Arizona’s Universal ESA Doesn’t Need Saving

In recent months, not one, but two competing ballot initiatives have been filed to alter Arizona’s Empowerment Scholarship Account (ESA) Program — the oldest ESA program in the country.

To qualify for the November ballot, both initiatives require 255,949 valid signatures by the petition filing deadline on July 2, 2026.  Each initiative claims it will make Arizona’s ESA program more accountable.

But that isn’t the case. In fact, if either initiative becomes law, the ESA program would become more restricted, making it harder for students to benefit from hard-won school choice success.

In his 1929 book, The Thing, G.K. Chesterton introduced the parable now commonly known as Chesterton’s Fence. The idea is simple: before tearing down a fence, we should understand why someone placed it there. It implies that what is already in place likely serves a purpose that may not be immediately obvious. Put another way, “if it ain’t broke, don’t fix it.”

Arizona ranks first in the 2026 edition of the EdChoice Friedman Index, and the program is one of the most popular in the country, serving over 100,000 students this year. Moreover, Arizona is one of only five states that meet EdChoice’s definition of true universality, meaning every eligible student can receive consistent funding to access a marketplace of educational options.

The light and appropriate regulatory touch of Arizona’s ESA program is a feature, not a bug. The program does not impose an income limit, an enrollment cap, or testing requirements that would serve as barriers to families participating. It also does not require schools to be accredited, which would stifle the supply of educational options available to participating students.

Yet both ballot initiatives would modify the very elements that make the ESA program distinctive in its pursuit of educational freedom, all at the expense of student flourishing.

What follows is an overview of both initiatives and why they would negatively impact a family’s freedom to choose.

The Protect Education Act

The Protect Education Act, supported by the union-backed Save Our Schools Arizona, would impose an income limit on eligible participants, require participating schools to meet accreditation or assessment-related conditions, and prevent families from rolling over ESA funds for future expenses, among other changes.

Prior to 2022, eligibility for Arizona’s ESA was limited to certain student groups, such as those with disabilities and those attending underperforming schools. Then, in 2022, the universal expansion made every K-12 student in the state eligible. Starting in the 2027–28 school year, the Protect Education Act would reverse course, requiring any student not covered by the pre-universal expansion criteria to live in a household with an annual family income at or below $150,000 — adjusted annually by 2% or by the GDP price deflator, whichever is lower. According to EdChoice’s Director of Fiscal Research and Education Center, Dr. Marty Lueken, “the $150,000 income cap would strip tens of thousands of ESA families of their ability to participate in the program and lock hundreds of thousands of Arizona families with children out of the program in the future.”

Private schools would also be adversely affected, since they would have to jump through unnecessary hoops to accept ESA funds. Under the proposal, most private schools that want to participate must be regionally or nationally accredited (schools that serve only students with special needs are exempt). If the school doesn’t want to pursue accreditation, its alternative is to pay for and administer an assessment adopted by the State Board of Education to all students without special needs and then publicly disclose the aggregate scores. Existing schools would be disqualified from participating in the program if they do not want to do so. These requirements could also reduce the growing number of schools entering the market and, according to an ExcelinEd analysis of other states with accreditation requirements, significantly decrease the number of available options for families.

Furthermore, the ballot initiative confiscates 100% of a family’s unspent ESA balance — SPED students have two years to use funds before confiscation — and reroutes 90% of those dollars to the Classroom Site Fund, which exclusively supports public-district and charter schools rather than ESA families. Eliminating rollover funding would create perverse incentives that nudge families to spend their accounts at year-end on whatever the system will approve, rather than on what their children need. Letting families keep unused funds year over year allows them to save for college or other transformative educational purchases that might be useful down the line. This change is particularly counterproductive to the stated purpose of the Protect Education Act, as rollover provisions in ESA programs hold families financially accountable by incentivizing them to choose less expensive options and save funding for the following year.

Additionally, creating a statute that automatically confiscates and redirects unused state funds sets a terrible precedent. Imagine the state confiscating any unused funds from one district and just giving it to another. The reality is that funding many schools — both public schools directly and private education options through ESA family choices — is not mutually exclusive. After all, Arizona has successfully increased public K-12 funding even as the ESA program has grown.

The Reform and Accountability Act

The other ballot initiative, known as The Reform and Accountability Act, supported by the American Federation for Children and other choice supporters, is sold as a more common sense approach that saves the program by preserving universal eligibility while addressing fraud and accountability issues.

The truth is that claims concerning a severe lack of accountability in Arizona’s ESA program are false. Every single dollar in the program is tracked by the program manager and the Arizona Department of Education, and unallowable spending rates remain below 2% according to both an EdChoice analysis and an analysis done by the Department of Education. Still, the language in the ballot initiative tries to fix a non-issue by proposing self-defeating measures that reduce the flexibility and provisions already in place that make the ESA work.

The Reform and Accountability Act proposes, beginning July 1, 2027, that all ESA expenditures be processed through direct pay and the marketplace system — ironically, “Marketplace” is currently the payment channel where the most documented unallowable spending has occurred. The other purchasing channels — reimbursement and debit cards — as they currently operate, would effectively be eliminated. For families, this means losing the option to pay any education service provider not registered with the state. Homeschooling families, especially, would be disproportionately impacted as they are more likely to rely on multiple education providers.  

The Reform and Accountability Act would also impose a host of compliance obligations on schools, tutors, and qualified tutoring organizations, limiting the supply of education providers. For instance, qualified schools would have to pay for, administer, and report the scores of an approved examination for every full-time ESA student, with conditional exceptions for students who meet certain disability criteria. ESA students not enrolled full-time at a qualified school, such as homeschoolers, must also take an approved examination and submit their scores to the Arizona Department of Education. In practice, families who chose an alternative learning option precisely to avoid test-driven instruction would find themselves back inside it.

According to EdChoice’s Schooling in America Survey, over 80% of parents support ESAs that allow parents to use funds for various educational options. In recent years, smaller education providers — such as microschools and homeschooling co-ops — have proliferated to meet demand from ESA families. However, smaller providers operating as qualified schools would be least equipped to absorb the costs of administering an approved exam and complying with annual data submissions, incentivizing these schools to pass the costs on to families through tuition hikes.

Arizona’s ESA Program Has Guardrails That Work

The current design of the ESA program gets accountability far more right than wrong.

Returning to the lesson of Chesterton’s Fence, further inspection shows that Arizona’s ESA was thoughtfully designed to empower parents to define what success looks like for their children while ensuring that the program manager and the Department of Education have the tools needed to address any issues. The ESA program already includes safeguards, such as prohibiting the use of ESA funds for “luxury” purchases. The Arizona Department of Education also conducts audits, and the state board already has a mechanism to refer cases of substantial misuse to the Attorney General for collection or criminal investigation.

Adding burdensome provisions in the name of accountability would only harm a program with very minor problems that can be addressed through basic administrative improvements. Partners, policymakers, and the press should say so plainly and avoid creating any superfluous changes to an ESA program that is best in class.