Education Choice, Access & Transparency How a Proposal to Regulate Arizona’s K-12 Scholarship Program Would Impact the State’s Families

In 2011, Arizona lawmakers created the state’s Empowerment Scholarship Account (ESA) program, which provided eligible private- and home-school K-12 families with access to a portion of their state formula funding. Originally open to students with a disability, eligibility was quickly expanded to various categories of students (either disadvantaged or attending poorly performing schools); by 2020 approximately one quarter of Arizona public school students were ESA-eligible. However, less than 1% participated.

During the pandemic, K-12 education changed rapidly. Extended school closures, remote learning, and frustration with classroom standards and curriculum led to the largest drop in district public school enrollment ever observed; Arizona’s traditional K-12 enrollment fell by 50,000 students. Today, there are about 75,000 fewer students in Arizona’s district classrooms than there were in 2019-2020. Fewer than 70% of Arizona’s school-aged children are in district schools, down from 80% ten years ago. Meanwhile, private- and especially home-school enrollments grew rapidly. 

In response to lagging demand for traditional public schools and public desire for alternatives, in 2022 Arizona lawmakers expanded eligibility for Arizona’s ESA program to every school-aged child in the state and removed any requirement for prior public-school attendance. Participation grew rapidly. Today, there are over 100,000 students participating (almost 10% of Arizona’s K-12 population) and annual scholarship awards total about $1.1 billion. Notably, one quarter of students are in non-universal eligibility categories (accounting for half of program costs); the Students with Disabilities category has grown especially quickly since 2022.

Accompanying this rapid student growth has been growing controversy surrounding the sustainability of the program and appropriateness of participant spending. In March, the Protect Education Act was filed by a coalition of public-school advocates and public employee union groups. The bill would limit participation in, and regulate the use of, Arizona’s ESA program going forward. Among other things, the Act if approved by voters would:

• Limit universal-eligibility scholarships to families earning less than $150,000 per year
• Require participating private schools to be registered, accredited, and/or engage in mandatory state testing
• Further restrict the use of ESA funds relative to current law, by more strictly defining “noneducational” and “luxury” items

Figure 1

Figure 2

What K-12 Education Looks Like in Arizona Has Been Changing for Years

Arizona’s K-12 landscape was changing long before universal ESA. Falling birth rates matter, but a growing number of families were already leaving the district system for charter, private, homeschool, and other nontraditional options—and ESA expansion followed that trend rather than causing it.

• District classroom sizes in Arizona peaked in 2008; district schools have been shrinking ever since, and all public growth was in the charter sector.
•  Today, one quarter of all Arizona’s public-school students are in charter schools; district enrollment has fallen by about 25%, and at least 15% of the remaining students are open-enrolled somewhere other than their assigned district.
• More recently, there has been rapid growth in private, nonpublic enrollment, particularly in the homeschool and nontraditional “microschool” spaces. Today, 10% of Arizona’s school-aged kids are estimated to attend nonpublic school.
• This trend is ongoing. A decade ago, 80% of Arizona’s kindergarten-eligible population was likely enrolling in a district classroom; today that figure is closer to 60%. The decline in district enrollment cannot be explained by birth rates alone and predates universal ESA.

Figure 3

Figure 4

The Protect Education Act Significantly Impacts Nontraditional K-12 Families

Over 102,000 children use Arizona’s ESA program, as of April 2026. Combined with another estimated 20,000+ receiving School Tuition Organization scholarships, there are over 120,000+ Arizona private- and home-school children getting public support for their K-12 education costs. For many families, this bill is not an abstract policy fight; it is a direct threat to how they afford their schooling.

• Based on a detailed analysis of ESA families’ demographic data revealed by their residence ZIP codes, CSI estimates that about one quarter (24%) of current ESA users have annual household incomes over $150,000. Assuming the geographic/income distribution of universal-eligibility families mirrors that of all participants, 20,300 current universal ESA families would immediately lose eligibility due to the income limitations.
•  Statewide, CSI estimates 30% of families with children had incomes over $150,000 in 2024, meaning an estimated 400,000 Arizona school-aged children would be excluded from ever participating in universal ESA. Based on reasonable assumptions about higher family sizes in income, up to 40% of all Arizona 5–17-year-olds may be excluded.
  •  Notably, CSI estimates 25% of public-school families have household incomes above $150,000, using the same ZIP-code-based American Community Survey (ACS) data. That number represents approximately 260,000 kids receiving the equivalent of $3.9 billion in K-12 funds from taxpayers.
•  Although the income cap has an annual inflation adjustment, it limits that adjustment to 2% per year. On average and over time, incomes in Arizona rise closer to 4% per year. This discrepancy means that over time an increasingly large share of families will become subject to the program’s income limits. CSI estimates that over 20 years (by 2045), over half (52%) of Arizona’s families with school-aged children would become income-excluded from universal ESA eligibility.
• Even though the income cap applies only to universal-eligibility families, other provisions of the Act and lower universal participation is likely to depress participation in other eligibility categories, too.
  • Over the four years after universal eligibility, growth in ESA students with disabilities was 10% faster than over the four years prior— even though eligibility criteria for these students hadn’t changed.
  • Without universal eligibility, there might have been approximately 10,000 fewer non-universal ESA users, or nearly all students with disabilities. Meaning, slower or fewer universal-eligibility students going forward could mean fewer ESA’s going to students with special needs, foster students, etc.

Figure 5

Figure 6

Figure 7

Universal ESA Aligns K-12 Funding with Expected Student Growth

On average, a universal-eligibility ESA costs state taxpayers about $7,700 per year. A public-school student costs taxpayers almost $15,000 per year, including $13,300 in state and local funding alone. Further, adding up participation in all of Arizona’s publicly funded K-12 options, the total number of students being educated (including 100,000 ESA users) is about the same as we expected in 2019. The total has not changed; how they are funded has changed.

• According to the most recent K-12 All Funding report, a public-school student costs taxpayers almost $15,000 per year; $13,325 of that is state and local money. An average ESA student costs about $10,700, but a universal-eligibility ESA student is closer to $7,700.
•  Even in terms of the state budget (General Fund), a typical ESA recipient costs less than a typical charter student. Further, because of how the funding formulas work, most district students are 100% funded at the margin—the last student in or out is almost entirely General Fund cost or savings, and property taxes provide a baseline.
• Even with 100,000 students participating, the total number of publicly funded K-12 students (district, charter, and ESA combined) is really no higher than state budget analysts expected back in 2020; there are just much fewer public-school students than expected, and the ESA program is much larger.
  • For example, in early 2020, the Legislative Budget Office (JLBC) forecasted 1,132,000 public school students alone by mid-2024; actual public-school enrollment was closer to 1,092,000.
  •  Enrollment under universal eligibility that year? About 45,000 kids, offset almost 1:1 by 40,000 fewer public-school kids than expected. Most of those enrollment declines occurred in 2021, before universal ESA eligibility was enacted.
• There have been large increases in K-12 education funding over the last decade, and only a fraction of that increase has gone to ESA growth. On an inflation-adjusted basis, Arizona’s is spending 30% more per public-school-pupil than it was a decade ago.
  • While funding is going up, where the money is going is changing. Spending in the classroom has fallen from 55% of funding in 2012, to 52.7% today. What is growing? Support services have gone from 40.1% to 42.4%.
• On the other hand, moving 20,000 universal ESA users back into district classrooms would have fiscal implications, increasing state and local taxpayer costs by $115 million per year. The ESA program not only aligns state funding with parent preferences, it does so at a net-cost-savings.

District Outcomes Are Weak by the State’s Own Tests—and There is Evidence Private Outcomes Are Better

While most nonpublic Arizona K-12 students are not taking the state assessment, they are often taking some form of standardized testing—and evidence suggests they generally perform better than their public counterparts on equivalent tests. On the other hand, Arizona’s district students fare very poorly on the state’s own public-school assessments. Just 39% of students are at least “proficient” in reading, 32% are proficient in math, and only 27% are meeting proficiency or above in science.

• Of Arizona private schools surveyed by Common Sense Institute, 84% already use some kind of standardized academic assessment/testing. Two-thirds are accredited by a “recognized accrediting body.”
•  Composite ACT scores for public, private, and homeschool students were published by the American College Test group. These standardized tests are administered to college-bound students every year. Based on these results, the average private school ACT test-taker performs 19% better than their public-school counterpart; the average homeschool test taker performs 12% better.
• Though not administered to homeschool students, the National Assessment of Educational Progress (NAEP) is a national standardized assessment of K-12 student learning and academic performance, including private students. Based on these national results, private school students appear to perform significantly better than their public-school peers.
• Where data is available, it appears the typical private school student appears to perform better than roughly 70% of their public peers on NAEP subject exams or ACT college-entrance exams.

Figure 8

The ESA Program Already Has Strong Oversight and the Act Would Add New Burdens

There are detailed statutes, rules, and administrative language regulating Arizona’s ESA program, including detailed lists of expenditures presumed unallowable. Additionally, processes are in place to enforce them, including statutorily required risk-based audits. Private schools operate under varying degrees of private, federal, state and local oversight, and are generally considered safer than public alternatives.

Figure 9

• When the ESA program has been subjected to random audit, apparent misspending rates are much lower than they are in other public programs, and much lower than reported or implied. For example, the Department of Education’s March 2026 technical report found 1.9% of samples ESA spending was “unallowable” and just 0.3% was “egregious”.
•  Media reports of higher or sensational misspending are often based on either anecdotal review or apparently risk-based samples.
• High standards of review specific to the ESA program are apparent in the user experience. Before 2025, nearly all transactions and reimbursement requests were typically subject to review and approval before they are processed. Additionally, before the Department began immediately processing all transactions under $2,000 (subject to later review), payment processing times took over 20 days—significantly longer than in comparable programs.
• Private schools, in turn, maintain their own processes—either independently or through affiliated accreditation programs—to maintain student safety and academic rigor.
  • All private school respondents to CSI’s survey reported employing at least one of various methods to ensure qualified staff without criminal histories, including fingerprint clearance cards and minimum educational standards.
• All respondents warned that complying with the Act’s new requirements would impose new administrative burdens and requirements. Additionally, three quarters said it was at least possible the Act would lead to tuition increases and 1 in 5 said it was at least somewhat likely they would no longer accept ESA-funded students if the Act becomes law.
• This change would impact over 4,600 reported ESA-using students, potentially requiring them to withdraw and move to another school.
  • This change would impact over 4,600 reported ESA-using students, potentially requiring them to withdraw and move to another school.

Figure 10