The hidden cost of customer incivility: why employees hide knowledge? Open Access

In today's knowledge-driven economy, organizational success increasingly depends on effective knowledge management (Fauzi, 2023; Peng et al., 2019), which requires a culture of knowledge sharing. To foster this culture, organizations invest in enhanced compensation, knowledge management systems, value-driven training, and collaborative work environments (Xiong et al., 2021; Connelly et al., 2012). Despite these efforts, knowledge-sharing cultures remain rare due to persistent knowledge hiding (KH), which continues to pose serious challenges for service industries (Zhao et al., 2024a). Over the past decade, KH has emerged as a central issue in knowledge management research (Zhao et al., 2024b). KH is defined as an intentional concealment of requested knowledge, manifesting in three forms: evasive hiding, playing dumb, and rationalized hiding. Evasive hiding involves providing incorrect information or delaying it, playing dumb is feigning a lack of understanding, and rationalized hiding entails justifying non-disclosure or shifting blame (Connelly et al., 2012). KH can significantly undermine an organization's financial, social, and operational performance (Zhao et al., 2024b; Offergelt et al., 2019), making it critical to identify its antecedents (Yuan et al., 2021).

Extant literature has primarily examined the antecedents of KH at the individual, organizational, and interpersonal levels. At the individual level, KH has been linked to factors such as psychological ownership (Ghani et al., 2020), perspective taking, reciprocity beliefs, self-efficacy (Yuan et al., 2021), and prosocial motivation (Škerlavaj et al., 2018). Organizational-level antecedents include justice (Ghani et al., 2020), politics (Kaur and Kang, 2023), injustice (Jahanzeb et al., 2021), and harmful conditions such as dehumanization (Muhammad and Sarwar, 2021), work overload (Kmieciak, 2024), incivility (Aljawarneh and Atan, 2018), bullying (Bari et al., 2023), and ostracism (Shah and Hashmi, 2019). Interpersonal level research has primarily focused on leadership styles such as exploitive (Guo et al., 2021), self-serving (Peng et al., 2019), ethical (Koay and Lim, 2022), and empowering (Yan and Teng, 2025), along with abusive supervision (Ghani et al., 2020). Recent studies on KH have begun to explore coworker-based factors, such as incivility and social undermining (Fatima et al., 2022). While literature acknowledges that KH can be both voluntary and situational (Rasheed et al., 2022; Anand et al., 2022), research has predominantly focused on internal drivers, leaving a gap in understanding the role of external factors (Rezwan and Takahashi, 2021; Fauzi, 2023).

Recent research emphasizes the growing role of extra-organizational factors and calls for examining the socio-psychological mechanisms through which contextual emotional experiences shape KH (Joshi et al., 2025; Xiong et al., 2021). KH is not merely transactional or cognitive but is fundamentally influenced by affective responses and social interactions (Connelly et al., 2012). Among these, customer-based stressors have emerged as a key focus, particularly customer mistreatment (Rasheed et al., 2022; Hayat et al., 2021). Scholars highlight that such external pressures can prompt employees to withhold knowledge from coworkers (Rasheed et al., 2022). A notable example is customer incivility (CI), defined as low-quality interpersonal treatment received by frontline employees (FLEs), including verbal abuse, unreasonable demands, and aggressive behavior (Kiffin-Petersen and Soutar, 2020). CI is a critical yet underexplored construct in understanding interpersonal dynamics related to KH. It provides a distinct lens to examine social-psychological mechanisms influencing knowledge behavior, helping to bridge a gap in knowledge management literature that often prioritizes structural or cognitive explanations (Garg et al., 2022; Joshi et al., 2025). Alarmingly, CI affects over 70% of service personnel, with 76% reporting frequent exposure (Agnihotri and Bhattacharya, 2024), highlighting its prevalence and the need for deeper investigation into its impact on organizational dynamics (Lages et al., 2023).

In service settings, employees are expected to maintain composure even during belittling encounters (Cheng and Liu, 2024), often suppressing their emotional responses. Adhering to the expectation of “the customer is always right,” even in the face of CI, intensifies emotional labor, which could lead to exhaustion, reduced performance, psychological withdrawal, and diminished well-being (Xie et al., 2023; Pu et al., 2024; Sliter et al., 2011). Despite this, CI has received little attention in knowledge management research and is increasingly recognized as an added emotional demand that results in negative affective responses in FLEs. As a form of external job stressor (Cheng et al., 2020), it leads to emotional exhaustion (EE), a resource-depleting condition that may prompt defensive responses (Grandey, 2000), such as KH. Therefore, it is crucial to understand the psychological mechanisms of how the interplay of CI and EE affects KH. Based on this observation, we pose our first two research questions: (1) What is the effect of CI and EE on KH? and (2) Does EE mediate the relationship between CI and KH?

Employee responses to CI differ based on individual emotional and interpersonal capabilities. Emotional intelligence (EI), a key personal resource, helps individuals manage emotional strain and mitigate emotional exhaustion (De Geofroy and Evans, 2017). Yet its role in the context of KH remains largely unexplored (Tian et al., 2022). EI contributes to understanding the social-psychological mechanism underlying KH by shaping how employees perceive and regulate emotional responses to interpersonal stressors. Employees with higher EI are less likely to engage in defensive behaviors such as KH when faced with incivility. As such, EI is an essential individual contextual factor that influences the CI-EE link to determine KH, and a key to understanding the social-psychological mechanism of KH. This leads to our third research question: (3) How does EI affect the relation between CI and KH?

To answer these research questions, this study uses the theoretical lens of Conservation of Resources Theory (COR) to understand the interplay of CI, EE, EI, and KH. COR theory contends that individuals strive to acquire and safeguard resources they value (Hobfoll et al., 2018). When stressors like CI threaten such resources, they may resort to protective behaviors such as KH to prevent further loss (Fatima et al., 2022). In this context, EI acts as a key personal resource that helps individuals regulate their emotional responses and buffer the impact of stress. By integrating CI, EE, and EI within the COR framework, this study provides a unified explanation for how external stressors translate into defensive knowledge behaviors and how individual resources can mitigate these effects.

The contributions of this study are four-fold. First, we respond to the calls on social-psychological aspects of KH, and introduce CI and EI as key drivers and moderators in KH. Second, it shifts the focus beyond internal workplace dynamics by identifying CI as an external, yet influential, driver of KH. Third, it advances understanding of the emotional pathway to KH by establishing EE as a mediating mechanism. Lastly, it highlights the buffering role of EI, offering a nuanced view of how individual emotional capabilities shape the social-psychological processes behind KH.

There has been a growing interest in understanding how incivility could determine KH; however, the understanding has been confined to workplace incivility (Ballekura and Vilvanathan, 2025; Iram et al., 2024; Shah and Hashmi, 2019) whereas the role of CI remains largely unexplored. CI is a frequent stressor for service employees (Andersson and Pearson, 1999), which includes rude remarks, dismissive gestures, or unrealistic demands. These daily microaggressions require employees to expend emotional labor to remain courteous and professional (Sliter et al., 2011). Supplementary materials (Table S1) summarize the conceptual differences between customer incivility and related construct. COR theory has been widely employed to explain personal conflicts in employee–customer service interactions (Hobfoll et al., 2018; Kim and Qu, 2019). It posits that individuals strive to acquire, retain, and protect valued resources such as emotional energy, self-esteem, and well-being (Hobfoll et al., 2018; Hobfoll, 1989). COR theory predicts that employees' effort for resource acquisition and conservation guides their attitudes and behaviors. A threat to a resource or its actual loss generates substantial emotional stress, compelling individuals to avoid potential future resource loss (Hobfoll et al., 2018). When individuals face stressors that threaten or deplete these resources, they experience strain and burnout. From the perspective of COR theory, such repeated interactions drain emotional resources (Hobfoll et al., 2018). In the service sector, employees must suppress their genuine emotional responses (e.g. frustration, anger) and instead engage in surface acting, which is emotionally and psychologically taxing (Grandey, 2000). When such depletion is sustained and unreciprocated by recovery opportunities, it results in EE (Maslach et al., 2001). EE refers to a prolonged state of mental and physical depletion resulting from sustained work-related or personal demands and ongoing stress (Xu et al., 2019). It reflects a feeling of being emotionally drained by one's occupational responsibilities. Repeated exposure to uncivil gestures accelerates what COR theory terms a “loss spiral,” wherein employees with fewer coping resources experience greater vulnerability to emotional strain (Hobfoll et al., 2018).

Building on the COR theory, prior research determines that customer verbal aggression (Grandey, 2000), customer-based social stressors (Dormann and Zapf, 2004), and customer interpersonal injustice depletes employee resources. Moreover, EE is triggered owing to employees' experience of resource depletion and a lack of adequate personal resources required for effective handling of the confronted stressors (Hobfoll et al., 2018). Empirical studies in service and hospitality contexts have demonstrated that CI significantly predicts EE, which in turn impacts turnover intentions and service performance (Sliter et al., 2011; Karatepe et al., 2021). Within the hospitality industry, where emotional labor is inherent to service delivery, understanding this relationship is critical for mitigating burnout and promoting employee well-being. Thus, based on these arguments and drawing upon COR, this study hypothesized:

According to COR theory, individuals are motivated to protect and preserve valued resources such as energy, time, emotional stability, and social capital. Employees experience a depletion of these resources when they are exposed to demanding or stressful workplace environments. EE, the core dimension of burnout, arises when individuals feel emotionally overextended and drained due to sustained job-related stressors (Maslach et al., 2001). It refers to a chronic psychological state characterized by feelings of being emotionally overextended, fatigued, and mentally drained due to sustained interpersonal demands (Wen et al., 2019). In the hospitality sector, FLEs routinely face emotionally demanding interactions, particularly with uncivil or aggressive customers (Sliter et al., 2011). Suggested by COR theory, individuals adopt defensive strategies when they perceive a threat to their remaining resources (Wright and Hobfoll, 2004), such as KH. In the face of knowledge requests, KH may serve as a self-protective response to emotional strain. As EE reflects a state of diminished emotional and physical capacity (Halbesleben and Bowler, 2007; Kammeyer-Mueller et al., 2016), individuals experiencing it may lack the cognitive and affective resources necessary to engage in knowledge sharing (Halbesleben and Bowler, 2007), therefore, KH serves as a resource conservation tactic from the COR perspective. Rather than expend further cognitive or emotional effort, they protect their remaining resources by withdrawing from knowledge exchange.

Empirical evidence supports this notion; for example, Lee et al. (2018) found that EE employees are less willing to share knowledge. Given that knowledge itself is a valued resource, its concealment may be further amplified when employees are resource-deprived (Škerlavaj et al., 2018). Moreover, an emerging body of knowledge has started exploring how EE could lead to KH (Islam and Chaudhary, 2024; Zhao and Jiang, 2022). These studies conclude that emotionally exhausted individuals often lack the resources to engage in knowledge sharing, prompting self-protective behaviors such as KH (Zhao and Jiang, 2022). This positions EE as a key driver of dysfunctional knowledge behavior in organizations. Based on COR theory and the above arguments, we propose the following hypothesis:

According to the COR theory Hobfoll (1989), Individuals strive to retain and protect valuable resources, including emotional energy, cognitive capacity, and social capital, when threatened by stressors. In service contexts, CI represents a frequent and unpredictable stressor that depletes employees' emotional and psychological resources (Sliter et al., 2011; Karatepe et al., 2021). Prolonged exposure to CI can result in EE, characterized by fatigue, detachment, and reduced coping capacity (Maslach et al., 2001). As a signal of resource depletion, EE may prompt employees to engage in KH as a defensive mechanism to conserve remaining resources (Halbesleben and Bowler, 2007). Employees may rationalize KH to counterbalance negative customer interactions and maintain self-worth and personal control (Koon and Pun, 2018). Consequently, CI constitutes a resource-threatening situation that emotionally drains employees, increasing the likelihood of counterproductive behaviors such as KH. Research indicates that emotionally exhausted employees are less willing to share knowledge, as doing so requires additional emotional and cognitive investment (Lee et al., 2018; Zhao and Jiang, 2022). EE thus serves as a key mediating mechanism linking CI to resource-conserving behaviors like KH.

EE has been widely established as a mediator in the relationship between undesirable workplace situations (e.g. workplace bullying, workplace gossip, workplace incivility, exploitive leadership, abusive supervision, and workplace ostracism) and negative work outcomes (e.g. interpersonal deviance, counterproductive behavior, and turnover intentions) (Chou et al., 2020; Kim et al., 2025). However, limited research has examined how CI provokes counterproductive behaviors toward coworkers, such as KH (Rasheed et al., 2022). Extending prior work, this study positions EE as the psychological mechanism connecting CI to KH. Drawing on COR theory, we argue that CI depletes employees' socioemotional resources, heightening EE and increasing their propensity to withhold knowledge. Based on this reasoning, we propose the following hypothesis:

Individuals with greater personal resources are better able to withstand the emotional toll of stressors, whereas those with fewer resources are more vulnerable to a “loss spiral,” in which repeated exposure to incivility accelerates EE. EI, the ability to perceive, understand, regulate, and use emotions effectively (Mayer, 2003), constitutes a critical personal resource that can buffer the detrimental effects of CI. Employees high in EI are better able to manage negative emotions during hostile customer interactions, reducing susceptibility to EE.

Prior research has examined psychological capital, optimism, political skill, and benevolence as buffers against workplace stressors (Jahanzeb et al., 2021; Kaur and Kang, 2023; Yuan and Yan, 2025). These constructs primarily reflect cognitive appraisals, motivational orientations, or social influence capabilities. In contrast, CI in hospitality represents an emotionally charged, interpersonal stressor that directly depletes employees' affective resources through sustained emotional labor (Hochschild, 1983). EI is therefore theoretically more proximal to EE, the key mechanism linking CI to KH, than broader personal resources. Despite this relevance, EI has received limited attention as a moderating resource in KH research (Garg et al., 2022; Tian et al., 2022). By foregrounding EI, this study advances an emotion-centered account of how frontline employees regulate resource loss and maintain cooperative knowledge behaviors under CI.

Empirical evidence demonstrates that EI mitigates stress responses and emotional depletion in service contexts (Karatepe et al., 2021). Its buffering role has been documented across various stressors, including burnout and emotional labor (Prati and Karriker, 2010), interpersonal conflict and counterproductive behaviors (Jordan et al., 2002), job insecurity (De Geofroy and Evans, 2017) and emotional job demands (Grover and Furnham, 2021).

Consistent with COR theory, this study posits that EI attenuates the negative impact of CI on EE (Hobfoll et al., 2018). Employees high in EI can regulate emotions and conserve resources during CI, reducing the likelihood of engaging in KH. Conversely, employees low in EI are more prone to emotional depletion and may resort to KH to protect remaining resources. Thus, EI moderates both the direct effect of CI on EE and the indirect effect of CI on KH via EE, limiting the translation of emotional strain into defensive knowledge behaviors (Lee et al., 2018). Based on these arguments, the following hypotheses are proposed:

The conceptual model of the study is shown in Figure 1.

KH is a global issue reported in North America, the Middle East, Turkey, the UK, and the US (Lee, 2022), but it is particularly prevalent in Asia, where power distance and collectivist norms amplify its effects (Shah and Hashmi, 2019). Evidence from China, India, and Pakistan highlights KH as a serious organizational concern (Jahanzeb et al., 2021; Abdullah et al., 2019). The service sector, characterized by task interdependence and emotional labor, is especially vulnerable (Ghani et al., 2020). In Pakistan, frequent customer-based stressors such as aggression, mistreatment, and verbal abuse toward frontline employees make it a compelling context for examining the emotional and interpersonal antecedents of KH (Nawaz et al., 2020), particularly in hospitality, where knowledge sharing is critical to service quality and customer trust (Lee et al., 2018).

To empirically test the proposed hypotheses, this study employed a two-wave, time-lagged survey design among FLEs in the hospitality sector of Pakistan. This design minimizes common method bias (CMB) and social desirability bias by separating the measurement of predictor and criterion variables across different time points (Podsakoff et al., 2003, 2012). The temporal ordering of variables enables a better capture of the hypothesized causal sequence, strengthening internal validity (Zhao and Xia, 2019). A four-week interval was used, allowing sufficient separation to reduce short-term memory effects while minimizing participant attrition.

This study received formal ethics approval (Date: 1st July 2025/No: 03-ACD-02-AI) from Lahore Leads University and complied with internationally accepted guidelines for research involving human participants. Procedures followed ethical principles, including voluntary participation, confidentiality, and participant welfare. Participants were informed of the study's purpose, their right to refuse or withdraw at any time, and provided informed consent before completing questionnaires. Responses were anonymous and confidential; no personally identifiable information was collected. Unique codes were used to match two-wave responses while preserving anonymity. The study posed minimal risk, involving only self-reported perceptions of workplace experiences without psychological, physical, or financial harm.

This study employed convenience sampling to survey frontline hospitality employees in Pakistan. This approach is widely used in service research because shift-based staff are difficult to access through probability sampling and have limited availability (Muhammad and Sarwar, 2021). Since FLEs interact directly with customers and are most likely to encounter CI and emotional labor, convenience sampling was both practical and theoretically appropriate for the study's focus. The study targeted FLEs in hotels across Lahore, Islamabad, and Karachi, Pakistan's major hospitality markets. A hotel list was obtained from the Pakistan Hotel Association, from which the top 15 chains and branches were approached with managerial approval for voluntary participation.

G*Power 3.1.9.7 indicated a minimum sample size of 146 (Hahs-Vaughn and Lomax, 2020). The final matched sample comprised 234 responses, exceeding the required threshold and ensuring adequate statistical power.

Data were collected in two waves following ethical standards of informed consent, anonymity, confidentiality, and voluntary participation. Participants were informed about the two-stage design in advance. A pilot study with 15 hospitality professionals and 3 academicians confirmed item clarity and contextual relevance, leading to minor wording refinements. In Wave 1,450 questionnaires measuring demographics, CI, and EE were distributed, yielding 325 valid responses (72.2%). After four weeks, Wave 2 measured EI and KH, resulting in 234 matched responses (71.9% retention), paired using unique codes to maintain anonymity. Follow-ups were conducted through telephonic calls, emails, and personal visits to increase the response rate. The demographic profile of respondents and hotels is presented in Table 1 and Table 2, respectively.

All constructs were measured using validated scales on a five-point Likert scale (1 = strongly disagree, 5 = strongly agree). CI was measured using (Wang et al., 2011) 4-item scale. Emotional exhaustion was assessed with the 9-item Maslach Burnout Inventory subscale (Maslach et al., 1997). EI was measured using the scale of (Wong and Law, 2002), covering self-emotion appraisal, others' emotion appraisal, use of emotion, and regulation of emotion. KH was measured using the 12-item scale by Connelly et al. (2012), capturing evasive hiding, playing dumb, and rationalizing hiding.

Data were analyzed using AMOS 24, following a covariance-based structural equation modeling (CB-SEM) approach suitable for theory testing (Hair et al., 2017). Preliminary screening was conducted to check for missing values, outliers, and normality. A two-step procedure (Anderson and Gerbing, 1988) was employed, whereby confirmatory factor analysis (CFA) was first used to assess the measurement model in terms of reliability and validity, followed by structural model analysis to test the research hypotheses.

Multiple procedural and statistical techniques were applied to assess common method bias (CMB). Harman's single-factor test showed the first factor accounted for 30.2% of variance, below the 50% threshold (Podsakoff et al., 2012). CFA of a single-factor model indicated poor fit (GFI = 0.704; AGFI = 0.648; NFI = 0.639; IFI = 0.685; TLI = 0.652; SRMR = 0.170; RMSEA = 0.118), suggesting CMB was not a major concern. The Common Latent Factor test further confirmed this, as differences in standardized loadings were below 0.20, further confirming that CMB did not pose a substantive threat to the validity of the findings.

Confirmatory factor analysis using AMOS 24 was conducted to assess reliability and validity, consistent with the theory-driven CB-SEM approach (Hair et al., 2017). Items with standardized loadings of 0.50 or higher were retained (Fornell and Larcker, 1981; Anderson and Gerbing, 1988). The Cronbach alpha and Composite reliability (CR) values for all constructs exceeded the recommended minimum of 0.70 (Fornell and Larcker, 1981) and average variance extracted (AVE) values were greater than the recommended threshold of 0.5 validity (Hair et al., 2019). For all cases, AVE was greater than the maximum shared variance (MSV), thereby confirming discriminant validity (Hair et al., 2019). The measurement model demonstrated an excellent fit to the data (χ²(428) = 636.958, χ²/df = 1.488, CFI = 0.958, SRMR = 0.047, RMSEA = 0.046, PClose = 0.825) (Hu and Bentler, 1999). Details of the measurement model results are shown in Table 3.

Structural equation modeling was subsequently employed to test the hypothesized relationships. CI was positively and significantly associated with EE (β = 0.569, p < 0.001), indicating a large effect size and supporting H1. This suggests that increases in customer incivility are associated with substantial emotional exhaustion among frontline employees. EE was positively associated with KH (β = 0.272, p < 0.001), reflecting a moderate effect size and supporting H2. Mediation analysis indicated that EE significantly mediated the association between CI and KH (β = 0.155, p < 0.001), thereby confirming H3 as shown in Tables 4 and 5. To reduce omitted variable bias, demographic controls (gender, age, education, and experience) were included with direct paths to knowledge hiding. None were significant, and their inclusion did not alter the hypothesized relationships.

Moderation analysis showed that EI significantly moderated the CI–EE relationship. The interaction term was negative and significant (β = −0.362, p < 0.001), indicating that higher EI attenuates the effect of CI on EE, thus supporting H4. The effects of CI on EE at high and low levels of EI are shown in Figure 2.

Finally, moderated mediation analysis was conducted using PROCESS Model 7 (Hayes, 2012) with CI as the independent variable, EE as the mediator, KH as the dependent variable, and EI as the moderator. Results revealed a significant CI × EI interaction (β = −0.2245, SE = 0.0536, t = −4.1912, p < 0.001, LLCI = −0.3300, ULCI = −0.1189). Conditional indirect effects indicated that the mediation of CI → EE → KH was strongest at low EI and weakest at high EI. The index of moderated mediation was significant (Index = −0.0242, BootSE = 0.0119, BootLLCI = −0.0514, BootULCI = −0.0043), supporting H5 (Table 6). These findings demonstrate that EI buffers both the direct and indirect effects of CI on KH.

This study addresses an underexplored dimension of workplace dysfunction by showing how an extra-organizational stressor, CI, triggers intra-organizational counterproductive behavior, KH, via EE. Prior KH research has primarily examined internal stressors such as abusive supervision, ostracism, and workplace bullying (Chatterjee et al., 2021; Khalid et al., 2022). By shifting the focus outward, this study demonstrates how customer behavior shapes internal workplace dynamics. It answers calls to examine the socio-psychological mechanisms underlying KH (Xiong et al., 2021; Joshi et al., 2025) by explaining how external aggression undermines coworker cooperation in service contexts.

Grounded in COR theory (Hobfoll, 1989), we proposed that CI increases KH, EE mediates this relationship, and EI attenuates it. As predicted, CI positively influenced EE. Hospitality employees facing uncivil customers experienced emotional depletion from regulating their reactions. Consistent with COR, resource loss prompted resource-conserving behaviors such as KH. Although employees suppressed negative responses toward customers due to power asymmetry, emotional exhaustion limited their ability to collaborate with coworkers. This extends KH research by showing that external stressors can drive intraorganizational dysfunction, complementing studies on internal stressors (Chatterjee et al., 2021; Bhatti et al., 2023; Khalid et al., 2022) and recent work linking customer mistreatment to KH through revenge attitudes (Rasheed et al., 2022; Hayat et al., 2021).

EE emerged as the central mechanism mediating the CI-KH link. While prior research identified EE as a mediator between internal stressors and KH, this study extends that logic to customer-induced stressors, underscoring affective depletion as critical in shaping knowledge behaviors in service-intensive roles.

EI further clarifies boundary conditions by attenuating the positive CI-KH relationship. Employees with higher EI were less likely to engage in KH under CI. EI facilitates effective emotion regulation (Mayer, 2003), enabling constructive appraisal of negative encounters and maintenance of psychological balance, thereby reducing EE, the mechanism linking CI to KH. Moreover, emotionally intelligent employees show greater empathy and responsiveness to coworkers (Davies et al., 1998), conserving emotional resources and reducing frustration that might otherwise lead to knowledge withholding. While prior research shows EI moderates customer-related stressors and negative outcomes (Grover and Furnham, 2021; Liao et al., 2022), its buffering role in KH remains limited (Tian et al., 2022). This study demonstrates that EI mitigates the translation of external emotional stressors into intraorganizational counterproductive behaviors.

Overall, these findings clarify the mechanisms and boundary conditions linking external stressors, emotional depletion, and KH, offering meaningful theoretical and practical implications for high-contact service environments.

This study advances the literature on CI and KH in hospitality contexts through several interrelated contributions. First, it identifies CI as a salient extra-organizational antecedent of KH, whereas prior research focused mainly on internal interpersonal stressors (Bhatti et al., 2023; Yao et al., 2020; Khalid et al., 2022). The findings show that CI spills over into coworker relationships, promoting knowledge hiding, thereby expanding the nomological network of KH and addressing calls to examine socio-psychological mechanisms beyond organizational boundaries (Joshi et al., 2025; Xiong et al., 2021).

Second, the study establishes EE as the key mechanism linking CI to KH. While EE has been shown to mediate relationships between internal stressors, such as role stress (Zhao and Jiang, 2022) and workplace bullying (Bhatti et al., 2023), this research extends that logic to extra-organizational stressors, demonstrating that CI depletes employees' emotional resources and triggers defensive knowledge behaviors.

Third, EI is identified as a critical moderator buffering the CI–KH link. Prior research has examined personal resources such as optimism, benevolence, and political skill (Yuan and Yan, 2025; Jahanzeb et al., 2021; Kaur and Kang, 2023), but emotion-based abilities remain underexplored (Garg et al., 2022; Tian et al., 2022). By theorizing and testing EI, the study extends the personal-resources perspective and positions emotional competence as a protective boundary in high-contact hospitality contexts. Employees with higher EI better perceive and regulate emotions (Mayer, 2003), manage negative customer encounters constructively, and remain responsive to coworkers (Davies et al., 1998), reducing EE and the likelihood of KH.

Fourth, the study contributes to a contingency perspective on CI, demonstrating that its harmful effects vary across individuals. Emotional competencies determine whether exposure to CI escalates into harmful behaviors, refining CI models and underscoring how individual differences shape intraorganizational outcomes of external stressors (Anand et al., 2022; Fauzi, 2023; Lages et al., 2023).

Beyond these contributions, the study advances COR theory by conceptualizing CI as a resource-depleting stressor that initiates a loss process culminating in defensive behaviors such as KH (Hobfoll, 1989). KH is framed as a resource-conservation behavior enacted under emotional depletion. The study extends COR in two ways: first, it shows that resource loss from customers shapes internal organizational behaviors, broadening COR beyond internal stressors. Second, by identifying EI as a personal resource that interrupts the loss spiral, it explains why employees exposed to similar stressors respond differently.

Finally, situated in Pakistan's hospitality sector, the study highlights the cultural context. Its collectivist orientation and high-power distance (Shah and Hashmi, 2019; Hofstede, 2001) may intensify the perception of CI as norm violations, heightening EE and KH. These dynamics underscore the importance of accounting for cultural and institutional factors, and cross-cultural replications are encouraged to test boundary conditions in other hospitality contexts.

This study offers focused managerial implications. The findings show that CI leads to KH through EE, indicating that KH is a stress-induced defensive response rather than merely an attitudinal issue. Thus, managers should treat KH as a signal of emotional depletion and implement low-cost monitoring tools such as pulse surveys, emotional check-ins, or CI incident logs to enable early intervention before exhaustion escalates into dysfunctional knowledge behaviors. Beyond individual coping, organizational-level action is essential: firms should design service systems with clear customer conduct policies, visible managerial support during incivility incidents, and empowerment protocols that allow employees to disengage from persistently abusive customers, thereby conserving emotional resources; supportive supervisory climates and peer support structures can further replenish depleted resources and discourage KH. Although EI buffered the CI-KH relationship, it should not be viewed as a standalone solution but as a complementary personal resource within a broader resource-management strategy; managers may incorporate EI into recruitment and role placement for high-contact roles while offering targeted development focused on emotion regulation and post-encounter recovery, avoiding unrealistic expectations that training alone can offset structural stressors. Finally, in collectivist and high power-distance contexts such as Pakistan, CI may be perceived as a violation of social norms, intensifying emotional strain; managers in similar settings should prioritize culturally calibrated, emotion-focused recovery practices, including structured debriefings, social sharing opportunities, and psychologically safe spaces, to help employees process negative service encounters and prevent KH.

While this study employed a robust two-wave time-lagged design to mitigate CMB and enhance causal inference, several limitations remain. First, data were collected from hospitality employees in Pakistan, potentially limiting generalizability due to cultural and industry-specific factors. Future research should replicate these findings across diverse cultural contexts and sectors to improve external validity. Second, although EI was examined as a key moderator, other personal and organizational factors, such as resilience, social support, or organizational culture, may also influence the relationships among CI, EE, and KH. Future studies could explore these additional moderators and mediators. Moreover, EI was treated as a unified construct; examining its four dimensions, self-awareness, self-regulation, social awareness, and relationship management could reveal which aspects most effectively buffer KH under CI. Third, reliance on self-reported data may introduce response biases; incorporating multi-source or objective measures, such as supervisor ratings or behavioral observations, could strengthen validity. Fourth, while the two-wave design supports temporal precedence, longitudinal or experimental designs with multiple time points would offer deeper insights into causal relationships and the dynamics of EE and KH over time. Finally, convenience sampling, though facilitating access to frontline hospitality employees across multiple hotel types and cities, may limit statistical generalizability. Future research should consider probability-based or multi-site sampling to enhance external validity.

The supplementary material for this article can be found online.