Limiting school choice programs to low-income households is bad for low-income households. It hurts the rest of us, too, and that matters. But the people it hurts the most are the people who most need better schools, and that’s what matters most.

Nobody needs school choice more than poor people, because nobody is served more terribly by the government school monopoly. That’s why Milton Friedman used to say that school choice “ought to have been a Democratic issue” rather than a Republican one. It’s also why, in the past generation, it has increasingly become a Democratic issue—at least in many state capitals, where the teacher unions have less power, progressives and Democrats not in thrall to the unions have become critical allies in the school choice coalition.

Poor families don’t need a bigger, fatter government bureaucracy. No matter how much money we throw at the broken monopoly school system, year after year, increases in spending consistently fail to produce better results. Poor families need power over their children’s education. They need school choice, which puts parents where they belong—in charge.

Sometimes the worst thing you can do for the poor is “help the poor.” What we want to do is tear down the walls that prevent poor people from making themselves into non-poor people. That’s what “helping the poor” ought to mean. But all too often, it really means building walls between poor and non-poor people, reinforcing the divide rather than tearing it down.

Throwing middle- and upper-income people out of school choice programs is a classic example of hurting the poor by “helping” them. It creates a sharp, government-enforced division between two separate and very unequal populations. On one side of the wall are poor people, who receive school choice; on the other are non-poor people, whose tax dollars provide them with school choice.

The first and most important way this hurts poor people is by denying them access to educational innovation. The real promise of school choice, and the only serious hope for delivering better education to the poor, is not in moving students from existing public schools to existing private schools. It’s driving an educational revolution in which educational entrepreneurs reinvent school from the ground up.

As Milton used to point out, education is the only thing our society still does in essentially the same way it did it in the 19th century. Medicine, transportation, communication, manufacturing, entertainment—all have been radically transformed, on balance much for the better. But education? For the most part we’ve still got 30 kids in a room, in a building they were assigned to based on where they live, sitting at desks, in rows, looking at a teacher writing on a board.

These days it’s a white board instead of a black board. Compare that to the difference between a telegraph and a smart phone.

I’ll make one concession. The end of the government school monopoly’s policy of racial segregation was a huge step forward—although I would argue that’s less a change in education and more the removal of ethnic injustice from the delivery of education. And of course, the government school monopoly still perpetuates segregation de facto by assigning students to schools based on where they live. The empirical research shows school choice produces greater ethnic integration in schools; it is the only education policy that has reliably and sustainably done so.

Could the entrepreneurs of previous generations have invented and successfully brought to market the telephone, the car, the airplane, or the computer if they had known in advance that they would only be allowed to sell them to poor people? Even poor people with vouchers?

Other than that, educationally we’re stuck in the 19th century. That’s why existing private schools are, to be blunt, not that much better than public schools. The data show that they’re better, but only moderately so. They’re not, on the whole, dramatically better. The dominance of the government school monopoly (about 90 percent of all students) makes it seem implausible that anything radically different could be worth trying, so few try it.

And what did 19th century America do a really lousy job at? Providing opportunity to the poor.

The government school monopoly wasn’t created to facilitate social mobility; quite the opposite. Go back and read the history. The people who created the government school monopoly didn’t talk about opportunity and advancement for the poor. They talked about how the poor needed to be better prepared to spend their lives in the station into which they were born—better prepared to do the kinds of jobs that are appropriate for, you know, that kind of people.

Our rigid, standardizing, factory-like schools are well designed to teach young people to know their place and stay in it. That’s yet another reason we need a radical revolution to reinvent schools.

Educational entrepreneurs are ready for the challenge. But they can’t do it if they don’t have a client base to serve that is large, reliable, and looking for the kind of revolution they can deliver.

Could the entrepreneurs of previous generations have invented and successfully brought to market the telephone, the car, the airplane, or the computer if they had known in advance that they would only be allowed to sell them to poor people? Even poor people with vouchers?

The more students are eligible for school choice, the more stable that population is, and the more hungry for innovation it is, the more school choice will empower educational entrepreneurs to reinvent schools. And the people who will benefit most from that are the people who have the most to gain—the poor.

Another reason—almost as important—why it hurts the poor to throw middle- and upper-income people out of school choice programs is that it cripples the political coalition supporting and protecting choice. It limits the number of people who have strong reasons to invest in supporting school choice as voters and activists. And it ensures that such support as school choice does have is drawn from the weakest part of society.

Choice programs need to be well designed to succeed, as the failure of a badly designed program in Louisiana has recently reminded us. Implementation also matters; Florida’s A+ voucher program was badly sabotaged by ridiculous procedures that the state bureaucracy created for parents who might want to participate. Strong political support, from the start, is essential for high-quality program design and implementation.

This is what Milton had in mind when he used to say “show me a program for the poor, and I’ll show you a poor program.” Policy gets made in a political environment—there’s no escaping it. A policy designed to attract few supporters, and from the politically weakest part of society, is going to be badly hindered from the start.

That’s the short-term political problem. There is also a long-term political problem.

Milton Friedman used to say “show me a program for the poor, and I’ll show you a poor program.” Policy gets made in a political environment—there’s no escaping it. A policy designed to attract few supporters, and from the politically weakest part of society, is going to be badly hindered from the start.

Like all means-tested programs, school choice divides society into two groups whose interests are at odds with each other. One side gets all the benefits, and pays little or nothing—certainly it pays much less than it receives. The other side pays the bills and gets nothing, other than the power and control that comes with paying the bills.

This is not a formula for a society in which walls are torn down so poor people have the opportunity to make themselves into non-poor people. It is a formula for a society in which the walls only get higher and thicker. “If we’re going to pay the bills for these people who can’t take care of themselves,” the non-poor say, “we have a right to arrange things the way we want them. And what we want is for people who can’t take care of themselves to stay away from our kids.”

And that’s not the worst of it. This kind of system divides society into two groups whose economic interests are directly opposed to one another. The worst of it is rising levels of mutual hostility, much of it along ethnic, religious, or other tribal lines.

In the school choice programs that already exist in most states, poor parents have proven that they can take care of themselves. They make better choices for their children than the government monopoly does. All we need to do is tear down the dividing wall that keeps them from improving their lives. Means-testing school choice keeps that wall in place.

Greg Forster (Ph.D., Yale University) is a Friedman Fellow with EdChoice. He is the author of six books, including John Locke’s Politics of Moral Consensus (Cambridge University Press, 2005), and the co-editor of three books, including John Rawls and Christian Social Engagement: Justice as Unfairness. He has written numerous articles in peer-reviewed academic journals as well as in popular publications such as The Washington Post and the Chronicle of Higher Education.