Degrees of Decline: Why Growing Regions Sometimes Produce Shrinking Cities

If the population history of U.S. industrial cities was a Game-of-Thrones-esque television series, I think we’d all be able to guess how the “plot synopsis” would read for the episode lasting from 1970 to 2000:

Sir Buffalo, Lady Cleveland, and their best friend Saint Louis are at an exuberant high point as the episode begins. At a moment when it seems their kingdoms will prosper forever, the armies of Lord Globalization and the evil sorcerer Suburbia mount a surprise attack. Despite their best efforts, our heroes are wounded and bleed out…for over 30 years!

Rating: TV-MA (some sprawl and blight not suitable for children)

Now of course, this TV show wouldn’t tell the whole story of U.S. cities (see witty comedy It’s Always Sunny in the Sun Belt), but I think you get the message: there are scores of U.S. towns and cities whose population, human capital reserves, and economic prospects hemorrhaged for decades. Some of these places are referred to as “Legacy Cities“. Others are simply left for dead.

When we think about where these declining cities are generally located, our eyes gravitate “up and to the right” on the U.S. map. The following maps illustrate the spatial distribution of MSA and city growth and decline from 1970 to 2016.

MSA Pop. Chg. Map
(click for interactive visualization)
City Pop. Chg. Map
(click for interactive visualization)

Indeed, the “top-right” corner (the Northeast and Midwest) is where we see the most pink and red dots (less so with MSAs). Conversely, cities south of the Ohio River and west of the Mississippi River are spotted with blessed shades of blue. However, I think it’s wrong for the conversation to stop there.

A Quick Case Study

Let’s look at Syracuse and Rochester. Both cities have seen considerable population loss over the past four decades, yet the Metropolitan Statistical Areas (MSAs) to which they belong have grown!

Upstate, NY Populations.png
(click for interactive visualization)

What’s going on here? From 1970 to 1980, the City of Rochester declined by over 18% while the Rochester MSA grew by a measly one percent. This one seems straightforward enough: people were simply moving from the city to the suburbs.

However, in the next thirty years—1980 to 2010—the city’s population shrank by around 31 thousand while the region gained a whopping 108 thousand new residents! How can we explain this? Again, the short answer might involve another conversation about suburban expansion, yet the discussion cannot end there. We need to pay attention to the “stickiness of place” and carefully consider the factors that keep Rochester (the region) on the map despite Rochester (the city) declining so steadily.

To do this, we have to identify what put Rochester on the map in the first place. In the early 1800s, Rochester earned the moniker of “Flour City” for its world-class wheat production. This was followed by the nickname “Flower City” for its abundance of plant nurseries. Even though those industries eventually migrated elsewhere, the Rochester region has indelibly become most associated with precision manufacturing. Companies like Bausch and Lomb, Kodak Eastman, and Xerox were all founded in Rochester, creating a nationally prominent hub for the optical imaging industry.

Despite Rochester’s continued urban decline and the relocation of these companies’ corporate headquarters, the region still holds a disproportionally large share of employment in the optical imaging and computer manufacturing industries. Why is this? According to one study, “the ultimate source of regional resilience is not just the industrial legacy but the surviving industrial expertise of the regional workforce.” In other words, a region’s ability to retain its tacit, industry-specific knowledge and expertise what will buffer it against massive structural changes (such as globalization and technology). Furthermore, large research universities, such as the University of Rochester, can partner with local R&D firms to act as a further safeguard against regional decline.

Quadrants of Aberration

Below is a two-by-two matrix upon which the 1970-2016 growth rates of MSAs and their principal cities are plotted. The upper-right and bottom-left quadrants represent places where the city and MSA mirrored one another…growth-growth or decline-decline. More puzzling are the upper-left and bottom-right quadrants, in which the MSA and city behaved opposite of one another during that time period.

Pop. Chg. Constrained
(click for interactive visualization)

NOTE: This graphic constrains the axes to a maximum of 100% in order to preserve visual clarity. To see an unconstrained graphic, click here

Since 1970, there have only been two MSAs—Battle Creek, MI and Kokomo, IN—where the principal city declined and the MSA grew. These are the “aberrant of the aberrant,” and I have not yet delved into what might have caused such a combination of urban growth and regional shrinkage.

The more prevalent of the two is the lower-right quadrant, where MSAs grew and their principal cities shrunk. In addition to Rochester and Syracuse, some of the notable members of this quadrant are:

MSA Name 1970-2016 MSA Pop. 1970-2016 City Pop. 
St. Louis, MO-IL 11% -50%
Cincinnati-Middletown, OH-KY-IN 28% -34%
Hartford-West Hartford-East Hartford, CT 17% -22%
Chicago-Naperville-Joliet, IL-IN-WI 21% -20%
Philadelphia-Camden-Wilmington, PA-NJ-DE-MD 14% -20%
Spartanburg, SC 89% -15%
Washington-Arlington-Alexandria, DC-VA-MD-WV 95% -10%

In this list we see two extremes. In St. Louis, we see a city whose tremendous population loss has occurred alongside modest growth of the overall region. In Washington D.C., we see a metropolitan area whose immense regional growth came at the cost of moderate decline in the city-proper.  Perhaps the ‘burbs are just a fact of (U.S.) life.

The Urban Footprint

Another way of looking at the city-suburb continuum is to measure the principal city’s footprint in the MSA. The following maps show how that footprint changed between 1970 and 2016.

The first map shows the level by which an MSA’s population was supplied by its principal city. For example, 149,518 of the Lincoln, Nebraska MSA’s 182,432 residents lived in Lincoln, a value of 82%.

City Pop. Pct. 70
(click for interactive visualization)

The next map shows the same information for the [relatively] present day. In 2016, 280k of the Lincoln, Nebraska MSA’s 327k residents lived in Lincoln, a value of 86%.

City Pop. Pct. 16
(click for interactive visualization)

The final map shows the magnitude at which these two values have changed between 1970 and 2016. The Lincoln MSA example’s change from 82% to 86% represents a growth of 4.6% from 1970 to 2016.

City Pop. Pct. CHG
(click for interactive visualization)


For as long as I maintain this blog, I will never tire of thinking about the complex relationship between U.S. cities and their suburbs. Suffice it to say (for now) that the fate of one is not necessarily always related to the fate of the other.

In a 1992 study, economist Edwin Mills concluded that “central city problems have little effect on suburbanization of population and employment.” Perhaps this seeming “inevitability” is a partial explanation for the barrenness of the upper-left quadrant and plenitude of the lower-right. Furthermore, it might help explain why the third “footprint” map is mostly red (indicating declining share of the MSA population held by cities).

While there are very few cities—okay, only two—in which urban growth was accompanied by regional decline, there are plenty of metropolitan areas who did not follow their principal city’s pattern of decline. For these regions, it is important to remember the example of Rochester, in which regional resilience can be achieved through maintaining a legacy of technical expertise (the kind of “know-how” that can’t be easily transferred or transported) and fostering collaboration between local research institutions.

Finally, for anyone who though this post was “too much breadth, not enough depth,” don’t worry…I’m just getting started.

[cover image credit: Jan Liesegang]

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s