But there are reasons to believe this was no fluke. A similar experiment at a New Zealand company in 2018 produced similar results. And research by Stanford University economist John Pencavel on British munitions plant workers during World War 1 has found that past a certain point, working more decreases hourly output.
Unproductive long days
It’s not hard to think of some reasons why this would be the case. Long hours cause fatigue, both physical and mental. That fatigue affects not just the last few hours of a workday, but all the hours of the next day. An employee who drags back to work after only a few hours’ rest isn’t going to be very productive in the morning. Then the next long day tires them out even more, and the punishing cycle begins again. Eventually the worker starts making little errors, slowing down and failing to take initiative to fix problems and exploit new opportunities.
Another reason long hours lead to reduced productivity is that a worker can simply run out of things to do. Some tasks can be done at any time, but some have to wait on the schedules of others. A restaurant’s delivery truck can’t just decide to deliver more food in a day in the absence of orders. A salesperson has to wait on customers’ schedules. A lab researcher has to wait for experiments to run, and so on.
Workers can probably always find something useful to do with extra hours, but that usefulness diminishes with each additional hour. Eventually, they end up using Twitter or chatting with other unoccupied workers, waiting for the end of the day.
A third possible reason for unproductively long working days could have to do with social signalling. Since it’s impossible for managers to closely monitor their workers all day to make sure they’re on task, employees may try to show that they’re hard workers by staying late at the office. Someone who gets her work done early may stay at her desk until the boss goes home (or her coworkers go home) for fear of being singled out as a slacker.
The obvious solution is to tell employees to go home earlier. If workers are fatigued, this will give them more energy. If they’re trying to look busy, this will free them from the need to keep up appearances. And if they have more actual work than they can do at the office, they can always do some at home.
The obvious solution is to tell employees to go home earlier. If workers are fatigued, this will give them more energy. If they’re trying to look busy, this will free them from the need to keep up appearances.
Four-day work weeks like the one at Microsoft’s Japan office are one way of sending workers home early. Another is to make companies give workers lots of paid leave days. A third is to simply shorten the workweek, allowing companies to decide how to allocate hours across the week. Some countries use a combination of these approaches — France, for example, has a 35-hour workweek and 36 total days of paid annual leave. It’s not surprising, therefore, that France has one of the higher levels of per-hour productivity in the developed world.
Japan and the US are two countries that need this approach. Americans work longer hours than their counterparts in more productive North European countries, and have no federally mandated paid vacation days whatsoever.
Japan, meanwhile, has seen its official working hours fall below those of the US, but many Japanese companies continue to make their workers stay at the office for unpaid overtime off the books. The government of Prime Minister Shinzo Abe has tried to change the culture of overwork, but is having a hard time doing so.
A plan to pay workers for results rather than hours worked has been withdrawn after data showed it having little effect, while a measure to punish companies for requiring unpaid overtime hasn’t been passed yet.
For the US and Japan, as well as other countries where employees are working inefficiently long days, mandating shorter hours – through four-day work weeks, shorter work days, or more paid leave – could be just what they need to raise productivity while improving work-life balance.
Noah Smith is a Bloomberg Opinion columnist. He was an assistant professor of finance at Stony Brook University, and he blogs at Noahpinion.