Political polarization in the United States has grown substantially over the past several decades, whether you measure it through how legislators vote, how voters feel about the opposing party, or where people choose to live. The trend is not a media invention or a matter of perception. Voting records, survey data, and geographic patterns all point in the same direction, though the forces driving the widening gap are more tangled than most commentary suggests.
What the Legislative Record Shows
The most straightforward way to track polarization over time is to look at what elected officials actually do. When researchers analyze roll-call votes in the U.S. House and Senate, a clear pattern emerges: the gap between the two parties’ voting behavior is wider today than at any point in the past three decades, and by most measures wider than at any point since Reconstruction-era Congresses in the late 1800s.1PubMed Central. The evolution of polarization in the legislative branch of government The era of routine bipartisan coalitions that characterized much of the mid-twentieth century, when conservative Southern Democrats voted with Republicans and liberal Northeastern Republicans voted with Democrats, has largely vanished.
Voting records tell only part of the story. Bill co-sponsorship data, which reflects whom legislators are willing to collaborate with before a bill ever reaches the floor, confirms the same trajectory. An analysis of co-sponsorship networks from 1973 through 2016 found that both the House and Senate show rising “weak” polarization, meaning the two parties increasingly avoid working together, and rising “strong” polarization, meaning legislators are forming tighter, more exclusive clusters within their own party.2Social Networks. A sign of the times? Weak and strong polarization in the U.S. Congress, 1973–2016 In other words, it is not just that Democrats and Republicans disagree more. They have also stopped interacting across party lines in the routine legislative work that used to hold institutions together.
Affective Polarization and the Power of Partisan Identity
Among ordinary voters, the story is a bit different from what happens on Capitol Hill. Americans have not moved as far apart on specific policy questions as their representatives have. Instead, the sharpest increase has been in what researchers call affective polarization: the degree to which people dislike, distrust, and feel cold toward members of the other party. Surveys asking respondents to rate the opposing party on a “feeling thermometer” show a dramatic cooling over the past few decades. People increasingly describe those on the other side as closed-minded, dishonest, or even a threat to the country.
What drives that hostility? Research disentangling policy disagreement from party identity found that partisan identity itself is the principal engine of affective polarization, and that policy preferences factor in largely because they signal which team someone belongs to. At the same time, genuine policy disagreement does independently drive how people feel about each other, so it is not purely tribal. The overall picture, though, is that partisanship functions more like an emotional attachment than a rational scorecard of issue positions.3American Journal of Political Science. The Nature of Affective Polarization: Disentangling Policy Disagreement from Partisan Identity This helps explain why two people who actually agree on most policies can still feel intense animosity if they identify with opposing parties.
Geographic and Social Sorting
One reason polarization feels so entrenched is that Americans are increasingly sorted, not just in their beliefs but in where they live. Within states, partisan geographic sorting roughly quintupled between 1976 and 2016, meaning that counties and precincts have become far more politically homogeneous. That within-state sorting measure tracks almost perfectly with the well-known index of congressional polarization, with a correlation of about 0.95.4Journal of Public Economics. Partisan spatial sorting in the United States: A theoretical and empirical overview Whether people are self-selecting into like-minded communities or whether shifting local economies and demographics are producing the pattern remains debated, but the result is the same: your neighbors are more likely to vote the way you do than they were a generation ago.
Sorting extends beyond geography. Religious identity, for instance, has increasingly mapped onto partisanship. Research tracking the period from 1984 to 2016 found that religious belief, belonging, and behavior all constrain how tightly people link their ideological identity to their partisan identity. The effect is particularly pronounced among evangelicals, whose affiliation acts as a primary conduit through which religiosity shapes partisan sorting, and the impact has been asymmetric, producing stronger sorting among those with right-leaning identities.5Social Science Quarterly. Religion and Partisan‐Ideological Sorting, 1984–2016 The upshot is that political identity now overlaps more heavily with religious identity, neighborhood identity, and cultural identity than it did in previous decades. When so many aspects of life align along the same partisan axis, the stakes of political disagreement feel much higher.
How Media Fragmentation Widened the Gap
A generation ago, the big three broadcast networks served as a rough common baseline for political information. That baseline has eroded. An analysis of news coverage over time found that while broadcast television news still covers similar topics with similar language across outlets, cable news stations have become increasingly distinct from broadcast news and from each other, diverging in both the subjects they cover and the language they use to cover them.6Scientific Reports. Unpacking media bias in the growing divide between cable and network news A viewer watching one cable channel and a viewer watching another may end up not just with different opinions but with different factual universes.
Social media has added another layer. A recent large-scale experiment used an algorithm to rerank posts that expressed partisan animosity and antidemocratic attitudes in users’ feeds. Decreasing or increasing exposure to that kind of content shifted how people felt about the opposing party by more than two points on a 100-point feeling thermometer, with the effect appearing equally on both sides of the political spectrum.7PubMed. Reranking partisan animosity in algorithmic social media feeds alters affective polarization Two points may sound modest, but the experiment demonstrates a causal link: the content you see in your feed doesn’t just reflect your feelings about the other side, it actively shapes them. Multiply that effect across billions of daily interactions and years of cumulative exposure, and the algorithmic sorting of information becomes a meaningful contributor to the polarization trend.
Is the Trend Symmetric?
A common question is whether both parties have moved equally toward their respective poles. The evidence suggests they have not, at least not at the elite level. A modeling study of the self-reinforcing dynamics of legislative polarization found that subtle differences in how often and how sharply public opinion shifts left versus right over time have a differential effect on elite behavior, causing Republicans to polarize more quickly than Democrats. The researchers describe a critical threshold of self-reinforcement beyond which polarization accelerates. Republicans appear to have crossed that threshold, while Democrats are approaching it.8PubMed Central. The nonlinear feedback dynamics of asymmetric political polarization
This does not mean Democrats have stayed still. Both parties’ officeholders have moved away from the center compared to mid-century norms. But the magnitude of the shift is not equal, and the mechanisms driving it differ. Primary election dynamics, the structure of party coalitions, and the feedback between media ecosystems and voter preferences all contribute to the asymmetry. The religious sorting described earlier, which disproportionately affects right-leaning identities, is one piece of that puzzle.
Income Inequality and the Feedback Loop
Polarization does not happen in an economic vacuum. An analysis spanning nine decades of U.S. data found a bidirectional relationship between income inequality and political polarization: rising inequality predicts rising polarization, and rising polarization predicts rising inequality.9Review of Income and Wealth. Income Inequality and Political Polarization: Time Series Evidence Over Nine Decades The mechanism is intuitive once you see it. When economic gains concentrate at the top, different income groups develop more divergent political priorities. Polarized legislatures, in turn, find it harder to pass redistributive policy, which allows inequality to persist or grow. The two trends reinforce each other in a cycle that is easier to start than to stop.
The Urban-Rural Divide in International Context
It is tempting to treat extreme polarization as a uniquely American problem. It is not, though the U.S. is an outlier in degree. A comparative study of fifteen Western democracies over roughly five decades found that long-term increases in the urban-rural electoral divide have been most pronounced in the United States, the United Kingdom, and Canada. But the pattern has also emerged in several European multiparty systems in recent decades, largely driven by the growth of smaller parties that draw support primarily from either urban or rural areas. Overall urban-rural divides remain lower in those multiparty systems because mainstream parties with geographically diverse support bases still compete effectively.10Comparative Political Studies. The Great Global Divider? A Comparison of Urban-Rural Partisan Polarization in Western Democracies The implication is that electoral structure matters. A two-party system like the American one may be more vulnerable to geographic polarization because there is no middle-ground party to absorb voters who feel ill-served by either pole.
The Psychology That Keeps Us Stuck
Once polarization takes hold, psychological mechanisms make it self-sustaining. Motivated reasoning, the tendency to process information in ways that protect your existing beliefs, plays a central role. The intuitive assumption is that people who think harder about political news will be better at distinguishing true claims from false ones. But research on motivated reasoning suggests the opposite can happen: cognitive effort is often directed toward defending a prior identity rather than pursuing accuracy, so more thinking does not necessarily produce better judgment on politically charged topics.11SHAREOK. Motivated Reasoning and Cognitive Effort in Political News Information Processing
The bias starts even earlier than interpretation. When people are given the freedom to choose which information to look at, they reliably begin by sampling information about their own political group and continue to sample more of it overall. The resulting skewed information diet then interacts with motivated interpretation to produce evaluative biases. In a series of studies, researchers found this pattern held across different sampling environments: participants consistently over-sampled information about their ingroup, a tendency linked to group-based identity motives.12PsyArXiv. Partisan-Motivated Sampling: Re-Examining Politically Motivated Reasoning Across the Information Processing Stream You seek out confirming information, receive a lopsided picture, and then interpret what you see through a partisan lens. Each step compounds the last.
How Polarization Bleeds Into Everyday Life
Polarization is no longer confined to election season or cable news arguments. It has started showing up in hiring decisions. A field experiment in Sweden submitted over 11,000 fictitious job applications after the 2022 general election. Applicants who listed a political party affiliation on their résumé received a positive employer response about two percentage points less often than otherwise identical applicants who did not.13Political Psychology. Political discrimination in hiring: Evidence from a large field experiment A separate experimental study in France similarly found that political value similarity influenced hiring decisions, particularly when candidates held extreme rather than moderate affiliations.14Journal of Business and Psychology. Evaluating Potential Political Bias in Recruitment: Effects of Extreme vs. Moderate Party Affiliaton These are not enormous effects, but they represent a form of discrimination that barely existed as a research concern a generation ago.
Corporations have also been pulled into the polarization vortex. Firms increasingly face pressure to take public stances on politically charged issues, and those decisions carry financial consequences that depend on the polarization of their customer base. Research on corporate activism found that companies taking a liberal-leaning public stand tended to see a positive response, while conservative-leaning stands had a negative effect, though the audience’s own political orientation was a powerful moderator: liberal consumers responded more strongly and positively to corporate activism in general than conservative consumers did.15PubMed Central. Understanding reaction to corporate activism: The moderating role of polarization Meanwhile, firms in highly polarized environments were more likely to suspend and then resume political donations, suggesting that the corporate response to polarization is often tactical rather than principled.16Journal of International Money and Finance. Political polarization and corporate political advocacy
The Nationalization of Local Elections
Polarization at the national level has a gravitational pull on state and local politics. When the gap between the national parties is wide, state legislative election outcomes increasingly reflect how a state leans in national politics rather than the positions of the state-level candidates themselves.17American Politics Research. Polarization and the Nationalization of State Legislative Elections A moderate Republican running for a state house seat in a blue-leaning state faces headwinds that have little to do with their own platform. This nationalization makes it harder for local candidates to distinguish themselves and rewards those who most closely mirror the national party brand, which pushes state legislatures toward the same partisan extremes visible in Congress.
The Generational Ratchet
There is a generational dimension to polarization that rarely gets enough attention. Research tracking affective polarization across birth cohorts found that each new cohort enters the electorate more affectively polarized than the one before it.18Political Behavior. Affective Polarization: Over Time, Through the Generations, and During the Lifespan Young people today are not simply absorbing the polarization around them and gradually mellowing. They are arriving at voting age with a higher baseline of partisan hostility than previous generations did at the same age. As older, less-polarized cohorts leave the electorate and younger, more-polarized cohorts grow as a share of it, the aggregate trend has a built-in upward momentum that would persist even if no other driver were pushing polarization higher.
Why younger cohorts start more polarized is not entirely clear, but several of the mechanisms discussed earlier converge on this group. They are the first generations to come of age with algorithmically sorted social media feeds. They have grown up in an already highly sorted geographic and cultural landscape. And the partisan identity they formed in adolescence was forged in a more polarized environment than the one their parents entered as young adults. Each of these factors likely contributes to the ratchet effect.
What Might Actually Help
Given the self-reinforcing nature of the trend, the question of what could reverse or slow it matters enormously. A few approaches have shown promise, though none is a silver bullet.
Electoral reform is one area of active experimentation. Ranked-choice voting, in which voters rank candidates by preference rather than picking just one, is designed to reward candidates who can appeal beyond a narrow base. A legal and policy analysis argues that the current system of plurality-winner primaries actively incentivizes polarization, because candidates can win by energizing a small, passionate fraction of the electorate rather than building broad support.19University of Pittsburgh Law Review. Depolarizing America with Ranked-Choice Voting Alaska offers a real-world test case. After adopting a top-four open primary followed by a ranked-choice general election, the state saw winning candidates who were more likely to be centrists willing to work across the aisle and who espoused more moderate policy positions than those elected under the old system, despite more ideologically extreme candidates standing for election.20Journal of Political Institutions and Political Economy. Assessing Alaska’s Top-4 Primary and Ranked Choice Voting Electoral Reform: More Moderate Winners, More Moderate Policy
On the social media front, the algorithmic experiment described earlier suggests that relatively modest changes to what platforms surface in people’s feeds can shift affective polarization in measurable ways. The challenge is implementing such changes at scale when platforms’ business models often reward engagement, and partisan animosity tends to be engaging content.
Perspective-taking interventions have also been tested. One study found that simply exposing people to social media feeds from the opposing party increased engagement with that content but did not improve their understanding of why others hold different views. However, when the experience was framed in empathic terms, such as prompting participants to recall a personal disagreement with a friend before viewing opposing content, it boosted their ability to understand the other side’s reasoning.21arXiv. Perspective-taking to Reduce Affective Polarization on Social Media The distinction is subtle but important: raw exposure to opposing views without any empathic framing can actually backfire, while exposure paired with the right psychological scaffolding can help.
Brain Structure, Risk, and Political Orientation
A more speculative but intriguing line of research examines whether political orientation has biological correlates that might help explain why polarization is so persistent. Brain imaging studies have found structural and functional differences between people who identify as liberal and those who identify as conservative. One study of young adults found that greater liberalism was associated with increased gray matter volume in the anterior cingulate cortex, a region involved in conflict monitoring, while greater conservatism was associated with increased volume of the right amygdala, a region involved in processing emotional and threatening stimuli.22PubMed Central. Political Orientations Are Correlated with Brain Structure in Young Adults
A functional imaging study extended this finding by looking at brain activity during a risk-taking task. Although Democrats and Republicans took equivalent risks in the task itself, their brain activity differed: Democrats showed greater activity in the left insula, while Republicans showed greater activity in the right amygdala. A simple two-variable model using activation in those two regions actually predicted party identification more accurately than a well-established model based on parental socialization of party identity.23PubMed Central. Red brain, blue brain: evaluative processes differ in Democrats and Republicans These findings do not mean people are born into a party. Brain structure is shaped by experience, and the causal arrow could run in either direction or both. But they suggest that political orientation is intertwined with deep-seated differences in how people process uncertainty and threat, which may help explain why political disagreements feel so visceral and why persuading someone to change their mind is so much harder than it seems like it should be.