Practice ・ 7 min
Do longer working hours mean more output? WWI factory data
In British World War I munitions factories, output rose with hours up to about 49 a week, then slowed; 70 hours produced little more than 56. The data are observational.

The evidence behind this article
- Research synthesis1
- Observational1
- Estimate1
Includes an analysis that pools many studies, which is less swayed by chance in any single one.
3 sources・Countries studied: United Kingdom, Japan, Several countries
How to read the kinds of evidence →What to do with this
Set a time and place for tomorrow's most important task, and at the end of the day note one line on what moved forward rather than how long you worked.
Do longer hours mean more output? Data from British munitions factories during World War I suggest not beyond a point. Up to about 49 hours a week, output rose in step with hours. Beyond that, gains slowed, and a 70-hour week produced little more than a 56-hour week. These are century-old factory data, so they may not carry over directly to today’s jobs.
Key points
- A study of British munitions workers during World War I, most of them women, found that below 49 hours a week, weekly output was proportional to hours. Above 49 hours, output rose more slowly, and output at 70 hours differed little from output at 56 hours.
- In the same data, weeks without a Sunday rest day had about 10% lower output than weeks with one, holding weekly hours constant. The data are observational, and the point where gains slowed differed across workers and types of work.
- A meta-analysis of about 600,000 people in Europe, the US and Australia found that people working 55 hours or more a week had a 1.33 times higher risk of stroke and a 1.13 times higher risk of coronary heart disease than people working 35 to 40 hours. No Japanese data were included.
What happened to factory output when hours went up
Economist John Pencavel analysed records collected by Britain’s Health of Munition Workers Committee during World War I. When the war began in 1914, the British government suspended regulations on working hours, and some people worked more than 100 hours a week.
The main data come from a large factory in the English Midlands in 1916. For about a year, they record weekly hours and output for groups of workers: about 100 women turning fuse bodies, 40 women milling screw threads, 56 men and a small group of youths. Weekly hours averaged around 51, ranging from 24 to 72.5.
Below 49 hours a week, output was roughly proportional to hours worked. Above 49 hours, each extra hour added less, and output peaked at about 63 hours. Output at 70 hours differed little from output at 56 hours.
Pencavel also looked at Sunday rest. Holding weekly hours constant, weeks in which people also worked on Sunday had about 10% lower output. His estimates imply that a 48-hour week with Sundays off produced slightly more than a 70-hour week with Sunday work, by 1.4% of the 70-hour output.
How long working hours are linked to health
Overwork can also show up in health. Kivimäki and colleagues combined 25 studies from 24 cohorts in Europe, the US and Australia. The analysis of coronary heart disease covered 603,838 people, and the analysis of stroke covered 528,908, with average follow-up of 8.5 and 7.2 years.
After adjusting for age, sex and socioeconomic status, people working 55 hours or more a week had a 1.33 times higher risk of stroke and a 1.13 times higher risk of coronary heart disease than people working a standard 35 to 40 hours. For stroke, risk rose with hours: 1.27 times at 49 to 54 hours, and 1.10 times at 41 to 48 hours, which was not statistically clear. The authors note that the link with coronary heart disease is weaker than the link with stroke.
These are also observational studies. Nobody was assigned to long or short hours by lottery. The figures describe average tendencies across large groups, not any one person’s risk. If you are concerned about your health, talk to a doctor.
How to read research on long working hours, including for Japan
The factory study is observational too: hours were not set as an experiment. The point where gains slowed also differed across workers and tasks. For 43 individual women workers, output stayed roughly proportional to hours at every level observed. These were factory jobs, so we do not know whether thinking work or work with people follows the same pattern. The study did not look at pay.
In Japan, among employees who work 40 hours or more a week, the share working 60 hours or more was 8.0% in 2024, down 0.4 points from the year before and on a downward trend, according to the government’s white paper on karoshi (death from overwork). We did not find a study that directly measures the link between hours and output for workers in Japan. Korean research was outside the scope of this review.
A small step for today
The research points toward protecting time for your most important work, not stretching the day.
- Use “Set when and where to do tomorrow’s key task” to reserve the hours when your mind is sharpest.
- Use “Wrote one line on today’s progress” to track what moved forward, not how long you stayed.
- Aim for “Slept about 7 hours,” and keep one day a week for real rest.
FAQ
Does working longer hours increase output?
Only up to a point, according to one well-known study. In British munitions factories during World War I, output gains slowed above about 49 hours a week, and 70 hours produced little more than 56. These are century-old observational data, and the turning point varied across workers and tasks.
How many working hours a week affect health?
In a meta-analysis of studies from Europe, the US and Australia, people working 55 hours or more a week had a 1.33 times higher risk of stroke than people working 35 to 40 hours. This is an average tendency, so ask a doctor about your own health.
What happens to output without a day off?
In the British factory data, weeks that included Sunday work had about 10% lower output than weeks with the same total hours and Sunday off. A weekly rest day went with more output from the same hours.
Sources
- [X-pencavel-2015] Pencavel (2015) The Productivity of Working Hours. The Economic Journal.
- [X-kivimaki-2015] Kivimäki et al. (2015) Long working hours and risk of coronary heart disease and stroke: a systematic review and meta-analysis of published and unpublished data for 603 838 individuals. The Lancet.
- [X-mhlw-2025] Ministry of Health, Labour and Welfare, Japan (2025) White Paper on Measures to Prevent Karoshi, 2025 edition.
Studies describe average tendencies. This article is not a promise that a habit will lift your income, and it is not medical or investment advice. About these articles ・ Search & sources
Updated:
← All articles