The five-day week, the eight-hour day, the weekend, the evening that belongs to you rather than your employer: these feel like permanent features of the world, almost laws of nature. They are nothing of the kind. Every one of them was won, recently, against resistance, by people who organised and legislated to take time back. The history matters, because it tells you exactly how productivity gains become free time, and the answer is not by themselves.
Were shorter hours ever simply granted?
Almost never. The dominant pattern across two centuries is that working hours fell not because employers grew generous but because workers, unions, and eventually governments forced the change. The nineteenth-century eight-hour-day movement crystallised the demand in a slogan still worth repeating: eight hours for work, eight hours for rest, eight hours for what you will.1 It was fought for through organised labour, strikes, and political pressure, sometimes at real cost, as in the violence around the 1886 Haymarket affair. The eight-hour day was a victory, not a gift.
The forty-hour week followed the same path. In the United States it became law through the Fair Labor Standards Act of 1938, which capped the workweek and mandated overtime pay, and most other wealthy economies reached a forty-hour standard between the late 1940s and the 1970s.2 These were legislative achievements, won and then written into law so they could not be quietly withdrawn. The weekend you enjoy was, in effect, regulated into existence.
What about employer-led experiments?
They happened, and their fate is instructive. The Kellogg company famously ran a six-hour day for decades from the 1930s, and workers valued it enormously. But it was eventually abandoned, finally disappearing in the mid-1980s.3 The lesson recurs throughout the history: reforms that depend on a single employer’s goodwill are fragile, vulnerable to a change of management or of mood, exactly as Best Buy’s results-only experiment was killed by a new chief executive decades later. Durable reductions in hours have tended to come not from individual firms but from broad coordination, collective bargaining, and law, which change the rules for everyone at once and make the change stick.
Why did the progress stall?
This is the part that should focus the mind. For roughly a century, hours fell steadily. Then, around the 1970s, in much of the rich world, the decline stalled. Productivity kept climbing, but the hours stopped falling, and the gains were taken almost entirely as more output and income rather than as time.2 The historical engine of shorter hours, organised pressure converting productivity into leisure, lost momentum, and without it, rising productivity simply flowed into rising consumption. The stall is not a mystery. It is what happens when the mechanism that turns gains into time stops running.
What does this mean for the AI dividend?
Everything. AI is delivering, or will deliver, a productivity windfall on the scale of past technological revolutions. The history tells us, unambiguously, that such windfalls do not become free time on their own. They became free time, in the past, only when collective action and law deliberately converted them, and they stopped becoming free time the moment that conversion stalled. There is no reason to expect AI to behave differently. Left to default market forces, its gains will likely flow first to capital and to more output, not to shorter hours, exactly as the gains of computing did after the 1970s.
The encouraging half of the lesson is that this is a choice, not a fate. The weekend proves that ordinary people, organising and legislating, can claim a share of productivity as time. The contemporary movement, the four-day-week pilots, the legislated reductions, the thirty-two-hour-week proposals, is the same engine restarting. Whether the AI dividend is taken as time or as more work is therefore not a technological question at all. It is a political and collective one, and it will be decided the way it was always decided: by whether enough people insist on it together. The history is not a record of gifts received. It is a record of time taken back, and it can be taken back again.
Frequently asked questions
How did we get the eight-hour day and the weekend?
Through collective action and legislation, not employer generosity. The nineteenth-century eight-hour movement, organised labour, and laws such as the US Fair Labor Standards Act of 1938 capped hours and established the standards we now take for granted.
Were shorter hours ever given freely by employers?
Rarely. Reductions in working hours were overwhelmingly won through organised pressure, strikes, and law. The few employer-led experiments, like Kellogg’s six-hour day, were often later reversed.
Why does the history of shorter hours matter for AI?
Because it shows that productivity gains become free time only when collective action or policy converts them. The AI dividend will not automatically shorten hours; whether it does is a matter of bargaining and law, as it always was.
About the author
Tom Goodwin is the author of Don’t Work Harder, a book about taking the time AI gives back as time rather than more work. He is a co-founder of GAMEPLAN and writes on productivity, technology, and the economics of the working week.
Footnotes
Footnotes
-
The eight-hour movement and its slogan (“eight hours for work, eight hours for rest, eight hours for what you will,” associated with Robert Owen); the 1866 National Labor Union call and the 1886 Haymarket affair. ↩
-
The US Fair Labor Standards Act of 1938 capped the workweek (phased to forty hours by 1940) and mandated overtime pay; most other wealthy economies reached a forty-hour standard between the late 1940s and the 1970s, after which the decline in hours largely stalled. ↩ ↩2
-
Benjamin Kline Hunnicutt, Kellogg’s Six-Hour Day (Temple University Press, 1996): the six-hour day ran for decades and was finally abandoned in 1984 to 1985. ↩