When my book The Undercover Economist was published five years ago, I would occasionally be asked whether I was in favour of sweatshops in developing countries. Not at all, I would reply. But I could see where the question was coming from, because I was certainly worried as to whether campaigning against them would do any good.

My argument had a logic that will be familiar to economists. Unless sweatshop workers are literally slaves, they are presumably working long hours in horrible conditions for low pay only because the alternative ways of making a living are worse.

When a well-meaning group of activists launches a campaign against sweatshop labour among, say, Nike suppliers in Indonesia, the obvious risk is that the sweatshops are closed, workers are tossed out on to the street, and the work is shifted to computerised sewing machines in Osaka. This is surely not the aim. The only alternative is economic growth: while it may be frustratingly slow, it finishes off sweatshops by producing far more attractive jobs.