Arizona’s school voucher law

Although the effectiveness of school vouchers is still unproven, there is plurality support across parties for these programs and the equity and legal issues seem manageable

The largest school voucher program in U.S. history recently took effect in Arizona. The new law will give parents who send their child to a private or parochial school or who homeschool their child the right to receive 90% of the state money that would normally go to their local public school. Parents are expected to receive about $7,000 for a nondisabled student.

The new law is the latest battle in a more than 20-year war between school voucher supporters and opponents in Arizona. Pro-voucher legislators have tried to pass similar legislation to the new law every year since 2017, but prior attempts had always either failed to secure enough support in the legislature or been prevented through ballot measures. Voucher opponents again tried to block the new law from taking effect, but could not obtain enough signatures in time to put the measure on this year’s ballot.

The Arizona law may still be scaled back or repealed altogether through a future ballot measure. But the school voucher debate will continue across the country, so it is worth considering whether the Arizona law and other similar voucher laws are good policy.

School voucher basics

The idea of school vouchers is simple enough. All states offer free public K-12 education to all children in their state. The state’s average cost per student can be calculated; the average cost per K-12 student is about $13,200, but that cost varies from $8,000 in Idaho and Utah to $25,100 in New York. Instead of paying the cost of a student to the student’s public school district, the state can issue the parent a voucher for that amount that the parent can apply toward the expense of sending the student to a private or parochial school.

In addition to Arizona, 14 other states and the District of Columbia have school voucher programs. However, all of the other voucher programs are limited to certain children. Eleven states have voucher programs for children with special needs. Seven states and the District of Columbia have voucher programs for families below a certain income level. Five states have voucher programs for students attending low-performing public schools. And three states have voucher programs for students who do not have a public school in their community.

In addition to voucher programs, some states have adopted other programs that help parents pay the cost of sending their children to a private or parochial school. Seventeen states have tax credit scholarship programs that incentivize people to contribute to private organizations that provide scholarships to families that function similarly to vouchers. And six states offer education savings account programs, where the state sets aside money in individual accounts for participating children, which parents can withdraw to pay any approved educational expenses, including tuition at a private or parochial school, but also including online courses, transportation, tutoring, and services for children with special needs.

The National Council of State Legislatures has a page explaining more about the history of school vouchers and the arguments for and against them. Briefly, voucher proponents argue that vouchers help individual students by allowing them to escape from low performing schools and help the public school system by forcing low performing schools to either improve or lose funding. Voucher opponents argue that vouchers help relatively few students and the students who are left behind in low performing schools will suffer even more because of reduced funding to those schools. However, school voucher proponents and opponents often are less motivated by educational concerns than by political concerns. Public schools employ public school teachers, who belong to public unions, who contribute overwhelmingly to Democratic candidates. So, Democratic elected officials have an interest in maintaining the size and influence of public schools and Republican elected officials have an interest in reducing that size and influence.

Research on the effectiveness of school vouchers

There have been many studies of the effectiveness of individual voucher programs in the United States and about a dozen reviews of those studies. Most of that research, however, focused on only four U.S. voucher programs in Cleveland, Milwaukee, New York City, and Washington, D.C.  (Some studies also analyzed voucher programs in other countries, but we do not find those studies to be helpful in considering U.S. voucher programs.) Therefore, it would be difficult to generalize from the results of studies of those four programs, even if those results were conclusive and researchers agreed on those results. However, almost all reviews of studies of U.S. voucher programs have concluded that the results are inconclusive. Of course, that does not mean that voucher programs are not effective; it means only that the existing research does not conclusively support either their effectiveness or their ineffectiveness.

One of the most recent reviews of voucher research, a 2021 article in School Effectiveness and School Improvement (an earlier version of which is available here), reviewed not only the prior studies of individual voucher programs, but also the prior reviews of voucher research. They summarized the diverse findings in ten prior reviews of the research: “Based on these reviews, school vouchers have no effect on student achievement, consistently improve achievement, or produce some mix of positive effects and no significant effects that is either encouraging or disappointing.” (citations omitted). For their part, they limited their review only to randomly controlled trials and concluded that U.S. voucher programs did not have statistically significant effects on reading scores. Their findings on math scores for U.S. programs were mixed, as one method (intention to treat) found positive effects that were barely statistically significant and another method (treatment on treated), which included one additional study, did not find statistically significant effects.

Public opinion on school vouchers

Education Next conducts an annual survey of American public opinion on various education topics, including school vouchers. They ask both about universal vouchers that would apply to all families with children in public schools and more limited vouchers that would apply only to low-income families. The table below shows the results from the 2022 poll, both for all respondents and broken down by partisan affiliation.

The table shows that about half of Americans support both universal vouchers and low-income vouchers, while about 40% oppose both types of vouchers and 10% are indifferent. Surprisingly, those percentages are almost identical for Democrats and Republicans. So, although there are significant partisan differences among elected officials on school vouchers, opinion is quite similar across parties among the general public.

The table above shows consistent plurality support for school vouchers, but not majority support. Considering that there are almost no differences in opinion across parties, one would not expect there to be significant differences across states either. Some groups have claimed to find strong public support either for or against vouchers in particular states. However, unless the poll asks specifically about school vouchers, it can be misleading to interpret the results as applying to vouchers. Polls about support for the public schools generally, about school choice generally, or about other school choice options such as tax credit scholarships or education savings accounts may produce quite different results than polls about vouchers would. Two recent state polls that did ask directly about vouchers produced similar results to the national poll discussed above. A Texas poll found that 46% supported vouchers, 43% opposed them, and 11% were unsure. And a Georgia poll found that 43% supported vouchers, 31% opposed them, and 26% were unsure.

Equity considerations

It is important to also consider how school vouchers affect different groups, especially disadvantaged groups. Voucher critics have often expressed concerns about “cream skimming” with vouchers, where the “best” students from low-performing public schools (whether that is measured by ability, grades, family income, or parental education) would take advantage of vouchers to move to private schools, leaving the public schools with less funding and a more challenging pool of students. One review of the voucher literature explicitly focused on these equity issues. They summarized results from three studies of two large U.S. voucher programs. A study of an earlier version of Milwaukee’s voucher program found that higher ability students were more likely to apply to the program and higher income students were more likely to use vouchers if they were selected. However, a study of a later version of Milwaukee’s program (after the income limit was raised and additional tuition costs were imposed on higher income families) found that students using vouchers had lower family incomes, were more likely to be Black or Hispanic, and had lower overall parental education levels (but higher college participation rates) than students remaining in the public schools. Finally, a study of Washington, D.C.’s voucher program found that voucher students were more likely to be Black or Hispanic, but that there were no significant differences in test scores, family income, or mother’s education. Therefore, although there is some evidence that voucher programs result in non-random sorting of students, there is not a consistent pattern of “cream skimming” by ability, family income, parental education, race, or ethnicity. However, the two studies of Milwaukee’s program show that program design may matter.

Legal considerations

School voucher recipients often use their vouchers to attend religious schools. Voucher opponents have often seized on that fact to argue that vouchers represent an unconstitutional state support of religion. In 2002, the U.S. Supreme Court ruled in Zelman v. Simmons-Harris that Ohio’s voucher program did not violate the First Amendment to the U.S. Constitution because it was administered neutrally, did not favor any specific religion, and did not advance religion. Therefore, state and city voucher programs should generally be permissible under the U.S. Constitution.

Many state constitutions, particularly among Western states, have “Blaine amendments,” which explicitly prohibit using state funds to support religious schools; indeed, the U.S. Congress passed a law in 1875 requring newly admitted states to include Blaine amendments in their constitutions. The U.S. Supreme Court effectively deferred to the State of Washington’s Blaine amendment in a 2004 case, Locke v. Davey, regarding a state scholarship program that did not allow students to use the scholarships to pursue degrees in theology. However, in a 2020 case, Espinoza v. Montana Department of Revenue, the Court held that a Montana rule that prohibited people from using tax credit-supported scholarships to attend religious schools violated the First Amendment, despite the fact that Montana’s Blaine amendment seemed to require that prohibition. The Court held that there was a difference between using state funds to study religion and using state funds merely to attend a religious school. Therefore, it seems likely that, even in states with Blaine amendments, state and city voucher programs may include religious schools, but further litigation is likely until the interaction between the First Amendment and state Blaine amendments is more clearly defined.

Conclusion

Overall, school vouchers seem like an area that will benefit from continued experimentation by states and cities. The existing research is inconclusive, public opinion shows plurality (but not majority) support, and equity and legal issues seem manageable to this point. Therefore, although it is too early to determine whether voucher programs like the new Arizona program are good or bad policy, it is useful to continue to try different versions of these programs, as long as data from those programs are made available to researchers to assess their effectiveness. And it is important that decisions about voucher programs are ultimately driven by educational considerations, not political considerations. Perhaps if the research reaches a clearer result in determining whether and under what circumstances voucher programs are effective, public opinion will also shift to a more convincing result.

Our grades for voucher laws generally:

Effectiveness: Incomplete (C at this point)

Equity: B-

Legal Feasibility: B

Social Acceptability: C+

Overall: C+