Many firms announced hard return-to-office policies to reclaim a sense of control. The market answered with attrition risk, disengagement, and quiet noncompliance. The best current data shows most executives are not doubling down on mandates, and where policies are rigid, senior people are the first to vote with their feet.
The mandate wave that wasn’t
The picture of a sweeping return-to-office movement is overstated. In a Stanford and Atlanta Fed survey of more than 1,000 US executives, only 12 percent of leaders with hybrid or fully remote workforces said they planned a mandate in the next 12 months (Bloom, Davis and Barrero, 2025). Even if those plans were fully executed, the share of paid days worked from home in the United States would fall by less than half a percentage point, from 21.2 to 20.8 percent (Bloom, Davis and Barrero, 2025). Workers’ stated willingness to comply is fragile. In the December 2024 wave of the Survey of Working Arrangements and Attitudes, only 44 percent of US workers said they would obey a fully in-person policy. The rest said they would resign or look for another job (Bloom, Davis and Barrero, 2025).
This is a power story, not a facilities story. Managers who equate presence with authority are discovering that labor markets, especially for experienced talent, now treat location demands as a negotiation variable. The result is selective noncompliance and, when forced, exit.
When presence is policed, who leaves
A working paper that tracks 260 million résumés across technology firms offers a sharper view of who departs after mandates. Studying Microsoft, Apple, and SpaceX, Ding and Ma (2024) use distributional synthetic controls to estimate causal effects of return-to-office policies. Their core finding is not subtle. Mandates triggered a disproportionate loss of senior employees, a pattern that threatens productivity and innovation. At Microsoft, which moved to a 50 percent in-office policy in April 2022, average tenure in the upper deciles fell by about two months, and the share of employees with 0 to 3 years at the firm rose by 6 percent, indicating a relative shift toward less-experienced staff (Ding and Ma, 2024). The effects scale with rigidity. Apple’s one-day-per-week policy was associated with a smaller seniority shift, under 4 percent, Microsoft’s 50 percent presence showed a larger one, and SpaceX’s full-time presence policy produced the strongest displacement of seniority (Ding and Ma, 2024). The study is a working paper rather than a peer-reviewed article, but the design is careful and the pattern consistent across firms.
The implication is straightforward. Presence-as-power trades institutional memory for badge counts. Rigid mandates do not just raise attrition risk in the abstract. They tilt a company’s experience curve toward the shallow end.
Hierarchy, status, and the office’s invisible script
Why does proximity get treated as power in the first place? Because traditional hierarchy manages by supervision and status. A broad review by Lee and Edmondson (2017) defines self-managing organizations as those that radically decentralize authority and eliminate the manager–subordinate relationship as a formal feature of the design. Their synthesis argues that conventional hierarchy is well suited to reliable execution in stable environments, yet it tends toward rigidity, reinforces status differences, and inhibits cross-boundary problem solving and development. Offices, organized around lines of sight and layers of approval, encode that script.
A different mechanism of power is available: credibility earned through role clarity and learning. Edmondson (1999) showed in a field study of 51 work teams that psychological safety, the shared belief that a team is safe for interpersonal risk taking, is associated with team learning behavior. Learning behavior then mediates the link to performance. The study is cross-sectional and focuses on one company, so it is not a grand causal law. It is, however, the seminal construct in a literature that has repeatedly shown why people speak up, share concerns, and surface errors. If presence is used to reassert status, it taxes that learning channel. If presence is used to lower interpersonal risk, it raises the odds that proximity pays off.
The engagement backdrop heightens the stakes. Global employee engagement fell to 20 percent in 2025, the lowest since 2020, with Gallup estimating roughly 10 trillion dollars in lost productivity, about 9 percent of global GDP (Gallup, 2025). Manager engagement is sliding as well, down nine points since 2022, with the sharpest annual drop between 2024 and 2025, from 27 to 22 percent (Gallup, 2025). The spread between typical and best-practice organizations is stark. In top performers, 79 percent of managers are engaged compared with a global average of 22 percent, a gap that Gallup highlights as evidence that engagement reflects organizational design, not fate (Gallup, 2025). A presence policy that reactivates status games rather than clarifying roles is fuel on a cold fire.
Power that travels
The office still matters. What must travel is the basis of power. Three moves follow from the evidence and its limits.
-
Make presence a resource, not a requirement. The executive survey data suggest limited appetite for new mandates and limited compliance when they appear (Bloom, Davis and Barrero, 2025). Treat the office as a convening tool for complex coordination and development, not as a universal attendance policy. If you do mandate, pressure-test who you might lose. Ding and Ma (2024) show the risk is concentrated among senior contributors.
-
Replace status with explicit roles. If hierarchy’s strengths are reliability and speed of routine execution, be explicit about where you want that and where you want experimentation. Lee and Edmondson (2017) describe a path where authority is distributed by design rather than informally hoarded by proximity. Even without adopting a full self-managing model, teams can clarify accountabilities and decision rights so that face time does not masquerade as authority.
-
Protect the learning loop. Use proximity to raise psychological safety rather than to monitor. Ask teams to surface what they will try next, what failed last sprint, and what help they need. That is the channel that Edmondson (1999) links to performance through learning behavior. If people fear the meeting more than the market, the office is working against you.
A final, practical read. The costs of disengagement are already visible in Gallup’s estimate of global productivity losses. The costs of mandating proximity at all costs are visible in the seniority shifts that follow hard policies. Power anchored in surveillance is brittle. Power anchored in clear roles and an environment where people can learn aloud is portable. In this cycle, it is also the safer bet.
Sources
- Bloom, Davis and Barrero (2025). Stanford Report — Return to office? Not everybody is doing it. Executive surveys show few new mandates planned and limited impact on national WFH rates, with workers reporting low willingness to comply with fully in-person policies. https://news.stanford.edu/stories/2025/03/return-to-office-not-everybody-is-doing-it
- Ding and Ma (2024). Return-to-Office Mandates and Brain Drain (Microsoft, SpaceX, Apple). A working paper using distributional synthetic controls finds rigid RTO policies disproportionately reduce seniority, with dose-response effects by mandate strictness. https://arxiv.org/pdf/2405.04352
- Lee and Edmondson (2017). Self-managing organizations: Exploring the limits of less-hierarchical organizing. A peer-reviewed synthesis arguing that conventional hierarchy trades flexibility and development for reliability, while self-managing designs redistribute authority by design. https://www.sciencedirect.com/science/article/abs/pii/S0191308517300059
- Edmondson (1999). Psychological Safety and Learning Behavior in Work Teams (Administrative Science Quarterly 44). A seminal field study linking team psychological safety to learning behaviors that mediate performance, clarifying why environments safe for voice matter. https://journals.sagepub.com/doi/10.2307/2666999
- Gallup (2025). State of the Global Workplace 2025. Global engagement at 20 percent, an estimated 10 trillion dollars in lost productivity, and a nine-point manager engagement decline since 2022, with best-practice organizations far outperforming the average. https://www.gallup.com/workplace/659610/state-global-workplace.aspx