Geographers Study Chicago

When the AAG returns to Chicago in 2015 for its third annual meeting in 20 years, participants will be revisiting the site of a lively field of inquiry for the study of cities.  Beginning with the famed “Chicago school sociologists” of the 1920s, the city has served as a model of urban form and function for generations of social scientists.  In the words of geographer Brian J.L. Berry, Chicago was “the classic laboratory for urban analysis in the United States.”  In 1963 Berry and his students, James W. Simmons and Robert J. Tennant, published “Urban Population Densities: Structure and Change” (Geographical Review, Vol. 53, pp. 389-405), a penetrating study of population density patterns found in Chicago and in cities around the world.

The focus of their study was “a single empirically derived expression

dx = d0 e-bx

“where dx is population density at a distance x from the city center, d0 is central density and b is the density gradient [p. 389].”  Variations on this expression had already appeared in dozens of previous studies they cited.  It would be modified and tested repeatedly in the years that followed and become a staple of geography textbooks.

Berry and his students had already gone well beyond the simple model.  They presented evidence showing that the gradient value, b, typically falls over time in Western cities, presumably as a function of urban growth.  They also proposed that the central density, d0 , typically increases then falls, but they presented no systematic argument as to why this should be so.  They concluded (p. 404) “regardless of time or place,” the model provided “a statistically significant fit to the distribution of population density within cities.”

How has the urban density model fared as a predictor in the decades since their paper appeared?   When they wrote,  the city of Chicago had just reached its maximum population to date  of 3.6 million.  With occasional decadal increases, the city’s population has fallen to around 2.7 million at present.  While such a decrease (25%) might be consistent with the urban density model, given the prediction that central densities fall over time, the behavior of the density gradient itself is difficult to reconcile with their model.


The density curve has taken on a very different shape during the past six decades.  In 1950 peak population densities averaging 35,000 persons per square mile were observed roughly 3 miles from downtown.  The “density crater” at the very center of the city, where land was devoted more to commercial, office, and transportation uses than to residences, was an assumed correction to be made to the model.   But beyond that, density decreased at a fairly uniform rate in 1950.

In 2010, the pattern of population densities beyond 8 miles from the city center is almost identical to what it was sixty years earlier.  But a massive reduction in population has taken place within 8 miles of downtown.  While many census tracts on the north side of the city grew substantially over this period, the typical census tract south of downtown Chicago declined.  This “unbalancing” of Chicago, with a shift of population, resources, and employment opportunities  to the north, away from the south side, has been a hallmark of the six decades leading to the present.

The heart of the city’s south side, bounded roughly by Roosevelt Road on the north, 79th St. on the south, and Western Avenue on the west, had a population of 800,000 in 1950, which was 45% African-American.  In 2010 that same area contained fewer than 500,000 inhabitants, 58% of whom were African-American.

High-density living, once a hallmark of Chicago’s many neighborhoods, African-American as well as Euro-American, declined markedly during the 60-year period.  In 1950, 2.8 million people lived in census tracts that had a density greater than 15,000 persons per square mile.  In 2010, only 1.69 million persons lived in census tracts with a density at least that high, a reduction of roughly 40%.   Living in more compact, denser environments, while often promoted as a characteristic of sustainable cities, has declined drastically in Chicago during the past six decades.

Did the urban population density model studied by Berry and his students fail the test of time?  The gradient is still there, out toward the edges of the city, and it has scarcely changed at all in six decades.   A reduction in central density accompanying depopulation of the inner city was not envisioned in their model, nor were the factors that brought it about.    Chicago’s experience offers good evidence that cities remain entities that merit the geographer’s close attention.

—John C. Hudson
Northwestern University

DOI: 10.14433/2014.0013