If You Build It (the Right Way), They Will Come

Texas and Idaho teach us about policy design and school choice access

Two new school choice programs recently closed applications for the 2026-27 school year.

Both Texas and Idaho have procedural and prioritization requirements in statute that require the states to gather income information from each applicant. Fortunately, from a policy analysis perspective, both states publicly released some information about the income distribution of students who will be participating in the program this fall.

While it’s easy to lump all universal programs together (and, to be fair, choice opponents make a habit of calling every school choice a program a “voucher” even though no traditional voucher programs have been created in quite some time), every program is unique. Legislative and administrative idiosyncrasies are not just window dressing. Policy design can radically impact the attractiveness or accessibility of a public program. Texas and Idaho, both being Republican-dominated states going from zero to 100 in terms of private school choice in the same school year, provide a nice case study for exploring how.

In Texas, the billion dollars allocated for its Education Savings Account (ESA) program must be distributed with priority to students with disabilities and low-income students. The most-prioritized group is students with disabilities from families with annual incomes below 500% of the federal poverty level ($165,000 for a family of four this year). The next group includes all students from families with household incomes below 200% of the federal poverty level ($66,000 for a family of four), which is just above the income threshold for free and reduced-price lunch. After that, all students with household incomes below 500% of the federal poverty level (FPL) receive priority. If funds are still available after these filters, the program becomes available to higher-income families so long as they are switching out of public schooling.

Ultimately, funding for this year was limited enough relative to the quantity of applications that all awardees had incomes below 500% of FPL. More than three-quarters have incomes below 200% of FPL. Because school choice demand was so high among lower-income families—around 101,000 low-income families applied for the program, more than any other eligibility category—Texas’s ESA will functionally be an income-based ESA for its first year.

Idaho’s program differs in several ways. First and perhaps foremost, it is a refundable tax credit (RTC) rather than an ESA. Rather than directing funds from a government-administered account to educational purchases, families receive tax relief for approved educational expenses. The “refundable” part of RTC means the government will still give a family full credit for education expenses even if that credit is higher than their tax liability. That refundability pays massive dividends for lower-income families.

But, without getting into all the potential pros and cons of RTCs versus ESAs, relying on reimbursement like this can pose a challenge for many families, especially lower-income families. If you’re not receiving financial help until later, you need to have enough money up front to fund your child’s education at the beginning of the school year. Needless to say, the lower your household income, the less likely you are to have extra thousands of dollars on hand to switch your child out of public school and pay your private school bills on time.

Idaho tries to overcome this problem by providing lower-income applicants advance payment for some or all of their qualifying education expenses. I’m not intimately familiar with how the state plans to administer this element of the program, but there is some precedent for advance RTC payments. As a COVID relief policy, families could temporarily receive advance payments for the Child Tax Credit, although there are no spending stipulations with the Child Tax Credit as there are with Idaho’s RTC. These advance payments did produce some added complexity come tax time, as families had to reconcile their advance payments with their actual incomes.

I’ve had a personal concern—one that only reflects my opinion and is not a statement from my employer—that housing school choice entirely in tax law will inhibit the effective accessibility of the program. Our income tax laws are complicated, costly, and time-consuming, such that millions of people leave “free money” on the table by not taking advantage of some of the most established and flexible tax credits available. Throw in reimbursements and reconciliations, and the transaction cost of school choice becomes quite high for many families.

Given that, how do all of these policy features actually affect program participation across different income groups?

In Idaho, about 2,700 out of roughly 6,000 applications had income levels low enough to qualify for advance payments (300% of FPL, or $99,000 for a family of four). That’s about 45% of applicants.

In Texas, 36% of applicants had family incomes of less than 200% of FPL.

As a percentage of respective student populations, low-income Texas students applied for their ESA at twice the rate of low-income Idaho kids—and that’s not accounting for the differing definitions of “low-income.”

Notably, Idaho has not yet received enough applications to distribute the entire $50 million fund for its program, while Texas reached its budget cap and was forced to turn away almost two out of every three kids who applied, including about 17,000 low-income children.

It’s also crucial to note the funding differences between the two states. Despite current public per-pupil expenditures being within a couple thousand dollars between Texas and Idaho, Texas has an average scholarship value twice as high as Idaho’s (roughly $10,000 and $5,300, respectively).

Of course, there are countless variables that would affect program take-up. I’m not trying to account for all of them, and I’m not establishing any causal relationships here. But the difference in low-income applications across the two states is striking enough that it’s worth thinking about why two universal-eligibility programs might have such different launches.

My takeaway from all these numbers is, simply, incentives matter. Beyond any geopolitical distinctions between states, the structure of a program can impact who is willing and able to participate. And if we are shooting for an education system that provides opportunity for all students, we should care about the finer details.

This was originally published to our Substack.