The Young Trust Each Other Least. They’re Also the Only Ones Getting Better.

Ask Americans born before 1960 whether most people can be trusted, and 77% say yes. Ask Americans born in the 2000s, and 35% say yes.

That is the finding the Marquette Law School Poll published in July 2026, and it is one of the starkest generational divides in American public opinion. Every younger birth cohort trusts other people less than the one before it: 63% for those born in the 1960s, 55% for the 1970s, 45% for the 1980s, 43% for the 1990s. Marquette’s release lays out the full ladder.

It is tempting to stop there and declare that the young have given up on each other. That would be wrong, and the reason it is wrong is the most important part of this story.

The headline hides a recovery

Marquette also tracked overall trust over time. It was 55% in 2021, 55% in 2022, 54% in 2023, 52% in 2024, 51% in 2025 and 54% in 2026. That is not a free fall. It is a country that has held roughly steady for five years.

And the youngest adults, the group with the lowest trust, are the ones moving fastest in the right direction. Among Americans born in the 2000s, trust rose from 23% in 2021 to 35% in 2026. Twelve points in five years.

So two things are true at once. Young adults trust other people far less than their grandparents. And young adults are more trusting today than they were at the start of the decade. Any account of “declining social trust” that reports the first fact and buries the second is not analysis. It is a sales pitch for despair.

The question changes the answer

Social trust is one of the easiest numbers in the world to manipulate, because small changes in wording produce big changes in results.

The Pew Research Center tested this directly. When Americans chose between “most people can be trusted” and “most people cannot be trusted,” 55% picked the trusting option in February 2025, compared with 52% in 2018. No decline. When the alternative was “you can’t be too careful,” trust came out lower, because caution sounds reasonable in a way that distrust does not. On a four-point scale, about 60% said they trust people “a great deal” or “a fair amount” in both 2018 and 2025. Pew’s methodology report walks through each format.

Here is the point that matters: the level moves with the wording, but the pattern does not. Across every format, Pew found the same groups report lower trust: younger adults, women, racial and ethnic minorities, people with less education and people with lower incomes. That consistency is the real signal. Trust tracks vulnerability. People who have had more reasons to be let down report less confidence that others will treat them fairly.

That changes the moral question. “Why won’t people trust each other?” becomes “What have people experienced that taught them not to?”

Is it the phones?

Everyone has a favorite villain, and for social trust it is social media. Marquette tested it. Among heavy users of five or more platforms, 46% said most people can be trusted. Among minimal users, 59% did. A 13-point gap is real, but it is far smaller than the 42-point gap between the oldest and youngest cohorts, and Marquette found social media use had inconsistent effects within age groups. The phone is part of the story. It is not the whole story, and anyone who tells you otherwise is selling a simple answer to a complicated problem.

Time together is disappearing

If trust is built anywhere, it is built face to face, and the time Americans spend face to face has cratered.

The U.S. Surgeon General’s 2023 advisory on social connection, drawing on the American Time Use Survey, found that the average American’s in-person time with friends fell from about 60 minutes a day in 2003 to about 20 minutes a day in 2020. For people ages 15 to 24, it fell from roughly 150 minutes a day to 40, a drop of about 70%. Time spent alone rose over the same period. About half of U.S. adults reported experiencing loneliness. The advisory also cites research finding that a lack of social connection carries a mortality risk comparable to smoking up to 15 cigarettes a day. Read the Surgeon General’s advisory.

You cannot trust people you never spend time with. You cannot learn that strangers are mostly decent if you never have to depend on one.

The structure of daily life has shifted underneath us. The same advisory reports that single-person households rose from 13% of all U.S. households in 1960 to 29% in 2022, and that only 39% of U.S. adults said they felt “very connected” to others in a 2022 survey it cites. None of that is a character flaw. Living alone can be a choice, and a good one. But it means the ordinary, unplanned contact that used to build familiarity between neighbors now has to be chosen on purpose. If nobody chooses it, it does not happen, and trust has nothing to grow from.

But people still show up

Here is the counterweight, and it is substantial. According to the Census Bureau and AmeriCorps, 28.3% of Americans 16 and older formally volunteered through an organization between September 2022 and September 2023. That is about 75.7 million people giving an estimated 4.99 billion hours, worth more than $167 billion. It was a strong rebound from 23.2% in 2021, though still below 29.0% in 2019. Even more striking, 54.2% of Americans helped or exchanged favors with neighbors, up from 51.7% in 2019. Census Bureau summary.

There is a catch. The average volunteer gave 70 hours in 2023, down from 96.5 in 2017. More people are helping, but each is giving less. Participation is broad and thin. That is a pattern worth watching.

Where the obligation actually sits

This site is built on the idea that moral decline is real and measurable. That is exactly why we refuse to exaggerate it. If the youngest generation’s trust is climbing, we say so. A thesis that cannot survive good news is not a thesis. It is a mood.

The honest reading is this. Americans still help each other in huge numbers. But they spend less time together than any generation in modern memory, the young start adult life far more guarded than their parents did, and the people most exposed to exploitation trust the least, for reasons that are not irrational.

Trust is not something you can order people to feel. It is earned by the people and institutions asking for it. That means keeping promises that are specific, handling shared money in the open, applying the same rule to the boss and the new hire, and making it easy to raise a problem without being punished for it.

It also means something for you personally. Trust grows from repeated, low-stakes reliability: showing up when you said you would, returning what you borrowed, telling the truth when a lie would be easier. Every one of those is a deposit in a shared account the whole country draws on.

The data says that account is not bankrupt. It says the young are rebuilding it faster than anyone expected. The question is whether the rest of us will match them.

Sources

If this made you angry, good. Now turn it inward. The Moral Consistency Journal is a 30-day audit of the places where your own standards bend. More Case Files and data breakdowns live at themoraldecayindex.com. Books that go deeper are in our Bookshop.org store, which supports independent bookstores.

Editorial standards · Report an error · Subscribe