Shrinking cities are cities that are experiencing acute population loss. Deindustrialisation and out-migration are some of the common reasons that cities shrink. In the United States, this problem is most commonly associated with the Rust Belt, while parts of Eastern Europe also experience similar problems[1]. Since the infrastructure of such cities was built to support a larger population, its maintenance can become a serious concern[1]. A related phenomenon is counter urbanization.
Contents |
The current population of the world is 6.5 billion people. Currently about 3 billion live in cities. By 2030, population in cities is expected to be 4.1 billion. Though some cities are continuing to attract residents, others are losing residents at varying rates. Saskia Sassen's "global cities" theory forecasts these urban "winners" and "losers": the winners being those cities with agglomerated financial and specialized services and the losers being those with outdated industrial infrastructure and manufacture economies.[2]
In the last 50 years, about 370 cities with more than 100,000 residents have undergone population losses of more than 10%. More than 25 percent of the depopulating cities are in the United States, and most of those are in the midwest.
Among many challenges facing shrinking cities, infrastructure management is among the most intractable. Shrinking cities were once more dense, industrious places requiring vast infrastructure to support the resulting population and economic activity. However, they now face a vastly reduced population. The remaining legacy infrastructure was intended for a larger population.
Overcapacity in infrastructure can severely strain the fiscal situation of any cities, most particular those who are already in dire straits. The problem is relatively straightforward; a declining city with aging infrastructure has fewer people to share the cost resulting in a higher per capita cost to the residents who remain. Since infrastructure costs are usually fixed, reductions in demand do not reduce costs.
The dispersed and sparsely populated neighborhoods that inevitably characterize shrinking cities are a major source of fiscal distress. Services must still be provided to fewer and fewer citizens over a larger geographic distance, once again raising the per capita cost. This is an instance illustrating how a lack of population density can be a strain on already stretched municipal budgets. “At the neighbourhood level, residential density is directly linked to expenditures on neighbourhood infrastructure. The higher the density, the lower the per capita length of collector roads, water distribution lines or sewer collection lines. Below a density of 40 dwellings per hectare net urban land network-related per capita costs increase exponentially”[1].
Some policymakers have suggested that such neighborhoods might be consolidated at a higher density to increase their ability to pay for the required municipal services. Cities such as Flint, MI have explored this option, yet it is currently largely in a preliminary stage. Since such a policy has yet to be formally implemented in the United States , the preliminary evidence regarding is efficacy is mixed at best. Decommissioning public infrastructure is an elaborate, expensive process that requires a great deal of study. Currently, many infrastructure management professionals believe it more sensible to incur current maintenance cost at some minimal level rather than removing infrastructure that may later need to be reinstated. Since future growth can be hard to predict, this is a pragmatic argument that can only be countered with detailed research.
However, such detailed research regarding infrastructure management in shrinking cities is not widespread at this point. Kent State University’s Sustainable Infrastructure in Shrinking Cities report consistently affirms this point. Simply put, far more research is needed before we can confidently recommend policy. As the report concludes:
“When we began this research, we hoped to find a technology or strategy that would enable substantial cost savings by decommissioning large components of costly infrastructure that were no longer necessary due to declining population. We found no such thing.” [3]
Several approaches are being employed in various cities to attempt to address the problem, using a combination of financial incentives and regulatory controls.
Urban renewal is a broad set of attempts intended to renovate core urban areas to make them more attractive to businesses and residents. The methods and results have been mixed, with a wide spectrum of results from negative to positive.
An urban growth boundary is a legal mechanism for limiting sprawl, which necessarily increases density within the boundary. Its intent is to prevent the problem of urban shrinkage before it happens. The boundary generally encompasses the city and its surrounding suburbs, requiring the entire area to work together to prevent unlimited sprawl and consequent urban shrinkage. This method is being used successfully in manys (such as Portland, Oregon) to maximize returns on infrastructure investments.
The following cities have lost at least 20 percent of their population, from a peak of over 100,000, since 1950.
| City | 1950 population | Peak population (year) | 2010 population | Percent decline from peak population | Notes |
|---|---|---|---|---|---|
| Akron, Ohio | 264,605 | 290,351 (1960) | 199,110 | 34.5% | |
| Albany, New York | 134,995 | 134,995 (1950) | 97,856 | 27.5% | 2.3% increase from 2000. |
| Baltimore, Maryland | 949,708 | 949,708 (1950) | 620,961 | 34.6% | |
| Birmingham, Alabama | 340,887 | 340,887 (1950) | 212,237 | 37.7% | |
| Boston, Massachusetts | 801,444 | 801,444 (1950) | 617,594 | 22.9% | 4.8% increase from 2000. |
| Buffalo, New York | 580,132 | 580,132 (1950) | 270,240 | 53.4% | |
| Camden, New Jersey | 124,555 | 124,555 (1950) | 77,344 | 37.9% | |
| Chicago, Illinois | 3,620,962 | 3,620,962 (1950) | 2,695,598 | 25.6% | |
| Cincinnati, Ohio | 503,998 | 503,998 (1950) | 296,943 | 41.1% | |
| Cleveland, Ohio | 914,808 | 914,808 (1950) | 396,815 | 56.6% | |
| Dayton, Ohio | 243,872 | 262,332 (1960) | 141,527 | 46.1% | |
| Detroit, Michigan | 1,849,568 | 1,849,568 (1950) | 713,777 | 61.4% | |
| Flint, Michigan | 163,413 | 196,940 (1960) | 111,475 | 43.4% | |
| Gary, Indiana | 133,911 | 178,320 (1960) | 80,294 | 55% | |
| Hammond, Indiana | 87,595 | 111,698 (1960) | 80,830 | 27.6% | |
| Hartford, Connecticut | 177,397 | 177,397 (1950) | 124,060 | 30.1% | 2.0% increase from 2000. |
| Jersey City, New Jersey | 299,017 | 316,715 (1930) | 247,597 | 21.8% | 3.1% increase from 2000. |
| Minneapolis, Minnesota | 521,718 | 521,718 (1950) | 382,578 | 26.7% | |
| Newark, New Jersey | 438,776 | 442,337 (1930) | 277,140 | 37.3% | 1.3% increase from 2000. |
| New Haven, Connecticut | 164,443 | 164,443 (1950) | 129,779 | 21.1% | 5.0% increase from 2000. |
| New Orleans, Louisiana | 570,445 | 627,525 (1960) | 343,829 | 45.2% | 2.1% increase from 2008 Census Bureau estimate; first after Hurricane Katrina. |
| Niagara Falls, New York | 90,872 | 102,394 (1960) | 50,194 | 51% | |
| Philadelphia, Pennsylvania | 2,071,605 | 2,071,605 (1950) | 1,526,006 | 26.3% | 0.6% increase from 2000. |
| Pittsburgh, Pennsylvania | 676,806 | 676,806 (1950) | 305,704 | 54.8% | |
| Providence, Rhode Island | 248,674 | 252,981 (1930) | 178,042 | 29.6% | 2.5% increase from 2000. |
| Reading, Pennsylvania | 109,320 | 111,171 (1930) | 88,082 | 20.8% | 8.5% increase from 2000. |
| Rochester, New York | 332,488 | 332,488 (1950) | 210,565 | 36.7% | |
| Scranton, Pennsylvania | 125,536 | 143,333 (1930) | 76,089 | 46.9% | |
| South Bend, Indiana | 115,911 | 132,445 (1960) | 101,168 | 23.6% | |
| St. Louis, Missouri | 856,796 | 856,796 (1950) | 319,294 | 62.7% | |
| Syracuse, New York | 220,583 | 220,583 (1950) | 145,170 | 34.2% | |
| Toledo, Ohio | 303,616 | 383,818 (1970) | 287,208 | 25.2% | |
| Trenton, New Jersey | 128,009 | 128,009 (1950) | 84,913 | 33.7% | |
| Utica, New York | 100,489 | 100,518 (1940) | 62,235 | 38.1% | 2.6% increase from 2000. |
| Washington, D.C. | 802,718 | 802,718 (1950) | 601,723 | 25% | 5.2% increase from 2000. |
| Youngstown, Ohio | 168,330 | 170,002 (1930) | 66,982 | 60.6% |