Recent empirical studies of school choice have come out somewhat mixed. For example, according to the New York Times, a number of recent studies have shown that charter schools may actually yield worse test scores in some cases. Yet there are reasons to doubt the reports. Most directly, test scores are not the only metric along which the economic success of school choice should be judged. Parental satisfaction also has great importance, and the data suggests that parents are happier with school choice.

The empirical debate aside, I will argue that school choice is a moral imperative, regardless of whether it boosts educational outcomes. Assume for the sake of argument that moving to a robust school choice policy regime, where any family in America can receive a tax credit to home school or to send its children to a private school, or to any other public school, will produce no net improvement in educational outcomes. We should still have such a system based on two ordinary, widely acceptable moral principles. 

First, when two policies have roughly equal empirical promise, we should choose the policy that better protects liberty. Second, parents have a much stronger moral claim to decide how their children should be educated than government does, even when greater government control could produce more social and economic equality. Let’s call the first the Liberty as Default Principle and the second the Parental Autonomy Principle.

The Presumption of Educational Liberty

The Liberty as Default Principle is based on a simple idea: if government is going to coerce people, it must meet a presumption in favor of liberty. Personal freedom is not absolute, just the default, but this means that there must be a good reason to override it. Many political philosophers from across the political spectrum hold to this principle (Joel Feinberg, Stanley Benn, John Rawls, Jerry Gaus), for it is based on the even more fundamental idea that all personal interference with the choices of others requires a justification; the interferer bears the burden of proof. The presumption in favor of liberty can be overridden by lots of considerations, such as protecting people from harm or exploitation. But in the absence of a justification, liberty should be the rule, not control or domination.

The Liberty as Default Principle only specifies who has the burden of proof in proposing legal policy. The Parental Autonomy Principle is much stronger; it holds that in cases where parental liberty is at stake it is especially hard to override. And the Parental Autonomy Principle is just as compelling. The family unit, however much it has varied across human history, is of greater moral importance than government. Parents have a much stronger moral claim on how to provide for their children than government. This is plain when we ask what could morally justify government taking children away from their parents. We generally think that parents must have seriously harmed or neglected their children.  Our reasons for deferring to parental autonomy (again, as a default) are empirical and normative—empirical because parents will better provide for their children than governments, and normative because parents simply have a strong moral right to raise their children as they choose.

However governments get their authority, governmental authority over children is hard to acquire. This does not mean that government cannot compel parents to feed or clothe their children or to educate them in general. But it does mean that when governments and parents disagree about how to raise the parents’ children, the parents should almost always prevail.

Parental liberty includes not only the right to choose how to feed, clothe, and cure children. It also includes the right to teach children just about anything, especially when it comes to moral, political, and religious matters. We already acknowledge that this right extends to educational choices on at least some margins. After all, even staunch public school advocates favor democratic governance of schools; and they probably support their local PTA. Parents should at least collectively have a say in how their children are educated. Furthermore, few people in the United States want to ban all private schools or home schools.

We also do not normally think that a governmental interest in ensuring equality of opportunity or equal outcomes overrides parental choices. We do not allow the government to pursue its interest in equality (insofar as it has such an interest) by redistributing children from bad parents to good parents, or even by forcibly removing rich children from private schools or home schools in order to increase the quality of under-performing schools. We do not coercively forbid committed parents from giving their children almost any advantage, and we’re even inclined to allow parents incredible liberty to pursue positional goods, goods whose consumption reduces the quality or value of other forms of the good.

The Nub on Rump Schools

Egalitarians are likely to respond by pointing to the potentially negative dynamic effects of school choice, arguing that it might cause schools to collapse in some areas and so actually reduce choice for the very poor. As far as I can tell, however, there’s as yet no evidence for this claim.

The more common argument against school choice is the fear of what I’ll call Rump schools. Rump schools are schools that parents would quickly flee under a regime of school choice, which would leave the remaining children, children whose parents may be unable to effectively exercise school choice, in a school with less funding and fewer teachers.

If there was evidence that school choice generally produced rump schools, that might count against school choice. But the risk of Rump schools could be reduced through ordinary market mechanisms, such as advertising and marketing to parents who would not otherwise remove their children from Rump schools. The Rump schools would lose all of nearly all of their students.

Moreover, it’s not clear that the overall effect of introducing competition wouldn’t create a better product for more poor children, even given the presence of some rump schools. For the prospect of rump schools to overturn the Liberty Principle, we’d need evidence that Rump schools were not only a general effect of school choice, but that their badness outweighs the other goods created by school choice.

But also recall the Parental Autonomy principle. Even if school choice produces some Rump schools, that is not an adequate justification for denying parents the right to make different educational choices for their children. Rejecting school choice on the grounds of equality is rather to level down by making some children and parents less free in order to help those who wouldn’t benefit from the autonomy.

Even opponents of school choice are unlikely to take this principle to its logical limit, because they don’t seem to be willing to coercively redistribute children from rich schools to poor schools, and they’re generally unwilling to ban private schools and home schools in order to increase the quality of public schools. If you’re really opposed to Rump schools, and you think they’re such a danger that government must interfere, then the forced redistribution of children from good school districts to bad school districts must be on the table. If you’re not willing to redistribute children, why are you prepared to trap them and their parents in bad schools?

Propaganda and Polarization and Segregation, Oh My!

One objection to my proposal is that we really don’t know how school choice will function as a full nationwide or even statewide program. Of course, in the absence of counter evidence, we should expand liberty, as I’ve argued. But that doesn’t mean we have to extend liberty to parents instantaneously or without concern for how quickly our society can accommodate such a change. Instead, the moral imperative of school choice should lead us to engage in experiments that gradually increase educational choice in order to help us carefully determine whether school choice will be better, on balance, for children and parents. However, it is worth noting that plenty of other countries have more school choice than we do. The Netherlands has four side-by-side school systems, so greater school choice is no utopian experiment.

A second objection to school choice is that it gives up on the public-school ideal of bringing children into contact with people and ideas not their own. With school choice, parents can propagandize their children not only at home but by assigning them to an ideologically congenial school as well. This not only reduces the autonomy of children as they grow up, but would arguably increase political polarization beyond its present level.

The problem with prohibiting school choice on these grounds is threefold. First, public schools are not neutral parties. They, too, can propagandize and polarize. The propaganda is especially problematic when it comes to promoting historical interpretations and scientific beliefs as obviously true, even though history is an interpretive art and the interpretations are just as controversial as are various ideas of the good; and science is a fallible enterprise whose findings are constantly being overturned. Public schools already polarize by creating conflicts over how children should be educated between parents with conflicting and irreconcilable values and views of history. Instead of letting parents make choices for their own children, public schools require parents to fight each other for control over the curriculum taught to all children. 

Second, the liberty and parental autonomy principles override the state’s concern for stopping propaganda and polarization. We already allow, and think we rightly allow, for parents to propagandize and polarize their children at home. It’s not clear that school choice would make these problems so much worse that it would justify overriding the Liberty and Parental Autonomy principles. 

Third, one wonders if polarization and propaganda efforts couldn’t be ameliorated rather than aggravated by a school choice system. Perhaps the government could provide additional subsidies for students to go to ethnically and ideologically heterogeneous schools. School choice need not be laissez faire.

A third objection to school choice is that it is likely to recreate historical patterns of racial and class segregation, since more highly educated white parents are likely to withdraw into private enclaves, leaving more vulnerable minority populations to schools with fewer resources. In reply, note that this is already a reality in the United States with the way we restrict who can go to which schools, which ties real estate – a notoriously regulated and inegalitarian part of the economy – to education. The promise of school choice is that it can break the connection between location, real estate, and schooling. While there would undoubtedly be some racial sorting, school choice is again not laissez-faire. Anti-discrimination law and racial equality promoting initiatives could be applied to a school choice system.

The Bottom Line

School choice is a moral imperative because it respects personal and parental freedom. Opponents of school choice disrespect the liberty of parents and may well make children worse off in the process. This is unjustifiable unless school choice can be shown to have significantly worse educational effects on children than does the status quo.

This is not to say that we must immediately have school choice, but rather that the principles of liberty and parental autonomy should serve as shared, public moral ideals that educational policy should, over time, institutionalize.

Kevin Vallier is Associate Professor of Philosophy at Bowling Green State University, a Niskanen Center adjunct fellow, and author of Liberal Politics and Public Faith: Beyond Separation.