Is College Still Worth It?

Net price is lower than the published price before grants, and lowest for the poorest students. At public four-year colleges in 2021–22, first-time full-time in-state students who received federal aid paid an average net price of $9,700 if their family income was $30,000 or less. Those with family incomes over $110,000 paid an average of $24,200 (both in 2022–23 dollars), because grant and scholarship aid shrinks as income rises.

But tuition has still grown far faster than household income. Public in-state tuition and fees doubled after inflation between 1995–96 and 2025–26, while median household income, also after inflation, grew 31% between 1995 and 2025.

Download this conclusion: tuition

The tuition finding as one image, all after inflation. It shows published tuition and fees at public four-year colleges in 1995–96 and 2025–26 ($5,940 and $11,950). It also shows the average net price after grants at its 2012–13 peak and in 2025–26 ($4,450 and $2,300).

Download the card (PNG, 1,200 × 630)

The 2025 Bureau of Labor Statistics figures are medians for full-time wage and salary workers aged 25 and over. They are averaged over eleven months, because October 2025 was not surveyed during the federal government shutdown.

The College Board finds that among full-time workers, four-year graduates have earned about 60% more than high school graduates for decades. The New York Fed puts the premium at about 68%, near its all-time high. Counting tuition, fees, books and the wages given up while studying, the New York Fed estimates the return on a degree at 12.5% a year in 2024.

These rates are for first-time, full-time students who started a bachelor’s degree at a four-year college in fall 2017. They count only degrees from the college where they started (National Center for Education Statistics). The National Student Clearinghouse counts students who finish anywhere. Of everyone who first started college in fall 2019, full-time or part-time, at any kind of college, 61.1% had earned a degree or certificate within six years. By then 9.0% were still enrolled, and about 30% had left without one.

Non-completers who borrow fare worst. In the Federal Reserve’s 2017 to 2019 surveys, 52% of people who had borrowed for college and left without a degree said they were doing at least okay financially. That compared with 67% of those who never went beyond high school and 77% of bachelor’s graduates who borrowed.

Among people who borrowed for a bachelor’s degree, not finishing was associated with a 34-percentage-point lower chance of doing at least okay. Lockwood and Webber, whose Federal Reserve note analyzes these surveys, stress that these gaps are not causal, since whatever makes students drop out may also hurt their finances.

The Foundation for Research on Equal Opportunity (FREOPP) estimates that for students who drop out, the median bachelor’s program returns negative $99,000.

“College” is not one investment. Electrical engineering and social work are both bachelor’s degrees, but they are not remotely the same financial bet. The New York Fed finds the return on a degree highest in engineering, math and computers, and business and economics. It is lowest in fine arts, liberal arts, leisure and hospitality and, above all, education, where pay often reflects a nine-month school year. Even so, it judges college a sound investment for the typical student in most majors.

Within every field, half of graduates earn less than the median. The New York Fed also looks at the 25th percentile of college earners (a quarter of graduates earn less). There the premium over the median high school graduate is well under $10,000 a year, and the return was 2.6% in 2024. So for at least a quarter of graduates, college does not appear to pay off.

FREOPP estimates that 23% of bachelor’s programs, weighted by enrollment, have a negative return on investment. Allowing for the chance of not finishing, their extra lifetime earnings do not cover tuition, books and the wages given up while enrolled. The median bachelor’s program returns about $160,000, and some programs at elite and flagship public universities return more than $3 million, such as business at the University of Virginia.

Field of study matters most, but completion matters too. Colleges with the lowest on-time completion rates have a median return below $50,000, against $367,000 where more than two-thirds finish on time.

In December 2025 the Department of Education began flagging, on the federal student aid form, colleges whose graduates earn less than typical high school graduates in their state. Inside Higher Ed reported that 1,365 institutions were flagged. That is about 23% of the nearly 5,900 in the Department’s database. It also reported that together they enroll fewer than 3% of undergraduates.

Recent graduates are those aged 22 to 27 with a bachelor’s degree or higher. The New York Fed counts a graduate as underemployed when working in a job that fewer than half of the people doing it say needs a bachelor’s degree.

People who go to college differ from those who do not in ways that would affect their earnings anyway: family background, academic ability, demographics and where they live. Allowing for those differences, FREOPP puts the premium at age 27 for the median bachelor’s program at 39%: $52,000 against $37,000. The unadjusted figures give 68%.

Dale & Krueger’s 1999 study found that students who went to more selective colleges earned no more than students admitted to comparable colleges who went to less selective ones. So much of the prestige premium reflects whom elite colleges admit. A study of about 700 identical twins, who share genes and family background, estimated a return of 9% for each year of schooling (Ashenfelter and Rouse).

The return to finishing a degree remains positive on average once these differences are allowed for, but it is smaller than raw comparisons suggest. It also varies enormously. This means both “college always pays” and “the premium is all sorting” are overstated.

Totals differ by source and measure. For June 2026, the Federal Reserve’s G.19 consumer credit series gives about $1.86 trillion for federal and private loans combined. That month, Federal Student Aid gives $1.72 trillion in the federal portfolio, across 42.3 million recipients. The New York Fed’s household debt report, built from consumer credit reports, gives $1.65 trillion for that month. They are measuring different things rather than disagreeing.

The mean and the median are also doing different jobs. The mean is the federal total divided by recipients, and a minority of large balances pulls it up to about $40,800. The typical borrower in the Federal Reserve’s 2024 survey owed $20,000 to $25,000. In 2020 a Brookings analysis found that the quarter of borrowers who went to graduate school held about half of all student debt.

Default risk is highest not among the biggest borrowers, but among those with small balances and weak completion outcomes. Among students who first entered college in 2003–04, 37% of those who borrowed $6,125 or less for undergraduate study had defaulted within 12 years. Of those who borrowed more than $24,000, 24% had.

The for-profit sector is a distinct risk profile, with much lower completion rates, much higher borrowing rates, and much higher default rates than the nonprofit sector. Among the same 2003–04 entrants, for-profit entrants defaulted at 47% within 12 years, against 13% for public two-year entrants. In other sectors, 12 to 13 of every 100 entrants defaulted. Counting only those who borrowed, the rates were 52% for for-profit students and 26% for public two-year students (Judith Scott-Clayton, Brookings).

Blanket anti-college arguments can mislead: much of the worst debt/default data comes from for-profits, not the system as a whole. The Federal Reserve’s 2024 survey shows the pattern.

In the Federal Reserve’s survey, 57% of borrowers said they were required to make monthly payments as of October 2024 (up from 37% in 2022). 20% said they were behind on payments or in collections, up from 16% in 2023. In June 2026 the Department of Education reported about 2,000 institutions where at least 25% of borrowers were more than 90 days behind on their federal loans. It counted borrowers who entered repayment between January 2020 and May 2025.

Electricians and plumbers usually train through apprenticeships. Most electricians’ apprenticeships last four or five years, with about 2,000 hours a year of paid on-the-job training. So they earn while they learn instead of giving up four years of wages. HVAC technicians usually need a postsecondary certificate and dental hygienists an associate degree.

The median electrician or plumber earns more than the bottom quarter of four-year graduates working full time. In 2024, 25% of women graduates earned less than $51,200. For men, 25% earned less than $62,300. But the median four-year graduate still out-earns most of these trades: $71,800 for women and $95,400 for men working full time, year-round, in 2024.

Electrical work is also physically demanding, with long periods of standing and kneeling, work at heights, and shocks, falls and burns among its common injuries.

The “skills-first revolution” has been real in rhetoric but modest in practice. The Burning Glass Institute and Harvard Business School found that dropping degree requirements raised the hiring of people without degrees by only 0.14 percentage points across all hiring. That is about 97,000 of 77 million hires a year, fewer than 1 in 700.

As of November 2025, about 19.3% of job postings on Indeed required a bachelor’s or more; 51% had no formal requirement. The share asking for a degree is slightly below its pre-pandemic level but up from 16.6% in November 2023, and Indeed’s posting data suggest that degrees still matter.

Degree holders live longer, by the estimates of the Institute for Health Metrics and Evaluation (IHME).

Graduates also vote and volunteer more. In the 2024 presidential election, 76% of 25- to 44-year-old citizens with at least a bachelor’s degree voted, against 42% of high school graduates the same age. In 2023, 38% of bachelor’s holders volunteered, against 17% of high school graduates (College Board). The College Board notes that not all of these differences are caused by education, though reliable studies support its significant role.

Students likely to finish, from any income background, who choose high-return fields such as engineering, computer science, nursing and economics, at colleges where most students finish on time. Price is not the main thing. The Foundation for Research on Equal Opportunity (FREOPP) finds that the colleges with the highest net tuition also have the highest median return. But a high price can turn a weak program negative.

Students unlikely to finish, borrowing for low-earning majors at low-completion or for-profit colleges with a high net price. The worst bet is not “college”. It is borrowing for a program you are unlikely to finish and that has weak earnings even if you do.

The decision is not “is college worth it?” It is: Is this college worth this price for this student in this major with this completion probability? The key questions: What is the 6-year graduation rate? What do graduates in this field earn? How much will I borrow? What share of borrowers repay? What would I do instead? What is the downside if I don’t finish?

Yes, for most people who finish. In 2025, full-time workers with a bachelor’s degree earned a median $1,578 a week, 63% more than high school graduates. Counting tuition and the wages given up while studying, the typical degree returns 12.5% a year, by the New York Fed’s estimate.

The average hides a large minority that a degree does not pay. The Foundation for Research on Equal Opportunity (FREOPP) finds that 23% of bachelor’s programs do not repay their cost once the chance of not finishing is counted. The New York Fed finds that for at least a quarter of graduates, college does not appear to pay off. The field matters most, then the college’s completion rate.

The clearest loss is starting and not finishing. About 30% of students who start college have left without a credential six years later. For them the median bachelor’s program returns negative $99,000 (FREOPP).

Download the conclusion

The verdict above as one image, with the two numbers that frame it. They are median weekly earnings in 2025 of full-time workers aged 25 and over: $1,578 with a bachelor’s degree and $966 with only a high school diploma. The image is 1,200 by 630 pixels, the size social networks use for link previews, and anyone may download it and post it.

Download the card (PNG, 1,200 × 630)