Recent Study Sheds New Light on Racial Segregation 

There Is More To Segregation Than Housing 

When talking about racial segregation, we often think of it in terms of where people live. But modern GPS technology has allowed scientists to look at segregation from a different perspective – questioning this old premise. Instead, new research from a Stanford team suggests that people’s daily experiences and interactions matter more than the nationality of their neighbors. This way of viewing segregation opens up for alternative solutions to a problem with complex roots in history, socioeconomics, and culture. 

In this article, we’ll look at the difference between residential and experienced segregation, the history of racial segregation in the US, and the Stanford study’s four main findings. 

Residential Segregation vs. Experienced Segregation 

Residential segregation is how we traditionally measure segregation. But where we live only tells part of the story. What if you live in one neighborhood but spend most of your days in another one? What are our daily habits’ societal implications, and how can this knowledge be used to create more integrated societies – faster?  

To answer these questions, we need a way to track not only where people live but also how and with whom people spend their days. This is precisely what researchers at Stanford have been trying to do. Economists Susan Athey and Matthew Gentzkow have used GPS data as a new way of measuring segregation. A method that hopefully can provide us with ideas on how to combat its harmful effects.  

What they aim to measure, as opposed to residential segregation, is “experienced isolation.” In their study, this is defined as people’s “exposure to diverse others in the places they visit over the course of their days.” The research is based on GPS data from a sample of US smartphone users, covering around 5% of the US population, during the first four months of 2017.  

A History of Racial Segregation 

America’s dark and sinister history of racial segregation still very much affects the urban landscape and the lives of the people who live there. In the early 1900s, racial housing covenants made it “impossible for people of color to buy homes freely. Covenants were embedded in property deeds all over the country, keeping anyone who was not white from buying or even occupying land.” This has been thoroughly documented in St. Louis; Seattle; Chicago; Hartford, Connecticut; Kansas City, and Washington DC. Sadly, the patterns of inequality that the covenants created persist.  

Racial segregation also correlates closely to income segregation. In Minneapolis, for example, where the tragic killing of George Floyd took place, the median Black family income in 2018 was $36,000, compared to nearly $83,000 among White families.  

The combination of historical and socioeconomic factors makes residential segregation challenging to tackle, and any changes in where people decide to live are by nature pretty slow. Add to this the tendency we as humans have to choose to live near those we find similar to ourselves and change becomes even slower. This tendency is described in the Schelling model, an agent-based model developed by economist Thomas Schelling. It does not include outside factors that pressure people to segregate (such as the covenants or the Jim Crow laws). Instead, Schelling’s work demonstrates how people’s “mild” in-group preferences towards their own group can lead to a highly segregated society when many individuals’ choices are aggregated.  

Considering the inertia that characterizes residential segregation, it becomes very interesting to look at segregation from other perspectives, as in the Stanford study. In their paper, Susan Athey and Matthew Gentzkow present four main results. 

#1 Experienced Segregation Lower Than Residential Segregation 

One of the most important findings in this paper is that people spend a notable amount of time outside the areas where they live. When doing so, the GPS data indicates that they’re likely to encounter a much greater diversity than the one in their own neighborhoods. Racial integration seems to be highest in entertainment, retail, and eating venues. Locations like churches and schools, on the other hand, are somewhat more racially isolated. 

#2 Experienced And Residential Segregation Correlate 

Nevertheless, the study finds that experienced isolation and residential segregation correlate to quite an extent. Both are generally higher in the South and the Great Lakes region while lower in the upper Midwest and Northwest. This leads us to result #3: that there seem to be specific population characteristics and types of cities that show less isolation and higher levels of exposure to diversity. 

#3 Public Transportation, Density, and Education Are Important Factors

The study shows that denser and wealthier cities with a higher education level are the ones where people’s experienced segregation is relatively lower. Another factor that plays in is to what extent people are using public transit. The higher the level of public transit use, the lower the level of experienced segregation. In other words, the more people mix and mingle because of their daily logistics, the lower the experienced isolation. This could be seen as a sign that by altering people’s habits through, for example, urban planning, experienced segregation could be affected.  

#4 People’s Movements Give Clues To What May Increase or Decrease Segregation 

By tracking when and where integrated interactions occur, the Stanford researchers hope to provide a helpful view on what efforts would be most efficient to combat segregation. Or, in their own words, “policies which affect the spatial distribution of commercial or leisure activities, or the transportation cost of accessing these activities, may be as or more effective than policies explicitly targeting housing,” as Athey and Gentzkow state in their paper.   

Implications For the Future 

There are of course limitations to the conclusions we can draw from the Stanford findings. One of the key limitations is that GPS data only captures exposure to other people, and not actual interactions between people. Is a city really more integrated because people mix on the subway? Still, without exposure there are definitely no interactions, so it makes sense to consider diversity in our daily environment a step in the right direction. 

And while creating a change in where people choose to live is a process that is inherently difficult and slow, these new findings show that other measures can have positive effects as well. Actions that may take much less time and effort, while bringing positive effects to people’s lives. All aspects of our daily lives matter, and not only where we live. Where we eat, work, study, and relax all affect how integrated we are. And these are behaviors that are easier to impact with policy decisions—an insight that shines a more hopeful light on the prospects of change.