There is no perfect way to fund higher education. Relying chiefly on tuition favors wealthy students and results in underfunding of the research mission of most colleges and universities because students who see themselves as customers understandably would prefer that the money they spend produce concrete benefits to them in the form of teaching and administrative services rather than add to the general storehouse of knowledge available to all of humanity.

Relying chiefly on donors risks directing an institution’s mission to the donors’ priorities, which will not always align with the priorities of faculty, students, administrators, and other stakeholders. A wealthy alum who is the CEO of an oil company might be happy to fund a new but mostly unnecessary stadium that bears his name or to endow a chair in economics but reluctant to fund the climate science initiative favored by the faculty, students, and administration. To be sure, canny university leaders gently nudge donors to support projects the former would fund in any event and then, because money is fungible, direct additional resulting resources where most desired, but this tactic is not always available.

Government funding of research carries the same risk of mismatch with internal priorities that donor funding does, and, as we have seen since the start of the second Trump administration, it carries additional risks because of the sheer size of the numbers involved. The single largest gift to a university was Michael Bloomberg’s 2018 one-time donation of $1.8 billion to Johns Hopkins University, his alma mater. But for decades Johns Hopkins has also been the recipient of the most federal grant money of any university, collecting $3.6 billion in just fiscal year 2024. Such largesse creates dependency. Thus, in 2025, in response to Trump administration cuts to the U.S. Agency for International Development and other cutbacks, Johns Hopkins lost $800 million in revenue, which led to thousands of layoffs, followed by a hiring freeze and other austerity measures.

To some small-government libertarians, the remedy is to greatly reduce the federal government’s role (and perhaps the role of state governments as well) in funding higher education (and much else). If money provides leverage, their argument goes, a substantial reduction in government spending results in a substantial reduction in leverage. The libertarian claim is not entirely wrong, but it is only a partial solution to the problem of government pressure, and it would throw the baby out with the bathwater.

The libertarian approach is at best a partial solution because the government has forms of leverage independent of direct funding. The government can increase taxes on endowment income, as it did in the “one big beautiful” tax-and-spending changes Congress enacted in 2025. It can deny visas to foreign students, as it attempted with respect to Harvard University in May 2025. Although a federal district judge preliminarily enjoined the Trump administration’s visa denials as unlawful retaliation against Harvard, the government has appealed that ruling, and, as we have seen in multiple settings, the likelihood of ultimately prevailing against the Trump administration is no guarantee that the administration will not cause targeted actors substantial interim damage. Beyond taxation and immigration, the government has other powers it could abuse to threaten universities and other knowledge-producing institutions. Ending or substantially reducing government funding would not prevent such abuse.

Moreover, ending or substantially reducing government funding of universities and other knowledge-producing institutions would be ill-advised quite apart from its impact on leverage that could be abused. Incentives to maximize shareholder value lead the private sector to underinvest in the kind of basic research that ultimately redounds to the benefit of the broad public. As noted above, other sources of funding (chiefly tuition and private donations) do not adequately fill the gap. Thus, as a coalition of leading public and private research universities put the point: “Since World War II, public sector investments in basic scientific research have formed the backbone of American innovation and ingenuity—helping our nation lead the world in scientific discovery, technological advancement, and economic growth and prosperity.”

Is there a way to provide government funding without risking undue governmental influence? Maybe. The key is to create dedicated funding streams over which the executive branch lacks the ability to exert influence.

In the past, that might have meant creation of an independent agency with disbursal responsibilities, such as the research councils one sees in the United Kingdom. By law (under the so-called Haldane principle), their “funding decisions are made independently of government.” However, this model will very soon be unavailable in the United States, because U.K. research councils, while acting independently of political interference, are nonetheless governmental entities. Given the very high likelihood that the Supreme Court will invalidate removal restrictions for the principals of nearly all independent agencies in Trump v. Slaughter, some other mechanism must be found.

Luckily, several alternative models are available. Perhaps the most straightforward would be for funds to be appropriated to a privately held escrow account or trust, much in the way that municipal bonds are made payable through irrevocable arrangements.

The Supreme Court has offered only the slimmest of grounds for distinguishing the Federal Reserve Board of Governors from other independent agencies, but it has pointed to the Fed’s “quasi-private” status. We can understand the significance of that status by reference to the ostensible constitutional harm caused by independent agencies: They assign executive power to government personnel and entities not answerable to the president via removal, but where power is given to a quasi-private entity, that power is more difficult to characterize as executive in nature. It follows a fortiori that empowering a truly private entity to make only ministerial disbursements would not undercut Article II’s Vesting Clause or its Take Care Clause, both of which are typically invoked to challenge independent agencies.

There are already some federal models for private entities to disburse public funds, such as Federal Home Loan Banks. To completely eliminate the risk of invalidation in the event that the Court articulates a very restrictive view in Slaughter, state-chartered entities could be deployed, although direct or indirect control by states could place funding at risk for those knowledge-producing institutions in states led by politicians who share the current presidential administration’s authoritarian proclivities.

I conclude with two caveats. First, I recognize that legally insulating federal funding streams from executive oversight would not eliminate all forms of leverage, just as the libertarian approach of eliminating such funding streams would not. But, as noted above, preservation of such funding is valuable in itself. A partial solution that keeps the baby alive is better than a partial solution that throws the baby out with the bathwater.

Second, one might worry that without the government’s ability to withhold appropriated funds, a future enlightened presidential administration would be unable to enforce Titles VI and IX of the Education Amendments of 1972. That need not be, however. For one thing, prior to the Trump administration, the Department of Education (DOE) had never deployed the ultimate sanction of a funding cutoff in response to Title VI or IX violations by a college or university. For another, Congress could amend federal law to switch the default—so that the administration would need to go to court and bear the burden of proving prove civil rights violations in order to obtain a court order to a disbursal entity to curtail payments.

That kind of a change might even garner support from Republicans in Congress who have sometimes complained about heavy-handed actions by the DOE under Democratic administrations. Although Republicans’ ire was mostly focused on the substance of DOE actions—involving rules governing the investigation of sexual assault allegations during the Obama administration and rules governing transgender status discrimination during the Biden administration—there could be bipartisan support for curtailing unilateral executive power in this domain more broadly.