The easy version of this argument goes like this: the government offered more student loans, colleges raised tuition, therefore student loans caused expensive college.
Reality is more complicated. But the mechanism deserves serious attention, because where it works, it works exactly the way the incentives predict.
What the research actually found
A study published by the Federal Reserve Bank of New York (Lucca, Nadauld and Shen) looked at the 2007–08 increase in federal student-loan limits. Schools whose students were more likely to be at the old limits raised their sticker tuition by about 60 cents for each additional dollar of subsidized loan capacity, and by about 40 cents per dollar for unsubsidized loans. Institutional grants didn’t rise to offset it.
Two precise caveats matter. It measures sticker price, not what the average student actually pays after grants. And it comes from one policy episode, with the strongest effects at private and vocational programs.
Other studies fill in the picture:
| Where aid goes | What researchers found |
|---|---|
| For-profit colleges | Aid-eligible for-profits charge about 78% more than comparable for-profits whose students can’t get federal aid — roughly the size of the aid itself (Cellini & Goldin, 2014) |
| Graduate programs | The 2006 Grad PLUS expansion raised borrowing and prices in previously constrained programs, without significantly improving completion (Black, Turner & Denning, 2023) |
| Pell Grants at public schools | Public institutions capture under 5 cents of each Pell dollar; selective private nonprofits offset much more through lower institutional aid (Turner, 2017) |
That doesn’t mean loans explain every tuition increase. State funding changes, labor costs, amenities and administration matter too. And loans help students attend college who otherwise could not. On average, college still produces substantial economic returns, though averages hide wide differences between programs.
There’s also a genuinely surprising number in the other direction. According to the College Board, the average inflation-adjusted net tuition paid by in-state students at public four-year universities fell from $4,450 in 2012–13 to about $2,300 in 2025–26, even as sticker prices kept rising. “College keeps getting more expensive” is true of the listed price. It isn’t true of what many public-university students actually pay.
Why the incentive works the way it does
Imagine selling a product to an 18-year-old buyer.
The buyer does not pay the full cost today. A third party lends the money. The loan follows the buyer to almost any accredited, aid-eligible school, and only a small share of programs have ever lost access because of poor outcomes.
And unlike most consumer debts, student debt faces an unusually high hurdle in bankruptcy: federal law (11 U.S.C. §523(a)(8)) generally requires the borrower to prove “undue hardship” in a separate court proceeding.
What incentive does the school face to lower its price?
Where students compare prices and have cheap alternatives, quite a lot — which is why public universities capture so little of Pell aid. Where the buyer is borrowing-constrained and the loan limit is the binding number, much less.
Skin in the game
Now change one element. Suppose institutions themselves absorbed part of the losses when graduates of their programs consistently couldn’t repay.
Suddenly the school cares more directly about tuition, graduation, employment, earnings and program value.
Current law is moving partway in that direction, through eligibility rather than loss-sharing. Schools already lose loan access if default rates are too high for too long. The 2025 federal budget law (the One Big Beautiful Bill Act) adds an earnings test: degree programs whose graduates fail it in two of three years lose federal loan eligibility. The same law closes Grad PLUS to new borrowers on July 1, 2026 and caps graduate borrowing. Whether graduate prices fall as a result is one of the most useful natural experiments to watch.
Subsidize the buyer, or create another supplier?
There is another possibility worth exploring. If taxpayers already subsidize higher education heavily, why not also compete with expensive institutions directly?
Imagine a national low-cost public university infrastructure: excellent professors teaching core material nationally, digital libraries, online lectures and rigorous examinations, with local laboratories, apprenticeships and clinical training where physical presence matters. Students could still choose traditional universities. But the subsidized system would also create a low-cost competitor.
That is a proposal, not a finding. Low-cost suppliers already exist: average published tuition at public two-year colleges is $4,150 in 2025–26, compared with $11,950 at public four-year schools (in-state) and $45,000 at private nonprofits. The open question is whether a credential from a cheaper supplier can earn the same trust from employers.
Subsidizing the buyer increases purchasing power. Creating another supplier increases supply. Those policies do very different things.
That’s the same mechanism behind subsidizing demand while restricting supply — and you can watch it play out in the ten-houses simulator.
The Zesht box
Follow the incentive
- What behavior does this reward?
- Enrolling and borrowing now, repaying later. For schools, pricing up to what students can borrow, especially in programs where students hit loan limits.
- What does it discourage?
- Shopping hard on price, and schools cutting costs to compete for students who can simply borrow the difference.
- What becomes more expensive?
- Sticker tuition at the most exposed schools: private, for-profit and vocational programs, and graduate programs after the 2006 Grad PLUS expansion.
- What becomes scarcer?
- Price discipline in programs where aid is plentiful and outcomes are weak. Cheap options still exist: average public two-year tuition is $4,150 in 2025–26.
- Who receives the benefit?
- Students gain access and financing. Schools capture a share of the aid — large at some for-profits and selective privates, small at public universities.
- Who ultimately pays?
- Borrowers, through debt that is unusually hard to discharge, and taxpayers, through subsidies, defaults and forgiveness.
- What happens when people have another option?
- Where low-cost substitutes exist and students compare net prices, capture falls. That is consistent with public schools capturing under 5 cents of each Pell dollar.
Goal → mechanism → result
The scorecard
- 1The goal
- Let students who lack cash or credit pay for college, widening access to an education that raises lifetime earnings on average.
- 2The mechanism
- Federal loans with annual and lifetime limits, available at any aid-eligible school, backed by the Treasury and hard to discharge in bankruptcy.
- 3The incentive
- When a third party finances the purchase and the student repays later, schools face buyers who are less sensitive to price.
- 4The observed result
- After the 2007–08 loan-cap increase, more-exposed schools raised sticker tuition about 60 cents per subsidized-loan dollar; aid-eligible for-profits charge about 78% more than ineligible peers.
- 5The unintended consequences
- Part of the aid shows up as higher prices rather than lower costs to students, and debt accumulates in programs with weak earnings. The Grad PLUS expansion raised borrowing and prices without improving completion.
- 6What supporters say
- Loans expand access for students who could not otherwise attend, net prices at public universities have fallen in real terms, and tuition growth also reflects state funding cuts and labor costs.
- 7What critics say
- Rising or uncapped loan limits let schools raise prices while students and taxpayers carry the risk; schools bear little of the cost when graduates can't repay.
- 8What the evidence actually shows
- Capture is real but uneven: strongest at for-profits, selective privates and graduate programs; weak at public institutions. Headline estimates come from specific policy changes and measure sticker prices.
- 9What remains uncertain
- Whether ending Grad PLUS for new borrowers (July 1, 2026) and the new program earnings test will lower graduate prices, and how far the 2007–09 results generalize.
Questions people ask
What is the Bennett hypothesis?
It is the idea, associated with former Education Secretary William Bennett, that increases in federal student aid allow colleges to raise prices. The evidence varies by sector: aid-eligible for-profit colleges charge about 78% more than comparable ineligible ones, while public universities capture very little of Pell Grant aid.
Do student loans raise tuition?
A New York Fed staff report found sticker tuition rose about 60 cents per $1 increase in subsidized loan limits after the 2007–08 cap increase, and about 40 cents for unsubsidized loans, with the largest effects at private and vocational schools. It measures sticker prices during one policy change, not all tuition growth.
Can student loans be discharged in bankruptcy?
Yes, but only if the borrower proves in a separate court proceeding that repayment would impose an undue hardship (11 U.S.C. §523(a)(8)). Since November 2022, the Justice Department has used a standardized attestation process and may support full or partial discharge.
Is college actually getting more expensive?
Sticker prices keep rising; the 2025–26 average is $11,950 at public four-year schools for in-state students. But average inflation-adjusted net tuition at public four-year schools fell from $4,450 in 2012–13 to about $2,300 in 2025–26, according to the College Board.
Sources
- Credit Supply and the Rise in College Tuition: Evidence from the Expansion in Federal Student Aid Programs (Staff Report 733)
- Does Federal Student Aid Raise Tuition? New Evidence on For-Profit Colleges
- The Economic Incidence of Federal Student Grant Aid
- PLUS or Minus? The Effect of Graduate School Loans on Access, Attainment, and Prices
- Trends in College Pricing 2025: Highlights
- GEN-14-03: Impact of Cohort Default Rates on Institutional Title IV Program Eligibility
- Federal Student Loan Program Provisions Effective Upon Enactment Under the One Big Beautiful Bill Act
- How Do College Programs Measure Up Against the One Big Beautiful Bill Act's New Accountability Standard?
- 11 U.S. Code § 523 — Exceptions to discharge
- Department of Justice's New Process for Student Loan Bankruptcy Discharge Cases (fact sheet)
Cite or link to this article
See something wrong? Zesht changes the theory when the evidence changes. Send a correction · Our standards