Inherent illiquidity

Consider a book like, One Billion people, from Yglesias.

From the economic point of view, we can adjust economies of scale in a stable fashion a N grows to the Fuzzy Constant.  We can reach two deep into the value chains in order to adapt? How often?

Value chain theory gives you the stop and start points, and the memory of the system is seldom longer than a working career.  So the Yglesias and Caplans and the N => grow unlimited, are not accounting for the moments when real N drops and many N are shoved off the end of the chain, becoming Antificants. 

It is not central planning, it is the Bayesian ratios that appear to be optimum at any given point. And the amount of reorganization after ech change in scale can be estimated, as if we can say: 'There are not just enough groups in N to organize a supply chain of depth m. Partial equilibriums are real, they are a sparse set of maximum productivity points. points where the Antificant crowd is small, you are doing well matching economy of scale to N.  We note how to size a Walmart store over the aggregate of Walmart stores.