Recently, Hillary Clinton promised that, if elected, she will name women to half of her presidential cabinet posts. If that happens, Clinton would join a handful of other global leaders who have appointed equal numbers of men and women — including, most recently, Canada’s Prime Minister Justin Trudeau.

Why would Clinton promise a gender-parity cabinet? And how likely is she to achieve this goal? Cross-national research on women in the executive branch suggests that obtaining parity is difficult but not impossible.

Why do politicians promise gender-balanced cabinets?

The decision is partly ideological. Left-leaning leaders in Latin America and Western Europe typically appoint more female ministers than right-leaning governments. This is consistent with these parties’ broader policy commitments to women’s rights and feminist issues. Promising a 50/50 cabinet can also be politically expedient. Because these governments are still rare, leaders can use them to establish their pro-women or feminist credentials. Spain’s José Zapatero, Italy’s Matteo Renzi, and France’s François Hollande were each widely commended for their parity cabinets. For Clinton, appointing an equal number of women and men would not only fit her ideology, but would also showcase her loyalty to the large number of women who have organized to support her.

How likely are leaders to have gender-equal cabinets?

What do trends from other presidential systems suggest about the likelihood of achieving a parity cabinet? Catherine Reyes-Housholder’s forthcoming study finds that gender-balanced cabinets remain extremely unusual. Her analysis of almost 2,000 appointments by 54 presidents across 18 Latin American countries shows that female chief executives are more likely to name women to their cabinets. But that’s most likely for inaugural cabinets and for traditionally “feminine” posts, like education and health.

Reyes-Housholder argues that female presidents may appoint more women to their cabinets for at least two reasons. First, they believe that’s what their constituents expect. It’s easy to see why a female president — particularly the first woman in the post — would believe that voters want to see more women wielding power in the executive branch.

Second, female leaders are more likely than men to network with other female politicians. As a result, they feel closer, both personally and politically, and more loyal to each other— two qualities that chief executives regularly want in their ministers.

Both of those appear to be true for Clinton. She could easily feel that her voters expect her to appoint women — and her network seems to contain a larger number of elite women with ample political capital.

So why don’t female leaders always appoint gender-equal cabinets?

But it won’t be as simple for Clinton as for those Latin American presidents; she’ll face some added constraints. Consider, for instance, that Mona Lena Krook and Diana Z. O’Brien’s global study finds no relationship between female chief executives and women’s cabinet appointments.

Parliamentary systems seem to make it especially hard for female leaders. Research on advanced industrialized parliamentary democracies by O’Brien, Matthew Mendez, Jordan Carr Peterson and Jihyun Shin found that male-led, left-leaning governments actually appoint more women to cabinets than those with female prime ministers or female-led coalition parties. Why?

O’Brien and her colleagues argue that unlike presidents, female prime ministers and party leaders always face the possibility that they could be removed from their post. Female leaders may feel especially susceptible to male challengers — and try to cultivate their loyalty by appointing these men to cabinets. These women also need to be conscious of how their actions will be portrayed in the media and perceived by voters. Appointing too many women in high-profile posts may be interpreted not as progressive but as driven by favoritism or sullied by identity politics.

Of course, Clinton will be a president rather than a prime minister — but the constraints on her are somewhere between those facing her European and Latin American counterparts. Clinton would enjoy a wider pool of potential female nominees than European prime ministers, who often must select appointees from among those elected to parliament. But Latin American leaders don’t need Congress to approve their nominees — as Clinton will.

Some evidence suggests that Clinton may also face backlash. Chile’s first female president, Michelle Bachelet, the only woman in Latin America to name a 50/50 cabinet, encountered harsh criticism from both the right and from members of her own coalition. It’s no surprise that Republican candidate Donald Trump said Clinton relies on the “woman card” right after her parity promise.

We think Clinton’s cabinet will be significantly female, if not 50/50

Given our research, we think Clinton would appoint women to a significant number of cabinet posts. But female chief executives with 50/50 cabinets are rare. Bachelet is actually an exception: globally, men have named almost all of the parity cabinets.

Even if Clinton were to appoint a gender-balanced cabinet, men could still wield more power within it than women. Chief executives around the world tend to name women to “feminineportfolios, which often have smaller budgets and attract less media attention. This means that women usually remain excluded from the most powerful positions, including finance, defense, interior affairs and foreign affairs (although the United States is an exception for that last, having had a female secretary of state for 12 of the past 19 years).

If Clinton appoints a 50/50 cabinet — and places women into posts usually held by men — it will be a major step forward for women’s political representation.

Diana Z. O’Brien is an assistant professor of political science at Indiana University, where she researches the causes and consequences of women’s access to political power.

Catherine Reyes-Housholder is a PhD candidate in Cornell University’s department of government, where she is researching the effects of female chief executives.