Reversing the Queue: Performance, Legitimacy, and Minority Hiring

Studies of minority hiring have found that poor-performing firms or firms in highly competitive contexts are more likely to hire minority candidates. However, most work has examined hiring for entry and mid-level positions, not senior management. Management positions differ in terms of the amount of uncertainty in identifying candidates qualified for the job; in the intensity of external evaluations of both managerial and firm performance; and in the level of accountability for that performance. Furthermore, the influence of senior minority managers on hiring practices may differ substantially, depending on where a manager sits in the firm’s hierarchy. Examining hiring practices on coaching staffs of teams in America’s National Football League, from 1970-2007, we find that betterperforming teams are less likely to hire minorities to fill lower-level and mid-level coaching positions (as predicted by prior literature on labor queues), but that such teams are more likely to hire minorities into leadership positions. We also find that minority head coaches hire more minorities for subordinate coaching jobs, but that the presence of a minority offensive or defensive coordinator (with a white head coach) is a significant, negative predictor of minority hiring in junior and mid-level positions. The authors wish to especially thank Saleem Choudhry, of the Pro Football Hall of Fame in Canton, Ohio, without whose generous assistance this study would not have been possible. Thanks also to Michael Huygue, Andy Reid, Steve Mariucci, Ken Whisenhunt, Bill Cowher, Charley Casserly, Chad Lewis, Vai Sikahema, and Greg Comella, all of whom provided tremendous insight into the NFL environment. We are grateful for the input of Frank Dobbin, Robin Ely, and our reviewers from the Academy of Management. 3 INTRODUCTION Since the passage of the Civil Rights Act of 1964, the steady rise in black managerial representation in corporate America has been taken as an indicator of racial progress (Stainback and Tomaskovic-Devey, 2009; Krymkowski and Mintz, 2007). However, research in this area is unequivocal that a white male advantage in access to managerial jobs persists in both job status and level of authority (Maume, 1999; Stainback, et al. 2005; Stainback and Tomaskovic-Devey, 2009; Elliott & Smith, 2001). Around seven percent of all corporate managers are African-American, but only five blacks are CEOs of Fortune 500 companies, and blacks represent less than one percent of corporate officers in those corporations. It is therefore hardly surprising that (with few exceptions, see Thomas and Gabarro, 1999) examinations of racial minority representation in management have tended to aggregate across managerial positions. This paper explores 1) macro-level (the impact of firm performance) and 2) micro-level (the impact of minority composition of senior management) determinants of minority composition. In the first case, studies have found that minorities are more likely to be hired by lower status firms and firms in tight competitive environments (Reskin and Roos, 1990; Wallace and Chang, 1990; Smith, 2002; Tomaskovic-Devey, et al, 2006), Furthermore, within firms, high status jobs are often closed to minorities (Tomaskovic-Devey, 1993). In the second case, prior research has found that managers prefer to hire similar others: minority subordinates are more likely to be hired by minority managers (Kanter, 1977; Elliott and Smith, 2004). This paper responds to a call for research that explores contingent relationships in racial composition, and that investigates the causal mechanisms driving minority hiring decisions (Reskin, et al, 1999). It poses two questions: 1) How does firm performance influence the hiring of minorities into senior and subordinate management positions? 2) How does the presence of minorities in senior management positions affect the racial composition of subordinate management? We examine this 1 http://www.blackentrepreneurprofile.com/fortune-500-ceos/ 4 impact at different positions in the managerial hierarchy, and link prior research on homophily to the phenomenon of “moral credentialing” (Monin and Miller, 2001). Although minority representation in senior management remains low, it is increasing, and we have reason to question whether the forces that shape minority hiring at the lower levels of firm hierarchy have the same efficacy in the hiring of senior managers. At the risk of stating the obvious, not all management positions are equal. They differ in terms of the amount of uncertainty in identifying candidates qualified for the job; in the intensity of external evaluations of both managerial and firm performance; and in the level of accountability for that performance. We expect that poor performing firms are more sensitive to each of these characteristics, and that their presence moderates the relationship between firm performance and minority hiring. Thus, in addressing the first question, we hypothesize that poor performing firms are more likely to hire minority candidates into lower-level management positions, but less likely to hire black senior executives. Furthermore, once an organization has a minority leader in place, the overall dynamics in hiring minorities may be significantly altered. In addressing the second question, we hypothesize that black senior executives are less subject to attribution biases, and better able to evaluate candidate potential irrespective of race; they are therefore more likely to hire black managers at all levels, particularly into positions where minority representations has been lowest. We take advantage of a very specific industry context to study these dynamics, the US’s National Football League (hereafter called the NFL, or the League). We use data gathered between 1970 and 2007 on the selection of head coaches, their direct reports and lower level coaching staff. This follows the choice of other labor market scholars who have found the professional sports industry amenable to examining questions related to labor market dynamics (Fee et al., 2006). The context provides complete