In 2022, Arizona captured the nation’s attention by adopting the country’s first fully universal education savings account (ESA) program, giving every child in the state the opportunity to access an “Empowerment Scholarship Account” of his or her own. Yet as the program began serving tens of thousands of new Arizona families, teachers union-allied opponents of school choice launched a massive media campaign to undermine its success, both in Arizona and nationally.
But as with much of the media’s reporting on polarizing topics over the past two years, the hostile coverage of Arizona’s ESA program has often more closely resembled a desperate partisan narrative rather than fact-based reporting.
Indeed, while union-aligned journalists and advocacy organizations have painted Arizona’s ESA program as excessively costly to taxpayers and responsible for triggering a budgetary shortfall, the two years of the universal ESA program’s history—and a new report from Arizona’s nonpartisan state budget analysts—suggest otherwise. Specifically, the hard data and recent state financial history offer the following takeaways:
- In the two years since Arizona’s universal ESA expansion took effect, the state budget enjoyed a massive overall surplus one year, and an overall net savings in its education funding formula (including the cost of ESAs) the second.
- In 2022-2023 (the first year with full ESA expansion), the state of Arizona posted a $2 billion surplus—even after accounting for the enrollment surge that occurred in the ESA program, which included the majority of students who have joined under the universal expansion.
- In 2023-2024, even as total ESA awards exceeded projections and reached over 75,000 students, corresponding public school enrollment declines led the total cost of Arizona’s education funding formula to actually decline relative to initial budget projections, producing a modest overall state savings.
- Arizona did experience an overall state budget deficit in 2023-2024, yet this was in spite of—not because of—the net savings generated by its education system (ESAs included). As the state’s nonpartisan budget analysts concluded in August 2024, total ESA awards did end up $92 million above the forecasted budget levels, but at the same time, the costs associated with public charter and district school enrollment dropped by an even larger figure: $93 million. As the analysts summarize:
“With the above forecast adjustments, we estimate the total combined district/charter/ESA enrollment will generate savings of $(352,200) in FY 2024 relative to the enacted budget” (emphasis added).
(Note: this analysis reviewed only “basic state aid formula” impacts, and did not address millions in additional ESA savings from all other sources of funding, such as the “Classroom Site Fund,” the “Instructional Improvement Fund,” or any other state, local, or federal funding that public school students receive that ESA students do not.)
Arizona’ s 2023-2024 Education Funding Formula (District, Charter, and ESA Students) Generated a Net Savings to the State Budget
Source: Arizona Joint Legislative Budget Committee FY 2025 Appropriations Report. August 2024.
- Arizona’s budget deficit in 2023-2024 came only after the state’s progressive governor, Katie Hobbs, vetoed the original budget that would have left over a billion dollars in reserves (even after fully funding the universal ESA expansion). Hobbs instead signed a budget that increased state spending by an additional $2 billion to the highest level of all time and exhausted the state’s surplus financial cushion, leaving it unable to absorb lower than projected revenue collections.
- Universal ESA expansion has lowered per-pupil K-12 enrollment costs in Arizona by shifting new student growth from higher-cost public schools to lower-cost ESA awards. Prior to the universal ESA expansion, 100% of growth in Arizona’s public K-12 student population each year was from public charter schools, which grew by approximately 10,000 students per year, and which receive well over $10,000 of state funding per child per year, compared to universal ESA students, who receive just $7,500.
Since the universal ESA expansion, however, public charter school enrollment has been flat, with growth shifting to the lower cost universal ESA program. In 2023-2024, for example, public charter school enrollment growth dropped from 12,000 (projected) to virtually zero (actual) as ESA enrollment from the universal expansion grew by over 10,000.
This shift contributes to a long-term flattening of the K-12 cost curve in Arizona. For example, while the 2024-2025 budget includes $95 million in projected cost increases associated with K-12 enrollment growth (for districts, charters, and ESAs combined), the 2022-2023 budget projections (the last to be made before enactment of the ESA expansion) included annual enrollment growth costs of $141 million.
In other words, annual state costs were poised to increase $45 million more (to pay for new students entering the K-12 education system) in the final year before ESA expansion than in the upcoming year.
- The entire cumulative increase in spending on ESAs under universal expansion is smaller than the increase in state spending on public schools that state lawmakers authorized over the same period, despite enrollment in Arizona’s traditional public district schools declining in that time. The total award value of all students who’ve joined under the universal eligibility category (even before taking into account the offsetting savings from them leaving public school) is approximately $420 million as of June 2024. In comparison, the state legislature added over $600 million in discretionary funding for public schools during the first year of the universal ESA expansion alone (in addition to hundreds of millions more for mandatory inflation increases, etc.), even as public district school enrollment was flat.
- Adding up the total award value of every child who’s joined the ESA program under the universal expansion amounts to a sliver (3%) of the total spent on Arizona’s public K-12 schools from state and local taxes alone, despite these students making up over 5% of the state’s K-12 population.
- As illustrated in several examples below, critics’ attacks against the ESA program have relied on ideologically motivated, often factually dishonest misrepresentations of the program and its finances. Given the frequency of these claims, however, it is worth addressing several of them directly:
Claim 1: ESAs have grown exponentially beyond budget projections.
Claim 2: ESAs have caused a budget deficit.
Claim 3: ESAs cost taxpayers more per student than public schools.
Claim 4: Other states risk financial crisis following Arizona’s footsteps.
Claim 5: Arizona’s ESA program misallocates resources to wealthy, undeserving families.
Claim 1: ESA Costs Have Grown Exponentially Beyond Budget Projections
Despite the facts above, ESA opponents and allied media have made several related claims, suggesting that Arizona’s universal expansion has proven exponentially costlier than projected and consequently caused a state budget crisis.
The director of the leading anti-ESA organization Save our Schools (SOS) Arizona declared, for instance, “when universal vouchers were passed, [state budget analysts] projected that the program would cost $33 million…Now projections from both Superintendent Horne and Governor Hobbs—so bipartisan agreement—the program is going to cost nearly $1 billion by next year.”
Columnist Laurie Roberts of Arizona’s USA Today affiliate, The Arizona Republic, similarly echoed earlier this year: “so the program that was originally projected to cost us $65 million this year has now pushed past $800 million.”
Such statements paint an alarming narrative of Arizona’s ESA program, but there’s just one problem. Such claims are simply false and represent either basic numerical illiteracy or willful misrepresentation of fact.
In both examples above (and in those similar), the numbers that are being compared are two entirely different measures—to compare them would be analogous to comparing someone traveling 60 miles per hour to someone else who has driven 600 miles and concluding that one is ten times larger than the other. In the same way, these critics’ are comparing 1) an initially speculated net change in state general fund costs associated with the universal expansion to 2) the overall total award value of all ESAs (including the hundreds of millions that were already awarded each year even before the universal expansion).
Specifically, to illustrate how spurious these comparisons are, consider the fact that Arizona’s ESA program spends approximately $250 million a year supporting special education students, who were already eligible and using an ESA before the 2022 universal expansion. That means that $0 of this cost was included in the cited estimates about how much more the universal expansion would cost the state General Fund, yet opponents include it in their latter estimates of the total cost of the ESA program—artificially suggesting an increase in hundreds of millions of dollars of cost where none occurred.
(It is worth noting that even the director of Save Our Schools Arizona is also on record explicitly declaring to the Arizona State Board of Education: “We fully support the original ESA program.” Despite such self-avowed support, the organization lumps in the $250-$300 million spent on these original ESA students as part of the supposedly unsustainable award values granted through the universal ESA expansion that they oppose).
Moreover, contrary to claims that the ESA program would reach a billion dollars in 2023-2024, the final estimated total(even including the cost of special education and other original ESA students) was $718 million in 2023-2024. Of this total, only approximately $420 million is actually associated with students who joined under the universal expansion (56,000 of the 75,000 total ESA population, with a median ESA award for non-special education students of $7,400).
Moreover, this far lower number still represents the gross total award values of these students (one measure), not their net general fund impact (an entirely separate measure). Conflating these two—as ESA critics have done in their attempted comparison—remains mathematically nonsensical.
Specifically, consider that according to the Arizona Department of Education, 18,000 students switched from public schools to the ESA program since the universal expansion. At a cost of roughly $7,400 per student, these students’ total gross ESA award cost is over $130 million in aggregate. Yet the actual net general fund impact of these students is almost, if not entirely, offset by the corresponding decrease in costs from them no longer being funded in the public school system. Thus, while these very same students constitute nearly zero net cost in the first half of ESA critics’ comparison, they appear as $130 million in the second half of their comparison, again demonstrating the completely incompatible points of reference used by opponents.
Taking even these basic realities into account, the actual net change associated with the ESA program would be less than $300 million. (As discussed above, this amounts to less than half of the cost of the discretionary increase in state public school spending showered on Arizona public schools in 2022 alone, even as public district school enrollment was flat).
Yet even this amount overstates the net financial impact of the program. As described above, state enrollment patterns now indicate thousands of students each year who would have been joining the state’s public charter schools (without ever having first attended one) who are now enrolling in the ESA program instead. These students all show up as a net cost under the ESA program, even as they cost taxpayers less than if they had enrolled in the public school system. Indeed, any student entering kindergarten, who moved from out of state, or even who once attended a public school but left during the COVID-19 lockdowns and mask mandates would appear as a net cost to the ESA program, even as these children would have been funded at typically higher levels in the public school system.
(Specifically, consider that over 260,000 people move to Arizona each year from other states, so the number of school-aged children not previously attending an Arizona public school—but who now qualify for an ESA—is significant. It would be erroneous to consider any such child now participating in the ESA program as a “net cost” of ESA expansion, yet this is precisely how opponents characterize them).
Finally, even the very earliest estimates cited by opponents that hypothesized initial net general fund changes of $33 million (2022-2023), $65 million (2023-2024) and $125 million (2024-2025) were provided with an explicit disclaimer by the authors that they were “highly speculative,” and these estimates were never included for use in the actual state budgets in any of the past 3 years. In contrast, the actual budgeted cost projections for the ESA program and the K-12 education system at large have proven remarkably accurate.
Claim 2: Arizona ESAs Caused a Budget Deficit
As the leftwing publication ProPublica declared in allied newspapers across the nation in July 2024, “in a lesson for these other states, Arizona’s voucher experiment has…precipitated a budget meltdown.” As the organization declared:
“[I]t remains an open question how much higher the cost of vouchers could go and therefore how much cash should be kept on hand to fund them. The director of the state’s nonpartisan Joint Legislative Budget Committee told lawmakers that ‘we’ve never really faced that circumstance before where you’ve got this requirement’ — that anyone can get a voucher — ‘but it isn’t funded.’”
Yet it turns out ProPublica was merely echoing false talking points from 12News Phoenix KPNX-TV reporter Joe Dana, who has repeatedly made that same claim, for instance saying:
“Here’s what the numbers gurus in charge of the financial projections at the capitol said to lawmakers last year about ESAs, ‘We’ve never really faced that circumstance before, where it’s sort of like you’ve got this requirement but it isn’t funded.’”
Despite the framing of this clip by activist journalists such as Dana and ProPublica, the budget director was not suggesting that ESAs posed a unique funding challenge that Arizona had not faced before. Rather he was answering a specific hypothetical question from a Democratic lawmaker who asked what would happen if the legislature decided to defy state law and refuse to fund its K-12 system (both ESA and public schools):
Lawmaker: “The direct question is this: what happens to our department of education if the dollars are not appropriated, what would then happen to our public schools to cover the request for new ESAs?”
Budget director: “Somebody gets sued potentially, because you’ve got law on the books that you’re not complying with. And so we’ve never really faced that circumstance before where it’s sort of like you’ve got this requirement but it isn’t funded.”
The budget director then also immediately clarified that the circumstances facing lawmakers was not unique, and that state law routinely handles any such difference between what was originally allocated for K-12 (based on initial enrollment estimates) and actual final costs:
“Now, we have a technical footnote in the budget…and this footnote has been there for a while…if you are short in [20]23, you can take and pay that with your [20]24 monies… So my answer has sort of evolved here as we’ve talked through this because I’ve remembered that provision.”
As noted above, this entire exchange took place amid the backdrop of a $2 billion state general fund surplus in 2022-2023 (even after fully accounting for the additional ESA costs). The attempt by Dana and ProPublica to suggest that ESA costs had thrown the state into uncharted waters—rather than acknowledge that this exchange between a lawmaker and budget director was about the process of appropriating the massive balance of available state funds to match actual expenses using a regular practice that had been in place for years—is emblematic of the deceptive agenda-driven anti-ESA reporting from such sources.
Claim 3: Arizona’s ESA Program Costs More Per Student than the Public School System
In addition to such deceptive misrepresentations above, self-described media “factcheckers” like 12News’ Dana—whom ProPublica and others have relied on for their anti-ESA reporting—have also made claims such as the following:
“The average the state gives for an ESA student is higher, not lower, than the average it gives to a public school district student.”
This is simply false.
Dana and others make this claim only by ignoring every part of state education spending on public schools other than what comes from a single source of funds through a single part of the funding formula.
Citing state budget analysts’ findings that a typical ESA student is funded in higher amounts from the state “general fund” by several hundred dollars compared to a public district school student under the basic funding formula, Dana declares that this means total state funding is higher for an ESA student.
Unlike the nonpartisan budget analysts that Dana cites, however—who explicitly recognize the significant additional state spending on public school students from other “major” sources that ESA kids don’t get—Dana simply ignores them as inconvenient to the narrative.
For example, the state budget analysts note that all Arizona public school students—but no ESA students—receive funding from the “Classroom Site Fund,” which is funded by a state sales tax and which alone allocates nearly $1,000 to every public school student in the state. Acknowledgement of this funding source alone puts the state cost per student higher for public school students than their ESA peers. (The funding difference grows to two to four thousanddollars per student when accounting for all sources of state and local funding, even before considering the additional public school costs paid for by Arizonans through their federal taxes).
Claim 4: Other States Risk Financial Crisis Following Arizona’s Footsteps
Among other sources cited by ProPublica and others, an analysis from the left-leaning Grand Canyon Institute has been used to prop up the claim that the leading culprit of Arizona’s (now resolved) 2023-2024 budget deficit was universal ESA expansion. As the organization declared: “The net cost of the universal ESA program is $332 million,” which “represents more than half of the projected state budget shortfall in FY2024 ($650 million) and is on pace to represent two-thirds of the projected shortfall in FY 2025.”
While the Grand Canyon Institute’s analysis at least improves upon the mathematical illiteracy of other ESA critics described above—attempting to recognize the difference between net and total costs—its analysis likewise arbitrarily assigns blame to the ESA program, which makes up just 3% of state and local spending on K-12 schools. Yet it ignores 1) the billions in public school and other non-educational spending increases over the past two years, and 2) the offsetting savings from reduced public school enrollment growth resulting from prospective students opting for lower cost ESAs over higher-cost public schools.
As noted above, the unprecedented increase in discretionary spending in the 2023-2024 enacted budget (on non-ESA related items) was more than double the size of the state deficit that arose that year.
Likewise, as described above, 100% of Arizona’s growth in public school enrollment over the past decade has come from public charter schools, but that growth has completely flattened in the wake of universal ESA expansion as new students entering the system disproportionately choose the less-expensive ESA program.
Moreover, contrary to the Grand Canyon Institute’s projections that the net costs of the state’s universal ESA expansion will rise by $100 million in 2024-2025, the state’s official nonpartisan budget analysts suggest otherwise. Specifically, as of August 2024, the analysts indicate a $100 million increase in total ESA awards (including the previously eligible special education category), which would be largely offset by 1) savings from the roughly half of ESA students who are switching from public schools and 2) savings from new Arizona K-12 students choosing an ESA instead of higher cost public school options.
Claim 5: ESAs Misallocate Resources to Wealthy, Undeserving Families
Finally, critics allege that by supporting families already pursuing private or home-based education, the ESA program siphons too much money to “wealthy” or “high-income” families.
Yet this charge reflects a major incongruity in the logic of anti-school choice activists. On one hand, they hold that scholarship assistance for children enrolled in private schooling options is too financially burdensome to the state. At the same time, however, they fully support increasing taxpayer spending on children from families of identical wealth in the public school system.
(This is why, for example, critics speak almost exclusively of the “costs” of ESAs students, but only of “funding” for public school students.)
As the Goldwater Institute has documented in a recent report, Arizona spends 10-20 times as much money subsidizing the education of “wealthy” students (those from families making over $150,000) in the public school system than it does supporting similarly financially situated students who’ve joined the ESA program from private or homeschool arrangements under the universal expansion.
Moreover, like their public school peers, ESA households represent families across the economic spectrum, as Goldwater and other organizations such as the nonpartisan Common Sense Institute have repeatedly documented. But opponents—who offer no objection to such families enrolling in large numbers in public district or charter schools (at taxpayer expense)—condemn these same families as an exceptional financial burden simply for choosing to instead direct their allotment of education funding to private or home-based learning options through the ESA program.
Such double standards are not the basis for sound public policy, nor are they a recipe for educational or financial prosperity within the states.
They are, however, apparently a sufficient rationale for those opposed to school choice to justify their positions. As the left-leaning Brookings Institute recently acknowledged:
“The Goldwater Institute…notes that Arizona spends a great deal of money to “subsidize public school instruction” for wealthy families. It accuses us (and/or others) of a double standard in how we object to using government funds to pay for wealthy students’ private schooling but not public schooling.”
Their response? Not even attempting to deny the charge:
“Americans have long accepted—in fact, embraced—a double standard for public and private schools.”
By simply proclaiming a national “consensus” in support of their own views—and ignoring an entire half of the nation seeking something better—advocacy organizations like Brookings suggest the education status quo should be preserved because…that’s how it’s always been.
Yet this same status quo failed families during COVID-19, locked children out of classrooms, has doubled inflation-adjusted K-12 costs over recent decades, and has failed to meaningfully improve student outcomes for generations. The proliferation of education savings accounts—like other school choice innovations such as charter schools—on the other hand, offers families and lawmakers the opportunity to expand the range of educational choices available to students and ensure that each child can pursue an education of excellence, not simply political convenience.
Matt Beienburg is the Director of Education Policy at the Goldwater Institute. He also serves as director of the institute’s Van Sittert Center for Constitutional Advocacy.
