Abstract
Scholars differ on whether the increase in minority managers represents real or vacuous progress toward the elimination of racial bias in the labor market. This study uses the National Study of the Changing Workforce to examine racial differences in work outcomes across the authority divide. On balance, this study finds more support for the pessimistic view of the minority presence in management, in that racial wage inequality is as large among supervisors as among nonsupervisors, and minority supervisors get less challenging job assignments and are more vulnerable to layoffs than white supervisors. Among subordinates, this study finds support for “bottom-up ascription” processes, in that minority workers who report to a minority boss earn less despite being more committed workers. The article concludes with a brief discussion of the implications of these findings and the need for further research on minorities in management.
As minority groups increasingly enter the ranks of management, some scholars claim that racial discrimination in the labor market is a vestige of the past (Brown et al. 2003). Others are more pessimistic, arguing that minority managers are concentrated in lower-paying service industries, often exercising authority over other minorities who themselves occupy devalued positions within firms (see, e.g., Elliott and Smith 2001).
Given the importance of this question, it is surprising that there is so little study of the effects of minorities in management positions. Organizational behaviorists and industrial psychologists have extensively researched how women in management positions affect workplace dynamics and firm performance (for a review, see Powell and Graves 2003), whereas sociologists have shown that firm and labor market factors determine minority representation in management (for a review, see Stainback, Tomaskovic-Devey, and Skaggs 2010). Generally lacking, however, are studies of how minority status affects the wages and working conditions among managers and the subordinates who report to them.
Although understanding the processes governing minority access to management positions is crucial to an assessment of the persistence of racial inequality in the labor market, it is also important to assess how minorities as a group fare when some of their members attain positions giving them the opportunity to reduce inequality. Thus, this study asks not only whether the jobs of minority and majority managers are similar but also whether minority bosses influence the job and career prospects of subordinates, and minority subordinates in particular. As I show in the review below, these questions have not been addressed in prior studies of minorities in management. Thus, by examining the influence of race on the job assignments and rewards of managers and nonmanagers alike, this study provides an answer to the larger question of whether diversity in managerial ranks represents real or vacuous progress toward eliminating racial bias in the labor market.
Background
Whereas diversity is associated with improved firm performance (Herring 2008), it has also long been known that senior managers tend to promote junior colleagues with similar backgrounds and worldviews, thereby reproducing the dominance of white men in management (Kanter 1977). For this reason, recent research has analyzed the political, economic, and organizational determinants of minority access to managerial positions.
Access to management
Minorities steadily increased their presence in management after the passage of 1960s-era antidiscrimination legislation. Drawing on employer reports to the Equal Employment Opportunity Commission (EEOC; the federal agency charged with monitoring and combating discrimination in the labor market), Stainback and Tomaskovic-Devey (2009, 802) showed that the proportion of all managers who were white men declined from 91 percent in 1966 to 57 percent in 2000; by comparison, there were virtually no African American managers in 1966, whereas in 2000, one in fourteen managers was African American. These figures imply that white men lost managerial jobs over time, but this was not necessarily the case in a growing economy. Indeed, Stainback and Tomaskovic-Devey’s analysis revealed that white men retained their advantaged positions in “old” goods-producing industries, whereas African Americans and white women rapidly increased their managerial presence in the growing (but lower-paying) personal and social service industries.
Nevertheless, political and legal pressure on firms was instrumental in diversifying the ranks of management. For example, Edelman (1990) showed that in response to EEOC lawsuits and affirmative action regulations, firms created human resource departments charged with ensuring compliance with antidiscrimination legislation, and these “responsibility structures” worked to increase the number of African Americans in management (Kalev, Dobbin, and Kelly 2006). Similarly, Skaggs (2009) found that supermarkets, industry-wide, increased their hiring and promotion of black managers a year after a discrimination lawsuit had been filed against an employer and in jurisdictions with more minority federal judges sitting on the circuit court.
Collins (1997) puts a human face on these findings in her interviews with black corporate executives, many of whom were the highest-ranking African Americans in their respective Fortune 500 firms. In looking back over their careers, many acknowledged that they were hired in response to the racial unrest of the 1960s. Some were hired into human relations departments and charged with ensuring compliance with affirmative action regulations and EEOC mandates. Others were hired as vice presidents of community and public relations, acting as liaisons to African American communities and charged with developing products for black customers or recruiting black employees into the firm. Although Collins’s informants earned comfortable salaries, they had been placed in “racialized jobs” that responded to the concerns of African Americans, as opposed to whites who worked in jobs with wider constituencies (e.g., sales, operations, marketing, and financing). As a result, by the end of their careers, African Americans were more limited than whites in how high they could rise within the firm, and they grew resentful as others were groomed to be CEOs of their respective firms (see Cose 1993). Moreover, many black executives sensed that their jobs could be eliminated in cost-cutting measures if the government relaxed its oversight of firms’ compliance with antidiscrimination legislation (see Leonard 1984). Of course, for lower-placed middle managers and semiprofessionals (e.g., social workers, counselors, corrections officers, etc.), government expansion in the postwar era employed African Americans in middle-class jobs, and efforts to shrink the size of government would directly threaten their livelihoods (Collins 1983; W. J. Wilson 1980).
Effects of minorities in management
The discussion above begins to highlight the difference between optimistic and pessimistic interpretations of an increased minority presence in management. On one hand, minorities have made demonstrable progress in entering managerial jobs that are among the most rewarding and prestigious in the economy (Stainback and Tomaskovic-Devey 2009). On the other hand, there is some evidence that in response to government pressure, firms reclassified some clerical and administrative jobs as managerial jobs (Jacobs 1992; J. Smith and Welch 1984) and created new managerial jobs filled by blacks to respond to African American demands for equality (Collins 1997). This pessimistic view is further supported by mobility studies showing that minorities lag behind white men in receiving promotions to top managerial jobs (Maume 1999) and that it is in government, not the private sector, where rates of minority ascendance to managerial jobs is most rapid (G. Wilson, Sakura-Lemessy, and West 1999). Furthermore, when minorities do get promoted into management, they typically manage minority subordinates who are themselves poorly paid and immobile (Elliott and Smith 2001; R. Smith and Elliott 2002; Tomaskovic-Devey 1993).
The degree of similarity in the managerial jobs minorities and whites hold is the main point of contention between those who are optimistic that a decline in racial inequality is evidenced by the growth in minority managers and those who are pessimistic about such a conclusion. Thus, the first aim of this study is to assess whether reaching managerial jobs matters for minorities, specifically by comparing the size of the racial wage residual between managers and nonmanagers. If minority managers hold jobs that are as rewarding as those of white managers, then the racial wage residual should be lower among managers than among nonmanagers—a finding that would support the optimistic view of minorities in management. Whereas Jacobs (1992) assessed the degree of similarity in managerial jobs held by men and women, I am unaware of a study that makes this assessment by race.
Another approach in assessing the effects of minorities in management would be to determine if, among managers, minorities differ from whites in the responsibility, freedom, job security, and interpersonal relations characterizing their jobs. The pessimistic view suggests that minorities are given less challenging and more unstable managerial jobs, and that in interpersonal encounters they are often reminded of their differences with majority managers (Collins 1997; Cose 1993; Kanter 1977). If so, then minority managers will differ significantly from their white counterparts in holding lower-quality and devalued managerial jobs. Of course, if minority managers hold jobs that resemble those of whites, then the evidence would support the optimistic view of minorities in management.
Some scholars suggest that another way minority managers are devalued is by their placement over minority workgroups. After majority groups reserve the best jobs for themselves (Reskin, McBrier, and Kmec 1999; Tomaskovic-Devey 1993), the challenge for the organization is obtaining desired work behavior and commitment from those who occupy devalued and immobile jobs. To solve this problem, organizations engage in “bottom-up ascription,” that is, placing minority superiors in charge of minority-dominated workgroups (Elliott and Smith 2001; R. Smith and Elliott 2002). Not only does this signal a firm’s compliance with external legal pressures to diversify the ranks of management, but it also creates the impression among minority subordinates that via hard work, commitment, and loyalty to their employers, they too can advance to supervisory positions.
Given these considerations, a second aim of this study is to assess whether having a minority manager matters for the rewards, experiences, and career prospects of minority subordinates. A pessimistic view of minorities in management is that they are powerless to affect the careers of minority subordinates, providing no more career support or mentoring to minority subordinates than white supervisors provide to white subordinates. Furthermore, because minority workers are devalued within the firm, they are paid less than whites irrespective of the minority status of their immediate superior. Such findings would suggest that minority managers are mere “cogs in the machine” whose role is to provide a symbolic image of potential mobility to minority subordinates, who otherwise occupy low-paying and dissatisfying jobs.
A more optimistic view of minorities in management would suggest that they act as “change agents,” taking an active role in advancing the careers of minority subordinates. One study shows that when given the opportunity, minority supervisors tend to promote minority subordinates (e.g., see Elliott and Smith 2004). Furthermore, bonds between minority supervisors and subordinates may be strengthened by shared discrimination experiences (e.g., see Cose 1993), and supervisors may not want subordinates to face the same career obstacles that they did. If so, then minority supervisors may seek to advance the careers of minority subordinates by providing job-related advice, support, and mentoring and by ensuring that they are paid similarly to whites. Some attempts have been made to assess whether female managers advance the careers of female subordinates (Cohen and Huffman 2007; Maume 2011; Stainback and Kwon this volume), but no one has yet made this assessment with regard to minorities in management.
Data
Sample
The data used in this study are from the 1997 and 2002 National Study of the Changing Workforce (NSCW) surveys, both of which are representative samples of the employed civilian labor force (for details on the design and implementation of the NSCW, see Bond et al. 2003). For theoretical and practical reasons, two restrictions were placed on the composition of the sample. First, from a total of 5,687 wage and salary workers, the sample is limited to adults ages 18 to 64 who worked full time (N = 4,535), on the grounds that these workers are fully committed to career pursuits (although the full-time restriction will be relaxed when analyzing subordinate work outcomes; see below). Second, including women in the analytic sample would complicate attempts to assess race effects on minorities in management, given that race and gender are distinct axes of inequality (Tomaskovic-Devey 1993). The most conservative estimate of the effects of minorities in management would be to contrast the rewards and job experiences of minority men with those of white men, and for this reason only men (N = 2,245) are included in the analytic sample (see Maume [2011] for an analysis of the effects of female supervisors on the career prospects of subordinates).
Who is a minority? Who is a manager?
The NSCW asked whether respondents identified themselves as Hispanic and also asked them to identify their race as either “white,” “black,” or “other.” Preliminary cross-tabulations indicated that most Hispanics saw themselves as members of a minority group (i.e., nearly 8 in 10 Hispanics claimed to be other, and only a handful of whites identified as Hispanic). Furthermore, when blacks and Hispanics were contrasted with non-Hispanic whites on outcome measures, the effects of these separate statuses were often similar in direction and magnitude but nonsignificant (likely because of small subgroup sample sizes). For the sake of parsimony, then, blacks and Hispanics are scored 1 on a binary measure of minority (non-Hispanic white men compose the reference group).
Regarding managerial status, the NSCW asked respondents to report on their usual duties and kind of work they did, and responses were classified using the 1990 standard occupation codes (SOC) developed by the Census Bureau. But by convention, the Census Bureau defines “executives and managers” as largely consisting of professionals (e.g., business executives, financiers, lawyers, etc.) who have substantial control over a firm’s fiscal and human resources (Maume 1999). On one hand, this is advantageous in that minorities in census-defined managerial occupations hold the most prestigious and rewarding jobs in the economy. On the other hand, many white-collar jobs (e.g., technical, sales, and clerical jobs) as well as blue-collar and service jobs require incumbents to supervise, evaluate, and reward the work of subordinates, but these jobs are often classified by their relevant job category rather than as managerial. To capture these individuals, the NSCW asked respondents if “supervising other people is a major part of your job”; those who answered affirmatively were defined as supervisors. A cross-tabulation of managerial and supervisory statuses showed that seven in ten managers supervised others, but only one in four supervisors was a manager. Thus, the analyses below distinguish between those in census-defined managerial occupations and self-reported supervisors.
Measures
Most of the measures described below were included in both the 1997 and 2002 surveys, but some were asked only in the 2002 survey. The appendix provides descriptive statistics on all measures used in the analyses.
Outcomes for managers
The most basic job reward is annual pay (including bonuses and overtime pay) in the year of the survey. After inflating the 1997 salary to 2002 dollars, the natural log of annual income was calculated to correct for its rightward skew. Approximately one in twelve respondents refused to divulge his earnings and was assigned the mean value on income; a binary measure was created to control for mean assignment.
The working conditions of managers were assessed by a series of composite indices. Except when noted, all component items were measured on a five-point Likert scale, and preliminary factor analyses confirmed that individual items tapped a single underlying construct (all composite indices had acceptable alpha reliabilities of between .6 and .8). First, the high-pressure job index was calculated as the mean of three items assessing the need to work fast, hard, and under tight deadlines. Second, the job autonomy index was constructed as the mean of four items asking about the respondent’s freedom to decide what to do on the job and when to take breaks, responsibility for deciding how the work is done, and ability to have “a lot of say” in what work gets done. Third, being disrespected on the job was assessed as the mean of two items asking about “having to do things on the job that go against my conscience” and “being treated with respect” on the job (reverse-coded). Finally, because of extreme skew on the component items, the job insecurity index was assessed by a count of answering affirmatively that they had been previously laid off because work was slow and saying it was “very likely” that they would lose their job and need to find another one.
Outcomes for subordinates
Testing the argument that minority managers oversee devalued minority subordinates requires information on the race of the immediate supervisor, the superior’s work-related behavior, and subordinates’ reactions to their work situations. First, subordinates were asked whether their supervisor was of the same racial or ethnic background, enabling the construction of a binary measure for reporting to a minority supervisor. Second, subordinates’ annual income (logged) was calculated in the same manner described above. Third, the subordinate’s perceived advancement opportunities were measured by a single item (“How would you rate your own chance to advance in your organization”; responses ranged from 1 = poor to 4 = excellent). Fourth, the supervisor’s job support index was calculated as the mean of four Likert items tapping supervisors’ keeping subordinates informed about job matters, having realistic expectations of job performance, recognizing when work is done well, and being understanding and supportive about problems at work. Finally, organizational commitment was calculated as the mean of two standardized items assessing how loyal the respondent was to his employer and whether he worked harder than he had to for the benefit of the company.
Individual controls
Besides race, the analytic models include a continuous control for years of completed education and binary controls for being single and/or a parent, living in the South, and sample year (1 = 2002; 0 = 1997). In addition, the models included controls for years of pre-employer work experience (indicative of the accumulation of general skills) and years of tenure with the employer (tapping the stock of firm-specific skills as well as employment rights and benefits conferred by seniority); years of pre-employer work experience was calculated by subtracting employer tenure from years worked full time since age 18. To capture potential nonlinearities in their relationships with the outcome measures, squared terms for experience and tenure were entered into all analytic models but were significant and retained only in the models of annual income.
Job-related controls
Prior studies show that minority presence in management is higher in government and social services as well as in those firms that created human resource departments to monitor compliance with antidiscrimination laws and affirmative action (Edelman 1990; Stainback and Tomaskovic-Devey 2009; G. Wilson 1997). Thus, the models include a binary control for working in government or the nonprofit sector (with the private sector serving as the reference category). Although the NSCW lacks information on the formalization and implementation of human resources policies in response to EEOC oversight (e.g., codified job descriptions, posting of job openings, formal written evaluations of employees’ job performance, etc.), the NSCW does have a measure of firm size (an ordinal scale, ranging from 1 = under 25 employees to 10 = 10,000+employees), which is a strong correlate of firms’ creation of bureaucratic structures and procedures to ensure due process (Marsden 1996).
A key measure in testing for bottom-up ascription processes is the racial composition of subordinates’ coworkers. This was assessed by the question “About what percentage of your coworkers are people from your racial, ethnic, or national background?” This measure (available only in the 2002 survey) has ordinal responses ranging from 1 = 100 percent of coworkers to 6 = 0 percent of coworkers; at the maximum value on this measure, the respondent is a racial token in his workgroup. In addition, the bottom-up ascription argument suggests that minorities are given managerial authority over devalued workers, a key indicant of which is the strength of their job attachment; thus the analytic models of subordinate outcomes will include a binary control for working part time. Finally, it is possible that subordinates’ opinions about their bosses and their own career prospects may be affected by whether they exercise authority on the job; thus, the models of subordinate outcomes include additional binary controls for exercising supervisory authority over other workers (self-reported by respondents) and holding a managerial occupation (as defined by census occupation codes).
Results
Pay inequality and the managerial divide
I first consider racial pay inequality, comparing across the authority divide. Table 1 presents unstandardized coefficients from an ordinary least squares (OLS) regression analysis and can be interpreted as the proportionate decrement in minority men’s earnings relative to white men. The results suggest that empirical support for the optimistic or pessimistic views of minorities in management depends on how authority status is defined.
Unstandardized OLS Effect of Minority Status on Annual Salary (Log), by Supervisor/Manager Status, Full-Time Employees, Men Ages 18 to 64, 1997 and 2002 NSCW
NOTE: Models control for education, pre-employer experience, experience squared, employer tenure, tenure squared, married, parent, Southern resident, sample year, and income assignment. In panel 1, the models explained between 26 percent and 33 percent of the variation in pay (for nonmanagers and supervisors, respectively; r-squared values were slightly higher in the panel 2 and 3 models due to smaller sample sizes.
The sum of subgroups may be less than the total N in panel 1 because of missing data.
p < .05.
When census occupational codes are used to distinguish managers from nonmanagers, the results tend to support the optimistic perspective that when minorities ascend to privileged managerial positions they are paid similarly to white male managers. That is, in panel 1, minority nonmanagers earn 12.3 percent less than white nonmanagers, but among managers the minority pay gap shrinks to 4.1 percent and is no longer statistically significant. Furthermore, this pattern of results is replicated in large firms (panel 3). Firms with 100+ employees are required to report to the EEOC on the diversity of their payrolls, and among these employers the racial pay gap declined from 11.8 percent among nonmanagers to a nonsignificant 2 percent among managers (similar results were observed when two other thresholds for large firm size were defined—250+ employees and 500+ employees). In smaller firms, however, black managers earn 14.5 percent less than white managers, which is slightly higher than the 13.9 percent racial pay gap among nonmanagers employed in small firms. The pattern of results is less clear by sector (panel 2). The 12.5 percent racial difference in pay among nonmanagers in the private sector is the only residual that reaches statistical significance; the remaining racial pay gaps are insignificant, although hardly trivial.
When the authority divide is defined by self-reported supervisory status, the findings tend to support the pessimistic view of minorities in management. That is, among all men (panel 1), the minority wage residual barely declines when workers exercise supervisory authority (11.9 percent) compared with nonsupervisors (12.8 percent). Similarly, minority wage gaps are sizable and significant among private-sector nonsupervisors (14.2 percent) and supervisors (10.8 percent) alike. Yet similar to census-defined managers, minority nonsupervisors and supervisors are paid more similarly to their white counterparts when employed in the public sector.
In contrast to the declining pay gap across the managerial divide among large-firm employees, the minority wage gap increases across the supervisory divide in large firms (panel 3). That is, in firms of 100 or more employees, minority supervisors earn 14.4 percent less than white supervisors—a pay gap that is two-thirds larger than the minority pay gap among nonsupervisors. The differing results in panel 3 for large-firm employees (i.e., the pay gap increases across the authority divide for supervisors, but decreases for managers) likely stem from the fact that census-defined managers are professionals employed at the apex of their employers’ hierarchies, whereas supervisors are more likely to be middle managers in large firms. As other research shows, it is in large firms where job titles proliferate (Baron, Davis-Blake, and Bielby 1986; Marsden 1996), and it is more difficult to codify skills requirements and define successful job performance (and thus set pay levels) among middle managers than among line workers (Baron, Davis-Blake, and Bielby 1986; Kanter 1977). Thus, despite the presence of human resources policies and government oversight of their implementation, supervisors tend to proliferate in large firms, and the difficulty in evaluating their performance allows organizational biases against minority supervisors to be reflected in pay scales (McGuire and Reskin 1993; Reskin, McBrier, and Kmec 1999). By contrast, pay gaps are smaller among managers employed in large firms because minority managers are fewer in number and bias against them is more easily detected by the human resource practices implemented in large firms (Kalev, Dobbin, and Kelly 2006).
Informal human resource practices and visibility in small firms may also explain why minority supervisors are paid only 6.9 percent less than white supervisors (a difference that is not significant), in contrast to the 19.9 percent pay gap among nonsupervisors. That is, lacking internal labor markets to motivate subordinates, a supervisor’s personality and charisma become even more important in motivating subordinates to work hard on behalf of the firm (Edwards 1979), and minorities who can effectively manage work groups are paid similarly to white supervisors in smaller firms.
Working conditions of managers and supervisors
Additional evidence for adjudicating between the optimistic and pessimistic views of minorities in management may be found in comparing the jobs held by minorities and whites who exercise authority. Table 2 presents standardized OLS effects of minority status on the working conditions of managers and supervisors.
Standardized OLS Effect of Minority Status on Working Conditions of Full-Time Supervisors and Managers, Men Ages 18 to 64, 1997 and 2002 NSCW
NOTE: N = 332 and 885 managers and supervisors, respectively. Models control for education, pre-employer experience, employer tenure, married, parents, Southern resident, firm size, government/nonprofit sector, and sample year. R-squared values ranged from a low of .04 in predicting being disrespected on the job among supervisors to a high of .11 in predicting job insecurity among managers.
Job requires respondent to work hard, fast, and under tight deadlines.
Respondent has freedom to decide what to do on the job, is responsible for how work gets done, has a lot of say in what work gets done, and decides when to take breaks.
Respondent has been temporarily laid off or had hours cut when work was slow and expects to lose job and search for new job.
Respondent is disrespected at work and has to do things that go against conscience.
p < .05.
As was the case above regarding the pay gap, it appears that support for the optimistic and pessimistic views of minorities in management depends on how managerial authority is defined. That is, in three of four outcomes, minority managers (as defined by census criteria) are similar to white managers, specifically in regard to exercising autonomy on the job, feeling insecure about their jobs, and being disrespected on the job. The only case in which minority and white managers significantly differ is the former reporting less pressure on the job in the form of working fast, hard, and under tight deadlines (B = −.15). Whereas some view having a high-pressure job as stressful (e.g., Karasek 1979), Kanter (1977) argued that fast-paced and pressure-packed jobs offer individuals the chance to display their talents. In other words, challenging jobs may be visible jobs, offering incumbents a chance to signal their suitability for promotion based on successful performance in a demanding job, and on this dimension, minorities lag behind whites in placement in high-pressure jobs (among managers and supervisors alike).
Stronger support for the pessimistic view is found in the significant minority residuals for the remaining working conditions that typify supervisory jobs. For example, compared with white supervisors, minority supervisors exercise less autonomy on the job (B = −.07). Not only does job autonomy tend to lead to better psychic adjustment to the demands of one’s job (Karasek 1979), but making decisions and organizing the flow of one’s work and the work of others is another way to display one’s talent and skills (Kanter 1977). These results suggest that minorities are more likely to be placed in jobs that restrict their decision-making abilities, thus limiting their chances to demonstrate to their own superiors that they would make good leaders of the firm. Similarly, several analysts claim that more so than whites, minority supervisors are placed in positions that are more vulnerable to the ebb and flow of the market, and they must suppress their minority identities to further their careers in white-dominated firms (Collins 1997, 1983; Cose 1993). The results in lines 3 and 4 of Table 2 are consistent with these arguments; that is, compared with white supervisors, minority supervisors are more insecure in their jobs and more disrespected by others at work.
To summarize the findings thus far, the optimistic view of minorities in management receives its strongest support among those who reach upper-echelon management positions, as defined by the Census Bureau. Upon doing so, minority managers earn salaries that more closely approach those of their white counterparts; their reported levels of job autonomy, security, and getting respect from others also approach those of their white counterparts. But lower down in the occupational hierarchy where supervisors are found, compared with their white counterparts, minority supervisors are paid less and placed in less demanding jobs, with less autonomy and less job security and where they are more disrespected by others—findings that support the pessimistic view of minorities in management.
Racial segregation of subordinates and minority supervisors
The bottom-up ascription argument posits that a minority-dominated workgroup is likely to report to a minority supervisor. I evaluated this proposition in a multivariate logistic regression model using the controls shown in Tables 1 and 2 (including dummy controls for whether the worker is part time, supervises others, and/or holds a managerial occupation). The predictor variable of interest in this analysis is an ordinal measure of the racial composition of coworkers (available only in the 2002 survey). Because the results were straightforward, they are briefly described (rather than shown in a table) here. In race-stratified models, for whites and minorities alike, the odds of reporting to a minority manager significantly increase as one’s coworkers are increasingly composed of minorities. These results confirm a basic tenet of the bottom-up ascription argument: namely, that minority supervisors tend to exercise authority over minority workgroups.
Minority supervisors and subordinate job experiences and career prospects
Table 3 shows the partial standardized effects of reporting to a minority supervisor and the racial composition of the workgroup on subordinates’ work experiences and career prospects, by minority status. Among whites, in all cases the racial composition of the workgroup and the minority status of the immediate supervisor have no effect on their job experiences and rewards.
Standardized OLS Effects of Having a Minority Supervisor and Racial Composition of Coworkers on the Experiences and Career Prospects of Subordinates, by Minority Status, Men Ages 18 to 64, 2002 NSCW
NOTE: N = 949 and 248 whites and minorities, respectively. Models control for education, pre-employer experience, employer tenure, married, parent, Southern resident, government sector, firm size, supervises others, managerial occupation, and works full time. In panels 1 to 3, r-squared values ranged from .03 to .06 among whites and from .10 to .16 among minorities; in panel 4, the models explained 37 percent and 40 percent of the variation in pay among whites and minorities, respectively.
Boss keeps respondent informed about work matters, has realistic expectations, recognizes work well done, and is supportive when problems arise.
Respondent is loyal to employer and will work harder than he has to for firm’s success.
Chance to advance within organization (1 = poor; 4 = excellent).
p < .10. **p <.05.
Among minorities, however, the results unequivocally support the pessimistic view of minorities in management and, in particular, that bottom-up ascription processes motivate the matching of minority supervisors with minority subordinates. Panel 1 shows that minority subordinates in white-dominated workgroups report receiving significantly lower support from their immediate superior (B = −.16)—a result that might be anticipated from research on gender tokenism (Kanter 1977). Yet when minority subordinates report to a minority supervisor, they get no more job-related support than they would from a white supervisor. Furthermore, minority subordinates show more loyalty toward and work harder for their employers when they report to a minority supervisor (panel 2), and they are marginally more likely to expect a promotion when their boss is a minority (panel 3). These results must be viewed in the context, however, that minority subordinates get lower pay when reporting to minority supervisors (panel 4). Social closure perspectives on workplace dynamics suggest that minorities are devalued within organizations (Tomaskovic-Devey 1993), and they are assigned minority bosses to make the symbolic promise that via hard work and commitment, they too can advance to the ranks of management (Elliott and Smith 2001). The results in Table 3 show that when reporting to a minority boss, minority subordinates receive lower pay, yet are more committed to their employers and more likely to expect to receive a promotion—results that are consistent with the reasons for and the effects of firms practicing bottom-up ascription when placing minorities in management positions.
Summary and Discussion
Scholars have rightly analyzed trends in and determinants of minority access to managerial positions, but lacking is an understanding of the effects of such access on pay inequality, working conditions, and career prospects among supervisors and subordinates. Progress has been made in eliminating race as a criterion for entry into management, but if this represents real progress, the race gap in managerial rewards and working conditions should narrow, and minority subordinates should describe their jobs and career prospects in more optimistic terms when reporting to a minority boss. This study tested these propositions with recent surveys of the labor force, finding that support for the optimistic or pessimistic views of minorities in management depended on minority managers’ placement in firms’ hierarchies.
The strongest support for the optimistic view of minorities in management was found in analyses using census criteria to distinguish between managers and nonmanagers. In these analyses, the minority pay gap declined across the authority divide and was negligible in the public sector and in large organizations. Furthermore, among census-defined managers, minorities and whites reported similar levels of autonomy, job security, and respectful treatment from colleagues. The Census Bureau’s definition of managers as professionals who exercise authority tends to identify those at the apex of the occupational hierarchy, where their relatively few numbers and high visibility within organizations makes it more difficult for employers to make job assignments and set pay levels based on race.
Below the upper echelon of managers are layers of middle and lower managers, where many technical, clerical, sales, and skilled blue-collar workers exercise supervisory authority over subordinates. The pessimistic view of minorities in management is that these positions are “glorified” administrative jobs that involve many of the same duties as the jobs occupied by those they are supervising (Jacobs 1992) and that minorities are often tapped to fill these supervisory roles to ensure the firm’s compliance with federal mandates on affirmative action and equal opportunity laws (Elliott and Smith 2004, 2001). When comparing supervisors and nonsupervisors, the findings support the pessimistic view of minorities in management. That is, minority wage residuals are nearly as large among supervisors as among nonsupervisors, and in large firms (where greater occupational differentiation increases the potential to differentially assign workers to jobs based on minority status), the minority pay gap was larger among supervisors than among nonsupervisors. Furthermore, among those with supervisory authority, minorities were assigned to less challenging and more insecure jobs, given less autonomy, and shown less respect at work.
From the point of view of subordinates, the results more strongly support the pessimistic view and, in particular, the argument that firms practice bottom-up ascription when tapping minorities to manage other minorities. That is, because of lingering racial bias on the part of white superiors and coworkers, minorities tend to concentrate at the bottom of the queue of desirable workers and hold devalued jobs within the firm. Yet to obtain desired behavior and commitment from devalued workers, minorities are assigned as supervisors, offering subordinates the hope that via hard work they too can ascend to managerial positions within the firm. Thus, subordinates in minority-dominated workgroups tended to report to a minority supervisor, and in such cases minority subordinates were more committed to the organization and more likely to expect to receive a promotion despite earning lower pay than their white coworkers. These results suggest that minority bosses are not necessarily “change agents” who can foster the careers of minority subordinates but are more likely to be “cogs in the machine,” whose role it is to ensure the commitment of subordinates to the firm even as they are paid less than whites.
On balance, then, the findings of this study support those who are skeptical about the degree of progress minorities have made in reaching the most rewarding and prestigious jobs in the American economy. These findings resonate with Tilly’s (1998) contention that even after the state intervenes in the operations of the labor market with the aim of reducing inequality, privileged actors find new ways to sustain old distinctions. Bottom-up ascription appears to be one particularly effective way in which minorities are allowed to enter the ranks of management but in a way that does not threaten the status of white men. By placing minority supervisors over devalued minority subordinates, white men are freer to monopolize the more rewarding management positions within the firm. The findings reported here suggest that minority supervisors are aware of their devalued status, reporting that their supervisory jobs are less challenging and more constraining than those of white supervisors, and they continue to lag behind the pay of their white counterparts. What is needed now is further study of whether minority supervisors have the means and the inclination to collectively push for stronger measures to combat labor market inequality, or whether they will seek individual remedies (e.g., disengagement, turnover) to their unequal treatment by their employers.
There are several avenues for further research that address the limitations of this study. First, this study examined control over workers, but managers also control fiscal resources, such as by deciding which products and services to offer and how to market new and existing products, arranging financing for existing operations and planned expansions, and so on. Comparing minorities with whites on the degree of control each has over the full range of important decisions made within the firm would provide additional evidence on how highly placed minority bosses are in the organizational hierarchy relative to white managers.
Second, the cross-sectional design of this study necessarily suggests that longitudinal analyses will be valuable in shedding light on the antecedents and effects of minority representation in management. For example, this study implied that because they receive less challenging and more circumscribed job assignments, the careers of minority supervisors would be stunted compared with whites placed in more challenging and autonomous jobs. With longitudinal data, cohorts of managers could be compared on initial placement, success or failure in revenue-enhancing projects, mentoring subordinates, and so forth. Although social scientists have studied race differences in career trajectories over time, there is a need for more focused analyses of managers and their effects on subordinates’ careers, given that there are now more minorities in the ranks of management.
Third, longitudinal data would also further test some of the key mechanisms of bottom-up ascription and how these processes unfold over time. That is, do the satisfaction and commitment of coworkers vary by whether a minority manager is hired from outside the firm or promoted from the pool of subordinates? If minority supervisors are placed in authority positions over devalued minority subordinates, are minority bosses at risk of leaving the firm, or are they rewarded handsomely enough to ensure that they remain with the firm? If a minority supervisor symbolizes potential mobility to minority subordinates, do minority subordinates increasingly disengage from their work or leave the firm when they grasp fully that their careers have plateaued? These are but some of the questions that need answers before we have a complete understanding of the effects on bosses and subordinates of increased minority representation in management.
Footnotes
Appendix
Descriptive Statistics on Variables Used in Analyses, Men Ages 18 to 64
| Min | Max | Mean | SD | |
|---|---|---|---|---|
| Dependent variables | ||||
| Annual pay (logged) | 8.29 | 13.82 | 10.67 | 0.61 |
| High-pressure job a | 1.00 | 5.00 | 3.88 | 0.91 |
| Job autonomy a | 1.00 | 5.00 | 4.11 | 0.86 |
| Disrespected on job a | 1.00 | 5.00 | 1.72 | 0.90 |
| Job insecurity a | 0.00 | 2.00 | 0.48 | 0.60 |
| Reports to minority superior b | 0.00 | 1.00 | 0.17 | 0.37 |
| Advancement opportunities b | 1.00 | 4.00 | 2.66 | 1.02 |
| Supervisor job-related support b | 1.00 | 5.00 | 4.36 | 0.80 |
| Organizational commitment b | −3.46 | 1.07 | −0.06 | 0.82 |
| Individual controls | ||||
| Nonwhite | 0.00 | 1.00 | 0.24 | 0.42 |
| Years of education | 9.00 | 20.00 | 13.66 | 2.88 |
| Married | 0.00 | 1.00 | 0.70 | 0.46 |
| Parent | 0.00 | 1.00 | 0.48 | 0.50 |
| Southern resident | 0.00 | 1.00 | 0.43 | 0.49 |
| Surveyed in 2002 | 0.00 | 1.00 | 0.46 | 0.50 |
| Pre-employer experience | 9.00 | 63.23 | 31.12 | 9.96 |
| Experience squared | 81.00 | 3998.12 | 1067.88 | 698.00 |
| Employer tenure | 0.00 | 45.15 | 7.93 | 2.72 |
| Tenure squared | 0.00 | 2038.52 | 56.23 | 9.83 |
| Pay is missing | 0.00 | 1.00 | 0.08 | 0.27 |
| Workplace context | ||||
| Government/nonprofit sector | 0.00 | 1.00 | 0.31 | 0.46 |
| Firm size | 1.00 | 10.00 | 5.72 | 3.35 |
| Race composition of coworkers b | 1.00 | 6.00 | 2.77 | 1.35 |
| Works part time b | 0.00 | 1.00 | 0.09 | 0.28 |
| Managerial occupation b | 0.00 | 1.00 | 0.13 | 0.34 |
| Self-reported supervisor b | 0.00 | 1.00 | 0.43 | 0.49 |
Supervisors or managers only, 1997 and 2002 NSCW (see Table 2).
Subordinate workers surveyed only in 2002 NSCW (see Table 3).
This research was supported by grants from the National Institute of Child Health and Human Development (R03-HD42411-01A1) and the Charles Phelps Taft Research Center at the University of Cincinnati. I thank Julie Kmec, Matt Huffman, Ryan Smith, and George Wilson for their comments on earlier drafts of this article.