r Academy of Management Journal 2017, Vol. 60, No. 2, 771–797. https://doi.org/10.5465/amj.2014.0538 DOES DIVERSITY-VALUING BEHAVIOR RESULT IN DIMINISHED PERFORMANCE RATINGS FOR NON-WHITE AND FEMALE LEADERS? DAVID R. HEKMAN STEFANIE K. JOHNSON University of Colorado at Boulder MAW-DER FOO National University of Singapore WEI YANG University of Texas at Austin We seek to help solve the puzzle of why top-level leaders are disproportionately White men. We suggest that this race- and sex-based status and power gap persists, in part, because ethnic minority and female leaders are discouraged from engaging in diversity- valuing behavior. We hypothesize, and test in both field and laboratory samples, that ethnic minority or female leaders who engage in diversity-valuing behavior are penal- ized with worse performance ratings, whereas White or male leaders who engage in diversity-valuing behavior are not penalized for doing so. We find that this divergent effect results from traditional negative race and sex stereotypes (i.e., lower competence judgments) placed upon diversity-valuing ethnic minority and female leaders. We discuss how our findings extend and enrich the vast literatures on the glass ceiling, tokenism, and workplace discrimination. Women and non-Whites have made remarkable gains within organizations (Joshi, Son, & Roh, 2015). Despite in the workplace in recent decades. Non-Whites and non-Whites and women outnumbering and sometimes women outnumber White men in the U.S. workplace by outperforming their White male counterparts, only amarginoftwotoone(Burns,Barton,&Kerby,2012). rarely are they given the reigns of the most powerful However, a demographic status and power gap remains, organizations in society. Economists are perhaps most as recent data show that only 25 Fortune 500 companies disturbed by this phenomenon, as orthodox economic are headed by people of color and 21 by women theory would predict that it is suboptimal for society to (Catalyst, Inc., 2013; Diversity, Inc. staff, 2012). Likewise, select its top leaders from only 34% of the population corporate boards in the Fortune 500 are primarily com- (i.e., the White men; The Economist,2008).1 posed of White men (74.4%), followed by White women One way to potentially reduce this status and (13.3%). Among ethnic minorities, 6.8% of corporate power gap is to place women and non-White leaders board members are African American, 3.1% are Latino, and 2.4% are Asian American (Zweigenhaft & Domhoff, 1 The so-called “glass ceiling” is a major reason why non- 2011). The status and power gap between men and Whites and women are considered minorities, even though women persists despite meta-analytic evidence sug- together they comprise a numerical majority. The term “mi- gesting that women tend to be rated as better leaders than nority” does not refer to a smaller number of people compared men (Paustian-Underdahl, Walker, & Woehr, 2014), and to the dominant group, but rather refers to a group that holds any performance evaluation gap that may exist fails to few positions of social power (Schaefer, 1996). Affirmative action programs and corporate diversity offices have been put account for the highly visible status and power gap in place with the purported goal of helping minorities break through this glass ceiling and achieve greater organizational We would like to thank Russell Cropanzano and Yuchen status, power, and influence (Harrison, Kravitz, Mayer, Leslie, Zhang, as well as the entire leadership division of the Leeds &Lev-Arey,2006;Levi&Fried,2008).However,despite School of Business. We would also like to thank Roland Smith, the increasing emphasis on promoting diversity, ethnic mi- Michael Campbell, Shannon Bendixen, Xiuxi (Sophia) Zhao, norities and women are still underrepresented at the highest and Angeline Lim at the Center for Creative Leadership, and, organizational levels and overrepresented at the lower orga- last but not least, the RSG for their tireless support of this project. nizational levels (Leslie, Mayer, & Kravitz, 2014). 771 Copyright of the Academy of Management, all rights reserved. Contents may not be copied, emailed, posted to a listserv, or otherwise transmitted without the copyright holder’s express written permission. Users may print, download, or email articles for individual use only. 772 Academy of Management Journal April in high-status, high-power positions, in the hope that engage in behaviors that increase diversity in the these leaders will empower other women and non- highest organizational ranks are systematically pe- Whites (Ely, 1994; Ibarra, 1995; Ragins & Scandura, nalized with lower competence and performance 1999). However, promoting demographic minorities ratings. Correspondingly, this logic may explain why into top leadership positions has been shown to have there are so few leaders willing to publicly advocate some unfortunate side effects. In fact, powerful mi- for non-White or women leaders to be promoted norities have been found to actively oppose the ad- (Ibarra, Carter, & Silva, 2010), and why ethnic mi- vancement of their fellow minority group members norities and women feel threatened at the prospect of (Ellemers, Rink, Derks, & Ryan, 2012; Ely, 1994; hiring a fellow member of their demographic group Sheppard & Aquino, 2013), possibly because they feel (Duguid, 2011; Hansen, Ibarra, & Peyer, 2010). Non- threatened by fellow members of their demographic White and female leaders may be highly aware of groups (Duguid, 2011; Duguid, Loyd, & Tolbert, 2012; the personal danger that diversity-valuing behavior Ely, 1994), or because, once demographic minorities poses to their careers. break into higher-status ranks, they wish to retain their We also contribute to the literature on tokenism by status by denying it to others (Kanter, 1977). providing an additional explanation for why mi- We suggest that another reason ethnic minority norities and women may impede the advancement of and women leaders may impede the advancement of their fellow women (“queen bee syndrome,” Staines, other ethnic minorities and women is because they Tavris, & Jayaratne, 1973: 55) and non-White (“crab are penalized in the form of lower performance rat- mentality,” Mendoza, 2002: 57) coworkers. The ings when they engage in “diversity-valuing behav- tokenism literature suggests that token non-Whites ior,” or behavior that promotes demographic balance and women take on the values of White men, and are within organizations (e.g., behaviors such as hiring placed in positions of status and power to act as and promoting ethnic minorities and women). By gatekeepers to prevent the further dilution of those “balance,” we mean an organizational demographic values, as well as to create the appearance of social profile that comes closer to resembling the demographic inclusion and diversity (Kanter, 1977). Our concep- makeup of the broader region or geographic area. tual model helps explain that token non-White or Thus, perfect gender balance involves a 50/50 split women leaders’ decision to promote White men in- between men and women employees, and perfect stead of non-Whites or women could also emerge racial balance involves the same percentage of ethnic from such leaders’ awareness that diversity-valuing minorities working within an organization as there behavior is personally costly. are in the surrounding region. When non-White and female leaders engage in diversity-valuing behavior, DIVERSITY-VALUING BEHAVIOR, LEADER they are perceived as threatening the existing status DEMOGRAPHICS, AND PERFORMANCE and power structure (Chattopadhyay, Tluchowska, & RATINGS George, 2004) and violating the expectation that mi- norities should play a supporting rather than a leading Race and sex have been found to be two of the most role in society (Sheppard & Aquino, 2013). As a result, important demographic markers leaders use to de- those that perceive this to be the case take actions that termine the degree to which fellow leaders are dif- preserve the established status and power hierarchy ferent from themselves (i.e., more important than by negatively stereotyping diversity-valuing ethnic age, education, functional background, or leadership minority and women leaders as being incompetent experience; Zhu, Shen, & Hillman, 2014). In the and having low performance. Conversely, ethnic mi- United States, at least, most people consider White nority and women leaders avoid negative stereotypes men to be members of a high-status social group when they engage in low levels of diversity-valuing and ethnic minorities and women to belong to low- behavior. status social groups (Berger, Fisek, Norman, & Our theoretical rationale linking leader demograph- Zelditch, 1977). Accordingly, we follow the lead of ics and diversity-valuing behavior to perceptions of other organizational researchers in examining both competence and job performance extends and en- race and sex bias simultaneously (Hekman, Aquino, riches the vast and expansive glass ceiling literature Owens, Mitchell, Schilpzand, & Leavitt, 2010; Zhu (e.g., Powell & Butterfield, 1994; Ragins & Scandura, et al., 2014). 1999) by helping solve the puzzle of why the glass Although there is a great deal of evidence that ceiling persists despite its societal costs. It may persist, a status and power gap exists for ethnic minority and in part, because non-White and women leaders who women leaders at a macro level (i.e., ethnic minorities 2017 Hekman, Johnson, Foo, and Yang 773 and women are underrepresented at the highest Diversity-valuing