Narcissism is a complex, multidimensional personality construct, but the narcissist label is perhaps applied most frequently as a complaint against people who habitually engage in actions that bother others. Narcissists antagonize others with overt, self-aggrandizing claims of superiority (e.g., Buss and Chiodo 1991), offensive language (e.g., Adams et al. 2014; DeWall et al. 2011), aggression (e.g., Bushman and Baumeister 1998; Reidy et al. 2010), and selfish, unethical decisions (e.g., Brown et al. 2010; Williams et al. 2010). In short, the evidence that narcissists act in socially inappropriate ways is hard to dispute; however, it is less clear how narcissism predicts responses to other people’s bad behavior. The present investigation examined the relationship between narcissism and intolerance for others’ antisocial actions. Specifically, our research tested whether narcissists are unfairly intolerant in the sense that they behave badly but do not show proportional acceptance of the same bad behavior from others.

Defining Narcissism and Unjust Intolerance

In the context of this paper, narcissism represents a personality trait present in normal people to varying degrees; we use the “narcissist” label and its variants to describe people on the higher end of the narcissism continuum. Except where otherwise specified, we define narcissism narrowly as grandiose narcissism, a type of narcissism that differs in meaningful ways from vulnerable narcissism (see Miller and Campbell 2008; Pincus and Roche 2011 for reviews of the distinctions between grandiose and vulnerable narcissism). More specifically, the new and prior narcissism evidence we cite was derived from studies that assessed narcissism with the self-report Narcissistic Personality Inventory (NPI; Raskin and Hall 1979), which has been the most commonly used measure of grandiose narcissism (see Miller and Campbell 2011 for discussion of the merits and drawbacks of the NPI).

Our definition of unjust intolerance applies when reported levels of tolerance for antisocial behavior are proportionally smaller than reported willingness to engage in the same antisocial behaviors. For example, people with high intolerance for others’ antisocial behavior are not unjustly intolerant if they are also highly unlikely to personally engage in antisocial behavior. In the case of narcissists, we know that they behave badly, but our study investigated whether they show sufficient tolerance for others’ bad behavior to avoid the unjust intolerance indictment.

Mixed Evidence for Narcissists’ Social Intolerance

Examples of evidence linking narcissism with social intolerance are plentiful. Narcissism is associated with feelings of superiority (e.g., Krizan and Bushman 2011) and the tendency to denigrate others (e.g., Campbell et al. 2000; Horton and Sedikides 2009; Park and Colvin 2015). Compared with others, narcissists are more disagreeable (e.g., Holtzman et al. 2010; Paulhus and Williams 2002), adversarial (e.g., Rhodewalt and Morf 1995), and prone to reactive hostility (e.g., Kernis and Sun 1994). Narcissists like their social partners less than others do (e.g., Lamkin et al. 2014) and they have unrealistic expectations of others’ behavior (e.g., Sherry et al. 2014). Narcissists are more prone to feeling victimized (McCullough et al. 2003) and are less willing to forgive others’ transgressions against them (e.g., Brown 2004; Exline et al. 2004). Narcissism also predicts more self-reported trait anger (Penney and Spector 2002) and higher biological stress responses to everyday frustrations (Cheng et al. 2013; Edelstein et al. 2010). Nonetheless, despite this lengthy list of studies highlighting variants of narcissists’ social intolerance, other evidence indicates that the relationship between narcissism and social intolerance is not straightforward.

It is apparent that narcissists are more critical than others of people and behaviors that do not obviously warrant criticism. It is also evident that narcissists are more inclined than others to get angry at and attack people who directly threaten their grandiose self-image. But it is not clear whether narcissists are more prone than others to object to behavior that is socially inappropriate, yet is unlikely to be viewed as a personal attack or experienced as an ego threat. Our investigation focused on narcissists’ responses to the mundane antisocial behaviors that ordinary people routinely display. The evidence that sheds light on the relationship between narcissism and responses to this particular category of bad behavior paints a different picture of narcissists’ intolerance.

Consider the connection between the narcissism of person perceivers and the narcissism of the people being perceived. Narcissists are judged less harshly by people who are also narcissistic (Hart and Adams 2014; Wallace et al. 2015). This finding fits the well-documented similarity-attraction principle: Individuals tend to respond more favorably to the people they resemble in some way (Griffitt 1966; Montoya and Horton 2012). This principle accounts for evidence that narcissists show less interpersonal aggression after experiencing an ego threat if they recognize that they share something in common with the target of their aggression (Konrath et al. 2006). In the context of the present investigation, willingness to engage in bad behavior is a shared characteristic that could make narcissists less critical in their judgments of others’ bad behavior.

Similarity principle aside, narcissists’ bad behavior highlights a value system that places low priority on prosocial actions. Narcissists claim and strive for superiority in agentic domains such as those that allow demonstration of competence and achievement, but they seem to care less than others about their standing within more communal dimensions of life, such as social contexts in which moral behavior is expected and reinforced (Campbell and Foster 2007; Campbell et al. 2002; Jones and Brunell 2014). If narcissists are not bothered by their own bad behavior, perhaps they are not so bothered by the bad behavior perpetrated by others. This possibility received support from a study by Lustman et al. (2010), in which narcissists did not take offense to other people’s aggressive automobile driving (see Edwards et al. 2013 for confirmation that narcissists drive aggressively). Similarly, Adams et al. (2015) found that evaluator narcissism levels did not significantly predict their likability judgments of hypothetical people described as having engaged in concrete examples of narcissistic behaviors. In addition, Kammrath and Scholer (2011) reported that intolerance of others’ antisocial actions was positively correlated with agreeableness, a quality that is negatively correlated with narcissism. This result suggests that the people who are most offended by others’ antisocial actions may be those with strong moral convictions regarding the importance of following rules of social decorum (for related evidence, see Skitka and Morgan 2009). Narcissists readily admit to possessing loose standards for interpersonal morality (e.g., Buss and Shackelford 1997), so it seems unlikely that they would respond to others’ bad behavior with self-righteous indignation.

Narcissists are interpersonally insensitive in terms of their relative lack of concern for the impact of their actions on others’ well-being (e.g., Raskin and Terry 1988; Watson and Morris 1991). They may also be insensitive in terms of not paying close attention or registering an emotional response to negative events in their social environment that do not directly threaten their self-esteem or block avenues for self-aggrandizement (e.g., Adams et al. 2014; Collins and Stukas 2008). This perspective aligns with evidence that grandiose narcissism is positively correlated with self-serving attributional bias (Farwell and Wohlwend-Lloyd 1998; Stucke 2003), uncorrelated with envy (Krizan and Johar 2012) and negatively correlated with overall life frustration (Schnieders and Gore 2011). Narcissists have low tolerance for experiences that frustrate them, but they are apparently not more likely than others to feel frustrated by their experiences.

In sum, some prior research hints that narcissists may be relatively tolerant of others’ antisocial behavior, but this evidence is less consistent and strong than the prior documentation of narcissists’ willingness to act badly. This evidence imbalance supports the prediction of a positive relationship between narcissism and unjust intolerance. Unjust intolerance is by definition a violation of fairness, so additional support for the hypothesized link between narcissism and unjust intolerance could be drawn from evidence that narcissists are more willing than others to tip the scales of justice for personal gain (e.g., Brunell et al. 2011; Campbell et al. 2005). Narcissists’ willingness to claim benefits at others’ expense is facilitated by their inflated sense of entitlement (e.g., Emmons 1984), their low empathy (e.g., Watson et al. 1984) and low remorse (e.g., Strelan 2007), and their comfort with the relationship consequences of self-induced social conflict (e.g., Rose and Anastasio 2014).

Study Overview

Our study provided a straightforward test of the relationship between narcissism and unjust bad behavior intolerance. To date, research confirming narcissists’ willingness to act badly has generally been conducted separately from the research that has measured narcissists’ responses to others’ bad behavior. We took a different approach by asking participants to report their willingness to engage in a list of common antisocial behaviors, and then asking them to report the extent to which they would feel bothered upon encountering the same behaviors—perpetrated by someone else. Because participants had to reflect upon their own behavior tendencies first, those who reported both willingness to behave badly and intolerance of bad behavior were likely to be conscious of unfairness conveyed by their responses. Unfairness is frowned upon, so it should appear less often in self-report measures when people are induced to confront their unfairness (e.g., Batson et al. 1999). Of course, enhancing the salience of unfairness may only reduce unfair behavior for people who strive to maintain moral standards of fairness, or at least to give the impression of doing so. As we have noted, narcissists are comparatively unapologetic about not playing by the rules, in part because they care less about whether their views and actions are socially acceptable (Watson and Morris 1991).

Method

Participants

Undergraduate college students participated for introductory psychology course credit (N = 219; 65 % female; M age = 18.86; 75 % White, 11 % Asian, 10 % Hispanic). All statistics in this paper exclude six participants who started but did not complete the questionnaire.

Materials

Self-Report Trait Measures

Narcissism was assessed with the 40-item forced-choice version of the NPI (Raskin and Terry 1988; M = 14.52, SD = 7.03; Cronbach’s α = .85). We also isolated NPI subfactors, defined by Ackerman et al. (2011), of leadership/authority (11 items; M = 4.92, SD = 2.90; α = .77), grandiose exhibitionism (10 items; M = 3.34, SD = 2.50; α = .74), and entitlement/exploitativeness (4 items; M = 1.02, SD = 1.09; α = .49). In addition, participants completed measures of other personality dimensions germane to narcissism and possibly to the outcomes of interest, including the 10-item Rosenberg (1965) self-esteem scale (1 to 7 response scale; M = 50.43, SD = 10.06; α = .88), the 9-item Psychological Entitlement Scale (Campbell et al. 2004; 1 to 7 response scale; M = 29.95, SD = 8.92; α = .87), and the 20-item Mach IV (Christie and Geis 1970), an index of Machiavellianism, which entails willingness to deceive and manipulate others for personal goal attainment (1 to 5 response scale; M = 57.69, SD = 7.47; α = .70). The NPI typically correlates positively with self-esteem (e.g., Brown and Zeigler-Hill 2004), psychological entitlement (e.g., Campbell et al. 2004), and Machiavellianism (e.g., Biscardi and Schill 1985).

Willingness to Behave Badly

We created an index of willingness to behave badly by computing the mean response to 25 items that invited participants to state their willingness or tendency to engage in selfish, rude, hostile, or unethical behaviors, or their capacity to avoid feeling bad about bothering others. We tried to generate items that addressed examples of common minor offenses that most people would view as inappropriate, but not uncommon, outrageous, or unforgiveable. We did not strive to select or avoid offenses according to the extent to which they seemed narcissistic. Item responses consisted of agreement ratings on a 7-point scale (1 = strongly disagree, 7 = strongly agree) and some items were reverse-scored so that high ratings always reflected more willingness to behave badly (M = 2.84, SD = 0.65; α = .81).

Bad Behavior Intolerance

We created an index of bad behavior intolerance by computing the mean response to 25 items that invited participants to state the extent to which they would feel bothered in response to others’ antisocial actions. Each item included the prefix “It would really bother me if…,” followed by a description of an example of another person’s behavior that matched the behavior described in one of the bad behavior intolerance items. Intolerance item responses consisted of agreement ratings on a 7-point scale (1 = strongly disagree, 7 = strongly agree; M = 5.32, SD = 0.63; α = .86). The Appendix provides the text of each willingness to behave badly item and its companion bad behavior intolerance item, and it displays the mean responses and the correlations between item responses and NPI scores.Footnote 1

Unjust Intolerance

The unjust intolerance label was applied to participants who reported willingness to engage in bad behavior and intolerance of others’ bad behavior. To avoid this label, participants with high willingness scores would need to also have low intolerance scores, and participants with high intolerance scores would need to also have low willingness scores. The unjust intolerance index was created by computing the mean of participants’ willingness and intolerance scores, and then subtracting the response scale midpoint of 4 from the mean. Unjust intolerance scores near zero reflected justified intolerance; relatively high, positive scores reflected high levels of unjust intolerance. Relatively low, negative scores reflected the opposite of unjust intolerance: low willingness to behave badly, combined with low intolerance of others’ bad behavior. The mean unjust intolerance score was slightly lower than zero (M = −0.16. SD = 0.79), one-sample t(218) = −3.05, p = .003, indicating that participants overall did not report unjust intolerance. We did not standardize willingness to bother and bothered by others scores before computing unjust intolerance scores because the variance of willingness to behave badly scores and bad behavior intolerance scores was similar (3.9 % difference), and because miniscule differences were observed in the unjust intolerance results when the unjust intolerance index was based on the combined z-scores of bad behavior willingness and bad behavior intolerance scores.

Procedure

At the time and place of their choosing, participants opened a web link to access the online questionnaire. After giving consent, participants reported their age, gender, and ethnicity. Then they completed all of the willing to behave badly items, before proceeding to the bad behavior intolerance items. Finally, participants completed the NPI and the other individual difference measures.

Results

Table 1 provides a matrix of the zero-order correlations between the dependent variables and the individual difference traits. Narcissism was significantly correlated with self-esteem and entitlement, but not with Machiavellianism. Male participants were more narcissistic (M = 16.32, SD = 7.20) than female participants (M = 13.58, SD = 6.74). The three NPI subfactors were consistent in the direction of their relationships with each dependent variable and personality measure, with the exception of self-esteem. Self-esteem was positively correlated with leadership/authority and grandiose exhibitionism, but was negatively correlated with entitlement/exploitativeness.

Table 1 Zero-order correlations between judgment outcomes and individual difference variables
Full size table

Willingness to Behave Badly

As expected, narcissism was positively correlated with the willingness to behave badly index (see Fig. 1). Narcissism remained a significant, independent predictor of willingness to behave badly (β = 0.32, SE = 0.006, p < .001) in a multiple regression model (adjusted R 2 = .32) that included the other four individual difference variables (self-esteem, entitlement, Machiavellianism, and gender). The other independent predictors of willingness to behave badly in this regression model were gender (men were more willing than women; β = −0.24, SE = 0.082, p < .001), Machiavellianism (β = 0.23, SE = 0.005, p < .001), and self-esteem (β = −0.18, SE = 0.004, p = .008).

Fig. 1
figure 1

Relationship between narcissism and self-reported willingness to behave badly

Full size image

Each of the three NPI subfactors was also correlated with willingness to behave badly. When the subfactors were entered simultaneously as predictors in a multiple regression model (adjusted R 2 = .29), entitlement/exploitativeness (β = 0.46, SE = 0.037, p < .001) and grandiose exhibitionism (β = 0.19, SE = 0.017, p = .004) independently predicted willingness to behave badly but leadership/authority did not (β = 0.01). When the other four individual difference variables were added to the regression model (adjusted R 2 = .39), the grandiose exhibitionism and entitlement/exploitativeness NPI subscales remained significant independent predictors, along with gender and Machiavellianism.

Bad Behavior Intolerance

Neither narcissism nor its subfactors were correlated with the bothered by others’ bad behavior index (see Fig. 2). In a multiple regression model that included narcissism and the other four individual difference variables (adjusted R 2 = .11), only gender emerged as a significant independent predictor of bad behavior intolerance (women were more intolerant; β = 0.35, SE = 0.089, p < .001). None of the three NPI subfactors was a significant, independent predictor of being bothered when the three subscales were entered simultaneously as multiple regression model predictors (ps > .46).

Fig. 2
figure 2

Relationship between narcissism and self-reported intolerance of others’ bad behavior

Full size image

Unjust Intolerance

Narcissism was positively correlated with the unjust intolerance index (see Fig. 3) and it remained an independent unjust intolerance predictor (β = 0.34, SE = 0.008, p < .001) in a multiple regression model that included the other four individual difference variables (adjusted R 2 = .20). The other independent unjust intolerance predictors in this model were self-esteem (β = −0.21, SE = 0.005, p = .004) and Machiavellianism (β = 0.15, SE = 0.007, p = .030).

Fig. 3
figure 3

Relationship between narcissism and the unjust intolerance index

Full size image

Each of the NPI subfactors was separately correlated with the unjust intolerance index. When the subfactors were entered simultaneously as unjust intolerance predictors in a multiple regression model (adjusted R 2 = .22), entitlement/exploitativeness was found to be an independent predictor (β = 0.41, SE = 0.047, p < .001), but not grandiose exhibitionism (β = 0.11, SE = 0.022, p = .095) or leadership/authority (β = 0.05). When the other four individual difference measures were added to this regression model (adjusted R 2 = .24), the only significant independent unjust intolerance predictors were the entitlement/exploitativeness and grandiose exhibitionism subfactors.

Discussion

Two primary messages about narcissism can be drawn from the results of our study. One is that narcissists report being no more or less tolerant of everyday antisocial behavior than other people. This finding offers a caveat to the link between narcissism and disagreeableness: Prior research has shown that narcissists tend to view others unfavorably without cause, and to show hostility toward those who pose a threat to their self-esteem; however, our findings reinforce the results of recent studies showing that narcissists are not uniquely perturbed by others’ violations of social standards (Adams et al. 2015; Lustman et al. 2010) or others’ narcissistic traits (Hart and Adams 2014; Wallace et al. 2015).Footnote 2 Feeling bothered by commonplace antisocial behavior is a conventional response, not a symptom of narcissism.

The second and most novel major message of our study highlights an implication of narcissists’ responses to others’ antisocial behavior: In the context of narcissists’ own bad behavior proclivities, their level of intolerance for others’ bad behavior could be viewed as unjustifiably high. Narcissism did not simultaneously predict more willingness to act badly and more intolerance for others’ bad behavior, but narcissists were unfairly intolerant nonetheless in the sense that their intolerance was high in proportion to their level of willingness to engage in the same behavior. Furthermore, because the design of our study led participants to state their willingness to behave badly shortly before judging how bothered they would feel in response to the same behavior from others, narcissists confessed to unjust intolerance despite having access to the reality of their unjust intolerance. The relationship between narcissism and the unjust intolerance index remained strong even when controlling for other measures of personality traits related to narcissism that also predicted unjust intolerance.

Our research was designed to study outcomes of narcissism, not to identify different individual difference predictors of willingness to behave badly, intolerance of others’ bad behavior, or unjust intolerance; however, some ancillary findings merit further attention. Although narcissism was the strongest predictor of willingness to behave badly and of unjust intolerance (controlling for other variables or not), the effects of narcissism were closely mirrored by the effects of psychological entitlement and Machiavellianism—including a lack of relationship with claimed intolerance of others’ bad behavior. In contrast, the relationship between self-esteem and the dependent variables showed a very different pattern, despite the positive relationship between self-esteem and total NPI scores. Inflated self-esteem may account for some forms of narcissistic behavior, but it does not explain narcissists’ unjust intolerance. High self-esteem actually predicted the opposite of unjust intolerance: High self-esteem participants were relatively tolerant of others’ bad behavior even though they also reported unwillingness to behave badly. The opposing outcomes of self-esteem and narcissism can be traced to the combination of the negative correlation between self-esteem and the entitlement/exploitativeness narcissism subfactor, and the fact that this subfactor explained the most variance in the unjust intolerance scores.

Entitlement and exploitativeness have historically been viewed as the most maladaptive elements of narcissism, whereas the leadership and authority features are less problematic and sometimes advantageous (e.g., Reidy et al. 2010; Watson and Morris 1991). The entitlement/exploitativeness part of the NPI can be viewed as a small island (with inadequate psychometric validity) of vulnerable, interpersonally toxic narcissism within a narcissism instrument dominated by items that measure aspects of a more grandiose, high self-esteem version of narcissism (e.g., Ackerman et al. 2011). The relationship between narcissism and unjust intolerance may prove to be even stronger, due to heightened intolerance of others, if narcissism is measured with an instrument specifically designed to assess traits of vulnerable narcissism (for evidence that narcissistic rage only applies to vulnerable narcissism, see Krizan and Johar 2015). Similarly, the magnitude of the relationship between narcissism and unjust intolerance would surely increase if narcissists judged bad behavior that directly threatened their self-esteem (e.g., Exline et al. 2004; Schnieders and Gore 2011; Stucke and Sporer 2002).

The gender effects observed in our study were also noteworthy. Male participants were more narcissistic than female participants—a common finding (see review by Grijalva et al. 2015)—but gender was unrelated to the unjust intolerance index. Compared with men, female participants were far less willing to act badly than men, and they were equally less tolerant of others’ bad behavior. In our view, these effects should be interpreted narrowly as an indication of the nature of the antisocial behaviors that participants were asked to judge. Women have exhibited less antisocial behavior than men in most studies (e.g., Archer 2004; Gallus et al. 2014), but women are not inevitably less prone to bad behavior in all situations (e.g., Eagly and Steffen 1986). Moreover, women have been found to forgive relationship transgressions to an equal or greater extent than men (Fehr et al. 2010; Miller et al. 2008), though women are comparatively less tolerant of unethical behavior in workplace contexts (e.g., Borkowski and Ugras 1998; Mudrack and Mason 2013). In short, the substantial gender differences observed in our study probably signal that participants were asked to judge instances of bad behavior that are more typically perpetrated by men. Most of the bad behavior scenarios could be described as examples of excessive assertiveness, a quality that is more characteristic of men than women (see review by Feingold 1994). Furthermore, 9 of the 25 bad behaviors pertained to automobile driving—a domain with unusually strong gender differences in attitudes and behavior (Social Issues Research Centre (2004). More research will be necessary to clarify how the specific nature of the bad behavior in question may influence the relationship between individual difference variables and unjust intolerance.

It would also be ideal, albeit more difficult, to assess bad behavior inclinations and attitudes through third party observations or direct behavioral or physiological measurement rather than relying upon self-report measures. Participants may have misjudged or misreported the extent to which they would have actually taken offense to real-life exposure to some examples of others’ bad behavior. Indeed, our interpretation that ego threat was unlikely to have influenced narcissistic participants’ judgments is partly based on the fact that the scenarios were hypothetical, and therefore more psychologically distant. Although we selected antisocial behavior scenarios that seemed more annoying than offensive, some of these behaviors could induce anger or reactance if experienced in real life.

It is also possible that the magnitude of the relationship between narcissism and unjust intolerance was influenced by our method of always asking participants to report their own behavioral tendencies before judging how they would respond to others’ behavior. This fixed-order approach presumably increased the likelihood that participants would recognize unjust tolerance in their judgments, but less narcissistic participants may have felt less comfortable confessing to violating social standards of fairness. Narcissism has unsurprisingly been linked with more dishonesty in some contexts (e.g., Brunell et al. 2011), but in our study, social desirability bias may have been more evident in the responses of participants with low narcissism. Alternatively, the fixed-order procedure may have actually suppressed narcissism effects by pulling narcissists out of their default state of self-absorption. Perhaps only narcissists would require a prompt to consider the connection between their judgments of self and others (see Rauthmann 2012 for evidence that narcissists’ self-judgments are relatively independent of their judgments of others). Also, reporting their own behavior first could have activated dormant empathy within narcissists, whereas less narcissistic people may have shown empathy regardless (see Hepper et al. 2014 for evidence that narcissists are capable of empathy, despite rarely displaying it).

In summary, our findings fill gaps in the grandiose narcissism literature and provide a platform for novel future research. Our research demonstrates that narcissists are not uniquely critical of routine antisocial behaviors, and it provides the first direct comparison of narcissists’ willingness to bother and their tolerance of being bothered. Narcissists’ unjust intolerance of bad behavior may not deserve a high rank of infamy within the pantheon of narcissistic faults, but it helps to provide a clearer picture of the mindset behind narcissists’ bothersome behavior.