ANALYSISApril 13, 2026
Should public funding follow students to private schools through vouchers?
In July 2025, President Trump signed the 'big, beautiful bill' into law, establishing the first federal school voucher program in U.S. history through tax credit scholarships for private school tuition. Simultaneously, Texas enacted what is described as the largest universal voucher program in U.S. history, allocating $1 billion over two years for more than 5 million students. These federal and state developments have intensified a national debate already playing out across 30 states where 114 bills were introduced in 2025 to expand, revise, or limit private school choice.
When a child leaves a struggling public school for a private one, should the tax dollars attached to that child follow them — or does redirecting public money to private institutions, many of them religious, quietly defund the schools that can't turn anyone away?
Not familiar with this story? Get context →
Who vouchers actually serve
C
The equity argument for vouchers is simple: a child trapped in a failing school is already paying the price for a monopoly that serves adults before students. Milwaukee's original 1990 program was specifically designed for low-income families with no exit options — this was never conceived as a subsidy for the wealthy. The question is whether we allow institutional loyalty to override actual family needs.
L
Milwaukee 1990 is not what's being built in 2024. In Florida, 69% of new voucher recipients were already enrolled in private schools — not students escaping failing institutions, but families who had already made their choice and are now asking taxpayers to fund it retroactively. That's not a rescue program; it's a tuition discount for families who needed no rescuing.
C
That Florida number is a genuine design failure, and conservatives should own it — but it indicts this specific program's structure, not the principle. The fix is prior public enrollment requirements and income phase-outs, not preserving a system where zip code determines educational destiny.
L
When Oklahoma shows fewer than 10% of applicants came from public schools, the competitive pressure that justifies the entire Friedman model is simply absent — you can't fix your way to a program that exists mostly on paper.
Academic outcomes record at scale
C
Both sides selectively fish the research pool. Early targeted programs in Milwaukee and D.C. showed modest positive results — that's not a minor asterisk, it's proof of concept. The later failures in Louisiana, Indiana, and Ohio reflect design problems in those specific programs, not evidence that choice itself is wrong.
L
You're asking us to credit Milwaukee and ignore Louisiana, Indiana, Ohio, and Michigan — all of which produced state-commissioned evaluations showing voucher students performing no better and sometimes worse than public school peers. If the positive results are proof of concept, the negative results at scale are proof the concept doesn't travel.
C
The honest liberal position — which their own rebuttal concedes — is that well-designed, income-limited vouchers for students in failing schools can produce positive outcomes. If you've granted that, you've conceded the principle and we're arguing about implementation, which is exactly the right argument to be having.
L
Conceding that targeted programs might work is not conceding that the universal, uncapped, billion-dollar programs currently being enacted are sound policy — that conflation is doing a lot of work in the conservative case.
Fiscal cost and accountability structures
C
The fiscal concern about uncapped tax credit scholarships is real — a range of $4 billion to $51 billion in projected costs reflects genuine uncertainty about program scale, and no serious conservative should wave that away. But notice what that argument concedes: it's a case for better program design, not for preserving the status quo.
L
A thirteen-fold range of uncertainty — $4 billion to $51 billion — should disqualify a program at the federal level before design fixes are even on the table. And because these credits route through donor intermediaries rather than direct appropriations, they bypass the congressional budget oversight that governs every dollar spent in public schools. That's not a design flaw; that's the design.
C
Routing through intermediaries is a feature critics treat as a bug — it's precisely the mechanism that keeps the program outside direct government control, which is a legitimate policy goal, not a conspiracy to evade oversight. The answer is transparency requirements for scholarship-granting organizations, not eliminating the model.
L
Transparency requirements don't substitute for hard annual caps, and 'we'll add accountability later' is how programs that started as targeted pilots became the uncapped universal programs conservatives are now defending.
Obligation to serve all students
C
The conservative principle is that parents, not bureaucracies, are the primary educators of children — and a government that mandates school attendance while prohibiting meaningful exit is running a coercive monopoly. Individual liberty in education shouldn't be a privilege reserved for families wealthy enough to move to good school districts.
L
Public schools are legally required to take every child — the child with severe autism, the undocumented child, the family that can't navigate an English-language application. Private schools are not. When Texas funds private institutions without requiring them to meet the same baseline obligations, it's not expanding freedom; it's creating a two-tier system and asking the public to fund both equally.
C
Public schools' universal enrollment obligation is real, and voucher programs should include disability accommodation requirements as a condition of funding — that's a legitimate design constraint. But the existence of students with complex needs doesn't justify trapping all other students in institutions that are failing them.
L
The moment you require private schools to accept every applicant, provide IEP services, and meet civil rights audit standards, most of the practical 'choice' disappears — which is why voucher programs almost never include those conditions.
Constitutional and legal foundations
C
Zelman v. Simmons-Harris in 2002 settled the federal constitutional question — publicly funded vouchers at religious schools do not violate the Establishment Clause. The legal architecture for expansion exists; the remaining fights are about program design and state constitutions, not fundamental legality.
L
Zelman settled the federal floor, not the ceiling. South Carolina's supreme court ruled in 2024 that its ESA program cannot fund private school tuition under the state constitution — and that reasoning applies directly to similar programs in Arkansas, Montana, Ohio, Wisconsin, and Utah. The legal foundation is more contested than 'Zelman settled it' suggests.
C
State constitutional challenges are real, but South Carolina is one ruling, not a trend — and the states where courts have actually upheld programs vastly outnumber the exceptions. Legal uncertainty about expansion is not the same as a settled case against the model.
L
One ruling becomes a trend when it's grounded in 'Blaine Amendment' state constitutional language that exists in over 30 states — dismissing South Carolina as an outlier requires ignoring the legal terrain that case is mapping.
Conservative's hardest question
The Florida and Oklahoma data showing that the overwhelming majority of voucher recipients were already in private school — not fleeing failing public schools — is the most damaging empirical fact for the conservative case, because it suggests these programs, as currently designed, function primarily as subsidies to families who had already opted out rather than ladders for students who need escape routes. It is difficult to dismiss because it comes from state administrative data, not ideological projections, and it directly undermines the equity argument that is the most morally compelling case for vouchers.
Liberal's hardest question
The most honest challenge to this argument is that the Milwaukee and D.C. Opportunity Scholarship programs — the earlier, targeted, income-restricted voucher programs — did show modest positive academic results for low-income students, and it is genuinely difficult to dismiss those findings as irrelevant simply because later universal programs failed. A critic could reasonably argue that targeted, income-limited vouchers for students in demonstrably failing schools are a different policy than what is being critiqued here, and that conflating the two weakens the empirical case against vouchers as a category.
Both sides agree: Both sides agree that the Florida and Oklahoma enrollment data — showing 69% and fewer than 10% of voucher recipients respectively were previously in public schools — represents a genuine design failure, not a statistical artifact.
The real conflict: The core factual-and-values conflict is whether the appropriate benchmark for voucher programs is an idealized public school system or the actual performance of public schools serving the same low-income populations — a dispute that is simultaneously empirical and normative, because which baseline you choose determines whether vouchers look like an improvement or a regression.
What nobody has answered: If a well-designed voucher program — income-capped, requiring prior public school enrollment, with full IDEA compliance — were enacted and produced positive outcomes for low-income students in failing schools, would opponents accept it, or does the objection run deeper than design to a foundational belief that public education must remain a unified system regardless of individual outcomes?
Sources
- Search results: 2025 federal 'big, beautiful bill' tax credit scholarship provisions and Joint Committee on Taxation analysis
- Search results: Texas universal voucher program 2025 legislative details
- Search results: FutureEd analysis of 114 state bills in 30 states in 2025
- Search results: Florida Policy Institute data on Florida voucher funding shares 2021-2025
- Search results: Step Up for Students data on Florida new voucher recipient prior enrollment
- Search results: Oklahoma voucher applicant public school enrollment data
- Search results: South Carolina Supreme Court ruling on education savings accounts
- Search results: NEA President Becky Pringle statement on vouchers
- Search results: New Mexico Governor Michelle Lujan Grisham announcement on federal program non-participation
- Search results: Institute on Taxation and Economic Policy projection of $51 billion cost
- Background knowledge: Zelman v. Simmons-Harris (2002), Milton Friedman (1955), Milwaukee voucher program (1990)