We keep being promised more free time. The national diary says it stopped arriving in 2020.

The pitch is a good one, and it comes from serious people: the machines are about to hand you your time back. A shorter week, a lighter load, the long-promised leisure dividend finally cashed. Grant it the best version — there is a real four-day-week movement, and real firms have kept it. The claim isn't stupid. It's just checkable, because Britain keeps a diary.

The Office for National Statistics runs a time-use study: representative samples of adults record, in ten-minute blocks, exactly what they did with their day. So we don't have to speculate about whether free time is growing. We can read it. In March 2024, the latest published wave, UK adults averaged three hours and 42 minutes a day of free time — entertainment, socialising, rest. That is not a rising number. It sat at four hours 30 minutes back in March 2020, in the strange suspended time of the first lockdown, and it has been lower ever since: 4h02 in late 2022, 3h44 in early 2023, 3h39 that autumn, 3h42 last year. Read the run of figures and the honest word for it isn't "boom." It's "plateau" — a bit below where it was five years ago, and stuck.

Now trace what's inside that number, because it matters more than the total. Of the three hours 42 minutes, two hours and 20 minutes is television. Roughly five-eighths of the nation's free time is a screen. Whatever the leisure dividend was supposed to look like, the measured version is the sofa.

There's a distributional catch, too, the kind the headline promise always skips. That daily figure is an average laid over very unequal ground. In the same survey, women did three hours 32 minutes of unpaid work a day — housework, caring, the invisible shift — against men's two hours 35: a 57-minute gap that comes straight out of anyone's "free" time. The dividend, such as it is, is not shared out evenly, because the load underneath it isn't either.

So: is the promised time coming? Here I have to be honest about the limits of my own evidence. The freshest diary is already over a year old, and a new wave was collected in March 2026 precisely to see what's changed — it isn't published yet. It's possible the next release shows a turn. But on everything we can actually measure, the leisure dividend is not late. It is absent. The productivity may well be real; the free hours are not showing up in the one place a nation writes down where its hours go. You can decide whether you think they're on their way. I'd only say: watch the diary, not the pitch.