Is full-time WFH working out?

Ask a roomful of executives in 2026 whether letting people work from home paid off, and you'll get answers that flatly contradict each other, mostly because they're not picturing the same thing. Some are thinking of an employee logging in from the kitchen table two days a week. Others are thinking of someone they haven't laid eyes on since 2022. The most rigorous research available says both groups have a point. Hybrid work is holding up fine. Full-time remote, the version where nobody ever comes in, is where things get messy.

Hybrid and fully remote are not the same experiment

Most of the mandate debate lumps every arrangement short of five office days a week into one bucket labelled "not really working." The research says that bucket is doing a lot of unearned work.

Stanford economist Nicholas Bloom ran the largest controlled study yet of the question, about as close to a randomised drug trial as workplace research gets, of more than 1,600 employees at the travel company Trip.com, published in the journal Nature. Staff who worked from home two days a week were just as productive and just as likely to be promoted as their fully office-based peers, and resignations among the hybrid group fell by 33%.

"The results are clear: hybrid work is a win-win-win for employee productivity, performance, and retention," Bloom said of the findings, which Trip.com credited with saving the company millions of dollars in reduced turnover.

But Bloom is explicit that fully remote work is a different animal entirely. "There definitely are studies, there's one in Nature Human Behavior, there's a bunch of stuff showing that is damaging for innovation, that's damaging for mentoring, for connectivity," Bloom said in a Stanford Engineering interview, describing fully remote work specifically, not hybrid.

In his own randomised field experiment on remote call-centre workers, quit rates among the fully remote group fell by nearly half because people were happier at home, but so did their promotion rates. "The one big sting in the tail is their promotion rates were almost half of the ones at the office," Bloom told the American Enterprise Institute. "So it's a bit of a mixed blessing."

A separate body of research points to wellbeing costs too, with fully remote employees more likely to report feelings of isolation than colleagues who have some in-person contact. Two very different report cards, for what looks from the outside like the same policy.

Employers are responding with mandates, not nuance

That distinction is exactly what's missing from the current wave of five-day mandates, which mostly treat "not fully in office" as one problem to solve, the way a doctor might treat a headache and a broken leg with the same aspirin. Amazon recalled roughly 350,000 office workers to a five-day week, Truist moved to full-time office attendance in January, and Instagram followed with its own five-day policy in February, as part of the return-to-office mandate wave sweeping corporate America. KPMG research cited in that same coverage found that 83% of chief business leaders now expect a full return to the office across their organisations by 2027, an aggressive stance for a policy the data only half-supports.

That blanket approach is already costing them people. A TopResume survey found that a quarter of managers had lost at least one team member in the previous six months specifically because of return-to-office requirements, with another 23.8% reporting losses over the prior year, and more than a third of employers, 37.3%, said RTO mandates were fuelling a broader talent and hiring crisis. In Canada, an Angus Reid survey found that 32% of remote workers would consider quitting if ordered back to the office most of the time, and 27% said they'd act on it quickly. More than half of those told to return described themselves as "very upset" or "upset", part of a retention battle brewing across Canadian workplaces.

"Return-to-work mandates may achieve short-term compliance, but they are unlikely to deliver sustainable engagement or performance," Jayne Clow, head of people and culture ANZ at NTT Data, told HRD earlier this year. "From a people leadership perspective, mandates can undermine trust and increase attrition risk in competitive talent markets," she said, part of a broader argument that flexibility is becoming the baseline employees expect, not a perk employers can withdraw at will. Conviction, in other words, is not the same thing as evidence.

The training pipeline is the part nobody priced in

That mentoring gap picked a spectacularly bad time to open up. Recent graduates are facing one of the toughest entry-level job markets in years, with unemployment among new graduates reaching 5.6% in early 2026, up 1.6 percentage points from three years earlier, according to a Stanford Institute for Economic Policy Research brief that points to AI absorbing much of the routine research, analysis and writing work once handed to junior staff. The World Economic Forum has estimated that entry-level postings in the US are down 35% since 2023, with similar contractions reported across other advanced economies.

Fewer of those routine tasks means fewer chances for a junior employee to be corrected, mentored and gradually trusted with more, which puts far more weight on whatever in-person time remains. HR teams aren't watching these employees quit. They're watching something quieter and, in its way, more worrying: Kahoot! growth director Ahteram Uddin has described a "neutral middle" emerging among early-career employees, workers who show up, do what's asked, but remain emotionally checked out.

"Having the right mentor, a mentor who can actually coach them, someone who can give them the hand-holding that they need, even if it's remotely, that is a very important part of the learning, onboarding, as well as growth success," Uddin said, noting that hybrid and remote set-ups can still deliver that support if organisations design for it deliberately.

So, is it working out?

Not as cleanly as 2021's remote-work evangelists promised, for most companies. Hybrid, done well, looks like a genuine win on productivity, performance and retention. Fully remote trades some retention and happiness for real costs to promotion, innovation and mentoring, costs that are harder to wave away now than they were in 2021, given the mentoring gap AI has opened up for junior staff.

None of that means fully remote can't work. Bloom points to payroll, benefits and IT support as functions where it runs smoothly. But for roles built on coordination, judgement-sharing and on-the-job learning, the evidence suggests the real trade-off was never hybrid versus office. It's hybrid versus fully remote. And right now, only one of them is winning.