“Diversifying evil doesn’t change the meaning of the noun.”—Eddie S. Glaude Jr., Distinguished University Professor of African American Studies at Princeton University, January 27, 2023.

In the summer of 2020, when public support and protests for the Black Lives Matter movement surged following the murder of George Floyd by Minneapolis Police, companies scrambled to issue statements that condemned anti-Black racism and expressed their commitment to diversity, equity, and inclusion (DEI). Many of these companies, however, were called out for hypocrisy given their racist patterns of hiring, promotion, and leadership. Such companies and leaders included the usual suspects, like Amazon and its CEO Jeff Bezos and Nike and its CEO John Danhoe (Ho, 2020). Some women-led companies and their leaders were also called out, however, such as Band.do and its founder Jen Gotch and Refinery29 and its editor-in-chief Christine Barberich (Hinchliffe, 2020). In contrast to companies led by white men, racism in companies led by white women was often met with surprise: women, after all, are expected, and tend, to openly endorse and express more egalitarian, less racist, and more caring ethics than men (Derry, 2017; Eagly & Karau, 1991; Ekehammar et al., 2003; Ekehammer & Sidanius, 1982; Maxwell, 2015; Pullen & Vachhani, 2021; Sidanius et al., 1994). For this reason, women are often assumed to be better leaders than men when it comes to managing DEI at work (Ely & Meyerson, 2000; Ely & Padavic, 2007; Joshi et al., 2015). But what if the women who rise to leadership in corporate America are more likely than the “typical” woman to support racial inequality?

In this paper, we propose that women who express anti-Black racism are more likely than women who do not to be perceived as “having what it takes” to succeed in corporate America, and are conferred more status and power as a result. Drawing on research and theory on organizational culture and intersectional feminism, we argue that ongoing doubts about women’s leadership abilities (Braun et al., 2017; Derry, 2017; Eagly & Karau, 2002; Hogue & Lord, 2007; Ibarra et al., 2013) lead people to look for signs that a woman is capable of fitting and succeeding in corporate culture. In the US, corporate culture is rooted in racial and gender oppression, upholding masculinity and whiteness as ideal employee characteristics and norms (Acker, 1990; Bell, 2020; Berdahl et al., 2018; Logan, 2019; Rosette et al., 2008). Anti-Black racism may go against official organizational policy and mission statements, but such prejudice is widely and implicitly embedded within modern workplace cultures (e.g., Bell, 2020; Levin, 2017; Logan, 2019; Nkomo, 2021; Rosette et al., 2013). Indeed, corporations play an important role in shaping race relations (Logan, 2019), rewarding their members for displaying behaviors and beliefs in line with white male corporate culture (Berdahl et al., 2018; Glass & Cook, 2020; Roberts, 2005). Thus, just as women can gain status at work for engaging in sexist behavior that reinforces male dominance (Alonso & O’Neill, 2021; Derks et al., 2011; Kanter, 1977; Staines et al., 1974), we propose that women can gain status at work for engaging in anti-Black racist behavior that reinforces white supremacy.

To test this prediction, we examine how employees who engage in anti-Black racism within the workplace are perceived, focusing our attention on white women, who occupy most of the few leadership positions held by women in the corporate world (Catalyst, 2022). With a series of experimental studies, we test whether white women who express anti-Black racism gain status relative to white women who do not. We first conducted a pilot study to test whether anti-Black racist speech is viewed as less offensive when it comes from a white woman than from a white man (it is). In Study 1, we examined whether anti-Black racist speech leads to relative status conferral for a white woman at work (it does). In Study 2, we tested whether anti-racist/pro-Black Lives Matter speech leads to status loss for a white woman at work (it does). Finally, in Study 3, we examined whether perceivers’ own racist and sexist attitudes help to explain these effects (they do, but not in the way we expected).

Our paper expands current understanding of how workplace culture can shape ethical or unethical business practices. If the women who are most likely to gain status in corporate America reinforce not only gender but racial inequality, then prevailing assumptions about women as a solution to companies’ DEI efforts are sorely misguided, with important implications for business ethics. Such women are likely to undermine rather than support DEI in the workplace, to the detriment of organizations and their stakeholders, particularly their marginalized ones (e.g., Bell et al., 2018; Marimuthu & Scheepers, 2022; McKay et al., 2007; Roberts & Mayo, 2019; Singer, 2018; Singh & Selvarajan, 2013; Wang & Seifert, 2020). Furthermore, if the “queen bee syndrome” (cf. Derks et al., 2016; Staines et al., 1974) extends beyond sexist behavior in male-dominated environments to racist behavior in white-dominated ones, this is likely to fuel misogyny against women leaders, especially from those who are harmed by their leadership. Finally, if engaging in racist behavior helps women gain status at work, this may reinforce the glass ceiling for other women who do not wish to emulate racist behavior and may reduce their own leadership aspirations as a result.

Our research highlights how white supremacy is perpetuated in organizations through a system of reward and punishment and answers calls from critical management scholars to study how white privilege and power is institutionalized in organizations (Al Ariss et al., 2014; Liu, 2021; Nkomo, 2021). If racist women are more likely to succeed in business, this represents an insidious way in which racism is institutionally perpetuated, validating the notion that business culture is in need of serious repair and change (Berdahl et al., 2018; Derry, 2012; Ely & Thomas, 2020; Singer, 2018). We turn now to reviewing research relevant to our predictions.

Background

The business world has long been criticized for its lack of diversity, equity, and inclusion (DEI), especially at the top. Each new civil rights movement, including the BLM movement of 2020, brings renewed focus on the lack of progress made in the corporate sector with respect to gender and race, particularly in leadership. Almost 60 years after the 1964 US Civil Rights Act outlawed discrimination based on sex and race in the workplace, little has changed in terms of who controls the helm and the treasure of the corporate ship, so to speak. Indeed, progress in race and gender-based wage and leadership gaps have largely stalled for 30 years or more (England, 2010; Leonhardt, 2020; Shu & Meagher, 2018). In 2022, men still held 94% of CEO positions in the S&P 500 (Catalyst, 2022) and controlled 99% of its shares (Green, 2022). As for racial diversity, the Fortune 500 counted only four Black men and two Black women as CEOs in 2022. Responding to public scrutiny and pressure, the corporate world frequently tries to demonstrate its efforts and commitment to rectify this imbalance. For example, after George Floyd was murdered by Minneapolis Police, Nike issued a statement saying it would “continue” to eliminate discrimination, despite its long track record of racial and gender discrimination and inhumane treatment of workers (Sherman, 2020).

A common approach by companies to demonstrate their progress and commitment toward DEI is to showcase women in their leadership ranks. Though women are still largely outnumbered by men in executive roles, the vast majority of women in these roles are white (Catalyst, 2022), and white women outnumber Black men as CEOs by seven-to-one. White women are likely to benefit from the fact that leaders are presumed to be white by default (Petsko & Rosette, 2022; Rosette et al., 2008), and, compared to non-white employees, white women enjoy relative social capital and ties to white men (Ibarra, 1993; Mollica et al., 2003; Moon & Stuart, 2018). As members of a disadvantaged group by gender, white women are often assumed to be willing and able to represent and advocate on behalf of other women and marginalized groups (e.g., Ely & Meyerson, 2010). After all, women do, on average, openly endorse more egalitarian views than men do (Araujo et al., 2017), including statements in favor of social equality (Sidanius et al., 1994) and opposed to sexism (Glick & Fiske, 1996) and racism (Carter, 1990).

Though they outnumber people of color in leadership, still few white women make it up the corporate ladder, as they too face bias and obstacles to ascension (cf. Eagly et al., 2007). When people “think manager,” they still “think male” (Braun et al., 2017; Offermann & Coats, 2018; Schein et al., 1996), or more specifically, “white male” (Petsko & Rosette, 2022; Rosette et al., 2008). Because corporate culture is imbued with the ideals, norms, and values (Schein, 1983, 1990, 2010) associated with white men (e.g., Ely & Meyerson, 2010), employees who aspire to fit and succeed in the corporate world must convince others that they approach and serve these ideals, norms, and values. This presents a challenge to women and racialized employees (Berdahl et al., 2018). As the “Benchmark woman” used to define femininity (Deliovsky, 2008; Purdie-Vaughns & Eibach, 2008), white women in particular may need to overcome stereotypes that portray them as lacking qualities associated with leaders, such as assertiveness and a willingness to take a stand (Bem, 1974), ambition (Sandberg, 2013), work devotion (Blair-Loy, 2009), competence (Heilman, 2012), confidence (Kay & Shipman, 2014), and an identification with and loyalty to those in power (Kanter, 1977).

One way that white women can signal identification with and loyalty to those in power and demonstrate their willingness and ability to fit into corporate culture rooted in gender and racial oppression is to put down and distance themselves from other women and racialized minorities. Consistent with this, research has shown that women who engage in sexist behavior—sexist jokes, distancing themselves from other women, and expressing attitudes and supporting policies that legitimize gender inequality—are relatively successful at assimilating into and rising within in male-dominated organizations (Alonso & O’Neill, 2021; Derks et al., 2011, 2016). Also known as the “queen bee syndrome” (Staines et al., 1974), this behavior can be understood as an individually adaptive response to a sexist organizational culture that rewards those who support, rather than challenge, that culture. In contrast, women who openly challenge sexist organizational culture by expressing feminist beliefs, supporting gender equality, and directly competing with men, for example, experience a variety of forms of social and professional backlash (e.g., Berdahl, 2007; Holland & Cortina, 2013; Maass et al., 2003; Okimoto & Brescoll, 2010; Rudman, 1998).

We propose the “queen bee syndrome” extends to racist behavior in white-dominated corporate America. A willingness to express anti-Black racist attitudes can highlight a white woman’s “whiteness” (Abdulle, 2017; Ahmed, 2007; Hogg & Terry, 2014), showcasing her higher-status and default leadership identity (Petsko & Rosette, 2022; Rosette et al., 2008). It may also help assuage concerns that she is too “feminine” (e.g., shy, retreating, deferential, and unwilling to take a stand; Bem, 1974) to take charge, by demonstrating her willingness to make a controversial statement. Finally, expressing anti-Black racism signals loyalty to and a willingness to act in the interests of corporate powerholders, most of whom are white men. White men are more likely than other groups to explicitly endorse sexist and racist attitudes (Ahmed, 2015; Cabrera, 2014; Carter, 1990; Pratto et al., 1994), consistent with their positionality and self-interest (Plaut et al., 2011; Pratto & Stewart, 2012). Thus, women may gain status in contexts dominated by white men, such as corporate America, by not only distancing themselves from and disparaging other women, but by distancing themselves from and disparaging those defined as most distant from “whiteness”: Black people (Ahmed, 2007; Brown-Iannuzi et al., 2013; Carbado & Gulati, 2013; Rabelo et al., 2021).

Though explicitly anti-Black racist statements are now generally frowned upon and contrary to most corporate policy manuals and mission statements, anti-Black racism remains rooted in the history and structure of the corporate world (e.g., Logan, 2019) and in people’s implicit attitudes, political opinions, and public policy support (e.g., Dovidio et al., 2000; Swim et al., 1995). Those familiar with corporate culture, such as full-time employees, implicitly learn this culture through their daily experiences and observations about what is valued and rewarded; these cultural assumptions and patterns cease to reach explicit awareness and come to be taken for granted as “natural” (Van Maanen & Schein, 1977). Because corporate culture implicitly reflects and rewards anti-Black racism, those who express such racism may be perceived as fitter contenders for status and power in the corporate world—especially women, who need to overcome doubts about their fit and worth in this world. We also expect that anti-Black racist comments will be viewed as less offensive when a white woman makes them than when a white man does, given stereotypes of women as weak and in need of protection, and thus less able of inflicting harm upon others (e.g., Glick & Fiske, 1996).

Integrating our theoretical reasoning with an intersectional lens, which offers a framework for analyzing multiple identities and their associated experiences (Collins, 2000; Crenshaw, 1989), we note that white women simultaneously belong to a low-status group (women) and a high-status one (whites). Thus, oppression and privilege intersect within their experiences as people who are marginalized by gender but privileged by race (Nash, 2008). Intersectionality theory encourages us to further consider how white women are positioned vis-à-vis white men in the workplace. As Aída Hurtado noted, “Each oppressed group in the US is positioned in a particular and distinct relationship to white men, and each form of subordination is shaped by this relational position” (Hurtado, 1989, p. 833). When white women engage in anti-Black racism in the workplace, they may signal their proximity to and alignment with white men, in addition to their fit with corporate culture, leading others to see them as worthy of relative status and power in that context.

In sum, we predict:

Hypothesis 1

Anti-Black racist speech in the workplace is judged to be less offensive from a white woman than from a white man.

Hypothesis 2

A white woman who engages in anti-Black racist speech at work is conferred more status than a white woman who does not engage in such speech; the opposite holds true for a white man.

We started with a pilot study to test our first hypothesis. This study also served as a check of our racist speech manipulation. We then conducted a fuller study (Study 1) to test Hypothesis 2. From there, we designed additional studies (2 and 3) to better understand why white women are, in fact, conferred relative status by others for engaging in anti-Black racist speech in the workplace.

Studies

Pilot Study: Gender and the Offensiveness of Anti-Black Racist Speech

We first conducted a pilot study to test our first hypothesis that anti-Black racist speech is judged to be less offensive from a white woman than from a white man. This study also served as a check of our manipulation of offensive anti-Black racist speech, which involved a comment we found on the internet that had been told by someone in real life.

Method

Participants

We recruited 30 participants from Amazon Mechanical Turk, which provides a reliable population representative of working individuals in the US and provides data shown to have strong external validity for studying attitudes embraced by the general population (Berinsky et al., 2012; Paolacci et al., 2010). We required at least a 95 percent approval rating on prior MTurk tasks for participants and we used a series of preliminary questions to ensure that participants had full-time job experience. On average, participants were 33 years old (SD = 9.26). Seventy percent of our participants identified as men and 30 percent identified as women. Fifty-seven percent identified as white and 43 percent identified as non-white.

Procedure and Design

Participants were recruited to participate in a study on “decision-making.” We employed a 2 × 2 (gender × speech) between-subjects research design. Each participant was randomly assigned to one of four conditions, which varied by speaker gender (man, woman) and speech type (racist, neutral). Participants were presented with a vignette and then answered a series of questions, described below.

Racist Speech Manipulation

All participants were presented with a short description of the debate about “free speech” in the US and were told that we had conducted an opinion poll with American workers to ask them what they thought about the issue. Participants were then randomly assigned to read one worker’s (a man’s or a woman’s) response (racist or neutral). The worker’s gender was manipulated with a name (Molly/Scott) and a photograph (a white woman for Molly, a white man for Scott) selected from the Chicago Face Database and matched on perceived attractiveness and neutrality of facial expression (Ma et al., 2015). The speaker offered either an anti-Black racist comment or a neutral opinion in response to being asked what they thought about the issue of free speech. Participants who had been randomly assigned to the racist speech condition read the following:

“Molly (Scott) Baker lives in Michigan, where she (he) has worked at General Motors as a Regional Sales Manager for 7 years. Molly believes that people need to learn to take a joke, saying, ‘Ugh I’m so fed up! No one has a sense of humor any more. Like I made a harmless joke in the office the other day—Today a giant gorilla escaped the zoo and won the women’s US Open... oh was that Serena Williams? My mistake!—and some people got mad and called me a racist. Some of my best friends are black! But everything today is about oppression and racism and social justice warriors. It’s so tiring’.”

Participants who had been randomly assigned to the neutral speech condition read:

“Molly (Scott) Baker lives in Michigan, where she (he) has worked at General Motors as a Regional Sales Manager for 7 years. Molly doesn’t have a very strong opinion about the issue, saying, ‘I don’t know what to say’.”

Following this information, participants proceeded to answer a series of questions about the speaker (Molly or Scott). Multiple checks were included to identify participants who were suspicious of the design or were not sufficiently attentive.

Measures

Offensiveness

We used a one-item measure of offensiveness, which asked participants the extent to which they agreed with the statement, “I think Scott/Molly’s opinions are offensive” (1 = strongly disagree, 7 = strongly agree).

Controls

Social desirability response bias (Crowne & Marlow, 1960)—a tendency to adjust one’s responses in the direction of socially desirable ones—is good to control for in self-report studies (Hays et al., 1989), particularly when examining socially sensitive issues like racism (Dovidio & Fazio, 1992; Dovidio et al., 1986; Gaertner, 1976). Social desirability controls have been substantiated as a useful technique to significantly reduce variance caused by social desirability response bias (e.g., Dalton & Ortegren, 2011; Straus, 2004; Valentine & Fleischman, 2008). We thus controlled for social desirability by asking participants the extent to which they agreed with the statement, “How concerned were you about making a good impression while participating in this survey?” (1 = strongly disagree, 7 = strongly agree).

Results

See Table 1 for the means, standard deviations, and correlations between study variables. The anti-Black racist speech condition (M = 5.56, SD = 0.27) was judged to be significantly more offensive than the neutral speech condition [M = 2.54, SD = 0.27; F(1,25) = 28.59, p < 0.0001], validating our manipulation of offensive racist speech. There was a significant interaction between speech type and speaker gender [F(1,25) = 4.420, p < 0.05]: the white man was perceived to be significantly more offensive when expressing a racist [M = 6.05, SD = 0.40] compared to a neutral [M = 2.48, SD = 0.40] opinion, whereas the greater offensiveness of expressing a racist [M = 5.07, SD = 0.36] compared to a neutral [M = 2.69, SD = 0.37] opinion did not reach significance for the white woman (see Fig. 1).

Table 1 Pilot study descriptive statistics
Fig. 1
figure 1

Pilot study: offensiveness by speech type and speaker gender

This study provides preliminary support for Hypothesis 1, which predicted that anti-Black racist speech in the workplace is judged to be less offensive from a white woman than from a white man. Next, we tested Hypothesis 2, which predicted that anti-Black racist speech at work leads to status conferral for white women.

Study 1: Anti-Black Racist Speech and Status Conferral

Using the same between-subjects vignette design as our pilot study, we examined how speech type and speaker gender interact to affect status conferred to the speaker at work.

Method

Participants

We recruited 347 participants from Amazon CloudSearch (formerly TurkPrime). Seventeen participants were removed for either failing attention or suspicion checks, leaving a total of 330 participants. The average age of participants was 36.7 (SD = 11). Sixty percent of the participants were identified as men and 40% were identified as women. A majority (75%) of participants were white and 25% were non-white.

Procedure and Design

The same procedure and 2 × 2 between-subjects design were used as in the Pilot Study.

Measures

Offensiveness

This time we used a two-item measure of offensiveness for improved assessment, asking participants the extent to which they agreed with the statements “I think Scott/Molly’s opinions are offensive” and “I think Scott/Molly’s opinions are fair” (reverse-scored) (1 = strongly disagree, 7 = strongly agree) (α = 0.78).

Status Conferral

We used the four-item measure of status conferral developed by Brescoll and Uhlmann (2008) to study the consequences to women and men in different occupational ranks, from trainee to CEO, for expressing anger in a professional context. The measure includes items assessing how much status, power, and respect a target deserves (1 = none at all, 7 = a great deal) and whether the participant would hire the target for their workplace (1 = never, 7 = definitely) (α = 0.86).

Controls

We controlled for social desirability with the same measure used in the Pilot Study. Controlling for the race and gender of participants yielded the same results as not controlling for these variables, thus we do not include them as controls in our findings. Importantly, patterns of offensiveness and status conferral (by speaker and speech) were the same for men and women participants and for white and non-white participants, and there were no significant differences by participant gender and/or race in offensiveness or status conferral. This is consistent with other studies of implicit bias, which show that, in contrast to explicit attitudes, implicit judgements, such as those measured with our study design, show little or no differences between groups (e.g., Banaji & Greenwald, 1995; Devine, 1989).

Results

See Table 2 for the means, standard deviations, and correlations between study variables. This study largely replicated the results for offensiveness in our Pilot Study. Whereas anti-Black racist speech was generally seen as more offensive than neutral speech, this effect was not significant [F(1,324) = 2.609, p = 0.11]. The interaction between speech type and speaker gender, however, was highly significant [F(1,324) = 113.37, p < 0.001]: The white woman was judged to be significantly less offensive when she made an anti-Black racist comment (M = 3.07, SD = 0.17) than when she made a neutral one (M = 4.58, SD = 0.16). The opposite held true for the white man, who was judged to be significantly more offensive when he made a racist comment (M = 4.97, SD = 0.18) than when he made a neutral one (M = 2.92, SD = 0.16) (see Fig. 2). These results are consistent with Hypothesis 1, which predicted that anti-Black racist speech in the workplace is judged as less offensive from a white woman than from a white man.

Table 2 Study 1 descriptive statistics
Fig. 2
figure 2

Study 1: offensiveness by speech type and speaker gender

For status conferral, anti-Black racist speech resulted in lower status overall (M = 3.75, SD = 0.11) than neutral speech did (M = 4.10, SD = 0.10), F(1,324) = 5.36, p < 0.05. There was also a main effect for gender: the white woman was conferred higher status than the white man overall, F(1,324) = 69.56, p < 0.001. Speech type interacted with speaker gender to predict status conferral [F(1.324) = 83.83, p < 0.001], with participants conferring the highest status to the white woman who made an anti-Black racist comment (M = 5.04, SD = 0.15), the lowest status to the white man who did (M = 2.46, SD = 0.16), and moderate status to the man or woman who made a neutral comment (see Fig. 3). These results are consistent with Hypothesis 2.

Fig. 3
figure 3

Study 1: status conferral by speech type and speaker gender

Using the Hayes PROCESS Macro Model 8 to test moderated mediation and indirect effects, we found that offensiveness mediated the interaction between speech type and speaker gender in predicting status conferral (index of moderated mediation = 2.1955, CI: 1.68, 2.78), suggesting the model shown in Fig. 4.

Fig. 4
figure 4

Theoretical model

Discussion

These results demonstrate that, as predicted, a white woman who engages in anti-Black racist speech at work is conferred higher status than a white woman who does not engage in this racist speech, whereas the opposite holds true for a white man. This effect was mediated by perceived offensiveness of the speech. We hypothesized and found that racist speech is judged to be less offensive from a white woman than from a white man; surprisingly, and unlike our pilot study (and Study 3, below), neutral speech was judged to be more offensive than racist speech from a white woman. It is possible that expressing neutrality about a controversial social issue (free speech) was viewed by these participants as dishonest, but as diplomatic for a white man and as duplicitous for a white woman, given gender differences in racist attitudes (Derry, 2017; Maxwell, 2015; Pratto et al., 1997).

Next, we examined what happens when a white woman engages in anti-racist speech at work. Is a willingness to take a stand on race in and of itself a recipe for a white woman to gain status at work, or is it the expression anti-Black racism that leads her to gain status? If a white woman is penalized with lower status for anti-racist compared to neutral speech, then supporting racism and inequality, not merely taking a stand on race, leads to her status conferral. We offer the following prediction for our next study of the effects of anti-racist speech:

Hypothesis 3

A white woman who engages in anti-racist/pro-BLM speech is conferred less status at work than a white woman who engages in neutral speech, whereas the opposite holds true for a white man.

Study 2: Anti-racist/Pro-BLM Speech and Status Conferral

We examine how anti-racist speech—in particular, expressing support for the Black Lives Matter movement—is perceived depending on whether the speaker is a white woman or a white man. As opposed to neutral speech, anti-racist speech directly challenges racist hierarchies and acknowledges the existence of current racial inequality and injustices.

Method

Participants

We recruited 240 participants from Amazon CloudSearch. Five Participants were removed for failing attention or suspicion checks, leaving us with a total of 235 participants. The average age of participants was 35.7 (SD = 12.44). Forty-seven percent of the participants identified as men and 53% identified as women. Most (68%) of the participants identified as white and 32% identified as non-white.

Procedure and Design

As with the prior two studies, participants were recruited to participate in a study on “decision-making.” In a 2 × 2 (gender × speech) between-subjects design, participants were randomly presented with one of four vignettes that varied by speaker gender (man or woman) and speech type (anti-racist or neutral), after which they answered a series of questions.

Anti-racist Speech Manipulation

All participants were presented with a short paragraph saying that recently there had been debate on the Black Lives Matter movement in the US and that we polled American workers about how they felt about it. Participants were then randomly assigned to read one worker’s response. The worker’s gender (man or woman) was manipulated with the same names and photographs used in the prior two studies. The speaker offered either an anti-racist opinion or a neutral opinion in response to what they thought about the Black Lives Matter movement. In the anti-racist speech condition, participants read:

“Molly (Scott) Baker lives in Michigan, where she has worked at General Motors as a Regional Sales Manager for 7 years. Molly (Scott) has a Black Lives Matter sign in her (his) front yard and believes that people who don’t support Black Lives Matter are racist. Molly (Scott) says, ‘Police brutality against Black people is a real problem in our country. I mean, it seems like every week the news has some story about a Black person getting beaten or shot to death by the police. People who don’t support Black Lives Matter are just racist, pure and simple. They don’t want to see the problem’.”

In the neutral speech vignette, participants read:

“Molly (Scott) doesn’t have an opinion about the issue, saying, ‘I don’t know what to tell you’.”

Measures

Dependent Variables

We used the same items as used in Study 1 to measure offensiveness (α = 0.77) and status conferral (α = 0.91).

Control Variables

We controlled for social desirability with the same measure used in the prior two studies. As in Study 1, controlling for participant race and gender yielded the same results as not controlling for these variables, so we did not include them in our findings. Again, response patterns were similar for men and women participants and for white and non-white participants and there were no significant differences by participant gender and/or race in patterns of offensiveness or status conferral by speaker and speech type.

Results

See Table 3 for the means, standard deviations, and correlations between study variables. There were no significant main effects on offensiveness of speech type [anti-racist or neutral, F(1,230) = 0.18, p = 0.67]or speaker gender [F(1,230) = 3.40, p = 0.07], and there was no interaction between speech type and speaker gender on offensiveness of the speech [F(1,230) = 0.01, p = 0.91]. There was, however, a significant difference in status conferral by speech type [F(1,230) = 12.23, p < 0.001], with neutral speech resulting in lower levels of status (M = 3.92, SD = 0.13) than anti-racist/pro-BLM speech (M = 4.59, SD = 0.14). There was no main effect for speaker gender on status conferral [F(1,230) = 0.04, p = 0.84], but speaker gender interacted with speech type to predict status conferral [F(1,230) = 5.27, p < 0.05; see Fig. 5]: the white man was conferred significantly higher status when he engaged in anti-racist/pro-BLM speech (M = 4.78, SD = 0.19), than when he expressed no opinion (M = 3.69, SD = 0.19), but the white woman was conferred a similar amount of status regardless of speech type (anti-racist M = 4.38, SD = 0.19; neutral M = 4.16, SD = 0.19). These results are partly consistent with Hypothesis 3, suggesting that unlike racist speech, anti-racist speech does not yield higher status than neutral speech does for a white woman. Thus, it is not merely a willingness to take a controversial stand about race relations, but a willingness to take an anti-Black one, that results in higher status for a white woman in the workplace.

Table 3 Study 2 descriptive statistics
Fig. 5
figure 5

Study 2: status conferral by speech type and speaker gender

Discussion

These results indicate that, unlike racist speech, anti-racist speech does not result in relative status conferral for a white woman, but does for a white man. In other words, when a white man supported Black Lives Matter, people were more likely to respect him and think he was worthy of hiring, status and power; when a white woman supported Black Lives Matter, she did not receive these accolades Interestingly, people judged a pro-BLM statement to be as offensive as expressing no opinion about BLM. This may be an indication of ambivalence toward the BLM movement, which has received both positive and negative press; pro-BLM speech may therefore be considered by the general public to be neither more nor less offensive, on average, than expressing no opinion about BLM. Importantly, participant gender and race did not significantly influence perceived offensiveness of pro-BLM versus neutral speech.

Combined with our prior two studies, these results resonate with ambivalent sexism, which views women as harmless, pure, and in need of protection from men on the one hand (benevolent sexism), and as conniving and seeking to gain control over men on the other (hostile sexism) (Glick & Fiske, 1996, 2001). White women who espouse anti-racist/pro-BLM attitudes may trigger hostile sexism, or concerns that they are seeking to upend the social order and control men (e.g., police), and thus be seen as offensive and unworthy of status. White women who espouse racist attitudes, on the other hand, may be seen as fitting into and capable of succeeding in the corporate world, and as relatively unoffensive due to perceptions of their harmlessness. Ambivalent sexism views men, on the other hand, as either abusers who use their power to harm (“bad” men) or as saviors who use their power to protect (“good” men) (Glick et al., 2018). White men who espouse racist attitudes may be seen as “bad” men and thus as relatively offensive and unworthy of status, whereas white men who espouse anti-racist attitudes may be seen as “good” men and thus as unoffensive and worthy of status.

We conducted another study to examine whether perceivers’ explicit attitudes account for the interactions we observed between speech and speaker gender. In particular, we expected that those who explicitly endorse racist and sexist views are most likely to reward racist speech in a white woman, as such speech not only reinforces the perceiver’s racist views but is also likely to assuage their sexist concerns about women trying to upend the status quo and control men (Glick & Fiske, 1996). Put another way, those who endorse beliefs that justify male dominance and white supremacy—white male supremacists—may be especially likely to favorably view a white woman who supports this supremacy. Those who reject white male supremacy, on the other hand (i.e., those low in racism and sexism), may confer higher status to a woman speaker than to a man speaker, as long as her speech is not racist. Combined, these effects may account for the interaction observed between speech type and speaker gender on status conferral:

Hypothesis 4

People who more strongly endorse sexism and racism confer more status to a white woman who engages in anti-Black racist speech than to a white woman who engages in neutral speech; people low in sexism and racism do the opposite.

Study 3: Perceivers’ Racist and Sexist Beliefs

To examine whether those who endorse relatively racist and sexist beliefs are most likely to confer status to a white woman for anti-Black racist speech, we conducted a final study that employed the same design and measures used in Study 1, adding measures of respondents’ endorsements of racist and sexist beliefs.

Method

Participants

We recruited 320 participants from Amazon CloudSearch. Twenty-five participants were removed for failing attention or suspicion checks, resulting in a total of 295 participants. The average age of participants was 36.4 (SD = 12.06). Sixty-five percent identified as men and 34% identified as women. A majority (66%) identified as white and 34% identified as non-white.

Procedure and Design

The same procedure and between-subjects design were used as in Study 1.

Measures

Dependent Variables

We used the same items as in studies 1 and 2 to measure offensiveness (α = 0.77) and status conferral (α = 0.95).

Independent Variables

To measure sexist beliefs, we used the six-item ambivalent sexism scale (Glick & Fiske, 2001). Three items measure hostile sexism (e.g., “Women seek to gain power by getting control over men,” 1 = strongly disagree, 7 = strongly agree, α = 0.91) and three items measure benevolent sexism (e.g., “Women should be cherished and protected by men,” 1 = Strongly disagree, 7 = Strongly agree, α = 0.81). To measure racist beliefs, we used a four-item measure of modern racism (Swim et al., 1995), which includes items such as “Discrimination against Blacks is no longer a problem in the United States” (1 = Strongly disagree, 7 = Strongly agree, α = 0.86).

Control Variables

We controlled for social desirability using the same measure as in our prior studies. As in the prior two studies, results did not differ by participant gender and/or race in patterns of offensiveness and status conferral by speaker gender and speech type, and controlling for participant race and gender yielded the same results as not controlling for these variables, so we do include them as controls in our findings.

Results

As can be seen in the descriptive statistics in Table 4, hostile and benevolent sexism were highly correlated with each other (r = 0.60, p < 0.001) and with modern racism (r = 0.74 and r = 0.47, respectively, p < 0.001). ANOVAs revealed identical patterns in how ambivalent sexism and modern racism predicted participants’ evaluations of speakers by gender and speech, thus we combined them into one measure of “white male supremacism” (α = 0.91). Consistent with prior research showing that women are less likely than men to explicitly endorse racist and sexist attitudes (e.g., Ekehammar et al., 2003; Ekehammar & Sidanius, 1982; Hoxter & Lester, 1994; Qualls et al., 1992; Sidanius and Pratto, 1994), women scored lower than men did on white male supremacism (F = 6.14, p < 0.05). Interestingly, participant race did not significantly relate to white male supremacism or interact with gender to predict these beliefs.

Table 4 Study 3 descriptive statistics

Replicating our previous findings, there was a significant effect for speech type on offensiveness. Anti-Black racist speech (M = 4.68, SD = 0.15) was perceived to be more offensive than neutral speech (M = 3.16, SD = 0.15), F(1,315) = 48.19, p < 0.001. There was also a significant interaction between speech type and speaker gender on offensiveness [F(1,315) = 7.15, p < 0.01], with the gap between offensiveness in racist and neutral speech much smaller for a white woman than for a white man (see Fig. 6), consistent with Hypothesis 1.

Fig. 6
figure 6

Study 3: offensiveness by speech type and speaker gender

This study also replicated findings from Study 1 for status conferral. There was a significant interaction between speech type and speaker gender on status conferral [F(1,315) = 15.54, p < 0.001], again with a white woman receiving higher status for racist (M = 4.45, SD = 0.18) than for neutral (M = 3.75, SD = 0.18) speech, and a white man receiving higher status for neutral (M = 3.99, SD = 0.18) than for racist (M = 3.37, SD = 0.18) speech (see Fig. 7).

Fig. 7
figure 7

Study 3: status conferral by speech type and speaker gender

As in Study 1, we used the Hayes PROCESS Macro Model 8 to test moderated mediation and indirect effects. Offensiveness mediated the interaction between speech type and speaker gender in predicting status conferral (index of moderated mediation = − 0.22, CI: − 0.47, − 0.04), consistent with the model presented in Fig. 4.

Finally, we tested the effects of participants’ endorsement of white male supremacism (see Table 5). Results showed that those low in white male supremacism conferred higher status to the white woman (M = 4.04, SD = 0.16) than to the white man (M = 1.81, SD = 0.16) overall, whereas those high in white male supremacism conferred higher status to the white man (M = 3.77, SD = 0.15) than to the white woman (M = 2.51, SD = 0.15) overall. Analyses also revealed a significant three-way interaction [F(8,286) = 23.72, p < 0.001] between speaker gender, speech type, and white male supremacism on status conferral. Unexpectedly, those low in white male supremacism conferred the highest status to the white woman who engaged in anti-Black racist speech (M = 4.88, SD = 0.15) and the lowest status to the white man who engaged in such speech (M = 1.29, SD = 0.15). Those high in white male supremacism conferred status to men over women regardless of speech type (see Fig. 8). Thus, Hypothesis 4, which predicted that people high in racism and sexism are more likely to reward anti-Black racist speech in a white woman, whereas people low in racism and sexism are more likely to reward neutral speech in a white woman, was not supported.

Table 5 Study 3: status conferral by speech type, speaker gender, and perceivers’ white male supremacism
Fig. 8
figure 8

Study 3: status conferral by speech type, speaker gender, and supremacism

Discussion

The results of this study suggest that the status conferred to white women for engaging in anti-Black racist speech is driven by those who least strongly endorse white male supremacy on an explicit level, as opposed to those who most strongly endorse it. In other words, those with “moderate” as opposed to “conservative” views regarding race and gender appear to be behind the status conferred to racist white women. This brings to mind Reverend Dr. Martin Luther King Jr.’s quote (King, 1963), “I have almost reached the regrettable conclusion that the Negro’s great stumbling block in the stride toward freedom is not the White Citizen’s Councillor or the Ku Klux Klanner, but the white moderate.” Our study suggests this stumbling block includes non-white moderates as well. Moderates may believe they are being progressive by conferring status to a white woman, especially if they think she is likely to succeed in the corporate world. Moderates, who openly reject white supremacism, and conservatives, who openly endorse it, combine to reinforce white male supremacy, however, by conferring status to a racist white woman or to a white man no matter what he has to say, respectively.

General Discussion

Across a series of experiments, we found that a white woman who engages in anti-Black racist speech is judged by others to be more worthy of status and power in the context of the American workplace than a white woman who engages in neutral or anti-racist speech. The opposite held true for a white man, who was conferred higher status for anti-racist speech (supporting Black Lives Matter) than for expressing no opinion or anti-Black racism. These results contradict the conventional assumption that (usually white) woman leaders, more than white men, help advance diversity, equity, and inclusion (DEI) in organizations. Though women are, on average, less likely than men to explicitly endorse racist attitudes (as prior research and our Study 3 has shown), it appears that the few white women who do explicitly endorse such attitudes are conferred relatively higher status at work. Thus, while companies may showcase white women in their leadership ranks as a sign they take DEI seriously, these companies may be achieving a diverse image at the expense of racial justice within their institutions.

Indeed, the implications of our studies for business ethics are worrisome. Prior research has found that “queen bees” gain status in sexist organizational cultures by distancing themselves from and denigrating other women (Derks et al., 2011, 2016; Kremer et al., 2021). Our research suggests that “Queen Karens”—to adopt the “Karen” meme (Mishan, 2021; Williams, 2020) depicting a white woman calling the authorities on Black people in “white spaces” (Anderson, 2015)—gain status by expressing anti-Black racism in the workplace. “Queen bees” reinforce gender inequality in the workplace, inciting anger, sadness, and anxiety in junior women (Sterk et al., 2018) and lowering their leadership aspirations (Ellemers et al., 2012). Similarly, “Queen Karens” are likely to entrench racial inequality in the workplace, and incite anger, sadness, and anxiety in Black (and other marginalized) employees, lowering their leadership aspirations. Queen Karens may even motivate their followers to practice and justify racial discrimination themselves (Brief et al., 2000).

Importantly, our finding that racist behavior is rewarded with status conferral for white women shows how corporate America remains, at least implicitly, aligned with its origins in racial oppression (Logan, 2019). Postcolonial theory in management (Banerjee & Prasad, 2008; Nkomo, 2021) argues that colonialism and racial subjugation were central to the development of modern capitalism, the effects of which persist today. The invention of “race,” racial stratification, anti-Blackness, and the birth of white supremacism (Loomba, 2002; Prasad, 2003; Spivak, 1986) is central to the colonial legacy that undergirds the modern workplace. Modern corporate culture continues to marginalize those excluded from whiteness and masculinity (e.g., Bhattacharyya & Berdahl, 2023; Glass & Cook, 2020; Nkomo & Al Ariss, 2014; Roberts, 2005). The modern corporation maintains and perpetuates structural inequality by centering power in the hands of a few (white men at the top) through various checks and balances, including individual reward and punishment. This includes penalizing women and racial minorities for diversity-valuing behavior (e.g., Hekman et al., 2017; our Study 2). Adopting a postcolonial lens to interpret our findings pushes us to rethink racism through a structural lens, and examine how members of modern corporations may maintain and perpetuate white male supremacy, even if they are not white men themselves. Our studies illustrate that white women might navigate their relative disadvantage in the context of the modern corporation by engaging in anti-Black racism.

Our research advances intersectional theory, with its origins in studying Black women’s experiences at the margins of both gender and race (Crenshaw, 1989). Intersectional scholars have called for the study of how subordinated and dominant identities are co-constituted and negotiated (Collins, 2015; McCall, 2005), and how people “mobilize (or chose not to mobilize) particular aspects of their identities in particular circumstances” (Nash, 2008, p. 11). Our studies offer insights into how white women may mobilize their whiteness to negotiate their subordinated gender and dominant racial identities in the workplace. There are historical underpinnings to white women weaponizing their whiteness to gain power and protection, including through enslaving Black people (Jones-Rogers, 2019), participating in anti-Black violence (Gilmore, 1993), falsely accusing Black men (or boys) of sexual aggression (e.g., Tyson, 2017), and calling the authorities on Black people (Armstrong, 2021).

Our research also adds insight into how white men, privileged with dominant identities in both gender and race, might lose status at work by expressing anti-Black racism but might gain status at work by supporting civil rights movements. Unlike white women, white men may be rewarded for advocating on behalf of a marginalized group, as in the “white man’s burden” (Glick & Fiske, 2001; Kipling, 1899), further enhancing their already high status without threatening the system.

Further advancing intersectional theory, the current studies thus add a gendered lens to research on whiteness in critical race studies (Chrobot-Masan, 2004; Grimes, 2001; Nkomo & Al Ariss, 2014). The ubiquity of white power and privilege extends to organizational “diversity” research (Nkomo, 2021) through the “commodification of difference, denial of white power, and chasing racial comfort” (Liu, 2021). To study whiteness, one must not only study white people, but also study how whiteness and its power are institutionalized and perpetuated (Nkomo, 2021). Our studies show how white supremacy in the workplace works in tandem with male supremacy by rewarding anti-Blackness in white women. Black women may be subordinated through exclusion, rejection, and violence (e.g., Hurtado, 1989), whereas white women may be subordinated through inclusion and “seduction,” which includes rewards, support, and protection for “good” behavior (Hurtado, 1989). Whiteness in the corporate world needs to be interrogated more broadly, moving beyond “individual racists” to focus on systems and structures that uphold and legitimize white supremacism.

There are several implications of this research for improving DEI practice in organizations. First, rather than placing the burden of advancing DEI on women and employees of color, which is the usual practice (Fryberg & Martinez, 2014; Settles et al., 2021), companies should also enlist anti-sexist and anti-racist white men in these efforts. If white men are more likely than white women to be rewarded for anti-racist advocacy, as our Study 2 suggests, then white men who are committed to anti-racism may be more likely than white women to succeed in receiving support for combating it in the modern workplace. The irony, of course, is that promoting white men as “saviors” in DEI work may reinforce perceptions of their superiority, thus reinforcing white male supremacy. For example, Joe Biden entered the 2020 presidential primaries in the most diverse field of candidates in history, and won in part because he was believed to be the most “electable” of the candidates, consistent with our finding that a white man fares relatively well for advancing progressive views (Peacock et al., 2021).

Second, organizations must be careful to select employees, and especially leaders, who are not only able, but willing, to stand up to racism and advance DEI. When making hiring and promotions decisions, organizations need to evaluate individuals’ knowledge and understanding of, as well as track records and reputations for, advancing DEI. It does not suffice to write a good diversity statement in an application or to extoll commitment to DEI in an interview; organizations need evidence of behavioral commitment through candidates’ actions. For example, how often have they spoken up against racism, despite the risks involved? What price(s) have they paid for doing so? How much of their own power, status, and resources have they been willing to give up in order to challenge racial inequality and oppression? How much experience do they have working with or leading racialized employees, and what do those employees think about them? Do women of color and people with other marginalized identities consider them to be allies (Bhattacharyya et al., 2024)? And so on.

Moving beyond individual solutions, however, the broader solution to preventing “Queen Karens” is to identify and condemn systemic and structural inequality in organizations and the implicit norms and practices that sustain that inequality (e.g., Ely & Thomas, 2020). It is these systems and structures, after all, that encourage and reinforce individual behavior, including anti-Black racism (Ahmed, 2015; Anthias, 1995; Flynn, 2015; Swim & Hyers, 2009). It is easy to identify and condemn a racist individual but difficult to identify and condemn a racist employer and institution; it will also do little good to condemn individuals for engaging in racist behavior when they are rewarded for that behavior, and much good to stop those rewards from flowing in the first place. Rather than joining a misogynist chorus against women who do what they are rewarded for in the corporate world, we need to join the Black feminist chorus against institutional structures and norms that reward racism and sexism (Nkomo, 2021; Nkomo et al., 2019).

In the end, to apply an oft-used metaphor, organizations need to move beyond blaming bad apples for their DEI failures and beyond showcasing different apples to cover up those failures, and actively scrutinize and reform the barrels that contain these apples. Our findings illustrate that diversity alone (e.g., Phillips & Loyd, 2006), even in leadership, is not enough to make business more ethical and egalitarian, and emphasize the need to move beyond “body count” diversity and toward an ethos of corporate responsibility to race (Logan, 2019). Organizations need to actively identify the institutional barriers and patterns that keep racism and sexism entrenched (Berdahl et al., 2018; Collins & Bilge, 2016; Ely & Meyerson, 2000), including the cynical practice of appointing leaders who support white male supremacy but are not white men. Other strategies for reform include educating and training central employees in anti-racism, perspective-taking, and an understanding of their own privilege and participation in oppression (e.g., Erksine & Bilimoria, 2019; Hershcovis et al., 2021; Swan, 2017) as well as supporting marginalized group members in understanding their potential complicity in these oppressive systems and strategies for solidarity and support. Cluster hiring of women and racial minorities is another strategy for increasing not only their numbers (Muñoz et al., 2017), but also for potentially preventing status-seeking behavior among them for limited positions, which may include mobilizing their higher-status identities at the expense of lower-status groups.

Limitations and Directions for Future Research

Like all studies, ours have limitations that could be improved upon with future research. To keep our materials consistent across studies, we examined anti-Black racist speech with the same manipulation. This limits the generalizability of our conclusions and our understanding of consequences to white women and men for putting down different racialized targets (e.g., Asian women, Latinas, or racialized men). A promising direction for future research would be to alter the target of racist speech to determine if this changes the perceived offensiveness of the speech and status conferred to the speaker. Furthermore, it bears pointing out that a “neutral” comment on the topic of racism is not so neutral. As Desmond Tutu famously pointed out, “If you are neutral in situations of injustice, you have chosen the side of the oppressor.”

Our theoretical argument was built upon theories of organizational culture and intersectionality, reasoning that a white woman, who experience racial privilege but gender subordination, might be conferred status for exhibiting proximity and loyalty to white men through anti-Black racist speech. Whether this phenomenon extends to men of color, who also hold both a dominant identity (in gender) and a marginalized identity (in race), and who can exhibit proximity and loyalty to white men through misogynist speech or behavior, would also be interesting to examine. For example, Clarence Thomas, a US Supreme Court justice and the second Black person to serve on the court, was nominated by a conservative white male Republican president, despite credible accusations that Thomas had sexually harassed Anita Hill when she was his subordinate. Since his appointment, Thomas had supported legal decisions that uphold white male supremacy and corporate power. Researchers could also examine whether anyone with at least one key marginalized identity (e.g., gender, race, sexuality, or class) might gain status by denigrating members of their own or another marginalized group. For instance, women of color may also be able to gain status in contexts run by white men by similarly supporting the status quo, such as an Asian woman who expresses anti-Black sentiments, thereby distancing herself from Blackness and aligning herself are more proximal to whiteness.

Finally, it is worth noting that public awareness and condemnation of racist behavior, including by white women (e.g., the “Karen” meme), waxes and wanes with social movements, and people’s judgements of white women who make anti-Black racist comments may change over time. However, the roots of racism run deep in the corporate world, and bias made explicit often turns implicit. If rewards for explicitly anti-Black racist speech by white women decline, but incentives remain for them to buttress white male supremacism in the workplace, then more implicit forms of racism may take the place of the explicit ones studied in the current research.

Conclusion

As our opening quote proclaimed, “Diversifying evil doesn’t change the meaning of the noun.” Eddie S. Glaude Jr. said this in response to the killing of Tyre Nicols, a 29-year Black father, son, and community member in Memphis, by five Black police officers (Glaude, 2023). Glaude went on to explain, “You can put Black faces in these institutions that are fundamentally corrupt at their core [but] it doesn’t change the fundamental nature of the institution.” Our research concurs with his conclusion: If organizations increase the diversity of their leadership without changing the unethical behavior it takes for members of underrepresented groups to reach these positions, this does not advance ethical business practices. Our research shows that the white women who are most likely to gain status and power in the workplace are the few who are willing to vocalize anti-Black racism. This illustrates how deeply and implicitly anti-Black racism is embedded in modern organizations and highlights the urgent need for changing their cultures in order to advance ethics and racial justice in the workplace and beyond.