Video 3 of 12 of the Libertarian Public Policy guide (Download MP3 Version)

Education 26:32

It’s doubtful whether any government involvement in education is a net good, and it would be better for the state to finance vouchers than to run schools itself.

Miron: Today we’re going to talk about government policies toward education. That is of course one huge area of government intervention in many, many different ways. Governments own and operate public schools at K-through-12 level, state college and university systems. They offer tuition subsidies, some federal, some state, various grants, loan policies. There’s research funding from the government that helps support universities and colleges. There are laws like compulsory education which require every child up to a certain age to be in school for a certain number of days per year, required curriculums, teacher certification programs, high-stakes testing, and much, much, much more. So there’s a huge range of intervention. Government is extremely involved in trying to regulate, define, subsidize education.

Libertarians oppose much or all of this intervention. There are plausible arguments, at least, for some modest degree of government involvement in education, although there’s cost even of that. It’s a hard question as to whether there should be even this limited involvement. But certainly, libertarians will argue against the vast majority of what occurs or for substantially, enormously reduced degree of intervention.

I’m going to address these issues in the following way. I’m first going to talk about the question of why should government subsidize education at all. Why are we involved in any way in trying to promote education? And then I’m going to talk about the potential cost of those subsidies. That is looking at the general question of why are we distorting the market at all. Why don’t we treat education just like every other good and let supply and demand determine how much is produced of that good? And for talking about that, in concluding that, it’s not so clear that we should be subsidizing at all. Take the different question of given we’re subsidizing, if we are subsidizing, how should we subsidize and argue against the current approach? Go from that to talking about vouchers or tax credits as a way to encourage education to subsidize if government is going to be involved in some way, rather than operating public schools.

And finally, accept that maybe we’re going to have public schools for a long time. But are there ways that we could reduce the negative effects of public schools? Could we make public schools better? Are there some changes in the way those operate? And talk about potentially good ideas for making them better, and some bad ways that people are trying to make them better.

The first big question is why do we treat education differently than all the other things that people buy and sell? We let people make their own decisions about very important life decisions: how many children do they have, most of the decisions about how to raise your children, about savings, about whom to marry, about what occupation to go into, about everything. Why isn’t education just another thing that people should be able to decide do they want it, is it useful, how much should they get, how much should they get for their children?

There are three standard arguments as to why education should be treated differently than markets for supply and demand, for toaster ovens, or for all sorts of other standard goods and services. And they’re under the labels of externalities, myopia, and credit constraints. And I’ll talk about each of these a little bit.

The externality argument says that my education has a beneficial effect on you. That means that when I decide whether to get some education, I’m likely to only look at the private benefit and cost to me and not recognize that if I get more educated, I’m helping everybody else that I might work with or interact with or things like that. A lot of people have made that argument, including many libertarians like Milton Friedman. And it seems plausible at some level. If everyone reads and writes the same language, potentially, they can communicate, they can do business together, and things like that. If everyone knows the same rules of arithmetic or calculus, one can certainly imagine that there are ways in which that’s beneficial. So the notion that there are some positive spillovers, from some kinds of education, at least, I don’t think should be dismissed out of hand. And certainly, if we’re talking about K-through-3 or K-through-6 education, everyone getting at least the basics of the three R’s or something like that may make some sense. That doesn’t by itself mean we need government to make it happen, but it’s not a completely silly argument.

The second argument for subsidizing education is that some parents will not make good decisions on behalf of their children. Children, of course, are not in a position to decide whether or how much education to get. Some parents should be sending their kids to school but won’t, will instead let them sit at home or have them work in the fields if it’s an agricultural society or have them apprentice in the family business or something like that. That’s not necessarily the right decision for those kids. It’s selfish or short-sighted or myopic decision by the parents. Therefore, we need to subsidize to encourage such parents to send their kids to school by making it cheaper. That might incentivize them to go ahead and let the kids go to school, instead of using them to bring the crops or whatever.

The third argument says there are lots of people who know that education is useful, is beneficial for their children. They would like their kids to have education, but they can’t afford it. Education of course costs money. As we’ll discuss, it doesn’t need to be as expensive as it is. But certainly, it costs something. So people with very limited means may want to send their kids to school but have difficulty in doing so because they can’t access the money, they can’t easily borrow because they’re not likely to have enough income to pay back the borrowing that would entail and so on and so forth.

All of these arguments have some element of truth. I wouldn’t want to say that they’re evilly motivated. They’re reasonably motivated. They go to reasonable interest. Whether they are large or small is of course a very different question. Whether they apply to a broad range of education, all kinds of education, and so on and so forth is much harder. They give us some basis for thinking that there might be some reason to intervene, to subsidize, to encourage people to get more education than they would if it were entirely up to them, entirely in a free market. Yet there are going to be costs of subsidizing education. Perhaps most importantly, if government is going to subsidize education, it has to decide what is education. It has to define it. If it were going to, say, give people vouchers to go out and purchase education, you’ve got to say what those vouchers can be used to purchase? Can those vouchers be used to purchase something that looks like a standard elementary school now? Or can it purchase an education in a ski instruction school or a French immersion school or a science immersion school? Can it purchase having kids be trained just to bring the crops in from the field? What does it go for? That means that government has this huge power over the education marketplace because once it’s doing the paying, it’s inevitably going to have to be doing the defining. Otherwise, the policy just doesn’t make any sense. That has the grave danger that it promotes huge standardization. Government officials will get together, there’ll be committees, there will be studies, and they will say the right kind of education is whatever it is – maybe something that looks like today. And they will not accept that for many kids, the standard education isn’t especially effective, or at least not by itself. Maybe lots of kids should have half of what they currently get now. But half should be something that’s more fun or more interesting, more sports or more science or more language or a whole range of things we haven’t even thought of and that bureaucrats won’t ever think of, but individual parents might think of and do things that are beneficial to those kids. You might say that’s going to be more important when you get to be 13 or 15 or 20 than if you’re 7 – where pretty much every kid maybe should be mainly learning the three R’s, but still it’s going to apply in some measure everywhere.

So if the government is subsidizing education, that’s going to mean standardization. And there’s potential huge loss for matching people with different interests, skills, and needs to different kinds of education.

Closely related, government being in this business is going to discourage innovation. If there’s a marketplace full of people using their money, or possibly, as we’ll discuss, using vouchers to purchase education, then there’s an incentive for people who think they have something new to offer, maybe a school for K-through-6-year-olds that starts teaching computer science skills along with reading and writing on day one instead of only when you get to junior high school or high school. There could be a strong incentive to innovate and offer that. And maybe it will be popular and very useful for a lot of people. But if the government is defining what the subsidy can go for, your incentives to do that are likely to be quite limited.

In the worst case, of course, if government is defining an education, it has the power to engage in thought control. It’s not an accident that virtually all totalitarian regimes have tried to control their education systems as one of their primary goals. Extreme cases are communist China or Soviet Union indoctrinating everyone to accept the wisdom of the government, the validity of the government, and so on. We don’t see that in any serious way in places like the United States, Western Europe, modern economies now. But the potential is there. Slopes can be slippery. Even the fact that government is subsidizing education sends a small bit of thought control that says you need the government to have education. But that’s not true. You can have education easily without the government. So by getting in that business, the government is already taking a stand that tries to push people to think about education a certain way.

Government involvement, government providing a subsidy also forces it to take a stand on a bunch of messy issues. Should colleges, high schools, whatever have speech codes? Should they practice affirmative action? Should they have single-sex education? And so on. Those have been very controversial issues for decades now. If the government is subsidizing education, it’s in at a minimum and awkward position because on the one hand, you might think single-sex education could be a good thing. Maybe it’s better for some women to only be with women in college. And maybe it’s better for some men. Maybe affirmative action in hiring is actually a useful thing for creating a good school. And we would want some schools to be able to do it. And yet those types of actions are sufficiently controversial, and they run into constitutional issues that if the government is providing the subsidy, it’s at a minimum, awkward, and complicated to decide where you draw these lines and what is allowed and what is not. So that’s another cost of having the government be involved. And of course government subsidizing education has a direct cost where you spend tens of billions of dollars on education. So that’s part of the cost, plus the distortions caused by the taxation necessary to fund it. So all the subsidy comes with a bunch of negatives that need to be considered and compared to the possible benefits from spillovers from one person’s education to everyone else’s, from helping kids who were born with parents, who don’t want to send them to school, or helping people who can’t afford to send their kids to school.

So what’s the bottom line on that? I’m not going to take a strong bottom line on that, except to say the level of subsidy that U.S. and other places currently provide seems grossly excessive. It’s way beyond making sure that every 8-year-old gets a chance to learn to read and write. We have government subsidizing college, universities, business schools, dental schools, law schools, and so on. All of these issues might argue for some government subsidy. The beneficial spillovers, the myopia, they don’t apply with anything like the same kind of force to a 26-year-old who wants to get an MBA as they might do a 5-year-old who might happen to be born into a family where the parents aren’t doing the thing that’s in the child’s best interest. So at a minimum, I think thoughtful consideration suggests way less subsidy than we currently have.

It’s of course inevitable that for the foreseeable future, we’re going to continue to subsidize, probably substantially, but certainly to some degree. Then there’s the question of if government is going to try to promote more education than would occur otherwise. How should to do so? There are many current approaches now, especially government-owned and operated schools. That’s very widespread at the K-through-12 level. And it’s also widespread at the college level. About three quarters of enrollments in the U.S. are in public-owned and operated schools. You could also consider things like grants, scholarships, vouchers, and tax credit that we’ll get to.

The key point I want to make is that the subsidy argument, the idea that people are not getting as much education as would be socially desirable on the own, even if it’s right, in no way, shape, or form the government needs to be owning and operating the schools. There’s no market failure. None of these arguments says that no one knows how to create a school, that there are monopoly problems, or any sort of externality or inefficiency in building a school and advertising a school and enrolling students into school and hiring teachers and all of that. So if you want to subsidize education, you don’t need to have any government-owned or operated schools. It could all be accomplished by giving purchasing power to the people who are not getting as much education as you think they should be getting for themselves or purchasing for their children.

How can you do that? How can you transfer this purchasing power so that people have more ability to purchase education in the marketplace? There are two main ways: vouchers and tax credits. I’m going to talk about vouchers because it’s simpler. I don’t have a strong feeling about the two approaches, although I mildly favor vouchers because it helps us avoid making the tax code more complicated. But reasonable people certainly do disagree and can disagree on the exact method. I’m going to talk about vouchers for simplicity.

First, what exactly the vouchers mean? Take an example of K-through-12 or K-through-8 education. It would mean that every parent of a school-aged child gets a piece of paper that is official and has registration and stamp and something on it that says legitimate that they can use to go to a private school to purchase education. And it has some amount, some face value. Call it $5000 a year. So any parent can use that for each of their children to buy whatever kind of schooling the definition of the voucher allows. Now a crucial issue is we’re going to get the benefits of this approach much more if the voucher is defined very, very broadly. And that’s one thing that we’ll have to talk about. But that means that there is now competition because anybody can start a private school and accept these vouchers. It means there’s scope for innovation because someone who wants to start new types of schooling or education can hope to attract these voucher dollars from parents who have them.

Using vouchers makes it somewhat easier, not completely simply by easier, to avoid the issues about whether to have single-sex schools, speech codes, and things like whether to tolerate those because the definition of who can accept the voucher can be very broad. Any school that satisfies very broad criteria can accept them, although the issue still can come up.

One thing that vouchers may or may not do, and it’s important to recognize, is dramatically raise the amount of education relative to the cost by making the education better. It might be, and probably is the case in a lot of situations, that current public schools are doing an okay job, but they’re just doing it much too expensively. What the vouchers do is allow us to first not necessarily pay the teacher union wage premium, and secondly to allow for other kinds of innovation and cost saving that make education less expensive. Whether allowing the vouchers will lead to dramatic improvements in test scores or kids getting better educated, the evidence isn’t quite so compelling that you will see dramatic effects. The evidence is suggestive of some mild effects, and those are good. But the best hope is probably vouchers would give us similar results at many fewer dollars, not that it will give us dramatically better results for fewer dollars.

Now one possible risk of the voucher approach is that government does now have to say what constitutes a school, what type of entity can receive these vouchers from parents and then get reimbursed for them by the federal government. And some libertarians worry that if you do that, that puts the government into position of defining all schools and possibly gives it the scope to then impose very heavy regulation on all schools. I think that’s not the right way to think about it and is not a major risk, but it’s worth discussing. First of all, all of these voucher schools have an incentive to lobby now against those regulations and that standardization, all these rules that limit the types of schools – whereas the public schools don’t have that incentive. If you think you have something that’s valuable, if you can convince your legislator or some legislatures, you can make a case that says we should define the rules of what constitutes a school more broadly. So private efforts, the lobbying by schools or associations of schools have some ability to perhaps push back against excessively tight regulation of what constitutes a school.

Secondly, whether or not that first thing is true, as long as we have compulsory education, the government is already in the position of defining all schools. The compulsory education law is the most fundamentally bad for limiting scope, innovation, and variety of different kinds of schools because if the law says every child has to be in school, you have to define what constitutes a school. A huge example of this has been home schooling. Initially, when some people wanted to engage in home schooling, they had big fights on their hand with the local regulator saying, “Well, no, staying at home and watching Sesame Street, or even if you’re staying at home getting lectures on nuclear physics from your parents, that’s not being in school.” Government already had the power to say exactly what constitutes a school. Even setting aside home schooling, if you want to send your child to something that’s not the public school and it doesn’t satisfy certain state-mandated curricula and so on, the state can say you’re not complying with compulsory education laws; therefore, it already has the power to limit what constitutes a school. So I don’t think that vouchers make that issue worse, but it is something that many libertarians are concerned about.

The last main point is say we’re stuck with public schools. Can we do something to make them better than they are? And what is it that’s wrong with public schools? Well, it depends a lot on which public schools you’re talking about. Public schools in middle-class and upper-income suburbs in most parts of the country are pretty good. Tons of kids go there. They’re happy. Parents want to live in those neighborhoods. They’re happy paying high property taxes to be able to live in those neighborhoods and send their kids to those schools. There’s lots of competition between those public school systems in many states because they have small towns and districts that compete with each other. So in many cases, public schools are pretty reasonable and certainly, there are no huge or obvious negatives. But public schools in other cases are quite horrific. They’re dangerous, they’re violent, there’s not much education going on, and so on.

One approach that’s been discussed and is implemented and used quite widely now, both from state actions and more recently from federal policy under No Child Left Behind, is accountability in high-stakes testing. And this is a case that I think is useful for libertarians because it illustrates that you need to be careful of wolf in sheep’s clothing. The accountability mantle sounds good to libertarians until you think about the details in this particular application. Of course, libertarians like the idea that people who sell things in the marketplace are accountable to their customers, and if they make crummy product or of too high a price, people don’t buy it. So that suggests a kind of accountability. The high-stakes testing approach in the public school setting said we’re going to make everybody, every kid take these the standardized test. And we’re going to look at the scores and we’ll compare them to other schools, compare them over time, and so on. And schools that have really low average scores or schools whose scores go down or things like that, we’re going to make them accountable by taking away funding, taking away teachers, forcing new practices or curricula, whatever on them. So at one level, it sounds okay. We’re trying to give the schools an incentive to behave better, to spend more time teaching and less time doing recess and things like that. The problem is that the accountability high-stakes approach clearly changes incentives, but it changes a lot of the incentives that the public schools face. One incentive it creates is to do things like calling parents of kids who have low test scores and telling them to stay home the day of the test because then their test scores won’t be in the average. It creates an incentive for teachers to actually cheat, to take the Scantron sheets in which the kids have filled in their answers to all the questions on the standardized test in the little bubbles and go in and erase them and put in the right answers because that gets that teacher scores to be higher and encourages reclassification of students as not proficient in English language so that their scores might not be counted in the averages for the school. So the accountability approach is sort of a government version of accountability. It’s a government approach to bad government of public schools. It has worked quite imperfectly all of these possibilities I’ve just mentioned indeed have happened, all these adverse incentives created by the high-stakes testing. Libertarians think the right form of accountability is accountability to parents, the fact that parents can take their kids and send them to different schools if they don’t like what the school is doing. They need the purchasing power to do that, and that’s why the voucher approach has appeal. It allows you to say if you don’t like what’s going on, where you’re spending your voucher this year, send your kid to a different school next year, and you get to take your voucher with you.

There are some things that might have some promise in terms of improving existing public school system. They basically consist of getting rid of regulations and rules that didn’t make much sense in the first place. They’re adding cost without any obvious or tangible improvements in value. So barriers to entry in teaching is an obvious one. To be a public school teacher, you need to have gotten various kinds of accreditation in most states, taking certain amounts of education degrees, credits, and so on. There’s very little reason to think that those particular requirements make people better teachers. So that just makes it more expensive to become a teacher, reduces the supply, and raises the cost. And it may even create an adverse selection of people who want to become teachers because people who are really on the ball don’t want to sit through these mushy-minded education courses in order to get their teaching certificates. There’s lots of regulation of the way the schools operate in terms of bilingual education. That forces additional classrooms, it forces additional procedures, and almost certainly is counterproductive for kids whose first language is not English instead of immersing them as quickly and as much as possible in English – which all the data would show, and most people’s personal experience would show, gets kids to learn English really fast, especially when they’re young. We instead have these redundant and extra classes to try to maintain kids’ ability to keep learning in their existing language rather than just converting to English as quickly as possible.

Teachers’ unions is of course a huge thing. If you could change the laws in states that allow teachers to unionize and therefore reduce the union wage premium, it would save a lot of money. It wouldn’t automatically make the quality of the schooling better, but it would get you similar quality at lower cost by eliminating the teachers’ wage premium.

Finally, but very, very importantly, if all regulation of public schools were left purely to states, and better yet, if states left it to local cities, towns etc. as opposed to having the federal involved, that has the potential for huge benefit. The federal government is inevitably going to generate tons of cost, clearly has much more potential for thought control and excessive standardization and so on, relative to lower level of government involvement in education. So repealing No Child Left Behind, but also every prior federal government intervention back to the ‘60s, would certainly go in the right direction for allowing public schools to do what the local citizens wanted to do with their public schools rather than the federal government.

To sum up, there is a plausible case for some subsidy. Assuming that subsidy were much smaller, none of it was federal, it was controlled purely by local or maybe state levels, I don’t think libertarians can assert with 100 percent certainty that the optimal level of subsidy for education is zero. But it’s almost certainly much, much smaller than it currently is. There’s a very compelling case for eliminating government schools for subsidizing education simply by giving people who should be purchasing more, that we think might want to purchase more, the purchasing power via vouchers or tax credits rather than doing it by owning and operating the schools. If we’re stuck with public schools, there are some things that could be done to improve them – basically less regulation, and in particular, elimination of the federal role that has the most potential to do great harm relative to all the things that are happening at the state and local level.

So education is super important, matters to everybody. That doesn’t mean the government should be involved in it a lot. That’s a crucial thing that you’ll hear again and again in these lectures. Just because something is good doesn’t mean the government should subsidize it. And just because something is bad doesn’t mean the government should try to penalize it. To a much greater degree in education and elsewhere, we should be letting individuals make their own decisions. Thank you very much.