An example of biased separation, among 180 million voters

The broad outlines of America’s partisan divisions are visible on any national map. Republicans typically dominate in most southern and plains states, and Democrats in the northeast and west coasts. Democrats cluster in urban America, Republicans in more rural places.

But keep zooming in – say at the level of individual addresses for 180 million registered voters – and this pattern is repeating itself: within metro areas, in provinces and cities, even within parts of the same city.

Democrats and Republicans live apart from each other, even in the neighborhood, to an extent that raises challenging questions about how closely lifestyle preferences have aligned with politics and how even neighbors can influence each other.

As new research shows, it is not just that many voters live in neighborhoods with few members of the opposite party; it is that almost all American voters live in communities where they are less likely to encounter people with conflicting politics than we would expect. This means, for example, that in an environment where Democrats make up 60 percent of the electorate, only 50 percent of a Republican’s closest neighbor can be Democrats.

According to Harvard researchers Jacob Brown and Ryan Enos, Democrats and Republicans are effectively separated from each other, to varying degrees. And at least for the past decade, they believe that this biased segregation is becoming more and more pronounced.

The maps above – and throughout this article – show their estimates of bias towards the individual voter, colored by the best guess of the researchers based on public information, such as the registration of voters, or the voters at the party’s pre-election and participated in the demographic composition of their environment.

We can not know how someone voted individually. But these maps show how Democrats and Republicans can live in many different places, even in the same city, in ways that go beyond the urban-suburban-rural patterns visible in the composite election results.

“We know that bad things happen to groups in general,” he said. Enos said. ‘This is true with racial segregation, and religious and ethnic divisions – segregation patterns that make it easier to demonize each other, and harder to share resources or power.

“The question with political parties is whether it is enough like the other groups to worry us about it.”

By living apart, opposing partisans can reject help for each other (with a term like ‘blue-state rescue operations’) or be more likely to buy myths about each other (such as widespread voter fraud). Other processes such as racial segregation, Mr. Enos added, have the tendency to accelerate.

This growing segregation does not necessarily mean that partisans search cities – or neighborhoods or even individual streets – where the neighbors are politically similar. Several forces pushed them apart, including broad changes in who the two parties represented. The closer we look between all these points, the more difficult it becomes to explain these patterns.

For each individual voter, linked to an address, the researchers looked at their thousand nearest voters and weighed those next door heavier than those a mile away. Shown in this way, about 25 million voters – mostly urban Democrats – live in neighborhoods where at most one in ten meetings are likely to be with someone from the opposite party. Democrats in parts of Columbus, Ohio and Oklahoma City live that way. So do Republicans in the reddest parts of Birmingham, Ala., And Gillette, Wyo.

Even when Democrats and Republicans are more equal to the same zip code or census, Mr. Brown and Mr. Enos still traces of segregation. This means that the two groups do not appear randomly confused. A Republican in a more mixed suburb of Denver is likely to live more close to other Republicans than would simply suggest.

“If we drop to a very low level and we still see that this sorting work is going on,” he said. Enos said, “it probably means something fundamental is going on here.”

Educational shift, geographical link

So, what can explain these patterns?

Over time, the Democratic Party increasingly aligned itself with urban voters, and the Republican Party with out-of-town voters, deepening geographic polarization nationally.

Highly educated white voters also moved to the Democrats as the working class white voters moved to Republicans. Educational realignment also has geographical consequences, with the changes concentrated in highly educated suburbs and more working-class towns and rural communities, respectively. None of these voters need to “sort” effectively on a map; rather, their preferences change (in ways that may appear in their voting behavior before voters update their party registration).

“Party coalitions have moved in a direction that really matches spatial differences in a way that they did not use,” said Greg Martin, a Stanford political scientist who also studied these trends.

Racial segregation also feeds partisan grouping, as African-American voters in particular are overwhelmingly democratic and also residentially segregated (the map of Metro Milwaukee’s biased segregation looks like its map of racial segregation). But Mr. Brown and Mr. Enos finds that racial segregation alone does not explain the levels of biased separation.

Lifestyle preferences that seem to have little to do with bias are also increasingly correlated with it. If you like city living and use transportation, you are more likely to be a Democrat; if you prefer large houses and pickups, you are more likely to be Republican. And so voters from the same party can choose to live in the same places for such functions, not necessarily to be each other, and this will cause partisan grouping.

Even if Republicans and Democrats live in the same city or in the same part of the city – essentially the same kind of place – they still seem somewhat separated from each other. Much of it is probably about housing. Even within the same census, there may be apartments with apartment buildings (probably the home of Democrats) and streets with single-family homes (more likely the home of Republicans).

But housing also cannot explain the full effect. Over 98 percent of the census treaties nationwide, Democrats and Republicans live with at least some segregation. This is true even in suburban neighborhoods southwest of Kansas City, Mo., where the residents are almost all white and homeowners, and the homes are all single-family homes.

This leaves a more interesting question: do people really pay attention to the politics of their neighbors and do they act on it in some way?

“I do think something new is happening at the neighborhood level around biased politics,” said Nancy Rosenblum, a political theorist at Harvard who wrote a book on neighbors. Interactions between neighbors have long been impartial, she said, based on values ​​such as reciprocity. I will borrow my leaf blower, look after my child.

But she fears a more malicious kind of politics is seeping into neighborhoods: ‘The most interesting question to ask here is: How deep does it go? And the test for how deep it goes for me is: how do neighbors in neighborhoods react during disasters? ”

This is when we normally see neighbors’ reciprocity really come through, she said. “If we look at Covid – and we see Covid as a national disaster – you see something change,” she said. ‘And it’s really very discouraging. It can make you cry. ‘

Now even masks are transported as biased signals.

Local influencers

There is little evidence that people choose where to live with politics in mind. Political scientists Clayton Nall and Jonathan Mummolo argued that other concerns take precedence, such as finding a cheap home and being close to a preferred school.

And neighborhoods carry just one piece out of anyone’s social circle with co-workers, friends and family, to say nothing of the political influences of biased news and social media.

But there is evidence that local environments matter. Mr. Brown, a doctoral candidate in government at Harvard, also looked at what happens to voters who eventually stay at the same address as the biased composition of the community around them shifts. As an environment becomes more democratic or republican over time, he finds that voters are more likely to change their party registration.

In a neighborhood that has gone from a little more Republican to a little more Democratic, for example, it increases the likelihood of a non-partisan voter registering as a Democrat by several percentage points. It is modest, but mr. Brown said it was “a significant change to something that is mostly fairly stable.”

He also finds in surveys that voters are more likely to show their bias – wearing clothes with a message, sticking out a garden sign or bumper sticker – when the people around them share their politics.

Other research shows that billboards can increase candidates’ voting shares, and that neighbors can influence political donations. Ricardo Perez-Truglia, a political economist at the University of California, Berkeley, found that people in politically minded areas gave more when he sent them a letter reminding them that their neighbors could look up their donations. And donors to Barack Obama in 2008 would probably make more money in 2012 if they moved to a heavier Democratic community in the intervening years.

“This is just one example of many other contexts in which it can take place,” Professor Perez-Truglia said. “If you know that everyone is from the same party, you have nothing to lose. You can be very vocal; no one is going to disagree with you. ”

Together, these studies suggest that people there are more likely to adapt and indicate their bias in public as places become more politically homogeneous. Maybe no one says, “I want to move here because of all these Biden yard boards.” But maybe one neighbor is admired by the people who set them up, and another neighbor decides, “This is not the place for me.”

Many biased signals are not so subtle these days. They come from Trump boards that are the size of billboards, which are proud even outside of election years. Other signs – ‘love is love’, ‘no man is illegal’, ‘science is real’ – implicitly blame anyone who does not share the values. And over time, it has become easier for entrepreneurs to create and market such messages, said Donald Green, a Columbia political scientist who studied the site signs.

“It’s very easy to know who’s who, ‘he said, describing the equally divided Hudson Valley community where he has a home, and where signs against a state arms control are common. ‘If you’re watching’ Repeal of the SAFE Act ‘- I’ve seen four driving to the Post Office just now – you know, you absolutely know. ”

It is also possible that biased segregation increases because these patterns feed each other. Voters with a similar taste in housing and now reassembled by parties are casually grouped into space, and they bump into each other as they go, in a kind of self-reinforcing cycle.