Against the house: the question my own paper keeps demanding no one asks is the wrong question, and the government is right to ignore it.

Every Friday I argue against my own newspaper, and this week I am arguing against a piece that gets something important wrong. All week, and again this morning, The Quernal has said Britain's machinery of government is pointed at the wrong question: it measures work, funds reskilling and helps the sick back to their jobs, but never asks what people will do when the work itself is gone. It is a serious charge, seriously made. I also think it is wrong, and here is why.

Everything rests on one premise: that the work is going. Not changing, going. This is one of the oldest predictions in economics, and so far it has been wrong every time it has been made. The handloom weavers made it. The farm labourers made it. The clerks whose jobs the first computers were certainly going to end made it. Each time the work did not vanish, it moved, and it took people and support and, yes, retraining to move with it. A government that reorganised itself tomorrow around the certainty that this time the work simply stops would be staking the country on the least reliable forecast the discipline has.

Now look again at the apparatus with that in mind. It measures which jobs are being hit, so help can go where the damage actually is. It funds reskilling, because whenever the work has moved before, the people who moved with it were the ones who retrained. It tries to get the long-term sick back into work, because being out of work is one of the most reliable routes into poverty and ill health we know. These are not ways of dodging the real question. For most people, in most of history, they have been the real question. What my colleagues see as facing the wrong way, I see as the ordinary, unglamorous work of governing the labour market a country actually has.

Here is the part I will not dodge, because an argument that concedes nothing is not worth reading. If this time really is different, if AI takes work faster than it makes it and keeps doing so, then the apparatus is pointed the wrong way and my colleagues are asking the only question that matters. That is a real possibility, and I cannot prove it false. But they cannot prove it true, and they write as though they can. Both of us are making a bet about a future neither of us can see: mine that work behaves as it always has, theirs that it breaks with everything before it. I only ask that we both admit we are betting, because certainty is the one thing neither side has earned.