Nothing of the Sort
For more than a decade, the idea that Americans are geographically sorting— increasingly choosing to live in neighborhoods populated with people just like themselves—has generated considerable attention in the media and continues to be cited as a very real phenomenon across the nation. In discussing the outcome of the 2020 elections, The Economist stated that Biden surged in some places because of sorting and approvingly held that there is a trend for liberals and conservatives to live among their own kind such that “Americans are still sorting themselves into politically like-minded communities.” You see it further in trend stories like this one from 2016 that say the election came down to “Whole Foods vs. Cracker Barrel.”
It makes for clever headlines, but it’s an argument that fails to hold up to empirical reality. Earlier research has already shown that geographic ideological segregation is actually lower than a generation ago, and thanks to data from the newly released National Community and Civic Life Survey at the Survey Center on American Life, the idea that Americans are choosing to live in completely like-minded enclaves does not hold up. In fact, numerous stories circulated this election cycle about neighborhoods with very diverse political leanings. A Wall Street Journal piece on neighbors in Pittsburgh attempted to paint a narrative of civil disagreement this past election, and during the 2016 election even The New Yorker wrote a longform piece on the fact that political life and neighborhood life are often entirely different worlds and should not be considered one and the same.
With that in mind, the first new finding of note takes aim at the idea presented in The Economist that liberals have moved to “diverse, densely populated cities; for conservatives it is places that are mostly white, working-class, and where the neighbors are a .22 round away.” Such an idea is not only incorrect, it perpetuates stereotypes.
The reality present in the national data is that in our nation’s cities urbanites may be politically sorted and thus divided by partisanship (68 percent Democrat vs. 28 percent Republican), but when it comes to ideology just 24 percent of urbanites identify as extremely liberal or liberal and 17 percent as extremely conservative or conservative. Although there are more liberals in cities compared to conservatives, neither ideological position is anywhere near a majority; centrists and moderates are the dominant position. Moreover, research has shown that social and friendship networks of Americans are socially segregated along several dimensions including politics. Partisans and ideologues are roughly equally likely to have social networks that reflect their own political predispositions—about three quarters of each group’s social network are liked minded politically—but moderates and centrists are diverse and split with a fairly equal mix of those on both the left and the right.
Moreover, in the suburbs, there is more political diversity than in cities, as 56 percent of suburbanites identify as Democrats and 42 percent as Republicans. Ideologically, identical numbers (21 percent) of those in suburbia identify as either liberal or conservative. Again, the dominant group (56 percent) are those who are moderate and middle of the road with slight ideologues.
In rural areas, Democrats make up 43 percent of the population and Republicans 51 percent. (You can see how this plays out in states like Montana, which has a Democratic governor, a divided congressional delegation, and voted for President Trump.) There are fewer liberals (11 percent) compared to conservatives (29 percent), but again this is anything but monolithically conservative with the overwhelming balance (56 percent) being centrist. None of these numbers get into landslide territory—60 percent or more of one ideology—for the left or the right and it is hard to argue that there is some geographic divide based on ideology.
Going further, in 2020 if one looks at the 350 largest metro areas that have densely populated urban cores, Biden won 41 percent of metros, compared to 59 percent for Trump. Now, Biden ultimately captured a greater share of the overall metro vote, winning 54 percent to Trump’s 44 percent—but neither presentation of the results shows a clear landslide liberal outcome. To be sure, while Trump lost to Biden in the largest cities, there are many slightly smaller urban area areas where Republicans won, and the data again support the simple statement that urban areas are not monolithically liberal and this ideological sorting has not happened.
The second new finding is that people do not significantly consider the politics of their neighbors when thinking about where to live. In the survey, respondents were asked, “Regardless of whether each of the following is available where you currently live, how important is it to you, personally, to live in a community…” and were presented with a series of 10 possible features, including being close to one’s extended family and a variety of entertainment options.
The results are striking. More than 80 percent of Americans believe that it is important or very important to live in an area with good public schools and easy access to parks. Three-quarters of Americans care about having amenities like restaurants nearby. Almost two-thirds care about being close to extended family and that an area has strong, local communal traditions. And more than half say that it is very or somewhat important to be in an area that has a mix of people with different socio-economic backgrounds and is racially and ethnically diverse.
When it comes to belief systems, however, the numbers look appreciably different. Barely a third affirm that it is very or even somewhat important to be in a neighborhood where most share your political (38 percent) or religious beliefs (34 percent). Put differently, of the 10 possible features that one could want in their community, only 7 percent of Americans—the smallest figure on the list—state that it is very important to live in a community with most who share their political views. Simply put, politics is not a significant focal point whatsoever when Americans are thinking about neighborhood and residential life.
It is worth noting that those on the political extremes do see the world a bit differently. Among moderates and those who are slightly liberal or conservative, barely 5 percent say that it is very important to be in an area where most people share their political views; 2 percent of those who are slightly conservative care about this. By contrast, 20 percent of extreme liberals and 18 percent of extreme conservatives care about the political leanings of their neighbors—a significant difference from centrists, but not even close to a majority even among more extreme Americans. When the somewhat important category is included, the figures are higher such that 29 percent of moderates care about the politics of their neighbors and about 60 percent of extreme liberals and conservatives care. But, again, these extremes really only account for perhaps 10 percent of Americans who are on the ideological extremes in general, and in the survey here they account for just 9 percent. Meanwhile, there is almost uniform high support across the ideological spectrum for parks and schools.
Finally, it is important to note that even with the stories of the 2020 election and neighbors dealing with disagreement, the fact of the matter is that most Americans do not know their neighbors. In 2018, Pew revealed that most Americans say they know at least some of their neighbors, but only about three-in-ten say they know all or most of them. Pew found that discussion with neighbors was low as well. For those Americans who know at least some of their neighbors, 25 percent stated that they had face-to-face conversations with them at least several times a week and another 24 percent said they had these conversations about once a week for 40 percent. Two years later, even in the midst of COVID-19 as of June 2020, the numbers were comparable; just 47 percent spoke with their neighbors on a weekly basis or more often even during a global pandemic which theoretically should bring people together.
The myth of the “Big Sort” needs to be shattered. The constant stream of news stories about Americans moving into likeminded enclaves does not hold up to empirical scrutiny and stories like that of Lehigh Valley, Pennsylvania where 1 in 3 don’t know the politics of folks next door do not make waves. Yes, there are congressional districts around the nation in which one party may not field competitive candidates and some that have been gerrymandered, but but the facts are clear: Cities are not monolithically liberal, rural areas are not the exclusive domain of gun enthusiasts, and most people do not think about the political orientation of a neighborhood when they think about living there at all—compared with amenities and communal traditions, politics is largely irrelevant.
Samuel J. Abrams is professor of politics at Sarah Lawrence College and a visiting scholar at the American Enterprise Institute.