House under construction

Housing Perspectives

Research, trends, and perspective from the Harvard Joint Center for Housing Studies

Reconciling the Back-to-the-City Thesis with Sustained Suburban Growth

Are cities growing faster than their suburbs? Unfortunately, there is little consensus among urban experts on this question. Some have argued that a “back-to-the-city movement” is bringing back America’s cities. Others have disagreed, pointing instead to more rapid population growth in the suburbs. In a previous article in this blog, Rachel Drew suggested that this disagreement was, at least partially, due to the differences in analysts’ definitions of city vs. suburban boundaries.

In a new paper published in Urban Affairs Review, “Are Millenials Coming to Town? The Determinants of Location Choice of Young Adults,” I explore this issue by developing a straightforward but precise framework for describing differences in population growth between central cities and suburban areas. While the paper finds some evidence for the back-to-the city thesis, it also suggests solid population growth in suburban areas over several decades.

Instead of selecting boundaries for a dichotomous measure of cities vs. suburbs (an approach used by many analysts), I used the distance from the central business districts (CBDs) of the 50 largest U.S. metropolitan areas to analyze demographic shifts in three periods: 1990 to 2000, 2000 to 2010, and 2010 to 2011–2015 multiyear estimates (hereafter, 1990s, 2000s, and 2010s, respectively). More precisely, I converted block-level census data into 75 rings in one-mile increments. This approach allowed me to describe differences in population growth between city centers and increasingly distant inner and outer suburban areas.

I find that, throughout the periods, middle-ring suburban areas grew faster than the nation as a whole. Specifically, suburban areas located between 10 and 35 miles of the CBD grew by 1.8, 1.4, and 1.2 percent per year in 1990s, 2000s, and 2010s, respectively, when the nation as a whole grew by 1.2, 0.9, and 0.8 percent annually. Thus, suburban areas have continuously attracted a greater share of the population during the period (Figure 1).

Figure 1. 

Annual Population Growth Rates by Distance from City Center




The patterns are different for downtown and inner city areas within ten miles of the CBD. Compared to the national averages, the immediate city centers within one mile of the CBD had annual growth rates that were slightly lower in the 1990s (0.9 percent) and the 2010s (0.5 percent) but considerably higher in the 2000s (1.6 percent). The popularity of downtown areas was more pronounced among young adults aged 25-to-34 years old, especially those who were white, well-educated, and earning higher incomes.

In contrast, inner ring areas between three and ten miles from the CBD became ‘dead-zones’ in the 1990s and 2000s, when the share of the population living in those areas generally fell. (In some inner-ring areas, the total number of people fell as well). However, the patterns dramatically changed after 2010. With the exception of places located zero-to-one mile and two-to-three miles from the CBD, all areas within a 35-mile radius from the CBD grew faster than the national average of 0.8 percent in 2010s.

These findings help to reconcile the differing conclusions of the back-to-the-city and suburban growth theses. On the one hand, they make it clear that during the 1990s and 2000s, population grew very rapidly in city centers, sometimes even faster than national averages, but only – as those reporting the back-to-the-city theses tended to do – if cities are defined as immediate downtown areas.

On the other hand, the data also show that during the 1990s and 2000s, the growth within these close-in areas was offset by smaller rates of, or even negative, growth in the inner-ring areas located three-to-ten miles from the CBD. Consequently, analyses that used broader boundaries for the urban/suburban divide – such as principal cities or core counties – produced strikingly different results.

Notably, the annualized growth rate of the area within a 10-mile radius of the CBD (which is roughly comparable to analyses that used “central cities” to differentiate urban and suburban growth) was just 0.6 percent in the 1990s and 0.2 percent in the 2000s. Since this was significantly smaller than growth in the areas located 10-to-45 miles from the CBD, this approach tends to support the conclusion that America continues to suburbanize at a steady rate. This approach produces a more nuanced story for the 2010s when the gap between the two areas closed. As noted earlier, this was largely due to more rapid growth in areas located three-to-ten miles from the CBD (Figure 2).

Figure 2. 

Annual Growth Rates, by varying radius from CBD



10 mi

While these patterns offer some evidence to support the “back-to-the-city” thesis, especially within very urban cores, they also reveal the extent and speed of suburban growth. Most notably, the population growth in areas 10-35 miles from the CBD outpaced the national average in all three decades. Moreover, in 2015, these areas were home to 106.3 million people (33.6 percent of the U.S. population), more than double the 48.6 million people (15.3 percent) living in the 3-10 mile bands, and substantially larger than the 9.1 million people (2.9 percent) living in the zero-to-three mile bands. Taken together, these trends both confirm the presence of the back-to-the-cities trend and offer perspective about its size relative to suburban growth since 1990.