The people on the other end of this

Who it hurts

About 100,000 Arizona children use this program right now. Roughly 20,300 of them lose it in the first year. Here is who they are — including the parts that do not flatter our side of the argument.

100,208

students enrolled

As of 27 July 2026, for the 2026-27 school year. ADE cautions the number is in flux until September; the more stable June figure was 101,128.

Arizona Department of Education

19%

are students with a disability

About 16,500 students, against roughly 14 percent in district public schools. Fifty-two percent of them are autistic.

ADE FY25 Q3 Executive & Legislative Report

~20,300

lose eligibility immediately

Roughly 24 percent of current universal-eligibility students would be cut off in 2027-28 under the $150,000 cap.

Common Sense Institute, May 2026

57.2%

of new enrollees came from public school

Up from 21 percent in FY23. Most new ESA families are leaving a district school, not being subsidized in a private school they already attended.

ADE FY25 Q3

Where ESA families sit on the income ladder

Neither side has individual income data, because ADE does not collect it. Every claim you will hear this fall — ours included — is inferred from ZIP codes. Here is the best available estimate.

Common Sense Institute, Q3 2025, inferred from ZIP-code median family income. Paradise Valley (85253) — the wealthy-ZIP example you will see quoted — accounts for 678 ESA students, or nine tenths of one percent of the program.

Income band	ESA families	Comparison
ZIP codes with median family income $75,000–$150,000	56.7%	35.8% of all Arizona families
Estimated under $75,000	~25%	—
Estimated over $200,000	<3%	13%+ of all Arizona families
Rural ZIP codes	13%+	10.7% of Arizonans

Scroll the table sideways →

The best evidence on the other side

Wealthier neighborhoods do sign up at higher rates

In May 2024, Brookings found that ESA participation rises steadily with neighborhood affluence: about 74 recipients per 1,000 children in Arizona's highest-income decile against 20 per 1,000 in the lowest, with similar gradients by poverty rate and college attainment. Maps of ESA density and median income in metro Phoenix, they wrote, “look remarkably similar.” That finding is real, it is well executed, and anyone telling you the program is a poverty program is overselling.

Both things are true, and the difference decides the vote

Brookings measured the uptake rate — what share of children in a given area enroll. The Common Sense Institute measured composition — what share of enrollees come from each income band. Wealthy families can sign up at higher rates while still being a small minority of participants, because there are far more middle-income Arizonans than wealthy ones. Both findings hold at once. And it is composition, not uptake rate, that determines who a $150,000 cliff actually removes. Fewer than three percent of ESA families are estimated to be above $200,000. The roughly 20,300 students cut off in the first year come overwhelmingly out of the middle of the distribution, not the top of it.

The cap tightens every year on its own

This is the part that gets least attention and matters most over time. The $150,000 line rises by 2 percent annually, or the GDP price deflator, whichever is lower. Arizona incomes have been growing at roughly 4 percent. A threshold indexed below wage growth does not hold still — it ratchets down in real terms every year without a single further vote.

~400,000

Arizona children who could never participate

Roughly a third of all school-age children in the state.

52%+

of families with school-age children excluded by 2045

The projected result of a 2 percent cap indexation against ~4 percent income growth.

$115M

added annual cost if 20,000 students return to district schools

District per-pupil spending from all sources runs around $15,000 against a universal ESA award near $7,700.

Common Sense Institute ballot analysis, May 2026

Why families leave in the first place

It is worth being precise about this, because the caricature is that parents are chasing prestige. National survey data says otherwise: what moves a family is usually that something has gone wrong.

33%BullyingNow the single most common reason parents give for switching schools.
27%Excessive stress or anxiety
26%Difficulty with teachers
24%Academic needs unmet
46%Safe environmentThe top reason homeschooling parents give for their choice, against 25 percent of district public school parents.

EdChoice / Morning Consult, May and June 2026 polling; EdChoice, Schooling in America 2025

And one piece of history Democrats should sit with

In the spring of 2020, 3.3 percent of Black households were homeschooling. By that fall it was 16.1 percent — a fivefold jump, the largest increase of any group in the country, at a moment when white households went from 5.7 to 9.7 percent. Two caveats we will state rather than let someone else state for us: the Census clarified its survey wording between those waves to separate homeschooling from remote learning, and this was the pandemic peak, not a current figure. But reporting at the time consistently found the same explanation from Black parents, and it was not tuition or prestige. It was that their kids were being treated badly and they had run out of other options.

U.S. Census Bureau, Household Pulse Survey, March 2021

Where Democratic voters actually are on this

If you assume support for education savings accounts is a Republican position, the polling will surprise you. It is one of the few education issues where partisan gaps nearly vanish.

Finding	Result	Detail	Source
Arizonans supporting ESAs	66%	75% among Arizona K-12 parents	EdChoice / Morning Consult, Arizona, February 2026
Democrats supporting universal rather than needs-based eligibility	75%	Republicans, 74% — a one-point gap	EdChoice Polling Primer, 2026
Support for universal eligibility by race	77% / 76% / 73%	Black / white / Hispanic respondents	EdChoice Polling Primer, 2026
Higher-income respondents wanting need-based restrictions	46%	Against 63% of low-income respondents — the reverse of what you would expect	EdChoice Polling Primer, 2026

Scroll the table sideways →

Who paid for this

$7.5 million, and almost none of it from Arizona

Unions are entitled to spend money on politics, and a teachers union backing a school measure is ordinary politics. The question here is narrower: Prop 212 has been presented as a grassroots response by Arizona parents, and the campaign finance filings describe something else.

National Education Association, Washington DC	$6,565,438	87% of all contributions
AEA Fund for Public Education (Arizona)	$810,000	10.7%
Everything else combined	$168,228	2.2%
Paid to Fieldworks LLC for signature gathering	$6,531,162	90% of all spending

Scroll the table sideways →

Union money accounts for roughly 98 percent of the $7.54 million raised, and about nine of every ten dollars spent went to paid signature collection — over 75 percent of signatures came from paid circulators. There is also an unresolved complaint over the campaign's own disclosure: its website stated that 9 percent of funding came from out of state. The committee's position is that Arizona law requires disclosing the share of out-of-state contributors by headcount rather than by dollars, which may well be correct as a matter of law. It is not what a reader of that website would have understood.

Arizona Secretary of State campaign finance filings; Transparency USA

The legislature overrode your vote. That was wrong.

In 2018 Arizona voters rejected universal ESA expansion by 65 to 35 on Proposition 305, and the legislature enacted it anyway in 2022. If you are angry about that, you are right to be — overriding a referendum is a genuinely bad thing for a legislature to do, and no argument on this page depends on pretending otherwise. But the remedy for a legislature ignoring voters is not a ballot measure that a private committee may be empowered to enforce, that confiscates children's account balances annually, and that tightens itself every year without another vote. You can believe the 2022 override was illegitimate and still conclude that Prop 212 is the wrong correction.

Do the one thing that matters

Vote NO on 212

Pledge to vote NO