The logic of the free market, in the case of schooling, fails on its own terms.

Public school advocates have long argued that attempts to inject free market ideology and mechanics into education are misguided and destructive because the free market is a bad fit for universal public education. But in a new paper, Douglas Harris argues that the reverse is true: schooling is a bad fit for free market logic.

Harris is Professor and Department Chair of Economics at Tulane. He’s done a great deal of work with value-added measures for education (one more way to use standardized test scores to “measure” education). He also led research into New Orleans schools in the wake of Hurrican Katrina, the catastrophic event that led to reformers rebuilding NOLA schools as a system of charter schools instead of traditional public schools.

Critics of his analysis (eventually published in book form as Charter School City) found his approach “clinically distant” and failing to acknowledge some of the damage done by outsiders. At the same time, they noted that Harris did express some reservations, and Harris himself wrote that “some of the rhetoric of reform supporters has gone overboard.”

Those reservations find full expression in a new working by Harris for the Annenberg Institute at Brown University. In “How Free Market Logic Fails in Schooling— and What It Means for the Role of Government,” Harris explains how schools are ill-suited for a free market approach, and so, when looking at current school choice policy he finds, “the current direction of policy is off-track and apparently inconsistent with the main criteria on which we evaluate education policy and even with the values that voucher advocates themselves profess.”

Harris considers six conditions needed for an ideal efficient free market, and how in school “all of the assumptions fail to an unusual extent.”

Failure #1 Choices of individuals do not affect other people.

Whether Pat eats at McDonalds or Sardis for lunch does not affect anyone else’s lunch. But schooling affects everyone. When individuals learn basic functional content (reading, math, etc.) as well as democratic values, that benefits everyone (a “positive externality,” as Harris puts it in economist-speak). Harris also argues that when students all attend a school in their neighborhood, it strengthens community bonds.

Harris points out that “students affect each other’s academic knowledge as well as their expectations and beliefs” as education is produced in a community of peers. The individual’s choice affects that community. He points out that this is also affected by the many well-documented ways in which school choice becomes school’s choice, as schools seek to shape that community of students to the school’s advantage.

Failure #2 Consumers (and producers) have good information.

When we shop for a new car, we can collect information and develop a pretty good idea of what we might be getting. But what we expect from a car is pretty straightforward, while “we expect schools to do many things and goals like creativity, values, and socio-emotional skills are hard to define and even harder to measure.” Many educational goals are not observable until years later; even if we could follow children into adulthood to measure (somehow) if they became productive, responsible members of society, the information would come far too late for the school or teachers or parents of new students to act upon it.

Even if we accepted short term measures like the Big Standardized Test as useful, Harris points out that student outcomes in school are heavily influenced by outside factors.

On top of all of that, a free market encourages vendors to provide not clear and accurate information, but marketing materials that may or may not be information rich. Harris does still believe that parents are best positioned to make school choice decisions, but the “market will not, and even cannot, provide the information they need.”

Failure #3 Consumers have many options.

Harris points out that families have limited options. Geography constrains them in ways “unlike perhaps any other product or service.” Betsy DeVos compared choosing a school to picking a Uber, Lyft or taxi. It’s a troubling analogy for many reasons, but to be truly accurate, instead of each service coming directly to your home to pick you up, each would require you to go to a separate pick up spot miles away from the others.

Failure #4 No switching costs.

If Pat and Sam are tired of eating Big Macs, they can switch to Whoppers with little trouble or cost. Pat and Sam can vote with their feet easily. But switching schools comes with high costs, both academically and personally. New school may mean either missed material or wasteful redundancy. And starting over socially can be hugely taxing for students.

Failure #5 Demand is flexible.

In most markets, “we can reasonably assume that consumers can choose not to enter a given market at all.” Businesses have that additional pressure to serve customers well (or to make it harder to avoid the market). But school demand is largely fixed, by desire for education, mandatory attendance laws, and parental need for child care.

“This fixed total demand allows schools to persist in the market even when no one is satisfied,” says Harris, and this is a standard argument against public schools and in favor of school choice. But it also holds true for choice schools as well—arguably more so, since a charter or private school need only capture a sliver of that total demand. For a charter or private school, that constant demand means that families that walk away can be easily replaced without any need for the school to make any special effort. In fact, we find schools that encourage families to vote with the feet so that they can be replaced with more “suitable” families.

Failure #6 Technology can be improved and those gains can be monetized through intellectual property rights.

The Whopper is a proprietary product; other stores can’t start selling them. While certain aspects of education can be copyrighted (textbooks, software, etc.) and profited from, most of education is about implementation or delivery, and those cannot generally be copyrighted or patented. The notion of coming up with an educational innovation that one could then patent, package, and get rich selling is appealing, and repeatedly attempted, but it doesn’t fit well with education.

And Harris doesn’t even note the teaching ethic that favors sharing over hoarding intellectual property rights.

Harris sees larger implications here. If schools can’t maximize profit via money-making innovations, then it will have to find profit “in less productive ways” such as deceptive marketing and selecting students “who require fewer financial resources.”

Market Failure in Equity

Harris points out that not only are educational free markets inefficient and marked by the failures noted above, but that this inefficiency also leads to inequity.

It is not just that the assumptions underlying efficiency do not hold in the market for schooling overall, but that they are even further from the truth for marginalized, oppressed, and disadvantaged families.

Emphasis his. The features needed for free market efficiency are not anywhere in place for schooling, but they are most severely absent for those marginalized, oppressed, and disadvantaged families.

The “schooling market” has many “unusual features,” and Harris is careful to point out that these features are “owed to the very nature of schooling rather than any government policy.” These features “conspire against” competition, innovation, and efficiency, and produce the undesirable effects that we’ve been noting in school choice programs since inception.

To the degree competition does arise, it gets misdirected so that schools compete to select the best students and to provide the best marketing and customer service, not to provide the best instruction and curriculum.

What should government do?

Concluding that the schooling is incompatible with a free market approach, Harris argues that government has five roles to fulfill.

Funding, because public subsidies are necessary to prevent “under-education.”

Standards and accountability, because the market mechanism of voting with one’s feet will not operate effectively.

Requiring transparency from schools, to counterbalance misleading advertising and bad policies.

Ensuring accessibility for all students, so that discriminatory policies do not exclude some students from the full benefits of education.

Facilitate citizen engagement around the question of what sort of schools they want.

Harris notes that it’s “fairly obvious” that these government roles “argue against vouchers,” hut he also notes that they could require some public schools to shape up as well.

It is a curious feature of the modern school choice movement that it has been built around the unexamined assumption that school must go hand in hand with a free market approach, as if that is the only way for students to have educational choice. But “charter schools, intra- and inter-direct choice, and others” could allow choice within a public system that is owned and operated by the public and held accountable to the government.

To help get there, education reformers would have to decide whether they are more committed to providing students with education choice or imposing the free market on our school system. What Harris argues in this paper is that the latter choice is not a real option—even if it’s the one many reformers prefer.