Does a Four-Day Week Reduce Stress?

A major international study found less burnout, better sleep and improved wellbeing when employees worked fewer hours for the same pay. But the evidence raises questions about what produces the benefits — and who gets to enjoy the extra time.

TL;DR

• Burnout fell substantially in the largest trial to date. Across 141 organisations and 2,896 employees it dropped from 2.83 to 2.38 on a five-point scale, an effect size of 0.55. Control companies did not move.
• The physical health gains were much smaller. Physical health improved with an effect size of 0.29, job satisfaction 0.27 — modest by any standard.
• Nobody has run a large randomised trial of a four-day week. A Swedish study from 2017 provides randomised evidence on reduced hours, with real but small benefits over 18 months.
• The trial cannot tell you what produced the improvements. Fewer hours, eight weeks of work reorganisation, expectation effects and self-selected employers are tangled together by design.
• Routinely recorded sickness absence did not shift. Across 27 months at a UK council it showed no significant trial effect, while the council's own staff surveys reported gains.
• The extra day does not land equally. A 2026 review found women often using the additional hours for unpaid care and housework, with men's contribution rising only slightly.

The 2025 international trial

The flagship study was published in Nature Human Behaviour in July 2025 by Wen Fan, Juliet Schor and colleagues. It followed 2,896 employees across 141 organisations in Australia, Canada, Ireland, New Zealand, the UK and the US through a six-month four-day week with no cut in pay, preceded by eight weeks in which each organisation reorganised how it worked. Twelve companies that considered the trial but did not run it served as controls.

The hours really did fall. Average company-level working time dropped from 39.2 to 34.0 hours a week, individual hours from 39.0 to 34.3, and both were still down at twelve months. Burnout fell from 2.83 to 2.38 on a five-point scale, an effect size of 0.55 and the largest result in the study. Mental health rose from 2.93 to 3.32 (effect size 0.39), physical health from 3.01 to 3.29 (0.29) and job satisfaction from 7.07 to 7.59 out of ten (0.27).

Sleep problems and fatigue both fell, and self-rated work ability rose from 7.04 to 7.83. The control companies moved on none of it: their burnout went from 2.90 to 2.94, which was not statistically significant, and job satisfaction slipped slightly. That contrast between trial and control is the strongest thing in the study.

What the randomised evidence adds

There has never been a large randomised trial of a four-day week, and the Fan team say so themselves, calling for government-sponsored randomised studies. What exists instead is randomised evidence on reduced hours in general, and it is Swedish and nearly a decade old. Schiller and colleagues, in the Scandinavian Journal of Work, Environment & Health in 2017, randomised 33 public-sector workplaces covering 580 full-time employees to normal hours or a cut to 75% of full-time on unchanged pay, then followed them with sleep and stress diaries for 18 months.

It worked, and it worked modestly. The reduced-hours group slept about 23 minutes longer, reported better sleep quality, less sleepiness, lower perceived stress and fewer worries at bedtime, and the effects held on days off and persisted across the full 18 months rather than fading. But the effect sizes were small, with Cohen's f² below 0.08 throughout. Note the dose: a 25% cut, against roughly 13% in the 2025 trial, and still small effects. It is a test of working fewer hours rather than of any particular four-day-week model, so it supports the principle without validating the arrangements employers now run.

Hours, or the way the work was reorganised?

Here is the live question in this literature. In the 2025 trial, employees who personally lost the most hours improved the most: burnout coefficients of −0.271 for those cutting eight hours or more, −0.203 for five to seven, −0.098 for one to four. That looks like a dose-response, which is normally good evidence that the thing you changed is the thing doing the work.

At company level the gradient disappears. Firms cutting one to four hours saw burnout coefficients of −0.444 and those cutting eight or more saw −0.521, which the paper reports as statistically comparable. It is tempting to read that as proof that something other than the hours produced most of the improvement, but that goes further than the data allow: company-level comparisons involve far fewer units and much less variation in exposure, so they are less sensitive, and the study was never designed to separate the preparation from the shorter week.

What can be said is narrower. Participating workers reported substantial improvements relative to the comparison group. The trial does not establish how much came from fewer hours, how much from cutting meetings and low-value work during the reorganisation, how much from expectation, and how much from the fact that the employers who volunteered were the kind who volunteer. The authors are candid about the last, stating their effects are probably overestimated.

The objective outcomes from a UK council

South Cambridgeshire District Council ran the longest four-day week in the UK public sector, and an independent 27-month evaluation by Joshua Pink, Daiga Kamerāde and Brendan Burchell, covering January 2023 to March 2025, assessed 24 performance indicators. Adjusting for the COVID period, nine improved, among them calls answered, complaints handled on time and emergency repairs. Three got worse: housing rent collected, tenant satisfaction with repairs, and time to re-let housing stock. Twelve showed no significant difference, and without the COVID adjustment ten improved and none got worse.

Two findings matter most. Turnover was about one percentage point lower. And routinely recorded sickness absence showed no statistically significant trial effect across the whole 27 months — worth sitting alongside the council's own staff wellbeing surveys, which reported improvements against a working-population benchmark. An administrative measure and a self-reported one, pointing different ways. The evaluators state their analysis cannot establish the trial caused any of the changes, and they have separately co-authored work on the same council with the 4 Day Week Foundation and the Autonomy Institute. In July 2025 the council voted 26 to nine to make the 32-hour week permanent.

A 2026 American evaluation points somewhere similar. Over 18 months at a Colorado housing authority, participants in a voluntary reduced-schedule scheme reported significantly lower work overload than non-participants, while the differences in work stress itself, and in work spilling into non-work life, did not reach statistical significance. Participation there was voluntary too, so those who opted in may not resemble those who did not.

Who gets the extra day

A shorter working week does not give everyone the same kind of free time. A 2026 scoping review in the Scandinavian Journal of Work, Environment & Health by Mireia Utzet and colleagues mapped working-time reductions with pay maintained from April 2014 to May 2025: ten research papers and five grey-literature reports, covering cuts of 10% to 25%. Of the studies included, all reported positive effects on work-life balance and 81.8% on mental health — counts of studies reporting a benefit rather than effect sizes. Eleven took a gender perspective, and several found women using the additional hours for unpaid care and housework while men's contribution rose only slightly, which the reviewers warn could reinforce existing inequalities in domestic responsibilities.

The picture is not uniformly negative for people with caring responsibilities. A University of Cambridge study published in May 2025 interviewed 73 employees and line managers at South Cambridgeshire District Council who had disabilities, medical conditions or informal caring responsibilities, with fieldwork from October 2024 to January 2025. Participants were overwhelmingly positive about the four-day week and its effect on their wellbeing. That research was qualitative and involved people already inside the scheme, so it cannot quantify an average benefit, establish causation, or show how widely those experiences would be shared.

Both things can be true. For some workers the extra day creates room for recovery, appointments or caring that previously competed with paid employment; for others unpaid work expands to fill it. Neither study establishes that a day absorbed by domestic work cancels out the wellbeing gains — how time is used is related to, but not the same as, whether wellbeing improves. The variable that matters may be less the number of hours returned than how much say you have over them.

What you can use without a four-day week

Most people reading this do not get to decide their organisation's working hours. The 2025 study's mediation analysis found that fatigue, sleep problems and self-rated work ability sat on the path between fewer hours and lower burnout, which is a description of how the change travelled through the people in that trial rather than a demonstration that targeting those three factors on their own would reproduce the result.

With that caveat, the things the trial companies did in their eight weeks of preparation are reasonable to ask for in their own right: fewer and shorter meetings, dropping low-value work rather than redistributing it, protected blocks of focused time, clearer decisions about what does not get done. Practical possibilities, not proven substitutes. What to avoid is a compressed week dressed up as a short one — if the same workload is squeezed into four longer days then hours have not fallen, and that is not what any of these studies tested.

What This Actually Means

A four-day week is a promising organisational approach to stress, and the evidence is encouraging on the outcome people care about most: participating workers reported much less burnout, and the comparison companies did not. It is not proof that a shorter week will reduce stress for every worker or every workplace. The largest study was not randomised, not preregistered, relied entirely on self-report, ran in employers who volunteered, and its authors think their own numbers are inflated.

The rest is thinner still. The randomised evidence is small, Swedish, from 2017, and a test of shorter hours rather than any current four-day-week model. Routinely recorded sickness absence at a UK council did not shift across 27 months, even as the same council's staff surveys reported gains. And the benefit depends partly on what happens to the day you get back. So the sensible conclusion is about mechanisms rather than schedules: lower demands and more say over your work are promising targets for reducing stress, and a four-day week is one way of getting there.

Frequently Asked Questions

Does a four-day week reduce stress?

Participating workers report substantially less burnout, and that was not seen in comparison companies. What the research cannot yet establish is how much of that comes from the hours themselves rather than from reorganising the work, expectation, or the kind of employer that volunteers for a trial.

Is a compressed four-day week the same thing?

No, and the evidence does not cover it. Every study here cut actual weekly hours while holding pay constant — about 13% in the 2025 trial, 25% in the Swedish one. The same hours across four longer days is a different arrangement and has not been tested this way.

My employer will never consider it. Is any of this useful?

Yes. The trial's own analysis points to fatigue, sleep and work ability as the channels through which the change worked, and those respond to lower demands and greater control. Fewer meetings, dropping low-value work rather than moving it, and protected focus time are worth pushing for in their own right.

Does it improve physical health?

Modestly, on self-report: an effect size of 0.29 in the 2025 trial, which is small. Routinely recorded sickness absence at the UK council showed no significant change over 27 months. The clearer gains were in burnout, fatigue and sleep.

The Bottom Line

A four-day week is a promising organisational intervention with real holes in its evidence. Burnout fell sharply among the people who did it, and nothing comparable happened in the control companies. But the study was not randomised, not preregistered and ran in self-selected employers whose own researchers say the effects are probably overstated, and the one objective measure anybody tracked did not move. The honest summary is that reducing what work demands of you, and increasing your say over how it gets done, looks like a promising way to lower stress — and a shorter week is one way to arrange that, not a guarantee on its own.

This article is for general information and is not medical advice. If work-related stress is affecting your health, speak to your GP.

Sources

• Fan W, Schor JB, Kelly O, Gu G. Work time reduction via a 4-day workweek finds improvements in workers' well-being. Nature Human Behaviour, 2025. https://pubmed.ncbi.nlm.nih.gov/40691306/
• Fan W, et al. Accepted manuscript, NSF Public Access Repository. https://par.nsf.gov/servlets/purl/10636509
• Schiller H, et al. The impact of reduced worktime on sleep and perceived stress: a group randomized intervention study using diary data. Scandinavian Journal of Work, Environment & Health, 2017. https://pubmed.ncbi.nlm.nih.gov/27942734/
• Utzet M, et al. Exploring the health and well-being benefits of reduced working hours with maintained salary: A scoping review and evidence map. Scandinavian Journal of Work, Environment & Health, 2026. https://pubmed.ncbi.nlm.nih.gov/41582927/
• Pink J, Kamerāde D, Burchell B. South Cambridgeshire District Council Four-Day Week Trial: A 27 Month Key Performance Indicator Evaluation, 2025. https://scambs.moderngov.co.uk/documents/s139784/Appendix%20B%20-%20South%20Cambridgeshire%20District%20Council%20Four-Day%20Week%20Trial%20-%20A%2027%20Month%20Key%20Performance%20.pdf
• Valley MA, et al. Evaluation of a Reduced Work Schedule and Housing Authority Employee Wellbeing. International Journal of Environmental Research and Public Health, 2026. https://pubmed.ncbi.nlm.nih.gov/42652341/
• University of Cambridge. The Impact of Working Time Reduction on Disabled People, People with a Medical Condition and Informal Carers: Evidence from the South Cambridgeshire District Council Four Day Week Trial, May 2025. https://scambs.moderngov.co.uk/documents/s139844/Appendix%20G%20-%20University%20of%20Cambridge%20impact%20of%20reduced%20working%20time%20on%20disabled%20people%20those%20with%20me.pdf
• Robertson Cooper. South Cambridgeshire District Council's Health and Wellbeing 2025. https://scambs.moderngov.co.uk/documents/s139843/Appendix%20F%20-%20Robertson%20Cooper%20report%20on%20South%20Cambridgeshire%20District%20Councils%20Health%20and%20Wellbeing.pdf
• South Cambridgeshire District Council. Four-day Week Update and Future Arrangements. Decision of 17 July 2025, with minutes and supporting documents. https://scambs.moderngov.co.uk/ieDecisionDetails.aspx?ID=12804
• South Cambridgeshire District Council. Four-Day Week Equality Impact Assessment, July 2025. https://www.scambs.gov.uk/media/hpuifw4t/4dw-eqia-july-2025.pdf