Tyler Cowen noted that Cities and Suburbs Are Becoming Pretty Similar. Of course. A city-suburb dichotomy is much too simple. Nevertheless because Census and others report many data series this way, we keep using the two categories.
The idea of (suburban) "sprawl" is useless and specious. Qian An, Jim Moore and I showed some time ago that, using conventional travel time data, means as well as variances for work as well as shopping trips varied little no matter whether the traveler was "urban" or "suburban". In other words, most people and businesses have an interest arranging themselves spatially so as not to have to put up with irksome distances.
People and businesses in cities want two things: space and accessibility. We now say they want to network profitably. "Space" can proxy for a preferred amenities package. How can they get both? When large numbers of people and businesses consider the possibilities and the trade-offs available to them they make workable choices. The outcome is the cities we have.
There is enough movement of people and capital among cities that they must see themselves as competing. Uncompetitive packages of location and travel costs cannot survive for very long.
Robin Dunbar thought that a village of 150-250 would be as large a community that our brains could cope with -- in the interests of group survival. Our "villages" are now much bigger. The networks in our brains are networked with the networks in the brains of very many fellow citizens -- the ones with whom we interact physically and/or electronically. That's modern accessibility; we can seemingly manage the blend of networks we want whether in the "city" or the "suburbs."