A study published by the Federal Reserve Bank of New York on June 1 finds that the rise of remote work since the pandemic explains approximately 64% of the recent increase in unemployment among young college graduates — significantly more than the impact of artificial intelligence — because companies are reluctant to hire inexperienced workers they cannot easily train and mentor from afar.
The story that has dominated conversations about young workers and the labour market in recent years is one about artificial intelligence: that automation and generative AI are eliminating entry-level jobs, making it harder for new graduates to break into industries, and fundamentally reshaping the skills that employers value. That story is not without evidence. But a study published Monday by the Federal Reserve Bank of New York finds that it may be telling the wrong story.
New research from the Federal Reserve Bank of New York finds that younger college graduates have been sidelined by remote work in recent years, as companies may be reluctant to hire those needing more training and mentoring. The buzz on college campuses is that AI is disrupting the job market for young college graduates. But new research from the Federal Reserve Bank of New York finds that the culprit may be something else: remote work.
Youth unemployment has risen dramatically since the pandemic — as has the prevalence of remote work. The researchers’ analysis suggests that these trends are related, with remote work making it more difficult for managers to train and mentor new employees. Accordingly, companies may be reluctant to hire less-experienced workers in distributed work arrangements. They estimate that remote work can explain 64% of the recent increase in unemployment among young college graduates.
The Data: What the Numbers Show
The unemployment rate for young college graduates swelled to 5.6% in March 2026 from 3.6% in March 2019, the New York Fed economists wrote in a blog post published Monday.
Work-from-home arrangements, which have surged fourfold since the pandemic, can explain an estimated 64% of the recent increase in joblessness among young grads, New York Fed researchers said. As of March, the unemployment rate for recent college graduates was 5.6%, compared to 4.3% for all workers.
The gap — 5.6% youth graduate unemployment against a 4.3% overall rate — is substantial. It means that young people who have invested significantly in higher education are facing worse labour market outcomes than the general working population. That outcome contradicts the longstanding assumption that a college degree provides labour market protection during economic uncertainty.
The study finds that the unemployment rate among young college graduates in “remotable” jobs rose by about 1 percentage point from 2017-2019 to 2022-2024. Yet for older workers in those fields — those ages 29 and over — the jobless rate declined slightly, leading to a notably higher unemployment rate for younger college graduates in remotable occupations compared with older workers. Yet in non-remotable jobs, there has been little gap in the unemployment rates between older and younger college grads.
That final sentence is the key finding. In jobs that cannot be done remotely — nursing, construction, food service, retail — young and older college graduates face similar unemployment rates. The gap emerges specifically and significantly in remotable jobs: software development, financial analysis, data science, marketing. These are precisely the fields that attracted the most remote-work adoption during and after the pandemic, and they are the fields where young graduates are now significantly disadvantaged relative to their older counterparts.
Why Remote Work Hurts Young Workers Specifically
The mechanism the New York Fed researchers propose is straightforward and intuitive: training and mentorship.
“Remote work has weakened incentives to hire young workers by impeding on-the-job training,” the study says. “Employers may not want to hire fresh graduates onto distributed teams because it is more difficult to teach them the requisite skills from afar.”
Learning how to do a job — not the theoretical knowledge taught in university courses, but the practical skills, professional norms, and tacit knowledge that make someone genuinely productive — happens primarily through proximity to experienced colleagues. Watching how a senior analyst approaches a problem, getting real-time feedback on a piece of code, absorbing the unstated norms of a workplace through observation: all of these mechanisms work better in person than they do through a screen.
An analysis of federal employment data, paired with a deep dive into the flexible work arrangements at one unnamed Fortune 500 tech company, reveals that companies are less likely to hire recent college grads into occupations that can be done remotely. Researchers speculate that employers are reluctant to put such workers in a setting where it’s harder to absorb lessons from coworkers.
The researchers found that while remote work can boost output among experienced workers, it can be to the detriment of younger engineers. Feedback on coding work increased 18.3% when workers were in the office, improving the quality of output.
The 18.3% increase in coding feedback when workers were in the office is a concrete, measurable illustration of the mentorship mechanism. More feedback means more learning. More learning means faster development of the skills that make a new graduate valuable. Remote work reduces feedback; reduced feedback slows development; slower development makes new graduates less attractive hires.
The Company’s Response: Hire Experienced Workers Instead
The New York Fed found that unemployment rates in recent years have increased particularly fast among young workers in occupations that can easily be performed remotely, like software engineering. Digging into proprietary data from an unnamed Fortune 500 company, the researchers also found that people who had worked next to their colleagues received more feedback and more mentorship. Perhaps understanding “the pitfalls of distance for worker development,” the company hired fewer less-experienced workers when its offices were closed during the pandemic and consistently opted for more experienced workers when hiring for positions on distributed teams, even after reopening.
The unnamed Fortune 500 company’s behaviour — consciously or not — represents a rational response to the economics of remote training. If it costs more to get a junior employee to the same level of productivity in a remote environment, and if there is an alternative (hiring an already-experienced worker who requires less training investment), companies will choose the alternative. The result is a labour market that systematically disadvantages the workers who have the most to gain from entry into it.
The AI Comparison: A Clarification, Not a Dismissal
While the impact of artificial intelligence on entry-level jobs has received much attention recently, the expansion of remote work has likely had a greater impact on youth unemployment.
The New York Fed study does not argue that AI has no impact on the labour market. It argues that, in the specific context of youth unemployment since the pandemic, remote work is a larger explanatory factor than AI. The distinction matters because the policy responses are different.
If AI is the primary culprit, the solution involves education reform, reskilling programmes, and AI governance. If remote work is the primary culprit, the solutions are closer at hand: return-to-office policies, hybrid work arrangements that ensure sufficient in-person time for junior employees, or explicit commitments by employers to invest in remote mentorship infrastructure.
However, even companies that offer remote work typically require a few days on-site, which provides room for collaboration and mentorship, said Nicholas Bloom, an economics professor at Stanford University who studies remote work. “I don’t think there is any evidence this is slowing employment,” Bloom said. “Indeed, quite the reverse, as it’s easier for people to work and so labour supply looks to be rising.”
Bloom’s pushback — that hybrid arrangements provide sufficient mentorship opportunity and that overall employment is rising — introduces important nuance. The New York Fed study captures a pattern in the data; it does not claim that remote work is universally harmful or that every company is responding identically.
What Young Workers Actually Want
The irony of the finding is that it inverts the popular assumption about generational work preferences. AI-related job disruption is happening to workers who have largely embraced technology. Remote work advantages are being captured by older workers — while the generation most associated with digital fluency and comfort with technology is being systematically disadvantaged by the remote work arrangements it was assumed to prefer.
Young workers in general are likely aware that the popularity of remote work is mostly to their disadvantage, as a Gallup poll last year found Gen Z to be the age group least likely to prefer a fully remote workplace setup, citing in part the lack of interaction with coworkers.
Gen Z workers — the generation that has been characterised as demanding remote flexibility — are, according to Gallup polling, the least likely to prefer fully remote work. They want interaction. They want mentorship. They want to be seen and developed. The labour market, having adopted remote work for reasons that had nothing to do with young workers’ preferences, is denying them exactly what they want and need.
What Happens Next
Some 64% of the rise in jobless rates among college-educated workers under the age of 29 vs. prepandemic levels can be attributed to work-from-home trends.
The New York Fed study will enter a policy debate that has been ongoing since the pandemic ended. Employers considering return-to-office mandates now have a specific, data-backed argument in their arsenal: remote-first policies are systematically disadvantaging the youngest, least-experienced members of the workforce.
Whether that argument moves corporate behaviour — or whether the cost savings and experienced-worker satisfaction associated with remote work prove more powerful than the new graduate disadvantage — will be determined in hiring decisions made across thousands of companies in the months and years ahead.
For the class of 2026, graduating into a labour market with a 5.6% unemployment rate while their professors told them AI was the threat to watch, the study offers something useful: a clearer picture of the actual problem. That clarity is a necessary first step toward addressing it.
LoudFact.com is an independent global news and explainer platform. This report is based on the New York Fed Liberty Street Economics study published June 1, 2026, and reporting from NPR, CNBC, Fortune, Yahoo Finance, and Business Report as of June 1-2, 2026.
