Karoline Leavitt Didn’t Even Make It 90 Days Post Leave. She’s Not The Reason Why. That’s the Story.

Karoline Leavitt returned to the White House podium on July 16 after maternity leave for her second child.¹ By mid-August, she’d announced she was stepping down as press secretary, saying in her resignation post that she’d come to feel she couldn’t be the mother her two young children deserve while carrying the demands of the job.²

The coverage has largely framed this as a personal choice: a devoted mother prioritizing family. Trump’s own statement leaned into that story too, framing her exit as a decision to spend more time with her family.³ And to be fair, that’s almost certainly part of the truth.

But look at the pattern around her. Reporting this month put Leavitt as the sixth woman from Trump’s inner circle to depart in 2026 alone, following the earlier exits of DNI Tulsi Gabbard, AG Pam Bondi, DHS Secretary Kristi Noem, Labor Secretary Lori Chavez-DeRemer, and two press-office staffers.⁴ And her own exit didn’t even wait the full 90 days we see and retention research confirms as the danger window after parental leave. She was gone within about four weeks of returning to work.

If you work in HR, benefits, or leave management, you should recognize this shape immediately. I’ve spent years studying it. I call it the 90-Day Retention Trap: the leave was compliant, the employee returned on schedule, and shortly after, sometimes well within 90 days, they resign anyway.

Here’s the part that gets missed in almost every headline about women “choosing family”: leave length is not the exclusive variable that predicts whether someone stays.

That’s not my opinion. It’s what the evidence-based research says. Leslie Hammer and Ellen Kossek’s foundational work built and validated the concept of Family-Supportive Supervisor Behaviors, or FSSB, the specific things a direct manager does (or doesn’t do) that determine whether an employee’s work and family roles feel workable together.⁶

Kossek, Hammer, and colleagues just tested that foundation directly against leave, in a randomized field study published in the Journal of Applied Psychology this year. Managers at one organization got trained on six specific leave-supportive behaviors, what the researchers nickname the LEADER model: Listen and give emotional support, Educate employees on their options, Address stigma by creating psychological safety, use a Dual agenda to creatively problem-solve for both the team and the employee, Enhance the leave experience by respecting time off, and Role model leave use themselves. Managers at a comparison organization got nothing. Same paid leave policy, same eligibility, same benefits on paper, in both places.⁵

The employees whose managers got trained used more family leave, reported higher job satisfaction, and reported stronger organizational commitment than employees whose managers didn’t, even though the underlying policy never changed. In a follow-up exploratory analysis, the researchers also found that role modeling in particular stood out: managers who visibly used their own leave, who talked openly about taking time off for their own health or family, showed a measurable link to those improved outcomes that the other five behaviors didn’t show as clearly. It’s not the only behavior that matters, but it’s one worth paying close attention to. The researchers’ own conclusion: offering paid sick and family leave is not enough.⁵

In other words: the number of weeks on the books matter for retention, but they matter less than what happens in the weeks before and after someone walks back in the door, and less than whether that employee’s manager was trained to handle it, including whether that manager modeled healthy leave-taking themselves.

So What Did Leavitt’s Manager Actually Do?

I invite you to consider that every manager is your organization’s leave strategy whether they (or you) know it or not. The only question is whether they’re a good one. So it’s worth asking, plainly: on the behaviors the research says actually move the needle, what did Leavitt’s boss do?

Trump is, functionally, her manager. And by the reporting available, none of those behaviors were present in any protective way:

• No educating or reducing stigma. There’s no indication anyone talked openly with her about her options, normalized the request, or worked to take the fear out of asking for what she needed. Silence around a leave request is its own message.
• No emotional support. Reporting on the culture around her described a workplace where women who’ve worked alongside her say the expectation was to “be available 24/7,” with little room for the kind of validating support the research identifies as protective.⁴
• No role modeling, one of the behaviors the research flagged as especially important. That same reporting describes an around-the-clock president who rarely seems to sleep and expects staff to match his pace. You cannot model sustainable, leave-supportive behavior you don’t practice yourself, and on this one in particular, Trump fails badly.
• No dual-agenda problem-solving. No phased return, no interim ramp period, no restructured re-entry workload waiting for her. She was back at the podium within days of returning, at full intensity, with the news cycle exactly as demanding as she’d left it. There’s no public indication anyone sat down before her return and asked, “What would make the next six months workable for you?” The job simply resumed, unchanged, the day she walked back in.

That’s not a personal indictment. It’s a diagnostic. Run any manager through those questions and you’ll know, with real predictive power, whether they’re about to lose a great employee. This is part of the manager assessment I walked HR and leave management teams through in my presentation at the 2026 Annual DMEC Conference earlier this month, and do in-house with our clients: your leave strategy isn’t your policy. It’s whoever that employee reports to.

These Are Not the People Who Planned to Quit

I want to be precise about who we’re talking about, because this is where most coverage of “moms leaving high-power jobs” gets it wrong. The women showing up in the 90-Day Retention Trap are not the ones who took paid leave already intending to quit. They are the opposite: ambitious, career-identified, high performers who wanted to stay and had enormous amounts left to give.

Karoline Leavitt is a textbook example. She built her entire career, at record speed, around this exact job, becoming the youngest press secretary in U.S. history, a role she pursued deliberately since college.¹⁰ People who plan an exit during leave don’t do that. People who lose a job they spent a decade building toward, because nobody made the return workable, do.

That distinction matters enormously for how employers should read attrition data. If you assume every departing new mother “always intended to leave,” you’ll keep doing nothing. If you recognize that your highest performers are walking away from careers they wanted with you (and going to a competitor), you’ll realize the leak is fixable, because it isn’t about desire. It’s about design.

This Is Exactly Why Our Leave Leader™ Manager Training Series Exists

This isn’t a hypothetical connection to the research, for me. Drs. Kossek and Hammer are two of the most established manager-behavior researchers in the field, but their expertise is in supervisor behavior broadly, not parental leave specifically. That’s what they brought me in for: I served as the subject matter expert on parental leave to the team that ran this study, drawing on years of frontline work with managers and employees navigating that exact transition, to help shape the research’s approach to the manager-employee relationship during leave. My contribution is formally acknowledged in the published Journal of Applied Psychology article.

The Center for Parental Leave Leadership is also the exclusive licensee of the two trainings that came directly out of this research: Leaders and Leaves™, the actual intervention tested in the randomized trial above, and Leave Supportive Supervisory Behaviors™, built on their foundational FSSB research. Both were developed by Drs. Kossek and Hammer. Paired with Making Leave Manageable®, our own training built on CPLL’s RETAIN Coaching™ methodology for the parental leave transition specifically, that’s the full Leave Leader™ Manager Training Series: the only place a manager can get trained on both the peer-reviewed, trial-tested behaviors from this study and a coaching framework built specifically for the parental leave transition, in one certificate pathway.

There’s a core message I’ve been bringing to every conference presentation and every leadership team I sit down with this year: compliance is not a retention strategy. Managers are. You can run a perfectly compliant leave program, paperwork perfect, return date honored, pay sent, and still lose your best people, because HR cannot be in the room for the hundreds of micro-transitions that happen during what we define as the 3 Phases of Leave: preparing, during, and returning. The manager is in that room, every day (and in that employee’s thoughts). And in most organizations, that manager has had zero training on what to actually do in it.

Trump isn’t going to sit through a Leave Leader™ training, and the Oval Office is about as far from a typical HR context as it gets. But strip away the presidency and the podium, and what’s left is a familiar shape: an employee who wanted to stay, and a boss who never learned, or never practiced, any of the behaviors that would have made staying possible. That same gap plays out in break rooms, Slack threads, and Zoom calls everywhere. It just doesn’t usually make the news.

A few things worth saying plainly, because the discourse around this topic tends to flatten into fatalism:

It is not inevitable that women leave the workforce after becoming mothers. Maternal labor force participation has risen for decades. AND culture matters. Right now, arguably, it’s about the worst moment in recent memory to be a working mother in America: in the most recent jobs reports, women accounted for the entirety of the labor force’s net decline while men’s participation held steady or grew. And if we go back to January 2026, 845,000 women have left the workforce compared to 406,000 men, more than double the rate.⁹ Today’s working mother attrition is real, but it’s concentrated where support is absent, not baked into motherhood itself.

Longer leave helps (a lot), but it doesn’t appear to be the lever that moves the needle most. When the U.S. military doubled paid maternity leave from 6 to 12 weeks, attrition among pregnant servicemembers dropped from roughly 23% to 18%.⁷ Google saw an even bigger swing on the civilian side: when it extended paid maternity leave from 12 to 18 weeks in 2007, attrition among new mothers fell by 50%, according to former YouTube CEO Susan Wojcicki, who championed the change.⁸ Leave length matters, clearly, sometimes enormously. But length without a supported return is still a leaky bucket. In my experience, you can hand someone 12 or 18 weeks and lose them in week 16 or 22 if nobody restructured their re-entry workload, checked in, or modeled that it’s okay to leave at 5:30 for daycare pickup.

“She chose family” is often doing narrative work that “the job was never (re)designed to make staying possible” should be doing instead. Those are different stories with different implications for what a workplace should fix.

The paperwork was perfect. The leave was compliant. She returned on schedule. And still, well before 90 days, one of the most ambitious, capable people in that building was gone.

That should bother every employer watching this story, regardless of politics.

Sources

1. NPR, “Karoline Leavitt is leaving as White House press secretary,” Aug. 12, 2026. https://www.npr.org/2026/08/12/nx-s1-5929560/karoline-leavitt-white-house-press-secretary-leaving
2. NBC News, “Karoline Leavitt to step down as White House press secretary at end of month,” Aug. 12, 2026. https://www.nbcnews.com/politics/white-house/press-secretary-karoline-leavitt-step-down-end-of-august-rcna592223
3. ABC News, “Trump announces White House press secretary Karoline Leavitt stepping down.” https://abcnews.com/Politics/trump-announces-white-house-press-secretary-karoline-leavitt/story?id=135590424
4. The Daily Beast, “MAGA Ladies Spill on Brutal Reality of Working for Trump After Leavitt Exit.” https://www.thedailybeast.com/maga-ladies-spill-on-brutal-reality-of-working-for-trump-after-leavitt-exit/
5. Kossek, E. E., Allen, S. J., Lawson, K. M., Hammer, L. B., Bodner, T., Xu, J., Holladay, C. L., Woods Herron, A., Cavanaugh, K. J., & Perry, M. (2026). Offering paid sick and family leave is not enough: Leave supportive supervisor behavior training improves policy use, health, and work outcomes. Journal of Applied Psychology. Advance online publication, July 20, 2026. https://doi.org/10.1037/apl0001395
6. Hammer, L. B., Kossek, E. E., Yragui, N. L., Bodner, T. E., & Hanson, G. C. (2009). Development and validation of a multidimensional measure of family supportive supervisor behaviors (FSSB). Journal of Management, 35(4), 837–856. https://pmc.ncbi.nlm.nih.gov/articles/PMC3109661/pdf/nihms-295010.pdf
7. University of Nebraska–Lincoln, “Longer maternity leave linked to lower attrition in U.S. military,” Oct. 20, 2023. https://research.unl.edu/blog/longer-maternity-leave-linked-to-lower-attrition-in-u-s-military
8. Fortune, “Susan Wojcicki knew what working parents need, Elon Musk does not,” Aug. 12, 2024. https://fortune.com/2024/08/12/susan-wojcicki-knew-what-working-parents-need-elon-musk-does-not (citing Wojcicki’s own 2014 CBS/Fortune commentary on Google’s 2007 leave expansion)
9. National Women’s Law Center, “July Jobs Report: Women Accounted for 100 Percent of the Labor Force Decline,” Aug. 7, 2026. https://nwlc.org/press-release/july-jobs-report-women-accounted-for-100-percent-of-the-labor-force-decline/
10. Yahoo News, “Who is Karoline Leavitt? Gen Z Trump megafan turned youngest-ever White House press secretary.” https://www.yahoo.com/news/karoline-leavitt-gen-z-trump-