As the worlds’ population ages, more frontline employees (FLEs) will be required in the aged care industry. Notably, workers at residential aged care services are susceptible to workplace stress (Biggins, 2019) due to the long hours, low wages, and physical and emotional demands of helping the elderly and incapacitated (Proust, 2019). This poses challenges for the industry to manage the welfare of its employees.

Introduction Frontline employees (FLEs) play a very important role in service delivery due to the interactive nature of the service encounter. They span the boundary between the firm and its customers and service firms rely on their FLEs to deliver their promise and create a favorable company image to customers (e.g., Bitner et al., 1990). Considerable previous research addresses how effective management practices and supportive work environments will induce positive attitudinal and behavioral responses of FLEs, which will, in turn, contribute to the positive customer perceptions of the service providers (e.g., Hartline and Ferrell 1996). In other words, the relationship quality or the exchange quality of the employee-organization interface will have a spillover effect on that of the customer-organization interface. But the question examined here is: can this spillover effect occur in an opposite way? In other words, will firms’ treatment of customers shape the employees’ relationship with their firms? With a few exceptions, limited research has paid attention to this inverse relationship. The purpose of our research is threefold: (1) to investigate whether employee perception of customer injustice can influence employees’ psychological contract violation with the firm, (2) to examine whether role conflict mediates the relationship between customer injustice and psychological contract violation, and (3) to explore the moderating impacts of customer identification on the mediation effect of role conflict. Method We conducted an experiment using a 2 (customer injustice: high vs. low) x 2 (customer identification: high vs. low) between-subjects factorial design. Two hundred participants were recruited from Amazon Mturk. Twelve responses were deemed unusable and excluded from the study, resulting in a final sample size of 188 (53.7 % female; age ranging from 18 to 65). Each participant was randomly assigned to one of the four experimental scenarios that corresponded to a combination of the two manipulated factors at either high or low level. All manipulations worked as intended. To analyze the moderating effect of customer identification via role conflict, we used the procedure of Hayes (2013) to estimate a conditional process model. We also controlled for the effects of empathy, income, and ethnicity. To test whether the indirect effect of injustice on contract violation is moderated by customer identification, an index of moderated mediation proposed by Hayes (2014) was calculated. To test whether this index is statistically significantly from zero, a 95% confidence interval was calculated for this index by bootstrapping 5,000 samples. The confidence interval of this index is .0162 to 1.1105, indicating the indirect effect is significantly moderated by identification. The results showed that the indirect effect of customer injustice via role conflict on contract violation is only significant (p< .05) when customer identification is high. In other words, when customer identification is low, the effect of injustice on contract violation is not mediated through role conflict. Research implications Our research provides empirical evidence that FLEs are sensitive to the treatment of customers by the firm. The traditional wisdom in the sales literature is that “if you treat your employees well, they will treat your customers well.” Our study complements this “trickle-down effect” in the extant literature and demonstrates a “bottom-up effect” that the firm’s unfair treatment of customers will adversely influence employees’ relationship with their firm. Our research also offers important insights into why customer injustice may lead to FLEs’ perceived psychological contract violation with the firm. Previous sales research suggests that role conflict can be influenced by an organization’s structure and culture as well as salespeople’s job characteristics (Singh 1998; Barnes et al. 2006). Our study complements these findings and identifies perceived customer injustice as a new role stressor of FLEs. In addition, our research reveals that the mediating effect of role conflict is moderated by customer identification. Customer identification increases the likelihood that customer injustice would manifest in a psychological contract violation via increased role conflict. The findings of this research also have several managerial implications. First, service and sales managers should be aware of the negative consequence of unfair customer treatment by the firm and how it may eventually jeopardize employees’ relationship with the firm. Second, managers should consult with their FLEs when implementing any new customerfacing policies to understand how these policies would impact FLEs’ other duties of serving customers. Finally, FLEs may form strong identification with their customers, which may amplify the negative consequence of customer injustice on psychological contract violation. Managers should try to counteract FLEs’ over-identification with customers by increasing organizational identification.

This study aims to extend previous research on the structural relationships between organisational empowerment and frontline employees’ behaviors，and explores the role of the self-regulating process and its impact on service recovery performance by using fuzzy-set qualitative comparative analysis (fsQCA). Following the procedure of applying fsQCA (data calibration, truth table construction and fsQCA analysis) on the data of express mail firms in China, the results explore the combination of several causes as solutions that lead to both high and low service recovery performance. The findings show that the organisational empowerment is a sufficient antecedent for high service recovery performance, especially in case where frontline employee have full awareness and positive work involvement. Moreover, in the context of organisational empowerment, a reasonable level of emotional exhaustion represents a positive impact on performance in service recovery. This study offers more comprehensive insights for practitioners to empower frontline employees and monitor their emotions and behaviors using more appropriate approaches.