What causes some cities to grow while others decline?
There are multiple answers to this question: long-term population shifts, i.e., Rust Belt to Sun Belt; loss of headquarter companies; loss or gain from industry; preponderance of higher education opportunities, etc. Generally, a city declines in population due to a multiplication of factors, while a similar city prospers due to a different mix of the same factors.
Boston is a good example. In the 1990s, four major high-tech firms in the city’s metro area — Digital General, Data General, Wang and Apollo — failed, yet the city has maintained relative population consistency and prosperity. A couple of reasons for Boston’s good fortune: a huge university infrastructure and a dense, thriving downtown.
For decades after the decline of the steel industry, Pittsburgh was a city in a population freefall, but unlike other, older industrial cities, it has finally arrested the meltdown and stabilized once again. Part of the reason: Pittsburgh, too, has a very good university infrastructure, which like Boston, has created a fulcrum of employment in particular fields, i.e., health services.
One difference between Boston and Pittsburgh is that the former city has been building on its intellectual resources for a longer period of time. It’s only been in the past decade that Pittsburgh stopped losing so many of its university graduates.
Over the last 60 years, all metro areas have had to contend with a shifting population, mostly further away from the central business district to the suburbs, then exurbs. Over the past two decades, some cities have experienced a reverse migration as young adults and even young families found inner-city living more to their liking. Cities such as Denver and Seattle have been successful in attracting population back to the downtown.
Even the Phoenix metro, notorious for its massive sprawl, has seen a kind of redensification of its central business districts. I’ve lived in the Valley of the Sun, the colloquial name for the Phoenix-Mesa-Scottsdale metroplex, for more than three decades and the central business district (CBD) of Phoenix is the healthiest I’ve ever seen it.
In the modern world, a healthy metro area attains a kind of balance between the CBD and its suburbs. But, what happens when a city loses the balance, i.e., when population density declines at the core and there is no stabilization?
Kyle Fee and Daniel Hartley, two research specialists working for the Federal Reserve Bank in Cleveland, had been watching their city’s population drift downward for years and decided to undertake a study basically trying to understand whether density loss presupposes full economic decline.
Cleveland’s population peaked in 1950 at 914,808. In 2010, the U.S. census reported the city’s population had fallen to 396,815, a level not seen since the end of 1800s. Yet, the full Cleveland metro area was still fairly strong at 2.25 million people.
Fee and Hartley’s study looked at how population density in major MSAs (metropolitan statistical areas) grew or declined over recent 10-year blocks, essentially during the 1980s, 1990s and 2000s. Then it scrutinized how population density changed for growing and declining cities as a function of distance from city center.
The major conclusion from the study showed growing cities tended to be increasing population density near the center and further out, but the declining cities, especially during the 1980s and 1990s, were losing population at the core. In the 2000s, the effects seem to be somewhat mitigated by the recession.
Some highlights from the Fee and Hartley study:
- The peak increase in population density in MSAs that were growing during the 1980s occurred 10 miles from the central business district. In the 1990s, this pattern was even more pronounced.
- In the 2000s, for cities that were growing, the biggest increase in population density was near the CBD, while there was smaller growth in population density further away from the CBD. One guess is that this was due to gentrification and redevelopment of neighborhoods closer to the city center.
- For cities that were shrinking, the 1980s showed a big drop in population near the CBD. Surprisingly, during that decade, at distances between 20 and 30 miles from the CBD, population density was increasing during this same period. As Fee and Hartley note, "This pattern is consistent with home-buying habits where more affluent households upgraded to larger and new housing farther from the center of the city, while less affluent households took up the housing left behind."
- For cities that were shrinking, the 1990s showed the "biggest" loss of population density close to the CBD. Meanwhile, at distances farther than 20 miles, there was no longer any increase in population density. The change in population density was essentially zero.
Fee and Hartley suggest population density is correlated with productivity in general, and there was a study in the mid-1990s for the American Economic Review that reports a relation between county employment density and productivity at the state level, which, in some ways, illustrates how Boston was able to withstand a loss of four key employers: because there was a strong density in population and a similar dense technology employment cluster 10-20 miles outside the CBD. There was, of course, also the education factor.
However, Hartley does concede, "We still don’t know if density, particularly at the core, is important to the whole metro productivity numbers."
So what should we take away from Fee and Hartley’s study? The most important point, says Hartley, "is that certain cities that had been losing density happen to be the same MSAs that have been declining in population over a long period of time."
That’s actually very important for politicians and urban planners to note; if you don’t take care of the CBD, it doesn’t bode well for the future of your city.