Young and Less Happy

Since 1972, the General Social Survey has asked Americans a question of almost comic simplicity: Taken all together, how would you say things are these days — would you say that you are very happy, pretty happy, or not too happy? Three boxes, no nuance. And for most of five decades, the answers barely moved. Around 31 percent of adults called themselves "very happy," a clear majority settled on "pretty happy," and the averages by age traced the same gentle slope, survey after survey: the young a little lower, the old a little higher.

That stability is what makes the last few years so striking. Between 1972 and 1989, adults aged 18–29 averaged 2.16 on the 3-point scale — a touch below the 2.28 of those 60 and over, but comfortably in "pretty happy" territory. Through the 1990s and 2000s the young held almost exactly that line. Then it gave way. Across 2015–2024, the young-adult average fell to 1.97. In 2021 it touched 1.80 — the lowest reading for either group tracked here in the survey's 52-year history — and in 2024 it stood at 1.91, still below anything recorded before 2018. Just 20.7 percent of 18-to-29-year-olds now describe themselves as "very happy."

Older Americans dipped too — from 2.28 in the first era to 2.10 in the most recent — but far less, and they remain the happiest group in the data. The result is not a reordering of the age curve. It is something quieter and, arguably, more unsettling: the same old gradient, now rising out of a much deeper hole at the young end.

The shape that didn't change — and the level that did

A famous literature, associated with the economist David Blanchflower, holds that well-being follows a U over the life course: high in youth, bottoming near midlife, recovering in old age. Richard Easterlin and others have long disputed it, arguing the U is fragile — appearing in some surveys and specifications, vanishing in others. Honesty requires saying plainly which camp this data falls in: in the GSS happiness item, there is no midlife U. In every era of the survey, average happiness rises gently and somewhat noisily with age. Young adults have always been among the least happy Americans; people past 60 have always been the happiest. Nothing flipped, and no U broke. The midlife-dip pattern may well be real in other datasets and other measures — life-satisfaction ladders, daily-affect diaries — but it does not appear in this coarse, three-box question.

What changed is the level. Each curve below is one era of the survey, pooled so every age cell rests on hundreds of respondents. The three twentieth- and early-twenty-first-century curves sit nearly on top of one another — four decades of stasis. The 2015–2024 curve sits below all of them at every age, and it sags hardest exactly where the others begin: at 18–24, where the era average is 1.92, the lowest point anywhere on the chart, with only about 17 percent saying "very happy."

Two things reward a slow look. First, the three earlier curves are almost interchangeable — at ages 25–29, the eras from 1972 to 2014 land within a single hundredth of one another. Whatever else changed in American life between Watergate and the iPhone, this blunt instrument registered almost none of it. Second, the orange curve is not uniformly displaced: the deficit is largest among the youngest and narrows with age, so the slope of the 2015–24 curve is visibly steeper at its left end. Today's 75-and-overs sit only modestly below their predecessors. Today's 21-year-olds sit in territory the survey had simply never charted.

A useful summary number is the gap between old (60+) and young (18–29). In 1972–1989 it was 0.11 points in favor of the old. It then narrowed for decades — to 0.09 in 1990–2004 and just 0.06 in 2005–2014, mostly because older adults' happiness eased while the young held steady. Since 2015 the gap has widened back to 0.12, its largest value in the record, and for the opposite reason: this time it is the young who fell.

When the floor gave way

Pooling eras shows the shape; the year-by-year series shows the timing. For more than forty years the two lines below run roughly parallel — older Americans a shade above, young adults a shade below, both bouncing inside a narrow band. In 2016 and 2018 the young-adult line drifts near the bottom of that historical band — low readings, but no lower than the 2008 dip, which fully rebounded within two surveys. Then comes 2021: young adults at 1.80, the lowest point in either series, with older adults posting their own record low at just over 2.0.

A necessary caution sits right on top of that low point. In 2021 and 2022, COVID forced the GSS to abandon its traditional in-person interviews for a push-to-web design, and self-administered surveys reliably elicit gloomier answers to questions like this one — people shade toward candor when no interviewer is watching. Level shifts across that boundary are therefore partly artifactual, and the shaded band in the chart marks the affected years. But the mode change cannot carry the whole story: the 2024 reading of 1.91, collected under a mixed-mode design, sits below every pre-2018 value in the series — including that 2008 dip. Older adults, surveyed the same way, fell far less and partially rebounded.

The thinning of "very happy"

Averages on a 3-point scale compress a lot. A second cut makes the change concrete: the share of young adults willing to tick the top box. For three straight eras spanning 1972 to 2014 that share held near 28 percent. In 2015–2024 it fell to 20.7 percent — roughly one in five. The top of the distribution thinned at the same time the bottom swelled; this is not a story of mild reshuffling between "very" and "pretty."

What this is — and isn't — evidence of

The GSS finding lands amid a broader literature reaching the same verdict. Blanchflower, Bryson and Xu (2025), surveying data across dozens of countries, document a marked deterioration in youth mental health and well-being since the mid-2010s — severe enough, they argue, that the very age profile of well-being is being reshaped, with the young no longer enjoying the relative buoyancy earlier cohorts did at the same age. The GSS, with its half-century of identical wording, is one of the cleanest American windows onto that shift, and it agrees: something changed for young adults in the late 2010s, and it has not changed back.

What the GSS cannot do is say why. The usual suspects — smartphones and social media, pandemic-era isolation during formative years, housing and economic precarity, a documented rise in loneliness — are all consistent with the timing, and none can be tested with a single descriptive item. Nor should the size of the move be oversold: on a 1-to-3 scale, the young-adult average fell about 0.19 points between the first and latest eras. That is enormous by the glacial standards of this series — several times any era-to-era movement in the prior four decades — but most young Americans still say they are pretty happy. The signal here is not mass despair; it is a cohort entering adulthood reporting less happiness than any cohort the survey has ever observed at that age, on a question too blunt to exaggerate.

The old gradient offers one guarded reason for hope: in every era, this measure rises with age. If today's young adults eventually climb the same slope, their low start may read, decades from now, as a deep but survivable trough. Whether they climb it — or carry the deficit with them — is the question the next fifty years of this survey will answer.