Shrinking Cities: Understanding Urban Decline in the United States
By Russell Weaver, Sharmistha Bagchi-Sen, Jason Knight, and Amy E. Frazier
Cities like Detroit, St. Louis, and Cleveland have lost stunning percentages of their peak population since 1950. Yet these are all in metro areas whose regional populations are much higher than in 1950, even if not at their all time peak high in all of them.
Some cities like Youngstown have gone so far as to try to plan for shrinkage and a permanently reduced population future. Detroit did something similar with its Detroit Future City plan, though that has subsequently been scrapped.
How do we think about shrinkage? How do we define a shrinking city? What should shrinking cities do? These are questions that have been swirling around for the decade I’ve been writing about cities here.
The new academic book Shrinking Cities aims to put some rigor around those questions. The first half of the book is devoted to an examination of shrinkage in the United States. The authors note that any measure designed to identify shrinking cities has a note of the arbitrary about it. The measure they select is population decline of 25% or more over the 40 year period from 1970 to 2010. They look at both census tracts and cities. Surprisingly, they find the shrinking census tracts are very widespread in America, and are on the rise in the Sun Belt. Shrinkage in terms of population loss is not a Rust Belt only phenomenon, though shrinking municipalities as a whole are concentrated in that region.
The authors look at economic decline separately, examining census tracts where there were increasing levels of concentrated disadvantage (which include include such measures as female headed household, unemployment rate, and low educational attainment as disadvantage indicators). Unsurprisingly, economic decline is more common and more severe in shrinking tracts, though there is not complete overlap. They find that social distress is more associated with economic decline whereas physical distress (e.g., vacant housing) is associated more with population decline.
After a presentation of the data, the book reviews various theories of what causes shrinkage and decline such as suburbanization. One that was particularly interesting was the social capital theory. Is low social capital associated with shrinkage and economic decline? It is. One of the indicators of social capital is homeownership. The book observes:
In other words, relative to tracts that shrank and did not decline, tracts that experienced coupled shrinkage and decline witnessed a substantial drop in homeownership rates that cannot be explained by chance alone. If homeownership is a useful indicator of social capital, then his result implies that social capital may have decreased in these shrinking-declining tracts as well. Thus, tracts that are able to keep their homeownership rates steady during population shrinkage may be more resistant to decline via social capital.
The second half of the book looks more at potential strategies for countering or adapting to change. These include various pro-growth policies, “rightsizing”, regional government, community development initiatives, etc. While some of these have had success in isolated cases, none of them have had systemic, replicable success. This makes for a bit of depressing reading. As one person told me about this aspect of the book, “Shrinking cities theory is at a dead end.”
Because of its academic nature, and high price tag, this book is not for everyone. But policy developers in shrinking areas should certainly ground themselves in this summary of shrinking cities research and academic theories found in it.
Cover image photo of Brush Park, Detroit by Stephen Harlan, CC BY-SA 2.0
Chris Barnett says
Aaron, does the book address the fact that one of the underlying causes of “shrinkage” in fixed-boundary cities is the decline of average household size?
Since 1960, average US household size has fallen from 3.33 to 2.53, a decline of 25%. If nothing else changed, fixed-boundary city populations (i.e. St. Louis, Detroit, Cleveland, which were landlocked by 1960) would likely have fallen by 25% before anything else had impact.
I also wonder how deeply anyone examines the issue of consumer housing type preference. “Shrinking cities” have an abundant stock of 900sf 2br bungalows and 1-1/2 story, 1500sf, 3br Tudors and Cape Cods with 8-foot ceilings and a detached garage (the 1920-1940 streetcar suburban archetypes). It’s expensive and difficult to convert those into open-concept, big-kitchen homes with high ceilings and a 3-car garage (as seen on HGTV).
Aaron M. Renn says
The book talks about demographic factors, but I don’t remember what it says about household size per se.
Keep in mind that cities such as SF, NYC, and LA are at all time record population highs despite declining household sizes.
Chris Barnett says
Those may be the only major US cities that can claim population growth while locked into historic boundaries since 1960. Possibly also Miami. No wonder; those are the US’ “world cities”. Volumes are written annually on how they’re different.
Philly is 25% off its peak. Likewise Minneapolis and Chicago. Baltimore, a little more. Milwaukee and DC, a little less, and Boston around 15%.
Of course Detroit, Cleveland, Buffalo, Cincinnati, St. Louis, and Pittsburgh are all off a lot more.
Newer cities, and those that have continued to grow through annexation or consolidation, are hard to gauge without getting into historic census tract data. Indianapolis, Jacksonville, Columbus, Atlanta, Nashville, Charlotte, Dallas, OKC, KCMO, Denver, Phoenix, San Diego, Austin, San Antonio, Memphis, Orlando and others fit in this category.
On its face, though, allowing for the (background) nationwide change in household size does seem to account for a significant share of population decline in fixed-boundary core cities over the past 50-60 years. Ignoring that factor would do a disservice to any analysis.
I pointed out the issue of housing type and choice to a fellow citizen well into his cups. He was a native of “suburban” (outer 1920s) working man’s cottage Detroit, and he was downright insulted. 🙂 “Solid brick bungalows are the best”.
In some ways, I find those neighborhood more charming than “McMansion Hell”. I wonder how well the latter 3,500 square foot particle board monstrosities will age?
Chris Barnett says
bb, many of the sad pictures of decay in Detroit are streets of those (unwanted and unloved) brick bungalows and Tudors. I had those in mind when commenting.
Yet similarly-built Midwestern places like Bexley (inner suburban Columbus) and Meridian Kessler and Irvington (former streetcar suburbs of Indy) and chunks of Minneapolis survive and thrive.
But in Indy, the “not brick” worker bungalow tracts are not doing so well.
Rod Stevens says
At the Congress of New Urbanism’s Detroit meeting last year, I visited neighborhoods between downtown and midtown that, based on the quality of the remaining housing there, must have been some of the wealthiest in America at the turn of the 20th century. Today only one house stands in blocks that previously had 40. Neighborhoods to the east near Belle Isle, what was once the city’s crown jewel of open space, are now open space themselves, vast open fields of building lots. These places don’t simply lack social capital, they lack buildings.
The immediate question is whether and how to move the remaining people out of these places so they no longer need police service, new lighting, or pipe replacement. These places might be re-inhabitted sometime, but there is so little job or wealth creation that existing construction can meet most needs for the next 30 years or more. And moving people raises the question of how to make the remaining neighborhoods stronger, so that isolated people will want to move there. Closing down one place implies making the other more attractive, a push-pull relationship. Economic development obviously needs to be part of the overall solution, but in the short to mid term the question is how to use limited public funds better, to actually improve life in the remaining places.
Then there’s the question of what to do with these “left overs”. Urban ag probably can’t pay the wages that young people need to survive, and few are likely to be drawn to it. The best lessons might be from the former East Germany, and what places there have done with their Stalinist-era industrial complexes. (Which are not so different from our own Eisenhower-era industrial complexes.) Most of the re-used places there seem to have become parks, but not parks like Central Park or Golden Gate Park, but more like pathways among industrial wastelands where nature’s retaking of the landscape is a slow and interesting show.
One of the first books to come out after 2008 about our industrial melt down was “Life Without Us”, how nature would come back in the city. Most of our redevelopment actions involve big ticket, highly-engineered human interventions, usually involving some flavor-of-the-month tourism or business idea. Maybe, like Frederick Law Olmsted, we need to think in terms of re-sculpting the land, planting the right species, and then letting nature grow back.
One of the strangest examples of this is on the Upper East Side. Indian Village is a pretty amazing collection of larger, more expensive Tudor houses I am guessing from the 20s. Imaculately maintained, beautiful yards, cute ornamental street lights, “block club” and “historic district” signs on every main entry street. You literally go immediately outside these preserved-in-amber 20s suburbs and it is totally decayed urban prairie. Weird thing to see!
Sounds like a wonderful book, but it seems as though defining shrinkage by population is limited in scope. Cities are a connection between so many factors pushing and pulling against each other. Perhaps we should focus on how population play as a factor in a health of a city, what are positive and negative indicators? What is the “right/ideal/optimal” balance that needs to be maintained between housing, jobs, infrastructure spending, governmental size, private/public investments, etc. and population. What is healthy and how do we craft a plan, including a feedback loop that addresses changes, that can move towards that destination.
Matthew Hall says
The only metros actually losing population are Detroit, Cleveland, Buffalo, and Pittsburgh. Even Cincinnati, St. Louis, and Baltimore are sustaining their metro populations. The southern half of St.Louis is doing increasingly well and is sustaining it’s metro population while downtown and midtown Cincinnati are seeing investment and activity they haven’t seen in generations. This is really only an issue for a handful of American cities and metros.
Robert Cook says
Maybe somewhat related, but there is a fascinating graphic here looking at county-level income distribution. You can play with the numbers and the counties to see how things have changed over time. http://xenocrypt.github.io/CountyIncomeHistory.html