School Choice

The Case for Education Vouchers


Imagine a society in which children attend school because they enjoy it rather than because "the law says so"; in which parents carefully choose from a variety of alternatives the type of school best suited to each of their children and willingly paid for it, recognizing that such is their responsibility as parents. Imagine a society in which education is more than just sitting in rows in a classroom for 12 to 16 years, where it is an integral, lifelong part of living, made exciting and relevant by a widely diversified, creative industry. Sadly, you haven't been thinking of America.

In America, as Charles E. Silberman1 has observed:
"It is not possible to spend any prolonged period visiting public schools without being appalled by the mutilation visible everywhere: mutilation of spontaneity, of joy in learning, of pleasure in creating, of sense of self.…

Because adults take the schools so much for granted, they fail to appreciate what grim, joyless places most American schools are, how oppressive and petty are the rules by which they are governed, how intellectually sterile and aesthetically barren the atmosphere, what an appalling lack of civility obtains on the part of teachers and principals, what contempt they unconsciously display for children as children."2

A large part of the problem relates to the bureaucratic nature of the public school system, especially in the cities. In an article entitled "Is the Public School Obsolete?" Christopher Jencks, characterizing the "organizational sclerosis" of the public schools, described big-city schools as "a system of education whose first axiom is that everyone, on every level, is incompetent and irresponsible."3 As a result,"…the school board has no faith in the central administration, the central administration has no faith in the principals, the principals have no faith in the teachers, and the teachers have no faith in the students. Decision-making is constantly centralized into as few hands as possible, in the hope of reducing errors to a minimum. Of course, such a system also reduces individual initiative to a minimum, but that is the price which a publicly-controlled bureaucracy, whose aim is not profits but survival, usually seems willing to pay." 4

Jencks goes on to describe the ways in which the system tends to encourage conformity and docility, while discouraging innovation and creativity, among teachers, administrators, and students. Another factor suppressing innovation is public control—"the minute accountability to publicity-hungry elected officials" that puts a premium on being noncontroversial.

Despite recurrent hopes of reformers, given such constraints it is unrealistic to expect meaningful change to occur within the existing system. The natural tendency of a bureaucracy is to defend and perpetuate itself, to resist all fundamental change. A business which does this eventually goes bankrupt…unless as Jencks points out, it can get the State to protect it from competition through a legally enforced monopoly. This, of course, is precisely what the public school system is all about. The tax system, by forcing the poor and the middle class to pay for public schools even if they desire to patronize other schools, effectively prices other schools out of the market.

This analysis of the source of the public schools' problems is not new; Milton Friedman and other libertarian economists have been making these points for years. What is new, however, is for non-libertarians like Jencks and Howard Levin to be offering the same analysis, and getting widespread publicity (and OEO grants) in the bargain. The glaring deficiencies of the present system are becoming so obvious that even Time is touting the Jencks-Friedman thesis.5

To break the public schools' monopoly and restore competition in education, Friedman has proposed,6 and Jencks endorsed, a plan called the "voucher system." Under the system, the government in a given area would compute its average pre-student school expenditure, based on the existing tax structure, and issue to parents a voucher for each of their children. The voucher would have a cash value equal to the average per-student expenditure and could be offered to a school—any school—as tuition. The effects of such a system would be revolutionary. Overnight, a real market for educational services would be created, with all the usual advantages of a market—differentiation of product and services, specialization, a range of prices, etc. Moreover, because parents would have a choice to make, they would be motivated to examine the quality of education which various schools had to offer.

Even more important than changing the economics of education, vouchers would begin to change people's attitudes toward education. Education has been a virtual state monopoly for so long that we now tend to confuse schooling with education and to sanctify the former as the only conceivable means of becoming educated. Thus, we regard 12 to 16 years of schooling as the means by which a person acquires the knowledge and abilities required for his life. In reality, there may be at least two things wrong with this assumption: the years may have been spent primarily "serving time," fulfilling society's expectations; and, the education in an advanced technological society is a need continuing throughout one's active years of life, not something one "does" once and so gets it out of the way.

By restoring the connection between payment and services received, the voucher system may go a long way toward reorienting people's way of looking at education. Viewing education as a service—as something to be selected and purchased in response to one's particular needs—would be a profound cultural change.

Another advantage would be the partial restoration to the parents of the responsibility for seeing that their offspring were adequately provided for—by them. To be sure, under the voucher system, the money would still be collected by the State as taxes; but by making payments for each child's education, by evaluating cost and quality of the services thus purchased, parents would cease to regard education as a free good, provided by "society" without cost. Ultimately a non-coercive halt to overpopulation can only come about when all parents assume full responsibility for their children. Full responsibility would automatically limit parents breeding to their ability to feed, clothe, and educate their offspring. But parents cannot make such decisions unless they are fully aware of all the costs involved in child-rearing. In treating education as a free good, we hide its cost from parents, particularly poor parents. Full-cost pricing, by means of vouchers, is therefore an important first step in making the cost of education visible to parents on a personal, yearly basis. Only after people have become accustomed to paying for their children's education can we begin to talk to them about fulfilling their responsibility to provide all funds themselves.

Some Libertarian Objections

Despite these advantages, there are, from a libertarian viewpoint, several substantive objections to the voucher idea. First, it can be argued that vouchers don't really change the fundamental injustice of public education—taxation of everyone to pay for the education of the few. If any change should be worked for, these critics contend, it should be to give tax exemption to those who opt out of the public school system in favor of private schools (or to everyone except those with children in public schools). Morally, this plan is indeed preferable to vouchers, insofar as it would be a step in the direction of reduced coercion; a significant number of people would retain the use of a nontrivial portion of their income, to spend on private schools (if parents) or however they wished. Yet overall, this solution is a poor one, for several reasons. Its single most important defect is that it is politically unfeasible, given the political context of present-day America. The liberal and educational establishments are the holders of the effective power to determine the fate of any educational reform proposal, and their primary concern is to preserve the wealth-transfer aspect of public schooling, i.e., to assure that the wealthy and childless are taxed to provide schools for the poor and prolific.

By this criterion, a tax-exemption plan is doomed to failure, whereas a voucher system has a good chance of success. Although vouchers will not change the basic concept of educational funding, they will, for the reasons outlined above, cause a radical change in the nature of educational institutions which in turn will encourage a radical change in people's attitudes. It is the latter which may eventually make possible a change in the method of funding. The imperative is to break the State's educational monopoly and the voucher plan is the only politically feasible way of doing this.

Another possible libertarian objection concerns the issue of religious schools. Under a voucher system, education vouchers would be redeemable at any school, including present-day parochial schools. To those libertarians who have spent years fighting to preserve separation of church and state, this could prove to be unsettling. Yet the voucher is not, in and of itself, a subsidy of religion. It is the parent, not the State, who supports religion by his choice to spend his voucher at a religious school. The existing public school system rather significantly violates the "free exercise of religion" clause of the First Amendment by forcing parents who currently choose religious schools also to pay for public schools. Thus, a voucher system would be more consistent with the "free exercise" clause, while not violating the "establishment" clause.7

One further libertarian objection to the voucher plan is the issue of state control. It can be argued with some justification that even if a voucher plan can be set up, the governing authorities who formerly exercised power through control of public schools will seek to impose stringent regulations on the policies and curricula of free-market schools in the name of "minimum standards." Even if such standards are initially designed only to prevent obvious charlatans and racists from opening "schools" offering little, if any, education, the natural tendency will be for those in power to exercise increasing control over the schools. The threat of cutting off voucher eligibility will act, then, as a potent persuader.

The only answer to this objection is to acknowledge that it is a very real danger that must be strongly resisted. It should be noted, though that State control over current policies and curriculum is total, so that in no case could the situation under a voucher system be worse. In fact, it should be substantially better. Countering the increasing tendency toward greater control will be the inherent dynamics of competition and technological change. As new methods of teaching are developed and proven effective, there will be increasingly powerful consumer demands, acting to oppose State control. Although State regulation must be staunchly resisted, it should not be feared; the market should be sufficient largely to offset repression.

Hopefully, the preceding arguments have established the desirability of the voucher plan as the best available means to the end of establishing a market structure in education and thereby developing a general acceptance of the concept of education as a service industry, from which one purchases desired amounts of assistance. It has been argued that these changes in market structure and attitude are a necessary prerequisite for acceptance of the idea of full parental responsibility for children and the resulting impact on our concepts of welfare and the population problem. Given, then, the desirability of the voucher plan, what must be done to foster its acceptance by those in power?

The educational establishment fears all steps in the direction of a free market, not merely because they threaten its power but also because of some serious misunderstandings of the workings of the market. Thus, one hears arguments that free-market schools are undesirable because they will increase segregation, because some will be operated by fly-by-night hucksters, or that (even if competent) such schools will widen the gap between the rich and the poor. In response to these fears, some backers of the voucher, like Jencks, are attempting to attach a multitude of restrictions to their proposals,8 most of which will hinder the market mechanism and weaken the basic virtues of the plan. Hence, there is a need for those with an understanding of the free market—the libertarian economic community—to explain in some detail why the above fears are exaggerated or beside the point and how a free market in education would actually deal with these problems.

The segregation issue is relatively easy to dispose of. In the first place, there is a good deal of evidence that an official policy of racial segregation in a voucher program, or in a participating school, would be found unconstitutional. Several southern states have attempted to set up voucher-like programs as a way of avoiding court-ordered integration and have had the plans declared unconstitutional by federal courts.9 Obviously, since a government agency would be making the funds available, the courts could easily require that the vouchers not be cashable at openly racist institutions.

On the other hand, what about de facto racial segregation—situations in which certain schools "happened" to have an enrollment of largely or exclusively one ethnic group, due to housing patterns, teaching methods, type of programs, etc.? It is at this point that the libertarian must draw the line and part company with those ideologues who insist that a child is irreparably deprived if he does not attend school in a "racially-balanced" environment. De facto segregation could indeed occur under a voucher system, but there is no evidence to indicate that this would be harmful; on the contrary, it appears more likely that the opposite would be the case.

First of all, the Coleman report data comparing Negro pupil test scores found that integration produced mixed results; more often than not there was an increase of several percent for Negro pupils in schools with a majority of whites, but in some cases the scores actually went down several percent.10 These data, of course, are based on existing public schools, in which the black or largely black schools are generally older, understaffed, overcrowded, and uncreative. It is hardly surprising that children taken from an under-equipped, isolated ghetto school and placed in a comparatively healthy environment will show a small increase in test performance. But this sort of comparison is not directly applicable to the voucher plan. Under a voucher system, well-funded, professionally-run schools, both racially balanced and unbalanced, will exist; only under these conditions will a meaningful evaluation of the effect of racial balance be possible.

The advocates of racial balance seem still to be thinking in public-school terms when they criticize potential de facto segregation in voucher schools. By definition, a public school is supposed to provide a generalized education for all of the community—to be "all things to all people." In contrast, the hallmark of a free market is specialization, the tailoring of specific schools to specific needs. A public school generally holds what amounts to an exclusive franchise (monopoly) on providing education for a given geographical area; a free market, however, could provide a number of smaller schools serving the same area, each specializing in serving the specific needs of a segment of the market. Thus, there might be a school stressing black pride and black history (which would probably end up—de facto-black), one featuring the Montessori method, another for Hassidic Jews, another in which Spanish was the primary language, serving Puerto Ricans or Mexican-Americans—all in the same general area.

Powerful support for the voucher system may come from black leaders like CORE's Roy Innis who are stressing the need for black people to gain an increased sense of self and control over their own lives, resources, and schools. Support for the goal of integration into the white world is gradually becoming the exclusive property of the liberals. The (black) head of the Justice Department's community relations service now considers integration "unrealistic and unwise," arguing instead for "a multiracial society in which we have black neighborhoods and black communities and still live as one nation.11

Last spring when a federal judge ordered the Los Angeles school system to achieve racial balance by fall 1970, the loudest protests were from Mexican-American groups who had been painstakingly building up ethnic programs in the public schools of their neighborhoods. Now this specialization must give way to the court-ordered melting pot. Much the same thing is happening to the newly-integrated black students in many parts of the South, where the public school systems are, of course, still under the control of the white majority. The old hatreds and prejudice continue, but in less direct fashion—such as separate classrooms, lunch periods, and gym classes for blacks, or large-scale demotion or transfer of black teachers.12

In short, the goal of decent, quality education for all is not being met by the monolithic, monopolistic public school approach. It demands the decentralization and specialization that are the natural characteristics of a free market. Far from being a disadvantage of vouchers, such de facto segregation as will occur will be an improvement over present conditions for those involved.

A second fear is that many free-market schools would be of inferior quality, operated by fast-buck promoters who would fleece the poor. Those who raise these fears cite the generally higher prices and lower quality of merchandise sold in the ghetto today. There are several answers to this fear. One is to point out that the rising level of consciousness and community involvement in ghetto areas—the demands for local control—will mitigate strongly against outsiders being able to get away with any sort of educational hucksterism. Increasing mass-media involvement in the plight of the poor is sure to focus more attention on new ventures like free-market schools than has been focused on shoddy stores which have existed for decades as part of the background, thereby increasing the parents' access to information about what sort of schooling the market makes available in the area.

Another answer is that comparison with the sale of other goods and services is not completely legitimate, since the existence of the vouchers will produce a guaranteed demand and ability to pay for educational services. This type of market will attract competition for the voucher dollars—something which is all too often lacking in the ghetto because of the generally lower demand for goods and services there. Nothing can guarantee the absence of hucksters from the field, but, in short, fears on this subject seem to be greatly out of proportion to the actual danger.

The final (and most important) fear raised by the educational establishment is that under a voucher system the gap between the poor and the rest of society would increase. This fear is based partly on the contention that increased de facto segregation is necessarily harmful (shown above to be false) and partly on the contention that there would be fewer sellers of education in the ghetto, offering lower quality education than elsewhere, because the costs of providing education there would be higher. This objection has led Jencks and others to propose that vouchers given to the poor should have a higher redemption value than other vouchers and that parents be forbidden to supplement vouchers with cash payments out of their own pocket.13 These restrictions are both unwise and unnecessary.

The prohibition against parents' supplementing the value of the voucher is intended to prevent the development of higher-cost schools catering to middle and upper-income families. In fact, such a rule is economically absurd in the real world; how on earth could the costs of all schools, each using different methods and technologies and dealing with different problems, be identical, such that the voucher would just cover the cost at all schools? Precisely because of the diversity and specialization that would occur, the cost of education would vary widely, with some schools vying for the unsupplemented voucher market by concentrating on efficiency and cost-saving methods and technology and others specializing in higher-cost programs for problem children, gifted children, etc. Moreover, as Harvard educational economist C.S. Benson points out, the ability to supplement, even by small sums, the amount of the voucher to meet the particular needs of one child in a family, or in response to an increase in family income, is one of the prime advantages of the voucher system.14 Today, parents are faced with the choice of "free" public school or a very expensive private school. An unrestricted voucher system opens up a continuum of cost alternatives for parents, from the basic voucher level on up in steps of what ever amount they can afford.

The other contention—that vouchers should have a higher value in the ghetto because of higher costs—also bears examination. For one thing, most educators and psychologists who have studied the problems of ghetto schooling agree that ghetto children start their school career with much less "educational investment" from their parents and environment than non-ghetto children. Furthermore, Stanford education professor Henry Levin points out, the less education a child has, the more effect each additional dollar spent on education will have, other things being equal. This being the case, "the marginal returns in future productivity from investing in the schooling of the poor will be greater than those which might be derived from spending the same amount on basic schooling for the advantaged."15 Thus, educational dollars spent in the ghetto are relatively more powerful than the same dollars spent in suburbia. At present, American public schools generally spend an average of twice as much on educating children of the rich as on children of the poor16 (partly because of higher expenditures and partly because the rich stay in school longer). Under the voucher system, since the voucher's value will equal the annual per-student educational expenditure, the level of spending in the ghetto will automatically be higher than it is now. Moreover, since the marginal returns from these (higher) expenditures will be greater in the ghetto than elsewhere, the need to have a higher-value voucher in the ghetto seems less than compelling.

A more serious objection to this contention is that it implicitly assumes ("other things being equal") that educational quality is equivalent to the dollar value of expenditure. While it is undeniable that certain technological devices are expensive, there is no inherent reason why quality education in the ghetto need be particularly expensive (even though this is "obvious" to a school board bureaucrat). Certainly, if you define a school to be a certain size brick building with many classrooms, each housing a teacher, a blackboard, and pupils sitting in straight rows—then it may cost more to build staff, and operate "schools" in some areas than in others. This, however, says nothing about whether education need cost more in the ghetto.

Even within the public school environment there are occasional innovations which demonstrate that education can be better provided without the massive costs common to most schools today. John Bremer, a 42-year-old British educator, has set up a remarkable high school in Philadelphia known as the Parkway Program17—the "school without walls." Bremer contends that "learning is not something that goes on only in special places called classrooms or in special buildings called schools; rather, it is a quality of life appropriate to any and every phase of human existence."18 Consequently, Parkway students have no school building but rely on the city's museums, libraries, shops, and businesses for "classrooms." There is some formal instruction in offices and apartments of the instructors, but most of the learning takes place out in the real world. The average student-faculty ratio is less than 8 to 1, with an average class size of 15. Parkway students are given a large amount of freedom—to stay home, to smoke, to swear, etc.—with the result that not only are they learning and maturing more rapidly than most high schoolers, but discipline problems, racial incidents, and hard-drug use are practically non-existent. Yet all of this costs only a fraction of what conventional high schools costs, because of the absence of a massive, impressive school building.

Another example is the work of George Dennison, described in his book The Lives of Children19 Dennison, funded by, but operating outside the regulations of the public system, took on the task of educating a group of "hard-core dropouts," described by the school bureaucracy as "unteachable." Dennison's ability to understand the youths and to see things from their viewpoint (much as John Holt does) enabled him to succeed in motivating them to the point of voluntary perfect attendance. Needless to say, the children learned and rejoined society. All of this was accomplished on a shoestring, with facilities no more elegant than an old rented building.

The Parkway Program and Dennison's work serve to reemphasize the point that schooling and learning are not necessarily the same thing. Largely due to the State's monopoly on both the funds and the resources for education, we have allowed "schools" in the sense of rigid, formal institutions to assume increasingly the sole responsibility for teaching in our society. The more we do this, the more we discourage self-study and self-responsibility and the more we come to depend upon credentials and diplomas rather than demonstrated competence and ability. To a certain extent, as Ivan Illich noted,20 we have made universal, obligatory schooling schooling into a new kind of secular religion, complete with temples, priesthood, and initiation rites.

The education establishment has confused process with substance in fearing that ghetto children will not be adequately educated unless more is spent on ghetto schools than is spent elsewhere. The voucher system provides a chance to look "outside the box" for better solutions to the problems of teaching and learning. As Holt, Kozol, Silberman, and Dennison have found out, our traditional, mindless, authoritarian notions of schooling do not work, particularly in the ghetto. The last thing these children need is more and more of the same—more glass and steel temples patrolled by armed policemen, more teachers with advanced degrees spending most of their time trying to maintain "order" and stifle inquiry. What we do need, instead, is diversity, decentralization, and disestablishment.

As Illich puts it:

"Two centuries ago the U.S. led the world in a movement to disestablish the monopoly of a single church. Now we need the constitutional disestablishment of the monopoly of the school.…The first article of a bill of rights for a modern, humanist society would correspond to the first amendment to the U.S. Constitution: 'The State shall make no law with respect to the establishment of education.'"21

Education vouchers will not accomplish this overnight. But, if allowed to operate unhampered by either the educational establishment or by well-meaning advocates, education vouchers can provide the vital first step in freeing education from the state. In so doing, we can begin to restore education to its rightful place as an integral part of our lives.

Robert Poole, Jr., is an MIT-trained systems engineer. His articles have appeared in THE FREEMAN and REASON. He is a contributor to a forthcoming anthology of critiques of the political system, to be published by Harper & Row, and an associate editor of REASON.


1. Silberman, Charles E., CRISIS IN THE CLASSROOM (New York: Random House, 1970).
2. Silberman, Charles E., "Murder in the Schoolroom," Part 1, THE ATLANTIC, June, 1970, p. 83.
3. Jencks, Christopher, "Is the Public School Obsolete?" THE PUBLIC INTEREST, Winter 1966, p. 22.
4. Ibid.
5. Friedman, Milton, "The Role of Government in Education," Robert A. Solo (ed.), ECONOMICS AND THE PUBLIC INTEREST, (New Brunswick, New Jersey: Rutgers University Press, 1955).
6. Friedman, Milton, CAPITALISM AND FREEDOM (Chicago: University of Chicago Press, 1962), Chapter 7.
7. For a further discussion of the constitutional issues, see Appendix A of EDUCATION VOUCHERS (Cambridge, Massachusetts: Center for the Study of Public Policy, March 1970).
8. Jencks, Christopher, "Education Vouchers," THE NEW REPUBLIC, 4 July 1970, pp. 19-21.
9. Cf. Appendix B of EDUCATION VOUCHERS, op. cit.
10. Coleman, James S. (ed.), EQUALITY OF EDUCATIONAL OPPORTUNITY (Washington, D.C.: Office of Education, U.S. Department of Health, Education, and Welfare, 1966), p. 332.
11. "Does Integration Still Matter to Blacks?" TIME, 9 March 1970, p. 14.
12. "The Bad Side of Integration," TIME, 13 July 1970, p. 32.
14. Benson, C.S., THE ECONOMICS OF PUBLIC EDUCATION (Boston: Houghton Mifflin, 1961), pp. 327-8.
15. Levin, Henry M., "The Failure of the Public School and the Free Market Remedy," THE URBAN REVIEW (reprinted by The Center for Independent Education, Wichita, Kansas), p. 9.
16. Jencks, Christopher, "Is the Public School Obsolete?" op. cit.
17. "The Parkway Experiment," TIME, 23 March 1970, p. 55.
18. Silberman, Charles E., "Murder in the Schoolroom," Part 3, THE ATLANTIC, August 1970, p. 95.
19. Dennison, George, THE LIVES OF CHILDREN (New York: Random House, 1969).
20. Illich, Ivan, "Why We Must Abolish Schooling," THE NEW YORK REVIEW OF BOOKS, 2 July 1970, pp. 9-15.
21. Ibid., p. 11.