I’m a fan of Joel Kotkin, but I’m not sure he’s right about the following:

In the past decade, suburbia extended its reach, even around the greatest, densest and most celebrated cities. New York grew faster than most older cities, with 29% of its growth taking place in five boroughs, but that’s still a lot lower than the 46% of growth they accounted for in the 1990s. In Chicago, the suburban trend was even greater. The outer suburbs and exurbs gained over a half million people while the inner suburbs stagnated and the urban core, the Windy City, lost some 200, 000 people.

Rather than flee to density, the Census showed a population shift from more dense to less dense places. The top ten population gainers among metropolitan areas — growing by 20%, twice the national average, or more — are the low-density Las Vegas, Raleigh, Austin, Charlotte, Riverside–San Bernardino, Orlando, Phoenix, Houston, San Antonio and Atlanta. By contrast, many of the densest metropolitan areas — including San Francisco, Los Angeles, Philadelphia, Boston and New York — grew at rates half the national average or less.

It turns out that while urban land owners, planners and pundits love density, people for the most part continue to prefer space, if they can afford it. No amount of spinmeistering can change that basic fact, at least according to trends of past decade.

Are people choosing low-density metropolitan areas — or did rising prices in high-density metropolitan areas drive the population shift? “Urban land owners, planners, and pundits” might love density, but have dense cities allowed enough new construction to meet existing demand to live in dense, amenity-rich cities and regions? Ryan Avent’s The Gated City offers an alternate hypothesis that strikes me as more convincing. If regions like New York and the Bay Area allowed more construction, more people would choose to live in them. Density isn’t deterring potential migrants and encouraging emigration: high prices driven by constrained supply are to blame.  

P.S. Commented Joel Av makes a great point below:

Reihan: as a Chicagoan I have to add that Kotkin’s thesis probably works more here in Chicago than in NYC or in the Bay Area. Housing here is better than those places, thanks to fewer regulatory constraints on residential construction (as you have noted in your previous blog posts). Prices for middle class homes in dense neighborhoods are comparable with those of middle class suburban homes.

What Kotkin misses – or fails to address adequately – is that the overall value proposition of living middle class life here is much less attractive than it used to be. The city schools here are terrible (we have arguably the shortest school day in the U.S.), the police protect their incompetent, corrupt officers, and the quality of other public services continues to decline as taxes continue to increase. These trends will likely continue until local and state governments clean up their balance sheets and incorporate real measures of accountability into the public sector.

The decline of public services is a trend common to many cities, but might be overlooked by some as a primary driver of suburban migration because of the focus that they devote to housing costs and urban density. So speaking purely in terms of an educated guess, I think it’s this general value proposition rather than density per se. Middle class families are increasingly choosing the suburbs over the city – and the suburbs of Atlanta, Dallas etc. over the Chicago ‘burbs, I would add – because they think they’re getting a better bang for their (tax) buck there than here.

This is dead on. And it’s not inevitable that cities will be governed poorly, though unfortunately the problem suffers from path dependence and the mismatched party heuristics raised by David Shleicher.