Is a College Degree Worth It in 2026? An Honest Cost-Benefit Analysis

A bachelor’s degree remains a sound financial investment for most students: the New York Fed puts the return near 12.5 percent a year, well above stocks or bonds.

The outcome depends heavily on finishing, the net price paid, and the major chosen, and at least one graduate in four has not recouped the cost in recent decades.

The question has become harder to answer because the average and the individual experience have drifted apart. Public confidence in higher education sits at 38 percent, down from 57 percent in 2015. Yet 93 percent of current bachelor’s students call the investment worthwhile.

Both groups are reading the same market. The public is reacting to sticker prices, a tougher entry-level job market, and a debt system under visible strain. Students are reacting to what a degree is doing for them now.

The honest answer requires separating the typical outcome from the outcome a specific student can expect, because the gap between the two is where most financial damage happens.

What the Earnings Data Shows

The College Board’s latest Education Pays report puts median earnings for full-time bachelor’s degree holders at $81,800, against $50,600 for high school graduates. That is a gap of $31,200, or 62 percent.

The advantages extend beyond the paycheck: unemployment is 2.6 percent for degree holders versus 4.3 percent for high school graduates, and 66 percent of full-time graduates have employer health coverage versus 51 percent. Only 4 percent of graduates live in households in poverty, compared with 13 percent of high school graduates.

The premium also compounds. Graduates held a 27 percent earnings premium at age 25 in New York Fed data, rising to 60 percent by age 55. Among adults aged 35 to 44, 40 percent of bachelor’s holders earned more than $100,000, compared with 13 percent of high school graduates. Jobs open to workers without a degree tend to offer little wage growth, so the gap widens with age rather than closing.

The return calculation deserves a closer look. The New York Fed’s economists Jaison Abel and Richard Deitz found the return to college has held between 12 and 13 percent for three decades, against roughly 8 percent for stocks and 4 percent for bonds.

The method weighs four years of costs, including the wages a student forgoes by not working, against a career-long premium. Forgone wages, not tuition, are often the larger share of that cost, which is why anything that stretches the timeline hurts returns so sharply.

The Sticker Price Is Not the Price

Published prices overstate what most students pay. Average published tuition and fees for 2025-26 were $11,950 at public four-year colleges for in-state students and $45,000 at private nonprofit four-year institutions.

After grants and tax benefits, the average net tuition and fees paid by first-time, full-time students came to about $2,300 at public four-year colleges and $16,910 at private nonprofits. Inflation-adjusted published tuition at public four-year colleges fell 7 percent over the past decade.

Tuition is also only part of the bill. Net costs, including housing and food average $9,660 at community colleges and $32,830 at private four-year colleges. The New York Fed’s cost estimates do not include the expense of living on campus, and taking five or six years to finish can significantly raise the total.

The students who struggle most financially are rarely the ones at an in-state public university with moderate aid. They are more often the ones at a high-priced private school without a meaningful grant, or those who extend their enrollment by two years.

Out-of-state students face a different picture. Public four-year schools charge them an average of $31,880 in tuition and fees, which moves the economics closer to a private-school calculation.

Where a Degree Fails to Pay

Three conditions account for most of the cases in which college does not deliver.

Not finishing. About 61.1 percent of students who began college in fall 2019 earned a credential within six years, a rate that has held steady for four cohorts. Part-time starters completed at only 34.1 percent. The New York Fed economists have warned that students who start but do not finish bear some of the costs while collecting far fewer benefits.

A student with two years of tuition debt and no credential is the central failure case, and the risk is not evenly spread. Working adults, part-time enrollees, and students at open-access institutions face the lowest odds.

Major and program. The average conceals wide differences. Georgetown University’s Center on Education and the Workforce has long documented that earnings can differ by more than 300 percent between one major and another. However, that research is based on older Census data and the exact figures have shifted.

The direction has not. A degree in a field with strong employer demand and a licensure or technical pathway behaves differently from one with a diffuse labor market. This does not mean humanities majors should be written off. It means they should be paired with deliberate skill-building, internships, and a realistic view of starting pay.

Debt that outruns income. Total student debt reached $1.866 trillion in March, and 10.3 percent of balances were 90 or more days delinquent in the first quarter. About 9 million federal borrowers are in default, holding roughly $220 billion.

Repayment research consistently finds that borrowers who leave without a credential are among the most likely to default, which helps explain why the debt crisis and the “college pays off” finding can both be true.

The Entry-Level Squeeze

The strongest case against the degree is not about lifetime earnings. It concerns the first few years. The New York Fed reports unemployment for recent graduates aged 22 to 27 at about 5.6 percent, with underemployment near 42 percent.

Context matters here. In December 2019, recent-graduate unemployment was 3.9 percent and underemployment was 41 percent. Underemployment, meaning work in jobs that typically do not require a degree, has therefore hovered around 40 percent for years.

The unemployment rate has deteriorated, and the earlier cushion of a quick first job is thinner. Underemployment is also not always a dead end: in 2019, 36 percent of underemployed recent graduates held non-college jobs paying $45,000 or more.

Artificial intelligence is the open question. A Gallup and Lumina Foundation survey found 46 percent of Americans expect degrees to matter less over the next five years as AI spreads. Early academic work circulated this year argues that demand for college graduates relative to non-graduates has begun to contract, with AI exposure blamed for part of the decline.

Those findings are preliminary and contested, and current data still show a large premium. A prudent reading treats AI as a risk to entry-level, routine knowledge work specifically, which strengthens the case for choosing programs built on judgment, licensure, hands-on skills, or domain expertise.

New Federal Loan Rules Change the Math

The 2025 budget law, formally the One Big Beautiful Bill Act, rewrote the federal borrowing rules, and its main loan provisions began taking effect in mid-2026. Grad PLUS loans are closed to new borrowers, and a lifetime cap of $257,500 now applies to federal student loans, excluding Parent PLUS.

Professional students can borrow up to $50,000 a year with a $200,000 lifetime limit, while other graduate students face a $100,000 lifetime limit. Parent PLUS borrowing is capped at $20,000 a year and $65,000 per dependent student. The law also created the Repayment Assistance Plan and ended SAVE, PAYE, and ICR for new borrowing.

The practical effect lands on families who used Parent PLUS to close the gap at expensive private schools, and on students in costly graduate programs. Many will turn to private student loans, which carry no federal income-driven repayment or forgiveness pathways.

Anyone comparing private lenders should weigh fixed versus variable rates, cosigner-release terms, and the long-term cost of refinancing student loans, since refinancing a federal loan with a private lender permanently forfeits federal protections.

A second change targets program quality. The “Do No Harm” earnings test compares graduates’ earnings against a benchmark for people with lower education, and programs that fail in two of any three consecutive years lose access to federal student loans.

The Education Department projected that roughly 6 percent of programs would fail. The first results are expected in 2027. Prospective students can use the test’s logic now by checking program-level earnings and debt data in the Education Department’s College Scorecard before enrolling.

A Four-Gate Test Before Enrolling

A defensible decision can be made by running a specific program through four gates. Failing one does not necessarily rule it out, but failing two should prompt a different plan.

The first gate is net price over the full cost of attendance. The school’s net price calculator, not the published tuition figure, shows what a student with a given family income will pay, and the figure should include housing, books, and the realistic number of years to finish.

The second gate is completion odds. A school’s six-year graduation rate for students with a similar profile, including part-time or working students, is a far better predictor than the institution’s headline figure. An institution with a high published graduation rate may still fail students who match the applicant’s circumstances.

The third gate is program earnings. The median pay of the specific major at the specific school, measured several years after graduation, reveals more than national averages.

The fourth gate is a debt ceiling. A common rule of thumb among financial aid counselors holds that total borrowing should not exceed expected first-year salary. It is a heuristic rather than a law, but it keeps monthly payments within a manageable share of income.

When the Answer Is Probably No

A traditional four-year degree is hardest to justify at a high-priced school with minimal aid, in a program with weak earnings outcomes, for a student with a high risk of not finishing. In that combination, cheaper routes often win.

Community college with a transfer pathway sharply lowers net cost, and apprenticeships, licensure tracks, and certificate programs in high-demand trades can pay off within a few years. Analyses built on Georgetown’s return-on-investment data suggest that community college credentials tend to deliver faster payback, while bachelor’s degrees overtake them in cumulative earnings 15 to 20 years after enrollment in most fields. The tradeoff is speed against ceiling.

Two misconceptions muddy this decision. The first is that rising wages for non-degree workers prove college is unnecessary. The New York Fed has noted that wages for workers without degrees grew as labor markets tightened after the pandemic, but that reflected a cyclical shortage of workers rather than a structural shift.

The second is that prestige guarantees return. Georgetown’s work indicates that institutional prestige and ROI are not the same thing, and a modestly ranked public university with a strong program in a high-demand field can outperform an expensive name-brand school.

The Verdict

For a student who finishes, pays a moderate net price, and enters a field with real employer demand, the degree remains one of the better financial decisions available, with returns that outpace most alternatives and benefits that extend to health coverage, job security, and household stability.

For a student at risk of not finishing, borrowing heavily for a program with weak earnings, the same degree can be a net loss that compounds for decades.

The question worth asking in place of “is college worth it?” is whether this program, at this price, with this completion risk, is worth it. Courtney Brown, vice president of impact and planning at Lumina Foundation, has described the sector’s standing with the public as fragile, and the new loan caps and earnings tests suggest the market is moving toward that more specific question as well.

Students and families who answer it with their own numbers, rather than national averages or headlines, are the ones most likely to land on the right side of the 12.5 percent average.

What People Ask