I looked up what college actually costs a few months ago, the real number, not the sticker price with a vague promise that aid will make it manageable. The average private university is running about $58,000 a year. Public out-of-state schools average around $43,000. Even in-state publics, the historically affordable option, are clearing $25,000 to $30,000 after you add room and board. Multiply the private number by four and you are looking at a financial decision larger than most people's first home purchase, to be made at 17, with no financial literacy curriculum behind you and a Common App deadline in front of you.
I am in the middle of this process right now. I have been thinking about it as an economics problem, which is maybe the most useful frame I have. This is what I found when I worked through the actual data.
Tuition at American private universities has risen roughly 8 times faster than general inflation since 1980. That is not one bad decade or a pandemic anomaly. It is a 46-year compounding trend that shows no sign of reversing. Understanding how it happened is not just intellectually interesting. It tells you which parts of the system are exploitable and which parts are just going to keep getting worse.
The first driver is the federal student loan system. In the 1960s and 1970s, the federal government expanded access to student loans substantially, initially with the goal of making higher education accessible regardless of family income. The effect was correct on the access dimension. The unintended consequence was price. When any buyer can borrow essentially unlimited amounts to pay for a product, and that borrowing is guaranteed by the government regardless of whether the education produces enough earning capacity to repay it, sellers learn quickly that they can raise prices without losing applicants. The loans show up and cover it. Economists call this the Bennett hypothesis, named after a 1987 New York Times op-ed by then-Secretary of Education William Bennett. The empirical evidence for it has gotten substantially stronger since then.
The second driver is the U.S. News and World Report rankings system, which sounds like a small thing and is not. Since the early 1990s, U.S. News has ranked universities on a methodology that rewards research output, faculty credentials, endowment size, selectivity, and campus resources. It does not reward graduating students with low debt loads, good employment outcomes at their actual earnings levels, or tuition that has grown slower than inflation. So schools competed on what the rankings measured. They hired more researchers, built better facilities, became more selective, and charged more for all of it. The students paying those costs are buying a product optimized for its ranking position rather than its return on their specific investment.
The third driver is administrative expansion. The ratio of non-instructional administrative staff to students at American universities roughly doubled between 1990 and 2020. Chief diversity officers, vice provosts for student success, compliance departments, enrollment management teams, student experience staff. Many of these functions are genuinely useful. None of them teach classes. All of them are paid from tuition revenue. A university that enrolled 10,000 students in 1990 and 11,000 in 2020 might now have three times as many administrators. The students in 2020 paid for that growth whether they knew it or not.
The average lifetime earnings premium for a bachelor's degree holder compared to someone who stopped at high school is about 65%, according to the most recent Georgetown Center on Education and the Workforce analysis. Over a 40-year career, that gap accumulates to roughly $900,000 in present value. So in aggregate, across all majors and all schools, college still pays.
But that average conceals a range wide enough to make the aggregate number almost useless for individual decision-making. The premium varies enormously by field, by school, by career trajectory, and critically by what you actually paid.
| Major | Estimated lifetime earnings premium | Payback period at $232k total cost |
|---|---|---|
| Engineering | $1.2M+ | 3 to 5 years post-graduation |
| Computer Science | $950k | 4 to 6 years |
| Nursing / Health | $800k | 5 to 7 years |
| Business (Finance track) | $700k | 7 to 9 years |
| Education | $250k | 13 to 16 years |
| Humanities (median) | $150k | 18 to 25 years |
A humanities degree from a school where you paid $80,000 total is a completely different financial decision from the same degree at $232,000. The payback period calculation changes dramatically. Whether that matters to you personally depends on what you intend to do with the degree, but it is information worth having before you commit rather than after.
Community college and transfer is the most financially rational path that gets the least social respect among the kind of people I go to school with. Average community college tuition runs about $3,800 per year. You complete two years of general education requirements at that cost, transfer to a four-year institution, and graduate with the same credential. The diploma says where you finished, not where you started. Nobody checks. In most professional fields nobody cares. Realistic savings compared to starting at a private university: $80,000 to $130,000, which is also the amount of money that could go into a Roth IRA and compound for 40 years instead of into interest payments.
In-state public universities cost roughly half what comparable private schools charge for the same credential. For the large majority of careers the marginal employment premium of a private school over a well-regarded public is either zero or too small to justify the price difference. There are exceptions, specifically certain investment banking and consulting pipelines, specific graduate programs with strong legacy preferences, and a small number of elite research universities where the network value and faculty access are genuinely different. Outside of those narrow cases, the premium is mostly constructed reputation rather than measurable outcome difference.
Financial aid offers are also negotiable more frequently than students realize. If you have a competing offer from a school with similar rankings, bringing it directly to your preferred school's financial aid office and asking them to review your package works often enough to be worth the awkward conversation. The worst outcome is that they say no and your offer stays the same.
The framing that gets students in trouble is treating college as a binary: go or don't go. The actual decision has at least three variables that matter independently. Where you go, what you study, and what you pay are separable choices with separable financial consequences, and optimizing on one without thinking about the others produces suboptimal outcomes.
A more useful frame: what is the all-in total cost after every source of aid, what do graduates from this specific program at this specific school actually earn in the first five years, what loan payment does the remaining debt require, and does that math work given what I realistically want to do? These are answerable questions. You can look up graduate salary data for specific programs at most universities through the College Scorecard. You can model the loan payment on any federal student loan calculator. You can compare net price across schools once you have actual aid offers rather than sticker prices.
The thing I find genuinely strange about how this works is that teenagers are expected to make this decision largely in an informational vacuum, with status signaling from peers and parents playing a much larger role than data. The school whose name sounds most impressive at a dinner party is not necessarily the school whose graduates have the highest earnings-to-debt ratios five years out. Those rankings exist and almost nobody looks at them. I think that is worth changing, starting with the conversation you have with yourself before you click submit.