSEs’ perceived dysfunctional customer behavior) via M (FSEs’ empathy) was − 0.1336 with the bias-corrected bootstrapping confidence interval of (− 0.223, − 0.060). Again, this provided evidence of mediation effect of FSEs’ empathy. Finally, the indirect effect of X (FSEs’ internal CSR perceptions) on Y (FSEs’ perceived dysfunctional customer behavior) via M (FSEs’ self-efficacy) was − 0.111 with the bias-corrected bootstrapping confidence interval of (− 0.196, − 0.026), which did not include the value of zero.

Overall, these results confirmed our conceptual model with mediating effects of FSEs’ empathy and FSEs’ self-efficacy and provided empirical confidence to include these two constructs in the SEM model.

3.4 SEM results

Overall fit statistics suggested a good fit between the hypothesized model and the observed data. The proposed structural model exhibited acceptable levels of fit with χ2(391) = 735.40, GFI = 0.90; NFI = 0.91; CFI = 0.95; RMSEA = 0.061. Given the satisfactory fit of the structural model, the standardized coefficients were inspected. Table 4 lists the hypothesized test results.

Table 4 Hypothesis test results
Full size table

As proposed in H1a, the positive relationship between FSEs’ perceptions of external CSR and their empathy in the tested context was significant (γ = 0.40, t = 4.87, p < .001) and FSEs’ perceptions of internal CSR was found to be positively and directly related to their empathy in H1b (γ = 0.21, t = 2.39, p < .05). With respect to H2a, FSEs’ perceptions of the external CSR was found to be positively and directly related to their self-efficacy (γ = 0.18, t = 2.08, p < .05). Additionally, the positive relationship of FSEs’ perceptions of the internal CSR and self-efficacy in H2b was also significant (γ = 0.45, t = 4.89, p < .001). As proposed in H4, the negative relationship between FSEs’ empathy and perceived dysfunctional customer behavior was supported (β = − 0.27, t = − 3.20, p < .01). The proposed negative relationship between FSE’s self-efficacy and perceived dysfunctional customer behavior (H5) was shown to be significant (β = − 0.22, t = − 2.578, p < .05). Finally, in support of H6, FSEs’ perceptions of dysfunctional customer behavior directly influenced their revenge intentions (β = 0.54, t = 7.99, p < .001).

Following the procedures prescribed by Joreskog and Sorbom (2006), we next conducted a model invariance test for differences in the proposed different effects of FSEs’ perceptions of external and internal CSR on empathy and self-efficacy. The free model was compared with the constraint model in which each CSR stakeholder group perception was set equal to empathy and self-efficacy. As a result of analyzing the difference of χ2 value according to degree of freedom change, with respect to H3a, the difference in the effect of FSEs’ perceptions of the external CSR on empathy and self-efficacy was significant. Secondly, the free model that the FSEs’ perceptions of the internal CSR expected to bear a different impact on empathy and self-efficacy was better than the constraint model in H3b.

The model invariance test provided empirical evidence that the effects of FSEs’ external CSR perceptions on their empathy were stronger than the effects of FSEs’ internal CSR perceptions on this same relationship, with a significant difference in χ2 (Δχ2 = − 7.14, p = .008), supporting H3a. Furthermore, differences in internal CSR versus external CSR perceptions moderated their respective paths to self-efficacy (Δχ2 = − 5.19, p = .023). That is, the effects of FSEs’ internal CSR perceptions on their self-efficacy were stronger than the effects of FSEs’ external CSR perceptions on their self-efficacy. The model comparison results are shown in Table 5, and the difference between the values of each path coefficient is shown in Table 4.

Table 5 Hypothesis test: model comparison
Full size table

4 Discussion

This study examines the preventive mechanism of CSR and stakeholder perspectives within the context of boundary spanning emotional labor. This investigation offers insights regarding how FSEs’ perceptions of internal and external CSR stakeholder activities can help FSEs deal with dysfunctional customer behavior. FSEs’ CSR perceptions can indirectly weaken their perceptions of dysfunctional customer behavior, which in turn reduces FSEs’ revenge intentions toward customers displaying such behavior. Moreover, FSEs are influenced by the manner in which an organization treats its employees, customers, and other stakeholders. This result shows FSEs’ perceptions of CSR reduce negative emotional labor effects from dealing with abnormal customer conduct. It is important to note that the current model does not support a direct causal relationship between the absence of CSR and FSE revenge but rather the presence of CSR can strengthen empathy and self-efficacy. Empathy and self-efficacy help shield against FSEs’ retaliation in response to dysfunctional customer behavior.

The findings demonstrate that FSEs’ empathy and self-efficacy each play a significant mediating role in the association of FSEs’ perceptions of CSR and dysfunctional customer behavior. More specifically, the results suggest CSR can impact FSEs’ perceptions of dysfunctional customer behavior by simultaneously strengthening their empathy and self-efficacy, both of which better equip FSEs when dealing with unfriendly, demanding, aggressive, and sometimes even hostile customers. FSEs’ external and internal CSR perceptions showed differential effects of varying intensities on empathy and self-efficacy, respectively, which suggests a moderating effect of FSEs’ perceptions of CSR stakeholder initiatives on these respective mediating constructs. External CSR stakeholder initiatives aimed at benefiting customers and other groups of stakeholders residing outside the company have a stronger impact on customer empathy than on self-efficacy to perform work-related tasks. Internal CSR stakeholder activities are focused on directly benefiting employees of the service organization (e.g., policies for improving working conditions, employee education programs, and paying fair salaries). In terms of increasing FSEs’ sensitivity for seeing situations from the customer’s perspective, FSEs’ internal CSR perceptions are shown to have a stronger positive moderating influence on self-efficacy to gain customers’ credence than on their empathy toward customers.

4.1 Theoretical contributions

The current study expands our understanding on the value of CSR as a preventive precursor for improving FSE’s self-worth at their jobs as well as their customer orientation. This helps FSEs cope with dysfunctional customer behavior. We show how FSEs’ views of their company’s CSR performance impact their empathy toward serving dysfunctional customer behavior. FSEs are required to treat customers with empathy, even when dealing with unruly and disruptive customers (Hochschild 1983). This research study illuminates the pivotal roles of FSEs’ levels of empathy and self-efficacy for bridging the positive CSR associations shared by FSEs’ regarding their organization and helping FSEs cope with their negative perceptions of dysfunctional customer behavior.

In dealing with dysfunctional customer demands, CSR can play an important role in not only helping FSEs to think from and understand the position of their customers while providing customers with high standards of quality service. This study expands on extant CSR literature by showing a mediating process occurs between the FSEs’ positive associations of being part of a socially responsible enterprise and their perceptions of dysfunctional customer behavior. Much of the recent CSR research appearing in management journals has focused on the consumer effects of CSR initiatives and favorable returns of a company’s investment in CSR (De Roeck and Maon 2016). However, a few studies (e.g., Raub and Blunschi 2014; Kim et al. 2018) have begun to recognize that CSR can be considered as a viable strategy for employees who are coping with high emotional demands resulting from service delivery.

This research marks the first study to explore how the psychological mechanism of CSR differs by the FSE’s stakeholder perspective (e.g., internal versus external CSR stakeholders). The findings show that internal and external CSR focused initiatives have differential effects on FSEs’ empathy and self-efficacy. FSEs’ perceptions of external CSR have greater influence on empathy than self-efficacy, and perceptions of internal CSR produce a larger impact on FSEs’ self-efficacy than empathy. This in turn respectively reduces their perceptions of dysfunctional customer behavior. These results suggest the strategic role of CSR as a preventive mechanism in the context of FSEs’ boundary spanning emotional labor.

Based on social identity theory (Hogg and Abrams 1988), an FSE’s positive identification with their employer can help them sustain a positive self-definition, even when dealing with challenging work conditions. This study contributes to studies on the motivation of frontline service work (Nguyen and Leclerc 2011; Wilder et al. 2014). For successful service delivery, it is essential for FSEs to competently perform their work and remain motivated to solve problems with confidence. In this sense, this study suggests that FSEs’ CSR perceptions are strategically valuable in employee–customer interactions because CSR contributes to FSEs’ self-efficacy.

4.2 Managerial implications

Contemporary business practices call for service organizations to develop CSR strategies and apply them across key stakeholders. Though the intuitive importance of CSR to society is rarely questioned, the stakeholder value of CSR remains a largely unresolved managerial concern (Sen et al. 2006). As a hygiene factor, CSR serves as a preventive mechanism for service providers in order to reduce tension among their key stakeholders (Lacey et al. 2015). Especially in light of the wide variety of CSR initiatives that companies can engage in, managers should consider which social issues are most important to their employees.

It is a challenge for managers to design CSR strategies that directly benefit their employees (Bhattacharya et al. 2008; Lee et al. 2013). This finding offers particularly important implications for service managers responsible for supporting FSEs who deal with dysfunctional customers. Managers can lend greater support by investing in FSE training programs that raise both the customer orientation and self-efficacy of their work. Indeed, FSEs usually have practical coping tactics to alleviate the stress they experience when customers display deviant behavior (Echeverri et al. 2012). Furthermore, they expect organizational structures to support them in managing dysfunctional customer behavior (Gong et al. 2014). CSR can play an instrumental role in influencing FSEs who are struggling in stressful situations with dysfunctional customers.

This study also offers a valuable implication to enhance FSEs’ job performance. As reported by Gallup, 70% of frontline employees do not reach their full potential in their work performance (Korschun et al. 2014). In this sense, it is noteworthy that the benefit of CSR goes beyond simply encouraging FSEs to identify with an organization that does good things (e.g., employee-company identification). CSR can also enhance essential qualities required for FSEs to deliver quality service to customers, such as empathy and self-efficacy. Thus, companies should develop CSR initiatives as a means to enhance FSE performance in employee-customer interactions. Specifically, an important role for managers is to incorporate CSR stakeholder activities (i.e., external and internal CSR) to “job products” that are well-suited to the needs fulfillment of frontline employees.

Employee volunteerism may be a particularly effective philanthropic approach for strengthening FSEs’ CSR perceptions for both internal and external stakeholders. Employees regard having an opportunity to volunteer in their local community through the workplace as a valuable employee benefit. Employee volunteerism programs provide the opportunity for FSEs to serve their community, and thus contribute to the general well-being of society. As a hybrid internal–external CSR stakeholder initiative, employee volunteerism programs not only stand to improve FSEs’ confidence in their employer but also give them an intimate understanding of some of the things their organization is involved in that helps improve the social conditions of the communities in which the organization operates.

4.3 Limitations and directions for further research

The present study provides scholars and managers with new insights into how CSR impacts FSEs’ perceptions of dysfunctional customer behavior. However, this research should be evaluated in light of its limitations, which presents future research opportunities. First, the model’s hypotheses were tested using field survey data obtained from a large South Korean insurance company. More samples from other service industry contexts and from different countries should be investigated to further generalize the current findings. Second, the field work did not grant us robust controls over the series of empirical tests that can be achieved in experiential designed research studies. Because the data collection method was based on self-report measures from respondents, it is difficult to fully support strong evidence of causal effects. Future studies may need to clarify the causal relationship among the variables used in this study through controlled experiments via stronger manipulations among the variables of interest.

Researchers are encouraged to reexamine the hypothesized relationships over time. A longitudinal study examining the effect of FSEs’ CSR perception on dysfunctional customer behavior would expand on the present study, given the overall impact of various CSR activities are likely to last for longer time periods and be accumulated over a certain time period. It is also necessary to investigate the role of other mediating variables beyond empathy and self-efficacy for bridging the relationship between CSR and FSEs’ emotional labor (Van den Broeck et al. 2008). For example, future researchers are encouraged to consider examining other sources of employee motivations and prosocial behavior that CSR may affect employees’ psychological needs and impact their desire to help others (Bauman and Skitka 2012; Henkel et al. 2017).

The present study viewed customer empathy from a cognitive perspective. Future research could more thoroughly explore the multi-dimensional nature of empathy, including empathy as a personality trait (Kerem et al. 2001). Also, future research could explore the effect of individual FSE-related characteristics in relationship to their propensity for seeking revenge against customers. For example, the employees’ moral identification may moderate their respective sensitivity and reactions to dysfunctional customer behavior (Skarlicki et al. 2008). Finally, job characteristics may impact this study’s tested relationships, such as staff positions versus managers (Sok et al. 2019). Higher demanding jobs have been shown to enhance emotional exhaustion (Alarcon 2011).

In conclusion, CSR played a preventive role for supporting FSEs who routinely deal with unruly and disruptive customers. CSR initiatives aimed at internal stakeholders (i.e., employees) made a stronger impact on FSEs’ beliefs to successfully perform their jobs. CSR initiatives directed at helping charitable organizations and other groups of external stakeholders strengthened FSEs’ customer orientation. Much remains to be learned by examining CSR through a broader social science and service business perspective.