This is an American Planning Association book, so even in its paperback form (believe me, not available as an ebook. _PLEASE_ prove me wrong) it is over $30. The lone review on Amazon is dead accurate (by Michael Lewyn of Jacksonville, FL, if others show up later and you wonder which one I'm referring to): the data-heavy parts are excellent, the rest is iffy. In particular, Lucy's analysis is fairly crappy.
The first few chapters lay out something that I'd worked out on my own: starting around 1990, affluent whites started moving back towards central cores, at least in some cities (not Detroit. Conspicuously not Detroit), a trend which got notably stronger in the ensuing two decades. At the same time, new households were also forming out on the periphery, including households which didn't commute to the local core, but rather to an "edge city" elsewhere on the periphery. So: two trends: a trickle, widening to a stream headed back to the center, and a wider stream from the older suburbs to further out suburbs. But that second stream was so heavily encouraged by multiple administrations wanting to increase homeownership rates (for various reasons of their own) that when circumstances turned against it, it showed much less resilience than the movement to the core.
While Lucy's projections are in line with what I expect, I feel like he's missing a lot in describing closer-in/older suburbs that do well. He is writing in terms of suburbs experiencing teardown activity rebuilding single family homes on existing lot vs. suburbs that need to be helped to expand 1945-1970 small housing. As a side note, Lucy does _not_ explain why teardown monster houses are different from added-onto small housing (it's simple: monster house price justification requires a 3x; add-ons only require you to get the size up to what you need it to be, which is going to be more like 1.5-2x -- it would have been nice if he'd pointed this out, however no one ever does). There's one spot where he mentions suburbs that have townhomes being built in them, but it's very tangential; he doesn't get into lot combining _anywhere_, much less greyfields being redeveloped with residential.
Lucy has a very nice demographic analysis based on the generation after the boomers being small: they are in household formation years, and there are fewer of them (believe me, I am well aware of this, as a member of this cohort). Lucy does not draw a lot of attention to their comparatively high rate of lifelong childlessness, which just makes what he is pointing out more extreme. That is, through the high household formation years of the boomers, you could just keep building houses. But once they've all reproduced (or not) and bought houses to support that activity (or not), you really have to at least slow down (and possibly stop for a while).
Lucy says some incredibly silly things in analysis later in the book. He goes to some effort to show that foreclosure rates were low in places where house prices with respect to median income were really high -- and then he says policy interventions should work to bring house prices down with respect to median incomes. He also talks about how suburbs that don't benefit by location/transit/whatever will need help. I understand the idea is not to completely screw people, but there's a part of me that just goes, why? I thought the goal was to get people to move to some areas and away from others -- and now you want to dilute the effect? Create transit, encourage transit oriented development, and then _reinvest_ in the areas you _don't_ want people in?
Lucy also has a bunch of ideas about governance structures. He _really_ doesn't like the thicket of government at the local level. But this leads him to say this:
"A risk associated with required collaboration among local governments could be that suburbs' representatives will support highway spending that encourages more sprawl. With many central cities constituting small proportions of metropolitan populations, the possibility that MPOs [Metropolitan Planning Organizations] will promotion deconcentration is a significant danger. Collaboration is a means to an end. The goal should be more compact development, shorter routine trips, and more alternatives to driving alone. Collaboration itself is not the goal."
Clearly a huge fan of democracy here. (<-- Sarcasm.)
I can't really recommend it, altho I'm reluctant to call it a bad book. It is problematic. I'm still working on _Cities and Suburbs_ (which is both better and worse) and _Devil in the White City_ (book club pick that turns out to be weirdly appropriate for my current obsession, altho wow Larson's writing style is not one I care for).