I have described the socialist computation problem as consisting of two parts: first, to solve the system of linear equations (consisting of a vast number of entries and featuring numerous factors of production), generating shadow prices; second, to improve it while preserving and creating the knowledge of new prices.
I have concluded that this problem is utterly computationally intractable. Thus, the Mises’ calculation argument against socialism can now, with the development of computer science, be put in highly precise terms and proven with maximal rigor.
What then is the “knowledge” problem as described by Hayek? Perhaps it’s something that applies not just to socialism but specifically to interventionism. Mises assumes that “the crowd of experts and specialists which [the central planner] assembles in his offices provide him with perfect information and answer correctly all questions he may ask them. Their voluminous reports accumulate in huge piles on his desk.” (HA, 696) Mises does so in order to grant his socialist opponent his best case.
But perhaps the assumption is extremely generous in itself. Is it really possible for any human being to obtain and digest all those data? Even in merely regulating rather than running the economy, the interventionist is faced with a formidable problem. Could Hayek be interpreted as saying that “regulators” don’t and can’t know what they are doing?