The Economic Case for School Choice
Why the Public School Model Doesn’t Work
By David C. Rose

Those who favor school choice normally defend their position by pointing out the problems with public schools. For example, public schools are largely controlled by teachers’ unions, so school policies are set to improve the welfare of teachers at the expense of children. Public school boards, superintendents and principals are seduced by fads that come out of today’s schools of education. Since public schools don’t face meaningful competition, they don’t have to produce a high quality output to survive and teachers don’t have to be competent to keep their jobs. Public schools are inefficient, spending way too much money on bureaucracy. Public school students’ performance on standardized tests is falling over time, and so on.

These observations can be divided into two groups:

While the problems that occasion these two kinds of observations are very real, framing the school choice issue in these terms is a dangerous and counterproductive distraction. All of these problems are really symptoms of a deeper, more fundamental problem. Since there are many ways to design and implement school choice, it is essential we understand exactly what we are doing and why. If we don’t understand the precise nature of the problem we are trying to solve, it is unlikely we’ll adopt the best approach.

The most obvious problem is purely a matter of political practicality. When the case for vouchers is nothing more than a case against public schools based on impugned motives and accusations of incompetence, it alienates the very people who can provide the most powerful opposition or the most powerful support to school reform. We should spend our time making a case for vouchers independent of the failings of public schools, a case that would be compelling even if public schools were currently doing a good job. Such a case can be made and should be made.

The problem with public schooling goes much deeper than the coterie of problems we associate with them. These are only symptoms. It isn’t just a case of their not getting the job done right, it’s more than that. The real problem is that there is honest disagreement over what the job should be. Some want high octane academics, some want an emphasis on the arts, some subscribe to educating the "whole child," some are strong believers in using athletics to develop character, some want religion, some don’t and so on. I have found an astonishing amount of diversity in parental preferences regarding education. A short chat with your local public school principal will confirm this.

The fundamental problem with public education is, and has always been, that we all want different things. There is another problem with pointing to the failings of public schools when arguing for school choice. It implies that if public schools were performing well, there would be no reason to push for school choice. This has the effect of shifting the burden of proof on those who favor choice because, after all, public schools were good before, so to suggest privatizing public education now looks unnecessarily radical. We should, supporters of public schools argue, be spending our time trying to improve public schools rather than hastening their demise through self serving schemes that will reduce public school funding and in so doing make things worse. We take the bait, hook, line and sinker, and assume the burden of proving that vouchers are better when the case for public schools has never been made. They defend the higher ground for no other reason than the status quo.

The truth is that we never chose public schools, the public school system actually evolved from community schools that were, for all intent and purposes, private. In short, public schools have always been a bad idea. They look like they used to do a good job only because we cannot compare their historical performance to what might have been under a fully private system over the last century. The truth of the matter is that while a compelling case has been made for having an educated public, a compelling case has never been made for public school systems.

We have been told that one of the biggest problems in education today is the lack of parental involvement. I have always found this to be disingenuous, a way of honoring the importance of parents while blaming them at the same time. In any case, giving parents a greater voice in how their schools are run is now a popular sentiment even among teachers and principals. The point I will now try to make is that when it comes to education, having voice is inferior to having a choice. In a public school system, even if we strip away the problems associated with bureaucracy, we have the basic problem of offering a single variant of a good to people who have different preferences. To assuage the problems relating therefrom, we encourage parental input. This might make parents feel better, but the basic problem remains; different parents want different things from their schools.

Since there are many competing philosophies of education today, it is a mistake to think that parental control of how schools are run is the key to success. The fundamental problem is not parental control but the lack of parental choice that results from the way we finance public schools. The absence of parental control is not the root of the problem because when our public schools were performing well there was actually very little parental control. The root of the problem is the absence of choice from among a set of competing schools that specialize to provide parents with diverse preferences what they want.

When goods are delivered by markets, firms can do better if they cater to differences in customer preferences. In such a world, customers don’t have a voice, but they have choice. I, for one, would rather be able to choose among a number of different car models than have an equal say in how the single type of government car is produced. The greater the number of people and the greater are differences in their preferences, the more worth-less is the value of having a voice.

Let me illustrate this by way of an example. When you buy canned corn at your grocery store, you probably have a dozen or so varieties from which to choose. No one asks for your input on how canning firms should do their job because canning firms are not interested in your input. Canning firms simply do what all firms do in private markets — they produce a product that they think a large number of customers will buy. The customer has no real control, not even a voice. But we all know that this is not a problem, because if a customer doesn’t like the kind of corn a particular firm sells, he or she can simply choose another brand. Public schools are nothing like the firms that process and sell canned corn. Public education is analogous to having everyone pay a certain amount to a "corn fund" at the beginning of the year to produce a single kind of canned corn. Everyone is welcome to state their preferences and every effort will be made to satisfy everyone as best as possible. Of course, a few get what they want, but most people find that the kind of corn that is produced in this manner is not even close to their preferences.

The fundamental problem is that even public schools that try hard to be responsive to parents find that it is impossible to give parents what they want because parents want different things. A responsive school that invites parental input, then, produces an educational product that a plurality of parents want, so often the majority of parents don’t get what they want from their school. The better solution is for schools to specialize, just like canned corn firms do, and for parents to sort among the variants. Rather than generate a diverse set of schools over which parents can sort, public schools try to be all things to all people and, not surprisingly, end-up pleasing almost no one.

To see this point, suppose that 900 kids go to a particular school. As luck would have it, all parents agree on all aspects of education except for one thing: the best temperature for classrooms. About one hundred of the kids have parents who would like the temperature set at 69, about two hundred at 70, about three hundred at 71, about two hundred at 72 and the final one hundred at 73. If the parents are actively involved in their schools and therefore shape policy, we know the temperature will end up being set at 71, which leaves 73% of the parents unhappy. If, however, there were five schools that differed by the temperature they offered, everyone would be happy.

Now suppose parents differed not just on temperature but on other things like classroom size, athletics, spending on art programs and so forth. The greater the number of issues on which they might differ and the greater their differences on each issue, the greater the proportion of parents who will be unhappy and the greater the difference between their happiness level in a government school system relative to a private system, in which individual schools purposely specialize to carve out market niches. The point of these two examples is that it makes a lot more sense to let competing schools produce their output their own way and let parents choose among them than it does to force everyone to attend the same school or kind of school but to give them a voice in how the school is run.

Markets are wonderful at catering to diversity, government is not, because government tries to satisfy everyone with a single type of good. Even when government tries to be responsive and innovative it pales in comparison to market delivered education. In private markets, many individual firms experiment with many different kinds of innovations. Competition makes fear of failure moot because firms don’t know how good "good enough" will be in the future. In government, innovation tends to be along a singular path and, to make matters worse, that path is often chosen as much to minimize the likelihood of bureaucratic embarrassment as it is an honest attempt to maximize efficiency. So, parental control is not what it is cracked up to be.

When education is delivered by markets, parents have little say about how schools are run but they do have choice. When education is delivered by government, parents have only very expensive choice and the poor have no choice at all. It is not parental control of how schools are run that is important anymore than it is customer control of how canned corn is produced. What really matters is parental control over which schools their children attend. That’s why we aren’t frustrated by the fact that canned corn firms don’t ask us for our input on what kind of corn to produce.

As is always the case when there is a diversity in preferences, the best solution is a private market that thrives by filling niches that are created by such differences. When markets match parents with schools that share their educational philosophy, everyone wins, including teachers who are increasingly pressured to be all things to all students. Unless the number of parents is very, very small and/or preferences are very similar, sorting always beats voice, and this has nothing to do with education philosophy — it is a consequence of simple economics.

All this leaves one wondering if there is any case for government involvement in education at all. The economic theory of market failure strongly suggests that there is, but from this conclusion one does not end-up supporting public schools. Under a number of assumptions, it is mathematically demonstrable that the quantity of a good exchanged in a free market at the market clearing price is the quantity that best promotes social welfare. This is a well-known and powerful result. Sometimes the assumptions aren’t met, and in that case the theorem on which this result is based does not hold. Such cases are referred to as instances of "market failure" because the quantity exchanged at the market clearing price will not be the quantity that best promotes social welfare.

The most powerful argument that education is subject to market failure is the argument that those who are not educated subject us to negative externalities. In our society we expect people to make it on their own, so if they are not educated and therefore incapable of making a living for themselves, they can cause all kinds of problems for the rest of us. In my view, this is a compelling argument for mandatory education and it is becoming more relevant over time because it is becoming increasingly difficult to support oneself — let alone a family — without some minimal amount of education. There is no question that if the government played no role in education, there would be some kids whose parents either couldn’t afford to pay for their education or who would not choose to send their kids to school at all, so when those kids grew up, they would spend their time trying to steal our stuff.

The social welfare maximizing proportion of the population that attends school is 100% of those kids who are educable, but the market would produce something less than that (though I suspect probably not much less than that). So the government tells us our kids must go to school until they are 16 and provides a school, too, which it runs. But is it really necessary for the government to provide a school? Does the solution of this market failure problem really require government-run schools? We have already seen how markets give us little voice but lots of choice and, in so doing, give us what we want, which government schools can’t do unless we all want the same thing. Is there a way to have our cake and eat it, too? Is there a way to solve the market failure problem without replacing markets with government-run schools? The answer is yes — vouchers.

People who oppose vouchers but accept the wisdom of my corn example make the predictable argument that the analogy between canned corn and education is not valid. Education is very important, indeed the most important thing in the life of a child, whereas no one need eat corn. So while the market is an O.K. delivery system for corn, education is just too important to leave to the vagaries of the market. This is total nonsense. Education is important for long-term financial success and to have a happy and fulfilling life. Be that as it may, food is much more important. A couple of weeks without food and the child is dead — forget about longrun happiness and financial security. No serious person would argue that education is more important than food.

Let’s go a step further. Food is really important, even more important than education, and there are some people who can’t get enough of it with their own resources. Since starving people might shoot us for money to buy food, starvation generates powerful negative externalities for society. How does government address this negative externality? Do we have government-run farms? Government-run food distribution systems? Government-run grocery stores? Of course not. What we do have is an arrangement that lets us have our cake and eat it, too. We let the market handle the production, distribution and sale of food and we give people who might otherwise starve — vouchers. The poor don’t go hungry, nor do they have to suffer the indignity of shopping at a government-run food store. Based on our experience with the Post Office, those who receive food stamps would probably prefer redeeming them at their local grocer more than redeeming them at a government-run grocery store. Regardless of how you feel about food stamps, I am quite sure you prefer that program to one of prepaying a given amount each year for the privilege of being able to shop at the government-run food store.

To summarize, then, I personally believe that education is subject to market failure because when children don’t get at least a minimal education, they tend to produce large negative external effects on the rest of us. This constitutes a compelling case for government involvement. To infer that this means education should be delivered by government-run schools is, however, an enormous non sequitur. So the question should not be "Why should we switch to vouchers?" or "How can we be sure that vouchers are better?" The question is, given how markets allow parents to sort by their personal preference and government-run schools don’t, offering only a meaningless voice in return, what is the case for government run schools? I simply don’t see it and I doubt there is one, just as I doubt there is one for having government run farms and grocery stores.

The only sensible explanation for having government-run schools is a desire by those who control government to extend their control over education, too. One can certainly question the motives and competence of those in public education, but the truth of the matter is that the public school system in general and its teachers in particular never had a fair chance to succeed. When schools were more locally controlled, populations were more segregated and, hence, homogeneous, and traditional pedagogy was the only widely recognized education philosophy, the public school model, though flawed fundamentally, did reasonably well. But in the face of extreme (and growing) heterogeneity of parental preferences regarding education, satisfying everyone is impossible and is becoming more so time.

If I haven’t sensitized you sufficiently to this point yet, suppose the government took over every aspect of canned corn production and that you have been put in charge of corn farms, corn distribution systems, corn canning plants, etc. There will now be only one kind of corn. How successful do you think you’d be in satisfying corn consumers? You’d fail, of course, because the task is impossible. Our teachers need to understand this point and they need to know that we understand this point. Such an argument will ring true with our teachers who are all too frustrated by this very heterogeneity and will lay the groundwork for more subtle arguments that will demonstrate that the path to greater income and professional respectability is market delivered education. Of course, the reason more teachers haven’t already reached this conclusion on their own is that this is also the path to reduced influence and power of the NEA.