A new study suggests that racial sorting might have occurred even without the discriminatory housing policies we know bolstered segregation.
We already know that federal housing policies such as redlining, real estate practices like blockbusting, and the resulting “white flight” to suburbs in post-World War America caused cities to segregate, sealing generations of African Americans into poverty.
But even if these discriminatory policies hadn’t existed, the fate of our cities may not have been all that different, a new working paper published in the National Bureau of Economic Research suggests.
Economists Allison Shertzer and Randall P. Walsh at the University of Pittsburgh analyzed data from 10 large U.S. cities in the Northeast and the Midwest from 1900 to 1930 to isolate the role of white flight that occurred in that period—before the Federal Housing Authority, which instituted many of the discriminatory housing policies, was born. They found that the exodus of white people from a particular neighborhood following the arrival of black residents led to a 34 percent increase in segregation during the 1910s; In the 1920s, it resulted in a striking 50 percent increase.
Here’s how the authors summarize their findings in the NBER paper:
Our finding that sorting by whites out of neighborhoods with growing black population was a quantitatively important phenomenon decades before the postwar opening of the suburbs is novel. This analysis suggests that segregation would likely have arisen even without the presence of discriminatory institutions as a direct consequence of the widespread and decentralized relocation decisions of white individuals. Whites could simply have responded to policies that reduced barriers to black settlement in their vicinity by accelerating their departure for neighborhoods at lower risk of “encroachment.”
The authors chose to focus on the first three decades of the 20th century because this period saw a very sharp rise in segregation in the country. The chart below shows the increase in the dissimilarity index and the isolation index—two measures of racial segregation—for the 10 largest U.S. cities during this period:
They isolated demographic data for 10 U.S. cities—New York, Philadelphia, Pittsburgh, Boston, and Baltimore in the Northeast, and Cincinnati, Detroit, Chicago, Cleveland, and St. Louis in the Midwest—all of which had seen large influxes in black residents as a result of the Great Migration. They then designed a strategy to quantify the contribution of white flight to racial segregation. Here is how they explain it in the paper:
We identify patterns in the data that are consistent with the neighborhood “tipping" model of racial dynamics as first proposed by Thomas Shelling (1971). Taking this initial evidence as suggestive, we then adopt a more formal empirical strategy that identifies the causal link between black in-migration and white flight.
One important thing to note: when white people left their neighborhoods in response to black arrivals in this period, they didn’t go to the suburbs—because suburbs didn’t really exist until the second half of the 20th century. They went to neighborhoods pretty similar to the ones they left—at least in terms of tax bases and public spending. That means that the measurements of white flight here “may thus provide a better gauge of racial distaste than those using postwar data,” the authors write in the paper.
The paper’s conclusions by no means take away from the fact that the segregation of U.S. cities was a result of racist institutions, because those were put in place long before the 20th century. This is also not to say that the many discriminatory policies of the 20th century (many of which, Incredibly, persist today) shouldn’t be tackled. They should, and their lasting effects should be aggressively mitigated. This is what the paper is arguing:
Policies that reduce barriers faced by blacks in the housing market may thus not prevent or reverse segregation as long as white households have the ability and desire to avoid black neighbors.
In other words, they may not be enough.