The Research
Business leaders tend to focus their diversity, equity, and inclusion (DEI) efforts on addressing bias and discrimination because these practices are pervasive and discrimination is illegal. But in the process, they overlook situational factors that play a leading role — and are easier to address — in determining whether women and people of color get promoted and are among the higher paid.
Career and pay equity are inextricably linked. Women and people of color continue to be underrepresented among managers and executives, and often among senior professionals as well.1 While pay equity is improving for these demographic groups, the raw pay gaps (average and median differences between women and men, and between people of color and White employees) remain high in many companies. These pay gaps will persist until women and people of color occupy the career levels and roles that command the highest pay at a rate that reasonably reflects their representation in the workforce.
Inequity not only hurts individual employees; it also deprives businesses of the talent that is overlooked. If leaders understand how situational factors affect their employees’ prospects for advancement, they can apply proactive career management (PCM) — identifying and removing barriers to employee success — to make their workplaces more equitable and more productive.
To better understand the factors underlying these differences in career outcomes, I examined the results of analyses of multiple years of promotion, retention, performance, and pay data in 20 large and midsize companies that collectively employed more than 700,000 people. The findings confirmed that significant gender and racial differences in promotion probability for women and people of color are attributable to four factors that are consistently consequential to employees’ success: performance ratings, role assignments, reporting relationships, and flexible working arrangements.
With the exception of performance ratings, these factors are related to an employee’s circumstances, not their knowledge, skills, or behaviour. For instance, if career progress requires experience in certain roles but women are proportionally underrepresented in those roles relative to men, the net result is that fewer women will advance.
Certainly, bias and discrimination are also in play: In most of the organizations, we found large, unexplained gender and racial disparities in the likelihood of promotion that cannot be accounted for by any business-relevant factors. And we found universal and sizable racial disparities in performance evaluations, with people of color, particularly Black employees, less likely to receive high ratings. These are a logical focus for corporate diversity, equity, and inclusion efforts because they are evidence of bias or discrimination.
But because the explained differences often have the greatest influence on career advancement and pay equity, it’s important to address them directly. The good news is that these factors are often the easiest to address successfully. Proactive career management is a data-driven approach that offers a solution that works.
Four Factors Drive Who Advances at Work
Let’s look more closely at the four factors I identified that explain who receives promotions, high-performance ratings, and larger pay increases at work and why women and people of color tend not to see those benefits.
Performance ratings. Because performance ratings affect all other outcomes, such as promotion, retention, and pay, disparities in this area have a cascading effect. In 89% of the companies studied, Black employees were less likely to receive high-performance ratings. The differences can be sizable, with the majority at 25% or higher. Moreover, because performance ratings tend to persist from year to year — a high rating one year predicts the likelihood of a high rating the next — the tendency for people of color to get lower ratings becomes a virtual blockade to their progress. Interestingly, women are generally more likely to be highly rated than otherwise comparable men, but that doesn’t translate into a higher probability of promotion for them.
Access to accelerator roles. In every company, certain roles are career accelerators and others are dead ends, regardless of who occupies them. In this study, having a supervisor role at any level almost universally improved the chance that an employee would be promoted and receive high ratings. However, we almost always found that women and people of color were far less likely to be in such roles than men or White employees, respectively.
The differences are often stark and steep. For example, in a global financial services company we studied, U.S.-based employees who were supervisors were 50% more likely to be promoted and 55% more likely to receive high performance ratings than those who were not supervisors. Outside the U.S., supervisors were 89% more likely to be promoted and 37% more likely to get high ratings. These effects were even larger for female supervisors than for male supervisors. But across the company, women were half as likely to be supervisors.2
Further, because women are more likely than men to face career disruption during their child-rearing years, they may take on supervisor roles later. In such cases, they don’t experience the same compounding benefit of being in such roles as their male colleagues.
Reporting relationships. Whom one reports to can have a decisive impact on the trajectory of one’s career. We generally found that people who reported to supervisors with high-performance ratings were significantly more likely to receive a high rating as well. The probability that they would be promoted also increased.
We saw similar effects for employees whose managers were higher up in the corporate hierarchy than the norm for someone at their level and thus had more clout. To the extent that women and people of color are less likely to report to high-performing and higher-level supervisors, they are less likely to receive high ratings or promotions themselves.
They may also be less likely to benefit from these relationships when they do have them. At a global insurance company, those who reported to a top-rated supervisor were about 50% more likely to be promoted and 30% more likely to receive high-performance ratings. Those whose supervisors were at a level higher than the norm were 9% more likely to be promoted and 14% more likely to be highly rated. However, having a high-performing or higher-ranked supervisor did not help women and people of color as much as it helped White men.
Flexible working. Increasingly, organizations view flexible work arrangements as essential to attracting and retaining talent, particularly women and younger employees. Since the COVID-19 pandemic, more companies and employees have embraced such arrangements, but historically, they have been career killers. For example, those who work part-time or remotely or take family leave are less likely to be promoted, are paid less, and are more likely to leave their companies.
At a business I’ll refer to as ConsumerCo, opting for flexible work arrangements hurt both women and people of color. The percentage of women working part-time was three times that of men; for people of color, it was two times that of White employees. Overall, working part time reduced promotion probability by 50% and the likelihood of receiving a high rating by 70%. It also increased turnover probability by 25% compared with similarly situated full-time workers.
Likewise, leaves of absence were disproportionately taken by women and people of color. Those on leave were about 50% less likely to be promoted or receive high ratings and much more likely to turn over — at least 150% more likely in the U.S. and at least 80% more likely globally.3
Unless companies do something to mitigate the negative consequences of flexible working on employees, the phenomenon of higher levels of flexible work arrangements risks further undermining career progress for women and people of color.
The Case for Proactive Career Management
With the exception of performance ratings, these drivers of career progress, or lack thereof, are situational. This is telling. It indicates that whether employees advance is not just a reflection of the talents and skills they bring or what they do, but also of how well their employers position them to succeed. All too often, women and people of color simply aren’t positioned to succeed in the same way their White male counterparts are.
This is not to say that bias and discrimination are not factors, as mentioned above. However, employers can make measurable progress on DEI when they understand the extent of the positioning issues in their workforce and deal with them in a purposeful, systematic way. This is where PCM comes into play.
The idea behind proactive career management for DEI is simple: Identify and rank the factors that most influence career success and retention; if and where they skew significantly against certain demographic groups, intervene to eliminate or reduce the disparities. In other words, work to ensure that demographic characteristics bear no relationship to how well or poorly positioned an employee is to stay and succeed at work.
Every organization has talent flows that shape its workforce. Employers hire, train, and develop employees. And they value employees by giving them promotions, high ratings, raises, and other forms of compensation for the knowledge, capabilities, behavior, and experience they bring, the effort they deliver, and, ultimately, how they perform. Meanwhile, employees’ decisions to stay or leave are influenced by whether their careers are progressing.
These talents flows and rewards define the internal labor market of an organization and produce the supply of people with the attributes needed to meet the demands of the business.4 But as we have seen, different demographic groups within an organization do not have the same opportunities to acquire these attributes or be rewarded for them.
By not distributing opportunities equitably, companies effectively restrict their talent supplies. Women and people of color become more likely to leave or, if they stay, become less engaged because their careers are stalled. Knowing how the internal labor market functions for different demographic groups within an organization is thus crucial for a company to execute its DEI strategy.
To implement PCM, organizations should start by creating a success profile that identifies and ranks the factors that drive who advances. From there, they can examine how each factor influences the probability that employees from different demographic groups get promoted and whether the organization is valuing their capabilities and efforts equitably.
Small organizations might rely on the four factors described in the previous section. For them, controlled, statistical analyses of their own internal labor market data are likely infeasible. Such organizations might complement the research with qualitative input about the keys to career progress from their leaders and employees who have successfully risen in the ranks there.
Large employers with complex organizations will be best served by developing an objective success profile based on empirical evidence that identifies and ranks the critical individual and situational factors that drive success in the organization. The natural place to start is with the factors that drive promotions. But organizations might examine other markers of success as well, like pay, ratings, and retention, to see how well the results align with those from the analysis of promotions.
After creating the profile, the next step is to examine how the profile plays out for different demographic groups. There are two questions to answer.
First, are the factors that most influence employee success distributed similarly across demographic groups, or do they tend to be concentrated more in some groups than in others? These are quantity effects, and they reflect the relative frequency of the factors across groups. For instance, are women and people of color as likely as men and White employees to have the prized characteristics or to work in circumstances associated with success?
Second, when people possess the prized characteristics or work in favored circumstances, do they benefit equally from those attributes, or do some demographic groups benefit more? These are price effects. They reflect whether the employer values the attribute, behavior, or situation in question differently for one group than for another.
Objectively answering these questions is the foundation of PCM, as illustrated in the case of ConsumerCo.
Positioning Employees for Success at ConsumerCo
ConsumerCo’s leaders were committed to improving career and pay equity in both their salaried and front-line workforces.
The company recognized that having a diverse workforce was not a luxury or a concession to social good; it was essential to its success. With an increasingly diverse customer base and growth occurring mostly in markets outside the U.S. and Europe, having a leadership team and a workforce well attuned to the preferences and norms of local cultures was important to achieving growth targets.
HR and DEI leaders modeled their workforce data to uncover and assess the success profile for their salaried workforce. They identified 12 factors that were most likely to predict whether someone would be promoted. Situational factors dominated this success profile — a finding that stunned company leaders, who viewed the culture as a meritocracy. Of the top 12, only performance ratings and tenure were specific to the talents or accomplishments of an individual employee. The rest were purely circumstantial.
Being an expat (working outside one’s home country) was the most positively influential factor. Reporting to a supervisor at a higher level than is typical had the least impact among the top 12 factors, but the effect was statistically significant.
Other factors included three of the four described earlier: having a high performance rating, being in a supervisory role, and reporting to a highly rated supervisor. In addition, working in certain functions or business lines mattered a lot to how one’s career progressed.
Similar modeling showed that even the probability of receiving a high rating was inordinately influenced by situational factors. This finding suggested that performance ratings were not accurate measures of individuals’ contributions.
When ConsumerCo’s leaders mapped the success profile to women in the company, they found that when women were positioned in situations conducive to their future progress, they usually fared as well as men. But overall, they were less likely to be so positioned. (See “Mapping Women’s Experiences to a Success Profile.”) For instance, women who were expats were more likely to be promoted from those roles than otherwise comparable men, but women were less represented among the expat population.
Being a supervisor is also an important stepping stone to success at ConsumerCo, but the company learned that women were less represented in supervisor roles. When they did have those roles, it did not help their careers to the same extent as comparable men, for reasons that were not clear. ConsumerCo leaders could see that female employees faced a double whammy of sorts: Not only were they blocked from this accelerator role, but they also didn’t get the same lift to their promotion prospects as their male colleagues when they did have that role.
When the company looked at people of color, it found that they tended to be both less well positioned and less valued than White employees. ConsumerCo needed to change its internal labor market dynamics to achieve career and pay equity for women and people of color.
A Checklist for Change
Focusing on situational factors has a distinct advantage: They’re easier to address than individual characteristics. For instance, although tenure is an important driver of success, it is a fact of people’s employment history, not something that ConsumerCo could change overnight.
Still, ConsumerCo identified several areas where it could act reasonably quickly, such as changing reporting relationships for women and people of color, finding opportunities for them to gain more supervisory experience earlier in their careers, and addressing the disparities in performance ratings for people of color.
The company had found that taking a leave of absence thwarted career advancement, especially for women, and so it is also developing new leave and career development policies to address imbalances for that group.
Here’s where the “proactive” part of PCM comes in.
The analysis of ConsumerCo’s success profile made it possible to create a simple checklist of green lights (career advancers) and red flags (success blockers) that managers could use to evaluate the positioning of individual employees. By considering whether women and people of color on their teams were experiencing more green lights or red flags, line managers and supervisors could identify who was well-positioned to stay and thrive, as well as who might be at risk of stalling in their career or leaving the organization. (See “Green Lights, Red Flags.”)
Meanwhile, HR and DEI leaders used the analysis to evaluate at the aggregate level how employees from different demographic groups were positioned to succeed. For each factor, they could effectively evaluate the situational reality of different groups and tailor their interventions to address the most significant gaps.
For instance, they found that female, Black, and Hispanic employees were less likely to be reporting to highly rated supervisors, while Asian employees were more likely to have this advantage. And although women were less likely to be reporting to a manager one grade higher than usual, there were no such disparities for people of color. Also, while every non-White racial group was significantly underrepresented in one of the core functions that most contributed to advancement, women were more likely to work in that function than men.
ConsumerCo is in the early stages of rolling out interventions to address its disparities. But other companies I have worked with have pursued a variety of solutions.
Closing gaps due to flexible work arrangements often requires changing the culture. In many companies, part-time work is looked upon as evidence of lower commitment, and it can be difficult to shake these perceptions. However, companies can make quick progress on pay if they stop factoring leaves of absence into the statistical models that they use to assess pay equity and adjust wages and salaries to close gaps.
Situation-based racial disparities in performance ratings have been reduced in some companies through the calibration process used to align ratings by different supervisors. Some have also eliminated forced rankings that provide for a very limited percentage of employees to receive top ratings. These tend to favor people with more tenure, whom supervisors have known longer, and presumably better — and who are more likely to be White.
Some clients have addressed the importance of supervisor roles by expanding training, both to reinforce the importance of these roles and to help supervisors understand the impact they have on employees’ prospects. Leaders can offer training to help women and people of color gain the experience they need to advance — and close the gap between women and men in these positions.
Proactive Career Management Has Benefits and Risks
Because a success profile is built from an organization’s data, it is more likely to reflect reality than the assumptions that leaders make about how employees succeed. Leaders may be prone to thinking that what got them to the top could get anyone else there as well, and they may oversimplify their own path. Further, people tend to generalize based on their own experiences, even if they don’t represent the organizational norm.
A good empirical model can counter the tendency of people to engage in wishful thinking or to cling to belief. It can also account for the effects of being in different business lines, functions, geographies, and job families. This is something that’s hard to do based on personal observation and experience alone.
Another value of PCM is that it is grounded in what creates value for the business. If, for instance, a role becomes critical to growth, one would expect the employer to accord more value to that role and the employees who successfully deliver value in it. And to the extent that an organization’s DEI strategy helps ensure that high-potential women and people of color are able to access these newly valuable roles and perform them ably, it will serve the greater business need.
If there is a downside to PCM, it relates to the sharing of sensitive data and intelligence. PCM proceeds from objective modeling of the drivers of success and how they affect different groups. Invariably, the models will surface potentially damaging information about the existence of systemic gender or racial disparities in outcomes and opportunities, which creates legal risk.
PCM works best when managers understand the realities uncovered by the models and the reasons why specific actions are warranted. Thus, it requires a reasonable level of transparency. Corporate legal departments may, however, object to exposing what they may believe to be strong evidence of bias and inequity.
There’s a delicate balance to be achieved. Most organizations I work with accomplish it by sharing high-level information with business and line leaders, and sometimes employee groups, about the reasons for particular actions while restricting the more granular data to a smaller group of top executives and HR and DEI leaders.
Legal counsel almost always presides over such decisions; however, such advisers have become more understanding of the need for data-driven DEI strategies, appreciating that people can’t solve problems they don’t know about or understand.
The unexplained disparities in career opportunity and advancement at many organizations are important to address because they signal the potential influence of bias and discrimination.
But doing so should not override attention to differences that can be explained by legitimate, business-relevant factors. The surest, most sustainable way to create career equity and equitable representation across all levels of an organization is to address the organization’s career development practices (or lack thereof) that systemically impede the progress of women and people of color.
Of course, the work that employees do, their knowledge and skills, and how hard and effectively they work influence whether they will succeed. But success is also determined by how individuals are positioned to succeed — and organizations can do better at ensuring that all high-potential talent can gain experiences that will set them up for growth.
“The MIT Sloan Management Review is a research-based magazine and digital platform for business executives published at the MIT Sloan School of Management.”
Please visit the firm link to site
