Britain keeps running the experiment. It keeps looking away before the results come in.

Start with the number, because it is genuinely remarkable and almost nobody noticed it. Between 2022 and 2025, Wales paid 644 young people leaving care £1,600 a month, before tax, for two years, with no conditions attached. Ninety-seven per cent of those eligible took part — the highest uptake of any opt-in basic-income scheme anywhere, and the most generous such trial the world has run. Then, last summer, the payments stopped as planned. The pilot was always time-limited. It has not been extended, and there is no move to scale it. The full evaluation, run by Cardiff's CASCADE centre, reports in 2027.

This is not a new shape. Basic income arrives dressed as a Silicon Valley novelty — a fix for the age of the machines — but the idea is old and very British. You can trace the lineage to Speenhamland in 1795, where Berkshire magistrates topped up labourers' pay against the price of bread; through the mid-century argument over family allowances; to Milton Friedman's negative income tax. What is new is only the willingness to test it properly, and there the record is unkind. Finland ran a two-year trial to 2018 and chose not to continue it. Ontario launched one in 2017 and a new government cancelled it in 2018, mid-experiment, before it could report — its findings arriving to an audience that had already stopped listening.

That is the pattern worth naming, and it is not "basic income fails." The trials mostly do not fail; they find modest, real things — a little more security, better mental health, small effects on work in either direction. The pattern is that the evidence is hostage to political time. An experiment takes years; a political cycle takes months; and the careful answer turns up after the question has been asked, argued and abandoned. The myth is that we tried it and it did not work. The record is that we rarely stayed in the room long enough to find out.

Where does the Welsh trial diverge? Encouragingly, it will actually report — CASCADE is seeing it through to 2027. And its early findings already carry this paper's question in miniature. Some young people described the money as the first real autonomy they had known; a few, professionals observed, stepped back from work they had not needed to do; and almost all found the drop back onto Universal Credit, when the money stopped, brutal. That last detail is the one to hold. The hardest part was not the income. It was the cliff at the end of it.

So here is the perspective, not a verdict. Britain has, quietly, run the closest thing it has to a real answer to this paper's question — what do people do when they do not have to work? The answer is being written up now, and will be ready in 2027. The one thing history suggests we will not do is wait for it.