The effect of other in-group members’ organizational citizenship behavior on employees’ organizational deviance: a moral licensing perspective

Purpose – The purpose of this study is to investigate the moral licensing effect of other in-group members’ organizational citizenship behavior (OCB) on focal employees’ organizational deviance through moral selfconcept. This paper also examines the moderating role of in-group identification in the mediated relationship. Design/methodology/approach – The multilevel path analysis and bootstrapping technique are employed to analyze the findings of a sample of 340 employees in 56 workgroups in Vietnam. Findings – The results demonstrate that moral self-concept mediates the positive relationship between other in-group members’ OCB and focal employees’ organizational deviance. Furthermore, the findings indicate that in-group identification strengthens the indirect effect of other in-group members’ OCB on focal employees’ organizational deviance via moral self-concept. Practical implications – The findings suggest that managers should be aware of the potential negative consequences of OCB and the drawbacks of in-group identification in group contexts. In addition, practitioners should proactively prevent other in-group members’ OCB from resulting in employees’ organizational deviance. Originality/value – This is the first study to examine the moral licensing effect of OCB on organizational deviance through the moral self-concept mechanism and the moderating role of in-group identification in this mediated relationship.

in a vicarious licensing effect (i.e., individuals receive a moral license from other in-group members' ethical behavior and tend to behave unethically afterward). Drawing on moral licensing theory (Miller and Effron, 2010) and the vicarious moral licensing model (Kouchaki, 2011), this study aims to investigate the ethical decisions of employees in the context of other in-group members' behavioral history. Specifically, the present research investigates the vicarious moral licensing effect of other in-group members' organizational citizenship behavior (OCB) on employees' organizational deviance through moral self-concept. This study also examines in-group identification as a boundary condition of this mediated relationship.
Individuals with deep ethical beliefs and values have great potential for a positive influence on their own decisions and behaviors that enhance long-term effectiveness and corporate sustainability (Bernal and Edgar, 2012;Epstein, 2008;Middlebrooks et al., 2009;Nguyen et al., 2013). Additionally, Park and Shaw (2013) have argued that "developing OCB within each individual organization's context should be a focal point for any organization's success, especially if the organization wants to achieve its sustainability goals" (as cited in Lee and Ha-Brookshire, 2018, p. 940). Importantly, Lee (2020) found that OCB increased corporate sustainability performance; conversely, counterproductive work behavior (CWB) degraded it. Hence, given an overall business and economic sustainability approach, it is crucial to prevent OCB from resulting in organizational deviance in terms of its impact on business sustainability. The present study contributes to the development of a more complete understanding of how and when other in-group members' OCB may lead to employees' organizational deviance. These findings offer an empirical basis for organizations to actively prevent OCB from resulting in organizational deviance, thereby improving business sustainability.
The current study aims to make several contributions to existing research on OCB, the moral licensing literature and social identity theory. First, this research contributes to the OCB literature by exploring the potential negative consequences of OCB in a group context. Drawing on the effect of vicarious moral licensing, the author expects that other in-group members' OCB will license employees' organizational deviance. Although the positive relationship between OCB and organizational deviance at the individual level has been examined in previous empirical research (e.g., Yam et al., 2017), to the best of the author's knowledge, the current study is the first in the literature to examine the positive association between OCB and organizational deviance at the group level.
Second, this study contributes to moral licensing theory by examining the underlying mechanism in the relationship between other in-group members' OCB and focal employees' organizational deviance. Based on the vicarious self-concept theory (Goldstein and Cialdini, 2007), this study proposes that moral self-concept, defined as "an actor's self-perception of being ethical" (Ashforth and Lange, 2016, p. 2), mediates the vicarious moral licensing effect of other in-group members' OCB on focal employees' organizational deviance. Understanding the mediating role of moral self-concept advances our understanding of the psychological processes underlying the vicarious moral licensing effect.
Finally, the current research provides evidence of the potential drawbacks of strong group identification. This study examines in-group identification as an important contextual boundary condition of the impact of other in-group members' OCB on employees' organizational deviance via moral self-concept. Drawing on social identity theory (Tajfel and Turner, 1979) and vicarious self-concept theory (Goldstein and Cialdini, 2007), this study examines the potential detrimental effect of in-group identification on employees who have observed other in-group members' OCB. As such, a high level of in-group identification combined with other in-group members' OCB can result in an increase in moral self-concept and subsequent organizational deviance among employees.
The hypotheses were tested using multiphase and multilevel data from 340 employees working within 12 branches of a mobile service company in Vietnam. Given that the measurement of the predictor (i.e., OCB of other in-group members) was at the group level, this study conceptualized in-group members as all employees working within the same department. Each department is treated as a distinct and permanent workgroup. Employees in each department have mutual trust and shared values. Additionally, they were arranged to sit in the same physical work environment, where they could have face-to-face interactions on a daily basis to directly observe and experience each other's OCB. Thus, this workgroup environment is an apt context to examine the vicarious moral licensing effect of group-level OCB on employees' organizational deviance. The following section discusses the conceptualization of the study constructs and provides the rationales for the theoretical model and hypotheses.

Literature review and hypotheses
Moral licensing theory Moral licensing theory (Miller and Effron, 2010) suggests that individuals who have engaged in moral actions feel licensed to commit future immoral actions without concern about discrediting themselves (Klotz and Bolino, 2013;Yam et al., 2017). For example, if people have recently made a large donation to one charity, they may have a sense that they can deny another charity's request for a donation. Miller and Effron's (2010) study reveals two possible explanations for the moral licensing effect: the moral credits and moral credentials models. In the moral credits model, previous good deeds can establish moral credits in one's personal moral account, and they therefore can spend those credits to commit bad deeds without any fear of penalties. People can earn moral credits when they donate money to one charity, and those moral credits can license them to deny another donation request with no feelings of selfishness. In this model, the licensing effect via moral credits does not change the meaning of the morally questionable actions, which is the main difference from the second model. The moral credentials model of licensing has its roots in the casual attribution theory; previous moral actions change the meaning of morally ambiguous behavior (Miller and Effron, 2010). Others can interpret one's current ambiguous behavior based on the positive reputation created by past good deeds. That ethical background changes the meaning of the current ambiguous actions -"past behavior serves as a lens through which one construes current behavior, and when the motivation for current behavior is ambiguous, it is disambiguated in line with past behavior" (Merritt et al., 2010, p. 349). In other words, prior good deeds establishing credentials function like desirable character witnesses for individuals, such that these character witnesses make subsequent ambiguous behavior seem less immoral. For instance, always leaving a tip in a restaurant establishes one as a generous person. Subsequently skipping a tip may appear less selfish because this action came from a bountiful person. Similarly, previous unprejudiced behavior establishes an individual's credentials as an unbiased person, such that moral credential make subsequent discriminatory attitude seem less prejudiced.
Moral licensing has received increased attention from scholars in recent years, and numerous empirical studies have provided support for the moral licensing effect. Sachdeva et al. (2009) found that participants who wrote a story referring to their positive traits chose to donate less money to charity and to pollute the environment more than those who wrote a story focusing on their negative traits. In two studies by Zhong et al. (2010), participants were asked to make choices in a series of 12 moral dilemmas. The participants tended to make a more unethical decision after making a praiseworthy choice in the previous dilemma. Similarly, customers who chose environmentally friendly products over conventional products expressed an increased desire to cheat and steal in a paid task (Mazar and Zhong, 2010). In general, this stream of research supports the idea that good deeds may boost one's moral self-concept, which then licenses subsequent bad deeds (Sachdeva et al., 2009).

A moral licensing perspective
The moral licensing effect has been applied in the organizational setting. For instance, Klotz and Bolino (2013) have proposed that OCB grants employees a moral license to engage in subsequent CWB. Lin et al. (2016) found that engaging in ethical leadership behavior was positively related to an increase in abusive behavior via increasing moral credits. In addition, Yam et al. (2017) provided support for the credentials mechanism (i.e., psychological entitlement) in the moral licensing effect. Specifically, they found that employees who engage in OCB because of an external motivation can have a feeling of entitlement, and that feeling can license employees to engage in both interpersonal and organizational deviance. Recently, Kong et al. (2020) demonstrated that moral credentials mediated the moral licensing effect of work effort on unethical pro-organizational behavior. These investigations provide evidence for the moral credits and moral credentials mechanisms of the moral licensing effect.
Organizational citizenship behavior, organizational deviance and moral self-concept Organizational citizenship behavior refers to employees' discretionary actions that are outside their formal job description and that contribute to the effective social and psychological functioning of the organization (Organ, 1997;Organ et al., 2006). Deviance is defined as "voluntary behavior that violates significant organizational norms and in so doing threatens the well-being of an organization, its members, or both" (Robinson and Bennett, 1995, p. 556). Previous research has demonstrated a positive relationship between OCB and deviant behaviors (Bolino and Klotz, 2015). For instance, Klotz and Bolino (2013) have suggested that morally praiseworthy OCB may lead to CWB. Additionally, when employees are compelled to engage in OCB, they may experience a sense of entitlement, which licenses subsequent organizational deviance (Yam et al., 2017). Despite the valuable insights of prior research on the moral licensing effect of OCB on organizational deviance, our understanding is incomplete without knowledge of "how the presence of teams and groups in organizations may affect this relationship" (Klotz and Bolino, 2013, p. 303). In an effort to address this gap in the literature, the present study examines the vicarious moral licensing effect of other ingroup members' OCB on focal employees' organizational deviance through moral selfconcept.
Moral self-concept refers to "one's perceived moral standing at any given moment" (Monin and Jordan, 2009, p. 341). An individual's history of good or bad deeds has an impact on his or her moral self-concept: good deeds elevate moral self-concept; conversely, bad deeds deflate it (Sachdeva et al., 2009;Zhong et al., 2009). When individuals engage in good or bad deeds, their moral self-concept fluctuates, which then influences subsequent morally relevant actions. Miller and Effron (2010) have proposed that previous good deeds increase one's moral selfconcept, which then frees them to subsequently commit bad deeds. Previous research has suggested the role of moral self-concept in the moral licensing process. For instance, Sachdeva et al. (2009) found that participants who wrote a story referring to their positive traits (an increase in moral self-concept) then decreased their subsequent altruistic or prosocial behavior. Similarly, the activation of moral self-concept by recalling previous moral actions can license prejudiced behavior (Monin and Jordan, 2009).
This study's theorizing about the positive relationship between other in-group members' OCB and employees' organizational deviance is informed by vicarious moral licensing theory (Kouchaki, 2011). The vicarious moral licensing model has its roots in moral licensing theory (Miller and Effron, 2010), suggesting that other people's past moral behavior can boost one's moral self-concept, subsequently licensing immoral actions. In addition, the vicarious selfconcept theory (Goldstein and Cialdini, 2007) proposes that when experiencing a behavior of people who are psychologically close to them, individuals tend to incorporate the attributes inferred from that behavior into their own behavior. This vicarious self-perception process can lead to changes in one's self-concept, which, in turn, alter their own behaviors (Goldstein and Cialdini, 2007). Applying these theories in the context of OCB, this study argues that other in-group members' OCB can have a positive effect on focal employees' moral self-concept, which then licenses subsequent organizational deviance. This study proposes the following hypothesis: H1. Employee's moral self-concept mediates the relationship between other in-group members' OCB and employees' organizational deviance.
The moderating effect of in-group identification One of the most important assumptions underlying the vicarious self-perception process is that the observer and actor must be psychologically connected, which is defined as the perception of being psychologically close to someone or having a sense of shared identity (Goldstein and Cialdini, 2007). Several studies have demonstrated support for this assumption. Galinsky et al. (2008), for example, found that individuals are more likely to treat others' acts as their own when they take others' perspectives. Similarly, taking the perspective of others can lead individuals to experience the exhaustion of others' self-control efforts (Ackerman et al., 2009), feelings of pain (Jackson et al., 2006) and cognitive dissonance (Norton et al., 2003). At the group level, social identity theory (Tajfel and Turner, 1979) proposes that one's self-concept is mainly based on group membership (Hogg and Abrams, 1988). Thus, the vicarious self-perception process happens only when the observer and actor have a shared identity. In other words, the degree to which an observer identifies with an actor determines the degree to which the observer is affected by the actor's behavior (Goldstein and Cialdini, 2007;Tajfel and Turner, 1979). Using the logic of this theory, this study argues that the effect of other in-group members' OCB on employees' moral selfconcept depends on the identification between employees and other in-group members (i.e., immediate work colleagues or peers in the same department as the focal employees). Furthermore, in-group identification moderates the corresponding indirect effect of other ingroup members' OCB on focal employees' organizational deviance via moral self-concept. Therefore, this study hypothesizes the following: H2. In-group identification moderates the relationship between other in-group members' OCB and focal employees' moral self-concept, such that the relationship is more positive when in-group identification is high than when in-group identification is low.
H3. In-group identification moderates the indirect effect of other in-group members' OCB on focal employees' organizational deviance via focal employees' moral self-concept, such that the indirect effect is stronger when in-group identification is high than when in-group identification is low (see Figure 1).

Sample and procedures
Data were collected from a mobile telecom service company in the south of Vietnam through the author's alumni connection with this company's manager. Using a random sampling approach, the author selected 500 employees from this company's 12 branches located in different provinces or cities to ensure that the sample was representative. Data were collected in 2019 via a multisource and multiwave design at three time points by using the Qualtrics online survey. At Time 1, employees reported their OCB and in-group identification. One week later (Time 2), employees rated their moral self-concept. The final wave of the survey (Time 3) was sent to employees one week later to obtain the ratings on organizational deviance. The measurement of the predictor and criterion variables was A moral licensing perspective separated into three time points about one week apart to alleviate measurement context effects. Although this study relied on individual self-reported data, the multisource design (i.e., predictor was obtained from other in-group members and criterion was obtained from focal employees) can somewhat mitigate concerns about common method bias (Podsakoff et al., 2012). Of those invited, 428 employees completed the Time 1 survey, and 379 employees completed the Time 2 survey. At Time 3, 359 employees returned the useable questionnaire. The final matched sample included 340 employees, yielding a response rate of 68%. Although participants were chosen using a random sampling approach, not all the selected candidates accepted the invitation to complete the three stages of data collection. Therefore, the resulting sample is not completely random. Participant data were matched to workgroups based on the name of the department and branch that participants provided in their responses.

Participants
Of the 340 employees in the sample, 68% were male and 61% had a bachelor's degree. Their average age was 30.26 years (SD 5 4.47), and they had been employed by the company for 4.98 years (SD 5 3.07). They belonged to 56 workgroups, with 6.07 employees per group on average.

Measures
All items were measured on a seven-point scale ranging from 1 (strongly disagree) to 7 (strongly agree). The English questionnaire was translated into Vietnamese and then back to English to ensure translation equivalence (Brislin, 1970).
OCB of other in-group members (Time 1). To measure employees' OCB, Williams and Anderson's (1991) six items on OCB that benefited the organization as a whole (OCB-O) were used. A sample item is "Adheres to informal rules devised to maintain order." This study focused only on OCB-O rather than on OCB-I (i.e., OCB targeting specific individuals) because the relationship between OCB and organizational deviance was examined at the group level. Other in-group members' OCB is the mean of all other members' OCB in the workgroup. The alpha reliability was 0.90. Aggregation was justified by r wg 5 0.85; the reliability of individual assessment of the group mean, ICC(1) 5 0.51; and the group mean, ICC(2) 5 0.75.
In-group identification (Time 1). In-group identification was measured with a four-item global measure of identification with a social group adapted from Doosje et al. (1995). A sample item is "I identify with other members in my workgroup." The scale's reliability coefficient was 0.91. Confirmatory factor analysis (CFA) revealed that the single-factor structure provided an adequate fit to the data (TLI 5 0.93; CFI 5 0.94; RMSEA 5 0.05). All items exhibited significant factor loadings.
Moral self-concept (Time 2). Moral self-concept was measured using Aquino and Reed's (2002) five-item internalization scale of moral self-concept (α 5 0.89). Participants reported the extent to which they embodied a set of ethical characteristics (e.g., helpful, hardworking,

In-group identification OCB of other in-group members
Moral self-concept Organizational deviance

Group level
Individual level Figure 1. Theoretical model honest, kind). A sample item is "Having these characteristics is an important part of my sense of self." The CFA on this construct demonstrated an adequate fit to the data (TLI 5 0.90; CFI 5 0.91; RMSEA 5 0.06). All items significantly loaded on the factor. Organizational deviance (Time 3). Organizational deviance was measured using Bennett and Robinson's (2000) twelve-item scale (α 5 0.89). A sample item is "I have come in late to work without permission." A single-factor CFA revealed an adequate fit to the data (TLI 5 0.91; CFI 5 0.92; RMSEA 5 0.06). All items exhibited significant factor loadings. The four scales used in this study have been widely used in organizational behavior and psychology research (DeCelles et al., 2012;Eby et al., 2015;Hardy et al., 2014;Mitchell and Ambrose, 2007;Mount et al., 2006;Priesemuth et al., 2014;Turnley et al., 2003;Van Dick et al., 2008).
Control variables (Time 1). The design controlled for employees' gender (1 5 male and 2 5 female), age and years of work tenure (open-ended), as previous research has shown that these variables may influence unethical behavior (Aquino et al., 2001;Erdogan and Liden, 2002).

Analytical strategy
Given the multilevel structure of the data (employees nested within workgroups), the author conducted a multilevel path analysis in Mplus 7.4 (Muth en and Muth en, 2015). Furthermore, the author calculated the mediation by following the recommendations for multilevel models to implement a parametric bootstrap with 20,000 resamples (Preacher et al., 2010). To test Hypothesis 2, which involves a cross-level interaction between other ingroup members' OCB and focal employees' in-group identification, the author group-mean centered in-group identification to eliminate the potential confounding effect of betweengroup interaction (Hofmann et al., 2000). Additionally, the Johnson-Neyman technique was used to determine the value of in-group identification at which the relationship between other in-group members' OCB and employees' moral self-concept is statistically significant (Bauer and Curran, 2005;Preacher et al., 2003Preacher et al., , 2006. To test for moderation mediation, the author examined the conditional indirect effects at ±1 standard deviation of in-group identification.

Results
A multilevel CFA on the four study variables (OCB of other in-group members, in-group identification, moral self-concept and organizational deviance) revealed an adequate fit to the data (χ 2 5 616.66, df 5 318, TLI 5 0.94, CFI 5 0.95, RMSEA 5 0.04). The descriptive statistics, correlations and reliabilities for the study variables are displayed in Table 1. As expected, other in-group members' OCB was significantly and positively related to employees' moral self-concept (r 5 0.34, p < 0.01). In-group identification was positively related to moral self-concept (r 5 0.29, p < 0.01). Moral self-concept was significantly and positively related to organizational deviance (r 5 0.38, p < 0.01). Hypothesis 1 predicted that moral self-concept mediates the relationship between other ingroup members' OCB and focal employees' organizational deviance. Supporting Hypothesis 1, other in-group members' OCB was positively associated with employees' moral self-concept (γ 5 0.30, p < 0.01), and moral self-concept was positively associated with organizational deviance (γ 5 0.12, p < 0.05). Furthermore, as shown in Table 2, the indirect effect of other in-group members' OCB on employees' organizational deviance via moral selfconcept was 0.036, and the 95% confidence interval excluded zero (0.006, 0.073).
Hypothesis 2 predicted that in-group identification moderates the relationship between other in-group members' OCB and focal employees' moral self-concept, such that the A moral licensing perspective relationship is stronger when the focal employees have a higher level of in-group identification. The interaction between other in-group members' OCB and employees' ingroup identification was positive and significant (γ 5 0.09, p < 0.05). The relationship between other in-group members' OCB and employees' moral self-concept was statistically significant and more positive at values of in-group identification greater than 4.26. Figure 2 shows the interaction pattern, which further supports Hypothesis 2.
Finally, Hypothesis 3 predicted that in-group identification moderates the indirect effect of other in-group members' OCB on focal employees' organizational deviance via moral selfconcept, such that the indirect effect is stronger when the focal employees have a higher level of in-group identification. Supporting Hypothesis 3, the conditional indirect effect at high (þ1 SD) levels of in-group identification was positive and significant (indirect effect 5 0.058; 95% CI 5 [0.017, 0.084]), whereas the indirect effect at low (À1 SD) levels of in-group identification was non-significant (indirect effect 5 0.012; 95% CI 5 À0.018, 0.046). 0.058 0.017 0.084 Note(s): N 5 340 at level 1; N 5 56 at level 2. The first value is the unstandardized parameter estimate, and the value within parentheses is the standard error. Bootstrap sample size 5 20,000. LLCI 5 Lower limit confidence interval; ULCI 5 Upper limit confidence interval *p < 0.05, **p < 0.01 Table 1. Descriptive statistics, correlations, and reliabilities Table 2.

Discussion
Drawing on moral licensing theory and the literature on vicarious moral licensing, this study proposed and tested a dynamic model of the vicarious moral licensing effect of other in-group members' OCB on focal employees' organizational deviance. The results demonstrate that moral self-concept significantly mediates the relationship between other in-group members' OCB and focal employees' organizational deviance. Furthermore, the results illustrate that ingroup identification moderates the indirect effect of other in-group members' OCB on focal employees' organizational deviance via moral self-concept, such that high levels of in-group identification strengthen this mediated relationship. The next section discusses the theoretical and practical implications of these findings.

Theoretical implications
This study contributes to the current literature in several ways. First, it adds to the OCB literature by applying the moral licensing perspective to investigate the potential detrimental outcomes of OCB in a group context. The current work has theorized and demonstrated that other in-group members' OCB might grant employees a moral license to engage in subsequent organizational deviance. By identifying employees' deviant behavior stemming from other in-group members' OCB, the study reveals that although OCB is a form of proorganizational behavior, its consequences might actually be negative for an organization. While prior research has mainly focused on the negative consequences of individuals' OCB (Koopman et al., 2016;Yam et al., 2017), this study investigated employees' organizational deviance that stems from observing other in-group members' OCB. By doing so, the current work broadens our understanding of the costs of OCB in a group context. Second, this research contributes to the literature on moral licensing by offering more empirical evidence of the psychological mechanism underlying the vicarious moral licensing effect. The results indicate that focal employees experience an increase in moral self-concept after observing other in-group members' OCB, such that elevated moral self-concept may lead to subsequent organizational deviance. By exploring the mediating role of moral self-concept in the relationship between other in-group members' OCB and employees' organizational deviance, this work demonstrates that moral self-concept also accounts for the effect of vicarious moral licensing, as is the case with the moral licensing effect (Monin and Jordan, 2009;Sachdeva et al., 2009). Additionally, while previous research has provided support for the mediating roles of moral credits and moral credentials in the moral licensing effect (Kong et al., 2020;Lin et al., 2016;Yam et al., 2017), this study further demonstrates that moral self- Relationship between other in-group members' OCB and moral self-concept as a function of in-group identification A moral licensing perspective concept underlies the moral licensing effect of other in-group members' OCB on employees' organizational deviance. In doing so, this study advances our understanding of the mechanisms underlying the moral licensing effect in a group context.
Finally, the finding that employees who highly identify with their workgroup are more likely to incorporate other in-group members' OCB into their own moral self-concept and subsequently feel licensed to engage in organizational deviance offers more evidence on the potential disadvantages of in-group identification. In-group identification has been regarded as a positive perception that is associated with positive consequences for both employees and organizations (Ellemers et al., 1997). By contrast, this study examined group identification with an individual difference approach and demonstrated that strong in-group identification may facilitate the vicarious moral licensing effect of other in-group members' OCB on employees' organizational deviance.

Practical implications
Prior research has proposed an important role of individuals' ethical conduct in influencing businesssustainability (Bernaland Edgar,2012;Epstein,2008;Middlebrooks et al.,2009;Nguyen et al., 2013). To support businesses' efforts to enhance long-term effectiveness and corporate sustainability, organizational behavior scholars have focused on the role of OCB and deviant workplace behaviors in achieving sustainability (Park and Shaw, 2013;Uwem et al., 2019). Recently, Lee (2020) demonstrated that OCB augmented, whereas CWB attenuated corporate sustainability performance. In terms of enhancing business sustainability, the findings of this study offer an empirical basis for considering other in-group members' OCB as a critical antecedent of employees' organizational deviance, thereby informing effective intervention approaches for organizations to prevent OCB from resulting in negative consequences.
To be clear, despite other in-group members' OCB being associated with employees' organizational deviance, the author is not discouraging OCB in the workplace. Instead, the author encourages managers to support OCB but to adopt a more cautious and nuanced view of OCB in group contexts. Therefore, as part of managerial education and leadership development activities, organizations should incorporate this study's findings into their education and training programs to increase managers' and leaders' awareness of the potential negative consequences of OCB in group contexts. This awareness can help them proactively prevent other in-group members' OCB from resulting in employees' organizational deviance.
Furthermore, organizations should make managers aware of this study's finding that moral self-concept mediates the vicarious moral licensing effect of other in-group members' OCB on employees' organizational deviance. This outcome underscores the important role of moral self-concept and can inform effective intervention approaches for managers to prevent the possible negative consequences of OCB in group contexts. Managers should prevent the increase of employees' moral self-concept by stopping other in-group members' OCB from being morally praiseworthy in the eyes of employees, thereby hindering the development of a moral license (Klotz and Bolino, 2013). For instance, if other in-group members are rewarded for their OCB (e.g., an extra day off), that may somewhat prevent the increase in moral selfconcept of focal employees. In other words, focal employees' moral self-concept may not increase when other in-group members' OCB is compensated. Thus, organizations should inform managers that constraining the increase of moral self-concept may prevent other ingroup members' OCB from resulting in employees' organizational deviance. Additionally, organizations should also provide accessible resources (e.g., self-concept management training) to help managers and leaders better understand and cope with their subordinates' moral self-concept arising from other in-group members' OCB.
Last but not least, given that this study has revealed a potential drawback of high ingroup identification, which strengthens the indirect effect of other in-group members' OCB on employees' organizational deviance via moral self-concept, organizations are advised to keep a close eye on employees highly identifying with other in-group members. Although strong in-group identification promotes workgroup effectiveness, it may become a breeding ground for the vicarious moral licensing process. A potential way for organizations to mitigate the detrimental effect of a high level of in-group identification is to clearly highlight the importance of in-group identification and its role in promoting the vicarious moral licensing effect, which could, in turn, lead to deviant behavior.

Limitations and further research
Several limitations of the current study should be acknowledged. First, although the present research was a correlational survey study relying on self-reported data, this study relied on multiwave, multisource data collection to mitigate common method bias (Podsakoff et al., 2012). Future researchers may consider adopting more objective measures of the study constructs. Second, although the data used in this study were collected from the company's 12 branches located in different provinces or cities, this study focused only on a specific industry (service), which could limit the generalizability of the findings to other industries in Vietnam. Additionally, the study context (Vietnam) could limit the generalizability of the findings to other contexts. Future research in other industries and cultural contexts (e.g., Western culture) is needed to generalize the findings.
Third, this study examined only the mediating role of moral self-concept in the relationship between other in-group members' OCB and employees' organizational deviance. Future research should explore other mediators (e.g., psychological entitlement; Yam et al., 2017) to advance our understanding of the various mechanisms underlying the vicarious licensing process. Fourth, this study focused only on OCB-O rather than on OCB-I due to the group level of analysis. More specifically, the effect of other in-group members' OCB-I on focal employees' organizational deviance will differ depending on the target of this behavior (e.g., OCB-I that targets focal employees or OCB-I that targets other in-group members). Future research should explore this relationship at a dyad level to account for the influence of both OCB-I that targets focal employees and OCB-I that targets others on organizational deviance. Finally, this research identified only in-group identification as a boundary condition of the vicarious licensing process. Future research may investigate other contextual factors such as the role of (un)ethical leadership in the vicarious moral licensing effect of other in-group members' OCB on employees' organizational deviance.

Conclusion
Other in-group members' OCB is positively related to employees' organizational deviance via moral self-concept. This vicarious moral licensing effect is stronger when in-group identification is high, as in-group identification strengthens the positive relationship between other in-group members' OCB and employees' moral self-concept. These findings advance our understanding of how other in-group members' OCB may lead to employees' organizational deviance and shed new light on the potential negative consequences of OCB in the group context.