Diversity matters in the workplace. It is an important social issue, and a performance imperative: more diverse top-management teams appear to benefit from a richer decision-making dialogue, which can contribute to better financial performance.1
CEO transitions matter, too. Our research has shown that a CEO’s likelihood of outperforming his or her peers depends heavily on the mix of strategic and organizational decisions made during the first two years on the job. Management reshuffles—a critical piece of the performance puzzle for many new CEOs, according to our research—should create opportunities for new CEOs to boost gender diversity. Too few do so, however, suggesting that new CEOs, and the boards that hire them, should be asking tougher questions about diversity and asking those questions sooner than they normally do.
A missed opportunity
At the beginning of their tenures, new CEOs typically change the makeup of their management teams. Our research shows that more than two-thirds of chief executives replace at least half of the members of their top teams within two years of taking office.2 They may do so to strengthen the capabilities of those teams, to embark on new strategic directions, or simply to replace former peers they had competed against for the top job, who may have different ideas about the way ahead. The management reshuffles that happen during transitions hold the potential to serve as “unfreezing moments,” dramatically improving the representation of women at senior levels and sending a strong signal to the organization that this issue matters and that the CEO expects to increase gender diversity going forward.
Yet only a small number of new CEOs are taking advantage of the narrow window of opportunity a transition provides to boost the top team’s diversity (see sidebar, “About the research”). For example, we found that within three years, gender diversity in senior teams that new CEOs reshuffled increased by only two percentage points—raising the proportion of women in management to only 14 percent, from 12 percent. The picture of female representation didn’t improve when we expanded the time period to cover management reshuffles over the entirety of the CEOs’ tenures.
Discover and subscribe to McKinsey Quarterly Audio
This finding suggests that even if a dearth of women in the management pipeline limited progress during the transition period, those same CEOs didn’t change the pipeline and promotion picture during their tenures. The trend was consistent across time as well: CEOs who took charge in recent years were no more likely to promote women to senior roles than those who became corporate leaders 20 or 30 years ago. And though our data focused solely on gender, research by our colleagues on the additional difficulties faced by women of color suggests that top-team transitions do little to help on that front either.3 Behind all the apparent inaction—and missed opportunities—we found three complex underlying patterns.
Up from the bottom
First, new CEOs in the least diverse companies and industries seem to make the most significant improvements in gender diversity over the course of their tenures (exhibit). Chief executives who took over companies where women made up less than 15 percent of the senior-management team, for example, increased female representation, on average, to 14 percent, from 10 percent—twice the level of improvement achieved by all CEOs who undertook management reshuffles. While the sample size is unfortunately small, the same effects are seen when looking at female incoming CEOs specifically.
Digging deeper, we found that CEOs who take the helm of companies in historically male-dominated industries made the most significant improvements, although the sample size was small. For instance, new CEOs in heavy-industry sectors, which had the lowest levels of female representation at the start of their tenures, more than doubled it on their executive teams, to 13 percent, from an average of 5 percent. Although the companies these CEOs led started from a lower base and had the greatest room to improve, it is still positive that their companies are addressing major imbalances even when the talent pipeline doesn’t make this easy.
Would you like to learn more about our Strategy & Corporate Finance Practice?
The cost of complacency
Our second finding was that, eventually, diversity appears to hit a ceiling. New CEOs at companies with the highest percentage of women in senior roles at the point of transition were the least likely to improve gender diversity. On average, in fact, companies with new CEOs where women made up 15 percent or more of the management team actually saw a reduction in the proportion of women in senior roles during reshuffles.
We take this finding to mean that more diverse companies tend to become complacent over time: the arrival of a new CEO is more likely to result in stagnation or decline than to help the organization capitalize on its momentum or positive starting position. The evidence suggests that once companies reach a minimum standard of diversity, the perceptions of their leaders—and, as a result, their priorities—change. This conclusion is consistent with the finding that nearly 50 percent of men believe that women are well-represented in leadership roles in companies when they account for only one in ten executives.
The insider’s edge
Finally, as the exhibit shows, our research reveals that CEOs promoted from within companies increase their gender diversity to a much greater extent, on average, than those hired externally. The difference is stark: internal CEOs raised female representation on management teams by nearly six percentage points more than external CEOs, who kept gender ratios stable, on average. Again, this is also the case when looking only at the female CEOs in our data set.
This finding offers an interesting counterpoint to some conclusions of our earlier research on transitions more broadly. In that work, we found that CEOs hired from outside companies were typically bolder in the number of strategic moves they made early in the game. As a result, they outperformed other CEOs over their tenures, on average.
The apparent divergence between bold strategic moves, on the one hand, and a lack of corresponding boldness in addressing gender issues, on the other, may result at least partly from the difficulties some leaders face in overcoming unconscious bias among other members of the top-management team. CEOs promoted from inside tend to know where the talent is, and that helps them mitigate the impact of biases among other senior executives. External appointees are less likely to have the same richness of information and may therefore find themselves defaulting to male-skewing conventional picks recommended by other leaders or the board. Of course, both inside and outside CEO hires are also susceptible to—and must guard against—their own unconscious biases.
Still looking for room at the top: Ten years of research on women in the workplace
Tough questions
Even if CEOs do make progress on gender balance early in their tenures, when they have a mandate to undertake significant management reshuffles, the job isn’t finished. New CEOs who aspire to create an inclusive culture that drives significant progress on gender diversity must ask and answer several difficult questions:
- How do I communicate the economic and strategic imperative of creating a diverse top team and make this a shared goal throughout the organization?
- What specific measures to improve gender diversity are appropriate for my organization, and how will I ensure that they take effect lower down the ladder?
- How do I make sure that women are moving into roles with profit-and-loss responsibility, as well as roles overseeing support functions, to prepare them for broader executive roles?
- How can I accelerate the pipeline of female talent while ensuring that fast- tracked women are supported and helped to succeed?
Success in this context is perhaps best measured by the legacy that CEOs create for their successors: Will those who follow them be starting afresh from a disappointing position, or maintaining momentum on the back of real progress?
