Americans Think Their Neighbors Are Bad People

The moral contempt problem is worse than the trust problem. And it’s not even close.

Back in December, I wrote about the United States settling into a low-trust equilibrium and how that changes everything. The argument was structural: trust in government has ratcheted downward for decades, each crisis lowering the baseline, recoveries getting weaker each time. The country wasn’t experiencing a bad stretch. It was learning to operate inside a new normal.

That piece focused on institutional trust. Government. The state. The machinery of collective action.

Yesterday, Pew Research Center dropped something a lot uglier.

They surveyed adults in 25 countries and asked a simple question: Do the people in your country have good or bad morals and ethics? In nearly every country, majorities said their fellow citizens were morally good. Indonesians thought so. Kenyans thought so. Poles, Brazilians, Swedes, Indians. Even the French, who’ve elevated complaint to an art form, rated their compatriots as morally decent.

The United States was the only country where the number flipped. 53% of American adults said their fellow Americans have bad morals and ethics. Not “could do better.” Not “mixed.” Bad.

That’s not a trust problem. That’s a contempt problem.

Percentage of adults saying people in their country have good vs. bad morals and ethics. The United States is the only country surveyed where a majority says their fellow citizens are morally bad. Source: Pew Research Center, Spring 2025 Global Attitudes Survey

Source: https://www.pewresearch.org/religion/2026/03/05/in-25-country-survey-americans-especially-likely-to-view-fellow-citizens-as-morally-bad/

My earlier piece argued that low institutional trust warps political cognition. High-trust environments produce debates about effects: Will this work? Who benefits? Low-trust environments produce debates about motives: What are they really after? Who is this meant to punish?

All of that still holds. But the Pew data catches something the institutional frame misses. The contempt has migrated. It’s not aimed only at Congress or the courts or the executive branch anymore. It’s aimed at the people next door.

That’s a different animal. Distrusting your government is a political position. Concluding that your fellow citizens are morally deficient is closer to a civilizational verdict. It means you’ve stopped seeing the people around you as participants in a shared project who happen to disagree, and started seeing them as the reason the project is failing.

Political scientists call the milder version of this affective polarization: the tendency for partisans to dislike, distrust, and dehumanize members of the opposing party independent of any specific policy disagreement. It’s been climbing in the U.S. for decades. Pew’s own prior work shows growing numbers of both Republicans and Democrats describing people in the other party as immoral.

Growing shares of partisans describe members of the opposing party as immoral. Source: Pew Research Center, 2022.

Source: https://www.pewresearch.org/politics/2022/08/09/as-partisan-hostility-grows-signs-of-frustration-with-the-two-party-system/

The new data from yesterday takes it further. This isn’t just “I think Democrats are bad” or “I think Republicans are bad.” It’s “I think Americans are bad.” The target has generalized. The moral condemnation has leaked out of its partisan container and settled into the air everyone breathes. (And, well, that was in 2022.)

---

Here’s the part that should bother anyone who studies political cognition for a living.

Pew asked people in all 25 countries whether specific behaviors are morally unacceptable: affairs, marijuana, gambling, pornography, homosexuality, abortion, alcohol, divorce, contraception. If Americans held tighter moral standards across the board, you could explain the 53% as high expectations colliding with imperfect reality.

They don’t. On most questions, the U.S. lands in the middle of the pack. Only 23% call marijuana use morally wrong, well below the global median. Only 29% say the same about gambling. Americans are among the most accepting of those two behaviors. On homosexuality, abortion, drinking, the U.S. is neither especially conservative nor especially permissive. It sits in the fat middle of the distribution.

The one exception is extramarital affairs: 90% of Americans call it morally wrong, near the global top. But one item can’t carry the weight of the broader pattern.

So Americans aren’t unusually judgmental about behaviors. They’re unusually judgmental about people. They’ll shrug at marijuana and tolerate gambling, then look at their neighbors and conclude: bad morals.

That disconnect matters. It suggests the moral judgment isn’t primarily doing ethical work. It’s doing identity work. People aren’t cataloging their neighbors’ conduct and rendering a verdict. They’re projecting a generalized sense of moral failure onto a population they’ve already sorted into camps.

Often, the judgment lands first. The evidence gets recruited afterward.

That’s motivated reasoning operating at scale.

---

I’ve spent a good chunk of my career studying motivated reasoning, and what Pew captured here is textbook, of course. Motivated reasoning isn’t a failure of thinking. It’s thinking organized around a conclusion you’ve already reached. You don’t weigh evidence and arrive at a judgment. You start with the judgment, scan for confirmation, and experience the whole process as rational deliberation. It feels like analysis, but it functions like advocacy.

Applied to moral evaluation, this can produce a specific pathology. You don’t observe your fellow citizens’ behavior and then decide whether they’re good people. You know they’re good or bad based on tribal affiliation, and everything you observe gets filtered through that prior. The guy flying the wrong flag isn’t just wrong about politics. He’s a bad person. The woman posting the wrong opinions isn’t mistaken. She’s morally broken.

The Pew data maps this cleanly onto partisan identity. Democrats and Democratic-leaning independents rate fellow Americans as morally bad at 60%. Republicans and Republican leaners come in at 46%. In a vacuum, that looks like Democrats being more critical. But Pew finds the same dynamic in more than half of the 25 countries surveyed: people who don’t support the governing party judge their compatriots more harshly.

What’s unique to the U.S. is the altitude. Both sides are so high that the country tips negative overall. Other democracies have the same partisan tendency to project moral failure onto the out-group’s territory. But nowhere else does the effect push the national average past 50%.

America’s affective polarization appears to have reached a level where partisan moral contempt, aggregated across the whole population, yields a national majority that views its own citizenry as ethically compromised. No other surveyed democracy is there yet.

---

There’s an age dimension here that deserves separate attention.

Younger Americans, 18 to 39, are more likely than older adults to say their fellow citizens are morally bad: 57% versus 50%. Pew notes this holds even after controlling for party affiliation.

This tracks with something Pew documented in a May 2025 report on Americans’ trust in one another: each birth cohort entering adulthood since the 1960s has been less trusting than the one before it. Each recent birth cohort appears less trusting than the one before it. People born in the 1990s trust less than those born earlier, and those born in the 1970s trust less than older cohorts as well. The pattern is monotonic.

But there’s a distinction worth pulling apart. The standard trust question (”Can most people be trusted?”) captures something relatively instrumental. It’s about behavioral expectations. Will the stranger return my wallet? Will the contractor do honest work? The moral judgment question cuts deeper. It’s not asking whether people will behave well. It’s asking whether they are good.

Younger Americans are answering both questions in the negative. They trust less and they condemn more. Those are different postures with different consequences. Distrust makes you cautious. Moral contempt makes you hostile. Caution is adaptive. Hostility corrodes the connective tissue a democracy runs on.

The generational math is unforgiving. The most trusting cohorts are dying. The least trusting, most contemptuous cohorts are inheriting the electorate. This isn’t a phase. It’s a conveyor belt.

---

Let me tie this back to the framework I laid out back in December.

I argued that low-trust societies don’t collapse in dramatic fashion. They persist, sometimes for decades, while governance turns brittle and social cooperation quietly decays. The hollow state keeps its formal authority but loses the ability to command broad loyalty. Authority becomes performative. Citizens experience the state as a thing that constrains them, not a thing that coordinates them.

The moral contempt data puts a floor under that analysis, and the floor is lower than I thought.

In a low-trust environment where citizens doubt institutional competence, reform is at least conceptually possible. Better performance could restore some confidence. Accountability mechanisms could help. Competence, demonstrated visibly over time, could slowly rebuild legitimacy.

But when citizens believe their fellow citizens are morally bad, the reform logic collapses. Who runs the reformed institutions? Bad people. Who staffs the transparency commissions? Bad people. Who benefits from the accountability mechanisms? Bad people using them to punish other bad people.

Moral contempt undermines the cooperative premise that institutional repair depends on. You can’t rebuild trust between citizens and the state if the citizens don’t trust each other to act in good faith within any institutional structure you build. This is, at its core, a first-mover problem with no obvious first mover. Institutions can’t earn trust from a population that’s already written off the people who would staff them.

The trap works like this: institutional trust requires social trust. Social trust requires a baseline of moral regard. Moral regard requires either shared experience or, at minimum, the assumption that the other side is operating from defensible values even when you disagree with their conclusions. Affective polarization has dismantled that assumption brick by brick.

We’re not just in a low-trust equilibrium. We’re in a low-regard equilibrium.

---

I closed my piece in December by noting that the only way out is through, and that getting there means escaping the grip of our late postmodern moment. I was gesturing at something I planned to develop: the idea that the epistemic frameworks we’re living inside, the ones that treat every claim as a power move and every institution as a site of domination, have made it structurally difficult to rebuild shared commitment to anything.

The moral contempt data sharpens that considerably. Postmodern critique taught a generation to interrogate motives. Political polarization supplied permanent opponents whose motives needed interrogating. Social media built a frictionless infrastructure for performing that interrogation in public. And motivated reasoning, the engine underneath much of it, made that interrogation very likely to confirm what we already believed.

The result is a population that can’t accurately evaluate its own moral condition. There’s no good reason to think Americans are morally worse than Indonesians or Brazilians or Swedes. They’re people who’ve been soaking in a particular cocktail of political sorting, media fragmentation, and epistemological suspicion long enough that moral contempt has become the default setting.

Recognizing the motivated reasoning at the core of this is, at minimum, a starting condition for recovery. Not because self-awareness cures bias. It doesn’t. But because it reframes the problem. The question isn’t “Are Americans actually morally bad?” The question is “Why have Americans concluded their neighbors are morally bad, and what does that conclusion do to the system?”

The first part of the answer involves decades of political sorting, the collapse of shared media environments, and the weaponization of moral language for partisan ends.

The second part is simpler and grimmer: it makes everything harder. Every policy debate, every institutional reform, every collective action problem becomes a referendum on the moral character of the people involved rather than an assessment of the proposal on the table.

That’s where we are. Not just low trust. Low regard.

And the data, at least for now, suggests we’re the only surveyed democracy that has tipped into outright majority moral pessimism about its own people.

---

Also, if you enjoyed (hated) this piece, please restack or share it with someone who you think would appreciate (hate) it. Thanks. :)