School Choice, The Law, And The Constitution: A Primer For Parents And Reformers

by Clint Bolick1

The Heritage Foundation
Roe Backgrounder No. 1139

September 19, 1997

Members of Congress soon will consider legislation embodying school choice programs, including the provision of school choice for poor children in the nation's capital. Meanwhile, in state legislatures across the country, serious reformers also are readying legislative proposals to give parents an opportunity to send their children to the school of their choice.

School choice—which gives parents control over where the public dollars earmarked for their children's education will be spent—is the most promising education reform in the United States today. Among reform proposals, it alone transfers power over basic education decisions from bureaucrats to parents and provides poor children in the worst school systems an immediate chance to receive a high-quality education. It also creates a strong incentive for public school systems to adopt long-overdue reforms. For these reasons, the school choice movement has grown to encompass support from conservatives and libertarians, centrist Democrats and Republicans, and leaders of minority communities.

But school choice threatens powerful entrenched interests that oppose it with every means and resource at their disposal. As school choice victories multiply in the state legislatures, opponents of choice are forced to resort to the judicial arena. Hence, inevitably, every meaningful school choice victory involves a two-part process: the legislature or ballot box, followed by the courtroom. Most people, understandably, dread litigation. But there is a maxim for measuring the impact of empowerment reforms: Reformers can be sure that they have accomplished something important only if left-wing special-interest groups challenge it in court.

Until a definitive ruling by the U.S. Supreme Court is handed down on the constitutionality of school choice—which could occur within the next two years—it is impossible to predict with certainty how school choice plans will fare in court. Even if a strong decision upholding school choice is handed down, antichoice forces soon will develop new strategies to tie up the school choice programs in litigation. School choice supporters should not view litigation as a reason for inaction, but rather as an inevitable cost of success. They should move forward with the most aggressive efforts possible to expand educational opportunities, building legal defenses into their programs as best they can.

This study examines the legal landscape for reformers contemplating school choice, specifically:

(1) The range of school choice options;

(2) A distillation, in plain English to the greatest extent possible, of the applicable U.S. Supreme Court precedents implicating school choice;

(3) Rules of thumb for drafting school choice proposals; and

(4) State constitutional hazards.


The possibilities for school choice programs are bounded only by the imagination of those who are committed to expanding educational opportunities—and of course by political realities. Beyond the rules of thumb outlined in the preceding section and other limitations in state constitutions (discussed in the next section), the contours of school choice programs are purely questions of policy. No "model" or one-size-fits-all school choice program exists, nor should one exist.

Still, current operational school choice programs that encompass private school options tend to fit into three categories: tax deductions and credits, targeted scholarships, and child-centered education funding. A brief discussion of some existing choice programs may prove useful.2

Tax Deductions and Credits

As described in the context of the Mueller decision, Minnesota provides state income tax deductions for expenses incurred in private or public schools, including private school tuition. The deductions were increased in 1997, along with refundable tax credits for nontuition expenses incurred by low-income families. Arizona in 1997 enacted a tax credit3 for contributions to scholarship funds.4

Some choice advocates prefer tax deductions and credits because no funds are transmitted from the state to private schools. That degree of indirectness may increase the odds of constitutionality and reduce the likelihood of government regulation. The prime objection is that they do not provide immediate benefits for economically disadvantaged families. This objection may be overcome by providing refundable tax credits or, as in Arizona, making the tax benefits available for contributions to scholarship programs.

Targeted Scholarships

The greatest need for school choice programs exists for economically disadvantaged children mired in large urban public school systems. Milwaukee and Cleveland, of course, have the first two operational choice programs for low-income youngsters. The programs are similar, but have noteworthy differences. Milwaukee allows up to 15 percent of students enrolled in the public schools (children in lower grades in private schools are also eligible) who are economically disadvantaged to use their state share of public funds (roughly $3,800 per pupil) as full payment of tuition in participating private schools in any grades. Students are selected by the schools through a lottery. Cleveland also has a lottery, with a preference for low-income children, for scholarships worth 90 percent of tuition (up to $2,500) at participating private schools. No more than 50 percent of the children may have attended private schools previously. The program started in grades K-3 and expands this year to include fourth graders. The legislature appropriated funds for approximately 3,000 participants for the coming school year. Preliminary results from both programs are very encouraging. Litigation in both states is pending, however.5

Variations of targeted scholarships exist. Maine and Vermont provide tuition subsidies for children in rural school districts that do not have their own public high schools. Students receive tuition—up to the average amount of public per-pupil funding—to attend public high schools in adjacent school districts or private schools. At present, neither state provides tuition for religious schools, but both programs are in litigation because of the exclusion of religious schools.6 Some school districts, including Houston, have made private schools available to students due to overcrowded public schools. Some states, including Wisconsin, allow opt-outs for at-risk students to attend private schools. And, of course, the federal Individuals with Disabilities Education Act allows students to attend private schools at public expense when the public schools fail to provide an "appropriate" education.

Child-Centered Education Funding

A more comprehensive approach to education reform is to conjoin public and private school choice with the education funding system, a proposal championed most prominently by Arizona Superintendent Lisa Graham Keegan. Instead of exclusively funding schools or school districts, the state (or federal government in the context of existing funding programs) would provide an equal amount of funds that follow the educational choices of each student.

Governor Froilan Tenorio of the Commonwealth of the Northern Mariana Islands (CNMI) proposed the first such system in 1997. The program would transform a portion of the state's education budget into child-centered funding that would follow each child in the CNMI to the public or private school of the family's choice. Significantly, funds going to public schools in the program would be placed under the control of the particular school. In addition to expanding choices, the new funding system would foster decentralization, autonomy, and competition in the public schools. It also would create a system that is entirely neutral, as between religious and secular educational options.


The strongest critics of school choice argue that the moment a dollar of public funds crosses a religious school threshold, it violates the First Amendment. Of course that cannot be the case, for such educational benefits as Pell Grants, the G.I. Bill, and federal daycare vouchers all can be used in religiously affiliated entities. School choice works the same way: Parents choose where to direct their children's education funds. A careful review of applicable precedents demonstrates that well-designed school choice programs accord fully with the principles of the First Amendment.

From the text of the First Amendment, which states that "Congress shall make no law respecting an establishment of religion," it is difficult for normal human beings to fathom how giving parents control over educational dollars possibly could present a constitutional problem. But jurisprudence in this area is circuitous, complex, and confusing. Fortunately, recent U.S. Supreme Court precedents have produced some clarity and common sense.

The story starts nearly a quarter-century ago. In its 1973 Nyquist decision,7 the U.S. Supreme Court sounded the death knell for "parochaid" efforts by some state governments to subsidize religious schools both on equity grounds and as a way to absorb the overflow of baby-boom children. The Court struck down direct grants for private schools, tuition reimbursements, and tax deductions for private school families. The Court emphasized that the First Amendment "compels the State to pursue a course of `neutrality' toward religion."8 By making benefits available exclusively to private schools and families who patronize them, the Court concluded, the state created an incentive to choose private and religious schools, and the aid therefore had the impermissible "primary effect" of advancing religion. But the Court expressly left open the question of a "case involving some sort of public assistance (for example, scholarships) made available generally without regard to the sectarian-nonsectarian, or public-nonpublic nature of the institution benefited."9

School choice opponents have virtually no precedents other than Nyquist available to them. And, as the Supreme Court observed in its 1997 Agostini v. Felton decision, establishment clause jurisprudence has "significantly changed" over the past decade.10 Specifically, what has changed "is our understanding of the criteria used to assess whether aid to religion has an impermissible effect."11

Indeed, since Nyquist, the Supreme Court repeatedly has upheld government aid programs that include religious schools and activities among the range of options:

Opponents of school choice have pointed to two features of the Title I compensatory education program acknowledged by the Supreme Court in Agostini: that the teachers are not allowed to engage in religious instruction, and that no public funds are transmitted from the government to religious schools. School choice programs, by contrast, provide public funds to pervasively sectarian schools for their unrestricted use.

But Agostini presented the difficult case of public employees actually teaching in religious schools. Where funds are placed at the disposal of third parties—as in Mueller and Witters—the connection between the state and religion is more attenuated. The Supreme Court in Agostini made clear the applicable principle: There is no impermissible effect "where the aid is allocated on the basis of neutral, secular criteria that neither favor nor disfavor religion, and is made available to both religious and secular beneficiaries on a nondiscriminatory basis."21

The Supreme Court, using the First Amendment, limits and sometimes forbids direct subsidies to religious entities, as well as programs that create a financial incentive to patronize religious schools. But the Court has made plain time and again that "programs that are wholly neutral in offering educational assistance to a class defined without reference to religion do not violate" the First Amendment, "because any aid to religion results from the private choices of individual beneficiaries."22

All credible contemporary school choice proposals readily satisfy these criteria. They do not propose subsidizing religious schools, but merely include such schools within the range of educational options made available to a neutrally defined category of beneficiaries (usually economically disadvantaged families). No public funds are transmitted to religious schools except by the independent decisions of third parties. As the U.S. Supreme Court repeatedly has affirmed, such "`attenuated financial benefit[s], ultimately controlled by the private choices of individual[s]'...are simply not within the contemplation of the Establishment Clause's broad prohibition."23

In other words, the First Amendment does not stand as a bar to a program whose primary effect is not to advance religion, but to expand educational opportunities to children who desperately need them.


The applicable First Amendment precedents yield three main rules of thumb for designing school choice programs:

Rule #1: No public funds should be transmitted to religious schools except at the direction of third parties.

By giving families control over education funds, any benefit to religious schools is indirect. Some policymakers agonize over the mechanism employed to transmit funds to the schools (for example, checks made out to the parents). That exercise tends to elevate form over substance and does not seem to influence the outcome in particular cases. The relevant question is who should determine where the money will go.

Nor has the Supreme Court distinguished between such mechanisms as tax deductions (Mueller) and grants or scholarships (Witters). Conceivably, policymakers can purchase additional constitutional insurance by making the assistance more indirect, as in tax deductions or refundable tax credits versus grants or scholarships. But the tradeoff is that such programs may not work well for economically disadvantaged families who may need direct assistance.

Within a scholarship-type program, one option is to establish individual trust accounts for students, who would own the funds and may direct them for specified educational expenses. Although such a mechanism probably is unnecessary, it does mean that no government funds are paid to religious schools because the funds belong to the children.

The rule against direct funding does not foreclose contracting out educational services to religious schools. If the government contracts with a particular religious school, it probably should limit the use of funds to secular services (as with the Title I program in Agostini). But if it merely includes religious schools among a range of options—allowing religiously affiliated charter schools, for example—then the use of funds can be less restricted (as with the religious publication in Rosenberger). The constitutional concern diminishes to the degree that independent decisions guide the direction of public funds.

Rule #2: The program should extend benefits to a neutrally defined class of beneficiaries and create no financial incentive to choose private or religious schools.

The choice program should define its beneficiaries in neutral terms—that is, not as private schools or private school students (Nyquist), but in terms of objective criteria (such as income, residency, at-risk, all students, or some other broad class of eligible students). Agostini approved aid that was dispensed on the basis of "neutral, secular criteria that neither favor nor disfavor religion, and is made available to both religious and secular beneficiaries on a nondiscriminatory basis." 24 The broader the range of educational options (including, for example, public as well as private school choice), the more likely that a court will find the program neutral.

If the program is neutral and provides independent choices, the program's constitutionality will not depend upon the extent to which individuals choose religious schools. As the Supreme Court emphasized in Agostini, "Nor are we willing to conclude that the constitutionality of an aid program depends on the number of sectarian school students who happen to receive the otherwise neutral aid."25 In Mueller, the tax deduction was available for educational expenses in private or public schools. Not surprisingly, the vast majority of benefits were claimed by religious school families.

The program also should not create a financial incentive to choose private or religious schools. Based on the fact that many private schools cost less than public schools, some choice advocates have proposed allowing students who attend private schools to save for college purposes the difference between the public funds and the amount of private school tuition. Although this may be a legitimate policy objective, the courts could perceive such a provision as a financial incentive to choose private or religious schools.

Rule #3: The program should not impose regulations beyond those necessary to ensure that the government's educational objectives are accomplished.

The First Amendment forbids the use of public funds in religious schools if they are accompanied by extensive or intrusive regulation of religious schools.26 The government legitimately may apply objective standards (such as nondiscrimination or a core curriculum), and states in fact already impose such requirements on most private schools. Of course, where public funds are used, the government permissibly may ensure financial accountability and impose other conditions for schools that choose to participate. But it may not interfere with the school's mission or governance, or with the school's day-to-day operations. If it does, the result is "excessive entanglement" between the state and religion, which the First Amendment forbids. This constitutional restraint—along with the freedom of schools to choose not to participate—provides a strong assurance against excessive regulation of religious schools in choice programs.


In addition to federal constitutional issues, reformers promoting school choice usually will encounter state constitutional considerations as well. Because every state constitution is different—and each state's constitutional jurisprudence is different even when the language is the same—no substitute exists for an in-depth review of applicable state constitutional provisions in the context of state constitutional provisions. The inquiry should proceed in two directions.

State Religious Establishment Provisions

Most state constitutions contain religious establishment provisions that are more specific than the First Amendment, and many speak specifically to state funding of religious schools. But more specific does not necessarily mean more restrictive. For example, some state constitutions prohibit the use of public funds "for the benefit" of religious schools. Although more specific than the First Amendment, it is clear that the more general encompasses the more specific: The First Amendment, too, prohibits public funds "for the benefit" of religious schools, but not for the benefit of schoolchildren.

The real determinant, of course, is how state courts have interpreted the provisions. Yet, even then the first impression can be misleading. Most state cases date back to the 1960s and 1970s, when both federal and state courts were striking down efforts to provide assistance to religious school students. Accordingly, most state precedents appear harmful for school choice prospects. The threshold question, however, is whether the state courts have interpreted the state constitutional precedents in harmony with the First Amendment. If so, regardless of how cases were decided 25 years ago, the state constitutional interpretation is likely to follow the U.S. Supreme Court precedents as noted.

If state constitutional precedents adopt standards different from the First Amendment, school choice advocates should examine those precedents closely and conform their programs as best they can. For example, a tax-deduction program might be permissible in a particular state while a scholarship program might not. Of course, school choice advocates always can attempt to change jurisprudence, and a choice program crafted to the needs of disadvantaged children can give them the opportunity to do so (particularly in light of state constitutional provisions that provide a right to an education).

In the few states in which state constitutional provisions seem an insuperable barrier, constitutional amendments creating exceptions for certain types of educational programs probably are necessary.

Other Provisions

Opponents of choice will bring out every possible weapon in the legal arsenal, so school choice advocates must scour the landscape to anticipate every possible attack. The legal challenges to the Milwaukee and Cleveland programs have presented two additional state constitutional claims that may arise in other states as well.

Typically, state constitutions contain some sort of educational provisions: In Wisconsin, it is a guarantee of a "uniform" education in district schools. The Milwaukee and Cleveland lawsuits have alleged that their state constitutional provisions implicitly limit the use of public education funds to public schools. (Indeed, some states actually have explicit provisions relating to the use of public school funds; school choice programs in such states may have to draw from different budget sources.) So far, courts in the two states have ruled that the education provisions set the state's minimum obligations, but that states can go beyond those obligations (for example, school choice).

Also, both the Milwaukee and Cleveland programs were enacted as part of the state budget. Many state constitutions contain provisions requiring separate bills for "local" legislation. In such circumstances, choice programs either should (1) apply to categories of cities, rather than to a single specific location (like urban centers having more than a certain population size with specified educational problems), or (2) demonstrate statewide ramifications, such as educational experimentation. Constitutional provisions like the Wisconsin "private or local bill" clause can have serious ramifications. For example, when a state trial court invalidated the expansion of the Milwaukee Parental Choice Program in January 1997, its finding that the legislation was an impermissible local bill rendered invalid even the nonsectarian portions of expansion.


To put it mildly, school choice programs are not without legal risks. School choice advocates should do all they can to make their programs bulletproof; and even then, they are likely to have to endure two or more years of litigation and uncertainty.

But the potential rewards are breathtaking. No matter how many briefs reformers have to write, no matter how many arcane legal issues they have to research, no matter how many hours they have to spend listening to lawyers from teachers unions pontificating about the horrors of school choice—all of it and more are worth it to walk the hallways of the participating schools and look at the students' faces. No other reform promises to have such a constructive impact on children's lives or fulfill our country's sacred promise of equal educational opportunities. Revolutionary War hero Thomas Paine made a prophetic observation: "Tyranny, like hell, is not easily conquered; yet we have this consolation with us, the harder the conflict, the more glorious the triumph."

This is good advice. Members of Congress, state legislators, and parents dedicated to giving their children the best of educational opportunities should not forget it.

Copyright © The Heritage Foundation. Used by permission.


1 Clint Bolick is vice president and litigation director at the Institute for Justice in Washington, D.C.

2 The Heritage Foundation and the Center for Education Reform publish comprehensive guides to school choice programs in the states, including public school choice and magnet schools.

3 A tax credit is subtracted from the amount of tax owed. A tax deduction is subtracted from the amount of taxable income.

4 A legal challenge from the Arizona Education Association is anticipated. The Institute for Justice will represent prospective scholarship beneficiaries in defense of the program.

5 The Institute for Justice represents parents and children in both cities in defense of the programs' constitutionality.

6 The Institute for Justice represents the Town of Chittenden, Vermont. The town voted to include religious schools among the range of options, but was forbidden by the state to do so on the grounds that such action would violate the First Amendment. In July 1997, the institute filed a lawsuit challenging the constitutionality of Maine's exclusion of religious schools.

7 Committee for Public Education & Religious Liberty v. Nyquist, 413 U.S. 756 (1973).

8 Nyquist, pp. 755-756.

9 Nyquist, p. 782, n. 38.

10 Agostini v. Felton, 65 U.S.L.W. 4524, 4533 (U.S. June 24, 1997).

11 Agostini, p. 4529.

12 Mueller v. Allen, 463 U.S. 388, 399 (1983).

13 Mueller, p. 399.

14 Witters v. Washington Department of Services for the Blind, 474 U.S. 481 (1986).

15 Witters, p. 488.

16 Zobrest v. Catalina Foothills School Dist., 509 U.S. 1 (1993).

17 Zobrest, p. 13.

18 Rosenberger v. Rector and Visitors of University of Virginia, 115 S. Ct. 2510 (1995).

19 Rosenberger, p. 2521.

20 The Court overturned Aguilar v. Felton, 473 U.S. 402 (1985), and partly overturned School Dist. of Grand Rapids v. Ball, 473 U.S. 373 (1985).

21 Agostini, p. 4531.

22 Witters, pp. 490-491 (Powell, J., concurring). Justice Powell earlier authored the Court's decision in Nyquist.

23 Rosenberger, pp. 2541-2542 (Souter, J., dissenting).

24 Agostini, p. 4531.

25 Ibid.

26 See discussion of "excessive entanglement" in Agostini, p. 4532.