About 16 percent of Americans between 18 and 24 were neither employed nor enrolled in school in 2024, and that share was essentially unchanged as of December 2025, according to research published by the Federal Reserve Bank of St. Louis in July 2026. Economists call this group "disconnected youth"; internationally the label is NEET, for not in education, employment or training.
The figure has drawn new attention because it has stayed flat during a stretch when the broader job market looked healthy. The overall U.S. unemployment rate was 4.1 percent in July 2026, according to the Bureau of Labor Statistics. The unemployment rate for 16- to 24-year-olds was 9.4 percent in May 2026 — roughly double the national rate, and a level that has climbed over the past year.
The gap is not unique to the United States. In Britain, the figure that prompted a wave of commentary this year is close to a million people aged 16 to 24 out of work and out of school, with a youth unemployment rate of 15.3 percent in the three months to December 2025 — higher than the European Union average for the first time. A Telegraph column published Aug. 9 argued that British policy had made hiring young workers more expensive and that young people were responding rationally to the incentives in front of them. The American numbers are lower than Britain's, but they are moving in a similar direction, and several of the proposed explanations cross the Atlantic intact.
How the U.S. measures it, and why the numbers differ
There is no single official American NEET count, which is part of why published figures vary.
The St. Louis Fed measure covers ages 18 to 24 and puts the disconnected share at 16 percent. The World Bank and International Labour Organization, using a broader youth definition, put the U.S. NEET rate at 11.6 percent for 2024. The American Institute for Boys and Men, in a 2025 analysis, estimated that roughly 12 percent of young men were NEET in 2024.
Those are different denominators and different age bands measuring overlapping populations, not contradictory findings. What they agree on is scale: the group runs into the millions, and it has not shrunk.
The category also mixes very different situations. It includes people actively looking for a first job and not finding one; people caring for a child or a relative full time; people out of the labor force because of a disability or a health condition; and people who have stopped searching. Time-use research from the American Institute for Boys and Men found that people in the NEET group spend substantially more of the day on leisure — about 6.7 hours versus 4.3 hours for their peers — and also more time on caregiving and household work. Both patterns are present in the same statistic, which is why a single number tends to be a poor guide to what any individual is doing.
The entry-level problem
The clearest recent signal is not about dropouts. It is about graduates.
The Federal Reserve Bank of New York, which tracks the labor market for recent college graduates, reported an unemployment rate of 5.6 percent for that group in the second quarter of 2026 — about 1.5 percentage points above the national rate, an unusual inversion of the historical pattern in which a degree conferred a hiring advantage. The same tracker put underemployment among recent graduates at 42 percent, meaning that share was working in jobs that do not typically require a bachelor's degree.
Employers have pulled back on the junior end of the ladder specifically. A Census Bureau working paper released in April 2026 examined early-career hiring against firms' adoption of artificial intelligence, and analyses of private payroll data by Stanford economists this year found declines concentrated in entry-level roles in AI-exposed occupations. Coverage of those findings has been split: some economists read them as evidence that AI is absorbing the tasks companies used to hand to a first-year hire, while others attribute most of the slowdown to a general corporate hiring freeze and say the AI signal is not yet clean enough to separate from the business cycle.
A third explanation has less to do with technology. Researchers at the New York Fed argued in June 2026 that the spread of remote work has sidelined younger workers, because remote arrangements favor employees who already have experience, established networks and a track record, and make on-the-job training of a newcomer harder to deliver.
Young men and young women
The unemployment rates diverge by sex. Federal data put unemployment for men aged 16 to 24 at 10.5 percent in May 2026, against 9.3 percent for women in that age range in June 2026 — different months, because of how the series are published, but consistent with a gap that has persisted through this cycle.
The composition differs too. Analysts who study the male side of the gap point to disability claims and long-term detachment from both school and work; on the female side, caregiving and parenthood account for a larger share of time out of the labor force. Women now outnumber men on American college campuses, a shift documented in university and Federal Reserve research, which changes who is in the "education" half of the NEET definition in the first place.
Federal statistics do not currently publish a clean nationwide breakdown of why young adults in this group are not working — how many cite disability, how many caregiving, how many discouragement. That absence is itself a limitation on the policy debate.
What Congress is doing
The main federal vehicle is the Workforce Innovation and Opportunity Act, the law behind job centers, apprenticeships and training grants. A reauthorization measure backed by the House Education and the Workforce Committee majority has advanced out of committee but has not been enacted.
The bill would shift a larger share of federal training dollars toward programs tied to specific employer demand and toward skills training rather than administrative overhead. Democrats on the committee objected to funding levels and to changes in how local workforce boards are governed.
In Britain, where the numbers are worse, a parliamentary committee in July 2026 recommended cutting employer payroll taxes for all workers under 25 to lower the cost of hiring a young person. No equivalent proposal has advanced in Congress.
Methodology and definitions
What "NEET" means. NEET stands for "not in education, employment or training." A person counts as NEET if, during the reference week of a survey, they were not working for pay at all, not enrolled in school or college, and not participating in a formal training or apprenticeship program. It is a status, not a behavior: it says nothing about whether a person is looking for work, and it does not distinguish between someone applying to jobs daily and someone who has stopped.
How it is measured in the United States. There is no single official U.S. NEET statistic. The figures in this article come from three different sources using three different methods:
- The Federal Reserve Bank of St. Louis measure covers ages 18 to 24 and is derived from the Census Bureau''s Current Population Survey, the same monthly household survey behind the national unemployment rate. It produced the 16 percent figure for 2024. - The World Bank and International Labour Organization use a broader youth definition and internationally harmonized estimates, which yielded 11.6 percent for the United States in 2024. - The American Institute for Boys and Men uses Current Population Survey microdata restricted to young men, producing roughly 12 percent for 2024.
Different age bands and different survey adjustments produce different numbers for overlapping populations. They are not competing claims about the same quantity.
How "unemployed" differs. The 9.4 percent figure for ages 16 to 24 is an unemployment rate, which is a narrower measure. To be counted as unemployed, a person must be out of work, available to work, and have actively looked for work in the prior four weeks. Someone who has given up searching is not unemployed by this definition, but may still be NEET. This is why the two figures move independently and why the NEET share can stay flat while the unemployment rate falls.
How "underemployed" is defined. The New York Fed''s 42 percent underemployment figure for recent college graduates counts people working in occupations where fewer than half of workers hold a bachelor''s degree. It measures job-to-credential mismatch, not hours worked or satisfaction.
Limitations.
- The reference period is a single week. Someone between semesters, between jobs, or on medical leave can be counted as NEET in one month and not the next. - Federal statistics do not publish a nationwide breakdown of why young adults in this group are out of work. Caregiving, disability, illness, incarceration, discouragement and voluntary time off all land in the same bucket. - Survey response rates for young adults have declined, and household surveys undercount people without stable housing or phone service — groups likely to be overrepresented among the disconnected. - Attributing changes to artificial intelligence is contested. The Census and Stanford analyses cited here identify correlations between AI adoption and reduced entry-level hiring; they do not establish that AI caused the decline, and a general hiring slowdown could produce a similar pattern. - Cross-country comparison with the United Kingdom is directional only. Britain uses a 16-to-24 age band, a different benefits system and different school-leaving rules, so its figures should be read as context rather than a like-for-like benchmark. - Some sources are dated differently by publication month rather than by data collection month; where months differ in this article, they are stated explicitly.