Monday, June 13, 2011

Ruminations on Rightsizing, Act I

In this first part of a series of pieces about Detroit’s evolving land use plans, I want to weigh in on the issue of population density.  Issues based on this simple metric are at the core of the reasons for, and objectives of, the city’s rightsizing debate.  Despite the significance of this measure, it seems to suffer from overuse and misuse by both advocates and detractors of rightsizing.

The Free Press’ famous “Comparing Detroit to three other major cities” map is one of the most effective, visually engaging and well-done maps I’ve seen.  No other map seems to have had such an effect on the rightsizing discourse.  The map's premise is mentioned in many articles about the city, including those by the Economist and Forbes.  


With that in mind, however, I think it might make unfair comparisons.  According to the 2010 census data, Detroit is still a relatively dense city.  Even after losing more than 200,000 residents since 2000, Detroit has greater population density than Denver, Sacramento, Portland, Las Vegas, Grand Rapids, Ann Arbor, Cincinnati, Dallas, Atlanta, Tampa, Austin, Boise, Des Moines, Orlando, Indianapolis, Louisville, Salt Lake City, Nashville, and Kansas City.  Among major American cities of its size or larger, Detroit is the seventh densest.  


My other concern with the Boston, Detroit, Manhattan and San Francisco map is that Detroit has always had less population density than these other three cities.  This is not a new change.  The Free Press could have made the same map in 1950, and it would still make Detroit look comparatively empty.  Moreover, San Francisco, Manhattan, and Boston are three of the densest communities in the country.  Healthy communities such as Chicago, Philadelphia, Brooklyn, and Seattle look sparse in comparison.  Even Indianapolis, a growing Midwestern city, looks empty when compared to Detroit, Baltimore, and Miami. 


Using this technique, you can make many cities look highly vacant.



Urban observers who argue against Detroit’s rightsizing plans often use this high-density defense.  This view, however, is also problematic, as noted in Rethink Detroit.  This defense by density is misguided, because low population densities are only especially problematic – from a budget standpoint – when a city was built for a higher density.  Jacksonville and Indianapolis look vacant in the density comparison maps because many areas of those cities remain undeveloped.  The local governments haven’t installed infrastructure such as streets, streetlights, and sewers in these areas, so they’re inexpensive to maintain.  

To accurately compare the relative density of cities, then, we should measure the amount of infrastructure per resident, in addition to the amount of area.  

Of course, there are many ways to measure infrastructure.  Perhaps the most ubiquitous type of infrastructure, streets and highways, is the best single indicator.  By this measure, feet of street per resident – “FSR” – Detroit has a lower density than 9 of the 10 largest American cities.

When compared to the seven largest American cities, only Houston has more feet of street per resident.


If you assess neighborhoods by population density alone, many of the best neighborhoods in Detroit, such as Indian Village, Boston Edison, and Palmer Woods, seem relatively vacant because the homes are on larger lots.  By an infrastructure density assessment, however, these areas appear healthy, because the roads are further apart, as well. 

While Detroit may appear rather stable through the lens of population density, the city’s FSR suggests another view.  Of the city’s 3,115.91 miles of streets and highways, 581 miles front vacant lots, according to Detroit Residential Parcel Survey data.  If you strung all of these lot-fronting roads together, they would stretch to Philadelphia, or to Chicago and back.  This is all infrastructure that must be maintained, salted, and policed. 

According to the Detroit Residential Parcel Survey data, 10,683 of the city’s 32,913 blocks are comprised of at least 25% vacant lots. 


Of course, FSR is but one metric available to planners, and population density remains a highly effective tool.  Residents should use both population density and FSR in their land use decision-making processes.  While both tools have shortcomings, they are effective complements and paint a more complete picture when used in unison.  

7 comments:

  1. Nice job showing the shortcomings of the whole "Detroit could fit Boston, San Francisco and Manhattan" observation. Joe Grengs would be proud of this article.

    My question is, given some of your observations using FSR and pop-density, such as this abundance of infrastructure servicing vacant lots, what can we do to reduce the cost of this infrastructure that is underused?


    Decommissioning infrastructure can be controversial, but is there a way to "mothball" infrastructure, roads in particular, in the same way you would a building, basically preserving it until needed?

    ReplyDelete
  2. Great article! I think that the 'feet of street' measure that you employed was an insightful metric for geographically measuring service area. Keep of the good work!

    ReplyDelete
  3. Good article indeed. Having been in Philadelphia for nine months or so, it has become clear that a metric for measuring a city's health based on density, which compares a city like Philly to a city like Detroit, will not be particularly accurate. Just a look a quick look at housing and neighborhood arrangement points to radical differences - the dominance of brownstone and rowhome multi-family apartments in Philly vs single-family detached homes on individual lots in Detroit. Detroit, as you address above, was not developed as a high-density city in the style of mid-atlantic east coast cities.

    ReplyDelete
  4. I don't know what this would do, but I am visualizing the Jacksonville map being plastered on vacant houses in Detroit. At least we could call it art.

    BTW, great work!

    ReplyDelete
  5. Good stuff Rob! Finding the right metrics is important. Have you read the NRDC Switchboard post about right-sizing and talking about the suburbs? It might be interesting to look at this as a regional/metro area comparison. Regional anything would be tough to implement in any fashion though...

    Post here: http://switchboard.nrdc.org/blogs/kbenfield/which_part_of_detroit_needs_ri.html

    ReplyDelete
  6. Definitely interesting; my family and I were just discussing some of these ideas (relative density, and lane-miles per resident). Did you consider trying to calculate area of street per resident at all? My guess is that could get a little sticky, but I'd be interested to know the stat for Detroit and other cities.

    ReplyDelete
  7. Very interesting article. I notice that you're using linear feet of street and assuming that street widths are similar across cities. I'm not sure how well that works. Detroit does have some extra wide streets compared with newer cities. I asked DPW last week about the number of vehicle-lane miles in the city, but they didn't have that. I believe cities must report the number of vehicle-lane miles for Federal aid roads, but I cannot find that information

    ReplyDelete