How many cities and regions are basing their land-use and transportation plans on the notion that most Millennials want to live in dense cities? Officials in those areas should read a new report from the Urban Land Institute (ULI). Among other things, the report says that three-quarters of Millennials in the nation’s 50 largest metropolitan areas live in the suburbs.
That’s not much less than the 79 percent of everyone in those metro areas who live in the suburbs. Moreover, 76 percent of minorities live in the suburbs, again not much different from the overall average. The ULI has a history of promoting smart growth and other urban planning fads, so these conclusions aren’t based on any hidden agendas.
Most of the ULI report is spent dividing suburbs into one of five categories: high-end, middle-class, “economically challenged,” “greenfield lifestyle,” and “greenfield value.” Greenfield lifestyle refers to master-planned communities that include parks and other amenities while greenfield value tend to be ordinary subdivisions with fewer amenities. This kind of classification may be useful to realtors, but if urban planners attempt to use it, perhaps to “fix” the economically challenged areas, the result is likely to be a disaster.
For example, the ULI report argues, contrary to its own evidence, that people are willing to pay a “premium” to live in walkable neighborhoods, which supposedly means there is a “deficit” of such neighborhoods. Planners who believe this are likely to try to impose such walkability on the challenged and value neighborhoods, which will make them more expensive. As I’ve previously noted, this approach confuses cost with value. The fact that walkable neighborhoods cost more doesn’t mean the average person is willing to pay a premium to live there; it only means that a few people will pay that premium. The fact that there are so few such neighborhoods shows that the number willing to pay that premium is small.
The media reports about this study ignore the classification of suburbs and instead focus on the fact that suburbs remain the dominant American lifestyle. In reaching this conclusion, it is important to realize that the report defines “suburban” as “Well-populated neighborhoods where most of the housing stock consists of single-family detached homes.” This includes lots of neighborhoods in central cities, whereas some neighborhoods outside of central cities, which most people would call suburbs, might be classified as “urban” in the ULI study.
This approach is fine if your purpose is to find out how many people live in high- or low-densities. But much of the city/suburb debate is a struggle over who gets to tax residents of an urban area, and the ULI numbers fail to address this issue. The study, for example, says that just 17 percent of people in the San Antonio area live in an “urban” setting, even though 62 percent live in the city of San Antonio itself. Because it has strong annexation powers, San Antonio has high share of the residents of its metropolitan area; by comparison, the city of Portland’s much weaker annexation authority means it has only 26 percent of the residents of the Portland-Vancouver metropolitan area. Not surprisingly, sprawl is a bigger issue in Portland than San Antonio.
Another cautionary note is that the ULI study is based on metropolitan statistical areas rather than urbanized areas. An MSA includes all of the land in the counties in which a city and its suburbs are located. In most cases, most of the land in the MSAs are neither urban nor suburban but rural. I prefer to use urbanized areas, which are more-or-less economic boundaries. Metropolitan areas are political and thus somewhat arbitrary, which can lead to distorted results.
For example, according to the report, 82.8 percent of the Boston metro area is suburban while only 2.9 percent of the Las Vegas metro area is suburban. This sounds very unlikely, but it’s because 97 percent of the land in the Las Vegas metropolitan area is rural, compared with only 15 percent of the land in the Boston metro area. When using urbanized areas, 98.2 percent of Boston and 87.5 percent of Las Vegas are suburban, which sounds a lot more realistic.
Given that urbanized data are available for just about any census questions, it seems strange that researchers continue to use metropolitan statistical areas.