After Donald Trump prevailed over Hillary Clinton, many a think piece observed that there are two separate Americas: a conservative one located in red states and a liberal one located in blue states and cities. While Clinton won the popular vote, conservatives outnumber liberals in four out of five states. More than class or the culture wars, place itself is increasingly the critical fault line of American politics.
A new study puts an intriguing twist on that narrative. The study, which is co-authored by my colleague Matthew Feinberg at the University of Toronto’s Rotman School, finds that our political identification is not only shaped by where we live, it is relative to it. The labels “conservative” and “liberal” mean very different things in different places.
We know this intuitively: Someone who identifies as a moderate in a deep-blue Ithaca, New York could easily be to the left of someone who calls themselves liberal in small-town Texas, just as a self-identified conservative in Berkeley may be more liberal than a moderate Utah.
Many people feel pressured to conform to the political identity of the place where they live. But the key factor at work is what the study dubs the “political reference point”—a locally shaped gauge that people use to identify their own political leanings. Basically, if we live in a red place we may call ourselves moderate or even liberal just because our views are to the left of the prevailing conservative positions surrounding us. Similarly, blue-city-dwellers may think themselves moderate or even conservative just because their positions are right of many peers.
The study examines this relative effect of place on politics at the state level and the county level, looking at the relationship between our self-reported political identity and positions on different policy issues in light of the political tenor of the places we live.
At the state level, the study uses data from the American National Election Survey, which arrays political identity on a 7-point scale from “extremely liberal” to “extremely conservative.” The chart below shows the results of their analysis for 2012.
If political identity was the same across states, the lines would flat. The sloped lines indicate variations in the same political identity across states. The bluer the state, the more liberal the policy positions; the redder the state, the more conservative those positions are.
In other words, identifying as extremely conservative means something very different in Utah than it does in Hawaii. In Utah, extremely conservative people opposed abortion even in cases of rape; extremely conservative Hawaii residents were willing to consider legalizing abortion. As the study points out, “conservatives and moderates in blue states indicated more support for liberal policy positions than conservatives and moderates in red states, and the bluer the state was, the stronger their support was for liberal positions.”
Next, the study looks at the variation in political identity across counties. To get at this, the researchers collected their own survey data on political identity based on a 7-point scale (from “strong conservative” to “strong liberal”) and then across a ten-point scale (from “strongly oppose” to “strongly in favor”) on 10 key issues. The study polled people across the seven political identities in both blue and red counties to determine how identity aligned with issue positions, resulting in a sample of 1,269 people total.
In this graphic below, they present a sample of how political identities correspond with issue positions in different states. (Be warned: the graphic is flipped from the traditional “left-right” continuum.) The red Texas icon represents people in the 100 reddest counties in the country and the blue New York icon represents the 100 bluest counties in the country.
Here again we see that labels such as “strong conservative” and “strong liberal” are shaped by the political inclinations of the places people live. A strong conservative in a blue county registered less support for a strong military than a strong conservative in a red county, while a strong liberal in a red county had a more conservative position on the military than a strong liberal in a blue county. Indeed, moderates in blue counties effectively had the same views as strong liberals in the reddest ones.
These findings on the relativity of our political identities make the authors more optimistic about America’s political future. “[T]he animosity and disgust so commonly felt toward those on the other side of the political ideology spectrum may often be misplaced,” they write. “[I]f a person feels hatred toward others simply based on how they identify on the political ideology spectrum, then in some circumstances, that hatred is actually aimed at someone with the exact same policy stances.”
Indeed, they conclude, frequently “it is not the policy preferences or the values that differ between people, but simply the labels they give themselves—labels that shift depending on their political reference point.”
Our political differences, which have been so magnified by social media that if often seems as though Americans occupy two completely different worlds, may actually be less daunting than we think. At this fraught moment in American history, that would be heartening news.