The case for education outside state control

By Andrew Garber, founder of TheHomeschoolingCompany. Reviewed August 12, 2026.

Education Outside
State Control

New York City Public Schools reports $34,717 in per-student spending for fiscal year 2026. Its published measure excludes debt service, transportation, and food. That is a lot of money before those costs are counted.

School in a Box from The Homeschooling Company costs $149 per month per child, or $1,788 per year. These figures buy different things. Public schools provide staffed buildings and services our curriculum subscription does not replace. The price difference raises a question about how much academic instruction can cost; it does not measure the cost of replacing an entire school.

This essay argues the problem is structural: the incentives of a state school system reward expansion over results.

The Incentive Structure Is the Problem

Public education is funded by taxation. You pay whether you use it or not. The people paying for the service are not the people choosing it, and the people delivering it are not directly accountable to the families receiving it.

When customers cannot exit, providers have less reason to improve. Restaurants that serve bad food lose customers. Software that ships broken loses users. My objection is that a family unhappy with the result still has to fund the system. A request for more resources can leave the same arrangement in place, even when that family wants to try something else.

Teachers unions exist to improve teacher pay and job security. Those goals sometimes align with educational quality and sometimes conflict with it, and when they conflict a union is built to side with its members. That is not a moral failing. It is the union's function. The problem is a structure in which the interests of the adults providing education can run against the interests of the children receiving it.

Consider what expansion makes possible: more positions, more programs, and more decisions for administrators to make. My concern is that those gains are easier for an institution to recognize than the value of doing the same work with less machinery. A family paying directly for a course has a different reason to care about the cost.

The Performance Ratchet

Here is the ratchet I am arguing against. A poor result can become a case for more funding, smaller classes, better facilities, or higher salaries. Those changes might help. But if expansion is the only answer under consideration, the structure that produced the result never has to compete with a different arrangement.

The people with the power to change the system are often the people who benefit from it. Admitting that the structure itself is the problem would mean dismantling the bureaucracy that employs a great many people. They are unlikely to vote for that.

Hiring more capable people can improve individual schools. My objection is that it leaves the family's position largely unchanged: they still fund the arrangement through taxation, and leaving it does not give them that money back to spend elsewhere.

The Product Is Obsolete

Even if the incentives were fixed, a shared timetable would still be wrong for some children. When a class moves through one lesson together, the pace has to be a compromise. A child can need more time on one idea and much less on the next.

Every student is different: pace, interests, ways of understanding. A system that treats them identically serves the statistical average student, who does not exist. Students who are ahead get bored. Students who need more time get lost.

One-to-one instruction can react to the student in front of it. A tutor can spend the lesson on the missed step, connect an explanation to something the child knows, or move on when the child is ready. Providing that attention costs money. A class shares the cost, and the pace, across many students.

That constraint has loosened. An explanation can now be written from one child's current skills and interests, rather than chosen for an entire class. The economic case for the factory model is much weaker than it was.

The Moral Case

Families are required to fund a system whether or not it serves their child. When the people who control a budget also benefit from increasing it, when the people paying cannot exit, and when failure justifies expansion, the system tends to grow regardless of who runs it.

The remedy follows from the diagnosis: end the compulsion. Let families keep more of the money now collected in taxes and spend it on whatever works for their child. Let providers compete for those choices, and let the ones that deliver worse results for more money lose them.

This is how most services work. We do not have state-run grocery stores funded by property taxes that you must use regardless of quality. Competition drives improvement. The claim that education is too important to be subject to competition has it backwards: the things that matter most are the things that benefit from it.

That is why my argument reaches beyond improving a school's curriculum. A cheaper way to supply instruction gives families another option, but meaningful choice also depends on what they are required to fund and how much remains available for the education they choose.

At The Homeschooling Company, we begin from the child's skills rather than an age-based classroom sequence. You can inspect the work to see whether that starting point fits.

An example algebra lesson

Your parent dashboard

See today’s plan, completed lessons, grades, and feedback. Attendance is saved when your child checks in each day, and a weekly summary shows how the week went.