Youth Economic Inactivity Is a Structural Problem
262 million young people are NEET worldwide. Unemployment figures miss most of them. Here's what the data actually shows about youth economic disengagement.
Written by AI. Jonathan Park

Photo: AI. Marcel Dubois
Somewhere in a wealthy country right now, there's a young person who woke up this morning with no work, no school, and no plan to look for either. That's the opening image Economics Explained uses in a recent video on youth economic disengagement — and it lands, because it's not describing one person. It's describing 262 million of them.
That number comes from the ILO's global estimates on young people classified as NEET: Not in Employment, Education, or Training. One in four young people worldwide. And the uncomfortable part isn't the scale — it's that the official headline numbers don't show it.
The measurement gap that lets politicians off the hook
Youth unemployment, as typically reported, has actually been falling across most wealthy nations. So by the official scoreboard, things are improving. But the unemployment rate has a structural blind spot: it only counts people who are actively looking for work. Stop sending applications — whether because you've given up, you're sick, you're caring for someone, or the job simply doesn't pay enough to cover the commute — and you're no longer unemployed. You're "economically inactive," which is economist-speak for "not our problem anymore."
Economics Explained puts the gap in stark terms: "For every young person counted in official unemployment statistics, there are roughly three more who aren't counted at all."
In 2023, 6% of young people globally were officially unemployed. The NEET rate was 20.4%. That's not a rounding error. That's a measurement system failing to see the thing it was built to measure.
It's worth being precise about who's inside that 20.4%, because lumping everyone together is exactly how bad policy gets made. A minority — Economics Explained puts it at roughly one in five NEETs in Europe — are voluntarily disengaged: taking time out, holding out for something better, or building something independently. Some of those choices are reasonable. Some are probably miscalculated. Either way, they're not the structural story.
The structural story is the other four in five: people dealing with health conditions and disability, caretaking responsibilities, discouragement after sustained rejection, or simply living somewhere the jobs aren't. In the UK, the share of NEETs reporting a work-limiting health condition rose from 26% in 2015 to 44% in 2025. One in five NEET women are caring for children, compared to one in thirty young men. In the US, a quarter of NEETs come from families earning under $25,000 a year, and the rate among Black youth runs more than 50% higher than among white youth.
"Patterns like these," the video notes, "point much more towards structural disadvantage than individual choice."
How the entry-level ladder got pulled up
To understand why this is happening now, you have to understand what the economy used to offer young people that it increasingly doesn't.
For most of the 20th century, there was a functioning first rung. Factory work, admin roles, data entry, retail management — jobs that didn't require experience or connections, paid a living wage, and gave you enough of a work history that the next job became easier to get. They weren't glamorous. They were the way in.
Automation gutted the most routine of those roles first. AI is now doing the same to entry-level white-collar work. The specific claim in the Economics Explained video — that employment for new workers in AI-exposed jobs fell 6% between late 2022 and mid-2025 while older workers in the same fields saw employment grow — is consistent with a broader research picture that has been building since the ChatGPT moment in late 2022.
Work by economists David Autor, Caroline Chin, Anna Salomons, and Bryan Seegmiller, published through the National Bureau of Economic Research, documents how automation disproportionately displaces workers in routine and entry-level tasks. More recently, research from the Federal Reserve Bank of St. Louis and independent labor economists has flagged a measurable cooling in entry-level hiring in sectors most exposed to large language models — finance, legal support, content production, customer service — beginning in 2023. A 2024 analysis by economists at the University of Chicago found that recent graduates entering AI-adjacent fields faced materially worse hiring outcomes than cohorts from just two years prior, even as total sector employment held relatively flat or grew among experienced workers. The direction of the effect described in the video is well-supported; the specific 6% figure should be treated as an approximation pending more granular longitudinal data, but the asymmetry — younger workers bearing the displacement cost while experienced workers remain insulated — has genuine empirical backing.
What filled the gap left by vanishing entry-level roles was the gig economy. Which, as Economics Explained puts it, replaced stable employment with "flexible work that sounds appealing until you realize it offers no progression, no training, and no way up." You get availability on demand; the employer gets to keep all the risk off their books. Half of Gen Z workers report burnout. More than two-thirds still rely on parents or family for essentials like rent and groceries — including many who are working.
The degree trap
The standard policy response to disappearing entry-level work has been: go to university. And the degree is still effectively required for most professional jobs, whatever the job posting says. Companies made a lot of noise in recent years about dropping degree requirements — Economics Explained notes that fewer than one in 700 new hires actually benefited from those changes.
So the credential is still the price of admission. The problem is the cost-benefit math has collapsed.
UK tuition fees tripled in 2012. The average US student borrower carries nearly $40,000 in federal debt alone. Young people took on that debt on a reasonable assumption: a degree translates to a graduate-level job. But in 2024, the share of UK graduates landing in roles that actually require a degree hit its lowest level since 2014. For the first time on record, the unemployment rate for recent US graduates has started to exceed the overall unemployment rate.
You can't get in without the degree. The degree increasingly doesn't get you in.
Housing closes the last exit
Even for young people who navigate all of this and find work, geography is increasingly a trap. The best jobs cluster in specific cities — finance in London and New York, tech in San Francisco. Previous generations could move to those cities because rent took up less than a fifth of their income. Today it's closer to a third on average, and in the cities where the opportunities actually are, it's considerably worse.
In Sydney, the income needed to rent a typical apartment without financial stress jumped from A$88,000 in 2019 to A$130,000 in 2025. The average young worker in New South Wales earns around A$54,000. In the US, three in five Gen Z renters are rent-burdened — spending more than 30% of income on housing before anything else. In Los Angeles and San Diego, it's closer to three in four.
The rational response is to not move. Stay in the town you grew up in, where opportunities are thinner and NEET risk is higher. Or move back in with your parents — now the living situation of roughly one in four young US adults, up from one in nine fifty years ago.
"None of this," the video observes, "is particularly good for your mental health."
The UK share of young people reporting common mental health conditions rose from 19% in 2014 to nearly 26% today. In Australia, depression prevalence among young adults more than doubled between 2007 and 2021. In the US, people aged 18 to 24 are now more likely to report poor mental health than people twice their age. Which came first — the structural pressures or the mental health deterioration — is genuinely disputed. The evidence suggests both directions are real, which makes this harder to untangle than either a purely economic or purely clinical story.
What the evidence says actually works
Economics Explained closes with the part that tends to get skipped in these conversations: the variation across wealthy countries is enormous, and it's not random.
NEET rates run above 22% in Italy and 19% in Greece. In the Netherlands, they sit just above 5%. In Denmark, 6.2%. These aren't cultural differences — they're policy differences.
The Netherlands runs a dual education system where vocational students split time between classrooms and real workplaces. Ninety percent of recent vocational graduates were employed in 2024, well above the EU average. Sweden has made childcare affordable enough that nearly 80% of women are in the workforce and its NEET rate sits at 6.7%. Denmark's "flexicurity" model — flexible hiring paired with strong safety nets and active retraining — means losing a job doesn't become a life sentence.
Eurofound estimates NEETs cost European economies €142 billion a year in foregone taxes, lost productivity, and benefits. In the UK alone, matching every region's NEET rate to the best-performing region would add £26 billion annually. That's just the gain side; the costs of inaction are considerably larger and harder to calculate.
Japan illustrates what the longer-run looks like when none of this gets addressed. After the asset bubble burst in the early 1990s, stable entry-level jobs dried up and never fully returned. Today, one in four Japanese people in their 20s will experience NEET status at least once in any four-year period. The phenomenon of hikikomori — people, mostly young men, who withdraw from work, school, and eventually the outside world — now affects an estimated 1.5 million people, with an average duration of around ten years. Japan even has a name for what happens when that goes unaddressed for a full generation: the 80/50 problem. Parents in their 80s still supporting children in their 50s who never re-entered the workforce. Two people removed from economic participation. When the parents die, there's no income, no work history, and nowhere to go.
That's not a cautionary tale. That's a policy outcome, produced by specific choices, over decades. The choices that led there — a hyper-competitive education system, narrowing entry points, and a welfare structure that leaned entirely on families to absorb the strain — aren't unique to Japan.
The lazy generation story is easier to tell than the structural one. It's also more politically convenient: if the problem is individual failure, the policy response is motivational. Lecture the unemployed. Sanction their benefits. The structural story requires actual changes to education systems, housing markets, childcare costs, and labor protections. Those are harder.
But they're not mysterious. The evidence of what works is sitting right there in the countries that have done it.
By Jonathan Park, Business Desk Editor
We Watch Tech YouTube So You Don't Have To
Get the week's best tech insights, summarized and delivered to your inbox. No fluff, no spam.
More Like This
How APL Turned an NBA Ban Into a Brand Identity
APL's NJ Falk breaks down the six storytelling threads behind one of e-commerce's most unlikely brand-building moves—starting with an NBA ban.
Bill Ackman's Art of Finding Invisible Money
Explore Bill Ackman's strategy of finding hidden value in companies, beyond standard metrics.
Why Product-Market Fit is the Entrepreneur's True North
Explore why product-market fit is vital for business success and how to achieve it.
Dubai's Crisis Has a Worker Problem No One's Covering
Dubai's economic model was built on migrant labor with almost no rights. Now, with airstrikes and a blockaded strait, those workers face the worst of the fallout.
HTML Pub: Simplifying AI Content Hosting
Explore how HTML Pub simplifies hosting AI-generated content with easy integrations and analytics.
Unpacking Katie Keith's $1.8M App Empire
Explore Katie Keith's strategy for generating $1.8M annually with 19 apps, and its fit within industry trends.
RAG·vector embedding
2026-06-15This article is indexed as a 1536-dimensional vector for semantic retrieval. Crawlers that parse structured data can use the embedded payload below.