Is College Worth It in 2026? Reconciling a 12.5% Return With a 5.7% Unemployment Rate

I keep seeing two Federal Reserve Bank of New York statistics used as evidence for opposite arguments in the same week. One says college is a great investment. The other says the job market for new graduates is genuinely bad right now. Both come from the same research shop. Neither is wrong. And if you only see one of them, you'll walk away with a distorted picture either way.

Two numbers that seem to contradict each other

Here's the setup. In April 2025, NY Fed economists Jaison Abel and Richard Deitz published research estimating that a college degree returns about 12.5% annually as an investment — a rate they explicitly call well above the threshold for a sound investment, with the typical graduate earning more than $32,000 a year more than a high-school-only worker.

Meanwhile, the NY Fed's own quarterly tracker of the labor market for recent college graduates has been showing something rougher: unemployment among grads aged 22-27 running around 5.7% as of the first quarter of 2026, with underemployment — working in a job that doesn't typically require a degree — sitting around 41.5%. Both figures are meaningfully worse than the broader national numbers.

If you only read headlines, you'd reasonably wonder which one is true. They both are.

The 12.5% return: what it's actually measuring

The 12.5% figure is a lifetime-averaged return on investment. Abel and Deitz built it the way you'd evaluate any investment: they compared the cost of a degree (tuition, fees, and the wages you give up while in school) against the extra earnings a college graduate collects over a full working career relative to someone who stopped at a high school diploma. That premium — over $32,000 a year on average — compounds across decades of a career, which is why the annualized return looks strong even after accounting for four years of costs and forgone wages.

That bottom row matters. The 12.5% is an average, and the distribution underneath it is wide — roughly a quarter of college graduates land in a return closer to 2.6%, per the same research, which is a much weaker case for the investment. Major, school, debt load, and completion all move an individual far from the average.

The 5.7%: what that's actually measuring

The 5.7% unemployment and 41.5% underemployment figures come from a completely different exercise: a real-time snapshot of how the newest wave of graduates — people who left school in roughly the last one to five years — are doing in the labor market right now, this quarter. It's not a return calculation at all. It's closer to "how hard is it to get hired today if you just walked across the stage."

For comparison, national unemployment has been running closer to 4.2%-4.3% in 2026, and unemployment for college-educated workers overall (not just recent grads) is lower still, in the low 3% range. Recent grads specifically are running hotter than both.

Why both can be true at once

Think of it this way: the 12.5% return is measured across every graduating class going back decades, smoothed over full careers, in good hiring years and bad ones. The 5.7%/41.5% figures are a snapshot of exactly one narrow slice — the graduates trying to get hired this quarter, in whatever labor market currently exists. A degree can be an excellent 30-year investment while the specific cohort graduating into a soft, AI-and-slowdown-affected hiring market in 2026 has a genuinely harder first eighteen months than graduates did five or ten years ago.

It's the same logic I used in my earlier piece on the AI-era entry-level job market: a cooling labor market and a long-run positive return aren't in tension — they're two different questions answered by two different datasets, both real.

Here's an analogy I keep coming back to. Imagine judging whether homeownership is a good investment. Averaged over 30 years, housing has historically been a solid store of value in most US markets — that's the "12.5% return" kind of claim. But if you ask "is it a good time to buy a house this specific month, in this specific city, with these specific mortgage rates," you get a completely different, much more situational answer. Both questions are legitimate. They're just not the same question, and a single headline number can't answer both at once. College works the same way: the degree is a long-run asset, but when and how you convert it into your first job is its own, separate, much noisier problem.

Run your own numbers

The averages above are useful for context, but your actual return depends on your specific costs and your specific expected salary — not the national average. Plug in your numbers below.

To be direct about what this tool is and isn't: it's a simplified model, not a reproduction of the NY Fed's actual peer-reviewed methodology. It treats your total 4-year cost plus four years of forgone high-school-path wages as the "investment," and your salary premium (grown at a flat assumed rate) as the "return," then solves for an annualized rate. It's built to build intuition about your own numbers, not to replace the NY Fed's research.

If you're weighing how to actually pay for it, my pieces on Trump Accounts vs. 529 plans and the student loan payment shock heading into 2026 cover the financing side of this decision, which matters just as much as the ROI math — a strong ROI on paper can still be a rough financial experience if it's financed with high-rate private debt rather than savings, grants, or a lower-cost path like community college transfer.

One more honest limitation of any ROI-style framing, including my own calculator above: none of this captures the non-financial reasons people go to college, or the non-financial costs of not going. A degree can open doors that have nothing to do with the wage premium — some careers simply require the credential regardless of the math — and some people thrive in a trade or a business they built without one. Treat the 12.5% and the $32,000 premium as real, useful inputs to a decision, not the entire decision.

Here's What I'd Actually Do

1. Don't treat the 12.5% as a promise about your specific major or school. It's a national average across every field and institution. Look up wage data for your actual intended major before you assume you're near the average.
2. Take the near-term hiring numbers seriously for your first-year plan, not your 30-year plan. A soft market in year one is a real problem to plan around (extended job search, side income, flexible timelines) — it doesn't necessarily change whether the degree pays off over a career.
3. Prioritize finishing. The weakest financial outcomes generally belong to people who took on college debt and didn't complete a degree — the wage premium in this data is specifically for degree-holders.
4. Minimize the "invested" side of the equation. Since total cost is one of the two biggest levers in any ROI calculation (the other being your eventual salary), in-state tuition, community-college transfer paths, and scholarships do as much for your return as picking a higher-paying major.
5. Build a bridge for the first 12-18 months post-graduation. Given the 41.5% underemployment figure, plan for the real possibility of a slower start — a side income stream or a flexible living situation can matter more than people expect right after graduation.
6. Revisit both numbers periodically. The wage-premium research updates periodically; the recent-grad labor tracker updates roughly quarterly. Both are worth checking again before you assume this year's snapshot still applies next year.

This is educational content, not financial advice. I'm a researcher, not a career counselor or financial advisor, and I don't know your specific major, school, financing plan, or job market. Return on a college degree varies enormously by individual circumstances. See the full disclaimer.