The midlife crisis had a graph.
A real one, plotted from survey data. Ask people across dozens of countries how satisfied they are with their lives, chart the answers against age, and what came back was a shallow smile. High in your twenties. Sagging through your thirties. Bottoming out somewhere around fifty. Climbing again after that.
It was the rare research finding that matched what people already believed about their own lives. The forties are hard. It gets better. Hang on.
That curve has straightened out over the last decade. Which sounds like good news, right up until you look at which end of it moved.
The U-curve was replicated more than 600 times, and then it stopped showing up
The pattern was first described in 2008 by Dartmouth economist David Blanchflower and Andrew Oswald at Warwick, and it went on to become one of the most reproduced results in the social sciences.
Researchers found it in rich countries and poor ones. They found it in life satisfaction, in happiness, in stress, in worry. The low point sat at roughly age 50 almost everywhere anyone looked, and the mirror image showed up in the bad stuff too. Depression, anxiety, and despair all peaked in midlife and then eased off.
Blanchflower and his colleagues have since documented that the pattern has been replaced by a straight line, using US and UK survey data plus pooled results from 44 countries.
Ill-being no longer humps in the middle. It now declines steadily with age, which is another way of saying the youngest adults are reporting the worst mental health of anyone.
Blanchflower has been blunt about what happened.
He told Dartmouth’s news office that the shift started around 2013, and that the original finding wasn’t an error. In his words, it’s not that the pattern researchers saw was wrong, it changed.
Despair among young British women rose from 4.4 percent to 12.7 percent while the oldest group barely moved
The numbers are worth considering, because “young people are struggling” has become the kind of sentence that slides past without landing.
In the UK household panel, the share of women aged 18 to 24 scoring in the despair range went from 4.4 percent in 2009 to 12.7 percent in 2021. Most of that jump came after 2016. Among young men it went from 2.3 percent to 6.4 percent over the same stretch.
In the American data, drawn from a survey that runs more than 400,000 interviews a year, despair among under-25s roughly doubled for both sexes between 2009 and 2023.
Here’s the part that reframes everything. Over those same years, despair among older Americans stayed essentially flat. The researchers found it remained unchanged for everyone over 45.
Nobody at the older end of the curve got worse. The curve changed shape because one end fell through the floor.
More Bolde Stories
The right side of the curve never moved, which is the part almost nobody mentions
Coverage of this research tends to stop at the collapse, understandably. But the other half of the finding is that later life held.
Across the six most populous English-speaking countries, Jean Twenge and Blanchflower found that since 2020, happiness and life satisfaction now rise steadily with age rather than dipping and recovering.
Rise, all the way up. The oldest respondents in these surveys are reporting the highest wellbeing of any adult age group, and they’re doing it in the middle of a decade nobody would describe as calm.
Which means the thing your parents’ generation told you about midlife being a passage rather than a destination is still, as far as the data shows, true. People in their sixties are reporting some of the best years they’ve had.
The dip you might be feeling at 47 is real. It just no longer looks unusual compared to what a 24-year-old is reporting.
People in their sixties don’t just report fewer bad days, they report better ones moment to moment
One reasonable objection: older adults might simply be better at telling a survey they’re fine. Life satisfaction is a summary judgment, and summary judgments are easy to launder.
So it helps that there’s a different kind of evidence entirely.
Stanford psychologist Laura Carstensen ran an experience-sampling study that pinged adults five times a day at random for a week, then repeated the whole thing five years later and again ten years later. Rather than asking people to summarize a decade, it caught them in real time, in traffic, at dinner, mid-argument.
Her team found that emotional experience improved with age across the whole span from early adulthood into very late life, with more positive states overall and noticeably more emotional stability. The frequency of negative emotions dropped until roughly age 60, at which point it leveled off.
The effects held after accounting for personality, physical health, and demographics, which rules out the easy explanation that older people are just healthier or wealthier or temperamentally sunnier.
It’s a small sample, 184 people in the first wave, and it’s one lab’s work. But it points the same direction as the population surveys, using a method that’s much harder to fool.
Some of what shows up in that data is the accumulated skill of having done things before, the kind of social competence people build in their fifties and sixties without particularly noticing they’ve built it.
The collapse is concentrated among young people who work, not young people in college
The default explanation for the youth decline is smartphones, and Blanchflower himself has pointed at 2013 and the arrival of ubiquitous social media.
But his more recent analysis complicates that story in a way worth knowing about.
Reporting on a working paper he circulated in January, The Hechinger Report noted that the wellbeing collapse turned out to be concentrated among young adults who were employed, especially women under 25. College students and young people not in the labor force still showed something close to the old curve.
If phones were the whole answer, misery would have risen at roughly the same rate across everyone in that age band. It didn’t.
Some of what’s showing up looks like work itself. In 2025, job satisfaction among workers 55 and older sat at 72 percent, against 57 percent for workers aged 18 to 24. Young workers rate their jobs as lower quality across multiple dimensions and report more trouble making ends meet.
That paper hasn’t been peer reviewed yet, and Blanchflower is careful to say the cause isn’t settled. His own conclusion is that the youngest workers, particularly those without college, are hit hardest and nobody quite knows why.
More Bolde Stories
The U-curve was contested long before it flattened
None of this is uncontested. The U-shape has been argued about for the better part of two decades, mostly over statistical technique.
Sociologist David Bartram has argued that the U appears mainly when researchers add control variables for things like income and marital status, and that doing so distorts the question being asked. Others have found the relationship looks flat, or looks like a rise rather than a dip in early adulthood, depending on the data and the method.
Blanchflower has pushed back hard, and the weight of published work still sits on his side. But the summary is that this was a disputed finding before it changed, which is a reason to hold the newer result with the same looseness.
What isn’t in dispute is the underlying movement. Young adults across many countries are reporting substantially worse mental health than young adults did fifteen years ago. Whatever shape the curve was, that part shows up in every dataset anyone has pointed at it.
Midlife didn’t get better, the comparison did
If you’re 45 and reading this hoping the midlife crisis has been canceled, it hasn’t.
The middle-aged trough is still there in absolute terms. Divorce, aging parents, career ceilings, the physical arithmetic of a body that no longer bounces. None of that has been repealed by a change in a graph.
What changed is that midlife is no longer the low point of the human lifespan, because a different group moved beneath it.
Which leaves two things worth carrying out of this. The first is that the promise built into the old curve still appears to be intact. Whatever is hard about your forties, the people twenty years ahead of you are reporting that it gets better, and they’re reporting it while they’re living it, on surveys taken this decade.
The second is about the twenty-somethings in your life. The cultural script says this is supposed to be their easiest decade, and that script is now roughly a decade out of date. A 23-year-old who seems to be struggling more than you remember struggling at 23 probably is. Character and gratitude have very little to do with it.
The graph used to say the hard part was in the middle. It doesn’t say that anymore.
If any of this describes where you are right now rather than something you’re reading about, that’s worth saying out loud to someone, and a doctor or therapist is a reasonable first person to say it to.
