Alberta’s School Choice Revolution Is Only Half Finished

Here is how to complete it.

There is a sentence that ought to end every argument about who should control a child’s education, and it is this: parents know their children better than bureaucracies do. It sounds obvious because it is. Yet the entire apparatus of public education in most of Canada is organized around the opposite assumption: that credentialed experts, housed in school boards, teachers’ unions, and ministries, are better placed than mothers and fathers to decide how a child learns, where, and with whom.

Alberta has been pushing back against that assumption for at least thirty years. The province’s school choice architecture is the most pluralistic in Canada: public, Catholic, Francophone, charter, independent, and home-based learning all exist within a framework that treats public funding of education as distinct from its public delivery. That distinction matters enormously. It says, in effect, that the taxpayer’s investment belongs to the child, not to any institution that claims the right to serve the child.

Last February 11, a petition organized to reverse that logic failed. The “Alberta Funds Public Schools” initiative, launched by an Alberta Teachers’ Association district representative during the province’s first-ever province-wide teacher strike, collected just over 126,000 signatures in 120 days. It needed 177,732 to trigger a referendum. The shortfall, roughly 51,000 names, was not close. Albertans, having been offered a clear chance to vote for centralized institutional consolidation, declined.

That is a mandate. The question is whether the province is bold enough to use it.

Because the case for consolidation was not made by an obscure fringe. It was made with the organizing infrastructure of the province's largest teachers’ union during a labour dispute that gave the ATA maximum public sympathy and media attention. If the union could not persuade Albertans to restrict educational choice under those conditions, it is unlikely to persuade them under any. The government should read that result not as permission to hold ground, but as an invitation to advance.

The next step is Education Savings Accounts. And it is a bigger step than it sounds.

An ESA is not a voucher. That distinction is worth making clearly, because the two are frequently conflated by critics who benefit from the confusion. A voucher is a payment to a school, a subsidy that expands the institutions a family can afford to attend. The petition that just failed was partly sold on the promise of fiscal savings: stop funding independent schools, the argument went, and the money would flow back to classrooms. That claim was false. When an independent school student costs the province 70 per cent of per-pupil funding and a public school student costs 100 per cent, defunding independent schools and absorbing those students into the public system costs more, not less. The fiscal argument for the petition was the opposite of its proponents’ intentions. An ESA is something more fundamental: a portion of the per-pupil provincial funding allocation placed in a parent-controlled account, spendable on any approved educational expense. Private school tuition, yes, but also specialized tutoring, speech-language therapy, occupational therapy, assistive technology, accredited online curricula, concurrent enrollment at a college or university, or any combination of those things. The money stays in education. What changes is who decides how it is spent.

That is the hinge on which the entire argument turns. Under the current system, even in choice-rich Alberta, the purchasing agent is always an institution. A family chooses a school, and the school (or board for some homeschoolers) decides what services and programs the child receives. For the majority of children, whose needs fit reasonably well within what schools can deliver, this works tolerably well. For a substantial and growing minority, it does not work at all.

Think of what a school, however excellent and however well-intentioned, is built to do. It is built to serve a large number of children simultaneously within a fixed budget, schedule, and physical space. It is optimized for the median learner. The further a child’s needs depart from the median, in any direction, the more the institution strains. And the strain is not a failure of will or compassion on the part of teachers. It is a structural feature of institutions. An ESA does not ask institutions to be something they cannot be. It simply allows families to stop pretending that every child fits.

The evidence from jurisdictions that have introduced ESAs is instructive, though it requires careful reading. Arizona created the first ESA program in the country in 2011, initially restricted to students with disabilities. When the state expanded eligibility to all students in 2022, participation grew from roughly 12,000 to nearly 90,000 students over two years, surpassing 100,000 by early 2026. A RAND Corporation analysis confirmed that approximately 7 percent of Arizona’s school-aged population was using an ESA by the 2024-25 school year. Before universal expansion, students with disabilities represented 60 per cent of users. After expansion, as more families enrolled, that share fell to 18 per cent of a much larger total, while the absolute number of students with disabilities using the program continued to grow.

Opponents predicted fiscal catastrophe. Arizona’s own nonpartisan budget analysts concluded in August 2024 that while ESA awards exceeded projections by $92 million, corresponding public school enrollment declines reduced costs by $93 million, producing a modest net saving. A 2021 analysis of 40 school choice programs across multiple states found that such programs generated between $1.80 and $2.85 in fiscal savings for each dollar spent. The apocalypse did not arrive.

What did arrive, particularly after political interference complicated the program’s administration, was a cautionary note about implementation. A Heritage Foundation survey found that while 99 per cent of Arizona ESA families still supported the concept, two-thirds reported frustration with administrative red tape that had grown around the program after a hostile state attorney general reinterpreted its rules. The lesson is not that ESAs fail. It is that ESAs designed with clear rules and efficient administration succeed, and ESAs designed badly create unnecessary friction. Alberta, with the benefit of Arizona’s experience, can build a program from scratch.

Florida offers a complementary picture. The state now operates the largest school choice ecosystem in North America, with multiple ESA and scholarship programs serving over 500,000 students. Priority is given to low-income and foster families, with middle-income families next in line, a structure that directly answers the charge that these programs are subsidies for the wealthy. The average ESA is worth roughly $8,000 for general education students and $10,000 for those with identified learning differences, against a full public school per-pupil cost that runs considerably higher. As of 2025, eighteen American states have established ESA programs. The idea has moved well past the experimental stage.

Alberta already knows the fiscal logic. Budget 2026 made it explicit: it costs the province roughly $5,800 less per student enrolled in an independent school than in a public school. An ESA funded at 80 to 85 per cent of per-pupil costs sits between those two figures, saving money relative to full public school enrollment while giving families more than independent school funding alone provides. The math is not complicated, and it runs in the same direction as the pedagogy.

The fiscal case is plausible but contested, and depends heavily on program design, the number of switchers, and oversight. The policy architecture is understood. What remains is the human argument, and it is the most important one. There are three groups of Alberta children for whom this reform is not theoretical but urgent.

The first is neurodivergent students. Identified learning differences have grown substantially since the pandemic, and the school system’s capacity to serve them has not kept pace. A child with dyslexia needs structured literacy instruction delivered by a trained specialist, often in a one-on-one setting, using a specific methodology for an extended period. A child with autism may need behavioural supports, sensory accommodations, and social skills programming that no standard classroom can deliver without stripping resources from other students. A child with ADHD may need a fundamentally different pacing structure than that of any school day. These needs are not institution-shaped. They are child-shaped. An ESA allows a family to fund the reading specialist, the therapist, the modified curriculum, and the part-time school enrollment, assembling the combination that actually fits their child rather than the one their school can afford to offer.

This is precisely why, when Arizona’s ESA program was still restricted to students with disabilities before 2022, students with complex needs made up the overwhelming majority of users. They were the families for whom the gap between what the institution offered and what the child needed was most acute and most costly. Alberta has twenty accredited independent schools that exist specifically to serve students with complex learning needs. An ESA would extend the logic that those schools embody to every child who needs it, not only those whose families live near one.

The second group is rural families. The charter school expansion that Bill 25, now before the legislature, proposes to make permanent is a meaningful reform. But charter schools are, with rare exceptions, an Edmonton and Calgary phenomenon. A family in Lac La Biche, Fox Creek, or the Peace Country has access to the same provincial rhetoric about educational choice as a family in Glenora or Windermere, and almost none of its practical reality. Distance does not become less real because a government webpage describes an option that exists 400 kilometres away.

An ESA travels. It could fund a family’s subscription to an accredited online learning program. It could pay a subject-matter specialist to provide remote tutoring in mathematics or science. It could support a cooperative learning arrangement among a cluster of rural families who share resources and instruction. It could fund dual enrollment at a regional college for a student who has exhausted the local school's offerings. For rural Alberta, where the tyranny of geography has always constrained the promise of choice, ESAs are not an abstraction. They are a practical instrument.

The third group is gifted students, and they are the most politely ignored of the three. Systems under pressure, which is to say, all systems, always respond by directing resources toward students who are failing. That is a defensible moral priority in isolation. But it means that students who are excelling, and particularly those who are excelling by a wide margin, are chronically under-challenged, under-stimulated, and under-served. A mathematically gifted 13-year-old in a mid-sized Alberta city may sit through years of curriculum she mastered at ten, not because her teachers lack goodwill, but because no individual school can simultaneously serve her and the twenty other students in her classroom who are working at or below grade level. An ESA could fund concurrent enrollment in a university-level course, access to specialized online programming, or a structured independent study arrangement. None of those things requires waiting for the school district to design a program. They require only that the family be trusted to make the decision.

That trust is the core of the argument, and it is also the nerve the ATA’s campaign inadvertently exposed.

The petition’s implicit premise was that professional educators, organized collectively and funded by the state, are better stewards of education decisions than parents. That premise has a long pedigree in progressive educational thought. It holds that expertise should govern, that trained specialists know better than lay parents, and that the diversity of parental preference is a problem to be managed rather than a resource to be mobilized. It is, in short, an assimilationist argument: bring all children into the same institutional framework, staffed by the same union-credentialed professionals, delivering the same curriculum, and the result will be educational equity. They see diversity as their kryptonite.

The evidence does not support it. Canadian students have been subjected to this framework’s results for decades, and those results include stagnating literacy and numeracy scores, a growing gap between students with complex needs and the supports available to them, and a public school system whose most consistent institutional reflex is to request more funding without specifying how additional funding would produce different outcomes. The ATA’s response to the petition’s failure has been to call for more funding. It is always the same answer because the union cannot ask whether the structure itself is the problem.

ESAs do not answer that question by destroying the structure. They answer it by giving parents the power to route around parts of the structure that are not serving their children, while keeping the money in education and maintaining accountability for how it is spent. A properly designed Alberta ESA would include approved expenditure categories, audited transactions, and baseline assessment requirements. The money cannot be spent on anything other than education. The family remains accountable for outcomes. The difference is that accountability extends to what the child actually learns, not merely to whether a school received a cheque.

The government has already been moving in this direction. Budget 2026 commits $90 million to expand independent school spaces, acknowledging that the province spends nearly $5,800 less per student in the independent system and expects to recoup a $10 million capital investment within 6.6 years. Bill 25 proposes to make charter schools permanent for operators in good standing, removing the renewal requirement that has always carried an implicit threat over schools that chose to do things differently. Each of these steps follows the same logic: the public interest in education is served by expanding what parents can choose, not by defending the institutions that currently exist.

ESAs are the completion of that logic. They are what it looks like when a government stops asking families to fit their children into available programs and starts building a system in which the programs fit the children.

The ATA will call it privatization. It is not. The money is public, it stays in education, and it is audited. What changes is that the family, not the union, decides how it is deployed. That is not privatization. It is accountable to the people the system is supposed to serve.

Alberta has spent thirty years building the philosophical and legal infrastructure for exactly this reform. The Choice in Education Act, the independent school funding framework, the charter school model, and the home education provisions: each of these established, in law and in practice, that parents are legitimate decision-makers about their children’s education, not supplicants seeking the system’s permission. An ESA is the natural next step in that architecture. It is also the step that reaches the children the rest of the system has not yet managed to reach: the child with dyslexia in a classroom without a reading specialist, the gifted student in a small city whose options ran out in Grade 5, the rural family whose nearest charter school is a six-hour round trip.

The ATA-adjacent petition failed because Albertans were not persuaded that restricting choice would help children. Now the government must persuade them that expanding it will. That argument is true. The evidence supports a carefully designed ESA. And the moment to make it is now, while the institutional left is still absorbing the fact that it lost.

---