Rethinking remote work: surveys show people will give up 5 to 8% of their pay to work from home, but a 2026 study links remote work to more isolation, anxiety and depression — suggesting they may be underweighting the cost

Here is something I have been turning over for a while. Most of us, given the choice, will give up real money to work from home — not metaphorical money but actual salary, the kind that pays rent.

In a survey of more than 30,000 Americans, over half valued the option to work from home two or three days a week at a pay raise of 5 percent or more, and nearly one in five put it at 15 percent or more. Stanford economist Nicholas Bloom also points to survey evidence that hybrid work “is worth about an 8 percent increase in salary” People behave as though working from home is worth a raise.

It seems we will trade money for it; the question is what, exactly, we are buying. A study published this month complicates the answer.

A quick note before we go further. I am not a doctor or a psychologist, and this is one person reading the research, not advice for your situation. The studies here are findings from particular groups of workers, not settled science or rules that apply to everyone.

What the new study actually found

“Home alone: Remote work, isolation, and mental health” was written by researchers Natalia Emanuel of the Federal Reserve Bank of New York, Emma Harrington of the University of Virginia, and Amanda Pallais of Harvard. They drew on five nationally representative US surveys covering nearly 568,000 workers, then compared people in jobs that became much more remote after the pandemic with people in jobs that still required showing up in person.

Workers in newly-remote occupations spent, on average, an additional 1.2 hours working alone each workday. Their mental health got worse. The authors estimate remote work accounts for about a third of the post-pandemic rise in worker mental distress.

The harm landed hardest on people who live alone, which makes a grim kind of sense: if the office was where you saw other humans, and you live by yourself, then losing the office means losing most of your day’s company.

Why the two findings can both be true

People are choosing the thing the study links to harm, and willing to pay for the privilege. So are they just wrong?

Maybe not, or at least not in the simple sense. The benefits of working from home are immediate and obvious: no commute, lunch in your own kitchen, the laundry done by the time you log off. You feel them on day one. The costs the study describes are slow and diffuse — an extra hour alone today does not announce itself, but it accumulates over months into something you might never trace back to your work setup.

That is a hard trade to price correctly. Present-bias is the usual label: we weight what we can feel now heavily, and what creeps up on us lightly. I suspect that is most of what is going on. The 8 percent we will forgo is the part we can see; the isolation is the part we cannot.

The case for caution about the caution

The study is not the final word, and the authors say so. It is a clever quasi-experimental design, not a randomized trial, so it tells us about a strong pattern rather than proving cause beyond doubt. The authors are also candid about the timeline. Given that the data end in 2024, they write, they “cannot fully capture long-term adaptations among remotable workers”. People may be learning to build the social life that the office used to hand them by default. That adaptation, if it is happening, would not show up yet.

Emma Zhang and Yale sociologist Rourke O’Brien, writing a companion Perspective in Science, note that the results suggest the shift to home work “carries measurable costs at the population level”. Population level is the key phrase: it is a statement about the average across hundreds of thousands of people, not a prediction about you specifically.

The fix, in the study authors’ own framing, is not to march everyone back to a desk. Emanuel, Harrington and Pallais write that “across a range of remote work arrangements, both individuals and organizations may want to prioritize making remote work less isolating”. That reads to me as the sensible middle. The arrangement most of us are paying for is not the problem so much as the loneliness it can quietly carry along with it.

If any of this is hitting closer to home than it is interesting, and the isolation feels less like a research finding and more like your own week, talking to a counsellor or therapist is worth more than any article.