… is from page 214 of F.A. Hayek’s November 1941 Nature paper, “Planning, Science and Freedom,” as this paper is reprinted as chapter ten of the 1997 collection, splendidly edited by Bruce Caldwell, Socialism and War:
[T]he competitive price system makes possible the utilization of an amount of concrete knowledge which could never be achieved or approached without it. It is true, of course, that the director of any centrally planned system is likely to know more [about the economy at large] than any single entrepreneur under competition. But the former could not possibly use in his single plan all the combined knowledge of all the individual entrepreneurs that is used under competition. The knowledge which is significant here is not so much knowledge of general laws, but knowledge of particular facts and the ever-changing circumstances of the moment – a knowledge which only the man on the spot can possess. The problem of the maximum utilization of knowledge can therefore be solved only by some system which decentralizes the decisions.
I’m aware that I repeat myself – repeatedly. But I don’t apologize for doing so, for I’ll repeat the following question until advocates of industrial policy – which is socialism-lite – offer a substantive answer to this question: How will the officials charged with carrying out industrial policy solve or escape the problem, described here by Hayek, of the utilization of decentralized and ever-changing bits of detailed knowledge?
Solving or escaping this problem is necessary if industrial policy is to enrich anyone other than the relatively few officials who run it, and the producers who are protected by it. The commonplace move of assuming that a miracle occurs doesn’t work. Ignoring the problem – another commonplace move – also is inadequate.
So how? How exactly will industrial policy make successful use of at least as much knowledge as is every moment made use of by markets?