A chief executive announces that everyone is back five days a week, and the reasons offered are always about performance: collaboration, energy, the accidental hallway conversation. What the reasons rarely include is a number. That absence is the tell. When Stanford, the Atlanta Fed, and their collaborators surveyed more than a thousand US executives in early 2025, only 12% of those with hybrid or remote staff planned any return-to-office push in the following year. The people closest to the productivity question were, overwhelmingly, not treating the office as the answer to it.
The same survey found employers keeping hybrid arrangements precisely because they liked the results: less physical space, better productivity, lower turnover. Most executives said they would not change course even in a recession or with unemployment doubling. If presence were the productivity lever it is claimed to be, the firms with the data would be pulling it. Instead the mandate tends to arrive from the top of organizations that already run distributed and profitable, which suggests the policy is answering a different question than the one it names.
What the mandate actually measures
A schedule is easy to audit. Output is hard. For a manager who cannot easily say whether a given week's work was good, a badge swipe is a comforting substitute: it converts an ambiguous judgment about value into a clean binary about attendance. The mandate restores that binary. It does not measure whether the work got done; it measures whether the worker was where the work is supposed to happen. Those are not the same thing, and the gap between them is exactly what remote work exposed.
Ricardo Semler learned to distrust that binary the hard way. As he has told it, after he collapsed in a factory at twenty-five and a doctor called him the healthiest sick person he had ever seen, he began pulling apart the machinery of oversight: the org chart, the time clock, the six-day week. The company stopped counting hours and started asking a simpler question, whether the pump had been delivered. Absenteeism fell. The lesson was not that supervision is unnecessary but that counting presence and counting results are rival activities, and that the first one crowds out the second.
Fear is the cheapest tool
Control-by-attendance is attractive for a reason Semler named bluntly. When Semco made vacations compulsory, thirty days with the laptop locked away, he watched adults negotiate for the right to work more. He read that not as loyalty but as fear, which he called the cheapest management tool ever invented. A mandate runs on the same fuel. It does not have to improve anything to work as intended; it only has to make people visible, and visibility, under pressure, feels like leverage.
The trouble is that leverage of this kind is expensive in ways that do not show up on the attendance report. A study of return-to-office mandates at Microsoft, SpaceX, and Apple, using synthetic controls over a very large résumé dataset, found that the mandates drove disproportionate departures among senior employees, and that the effect scaled with the strictness of the rule. The people most able to leave, and most costly to replace, were the ones who left. A signal aimed at the workforce's discipline landed hardest on its experience.
The honest version of the policy
None of this means presence never matters. Some work is genuinely better in a room, and a leader who says so plainly, for specific tasks, is on solid ground. The problem is the blanket rule justified by a productivity claim the data does not support. If the real goal is control, the honest move is to say so and defend it, because control has costs a company can weigh: attrition, morale, the quiet withdrawal of discretionary effort. Dressed as a productivity policy, those costs go unpriced, and the mandate gets credit for benefits no one measured.
The firms that kept hybrid are running the experiment the mandate refuses to run. They are watching output, retention, and cost, and adjusting. That is not softness; it is measurement. The company that mandates attendance and then declines to check whether output moved has not made a productivity decision at all. It has made a statement about who is in charge, and it has chosen the one instrument that reveals nothing about whether the work is any good.
Sources
- Bloom, Davis & Barrero / Stanford (2025). Return to office? Not everybody is doing it. Only 12% of US executives with hybrid or remote staff planned an RTO push; employers keep hybrid for less space, better productivity, lower turnover. https://news.stanford.edu/stories/2025/03/return-to-office-not-everybody-is-doing-it
- Ding & Ma (2024). Return-to-Office Mandates and Brain Drain. RTO mandates at Microsoft, SpaceX and Apple drove disproportionate senior departures, scaling with mandate strictness. https://arxiv.org/pdf/2405.04352