School choice actually puts more money into the budgets of government schools, and saves money for state budgets as well. That may be hard to believe, since we so frequently hear the opposite—that choice “drains money” from government schools—but it’s established by a large body of empirical evidence. If Oklahoma wants to improve both school budgets and state budgets, it should give every family in the state an Education Savings Account (ESA) in 2022.

That choice “drains money” from affected government schools is a canard. If I choose to have my appendix taken out at St. Paul’s Hospital, I am not “draining money” from the budget at St. Peter’s—not robbing Peter to pay Paul, as it were. If the government sues my business for declining to make a wedding cake for a man who’s marrying a shoe, and I choose to hire Atticus Finch to defend me, I am not “draining money” from Perry Mason. The government school monopoly is the only place anyone ever thinks of people as “draining money” from every service provider they don’t choose to work with.

In fact, a well-designed school choice program actually delivers a better education while reducing the cost burden on both affected government schools and state budgets. The latest major study on this topic, published in March by Marty Lueken of EdChoice, examined every choice program that had existed for at least three years (not including century-old, revenue-neutral “town-tuitioning” programs in New England, or individual income tax provisions) and found that almost all of them save money. Over their lifetimes, the programs had saved at least $12 billion, and may have saved as much as $28 billion.

This is only the latest in a long line of empirical research. EdChoice counts a total of 70 studies on the fiscal effects of school choice programs. Of these studies, 65 found that the programs being studied save money. Four found that the programs being studied were revenue-neutral, and five found programs that cost money. (The numbers don’t add up exactly because a handful of studies contained more than one finding.)

It’s true that a small handful of the nation’s large number of choice programs—76 and counting, in 32 states plus D.C. and Puerto Rico, serving over 600,000 students—don’t save money. But it’s not hard to design a program that avoids this eventuality. A few programs are poorly designed, or are designed with other priorities at the forefront (such as giving parents financial parity with the bloated government system).

ESAs are an especially attractive form of program design if the predictability of fiscal effects is a concern. With an ESA, the state gives each participating family a fixed amount of money each year, which can be spent on private-school tuition and other eligible education services, with the remainder saved for college. This means there are no unexpected variations in how much money each participating family will take out of the program, as there can be with other program designs. The state can then set the amount each family gets at a level that will generate savings.

Choice saves money for government schools because when a family chooses to remove their child, the school’s costs go down more than its revenue does. A big portion of a government school’s revenue—in Oklahoma, the average is 42%—comes from local sources, mostly property taxes. These local funding streams don’t change much when enrollment changes. (You may have noticed that the tax assessment on your house or your car doesn’t include the enrollments of nearby schools when calculating what you owe.) So when a child leaves a government school, almost half its revenue is unaffected.

Meanwhile, a school’s spending is mostly made up of “variable costs,” which means costs that go up and down as student enrollments go up and down. By far the biggest cost in school budgets is staffing—from teachers to bus drivers—and schools need fewer staff when they have fewer students to serve. Only a small portion of school budgets are “fixed costs”—overhead, like the heating and air conditioning bills, that don’t change much with enrollments.

This isn’t just theory. Testing this hypothesis is exactly why there has been a lot of empirical research on the effects of choice programs. The evidence shows that schools are saving money.

Whenever a choice program is enacted, education special interests wail to the cameras about the enormous proportion of their budgets supposedly made up of fixed costs, which remain in place when students leave. It’s a funny thing, though; when state legislative committees get together every year to set funding formulas for state support that’s based on enrollment, those same lobbyists suddenly get a strange form of amnesia. They wail to the committees about their need for big per-student allotments from the state to cover their enormous variable costs.

Choice saves money for state budgets, too. In a typical choice program, the cost to the state to fund the program is less than the amount the state sends to its government schools through its per-student funding formula. A student who moves from the government school system into a choice program is moving from a program that costs the state more per student to a program that costs the state less per student.

Now, it’s important to note that the details of each particular program make a big difference here. For example, if a state sets the funding of the choice program to exactly the same amount it spends per student in the government schools, the program will be revenue-neutral for the state. It can choose to spend even more on choice, sacrificing savings to other priorities. Program design can also affect the balance of how much of the savings go to state governments versus local schools.

It’s also true that children who enter a choice program when they are beginning school for the very first time do not save government schools money, although they do save states money (because they never enter the more-expensive government school system). The evidence finds this isn’t a big enough effect to remove the net savings. Also, if a choice program gets high participation, we can expect local governments eventually to adjust their tax burden—but that’s very unlikely either to happen quickly or to remove all the savings.

These fiscal savings may be one reason choice tends to improve educational outcomes in affected government schools. A large body of research finds that choice delivers a better education, not only for the students who use it, but in affected government schools. Adding some margin in the budget may be one reason for this effect.

Delivering better results while saving money both for schools and states is about as good an outcome as you can ask for. The only people who would lose if Oklahoma gave every family an ESA are the special interests. But it’s amazing how easily you can tune out their wailing once you know how far it is from the facts.