Elizabeth Warren wants free college for every American. But what the Massachusetts senator doesn’t seem to realize is just how much more costly college would get if her “free” proposal passed.

Shortly after Valentine’s Day in 1987, Education Secretary William J. Bennett wrote a now-famous op-ed in The New York Times titled “Our Greedy Colleges.” In it, he suggested that “increases in financial aid in recent years have enabled colleges and universities blithely to raise their tuitions, confident that federal loan subsidies would help cushion the increase.”

This observation became known as the “Bennett Hypothesis.” As the years go by, it seems more apt to call it the Bennett truism.

In the last 20 years, the federal government’s total spending on student loans has skyrocketed, from $24.8 billion in the 1995-96 school year to $93 billion in 2017-18.

At the same time, the price of college tuition has soared. Between 1998 and the present, tuition at four-year institutions has roughly doubled, and at private four-year colleges tuition has gone up 58%.

The price increase is even more dramatic looking at the last 40 years. Since 1980, the cost of attending a four-year public university has increased 287%—an uptick rate surpassed only by increases in the cost of medical care.

Enter Warren.

On Monday, she published a proposal that includes the following:

  • Students with household incomes below $100,000 would have the first $50,000 of their student loan debt canceled.
  • For every additional $3 of income over $100,000, the amount of loan forgiveness offered would be cut by $1.
  • Borrowers from families earning more than $250,000 annually would receive zero debt cancellation.

On the whole, as Robert VerBruggen has pointed out, her proposal would cancel all loans for about 75% of borrowers and provide partial cancellation for 95% of borrowers.

This debt cancellation portion of the plan would cost taxpayers $640 billion, as Warren pointed out herself.

And that’s just the retroactive part of the proposal.

The plan would also provide “universal free college,” allowing students to attend a two- or four-year college “without paying a dime in tuition or fees,” as she says. The total tab? $1.25 trillion over just the next decade.

Warren suggests her “free” college and debt cancellation plan would be financed (again) by an “ultra-millionaire tax,” singling out the 75,000 families in America she estimates to have more than $50 million in assets.

This is a group she has already identified to finance her “free” childcare plan. Things are getting expensive in a hurry.

Her latest proposal is problematic for a host of reasons, not least of which is the exorbitant cost to taxpayers. But it would also fail to achieve the goal of greater equality in access to education. A similar proposal for “free” college was already tried in England, and it ended up benefiting the wealthy rather than the needy.

But beyond these failures, Warren’s proposal would likely expedite the rise in college tuition. It comes down to simply math: When colleges know the federal government is financing “free” tuition in perpetuity, they’ll have all the more reason to raise tuition and fees, which taxpayers will then absorb.

In fact, a growing body of literature has already shown that federal subsidies have this tendency to push tuition prices higher.

In one study, researchers Grey Gordon and Aaron Hedlund found that raising subsidized loan limits led to a 102% increase in tuition from 1987 to 2010. Absent that additional federal money, the authors estimate tuition would have only gone up by 16% on net.

Another study by David O. Lucca, Taylor Nadauld, and Karen Shen of the Federal Reserve Bank of New York found additional evidence of the Bennett Hypothesis at play. The authors found that credit expansion (increasing subsidized federal student loans) leads to a tuition increase of 60 cents for every additional dollar of subsidized federal loans. Their conclusion bluntly states:

… a credit expansion will raise tuition paid by all students and not only by those at the federal loan caps because of pecuniary demand externalities. Such pricing externalities are often conjectured in the context of the effects of expanded subprime borrowing on housing prices leading up to the financial crisis, and our study can be seen as complementary evidence in the student loan market.

As Carlo Salerno of CampusLogic points out, students choose to take on college loan debt, and are not assigned that debt. So loan forgiveness “unfairly rewards the person who borrows to get a Ferrari over the one who got a Kia.”

That inequity is underscored by the numbers. As Salerno calculated, a wealthy student who borrowed $100,000 a few years ago and has been delinquent on repayment would get more forgiveness than the low-income student who responsibly worked to pay down $40,000 in debt over the past 20 years and only has $10,000 remaining, which would be forgiven.

Some would clearly benefit from this scheme, but it would penalize students who choose to work while in college to minimize their debt, those who pursue an apprenticeship over an expensive degree, and those who take out debt, but live modestly post-graduation in order to fully pay back what they owe.

Moreover, as the Urban Institute found (in an analysis unrelated to the Warren plan), “the top 25% of American households by income hold nearly half of all student debt—and the bottom 25% holds just a tenth of it. Canceling all student loans would deliver $5 to rich Americans for every $1 given to poorer families.”

Proposals to make college “free” or to forgive vast amounts of student loan debt reward one entity more than any other: the universities.

Subsidizing the already-dysfunctional student loan system is not the solution. If we want to get serious about addressing the student loan issue, we must pursue structural changes to accreditation, along with innovation in financing through options like income share agreements. Making sure colleges have some “skin in the game” also holds promise.

But above all, Washington should get out of the student loan business. The federal government currently originates and services 90% of all student aid, leaving taxpayers greatly exposed when defaults occur or when loan forgiveness becomes more generous.

Getting the feds out of the student loan business would go a long way toward finally addressing the root causes of soaring tuition.