NYU's Solly Angel argues that city planners should do what they can to attain "orderly expansion." He wants major infrastructure trunk lines (water supply, sewage, highways, etc.) to be planned and laid out in advance of development. Sounds good and supposes the planners are good at predicting demand (the time and place) for development. If they are good at what they do and practice a light touch, they can let market savvy developers do much of the rest. The hard part is actually attaining this balance. Could we end up with "ghost cities" if we get it wrong? What are the likelihoods?
This brings up the difficult division of labor question when it comes to city development. Randy Holcombe suggested that infrastructure be the domain of public planners while land use development be handled by private planners (developers). The latter wouldd take published infrastructure plans as parts of the "rules of the game." I elaborated in my Spontaneous Cities chapter (of David Emmanuel Andersson's volume), suggesting that only the one-off mega projects be left to public authorities -- and noting the inevitable risks involved. Anything one-off will be risky but not likely to come into being via private risk taking.
Interesting but how do we get there? For starters, it would be nice if the question of a division of labor were included in planning curricula. Much of the literature presumes that markets are best characterized as "market failure" to be righted by public action immune to "government failure". But there is lots of evidence that the land use planning cherished by so many reduces housing affordability. Why not start there?