Parental-choice debates typically focus on whether private schools should receive public funds at all. This paper focuses on a question that inevitably follows when schools do receive them—the question of accountability. That is, what regulatory conditions ought to attend private schools’ receipt of public funds? This is an enormously complicated question. On one level, it is entirely reasonable to condition private schools’ participation in parental-choice programs on some government oversight. When the government acquires goods or services from the private sector, surely it is entitled to ensure that it receives what it is paying for. On the other hand, this simple logic masks enormous complexities: For what should private schools be held accountable? How should regulators determine whether their performance satisfies accountability standards? What ought to be the consequences of nonperformance? To what extent do—or might— accountability regulations threaten the autonomy and religious liberty of participating schools?
This report tackles these questions, focusing on academic accountability: How should regulators hold private schools that receive public funds accountable for their students’ academic performance? After reviewing current accountability regulations in private-school-choice programs, the paper discusses challenges that are presented when parental choice and accountability policies intersect. These challenges include opponents’ often disingenuous deployment of accountability demands as a weapon to undermine parental choice; the problem of selection bias; the risk that accountability regulations deter some schools from participating; and the limits of the standardized tests on which most accountability regulations rely.
I recommend two guiding principles for accountability regulations in the parental-choice context. First, regulations should aim to give parents the information needed to make wise choices, thereby enabling them to hold schools accountable through enrollment decisions. This means that information about school quality must be transparent, easy to interpret, and reflective of the school-quality criteria that matter to parents. Second, accountability regulations ought to advance the goal of providing parents with access to more and better schools. In most cases, this goal is better advanced by convincing good schools to participate in choice programs than by excluding bad ones from doing so. This means that regulators should take care to ensure that regulations are not unduly burdensome and do not threaten the autonomy of participating schools. It also suggests that increasing the level of funding in choice programs may itself serve as an accountability function.
Over the past 18 months, the disruption wrought by the Covid-19 pandemic to K–12 education in the U.S. stimulated a flurry of legislative and executive actions favoring parental choice. Already, 2021 has been the single most successful year in the history of the private-school-choice movement.  Prompted by frustration over public schools’ insistence on continuing remote instruction—and private schools’ willingness to adapt to enable a safe return to classrooms— state legislators introduced dozens of parental-choice proposals.
To date, 21 states have voted to create, expand, or improve school-choice programs, including three new programs in three states that previously did not have any private-school-choice options—Kentucky, Missouri, and West Virginia.  In 2021 alone, Arkansas, Florida, Indiana, Kentucky, Missouri, New Hampshire, and West Virginia have enacted new parental-choice programs, while Arizona, Georgia, Florida, Indiana, Iowa, Kansas, Maryland, Montana, Nevada, Pennsylvania, Oklahoma, and South Dakota expanded existing programs. Eligibility for two of these programs—the new education savings account program in West Virginia and the expanded voucher program in Indiana—has grown to nearly 80% of school-aged children. Debates continue over additional school-choice proposals in other states. 
These parental-choice victories are worthy of celebration. They promise to expand the educational opportunities available to the low- and moderate-income children who frequently have not been well served (historically and in the past year) by district schools and to extend needed public resources to the private and faith-based schools that have long served the same children well. They are also further evidence that private-school choice—long the third rail of educationpolicy debates—has moved in from the margins to the mainstream. Indeed, in the three decades since Wisconsin enacted the nation’s first school voucher program in 1990, parental choice has exploded onto the American educational scene. With the addition of three new states to the school-choice roster this year, 31 states, the District of Columbia, and Puerto Rico now have at least one private-school-choice program, and more states likely will join the private-schoolchoice ranks in the near term. 
Parental-choice debates have historically centered on the question of whether private schools should receive public funds at all. This paper focuses on what inevitably follows when schools do receive them—the question of accountability. That is, what regulatory conditions ought to attend private schools’ receipt of public funds? The question of accountability and parental choice is immensely complex. On one level, it is entirely reasonable to condition private schools’ participation in parental-choice programs on some level of government oversight. Most people of good faith agree that, when the government acquires goods or services from the private sector, it is entitled to ensure that it receives what it is paying for. On the other hand, this simple logic masks enormous complexities: For what should private schools be held accountable? How should regulators determine whether their performance satisfies accountability standards? What ought to be the consequences of nonperformance?
This paper primarily addresses questions about academic accountability: How should regulators hold private schools that receive public funds accountable for their students’ academic performance?  But accountability debates extend beyond the realm of academics, encompassing demands that private schools be subject to regulations that threaten their autonomy and religious-liberty interests, as well as imposing implementation costs that may exceed the financial benefits of their participation in a school-choice program. 
After briefly outlining the landscape of parental-choice policies in the U.S. today, this paper tackles the complex issue of accountability. The discussion describes accountability policies in existing choice programs; examines some challenges presented when parental choice and accountability policies intersect; and recommends principles to guide accountability regulations.
These recommendations reflect the belief that the goal of accountability should be to empower parents to make good choices for their children. Regulations promoting this goal should satisfy two criteria. First, they should be designed to provide parents the information that they need to make wise choices, which means that information about school quality must be readily available, transparent, easy to interpret, and reflective of the school-quality criteria that matter most to parents. This information will enable them to serve an accountability function by making informed decisions in school-choice markets.
Second, accountability regulations ought to advance the goal of providing parents with access to more and better schools. In most cases, this goal is better advanced by persuading good schools to participate in choice programs than by excluding bad ones from doing so. The exclusion of persistently failing schools from choice markets is perhaps a necessary element of an accountability regime, but it should not be the primary element. The long history of academic accountability efforts in the U.S. demonstrates that punitive government regulations rarely drive school improvement and that closing bad schools works as a reform only when better options are available. Moreover, and importantly, regulations that unduly constrain the autonomy of schools, as well as those that raise religious-liberty concerns, are particularly problematic because they may deter the entry of strong schools that need parental-choice resources less.
Parental choice has gained a firm foothold in the educational landscape. As it continues to expand, questions and debates about holding participating schools accountable will only intensify. Whether, when, and how the law ought to regulate the quality of the schools participating in parental-choice programs are important and vexing questions for the law of education. As the troubled experience with efforts to hold public schools accountable for student performance demonstrate, a host of public-choice, selection-bias, and institutional-design challenges haunt efforts to use the law to regulate school quality. Arguably, each of these challenges is compounded in the parental-choice context for interrelated reasons discussed below.
The Parental-Choice Landscape
Today, many parents can choose from a range of schooling for their children. Many (if not most) school districts permit students to attend a public school other than the one geographically assigned to them, including magnet schools. Forty-four states and the District of Columbia authorize charter schools, which are publicly funded but privately operated schools that are designated as “public schools” in all state and federal laws.  And, as discussed previously, 31 states, the District of Columbia, and Puerto Rico have at least one publicly funded privateschool- choice program. With the adoption of new programs this year, there are now nearly 70 private-school-choice programs in the U.S. 
While many commentators lump all private-school-choice programs into a single category (often derisively referred to as “vouchers”), there are a variety of school-choice mechanisms: (1) voucher or scholarship programs, through which students receive publicly funded scholarships to attend private schools; (2) scholarship tax-credit programs, which employ state tax policy to incentivize private donations to nonprofit scholarship-granting organizations; (3) education savings accounts (ESA), which give parents the option to use the education dollars allocated for their children on a variety of educational options, including tuition and fees, textbooks, and tutoring; and (4) individual tuition tax-credit programs, which provide tax benefits for private-school tuition. Many states have multiple programs employing different private-school-choice devices. In the 2019–20 school year, 331,000 children participated in scholarship tax-credit programs, 220,000 in voucher programs, and 23,000 in ESA programs. The largest school-choice programs—in terms of the total number of participating students—are voucher programs in Indiana, Ohio, and Wisconsin and scholarship tax-credit programs in Florida, Arizona, and Pennsylvania.  The number of students participating in choice programs likely will increase dramatically as new and expanded programs become available during the next school year.
Accountability and Parental Choice
While debates about accountability in K–12 schools are hardly new, private-school choice itself was conceived of, and championed, as the ultimate academic accountability device. At least since Milton Friedman’s seminal 1955 essay “The Role of Government in Education,” proponents of choice have argued that empowering parents to select their children’s schools will improve the academic achievement of students in both the public and nonpublic educational sectors. Friedman was not the first to argue that the government should fund private schools, but he was the first to articulate the device of publicly funded scholarships or “vouchers” redeemable at both public and private schools. His case for parental choice was, at its core, about accountability. “The interjections of competition,” Friedman argued, would “stimulate the development” of alternatives to public schools, “promote a healthy variety of schools,” and “introduce flexibility into school systems.”  Several decades later, John Chubb and Terry Moe amplified these arguments in their influential book Politics, Markets and America’s Schools. Chubb and Moe argued that interjecting choice and competition into the monopolistic American education system would prove so transformational as to obviate the need for all other education reforms, including accountability regulations. They wrote that “if choice is to work to greatest advantage, it must be adopted without these other reforms, since the latter are predicated on democratic control and are implemented by bureaucratic means. The whole point of a thoroughgoing system of choice is to free the schools from these disabling constraints.” 
In contrast to these early expectations, however, demands to use the law—in addition to parents’ choices—to hold chosen schools accountable across a variety of variables feature prominently in parental-choice debates today. There are at least two reasons. First, many accountability debates are not really about holding schools “accountable.” Rather, they represent opponents’ thinly veiled obstructionist efforts to kill choice proposals before they are enacted and to subject private schools to government control when such efforts fail. Second, the animating justification for parental-choice policies has shifted over the past 25 years. Somewhat ironically—since subjecting district schools to competition does appear to improve their academic performance—parentalchoice policies are no longer justified by the need to subject government-operated schools to competition but rather by the imperative of giving disadvantaged students alternatives to failing public schools. As Howard Fuller, the architect of the nation’s first private-school-choice program, has observed, parental choice is today “more of a rescue mission than a fight for broad societal change.” 
This shift has significant implications for debates about accountability. If the goal of parental choice is to “rescue” children, it becomes natural to ask: Are these programs working? Do participants learn more or better? Are the participating private schools actually superior to the public-school alternatives available to participating students? Unfortunately, parental choice alone has proved ineffective in weeding poorly performing schools out of private-schoolchoice programs. And in some cases, hoped-for academic gains have failed to materialize among program participants. Most studies of private-school-choice programs find modest positive effects on academic performance over time, as well as more significant longer-term effects on noncognitive variables, including high school graduation rates, college matriculation and persistence, and a reduced likelihood of involvement in the criminal-justice system.  Some studies have found that participants actually lose ground, at least in the short term, when they transfer from public to private schools (at least as measured on standardized tests), prompting commentators to announce that school choice does not “work.” 
Accountability in Current Choice Programs
Historically, private schools have not been subject to the accountability regulations governing district and charter schools. Private-school regulations vary across the states, and almost all are minimal. For example, many states require private schools to register with state education officials, secure approval to operate, or both. No state requires accreditation to operate, but a handful require a subset of schools (for example, high schools) to be accredited, and some permit schools to satisfy other regulatory requirements.
Similarly, most states exempt private schools from state curricular mandates, although 40 states do regulate curricula in some way. Some states mandate that private schools teach certain core subjects; others mandate that they adopt a curriculum approved by an association of private schools. Others are silent on the matter of curriculum. A handful of states require that a private school’s curriculum roughly approximate the public-school curriculum, although these requirements typically are under- or unenforced. (New York’s recent effort to enforce against Orthodox yeshivas the legal requirement that private schools’ curricula be “substantially equivalent” to public schools’ sparked significant controversy.)  Only two states, North Dakota and Nevada, require private schools to employ certified teachers, although just more than half the states require certification for certain categories of employees (for example, school leaders) and certain types of schools (for example, secular—but not faith-based—schools). 
Additionally, federal law regulates private schools in a number of ways: federal tax regulations prohibit racial discrimination by all tax-exempt nonprofit entities (including private schools).  Schools receiving federal funds (for example, those participating in the federal free and reducedprice lunch program) are prohibited from discriminating against students on the basis of race or sex (although an exemption exists for single-sex schools) and are required to make certain accommodations for students with disabilities under the federal Rehabilitation Act. Secular, but not religious, private schools are subject to the Americans with Disabilities Act as well. Subject to the nondiscrimination provisions described above, private schools otherwise enjoy substantial freedom to set their own admissions criteria, including academic criteria and—in the case of faith-based schools—preferences for coreligionists.  Private schools also are subject to state and federal employment and nondiscrimination regulations, although these laws generally require faith-based schools to consider religion in certain hiring decisions. Moreover, the Supreme Court has held that the First Amendment precludes government regulation or judicial scrutiny of employment decisions concerning “ministerial” employees—which include all teachers whose duties include religious instruction or advancement of the school’s religious mission, even if the teacher has neither a ministerial title nor formal religious training. 
Most private-school-choice programs impose additional regulations on participating schools, especially in voucher programs. (Scholarship tax-credit programs tend to be lightly regulated.) All private-school-choice programs regulate the quality of schools by mandating certain predictive “inputs.” For example, many limit participation to accredited schools and establish minimum qualification requirements for teachers—usually a bachelor’s degree or substantial teaching experience. A handful of programs establish basic curricular minimums beyond those required of nonparticipating private schools, such as the teaching of civics and character education.  Some programs mandate that schools communicate with parents about students’ progress. Some require participating schools to administer standardized tests and report the results to state regulators, although these results are not necessarily disclosed to the public. A handful of voucher programs require participating students to take the same standardized assessments as district and charter schools, although only Indiana requires participating schools to administer state-mandated assessments to students who do not receive vouchers. In Indiana and Louisiana, persistent underperformance on these exams can lead to the school’s exclusion from participation in the program.
Several voucher programs (but no tax-credit scholarship programs) regulate school admissions policies. Several require the random selection of scholarship recipients (but not all students). The voucher law in Washington, DC, prohibits schools (including faith-based schools) from considering religion in admissions, and Maryland’s voucher program prohibits discrimination on the basis of LGBTQ status in admissions and employment. (This requirement is being challenged, thus far unsuccessfully, on Free Exercise grounds.)  The Milwaukee Parental Choice Program, the nation’s oldest voucher program, requires schools to permit participants to opt out of religious exercise. Additionally, and not surprisingly, all programs regulate various aspects of school finances. Some programs limit scholarship amounts to the school’s tuition level, and a handful of voucher programs prohibit schools from charging tuition above the scholarship award.
The Challenges of Accountability in the Private-School-Choice Context
Significant challenges haunt accountability efforts in the context of private-school choice. I discuss four of these challenges here. First, opponents of parental choice frequently use the mantra of “accountability” to kill choice proposals and, when such efforts fail, to subject private schools to government control. Second, the twin goals of parental-choice programs—to increase the number and quality of options available to disadvantaged students—come into tension in the accountability context. Third, selection-bias difficulties pervade efforts to compare the performance of students participating in parental-choice programs with their district-school peers. Fourth, most accountability metrics rely heavily on standardized test scores, which fail to capture important noncognitive indicators of school quality, including many that parents value more than test scores.
1. The Political Economy of Accountability. The most significant challenge to designing optimal accountability policies for chosen schools is politics. Many private schools—and those people who advocate for parental choice on their behalf—mistakenly oppose even minimal accountability requirements. Proponents of private-school choice would do well to acknowledge that it is not unreasonable to condition the participation in private-school-choice programs on some greater level of oversight and to work with policymakers to craft accountability rules that attend to the pluralistic landscape of American private education. Subjecting participating private schools to the full panoply of regulations governing district and charter schools may well threaten the autonomy and identity of private schools by forcing them to conform operations to government norms, including monolithic curricular mandates. On the other hand, requirements that participating private schools administer standardized tests and report their results, both to regulators and parents, are entirely sensible.
Resistance to even reasonable regulation opens the door to bad-faith regulatory demands by parental-choice opponents. The reality is that many demands for accountability are not made in good faith. Assertions that parental choice “doesn’t work” pervade education-policy debates. Any study of a parental-choice program suggesting tepid or negative results presents an opportunity to bludgeon the very idea of parental choice, and examples of individual school failures inevitably are held up as an exemplar of the folly of parental choice. Relatedly, publicschool proponents frequently argue that school choice should be suppressed because these programs divert students and resources away from public schools. Opponents of private-school choice often seek to use the mantra of accountability to impose a chokehold on private schools, subjecting them not just to government oversight but government control.
Tellingly, opponents of school choice often advocate more stringent and more punitive accountability policies for private schools than those governing district schools. Mandatory closure and exclusion policies are cases in point. Public-school proponents vehemently object to closing public schools for any reason, especially academic failure. For example, the National Education Association takes the position that closing public schools—even persistently failing ones—always harms students.  In contrast, choice opponents routinely demand the automatic closure of failing charter schools and the exclusion of failing private schools from parental-choice programs. 
2. More and Better Schools. The second challenge of accountability in the parental-choice context is related to the first. Accountability policies focus on the quality variable in the parental-choice equation: they seek to ensure that the choices available to participating students are good ones, for example, by excluding persistently poor performers from participating. Unfortunately, this goal can come into tension with the goal of increasing the number of options available. In calling for more comprehensive accountability regulations in private-school-choice programs, parental-choice proponents Bruno Manno and Chester Finn have observed that there may be a “painful trade-off” between quality and quantity in parental-choice programs, which is endemic to the “vexing reality . . . that market forces alone can’t reliably generate academic effectiveness.” 
Within the school-choice movement, accountability proponents like Manno and Finn generally assume that accountability regulations’ primary effect is to suppress the participation of the poorly performing schools in school-choice programs. However, this is not necessarily the case. Stringent accountability policies may close or exclude schools that are struggling academically but that also are better than the community’s district schools. Limiting the participation of struggling private schools, in this case, may force some students into worse public ones. Moreover, accountability policies advance the goal of more and better schools only if they succeed at weeding bad schools out of private-school-choice programs without deterring the entrance of good ones.  To be sure, regulatory deterrence is a real risk, especially among high-performing schools. Anxiety about regulations (current and future) is one of the primary reasons private schools choose not to participate in a parental-choice program.  Since parental-choice programs, in a sense, offer school operators a bargain—financial support in exchange for educating children and complying with program regulations—the burdens of the exchange must be properly calibrated with the benefits for school operators to accept the government’s offer. If the regulatory burdens are too high—or the financial support too low—some schools will decline to participate.
This is particularly true because scholarship levels in private-school-choice programs tend to be only a small fraction of the funding provided to charter and district schools, often falling below private-school tuition or the cost to educate. The fact is that higher-performing schools are more likely to decline to participate in private-school-choice programs because they tend to have higher enrollments and sounder finances, as well as less need for public funding. They usually also have fewer available seats and may be disinclined to expand to welcome participants if compliance costs are high and the per-pupil funding is low.
On the other hand, lower-performing schools likely are more willing to accept regulatory demands in order to gain the resources they need to sustain their operations by filling empty seats. In speculating about the reason for the disappointing early results of the Louisiana voucher program, for example, researchers hypothesized that “it could be the case that a higher-quality set of private schools participated in earlier voucher and scholarship programs . . . in which more positive voucher experimental impacts were reported.” They further observed: “Less than onethird of the private schools in Louisiana chose to participate in the LSP [the voucher program] in its first year, possibly because of the extensive regulations placed on the program by government authorities combined with the relatively modest voucher value relative to private-school tuition.”  Regulations targeting factors other than academic performance, including those that regulate school admissions and are perceived to curtail religious liberty, are particularly likely to have a deterrent effect on participation.
These realities are suggestive of one reason that what may appear to be the superficially easiest solution to the accountability dilemma—to mandate that all private schools participating in choice programs comply with the testing and accountability requirements imposed on district and public schools—is not optimal. The best schools, with the fewest empty seats and the least need for public resources, may balk at such a requirement and decline to participate. The realities also highlight the need for better-funded parental-choice programs. At present, the level of funding available in most school-choice programs often is insufficient to incentivize the expansion of high-performing schools to accommodate choice participants, many of whom arrive in need of acculturation to community expectations and remedial help. Current funding levels have proved insufficient to spur the development of new schools that serve underprivileged students. While additional funding is difficult to attain politically, funding equity between private, charter, and district schools ought to be a priority for parental-choice proponents. Increasing funding for private-school choice will serve to enhance both access and accountability goals.
3. Selection Bias. Opponents of parental choice frequently accuse private and charter schools of luring the best students away from district schools and then taking credit for their success. If such “cream skimming” occurs, any “gains” resulting from parental choice are mere illusions: transfer students who do better in private schools do so because they are better students than the ones who remain behind.  Whether such cream skimming occurs is the subject of vigorous debate.  There is some evidence that better-educated, more motivated parents are more likely to take advantage of parental-choice opportunities.  On the other hand, selection bias may run in both directions. At least in the urban public-school context, some research suggests that the cream may already have been skimmed. That is, many of the most motivated families have already exited struggling public schools, resulting in significantly reduced levels of within- school heterogeneity. 
Moreover, many elements of parental-choice programs may disadvantage choice schools. For example, virtually all private-school-choice programs are means-tested or limited to specialneeds children; a handful of programs are means-tested and limited to students attending failing public schools or residing in failing schools’ catchment boundaries. These restrictions effectively limit eligibility for participation to students who are most likely to have fallen behind academically. Furthermore, within the eligible group, parents may be more likely to seek alternatives for children who are not doing well in their current school—because they are struggling academically or because they have developed behavioral issues. In either case, their chosen schools face the challenge of acculturating them to higher behavioral and academic expectations, as well as remediating past educational deficits.
4. The Limits of Standardized-Test-Based Accountability. Most accountability policies rely heavily on standardized test scores. As opponents of parental choice readily acknowledge in other contexts, these measures fail to capture many important indicia of school quality, including noncognitive educational skills (for example, character, generosity, and resiliency) and achievements (for example, high school graduation and college matriculation, and persistence) that predict long-term success better than test scores do. The available research—discussed previously— suggests that the most significant benefits of parental choice often are not reflected in standardized test results but rather in “noncognitive” outcomes that are not easily measured and may unfold only over time.
Test-based measures of student performance are, moreover, only one factor influencing a parent’s assessment of a school’s quality. While the available evidence suggests that parents participating in choice programs are more informed about their children’s schools than nonparticipants and that they value school quality, it also suggests that parents consistently consider factors other than test scores (including school culture, extracurricular activities, after-school care, safety, discipline, proximity, and high school graduation rates), along with academic achievement scores, when explaining why they chose their children’s schools.  In the private-school-choice context, some evidence suggests that a religious learning environment is a particularly important factor influencing parental choice, especially among lower-income parents. 
Moreover, the state-mandated tests used to measure the cognitive aspects of student performance often are themselves deeply flawed. Private schools typically administer “norm-referenced” tests, which compare students’ performance against that of their peers by reporting relatively straightforward percentile scores. By contrast, the now-defunct federal No Child Left Behind regime required state accountability regimes to rely on tests designed to measure whether students achieved “proficiency” in certain state-determined learning goals.  Most states continue to use such “criteria-referenced” exams, although federal law now allows test flexibility. Unfortunately, state curricular standards—and the tests measuring proficiency against those standards—vary dramatically both in terms of content and rigor.  Moreover, even in states with rigorous academic expectations and well-designed tests, the results of state standard-based tests are frustratingly opaque and confusing. Most state test reports employ tiers of performance without clear meaning (such as “not proficient,” “approaching proficient,” “proficient,” and “above proficient,” or, in the author’s home state of Indiana, “did not pass,” “pass” and “pass plus”).
Aggregating the performance of students at the school level can compound the confusion. Federal law now requires that states issue school-level report cards for all district and charter schools. Most states use an A–F school grading system or a numerical index to report overall school quality. The overall school score often aggregates numerous factors (in some cases, dozens) in a statistically complex way, making it difficult for parents and schools to interpret.  Another problem: measures of “proficiency” can mask important differences between schools. Consider two schools with identical proficiency levels—say, 50%. If 75% of the students at one of the schools were significantly below proficiency the previous year, while student scores at the second school remained stable, it would be reasonable to give the first school a higher accountability rating. For this reason, most academic studies comparing the performance of schools focus on student growth rather than proficiency.  So federal law now requires that district- and charter-school report cards include measures of student “growth.” Unfortunately, these measures are even more difficult to understand than measures of proficiency. Regulations often are opaque about the details of growth determinations, and reports of student growth mask enormous statistical complexity. 
Optimal Accountability for Chosen Schools
If the goal of accountability regulations is to ensure that parents have only good choices to make, these complexities dramatically complicate regulatory design efforts. If, on the other hand, the goal of accountability regulations is to help parents make good choices, the complexities are not insurmountable. Indeed, regulations in parental-choice programs, designed with potential complications in mind, can empower parents to serve the accountability function that early advocates of school-choice programs assumed that they would. By choosing wisely, parents can drive improvements not only in their own child’s performance but also in school quality overall, limiting the need for punitive regulatory interventions.
To help parents make good choices, accountability regulations should satisfy two criteria. First, information about school quality must be transparent, readily available, easy to interpret, and matched, to the greatest extent possible, with the indicia of school quality that matter to parents in the real world. Second, accountability regulations must promote educational pluralism; that is, they should aim to expand the number and variety of schooling options available to parents, in order to provide them with more and better schools. In most cases, the best way to do so is by attracting good schools to participate in choice programs, rather than by forcing bad ones to exit them. While the exclusion of persistently failing schools from choice programs is perhaps a necessary element of an accountability regime, it should not be the primary element. The long history of public-school accountability efforts in the U.S. demonstrates both that punitive government regulations rarely drive school improvement and that closing bad schools works as a reform only when better options are available. The remainder of this essay suggests a few broad design elements that can help accountability regulations accomplish these goals.
Encourage Transparency at the School Level
Today, school-quality information about charter schools and district schools is far more readily accessible to parents than information about private schools participating in parental-choice programs. Not all programs require participating schools to administer any standardized test, and those that do often do not require that they make the results publicly available.
There are four exceptions. In Wisconsin, all public schools and schools participating in the voucher program participate in the state accountability system and receive the same report cards. In Indiana, all private schools participating in the state’s voucher program, but not scholarship tax-credit program, are subject to the same academic accountability measures as district schools and charter schools. That is, they must administer the state assessment, which forms the foundation of an A–F grade, along with several other measures, including growth. Two other states, Louisiana and Ohio, report information about the performance of students participating in choice programs but not nonparticipating students—whose parents enrolled them without using vouchers—in the same school. In Ohio, private schools participating in one of the state’s voucher programs are required to administer the state’s assessment test, but nonparticipating students may opt out. The state department of education reports proficiency rates for all tested students as well as voucher recipients for each school.  Both the Louisiana Scholarship Cohort Index and the student assessment results reported in Ohio are imperfect measures of school quality, therefore, because only voucher recipients must take the test. 
The fact that so few states require private schools participating in choice programs to make achievement data available publicly reflects a political reality: proponents of private-school choice have successfully resisted efforts to condition participation in these programs on schoollevel reporting of standardized test scores. While private schools are exempt from most state reporting requirements, the increased transparency required of public and charter schools may, over time, provide an incentive for more private schools to release achievement data voluntarily. That said, private schools’ continued anxieties about transparency may mean that imposing these requirements may dissuade some private schools from participating in privateschool- choice programs. One survey of private schools in parental-choice states found that 21% of the schools opting not to participate listed mandatory disclosure of test results as a reason for eschewing participation. 
Permit Test Flexibility
The goal of promoting transparency among private schools is linked to the continued willingness to permit schools participating in choice programs to choose among a range of academic assessments. Most private schools choose not to administer state-mandated exams. Some simply object to standardized tests altogether; others believe that nationally normed tests provide a better picture of student performance. Many private schools worry that testing mandates will force them to align their curricular content with state exams, which may unduly constrain their ability to innovate and differentiate themselves from other schools. Moreover, since private schools do not always adhere to state curricular requirements, student performance on state exams is not necessarily an accurate reflection of student achievement and progress. 
The principal challenge of permitting multiple tests is comparative: since district and charter schools must use the same assessment, apples-to-apples comparisons across schools and sectors are possible. This allows regulators and parents to compare student achievement scores between and among schools. Such comparisons are impossible if private schools opt for other assessment devices. Unfortunately, mandating that private schools in a choice program administer the state assessment would likely deter some private schools’ participation altogether. A compromise position would be to permit private schools to administer the researched-based testing regime of their choice but require them to publicly report results—ideally, for all children enrolled in their school and not only those participating in the choice program. While multiple testing regimes complicate comparisons across sectors and between private schools administering different tests, a transparency requirement will empower parents with far more information than is currently available publicly while preserving schools’ curricular autonomy.
Develop Alternatives to Standardized Tests to Measure School Quality
Federal education law (the Every Student Succeeds Act) requires states to administer the same standardized test to all district and charter schools. It also mandates other measures of school quality, including graduation rates and “noncognitive” indicia.
States have opted for a range of other measures, including, inter alia, chronic absenteeism, access to a diverse curriculum, rigor of high school curriculum, postsecondary enrollment and persistence, and, in a handful of states, measures of school climate and parental satisfaction.  There are other plausible noncognitive measures of school quality, including attendance rates, student and staff retention rates, and (for high schools) information about college enrollment and persistence. In areas with multiple public and private high school options, schools might also provide information about where eighth-graders enroll in high school.
The development of alternative measures of school quality is particularly important for schoolchoice programs for three related reasons. First, as discussed previously, while parents clearly value academic performance, it is not the only factor influencing their decisions to choose a school for their children. Test scores are only one among a range of factors. Studies suggest that only a minority of parents rank test scores as the most important predictor of school quality.  Logically, if a goal of accountability regulations is to help parents make informed decisions, the regulations should rely on and require schools to report about the factors that matter to parents, not only those that are easiest to collect. Second, where schools administer a range of standardized tests, these alternative measures mitigate the apples-to-oranges comparison problem described above. Presumably, all schools can uniformly measure and report truancy and graduation rates. Third, the research discussed previously on the long-term effects of participation suggests that noncognitive measures of school quality may capture the benefits of private schools more than standardized test scores.
One possible compromise that advances each of these three accountability goals—transparency, flexibility, and testing alternatives—would be for regulators to preserve text flexibility but require the school to report test results publicly along with one or more noncognitive measures of school quality. As is the case under federal law, schools should be given the option of selecting between a range of such indicators, subject to the approval of state regulators. And, as with district and charter schools, private schools should be given a seat at the regulatory table during conversations about the range of acceptable measures. Additionally, states could provide incentives to encourage participating schools to administer state exams (in addition, or as an alternative, to the tests traditionally administered by private schools). For example, a state might offer modestly higher scholarship amounts for students attending schools that agree to administer the state test, or, alternatively, expand the income-eligibility range for participation in parental-choice programs for such schools.
Structure Penalties to Preserve Better Choices
Finally, and importantly, accountability policies in school-choice programs ought to be structured to prevent the exclusion of private schools that are better than the other educational options in a community. Thus, as discussed earlier, closing failing schools leads to academic improvements if—and only if—students transfer to higher-performing schools. Unfortunately, since many do not, the overall academic effects of school closures are negative. However, a recent study by Stanford University’s Center for Research on Education Outcomes found that students displaced by charter-school closures tend to do better than those displaced by the closure of a district school, especially in states that worked with authorizers to “choreograph” the closure of charter schools to ensure that students landed in a higher-performing school.  This suggests that accountability policies in school-choice programs should take this into consideration. In other words, if regulators penalize struggling private schools by limiting their future participation in parental-choice programs (as in Indiana and Louisiana), they should also consider the comparative strength of other local schools before they do so. 
Conclusion: Beyond Academic Accountability
This paper has focused primarily on the questions of whether, when, and how to hold private schools participating in parental-choice programs accountable for their students’ academic achievement. It has also touched briefly on accountability debates that focus on other issues— including demands that participating schools be subject to greater government control and oversight and to comply with regulations seeking to make them more equitable and inclusive. These other regulatory demands are more problematic than the ones discussed in this paper.
Regulations that interfere with the religious liberty of participating private schools are especially problematic. These regulations raise constitutional Free Exercise concerns and questions, some of them unresolved, that are beyond the scope of this paper. For example, can states condition faith-based schools’ receipt of public funds on a waiver of the First Amendment’s ministerial exception, discussed previously? The Supreme Court’s recent decision in Fulton v. City of Philadelphia suggests that the receipt of public funds alone does not negate Free Exercise rights but provides no definitive guidance about that particular question.
Fulton invalidated, on Free Exercise grounds, the City of Philadelphia’s refusal to certify Catholic Social Services as a foster-care agency because it would not place children with same-sex couples. However, the Supreme Court declined to subject all regulations imposing substantial burdens on religion to exacting constitutional scrutiny—although it left open the possibility that it might do so in a future case. Ultimately, Fulton leaves many important constitutional questions unanswered, questions that will be tested in future litigation. States would be well advised to avoid raising them in accountability regulations.
Constitutional issues aside, regulations that seek conformity of private-school options with public- school standards are also problematic—they risk undermining a central purpose of parental- choice policies—namely, to inject pluralism into America’s education landscape by offering alternatives to public schools to children of modest means. Finally, regulations that interfere with the autonomy of private schools—especially those that threaten the religious liberty of faith-based schools—are most likely to deter them from participating in choice programs, thus undermining the increasing access to more and better schools.
About the Author
Nicole Stelle Garnett is the John P. Murphy Foundation professor of law at the University of Notre Dame, where she also is a fellow of the Institute for Educational Initiatives and the senior policy advisor for the Alliance for Catholic Education, a program engaged in a wide array of efforts to strengthen and sustain K–12 Catholic schools. She is the co-author of Lost Classroom, Lost Community: Catholic Schools’ Importance in Urban America (2014) and the author of Ordering the City: Land Use, Policing and the Restoration of Urban America (2009) as well as numerous articles on education policy, urban development, and land-use planning. Before joining the faculty at Notre Dame in 1999, Garnett served as a law clerk for Associate Justice Clarence Thomas of the Supreme Court of the United States and Judge Morris S. Arnold of the U.S. Court of Appeals for the Eighth Circuit. She also practiced law at Institute for Justice, a nonprofit public-interest law firm in Washington, DC, where she helped to defend the inclusion of faith-based schools in private-school-choice programs. She holds B.A. in political science, with distinction, from Stanford University and a J.D. from Yale Law School. Garnett is an adjunct fellow of the Manhattan Institute.
Photo by Mod Quaint/iStock