By the middle of 2026, more than half of the Fortune 100 had their employees under a five-day in-office mandate — a number that sat at roughly one in ten a year earlier. That is not a drift. That is a coordinated repricing of how the largest companies in America believe work should look. And it arrived with press releases full of words like collaboration, culture, and innovation — and conspicuously empty of citations.
Here is the thing the announcements never say: there is no body of peer-reviewed evidence showing that five days of mandated presence produces more output than well-run hybrid work. The research that exists points, over and over, somewhere else entirely. The mandate is not a productivity strategy. It is a performance of control — and the science it refuses to cite tells you exactly who the performance is for.
What the research actually says about where work happens
The most rigorous work in this field belongs to Stanford economist Nicholas Bloom, who has been running controlled trials on remote and hybrid work since before it was fashionable. His 2024 randomized trial at Trip.com — published in Nature, the kind of venue mandate decks never quote — found that hybrid employees were just as productive as their five-day-office peers, received the same performance grades, and quit at a rate about one-third lower. His earlier call-center experiment found home-based workers measurably outperformed their in-office counterparts.
You can argue about how far those results generalize. What you cannot do is cite a comparable body of evidence pointing the other way — because it does not exist. Companies issuing mandates are not weighing competing studies. They are declining to weigh studies at all.
Productivity paranoia is a leadership problem, not a labor problem
Microsoft's Work Trend Index gave the real mechanism a name back in 2022: productivity paranoia. In its survey of global leaders, roughly 85% said hybrid work made it hard to feel confident their people were productive. In the same survey, the overwhelming majority of employees reported being productive. Both things were true. The workers were working. The leaders could not see it — and what a certain kind of leader cannot see, he does not believe exists.
The mandate resolves the leader's anxiety, not the company's output problem. It converts "I can't see you" into "you must be visible" — and calls the conversion culture.
Read the mandate memos with that lens and the language decodes itself. "Serendipitous collaboration" means I want to walk the floor. "Culture is built in person" means presence is the only instrument I trust. None of it is a claim about output, because output was never the variable being managed. Visibility was.
The mandate is a real-estate decision wearing a culture costume
Follow the balance sheet. The companies leading the mandate wave are, disproportionately, companies carrying long commercial leases and owned campuses that pandemic-era occupancy turned into visible dead weight. Office attendance in major American metros spent years stuck near half of pre-2020 levels — a fact any CFO staring at a ten-year lease experiences as a standing indictment. A full building does not make the lease cheaper. But it makes the line item look less like a mistake.
Add the civic pressure — downtown business districts, transit systems, and commercial landlords lobbying loudly for foot traffic — and you get a mandate wave that correlates with real-estate exposure far better than it correlates with any measure of productivity. The building is on the balance sheet. The badge swipe became the KPI.
What a mandate teaches your best people
Behavioral science has one more thing to say, and it is the expensive one. Decades of motivation research — from self-determination theory onward — converge on a simple point: autonomy is not a perk, it is a load-bearing input to sustained performance. People stay at hard things when effort and outcome feel connected, and when they are trusted to choose how the work gets done. An evidence-free mandate severs exactly that connection. It teaches your workforce that policy is set by anxiety, not by results — that what you produce matters less than where you sit while producing it.
And the people who absorb that lesson fastest are the ones with options. Researchers tracking résumé data after major tech RTO announcements found departures skewed senior — the experienced, hardest-to-replace tier left at elevated rates, taking institutional knowledge with them to more flexible competitors. Attrition from a mandate is not random churn. It is adverse selection, aimed at your top of the curve. The company keeps the people who had nowhere else to go and loses the ones it could least afford to lose — then reports the resulting headcount reduction as if it were discipline rather than damage.
Presence is not output
None of this is an argument that offices are useless. The honest research case for shared space is real and specific: onboarding, apprenticeship, the high-bandwidth early phase of creative work. A company that read the evidence would build policy around those moments — deliberate gatherings with a purpose, measured against outcomes. That is not what a five-day fiat is. A fiat is what you issue when you have confused presence with output and activity with work, and when no measurement exists that would change your mind.
That last part is the test, and it travels well beyond this topic. Ask the question of any policy handed down this year: what evidence would reverse it? If the answer is none — if attendance is the goal and attendance is the metric — then it was never a productivity strategy. It was theater, staged for the anxiety of the people who ordered it.
The chairs are full. The output is flat. Nobody is saying so in the press release.
