Letter: Statistical reliability

Sir: 'Without a shadow of doubt,' writes Gavyn Davies, 'the most frustrating aspect of economic forecasting is the fallibility of the raw statistics on which we work.' Why then do economic forecasters, who aren't exactly infallible either, tirelessly persist in presenting their offerings with a decimal-

point precision warranted only by an assumption of omniscience (not just prescience) on the part of all concerned?

Far better, surely, for them to make a virtue of uncertainty (not to say ignorance) - and, incidentally, improve their chances of being proved right, at least before the results forecast are revised - by hazarding guesses accurate to the nearest half, at most quarter, percentage point only.

Who, after all, knows or cares about distinctions (between, say, 1.2 and 1.3 per cent GDP growth) dwarfed by the margin of error?

Yours sincerely,

WALTER GREY

London, N3

11 October