When US Supreme Court Justice Ruth Bader Ginsburg died in 2020, Harvard Business School Assistant Professor Edward Chang noticed something interesting: To fill the vacancy, then-President Donald Trump replaced Ginsburg with another woman, Amy Coney Barrett, even though he had previously added two men to the bench.
“There was plenty of evidence showing Trump wasn’t particularly pro-diversity,” says Chang, noting that Trump had issued a memo telling federal agencies to halt diversity trainings because they were “un-American.” “It was interesting that he chose to nominate a woman. Did the fact that he was replacing a woman as opposed to a man affect his decision?”
Most likely, yes. In studying the appointments of more than 2,000 federal judges and more than 5,000 corporate board members, Chang found that leaders have a strong tendency to replace “like people with like people” by matching the race and gender of a new appointment to the demographics of the departing member of a group.
“People are otherwise likely to choose replacements who demographically resemble their predecessors.”
This preference for maintaining the demographic status quo might help explain why the numbers of women and people of color in leadership roles continue to remain scarce, perpetuating a pattern of sameness, with white men often replacing other white men in high-level positions, Chang says.
However, the research results also reveal a bright spot: Organizations that make progress in appointing women and people of color to senior roles tend to hire people with similar demographics to replace them when they leave, allowing for diversity to “stick.”
The findings also raise an important question at a time when many corporate diversity, equity, and inclusion (DEI) efforts are at a crossroads: Does this tendency to keep demographics the same from one leader to the next stand in the way of meaningful change?
“The data suggest that we’re unlikely to make serious progress on diversity without concerted effort, as people are otherwise likely to choose replacements who demographically resemble their predecessors,” says Chang, who coauthored the study with Erika Kirgios of the University of Chicago’s Booth School of Business. The findings were recently published in the journal Management Science.
The departing leader’s identity matters
To explore the demographics of leadership roles, Chang and Kirgios studied more than 2,000 judicial appointments to federal courts over 75 years. They found that both in the Supreme Court and lower courts, departing judges were much more likely to be replaced by people who looked like them.
For example:
- Women replaced women 37 percent of the time. That’s double the rate that women replaced men on the bench: 16 percent.
- Black people replaced Black people in 25 percent of successions. For other replacements, Black judges were appointed 7 percent of the time.
- White men replaced white men at a 76 percent rate. But when a predecessor wasn’t a white man, that number dropped to 45 percent.
The researchers also found that judges were replaced by someone from the same specific demographic group. “It’s also a specific race matching—a Black judge is not more likely to be replaced by a Hispanic or Asian judge, but instead they’re likely to be replaced by a Black judge,” Chang says.
Corporate boards still prefer men and white people
The researchers conducted a similar analysis with board seats from companies in the S&P 1500 between 2009 and 2014, looking at more than 5,000 selections of new directors. Again, the team found a stronger likelihood that a board member would be replaced by someone from the same demographic, though the correlation was not quite as strong, Chang says, in part perhaps because board appointments tend to be more in flux, without the strict one-to-one replacements found in the judiciary.
Although women and racial minorities had a better chance of being named to a board when they were replacing someone of the same gender and race, overall, diversification remains slow, with boards still preferring men and white people overall. In fact, when replacing departing women or racial minorities, boards chose men 65 percent of the time and white people 75 percent of the time.
“People ... seem to ignore the diversity of the remaining group, which is what you’d think they’d care about if they think that diversity is good overall.”
To further explore what they call the “demographic stickiness” phenomenon, the researchers conducted experiments online, telling participants the makeup of a board of directors, as well as the identity of a departing board member, and asking them to recommend a replacement from a list of available candidates of different races and genders.
Once again, the team found a preference for replacing like people with like people. When told a departing member was a white woman, for example, they found people were likely to replace her with another white woman 57 percent of the time, versus 43 percent of the time when they were not told the identity of the departing member. The same was true of replacing a departing Black man—42 percent replaced the person with another Black man when told the departing person’s identity versus 34 percent when the identity was withheld.
In addition, the researchers found that when a white man was departing, participants still showed a propensity to replace him with another white man, even when white men continued to hold the majority of the board. “People myopically attend to the identity of the person who is leaving,” Chang says. “They seem to ignore the diversity of the remaining group, which is what you’d think they’d care about if they think that diversity is good overall.”
What’s driving these decisions?
Chang speculates that a behavioral phenomenon known as “loss aversion,” in which people focus on avoiding losses rather than reaping potential gains, might be influencing board recommendations. In other words, people may be motivated to match outgoing and incoming demographics to preserve group dynamics. They may also place more weight on avoiding diversity losses than on seeking greater diversity gains, so they don’t take bigger steps to diversify, even when they have the opportunity to move the needle.
“There’s a pessimistic interpretation of the results, which is that we are identifying these human decision-making tendencies that explain why demographic change is so slow,” Chang says.
“Once demographic change has happened, it’s much stickier than people might expect.”
However, the researchers also found that once an organization made a change, the same phenomenon that perpetuated the hiring of white men also applied to keeping women and people of color in power.
“Once demographic change has happened, it’s much stickier than people might expect,” Chang says. Diversity in judgeships, for example, generally increased more under Democratic administrations; however, those diversity gains remained under Republican presidents. “For the same reasons demographic change is slow, it also doesn’t tend to backslide.”
Three steps to improving diversity
Chang says organizations can take some important steps to diversify their leadership roles:
Focus on smaller, short-term goals. While many organizations are working to change people’s attitudes in the long term, for example, by conducting diversity training to inform broad hiring decisions throughout the workplace, they should also attempt to make small changes in the short term, for example, by focusing on implementing one-time interventions to hiring processes with the goal of diversifying one role at a time.
Consider the overall diversity of the leadership group, rather than focusing on the identity of departing members. “Be conscious that you’re not just replacing like with like,” he says. That could also mean considering diversity more broadly than just replacing a departing woman with another woman or a person of a particular race with the same race. “Just because you are making progress on gender diversity does not mean that you will naturally make progress on racial diversity or vice versa,” he says.
Recognize that any progress is good progress. That is, when leaders take even small steps in the direction of diversity, they should keep in mind that those efforts may have significant staying power. “Even once you are no longer part of the organization or in a decision-making role,” Chang says, “you can have optimism that any progress you made will probably persist beyond your tenure.”
You Might Also Like:
- Career Advice for Minorities and Women: Sharing Your Identity Can Open Doors
- Beyond the 'Business Case' in DEI: 6 Steps Toward Meaningful Change
- Unpacking That Icky Feeling of 'Shopping' for Diverse Job Candidates
Feedback or ideas to share? Email the Working Knowledge team at hbswk@hbs.edu.
Designer's Note: Image created using photos from AdobeStock/Atlas.