Arizona Republicans just voted to disarm the people who teach your children, and they did it by writing a ban on union payroll deduction into the state constitution—a maneuver that has nothing to do with education and everything to do with making sure the loudest organized opposition to the voucher lobby runs out of money. The authors, Heritage Foundation fellows Corey DeAngelis and Jason Bedrick, present the case in “Arizona Republicans give teachers union an education lesson they won’t forget” as a victory for families, a clean shot across the bow of a grasping union that dared to challenge the state’s universal school-voucher program. Two amendments are heading to the November ballot. One shields ESA accounts for military families. The other forbids the state from using its payroll system to deduct union dues. The framing is “protecting kids from union power.” The mechanism is cutting off the union’s oxygen and hoping no one notices that’s the point.
Let me concede what’s true. The Arizona Education Association did circulate a ballot initiative—the Protect Education Act—that would impose new regulations and an income cap on the Empowerment Scholarship Account program. An income cap low enough to cut off a firefighter married to a nurse is a real policy choice, and reasonable people can disagree about where the line goes. Universal eligibility versus means-testing is a legitimate argument. The union’s push is not without real stakes: its proposed cap would strip scholarships from families already using them—parents who chose a school and do not want to be forced back. That is a concrete disruption to real lives, and the Heritage authors are right to name it. But acknowledging that disruption is not the same thing as accepting the framing that those families’ interests and the union’s interests are opposites.
Now look at the architecture of the Republican response. When the union threatened the voucher program, lawmakers didn’t beat them on the merits of the cap. They didn’t run a counter-campaign, make the case for universal eligibility, and trust voters to choose. They put a constitutional amendment on the ballot that kills the union’s funding mechanism. That is not a policy victory. That is a procedural mugging dressed up as a lesson.
The payroll-deduction ban is the tell. A payroll deduction is infrastructure—the same system that collects your federal taxes, your Social Security withholding, 401(k) contributions, health-insurance premiums, charitable giving, and credit-union deposits. It costs the state almost nothing to run. The transaction is between the worker and the organization; the state just processes the paperwork. Describing this arrangement as unions “tapping taxpayer resources” is like describing your employer withholding your mortgage payment as the bank tapping the Treasury. The money is yours. The system is already there. The cost is rounding error. The “taxpayer resources” framing is a conjuring trick: the resource is the payroll clerk’s software running one more line item, and the cost rounds to zero. What the state is actually withdrawing is the convenience. The union can still collect dues—it just has to bill each member individually, chase the checks, and lose a chunk of its revenue to the friction. That is the whole idea. Starve the opponent, then call it principle.
This is what the Suppressed Variable diagnostic would flag, if I were the kind of writer who named her tools. The op-ed points at the union’s political advocacy and says: why should the state help? The structural factor it ignores is that payroll deduction is a universal administrative convenience available to every other automatic payment a worker authorizes. Singling out union dues is not neutrality; it is a targeted defunding of the one organization in Arizona politics that reliably opposes the voucher lobby’s agenda. The money didn’t disappear. It changed hands, from the union’s treasury to the voucher movement’s prospects. Follow the cost down to whoever absorbs it—and it lands on every public-school teacher whose collective voice just got quieter in a state where that voice was already losing.
Consider what actually costs money. Arizona, under its existing right-to-work law—which already makes union membership voluntary and already makes free-riding automatic—pays its teachers well below the national average. Make every union dollar a separate act of initiative rather than part of an automatic system, and the bargaining power that might close that gap only gets weaker. When bargaining weakens, wages stagnate across the profession, not just for union members. That transfer—from workers’ pockets to the general fund by way of suppressed wages—is orders of magnitude larger than the processing cost of running a payroll deduction. The Heritage piece mentions the rounding error. It does not mention the transfer.
The military-families amendment is the decoy. Protecting ESA accounts for children of service members polls beautifully, and framing the union’s initiative as an attack on military kids is sharp politics. But the actual provision goes further: it declares that any ballot measure conflicting with those protections is void in its entirety. The union’s Protect Education Act regulates the whole ESA program; the amendment shields a subclass of it; the conflict voids the entire union measure. That’s not a shield. That’s a kill switch. The military families are the armor, not the cargo.
And then there is the ESA program these amendments also protect. The original program was pitched as a lifeline for disadvantaged kids stuck in failing schools. It is now a universal entitlement with no means-test—open to every K–12 student regardless of family income—and more than 100,000 students use it. The voucher movement won its big fight, and good for them: the program exists, and families are using it. The union’s Protect Education Act was a countermove, and the legislature had every right to fight it. But they fought it by changing the rules of the fight itself, permanently, in the constitution, so that the next time teachers organize to challenge the voucher expansion, the union’s bank account is already empty. That isn’t a lesson. That’s a rigged table.
Here is the thing the Heritage piece is carefully not saying: the people it calls “organizations that have fought parental empowerment every step of the way” are mostly teachers. They spend more waking hours with other people’s children than almost anyone else in those children’s lives. Most of them chose the work because they wanted to be there. Framing a union’s concern about unregulated universal vouchers as hostility to parents is an old move, and it is a dishonest one. It takes the genuine and complicated tension between funding universal private-school access and maintaining a public system that teaches nine in ten Arizona kids and flattens it into a cartoon villain and a cheering section.
If Arizona Republicans genuinely believed their voucher program was popular enough to beat the union at the ballot, they wouldn’t need to defund the union first. They’d run on the merits, make the case for universal choice, and let voters decide. Instead, they’re asking voters to amend the constitution to ensure the other side can’t afford the campaign. Ask yourself what kind of confidence that signals.
What makes the piece genuinely dishonest is the pretense that this was a contest of ideas. “Lawmakers advanced two constitutional amendments to the November ballot that defend parental choice and limit the ability of unions to tap taxpayer resources for their operations.” Defend parental choice. The phrase works like a handshake at a church picnic, but what it describes is using the state constitution to financially cripple the only institution that gives classroom teachers any leverage over their own working conditions. Parental choice, in this usage, includes the choice to ensure that the people teaching your children have no collective power to bargain for smaller classes, higher pay, or a say in what the voucher regime does to the schools most kids still attend.
So what’s the alternative? The union’s Protect Education Act has real problems I haven’t defended here. A better response would have been a counter-ballot measure that addressed those problems straight on—a competing regulatory framework for the ESA program, a more sensible income threshold, stronger accountability rules, something that let voters choose between two serious education proposals. Instead, voters get a poisoned choice: one amendment defending military kids’ scholarships (obviously yes, you’d have to be a monster) and another defunding the teachers union (cast as “payroll integrity”). The substantive question—how should Arizona regulate its voucher program?—never makes it onto the ballot in a form voters can actually decide.
There is a broader principle here, and it should be boring enough that it isn’t controversial: allow a worker to direct a portion of their paycheck to any lawful organization through the same payroll system without the legislature having to reauthorize it by constituency. Voluntary payroll deduction for unions, credit unions, political causes, churches, the PTA—whatever the worker chooses. The state processes it. The state does not choose whose cause is worthy. That is the actual small-government answer, and it is the one Heritage will never propose, because the point is not to remove the state from the arena. The point is to remove one set of players.
That payroll-pipe idea raises a deeper question: how does an organization earn its members in the first place? The Ghent system in Sweden, Denmark, and Finland makes the answer explicit—when union membership is connected to an actual benefit, like your unemployment-insurance payout, people join because it pays, and the union stays strong because it is useful. I am not suggesting Arizona copy the Ghent system. There is no union infrastructure in Arizona to administer it, no set of organized employers to bargain against it, and no social trust yet earned for it. Policy without institutions is a press release. But the principle—that an organization earns its members by delivering something real, rather than by being propped up or torn down by state action—applies regardless of context. The Ghent principle says: make membership pay by making the organization useful, and let the employer’s payroll system be a pipe, not a prize.
These amendments do the opposite. They protect one universal government program by constitutionalizing it against democratic oversight, and they defund the one institution that might have objected to the arrangement. If you want a functioning school system, you need teachers who can afford to stay in the profession, and you need them organized enough to push back when the legislature decides that a scholarship for private tuition matters more than the condition of the public-school classroom down the street. The union is the counterweight. The Republican amendments are an attempt to remove the counterweight from the scale and call the result a fair weigh-in. That isn’t a lesson the teachers will forget. It’s just not the one the authors think they taught.