Arrogance from favor: the better supervisor-subordinate guanxi, the more counterproductive work behaviors

Abstract

Purpose

Drawing on the attribution theory, this study re-examined the effect of supervisor-subordinate guanxi (SSG) on counterproductive work behaviors (CWBs) by highlighting the critical moderating role of job self-efficacy and the mediating role of psychological entitlement, with the aim of revealing the potential drawbacks of SSG.

Design/methodology/approach

Three-wave matched data were collected from 434 employees working in China. The authors adopted path analysis in Mplus 7.4 to test the hypotheses proposed in this study.

Findings

The results suggested that among subordinates with high job self-efficacy, SSG triggers psychological entitlement, which subsequently leads to CWBs.

Originality/value

This research challenges the prevailing consensus regarding the beneficial effects of SSG on subordinate management from the social exchange theory perspective by revealing when and why high-quality SSG can also engender subordinates’ psychological entitlement and subsequent CWBs. By doing so, this study provides a more dialectical view of the impact of establishing high-quality SSG in human resource management.

Citation

Xiayi, L., Yang, T. and Wan, W. (2025), "Arrogance from favor: the better supervisor-subordinate guanxi, the more counterproductive work behaviors", Leadership & Organization Development Journal, Vol. 46 No. 9, pp. 1-15. https://doi.org/10.1108/LODJ-07-2023-0403

Introduction

Supervisor-subordinate guanxi (SSG) is defined as the informal interactions between subordinates and their immediate supervisors outside the workplace, which are typically built on mutual interest and benefits (Wong et al., 2003). Since supervisors control scarce organizational resources, subordinates who have a high-quality guanxi with supervisors have additional access to inside information, higher performance evaluations and increased promotional opportunities (Cheung and Wu, 2011; Miao et al., 2020; Zhong et al., 2022). Drawing on the social exchange theory, subordinates within good guanxi engage in more beneficial (i.e. organizational citizenship behavior, voice behavior) (Miao et al., 2020; Wang et al., 2019) and fewer detrimental workplace behaviors (i.e. counterproductive work behaviors, CWBs) (Zhang and Deng, 2016) to pay back their supervisors’ favorable treatment. Hence, prior research has converged on the view that building and maintaining good guanxi with subordinates leads to successful management of them (Law et al., 2000; Wang et al., 2019).

However, this prevailing view may be premature when considering the Chinese idiom, “arrogance from favor (恃宠而骄)”, which means an individual becomes arrogant and conceited due to receiving preferential treatment from supervisors (Tracy et al., 2009). Given that such inflated self-perception is often associated with workplace deviant behaviors (Korman et al., 2023; Qin et al., 2020), this study contends that extant research might overemphasize the impact of SSG on positive workplace behaviors while ignoring its influence on negative workplace behaviors, which leads supervisors to mistakenly assume that they can manage subordinates effectively by building high-quality guanxi with them (Wang et al., 2019). Indeed, extant SSG literature, rooted in the norm of reciprocity has mainly adopted a top-down perspective, contending that subordinates should feel compelled to return the benefits they received from supervisors (Miao et al., 2020). Actually, it is subordinates’ attribution of causality to supervisors’ behavior rather than the behavior itself that ultimately determines their attitudes and reactions (Martinko and Gardner, 1987). For example, Shi et al. (2022) found that family-supportive supervisory behaviors have a stronger positive effect on work engagement through loyalty to the supervisor when subordinates attribute such behaviors to supervisor’s internal motivation rather than meeting organizational expectations. Therefore, this study adopts a bottom-up perspective to reexamine the influence of supervisors’ preferential treatment in a high-quality SSG on subordinates’ psychological and behavioral reactions. Drawing on the attribution theory, we challenge the SSG literature by proposing that SSG can elicit subordinates’ CWBs or actions that are intended to cause harm to organizations and/or organization stakeholders (Spector et al., 2006). Specifically, we contend that supervisors’ favorable treatment in high-quality SSG can induce subordinates’ CWBs through the sense of psychological entitlement, or the stable belief that one should receive favorable treatment without considering the actual deservedness (Harvey and Martinko, 2009), among those holding self-serving attributional styles. Given that individuals with high job self-efficacy, or the cognitive self-appraisal of one’s abilities to exert control over difficult job situations and perform well in their jobs (Schaubroeck et al., 2001), tend to make self-serving attributions for preferential treatment (Silver et al., 1995), we hypothesize that SSG interacts with job self-efficacy to predict subordinates’ psychological entitlement, which subsequently causes them to engage in CWBs morally licensed (Merritt et al., 2010) or with the aim to restore equity (Li et al., 2022; Qin et al., 2020) (see Figure 1).

Our study makes several contributions to the SSG and CWB literature. First, our study challenges the prevailing consensus that high-quality SSG elicits subordinates’ beneficial behaviors (Han et al., 2012; Miao et al., 2020; Sun et al., 2023; Wang et al., 2019; Zhang and Deng, 2016). By incorporating job self-efficacy as a critical moderator, our study reveals that SSG can also induce subordinates’ CWBs via psychological entitlement. Therefore, our study shifts the prevailing consensus regarding SSG to show when and why high-quality SSG can also lead to subordinators’ detrimental reactions, providing a more dialectical and balanced understanding of the effect of SSG on subordinate management. Second, our study contributes to the SSG and CWB literature by reversing the negative linkage between them. Although Zhang and Deng (2016) have found that SSG helps reduce CWBs by increasing subordinates’ job satisfaction, our study posits that high-quality SSG can also lead to CWBs among individuals with high job self-efficacy via the mediating mechanism of psychological entitlement. In doing so, our study addresses the call to further explore the impact of SSG on CWBs (Miao et al., 2020). Third, by introducing attribution theory (Kelley and Michela, 1980) as an overarching framework, our study examines SSG through a new theoretical lens. Extant research has mainly adopted social exchange theory and the norm of reciprocity to explain the influence of SSG (e.g. Han et al., 2012; Wang et al., 2019). However, such theoretical perspectives overlook the important role of subordinates’ attribution in the leadership process (Martinko et al., 2007), thus neglecting its potential costs. Hence, our study adds substance to SSG theorizing by introducing attribution theory as a new theoretical framework to explain why SSG may increase subordinates’ CWBs under certain conditions (i.e. among subordinates with high job self-efficacy).

Theoretical background and hypotheses development

SSG, job self-efficacy and psychological entitlement

Supervisors in Chinese organizations tend to exhibit particularistic treatment towards subordinates who possess high-quality guanxi with them. Specifically, in a high-quality SSG, supervisors prefer to provide subordinates with additional access to valuable information and resources (Wu and Ma, 2023). Besides, they are more likely to provide these subordinates with favorable participation in the organizational decision-making process and endorse suggestions from these subordinates (Burris, 2012; Wang et al., 2019). Furthermore, they give guanxi subordinates higher performance evaluations and increased promotional opportunities (Gu and Nolan, 2017; Miao et al., 2020). As a consequence, previous research mainly based on the social exchange theory contends that SSG contributes to effective human resource management by facilitating subordinates’ positive while inhibiting negative work outcomes via various psychological mechanisms (e.g. job satisfaction, psychological ownership, psychological empowerment and trust in the supervisor) (Han et al., 2012; Miao et al., 2020; Sun et al., 2023; Wang et al., 2019; Zhang and Deng, 2016). However, research has not yet considered that supervisors’ favorable treatment can also induce subordinates’ inflated perceptions of self-worth and uniqueness (Campbell et al., 2004), which brings about a sense of psychological entitlement (Campbell et al., 2004; Qin et al., 2020). Therefore, we posit that SSG has the potential to induce higher levels of psychological entitlement.

Given the influence of SSG on subordinates’ positive psychological reactions, it is noteworthy that not all subordinates are expected to develop a sense of psychological entitlement within a high-quality SSG. According to Jordan et al. (2017), self-serving attributional bias, described as individuals generally attributing positive events to internal factors and attributing negative outcomes to external factors, is of particular significance in upholding inflated self-perception. In this regard, we contend that SSG is more likely to engender psychological entitlement among subordinates with high job self-efficacy because they are particularly self-serving in their attributions (Silver et al., 1995). Specifically, job self-efficacy captures individuals’ cognitive self-appraisal of their capabilities to exert control over difficult job situations and perform well in their jobs (Schaubroeck et al., 2001). Since subordinates with high job self-efficacy are more likely to adopt a self-serving attributional style, they are more likely to attribute supervisors’ preferential treatment in good SSG internally to their unique capabilities and contributions at work. Consequently, these inflated self-perceptions of self-importance likely induce high levels of psychological entitlement (Harvey and Harris, 2010; Qin et al., 2020). Conversely, subordinates with low job self-efficacy are not confident in their own capabilities and contributions. Hence, they are less likely to make an internal attribution for the special treatment in a high-quality SSG (Schaubroeck et al., 2001; Schreurs et al., 2010). As a result, they are less likely to experience inflated perceptions of self-worth and self-capabilities and are thus less likely to develop a sense of psychological entitlement when supervisors treat them preferentially within a high-quality SSG. Taking this together, this study proposes the following hypothesis:

H1.

SSG will interact with job self-efficacy to predict psychological entitlement. Specifically, the positive relationship between SSG and psychological entitlement only exists among subordinates with high job self-efficacy rather than those with low job self-efficacy.

Psychological entitlement as a catalyst for CWBs

We further propose that psychological entitlement increases subordinates’ willingness to engage in CWBs. Psychological entitlement is the result of inflated perceptions of self-worth and importance and unrealistic expectations for preferential treatment regardless of their actual performance (Campbell et al., 2004; Jordan et al., 2017). Consequently, on the one hand, we argue that psychological entitlement can engender CWBs because highly entitled subordinates take it for granted to engage in such behaviors from a moral licensing perspective (Lee et al., 2019a; Merritt et al., 2010). Specifically, the inflated feelings of self-worth and importance help highly entitled subordinates establish their moral credentials, allowing them to morally rationalize their socially irresponsible CWBs (Lee et al., 2019b; Yam et al., 2017). In other words, highly entitled subordinates do truly believe that they are above organizational rules and thus have limited moral concerns about CWBs. For example, they might make fun of colleagues casually without discrediting themselves.

On the other hand, we argue that highly entitled subordinates’ unrealistic expectations can also lead them to engage in CWBs from the equity theory perspective (Adams, 1965). In particular, highly entitled subordinates generally expected preferential treatment regardless of their actual performance (Campbell et al., 2004; Harvey and Martinko, 2009; Priesemuth and Taylor, 2016). Obviously, their unrealistic expectations can hardly always be satisfied. Such unmet expectations will lead to their perceptions of mistreatment and job-related frustrations (Harvey and Harris, 2010). In this case, highly entitled subordinates may resort to CWBs purposely to obtain their deserved returns with the aim to restore equity (Li et al., 2022; Qin et al., 2020). For example, they might spend too much time daydreaming instead of working on a personal matter. In line with this argument, previous research has found that psychological entitlement positively influences individuals’ aggression, financial misconduct and workplace deviant behaviors (Campbell et al., 2004; Li et al., 2022; Qin et al., 2020). Thus, we propose the following hypothesis:

H2.

Psychological entitlement is positively related to CWBs.

In light of the proceeding hypotheses, where we propose the interactive effect of SSG and job self-efficacy on psychological entitlement, coupled with the positive effect of psychological entitlement on CWBs, we further argue that psychological entitlement can mediate the interactive effect of SSG and job self-efficacy on CWBs. That is, we posit that for subordinates in a high-quality SSG, compared with those holding low job self-efficacy, subordinates with high job self-efficacy are more likely to develop awareness of psychological entitlement due to their tendencies to attribute supervisors’ preferential treatment with a self-serving attributional style. The heightened psychological entitlement subsequently leads them to engage in more CWBs. Thus, we propose the following hypothesis:

H3.

Psychological entitlement mediates the interactive effect of SSG and job self-efficacy on CWBs, such that the indirect effect only exists among subordinates with high job self-efficacy rather than those with low job self-efficacy.

Method

Sample and procedure

Data were collected from 677 full-time working adults at four different enterprises located in Quanzhou, China. Before distributing the questionnaires, we first told participants about the voluntary nature of participation. Also, we identified the statistical and internal research purposes of the study and provided a guarantee that their survey information would be treated with the utmost confidentiality and anonymity. Then, with the assistance of trained research assistants, we administered the surveys to participants individually, providing each with an envelope. All the surveys were collected on the spot.

In order to reduce the impact of potential common method biases (Podsakoff et al., 2012), surveys were distributed over three separate time periods over two weeks. At time 1, 677 participants rated SSG, job self-efficacy, the emotional exhaustion scale and their demographic characteristics. At time 2, two weeks after time 1, the 677 participants filled out a follow-up questionnaire rating their psychological entitlement. Then, two weeks after time 2, at time 3, the 677 participants assessed their CWBs. We matched the responses across three time periods with the last four digits of each participant’s phone number (e.g. 0122). The final sample included 434 matched effective questionnaires, yielding a response rate of 64.11%.

Among the 434 participants, 50.9% were male. Age was coded into four categories (along with percentage of the samples in each band): below 25 years (16.4%), 26–35 years (44.9%), 36–45 years (23.5%) and over 45 years (15.2%). Concerning education, 22.4% held a high school degree or less, 36.6% held completed a college degree, 31.6% held a bachelor degree and 9.4% held a master degree or higher. Tenure was reported in four bands: less than 3 years (25.3%), 4–6 years (30.2%), 7–9 years (31.3%) and over 10 years (13.1%).

Measures

With the assistance of two bilingual human resource management researchers, we translated all English measures into Chinese following the back-translation procedure from Brislin (1980). All measures were rated using a five-point Likert response format from 1 (strongly disagree) to 5 (strongly agree) unless otherwise noted.

SSG (T1). Participants rated SSG with a six-item scale from Law et al. (2000). A sample item reads, “When there are conflicting opinions, I will definitely stand on my supervisor’s side.”

Job self-efficacy (T1). Participants measured their job self-efficacy with the six-item scale from Personal Efficacy Beliefs Scale (Riggs et al., 1994). A sample item reads, “I have confidence in my ability to do my job.”

Psychological entitlement (T2). Participants rated their psychological entitlement using the four-item scale developed by Yam et al. (2017). A sample item reads, “I honestly feel I’m just more deserving than others.”

CWBs (T3). Employees assessed their CWBs with the five-item scale from Fehr et al. (2017). A sample item reads, “Spend too much time fantasizing or daydreaming instead of working.” We adopted a self-report measure because many CWBs are done in private without anyone else knowing (Qin et al., 2020).

Controls. We controlled for employees’ age, gender (1 = male and 2 = female), education and tenure because these demographic variables may be related to workplace deviant behaviors (Berry et al., 2007). In addition, we also included employees’ emotional exhaustion as a control because of its great influence on CWBs (Bolton et al., 2012; Chen et al., 2020). We measured employees’ emotional exhaustion with a five-item scale developed by Schaufeli et al. (1996). A sample item reads “I feel emotionally drained from my work” (1 = never to 5 = everyday).

Results

Preliminary analyses

The descriptive statistics, reliability estimates and inter-correlations among variables are presented in Table 1. We first conducted a series of confirmatory factor analyses (CFA) in Mplus 7.4 version to assess whether the measures in our study have eligible discriminant validity. The results in Table 2 showed that our five-factor base-line model (SSG, job self-efficacy, psychological entitlement, CWBs and emotional exhaustion) provided a significantly better fit to the data (χ² = 446.421, df = 289, χ²/df = 1.61, comparative fit index (CFI) = 0.983, Tucker–Lewis index (TLI) = 0.980, root mean squared error of approximation (RMSEA) = 0.038 and standardized root mean error residual (SRMR) = 0.031) than the other alternative models, supporting the distinctiveness of key variables. Then, because all the data in our study were collected from the same source procedurally, we conducted Harman’s single-factor test to examine the common method bias statistically (Podsakoff et al., 2003). The Harman’s single-factor test drew out five factors, and the largest explains 22.826% of the variance, which suggested that there was no serious common method bias in this study.

Hypotheses testing

We formally conducted path analysis in Mplus version 7.4 to examine our proposed hypothesized model. Specifically, we centered on SSG and job self-efficacy to estimate their interactive effects on psychological entitlement (H1). Then, we examined the mediating effects (H2) and conditional indirect effects (H3) with 95% confidence intervals of the bootstrapping approach (5,000 iterations) (Preacher et al., 2010) (results are shown in Table 3). The model fit information of integrative path analysis model (χ² = 20.059, df = 15, χ²/df = 1.34, CFI = 0.968, TLI = 0.964, RMSEA = 0.028 and SRMR = 0.025) indicated that the data fit our hypothesized model well. H1 suggested that SSG and job self-efficacy interactively influence psychological entitlement. Results indicated that after controlling participants’ gender, age, education, tenure and emotional exhaustion, the interaction term of SSG and job self-efficacy was statistically positive and significant for psychological entitlement (b = 0.201, S.E. = 0.044, p < 0.001) (see Model 2 in Table 3). Furthermore, we plotted the interactions to better illustrate this interactive effect following Aiken and West (1991). As shown in Figure 2, simple slope tests indicated that the relationship between SSG and psychological entitlement was significant and positive among participants with high job self-efficacy (+1 SD: b = 0.435, S.E. = 0.062, p < 0.001) but was not significant among those with low job self-efficacy (−1 SD: b = 0.066, S.E. = 0.066, p = 0.322). Hence, H1 received support.

H2 hypothesized that psychological entitlement positively influences subordinates’ CWBs. The results indicated that there was a significantly positive linkage between psychological entitlement and CWBs (b = 0.154, S.E. = 0.041, p < 0.001) (see Model 1 in Table 3), thus supporting H2.

H3 proposed that psychological entitlement mediates the interactive effects of SSG and job self-efficacy on CWBs. We conducted a first-stage moderated-mediation analysis to examine the conditional indirect effects. Results in Table 4 indicated that the indirect effects of SSG on CWBs via psychological entitlement were statistically positive and significant when job self-efficacy was high (estimate = 0.067, standard error (SE) = 0.021, 95% confidence interval (CI) [0.029, 0.114]) but was not significant when job self-efficacy was low (effect size = 0.010, S.E. = 0.011, 95% CI [−0.007, 0.036]). The difference between these conditional indirect effects was also significant (estimate = 0.057, SE = 0.021, 95% CI [0.023, 0.107]). Thus, H3 received support.

Discussion

To date, it has been widely acknowledged that a good SSG can promote subordinates’ positive workplace behaviors and that the SSG should facilitate successful human resource management. However, our study challenges this consensus by revealing that SSG can also engender CWBs under certain conditions. Drawing on the attribution theory, our field study shows that for subordinates with high job self-efficacy, SSG can engender high psychological entitlement, which subsequently leads to increased CWBs. Our study has several important theoretical as well as practical implications.

Theoretical implications

Our research makes several theoretical contributions to the extant literature. First, our study contributes to the SSG literature by offering a more balanced and dialectic perspective to help people understand the influence of SSG on subordinates’ workplace behaviors. Extant research, rooted in the social exchange perspective and the norm of reciprocity, has reached a unanimous conclusion regarding the beneficial effects of SSG on subordinates’ workplace behaviors (e.g. Han et al., 2012; Miao et al., 2020; Sun et al., 2023; Wang et al., 2019; Zhang and Deng, 2016). However, our study challenges this consensus view by providing empirical evidence revealing that SSG may also engender negative workplace behaviors, such as CWBs. By doing so, our research provides a more comprehensive understanding of the influence of SSG on subordinates’ workplace behaviors.

Second, our study also contributes to the SSG and CWB literature by extending the discussion on the impact of SSG on CWBs. Prior research has suggested that SSG helps inhibit subordinates’ CWBs via the mediating role of job satisfaction (Zhang and Deng, 2016). In this study, we reveal that SSG can also engender subordinates’ CWBs through the mediating mechanism of psychological entitlement for individuals with high job self-efficacy. Thus, we extend the discussion regarding the linkage between SSG and CWBs, offering a more complete picture of the role played by social cognitions in linking SSG and CWBs. In this way, our study responds to the call for further exploration of the influence of SSG on CWBs (Miao et al., 2020).

Third, our study also contributes to the SSG literature by examining its impact from the attribution theory. Extant research has mainly adopted the social exchange perspective and the norm of reciprocity to illustrate the effects of SSG on subordinates (Wang et al., 2019; Zhang and Deng, 2016), which implicitly assumes that subordinates feel compelled to return the favorable benefits they received from supervisors (Eisenberger et al., 2002; Gouldner, 1960). However, this line of research might neglect the critical role of subordinates’ interpretations in determining the actual effects of leadership behaviors (Gardner et al., 2019; Martinko et al., 2007). For example, Liao et al. (2020) found that abusive supervision can elicit leader-directed deviance when subordinates attribute it to injury initiation motives, while it can also help improve task performance when subordinates attribute it to performance promotion motives. In this study, we also found that high-quality SSG, which traditionally yields functional results, can also lead to CWBs when subordinates attribute it in a self-serving way. In this regard, our study provides a new theoretical perspective for explicating the effects of SSG, echoing the call for more research to examine the influence of leadership from subordinates’ attribution perspective (Martinko et al., 2010).

Finally, our research contributes to the SSG literature by investigating job self-efficacy as a critical moderator in determining the influence of SSG. Prior research on SSG has converged on the idea that high-quality SSG reduces subordinates’ negative outcomes (e.g. turnover intention and CWBs) (Cheung et al., 2009; Zhang and Deng, 2016). While our study revealed that high-quality SSG can also lead to CWBs among subordinates with high job self-efficacy, enriching research suggested that positive leadership behaviors might also leads to subordinates’ deviant behaviors under certain conditions (Qin et al., 2020). As such, our study enriches the SSG literature by offering a more nuanced perspective to investigate the impact of SSG on subordinates, answering the call to identify more contingent factors that moderate the influence of SSG (Miao et al., 2020).

Practical implications

Our study also holds practical implications for organizations to caution against the potential detrimental effects of SSG among subordinates with high job self-efficacy. First, supervisors need to be aware of both the beneficial (e.g. increased job satisfaction and decreased CWBs; Zhang and Deng, 2016) and undesirable influence (e.g. enhanced psychological entitlement and CWBs) of SSG on subordinates. As such, supervisors should be cautious about favoritism and preferential treatment in the workplace, as it not only goes against the moral standard that everyone should be treated equally (Thau et al., 2013), but can also undermine subordinates’ felt obligation to reciprocity (Lee et al., 2019a). Hence, it is crucial to integrate our discoveries into management training programs in order to alert supervisors to develop a more dialectical and impartial understanding of how their preferential treatment impacts subordinates’ reactions.

Second, the reason why subordinates with high job self-efficacy are more likely to engage in CWBs within a high-quality SSG lies in their tendency to attribute supervisors’ preferential treatment using a self-serving attribution style. In order to address this problem, organizations should provide training and development programs that guide employees to enhance self-confidence based on their actual skills and achievements rather than relying on supervisors’ preferential treatment. By doing so, organizations can cultivate a more harmonious and productive workplace environment, where employees are valued for their merits rather than personal guanxi with supervisors. For example, it would be beneficial to enhance mindfulness training in the workplace, thereby reducing subordinates’ tendency to attribute supervisors’ special treatment solely to their own distinctiveness and contributions with a self-serving attribution style (Ruedy and Schweitzer, 2010).

Limitations and future directions

Despite the above theoretical and practical implications, our study also has several limitations that need to be addressed by future studies. Firstly, although we collected data at different time points, the data in this research rely on self-report measures and are cross-sectional data in essence, which limits the ability to establish causal relationships (Podsakoff et al., 2012). Therefore, we encourage future research to conduct experimental design and longitudinal studies to strengthen causality. Secondly, our study is limited to four different enterprises in Chinese, which suggested that our research conclusion might be influenced by Chinese culture. Therefore, we encourage future research to conduct this study in other cultures. Thirdly, although our study posits job self-efficacy as a critical contingency that moderates the effects of SSG, we cannot exclude the possibility that other personality traits (e.g. locus of control) can also condition the impact of SSG, which is worth further investigation. Fourthly, we encourage future research to take a temporal view of SSG research because supervisors might change their attitudes and reactions toward those entitled subordinates. One possible outcome might be that supervisors felt disappointed at those entitled subordinates because their efforts to provide preferential treatment did not bring them the expected outcomes. Conversely, those entitled subordinates may have high expectations of receiving special treatment and rewards, and they even develop resentment toward supervisors when their expectations are not met, which may eventually result in the breakdown of the guanxi. Therefore, future studies can benefit from investigating the dynamic change of SSG. Finally, our study offers a new theoretical perspective for future studies to investigate the linkage between SSG and other outcomes. Our results reveal that due to self-serving attributional bias, subordinates may not react to supervisors’ expectations from the social exchange perspective (e.g. increased job satisfaction and decreased CWBs). Therefore, we encourage future research to investigate the impact of SSG on subordinates’ psychological and behavioral reactions from an attribution theory perspective.

Conclusions

Drawing on the attribution theory, this study re-examines the influence of SSG on CWBs. Contrary to extant conclusions suggesting that high-quality SSG helps inhibit subordinates’ CWBs, our study found that high-quality SSG can also lead to CWBs via psychological entitlement among subordinates with high job self-efficacy. By revealing the potential drawbacks of SSG, our study shifts the consensus regarding the impact of SSG to show when and why high-quality SSG can also elicit subordinates’ negative psychological and behavioral reactions. We hope that our study can stimulate future research endeavors to adopt a dialectical and balanced perspective to further investigate the impact of SSG.

Figures

Figure 1

Conceptual model

Figure 2

Interaction between SSG job self-efficacy in predicting psychological entitlement

Table 1

Descriptive statistics, variable inter-correlations and reliability estimates

Variable	Mean	SD	1	2	3	4	5	6	7	8	9
1 Gender	1.49	0.50	–
2 Age	2.38	0.93	−0.129**	–
3 Education	2.28	0.92	0.308***	0.082	–
4 Tenure	2.32	1.00	0.020	0.392***	−0.041	–
5 EE	2.66	0.89	0.009	0.036	−0.003	0.063	(0.893)
6 SSG	3.26	1.06	−0.015	0.016	−0.006	0.028	0.031	(0.920)
7 JSE	3.55	0.92	−0.077	0.019	0.089	0.072	−0.097*	0.131**	(0.889)
8 PE	3.30	1.09	0.035	−0.036	0.018	0.001	0.139**	0.263***	0.183***	(0.903)
9 CWB	1.75	1.05	0.004	0.134**	−0.081	0.161**	0.362***	0.072	−0.093	0.204***	(0.932)

Note(s): n = 434. EE = Emotional exhaustion; SSG = Supervisor-subordinate guanxi; JSE = Job self-efficacy; PS = Psychological entitlement; CWB = Counterproductive work behaviors. Diagonal presents Cronbach’s alpha in boldface. ^*p < 0.05, ^**p < 0.01, ^***p < 0.001

Source(s): Authors’ work

Table 2

Results of the confirmatory factor analyses of study measures

Model	χ2	df	χ2/df	Δχ2(Δdf)	CFI	TLI	RMSEA	SRMR
EE, SSG, JSE, PS, CWB	466.421	289	1.61	–	0.983	0.980	0.038	0.031
EE, SSG + JSE, PS, CWB	2665.267	293	9.10	20198.846(4)	0.767	0.741	0.137	0.153
EE, SSG + PS + CWB, JSE	3474.043	296	11.74	3007.622(7)	0.687	0.657	0.157	0.172
EE + SSG + PS + CWB, JSE	5983.745	298	20.08	5517.324(9)	0.441	0.390	0.210	0.212
EE, SSG, JSE, PS, CWB	8159.772	299	27.29	7693.351(10)	0.227	0.160	0.246	0.256

Note(s): n = 434. EE = Emotional exhaustion; SSG = Supervisor-subordinate guanxi; JSE = Job self-efficacy; PS = Psychological entitlement; CWB = Counterproductive work behaviors; CFI = Comparative fit index; TLI = Tucker–Lewis index; RMSEA = Root mean square of approximation; SRMR = Standardized root mean square residual

Source(s): Authors’ work

Table 3

Summary of results of path analyses

Independent variables	Dependent variables
	Model 1		Model 2
	Psychological entitlement (S.E.) [95% CI]	CWBs (S.E.) [95% CI]	Psychological entitlement (S.E.) [95% CI]	CWBs (S.E.) [95% CI]
Intercepts	1.976***(0.327) [1.309, 2.590]	−0.196(0.301) [−0.791, 0.380]	2.726***(0.277) [2.154, 3.240]	−0.167(0.263) [−0.694, 0.339]
Control variable
Gender	0.061(0.105) [−0.153, 0.263]	0.080(0.098) [−0.117, 0.269]	0.121(0.103) [−0.082, 0.327]	0.060(0.098) [−0.136, 0.245]
Age	−0.051(0.059) [−0.168, 0.060]	0.116*(0.056) [0.004, 0.224]	−0.026(0.058) [−0.144, 0.087]	0.117*(0.054) [0.008, 0.220]
Education	0.020(0.059) [−0.100, 0.135]	−0.113*(0.054) [−0.218, −0.009]	−0.005(0.059) [−0.120, 0.114]	−0.097(0.053) [−0.199, 0.008]
Tenure	0.002(0.055) [−0.106, 0.111]	0.100(0.052) [−0.004, 0.200]	−0.024(0.053) [−0.129, 0.084]	0.108*(0.051) [0.004, 0.204]
Emotional exhaustion	0.162**(0.057) [0.046, 0.272]	0.387***(0.060) [0.264, 0.500]	0.188**(0.056) [0.076, 0.298]	0.377***(0.062) [0.251, 0.492]
Independent variable
SSG	0.268***(0.052) [0.167, 0.370]	0.012(0.043) [−0.069, 0.099]	0.251***(0.050) [0.153, 0.348]	0.026(0.043) [−0.054, 0.112]
Mediator
Psychological Entitlement		0.154***(0.041) [0.070, 0.230]		0.155***(0.043) [0.069, 0.235]
Moderator
Job self-efficacy			0.208***(0.055) [0.100, 0.310]	0.107*(0.054) [−0.216, −0.002]
Interaction
SSG *Job self-efficacy			0.201***(0.044) [0.113, 0.284]	0.069(0.043) [−0.015, 0.153]
R2	0.091**	0.190***	0.167***	0.200***

Note(s): n = 434. Unstandardized coefficients were reported. In addition to path estimate, we have also reported the standard error (S.E.) in the parenthesis and the 95% confidence interval [95% CI]

^*p < 0.05, ^**p < 0.01, ^***p < 0.001

Source(s): Authors’ work

Table 4

Conditional indirect effect of SSG on CWBs (via psychological entitlement) at ± 1 SD of job self-efficacy

Conditional indirect effect of SSG	Effect size	Boot SE	LL 95% CI	UL 95% CI
SSG	SSG → psychological entitlement → CWBs
High (+1 SD)	0.067	0.021	0.029	0.114
Low (−1 SD)	0.010	0.011	−0.007	0.036
Difference	0.057	0.021	0.023	0.107

Note(s): n = 434. Bootstrap sample size = 5,000. CI = confidence interval

Source(s): Authors’ work