Abstract
This experiment examined the intersection of socioeconomic status and racial identification in understanding Blacks’ reactions toward Black victims of racial discrimination. When a Black victim of racism was presented as wealthy, rather than non-wealthy, other Blacks viewed this individual as weakly racially identified and expressed little empathy for the victim. This occurred even when this Black individual faced blatant and undeniable discrimination, suggesting that for minorities, the possession of wealth can come at the cost of being perceived as disconnected from one’s racial group and unworthy of the group’s support. The effect of the wealth manipulation on empathy was mediated by the belief that wealthy Blacks are weakly racially identified. The present investigation represents one of the first experimental investigations into the intersection of socioeconomic status and perceived racial identification among Blacks and also provides insight into intragroup dynamics within minority groups.
In the summer of 2009, Black Harvard Professor Henry Louis Gates was arrested while trying to break into his own home in Cambridge, Massachusetts. Following his arrest, the professor took to the media and vehemently declared that he was a victim of racism. Although not everyone agreed with Gates’ construal of the arrest, some of those who disagreed with Gates came from a surprising group: other Blacks. For example, one Black writer reported experiencing little sympathy for Gates because, “the idea that he is the victim of the same racism that sends a poor Black man to jail simply doesn’t fly with me” (Watkins, 2009). Likewise, another Black critic suggested that Gates’ contention that his suffering is comparable to the millions of lower status incarcerated Black men in America is “laughable” (Booker Rising, 2009, p. 1).
More generally, Ellis Cose’s Rage of the Privileged Class (1994) pointedly documented the predicament of race for middle class and wealthy Blacks, including their frustration with fellow Blacks, who they perceived as trivializing their experiences with racism and who were reluctant to view them as “actual” victims of racism. Why would members of a group who have suffered extensive historical and contemporary discrimination and recognize its pervasiveness in all spheres of society (Bertrand & Mullainathan, 2004; Kessler, Mickelson, & Williams, 1999; Markus & Moya, 2010; Myrdal, 1944; Pager, 2003) assume that some in-group members are unworthy of the group’s support when they claim discrimination? In this investigation, we empirically explore this trivialization of wealthy Blacks’ experiences with racism and investigate whether and why it occurs. Specifically, we contend that Blacks view wealthy Blacks as weakly racially identified and that this inference in turn results in decreased empathy toward wealthy Blacks who experience racism.
Wealth and the Denial of Black Identity
Although tensions exist between the rich and poor across all groups in society, this division is particularly pronounced in Black communities (Harris & Khanna, 2010; Lacy, 2002, 2004; Neckerman, Carter, & Lee, 1999). As Blacks are vastly overrepresented among the poor (U.S. Census Bureau, 2012) and being poor is central to the stereotype of being Black (Devine, 1989), Blacks who are wealthy may be perceived by fellow Blacks as different from the prototypical member of their racial group (Turner, 1987). Indeed, derogatory labels such as “sell out” and “whitewashed” and statements to “remember where you came from” are levied by Black toward wealthy Blacks who work and live in predominately White upper class neighborhoods, suggesting that wealthy Blacks’ racial identity is viewed as suspect (Anderson, 1990; Fordham & Ogbu, 1986; Harris & Khanna, 2010; Neckerman et al., 1999).
The perception that wealthy Blacks are weakly racially identified is grounded in historical aspects of U.S. culture. Racial theorists contend that questioning the authenticity of fellow Blacks’ racial identity can be traced to the conflict between house and field slaves in plantation life and was reestablished during the Black Power movement in the late 60s (Austen-Smith & Fryer, 2005; Cole & Omari, 2003). Additionally, the increasingly limited contact between wealthy and non-wealthy Blacks that began following desegregation may have exacerbated the belief that wealthy Blacks are different from other Blacks. For example, children of the most elite Blacks socialize in highly selective Black clubs, such as Jack and Jill, and wealthy Black adults have access to exclusive Black social clubs that are not open to other Blacks (Graham, 1999). Thus, non-wealthy Blacks might conclude that wealthy Blacks perceive themselves as superior to other Blacks, and this division might lead non-wealthy Blacks to view wealthy Blacks as no longer sharing a common fate with Blacks and as out of touch with and unsympathetic toward the plight of Blacks more generally (Neckerman et al., 1999). Indeed, Harris and Khanna (2010) conclude from their interviews with middle-class Blacks that “Middle-class blacks are often perceived as having rejected their ‘true natures’ and therefore, they cannot truly be considered part of the extended black family.”
The reluctance of Blacks to view wealthy Blacks as racially identified falls under the theoretical scope of identity denial. Identity denial is a process “wherein an individual who does not match the prototype of an in-group sees that identity called into question or unrecognized by fellow group members” (Cheryan & Monin, 2005). Identity denial can occur independent of an individual’s actual identity. For example, even though wealthy Blacks may be firmly grounded in their racial identity, other Blacks may not recognize or believe these professions of affinity.
Consequences of Identity Denial
According to identity denial perspectives, only prototypical group members have the ability to construct the meaning of identity and publicly confer identity on other group members (Austen-Smith & Fryer, 2005; Harris & Khanna, 2010; Turner, Hogg, Oakes, Reicher, & Wetherell, 1987). Prototypical group members exert their ownership of the identity construct by rejecting in-group members who display non-prototypical behaviors (Austen-Smith & Fryer, 2005). To the extent that prototypical Blacks “own” the meaning of identity, wealthy Blacks will be viewed as outsiders by their own racial group. For example, in discussing identity denial among racial minorities, “As long as distressed communities provide minorities with their identities, the social costs of breaking free will remain high.” As this statement suggests, acquiring wealth and behaviors that correspond with wealth may come at a risk for Blacks, leaving them vulnerable to rejection from their own racial group. Further, once rejected from a group, rejected individuals reside psychologically outside the boundaries of the group and are not offered the psychological benefits of in-group membership, such as empathy (Cikara, Bruneau, & Saxe, 2011; Cuddy, Fiske, & Glick, 2007). For wealthy Blacks, this means that the judgment that they are weakly racially identified will lead them to experience little empathy from other Blacks.
Blacks’ reluctance to express empathy toward wealthy Blacks may become particularly salient when wealthy Blacks claim to experience racism. Blacks might express little empathy toward the Black elites’ claims of racism, as wealthy Blacks enjoy structural benefits that are out of reach for most Blacks. And, given the manifestation of discrimination in basic needs such as access to housing, health insurance, safe neighborhoods, and jobs, racism claims from those who “have it all” might be viewed as trivial. Further, wealthy Blacks may be viewed as fair-weather group members, attempting to connect themselves to their larger group when it is convenient (when they share a common discriminatory fate) and not in their otherwise everyday activities and institutions.
Present Goals
Despite the existence of survey and anecdotal narratives about wealth and racial identity denial (Anderson, 1990; Neckerman et al., 1999), psychologists have given little empirical attention to the existence and consequences of wealth-related identity denial within the Black community. Further, given the primacy of relationships with fellow in-group members in shaping psychological well-being (Postmes & Branscombe, 2002), it is particularly important to understand factors that lead people to embrace and reject members of their own group. In this investigation, we examine whether Blacks view wealthy Blacks as weakly racially identified and whether this in turn results in a lack of empathy for wealthy Blacks when they report racism.
Overview and Predictions
Black participants read about a high socioeconomic status (SES), low SES, or no SES information control Black plaintiff who claimed that he was the victim of employment discrimination. The attorney for the plaintiff presented blatant or ambiguous evidence of racism. We manipulated the clarity of discrimination, so we could more fully examine our hypotheses about wealth and identity denial. On one hand, one might assume that when a Black person experiences blatant discrimination, other Blacks will band together and perceive all members of the group as sharing a common connection to the group and express empathy toward group members (Branscombe, Schmitt, & Harvey, 1999). On the other hand, if wealthy Blacks are perceived as weakly racially identified, they will be viewed as outsiders to their own group, and will not receive a boost in perceived group identification or receive empathy, even when they experience blatant racism. Instead, once viewed as perpetual outsiders to their own group, even events that seemingly highlight the shared fate of all Blacks (blatant discrimination) will not result in Blacks seeing wealthy Blacks as connected to the in-group or worthy of the group’s emotional support.
Thus, the present study involved a 3 (Black Plaintiff’s SES: high, low, no information) × 2 (Racism Ambiguity: ambiguous, blatant) factorial design. Black participants reported their perceptions of the plaintiff’s Black identification as well as their empathy for the plaintiff. We predicted a main effect of plaintiff SES on racial identification and empathy, such that participants would perceive the wealthy plaintiff as less identified with Blacks and express less empathy toward him relative to the low SES plaintiff and control plaintiff. We also predicted that when the plaintiff was exposed to blatant, rather than ambiguous, racism participants would view him as more racially identified (Branscombe et al., 1999). However, this discrimination-induced increase in perceived identification should occur only for Blacks with low SES or about whom no SES information is presented. Thus, we predict an interaction whereby wealthy Blacks do not get a boost in perceived identification in the presence of blatant discrimination. Finally, the reluctance of Blacks to show empathy for wealthy Black victims of racism should be mediated by perceptions of their (low) racial identity.
Participants
Participants were 236 Black undergraduate volunteers (66% female, largely sophomores and juniors) enrolled at a predominantly Black public university in the southeastern United States. Participants were remunerated with class credit.
Procedure
Participants were instructed to read and make judgments about three passages. Two passages were racism-unrelated (i.e., a woman’s decision about leaving her boyfriend, a man’s choice regarding attending a school near home or taking a scholarship at a far-away university) and were included to reduce demand.
The racism-related passage focused on a racial discrimination lawsuit filed by a Black man, who was employed for 10 years as a manager at a company. The plaintiff claimed he was denied a promotion because of racism. The passage also contained the critical wealth manipulation. Participants learned that the plaintiff lived in an area where homes average $65,000 and that he drove a 2001 Ford Explorer (low SES) or lived where homes averaged $650,000 and drove a 2010 Cadillac Escalade (high SES). The control participants received no SES information.
The passage revealed that the plaintiff’s attorney cited documentation indicating he consistently received well above-average evaluations and was well liked by his fellow managers. A “private internal memo” written by a White senior vice president who made the promotion decisions, included one of two statements. In the ambiguous racism condition, he stated that “Despite his record, I am just not sure that I want to promote this guy to a senior management position.” In the blatant racism condition, he stated that “Despite his record, I am just not sure that I want to promote a Black guy to a senior management position.” Participants imagined being jurors in this lawsuit and completed measures of perceived Black identity of the plaintiff and empathy toward the plaintiff. Participants provided their perceptions of the plaintiffs’ Black identity with an adaptation of Luhtanen and Crocker’s (1992) identity subscale of the Collective Self-Esteem scale (e.g., Being Black is an important part of the plaintiff’s self-image; 1–7 scales with higher scores indicating stronger agreement; α = .87). 1 Participants next reported feelings of empathic concern for the plaintiff with the following 5 items from Batson (1991; Batson, Chang, Orr, & Rowland, 2002): empathy sympathy, compassion, warmth, softheartedness, and being moved (1–7 scales with higher scores indicating stronger agreement; α = .84). Participants were then debriefed.
Results
Manipulation Checks
Thirty independent Black raters read the ambiguous (15 raters) or the blatant (15 raters) racism scenarios. They were asked, “To what extent do the actions in this scenario represent racism? (1 = small extent, 7 = great extent). As expected, the blatant scenario was rated as more racist than the ambiguous scenario, Ms = 6.7 and 2.8, respectively, t(28) = 30.5, p < .001. Thus, racism ambiguity was manipulated successfully.
Thirty independent Black raters were given the low SES information or high SES information. They were asked, “Please give your impression of this person’s socioeconomic status?” (1 = low SES, 7 = high SES). Perceptions of socioeconomic status were higher for high SES person than the low SES person, Ms = 6.21 and 2.10, respectively, t(28) = 35.85, p < .001. Thus, socioeconomic status was manipulated successfully.
Preliminary Analyses
Preliminary analyses indicated that there were no main effects or interactions involving gender on any dependent variables. Thus, gender was not included as an independent variable in subsequent analyses. Primary analyses involved 3 (SES: High, Low, No information) × 2 (Racism Ambiguity: blatant, ambiguous) analyses of variance.
Perceived Black Identity
Consistent with hypotheses that wealthy Blacks are viewed as weakly racially identified, there was a main effect of SES on perceived Black identity, F(2, 230) = 68.96, p < .001. Tukey’s honestly significant difference (HSD) tests revealed that the high SES plaintiff was viewed as less racially identified than the plaintiff in the both the low SES (p < .001, d = 1.36) and control conditions (p < .001, d = 1.62); the latter two conditions were not different from each other (p > .75, d = .10). There was a main effect of racism ambiguity on perceived Black identity, F(1, 230) = 58.25, p < .001, d = .77. The plaintiff facing blatant racism was perceived as more Black identified than the plaintiff facing ambiguous racism. Thus, overall, participants perceive clear victims of racism as strongly connected to their larger racial group.
Also consistent with hypotheses, there was a significant interaction between racism ambiguity and plaintiff SES, F(2, 230) = 9.03, p < .001. We decomposed the interaction in two ways. First, we examined the impact of plaintiff SES within each level of racism ambiguity. Specifically, in the ambiguous racism condition, the wealthy plaintiff was perceived as less identified than the control, F(1, 230) = 22.53, p < .001, d = 1.34, and low SES plantiff, F(1, 230) = 8.02, p < .01, d = .92. Additionally, in the blatant racism condition, the wealthy plaintiff was also perceived as less identified than the control, F(1, 230) = 55.72, p < .001, d = 1.97, and low SES plaintiff, F(1, 230) = 63.45, p < .001, d = 2.11. The interaction highlights how the effect of wealth is stronger in the blatant racism condition.
Second, to more directly test whether the stronger effect of plaintiff wealth in the blatant racism condition stems from increased perceptions of the plaintiff’s identification level in the nonwealthy conditions, we examined simple effects of racism ambiguity within each level of plaintiff SES. These simple effects demonstrated that when racism was blatant rather than ambiguous, the low SES, F(1, 230) = 54.86, p < .001, d = 1.63 and control SES plaintiffs, F(1, 230) = 21.03, p < .001, d = 1.06 were viewed as more racially identified. However, in the high SES condition, there was no impact of racism ambiguity on perceived Black identity, F(1, 230) = 1.6, p > .20, d = .29. Thus, even when wealthy Blacks experience blatant racism, an event that seemingly highlights their similarity to Blacks as whole, other Blacks still viewed them as weakly racially identified.
Empathy
Consistent with hypotheses, there was a main effect of SES on empathy, F(2, 230) = 49.8, p < .001. As can be seen in Table 1, Tukey’s HSD tests revealed that participants reported less empathy toward the high SES plaintiff relative to both the low SES plaintiff (p < .001, d = 1.34) and the control plaintiff (p < .001, d = 1.28), which were not different from each other (p > .65, d = .10). This is consistent with the argument that Blacks withhold empathy from high SES in-group members who claim racism. There was also a main effect of racism ambiguity on empathy, F(1, 230) = 50.64, p < .001, d = .78. Those exposed to a plaintiff facing blatant racism reported greater empathy than those exposed to a plaintiff facing ambiguous racism. Thus, overall, increasing evidence of racism engendered greater empathy for an in-group member who claimed discrimination.
Perceived Black Identity and Empathy as a Function of Racism Ambiguity and Plaintiff SES.
Note. Standard deviations are in parentheses.
Further consistent with hypotheses (see Table 1), the Racism Ambiguity × Plaintiff SES interaction was significant, F(2, 230) = 4.82, p < .01. We first examined simple effects within each level of racism ambiguity. In the ambiguous racism condition, participants reported less empathy for the wealthy plaintiff relative to the control, F(1, 230) = 13.1, p < .001, d = .98 and low SES plantiff, F(1, 230) = 15.04, p < .001, d = 1.10. In the blatant racism condition, participants also reported less empathy for wealthy plaintiff relative to the control, F(1, 230) = 54.76, p < .001, d = 1.72 and low SES plaintiff, F(1, 230) = 67.24, p < .001, d = 1.86. The interaction revealed that the wealth effect was more pronounced in the blatant racism condition.
To more directly examine our hypotheses, we conducted simple effects tests examining the effects of racism ambiguity within each level of plaintiff SES. These tests demonstrated that participants expressed increasing levels of empathy when the in-group member experienced blatant compared to ambiguous racism when the plaintiff was both low in SES, F(1, 230) = 32.83, p < .001, d = 1.17, and when they had no information about the plaintiff’s SES, F(1, 230) = 26.28, p < .001, d = 1.16. In contrast, when the plaintiff was from a high SES group, participants were insensitive to the ambiguity of racism. Participants were no more sympathetic toward a wealthy in-group discrimination claimant when he faced blatant racism compared to ambiguous racism, F(1, 230) = 2.4, p = .12, d = .37. Thus, even when wealthy Blacks experience clear racism, fellow in-group members do not express empathy.
Perceived Black Identity Mediates the Effects of SES on Empathy
We tested for whether the interaction between target SES (coded as high SES = 1, aggregate low SES and control = 0) and racism ambiguity was mediated by perceived racial identity. Because there was no difference between the low SES and control conditions, we combined them for these analyses. Supportive of mediation, the interaction effect (which was entered into the equation after the main effects of target SES and racism ambiguity) predicted the dependent variable, empathy, in the first equation (β = .63, p < .01) and the potential mediator, perceived racial identity in the second equation (β = .66, p < .001). In the third equation, perceived racial identity predicted empathy (β = .61, p < .001). Further, the impact of the interaction on empathy was reduced to nonsignificance (β = .19, p > .25). Sobel tests demonstrated that the impact of the interaction was significantly attenuated when perceived racial identity was in the equation, (z = 3.59, p < .001). 2
Discussion
The present study converges around several conclusions. First, Blacks perceive wealthy Blacks as weakly racially identified compared to non-wealthy Blacks. This is consistent with perspectives on identity denial which contend that prototypic group members yield power over the construct of identification and do not confer identification to group members who are non-prototypic of the group (Austen-Smith & Fryer, 2005; Harris & Khanna, 2010). The racism ambiguity findings provide further evidence that wealthy Blacks experience identity denial. Specifically, as it becomes clearer that non-wealthy Blacks have experienced racism, other Blacks rally around those individuals, perceiving them as part of the in-group. However, this rallying around the in-group is remarkably absent when the victims of racism are wealthy. We believe that this occurs because Blacks perceive wealthy Blacks as outsiders to their own group, and are thus unwilling to confer a sense of identity on those not squarely within the prototypic boundaries of the group.
This study also demonstrates the consequences of identity denial for wealthy Blacks. Black participants expressed less empathy for wealthy Black victims of racism relative to non-wealthy victims of racism. Additionally, empathy was not enhanced when it was obvious that wealthy Blacks experienced racism, suggesting that wealthy Blacks are viewed as outsiders to their own group. Evidence for this comes from the data showing that perceived (low) racial identification mediated the effects of the wealth manipulation on empathy toward the plaintiff.
These findings converge with survey data from sociology and interviews with Blacks, both documenting wealthy Blacks’ perceptions that other Blacks see them as less racially identified and take their racism claims less seriously, compared to nonwealthy Blacks (Cose, 1994; Feagin & Sykes, 1994; Harris & Khanna, 2010). Further, while evidence from other fields focuses primarily on the narratives of wealthy Blacks, the present data experimentally point to the reality of this dilemma; namely that other Blacks do in fact discount wealthy Blacks’ experiences with racism relative to the same claims made by nonwealthy Blacks. While Black participants readily embraced non-wealthy Black professionals who filed a racism claim, they offered less support to wealthy individuals who experienced the exact same racist event. Further, the fact that this effect was observed with Black college students, a group that is on average significantly more privileged economically than other Blacks (Massey, Mooney, Torres, & Charles, 2007), suggests that this dilemma may be even more pronounced than we could observe with this sample.
Although the findings of this study are consistent with hypotheses, one might question whether members of other racial groups would also withhold empathy from wealthy Blacks and perceive them as less connected to Blacks as a whole. Although the reactions of racial out-group members are certainly interesting from an empirical standpoint, from a theoretical standpoint, it is the reactions of fellow in-group members that are central to understanding intragroup processes concerning wealth and race. Additionally, the psychological struggle to maintain authenticity in one’s Black identity is inherently an intragroup phenomenon, and Blacks’ identity-based concerns are unlikely to be the same factors that motivate racial out-groups’ behavior toward Blacks (Oyserman, Fryberg, & Yoder, 2007). Further, even if out-group members do withhold empathy from wealthy Blacks, the mechanisms guiding this behavior are unlikely to concern authenticity with Black identity.
The present data have implications for the intersection of wealth and race. While sociologists and historians have investigated the role of wealth in Black communities (e.g., Cose, 1994; Harris & Khanna, 2010; Lacy, 2002, 2004), psychologists have not. This lack of attention to wealth can result in erroneous conclusions about intragroup behavior among minorities, such as Blacks. Further, our research points to the precarious predicaments that wealthy Blacks and others with high status may experience. For example, wealthy Blacks may find themselves experiencing the need to provide evidence of their racial identification to other Blacks, and may be reluctant to claim discrimination because they are so well off (Harris & Khanna, 2010). As the suppression of racism is psychologically and physiologically stressful (Krieger & Sydney, 1996), this can have adverse consequences for these Blacks. Further, as discrimination occurs at every level of social class, this leaves wealthy Blacks potentially experiencing rejection from both Whites and fellow minorities when they claim discrimination (Feagin & Sikes, 1994; Harris & Khanna, 2010; Kaiser & Miller, 2001, 2003).
In addition to contributing toward a greater understanding of wealth and race, this experiment contributes to research on group-based empathy. At first glance, the finding that individuals do not always empathize with some members of their own group seems at odds with the robust findings in the empathy literature; specifically, that individuals show high degrees of empathy for in-group members across a wide variety of situations (Batson & Ahmad, 2009; Brown, Bradley, & Lang, 2006; Cikara, Bruneau, & Saxe, 2011; Mathur, Harada, Lipke, & Chaio, 2010; Roberts & Levenson, 2006; Xu, Zuo, Wang, & Han, 2009).
Although empathy toward one’s in-group is a dominant pattern, its emergence is based upon the notion that groups are cohesive and unified entities whereby all in-group members are treated as interchangeable group exemplars (Tajfel & Turner, 1986). However, as our data demonstrate, individuals do not necessarily psychologically include all other individuals who belong to their group as bona fide in-group members. It is perceptions of psychologically belonging to the in-group, rather than actual in-group membership, which are critical in understanding when minorities will respond with intragroup empathy and when they will not. Thus, this study demonstrates how broad conclusions about the prevalence of in-group-directed empathy are likely overstated. The literature would benefit by further incorporating perspectives highlighting how psychological group membership is more complex than simply belonging to a given category (Saperstein, 2006).
Conclusion
Although poverty is a major impediment to upward mobility among Blacks, wealth also poses problems for Blacks. While wealth can buy resources and make for a good material life, it can leave minorities vulnerable to rejection from other members of their group. As upward mobility is part of the American Dream, it will be important to identify strategies that facilitate this pursuit among minorities without the repercussion of being viewed as disconnected from their race. Further, given the divisiveness of wealth within minority communities (Harris & Khanna, 2010; Lacy, 2002, 2004), psychological science would benefit by understanding how wealth shapes the experiences of minorities.
Footnotes
Declaration of Conflicting Interests
The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding
The author(s) disclosed receipt of the following financial support for the research and/or authorship of this article: Kaiser’s effort was supported by NSF BCS-0749159.