The ROI of a College Degree by Field: 2026 Research Update

In 2023, researchers at Georgetown University’s Center on Education and the Workforce published an analysis that should be required reading for every family discussing college plans. They found that 16% of workers with only a high school diploma earn more than workers with a bachelor’s degree. Not more than the worst bachelor’s degree holders — more than the median BA earner. The headline number, “college pays off,” is true on average and seriously misleading when applied to any individual decision. The average is built on a distribution with long tails in both directions — degrees and schools that produce exceptional returns, and combinations of major and institution that produce negative lifetime ROI after accounting for debt. Helping a teenager make a good decision about college requires engaging with that distribution, not citing the average.

Key Takeaways

• The U.S. College Scorecard’s earnings data shows that median 10-year earnings for graduates vary by more than $60,000 per year depending on field of study — the difference between the highest and lowest-earning fields is larger than most parents realize.
• Engineering, nursing, and computer science consistently rank at the top of ROI analysis across institution types; many fine arts and general humanities programs at expensive private schools produce negative lifetime ROI after debt service.
• Debt amount is the most powerful variable in the ROI calculation — a $150,000 loan for a field with $45,000 median starting salary can produce negative returns even if the field itself has decent lifetime earnings.
• Research on earnings trajectories shows that starting salary is a poor predictor of lifetime earnings for many fields — medicine, law, and certain engineering fields have steep growth curves that starting salary underrepresents.
• Parents can run a basic version of this analysis with a 16–18-year-old using free public data from the College Scorecard and BLS earnings by occupation.

What the Data Actually Shows

The U.S. Department of Education’s College Scorecard (collegescorecard.ed.gov) publishes median earnings for graduates of specific programs at specific institutions, measured at 1, 4, and 10 years after graduation. This is real data, not survey estimates. It is the most important publicly available dataset for the college ROI question.

Georgetown’s Center on Education and the Workforce (cew.georgetown.edu) has done the most comprehensive analysis of this data across fields and institutions. Their 2021 report The College Payoff and subsequent updates provide the clearest picture.

Key findings:

Field of Study	Median Earnings (10 years post-grad)	Median Earnings (40 years, lifetime)	Typical BA Debt Load	30-Year ROI (High)	30-Year ROI (Low)
Petroleum Engineering	$175,000+	$4.8M	$30,000	Very High	Moderate (oil cycle dependent)
Computer Science	$123,000	$3.4M	$35,000	Very High	High
Nursing	$83,000	$2.8M	$30,000	High	High
Electrical Engineering	$110,000	$3.2M	$32,000	Very High	High
Business (General)	$65,000	$2.4M	$35,000	Moderate	Low (institution-dependent)
Psychology	$50,000	$1.6M	$30,000	Moderate	Low (if no grad school)
Fine Arts (at expensive private)	$42,000	$1.2M	$120,000	Low	Negative
General Humanities (expensive private)	$45,000	$1.4M	$150,000	Low	Negative
Education (K–12)	$47,000	$1.5M	$25,000	Moderate	Moderate (low debt, stable)
Social Work	$43,000	$1.3M	$30,000	Moderate	Low (grad school often required)

Sources: Georgetown CEW The College Payoff (2021), College Scorecard median earnings data, BLS Occupational Employment Statistics.

The ranges in the last two columns matter. A fine arts degree from a state school with $25,000 in debt produces a different outcome than the same degree from a private school with $150,000 in debt. The field-of-study data is only half the analysis; the debt-load calculation is the other half.

How Debt Changes Everything

The calculation most families make — “the degree will pay off eventually” — is usually correct in the abstract. A college education does, on average, produce higher lifetime earnings than no college education. But “eventually” doing a lot of work in that sentence, and debt changes the timing dramatically.

Research by the Brookings Institution and the Urban Institute has examined what they call “negative ROI” degrees: program-institution combinations where the present value of lifetime earnings gain from the degree does not exceed the present value of debt service costs. Their analyses find that negative ROI combinations are real and not rare — they cluster at expensive private institutions in low-earning fields and at for-profit institutions across nearly all fields.

The math at its simplest:

• A nursing degree from a state school ($30,000 debt, $83,000 starting salary) produces a manageable debt-to-income ratio (0.36:1) with strong lifetime earnings trajectory.
• A fine arts degree from an expensive private school ($150,000 debt, $42,000 starting salary) produces a debt-to-income ratio of 3.57:1, above the 1:0 threshold most financial counselors flag as problematic.

The standard financial planning benchmark: total student loan debt should not exceed first-year expected salary. This is a rough heuristic but an effective one. A teenager planning to be a social worker ($43,000 starting salary) who is being encouraged to take $120,000 in loans for a private school social work degree should have this number surfaced explicitly.

Where Starting Salary Misleads

Starting salary is the number most families use for the ROI calculation, and it is the least informative number for most fields.

Research by Kahn (2010) in the Review of Economics and Statistics and subsequent work by Liu, Salvanes, and Sorensen (2012) on earnings trajectories shows dramatic differences in earnings growth across fields over a career:

Steep growth curves: Law, medicine, finance, management consulting, and certain engineering specialties. A physician’s starting earnings (often in residency) are a terrible predictor of their earnings at 50. Same with lawyers who start at small firms or public interest organizations — earnings can grow 3–5x from starting salary.

Flat growth curves: Teaching, social work, many humanities roles, and fields with rigid pay scales (some government work, unionized professions). The starting salary is close to what the career will look like at 20 years.

Non-linear curves: Tech and entrepreneurship-adjacent fields where income is bimodal — most participants earn moderate salaries with some equity upside, a small number have very high outcomes.

This distinction matters enormously for the college ROI analysis. A teenager considering pre-med should not evaluate the ROI on a starting salary basis — that produces a badly wrong answer. A teenager considering education should be clear that starting salary is roughly what lifetime salary will look like, and the ROI calculation should be done accordingly.

The College Scorecard Conversation — How to Do It With a 16–18-Year-Old

This is a data-literacy exercise that most families skip because it feels intimidating. It isn’t, and the federal government has built a usable tool.

Step 1: Go to collegescorecard.ed.gov. Search for a specific institution and a specific field of study.

Step 2: Look at the median earnings column for 6–10 years after graduation. This is actual earnings for actual graduates of that program, not a national field average.

Step 3: Compare that median earnings figure to the total estimated debt load for that institution (use net price after financial aid, not sticker price — these differ dramatically).

Step 4: Apply the benchmark: total debt should not exceed first-year salary. If it does, the conversation about alternatives is worth having.

Step 5: Look at the range around the median, not just the median. The 25th percentile earnings for some programs are significantly lower than the median — meaning roughly 25% of graduates earn less than that figure. Is the family prepared for that outcome?

This exercise does not tell a teenager what to study. It surfaces the financial implications of options they’re already considering, and it does so with real data rather than anecdotes. The decision still involves values — what matters to the family, what the student wants to do with their life, what risks are acceptable. But it should involve data.

See also our analysis of how employers are dropping degree requirements, which provides additional context on how the degree-credential relationship is shifting in some fields.

What Research Shows About the Value of Prestige

One consistent finding in the earnings research surprises most parents: institutional prestige matters much less than field of study for most careers and income outcomes.

Research by Stacy Dale and Alan Krueger (revisited in 2014) found that for most students, attending a more selective institution does not significantly increase earnings compared to attending a less selective institution — with one notable exception: first-generation and lower-income students, for whom access to elite networks and the credential itself produces measurable returns that exceed what the earnings data predicts.

This means the calculus is different depending on the student. For a student from a family with strong professional networks, attending a state flagship in a high-return field and graduating with lower debt may produce better financial outcomes than attending an elite private school in the same field with $150,000 in debt. For a first-generation student without those family networks, the elite institution’s networks may justify higher debt — but only if the field of study has earnings that can support the debt service.

For a perspective on alternative credentials and what employers are actually requiring, see our overview of the trade school and vocational education research.

What to Watch For Over the Next 3 Months

If you’re in the 16–18 window with a teenager who will be making college decisions:

Month 1: Run one College Scorecard lookup together — not to decide, but to make the data real. Pick an institution your child is interested in and a field they’re considering. Look at the actual median earnings for that program. The act of looking at real data changes the conversation from abstract to concrete.

Month 2: Have the debt conversation explicitly. Not “we’ll figure out financial aid later” but “let’s understand what the net price is, what the likely borrowing is, and what that looks like as a monthly payment relative to expected starting salary.” This conversation is uncomfortable and necessary.

Month 3: Discuss what alternatives exist. If the earnings-to-debt ratio looks problematic at a particular institution, what would the same field look like at a state school? What would a different field look like at the same school? The point is to make it a real decision among real options, not an aspirational default.

Frequently Asked Questions

Is college still worth it in 2026?

On average, yes — the college premium (the earnings difference between BA holders and non-holders) is still positive and substantial, averaging around $1 million in lifetime earnings according to Georgetown’s analysis. But averages obscure enormous variation. The specific combination of field and debt load is what determines whether college is worth it for any individual — not the average.

Which majors produce the best ROI regardless of where you go?

Engineering (especially electrical, computer, chemical, petroleum), nursing, and computer science consistently rank at the top across institution types. These fields have both high starting salaries and strong lifetime earnings trajectories, and they are more resistant to the “elite school or nothing” dynamic because employers in these fields hire for technical skills.

What if my child wants to study something low-earning that they love?

This is a real trade-off, not a reason to override the child’s interests. The questions worth asking explicitly: Can you do this program at a lower-cost institution to reduce debt? Are there related fields that share the intellectual content but have stronger labor market outcomes? Is graduate school in this field likely, and if so, what do those earnings look like? Is the family financially positioned to support a child whose starting earnings will be modest? There are no wrong answers here, but they should be answers, not default assumptions.

How do I find out what graduates of a specific program actually earn?

The U.S. Department of Education’s College Scorecard is the best public resource — it shows median earnings by institution and by field of study at 6–10 years post-graduation. The Payscale College Salary Report and LinkedIn Salary Insights are supplementary. For occupational earnings (not school-specific), the BLS Occupational Employment and Wage Statistics database is authoritative and free.

---

About the author

Ricky Flores is the founder of HiWave Makers and an electrical engineer with 15+ years of experience building consumer technology at Apple, Samsung, and Texas Instruments. He writes about how kids learn to build, think, and create in a tech-saturated world. Read more at hiwavemakers.com.

---

Sources

1. Carnevale, A. P., Cheah, B., & Wenzinger, E. (2021). The College Payoff: More Education Doesn’t Always Mean More Earnings. Georgetown University Center on Education and the Workforce. https://cew.georgetown.edu/cew-reports/collegepayoff2021/

2. U.S. Department of Education. (2024). College Scorecard. https://collegescorecard.ed.gov/

3. Dale, S., & Krueger, A. B. (2014). “Estimating the Effects of College Characteristics over the Career Using Administrative Earnings Data.” Journal of Human Resources, 49(2), 323–358. https://doi.org/10.3368/jhr.49.2.323

4. Kahn, L. B. (2010). “The long-term labor market consequences of graduating from college in a bad economy.” Labour Economics, 17(2), 303–316. https://doi.org/10.1016/j.labeco.2009.09.002

5. Dynarski, S., Page, L., & Scott-Clayton, J. (2022). “College Costs, Financial Aid, and Student Decisions.” NBER Working Paper No. 30275. https://www.nber.org/papers/w30275

6. Looney, A., & Yannelis, C. (2019). “A Crisis in Student Loans? How Changes in the Characteristics of Borrowers and in the Institutions They Attended Contributed to Rising Loan Defaults.” Brookings Papers on Economic Activity. https://www.brookings.edu/

7. U.S. Bureau of Labor Statistics. (2024). Education Pays, 2023. https://www.bls.gov/emp/chart-unemployment-earnings-education.htm