If farmland is as scarce as Monbiot proposes then vertical farms could work. Most of the value of a $1000 a month apartment is in consumer surplus. Yes, a vertical farm in manhattan will probably never work, but one on some existing farm land in New Jersey probably will. Vertical farms currently won't work but any real scarcity in farmland will drive up prices until it becomes economical as the price of food is highly elastic.
Farmland isn't actually scarce, though. Right now the world has plenty of farmland for all its people, which is why food is so cheap and why we can afford to consume massively space-wasting crops. People may be starving in Africa, but it's for political reasons which have nothing to do with the amount of arable land there (significantly more than China or India which have similar populations to Africa).
If the world's population ever does increase to the point where farmland is scarce, there's a lot of other things which become economical before skyscraper farms do... even if fusion energy comes along to solve the lighting problem. Switching to land-efficient crops is one thing. Irrigating the world's deserts is another. Even farms on barges out in the ocean make a lot more economic sense than skyscraper farms.