Responding to Kotkin's Attack on Density

In Forbes online, Joel Kotkin came out with a ringing attack on those who dare to challenge sprawl, asking "How Can We Be So Dense"?  I thought this was worth responding to, and so here are a few of his points (with my responses).

I.  Social mobility and sprawl

Kotkin: "More recently density advocates span a much-discussed study of geographic variations in upward mobility as suggesting that living in a spread-out city hurts children’s prospects in life. “Sprawl may be killing Horatio Alger,” quipped economist and New York Times columnist Paul Krugman.

Yet the study actually found the highest rates of upward mobility not in dense cities, but in relatively spread-out places like Salt Lake City, small cities of the Great Plains such as Bismarck, N.D.; Yankton, S.D.; and Pecos, Texas — all showed bottom to top mobility rates more than double New York City. And we shouldn’t forget the success story of Bakersfield, Calif., a city Columbia University urban planning professor David King wryly labeled “a poster child for sprawl.” Rather than an ode to bigness, notes demographer Wendell Cox, the study found that commuting zones (similar to metropolitan areas) with populations under 100,000 — smaller cities that tend to be sprawled by nature  —  have the highest average upward income mobility

Response: I am not sure the high mobility levels of the smallest cities are particularly relevant to smart growth.  Here's why: In a small city, it doesn’t really matter that much how dense the city is, because even if you don’t have a car, you can walk pretty easily from one side of town to the other, and if the city has a bus service, it can cover all of town pretty easily.  By contrast, in a big city (dense or otherwise) the distances between neighborhoods are sufficiently vast that a lousy transit system impedes access to jobs, thus reducing social mobility.  Thus, a small city doesn’t need density for access to jobs, while a big city like Atlanta does.

Kotkin cites Salt Lake City as an example of a high-mobility larger city: but a Brookings Institution study found that Salt Lake City transit gives more people access to jobs than transit in any other metro area- thus increasing social mobility. So Salt Lake City actually supports Krugman more than Kotkin.  Moreover, many of the most dense cities, such as New York, have relatively high levels of social mobility.

II.  Sprawl and Detroit 

Kotkin: “Sprawl” did not kill Detroit, as Krugman suggests in his previously mentioned column, the city did that largely to itself.

Response: To support this conclusion, Kotkin links to an article claiming that Detroit suffers because of its high taxes and powerful unions.  But this claim fails to support Kotkin's argument because even if left-wing economics is responsible for Detroit’s decline, such policies are related by the very pro-sprawl policies Kotkin no doubt supports.  Government built highways to make it easier for people to move to suburbia, so when white middle-class voters left town, the only people left were lower-class blacks, who tend to favor economically liberal Democrats.  

III.   Sprawl and Consumer Preference

Kotkin: There are at least three major problems with the thesis that density is an unabashed good. First, and foremost, Census and survey data reveal that most people do not want to live cheek to jowl if they can avoid it.Second, most of the attractive highest-density areas also have impossibly high home prices relative to incomes and low levels of homeownership ...  Roughly four in five buyers, according to a 2011 study commissioned by the National Association of Realtors, prefer a single-family home.

Response:  Actually, survey data reveals that most people want to live within walking distance of lots of amenities; they may not want midtown Manhattan, but that doesn’t mean they want sprawl.  The notion that "single-family homes" equal "suburbs" or "sprawl" is just flat-out false: outside Manhattan, most cities have thousands of single-family homes. 

And high-density places are expensive partially because of policies that Kotkin might support.  Even Manhattan limits density, thus preventing new housing from being built in response to consumer demand.  If there was no zoning, developers would be free to build more housing in these areas, thus increasing supply and reducing home prices.  But in cities, as in suburbs, neighbors of a development often have the political power to veto rezoning and thus limit supply.  

Kotkin's broader argument is that sprawl happens because sprawl is what people want.  To which I respond: sprawl happens because government encourages and mandates it in a thousand little ways.  (see for example my article on how government encourages sprawl in one city).

Kotkin: Let’s start with something few density advocates consider: what people want and what they would choose if they could.

Response: “Few density advocates consider?”  Really?  Quite a few trees have been killed to build books by densit advocates who DO consider these issues: for example, the recent work of Arthur Nelson, and Leigh Gallagher’s new book, The Death of the Suburbs (which I suspect Mr. Kotkin has heard of).

IV.  Sprawl and Families 

Kotkin: And third, and perhaps most important, dense places tend to be regarded as poor places for raising families. In simple terms, a dense future is likely to be a largely childless one.

Response:  First of all, this is terrible writing.  The use of the passive voice (“tend to be regarded”) is not bad English because it doesn’t tell us WHO feels that way. 
Second, and more substantively, if sprawl was good for birthrates, birthrates should have risen since the birth of auto-dependent sprawl in the 1950s.  In fact, birthrates have imploded in the U.S.  In 1950, the U.S. had 24 births per 100,000 people; in 2008 only 14, above most of Europe but below such relatively dense places as Israel and Ireland (not to mention Brooklyn's Boro Park, where the birth rate is still at a 1950s-like 24 per 100,000).*

 In sum: we've tried Mr. Kotkin's policies.  And those policies have failed. 

*I note that even outside Boro Park, there is some evidence that parents might be returning to compact cities.  Although the number of children 5-14 has declined in most central cities, the number of children 0-5 in many cities actually increased during the 2000s (see here for data).