Return-to-Office Mandates Were Never About Productivity

A friend who runs analytics at a mid-size insurer told me her CEO announced a five-day return-to-office mandate two weeks after her team delivered a report showing remote output on every metric they tracked — tickets closed, code shipped, deals sourced — held flat or improved over the prior two years in office. The report went into the same meeting where the mandate was announced. Nobody referenced it.

That's not an isolated anecdote about one tone-deaf executive. It's the pattern the actual research keeps finding, and almost none of the RTO commentary is willing to follow the data to its actual conclusion: return-to-office mandates were never really being decided on productivity grounds. They just needed productivity as the thing to say out loud.

The predictor that actually explains who mandates RTO

If RTO were a productivity decision, you'd expect the strongest predictor of a mandate to be some measure of underperformance — falling output, missed targets, something in the numbers. A 2024–2026 study by researchers Yuye Ding and Mark Ma, examining return-to-office mandates across public companies, found something else entirely: the single strongest predictor of a firm issuing a top-down RTO mandate was having a powerful male CEO — specifically, one drawing a pay premium markedly higher than the company's next-highest-paid executive, a standard proxy researchers use for concentrated, unchecked authority at the top.

Not falling revenue. Not declining output. CEO power.

The same research found the consequences of these mandates split sharply along a line that has nothing to do with output either: female employee turnover after an RTO mandate ran nearly three times higher than male employee turnover, concentrated most heavily among senior, highly skilled women — precisely the group a genuine productivity crisis would least afford to lose. A policy sold as restoring performance was, by the data, disproportionately driving out the people whose performance the company could least afford to lose. That's not what a productivity fix looks like. That's what a status reassertion looks like, with an uneven bill attached.

Why "ideal worker" isn't a throwaway phrase

There's a specific sociological concept that explains why the fallout lands so unevenly, and it's worth naming instead of gesturing at: the "ideal worker" norm, the assumption baked into how most large organizations structure authority and visibility, that the model employee is someone whose job is their only major claim on their time — unencumbered by caregiving, available for the room, present because presence itself signals commitment. That norm was never gender-neutral in practice, even when it's gender-neutral on paper. It disproportionately describes men, and it disproportionately excludes women who still carry a larger average share of household and childcare labor, mandate or no mandate.

An RTO policy re-anchors "commitment" to physical presence — the one axis on which the ideal-worker norm has always been easiest to enforce and easiest to see. It's a much cruder, more visible metric than "did you deliver," which is exactly what makes it attractive to a leader who wants legible control rather than genuinely difficult measurement. You can see who's at their desk at 9am. You cannot, without real work, see who's actually doing the job well from a kitchen table forty miles away. Mandating the visible thing is administratively easier than measuring the real thing — and it has the added, probably not incidental, effect of reasserting exactly who gets to set the terms.

This isn't a story about remote work being universally better — the research on collaboration costs from fully distributed teams is genuinely mixed, and some in-person concentration correlates with more cross-team idea exchange in specific creative and R&D functions. That's a real, separate debate. It's just not the debate RTO mandates are actually settling, because the mandate-issuing pattern doesn't track the companies where that collaboration cost is highest. It tracks CEO pay-gap concentration.

The tell is which employees get exceptions

So actually, here's the reframe that makes the power explanation impossible to unsee once you notice it: watch which employees quietly get RTO exceptions. It is very rarely the most junior people, and it is very often whoever the CEO or his direct reports most want in the building — a detail that a genuine, uniformly-applied productivity standard would have no reason to produce. A policy justified by universal claims about collaboration and output, applied selectively based on organizational proximity to power, was never really about the universal claim. The exception list is the confession the press release won't give you.

None of this means every leader issuing an RTO mandate is consciously staging a power play. Most probably believe the productivity story they're telling, the way most people believe the stated reason for a decision that was actually made for reasons closer to identity and control. That's what makes the pattern durable — it doesn't require anyone to be lying. It only requires a stated justification that's easier to say out loud than the real one, and a workforce too polite, or too exhausted, to ask why the mandate arrived the same week the report proving it unnecessary landed on the same desk.

This is the same gap as every other leadership metric that surveys well and means nothing

I've written before about how psychological safety initiatives function as theater — a genuine-sounding value, implemented in a form that costs leadership nothing and changes nothing structural. RTO mandates run the identical playbook from the opposite direction: instead of a costless initiative standing in for a real commitment, it's a costly mandate standing in for a real conversation about who actually holds authority to decide how work gets done. Both moves share the same tell — the stated value (safety, productivity) never has to survive contact with the data, because nobody with the power to reverse the decision is the one paying its cost.

That symmetry matters because it means the fix for RTO overreach won't come from better productivity studies. Leaders issuing power-driven mandates aren't short on data — Ding and Ma's study existed, the insurer's internal report existed, and neither survived contact with a decision that was never actually about the numbers. The fix has to target the actual mechanism: making the cost of a status-driven mandate visible and attributable to the person who issued it, the same way visible attrition numbers, broken out by gender and seniority, are starting to do at companies where boards have finally started asking who's actually leaving and why.

The next time your organization announces a return-to-office policy citing "collaboration" or "culture," the useful question isn't whether those values matter — they might. It's who drew up the exception list, and whether the person who wrote the memo is on it.