Sir, Your article on self-checkout (Hot Topic, p3, 28 August) caught my attention, in particular the sentence used as the headline: "If self-checkout wasn't working, the supermarkets wouldn't be in such a hurry to roll it out."

May I suggest another interpretation of their haste? Money.

Obviously, paying one person to supervise a bank of six checkouts (the number suggested in a letter by John Curnow of NCR in the same issue on p22) is cheaper than paying six people to do the actual work of scanning and packing your shopping for you.

So a bank of self-checkout tills are installed in place of the row of cashiers, leaving (in my local Tesco Metro) two token manned checkouts, doubtless still there to allow them to say they give customers a choice, at which are formed two of the most formidable queues you are likely to encounter anywhere on the high street.

Thus, when choosing their preferred means of checkout, the consumer is left with a choice of the bank of six self-checkouts (actually 10 at the same branch of Tesco), or the two manned checkouts with the super-queues.

Does this prove that self-checkout is a more efficient means of payment? Not really.

Ten to two is hardly a fair test. The relevant comparison is with the previous arrangement which, as I recall, used to be quite speedy enough.

Furthermore, it seems that a large number of people actually are willing to join the two super-queues rather then contemplate self-checkout.

And my own view? Oddly enough, self-checkouts in Tesco have never bothered me that much.

It's to be expected. We're talking about Tesco, after all. M&S is a different story, but I don't shop there any more.

Philip Gaudoin, grocery shopper, London