I recall a rather amazing paper a while back arguing that markets cannot be perfectly efficient unless P=NP. There are too many brutal NP-hard optimizations involved. I suspect the same would hold for any attempt at central planning. If these things were possible we would live in a utopia.
> I suspect the same would hold for any attempt at central planning.
It would be worse for central planning, because in addition to the problem you mention, central planning has the problem of getting the necessary information to the central planner, which in the general case is impossible: the information goes up as the exponential of the population size, but the bandwidth of information channels to the central planner only goes up linearly with the population size.
They are at least in the sense that instead of a small group of individuals attempting to process all the information (impossible), you have a large number of individuals processing subsets of the information (possible, but can have it's own issues like missing the forest for the threes).
