Recalibrating Accountability: Education Savings Accounts as Vehicles of Choice and Innovation

Report Education

Recalibrating Accountability: Education Savings Accounts as Vehicles of Choice and Innovation

December 12, 2016 36 min read Download Report

Authors: Lindsey Burke, PhD and Jason Bedrick

Executive Summary

In order to foster a variety of innovative and high-quality education options for all students, universal access to education savings accounts (ESAs) should be the goal of policymakers in every state.

ESAs are flexible spending accounts that parents can use to purchase a wide variety of educational goods and services, including private school tuition, tutors, textbooks, homeschool curricula, online courses, educational therapy, and more. Parents can also save unused funds for later educational expenses, such as college tuition.

This Special Report explores how ESAs expand educational opportunity and hold education providers directly accountable to parents; it also explains several common types of regulations that can undermine the effectiveness of the program and how they can be avoided.

The potential of ESAs to foster innovation and improved quality depends on a robust market in education. Increasing demand will require a critical mass of potential students, so ESAs should be made available to all families. A robust education market will also require education providers to have the freedom to innovate and parents to have the freedom to choose the providers that best meet their child’s needs. International research comparing different types of education systems has found that the most market-like, least regulated systems consistently outperformed more centralized and regulated ones. Policymakers therefore should avoid well-intentioned but misguided regulations such as open admissions requirements, price controls, state testing mandates, and excessive reporting requirements. Although intended to guarantee access and accountability, these regulations produce unintended consequences that can reduce the effectiveness of ESAs and even undermine their intended goals.

The best way for policymakers in Texas and elsewhere to expand access to a high-quality education for all children is to provide all families with ESAs that give them the maximum possible freedom to choose the education providers that work best for their children.

Introduction: A New Era in Education

America’s education system is entering a new era.

For more than a century, most American children have attended the district school assigned to them based on the location of their home. District schools offer relatively uniform instruction, adhering to the state’s standards and accountable to the state’s performance metrics. Numerous states, recognizing that it is time to shift from an institution-based model of education to a student-based model, have begun to explore a new model of education that expands students’ opportunities and makes education providers more directly accountable to parents.

This new model of education will require redesigning the way we fund education and rethinking the way we hold education providers accountable. Milton Friedman, Nobel Laureate in economics, developed a proposal for this redesign. He argued that, in order to cultivate the informed citizens that are necessary for self-government, society has a legitimate interest in ensuring that all children had access to a quality education. However, the government is not necessarily the best equipped provider to deliver that education. Rather than the government running schools and assigning students to them based on where they lived, the state should give the funds directly to parents who could then choose the school that worked best for their own children. Education choice should be universally available and free of any “conditions…that interfere with the freedom of private enterprises to experiment, to explore, and to innovate.”[1] The increased choice and competition would then spur both a wider variety of schooling options as well as improved school quality.

As of 2016, 29 states have made significant strides toward the kind of education choice policy that Friedman envisioned. However, most of these policies have limited student eligibility and some include numerous regulations that reduce the program’s effectiveness. Nevertheless, the near-consensus of high-quality research indicates that these choice programs improve student outcomes for both participating students and—through increased choice and competition—for students who remain in their assigned district schools.[2]

These education choice policies include school vouchers, tax-credit scholarships, and publicly funded education savings accounts (ESAs). ESAs are a critical refinement of Friedman’s original idea. ESAs empower parents with a much greater ability to customize their child’s education. Whereas vouchers are coupons that parents can only redeem at private schools, ESAs are quite different. They are flexible spending accounts that parents can use to purchase a wide variety of educational goods and services, including private school tuition, public school services, tutors, textbooks, curricula, online courses, educational therapy, and more. Moreover, parents can save unused funds for later educational expenses, such as college tuition.

The Ashtons are one such family. When Arizona first enacted ESAs in 2011, Marc and Lisa Ashton decided to enroll their son, Max, who is legally blind, in a private school. With his ESA, Max was able to access 90 percent of the state funding that would have been spent on him in Arizona public schools to pay for private school tuition at Brophy College Preparatory Academy. In addition to paying for private school tuition with a fraction of what was spent on him in the public system, Max was also able to purchase all of his braille materials and other assistive technology, such as a talking computer, that he needed to be academically successful. Max is now using his leftover ESA money to pay part of his tuition at Loyola Marymount University.

Through the ESA program, the Ashtons achieved a level of education customization that had previously been unavailable to them for far less than the total of federal, state, and local dollars that had been spent on Max in the district school system. The Ashtons’ story is far from an isolated one in states that have ESA programs, and demonstrates what is possible when parents are able to direct their children’s share of education funding.

ESAs not only expand educational opportunities, but they also hold education providers directly accountable to parents. As one ESA parent said when pressed by a legislator on the issue of accountability: “The accountability is with me. I am responsible for my child. I am responsible for my child’s education. It lies with me, not some bureaucrat in some high tower somewhere.”[3]

In order to foster a variety of innovative and high-quality education options for all students, policymakers in every state should pursue universal access to ESAs. The best way for policymakers in Texas and elsewhere to expand access to a high-quality education for all children is to give education providers the freedom to innovate and to give parents and students the freedom to choose the education providers that work best for them. This Special Report explores how ESAs expand educational opportunity and hold education providers directly accountable to parents; it also explains several common types of well-intentioned but misguided regulations that can undermine the effectiveness of ESAs and how they can be avoided.


Education Savings Accounts: Enhancing Access and Accountability

The promise of public education in America is that every child will have access to a high-quality education that meets his or her learning needs. Unfortunately, the district school system falls far short of that promise.

America’s district school system guarantees all children access to a seat at a school assigned to them based on the location of their home. In theory, all of these schools must meet a certain standard of quality, but in practice, no such guarantee exists. America’s district schools have fallen behind their international counterparts. On the most recent PISA[4] exam, America ranked 32nd in math among participating nations in the industrialized world,[5] despite a significant increase in funding. Indeed, adjusted for inflation, the average amount spent annually per pupil at the nation’s district schools has approximately tripled since 1970 and yet the scores of 17-year-olds on the Long-Term Trend Assessments of the National Assessment of Educational Progress have remained flat.

Moreover, the quality of district schools can vary significantly, and the schools to which students from low-income families are generally assigned tend to underperform significantly on average. Variation in funding levels does not explain the variation in results. If funding were the primary driver of quality, then Washington, DC’s district schools, which spend almost $30,000 per pupil annually, should be among the nation’s best instead of among the worst.[6] A 2012 report on international and state trends in spending and student achievement found that “[j]ust about as many high-spending [U.S.] states showed relatively small gains as showed large ones…. And many states defied the theory [that ‘additional spending on education will yield gains in test scores’] by showing gains even when they did not commit much in the way of additional resources.”[7]

So if funding is not the primary driver of school quality, what is? There is no perfect system that will always produce high-quality results. Moreover, there is no “one best way” in education. Different children have different learning needs. What is achievable is an environment in which there are incentives to meet those various needs, to improve performance, and to scale up high-quality options. By making education providers directly accountable to parents, an education marketplace provides these incentives.

In a review of the international research comparing different types of education systems, Andrew J. Coulson of the Cato Institute found that that the most market-like, least regulated systems consistently outperformed more centralized and regulated ones.[8] The most market-like systems were defined as those in which parents chose their child’s school and bore some direct financial responsibility, and in which the private schools had considerable autonomy to determine their own curricula and pedagogy and set their own wages and tuition.[9] These more market-like education systems outperformed more centralized ones by a ratio of 15 statistically significant findings to one across numerous different measures of educational outcomes, including:

  • Academic achievement;
  • Efficiency (academic achievement per dollar spent);
  • Parental satisfaction;
  • Student attainment; and
  • Subsequent earnings.[10]
  • District schools are primarily accountable to elected school boards and unelected state education bureaucrats, not parents. Parents who want to effect change in the district school system often need the support of a majority of school board members, which means that parents who disagree must vie against each other in a political process to have their preferences expressed.[11] Since politics is a zero-sum game, a district-based school system creates winners and losers and can shut out minority voices. Moreover, if a policy is set at the state level, even a majority of parents may be powerless to change it at the local level.

    By contrast, in a market-based system, schools are directly accountable to parents. Using a voucher or ESA, parents can choose the school that reflects their preferences and values and meets their child’s learning needs. Because dissatisfied parents can leave and take their money with them, schools of choice have a powerful incentive to meet their needs. If different groups of parents have differing views about education, rather than fight to have their preferences reflected in a single school system, they can enroll their children in different schools that each reflect their preferences. In this way, a market-based system fosters a greater variety of options and cultivates social harmony.[12]

    ESAs expand access to a wide variety of educational options far beyond school vouchers. A voucher system might foster a variety of options among schools, but an ESA fosters a greater variety of educational options beyond the traditional classroom. In addition to, or even instead of, enrolling at a private school, ESA students might learn from tutors, take a course online or at a local college, procure services from a public school, study from a specialized curriculum, or use some combination of these. ESAs both empower parents to completely customize their child’s education and provide a platform for innovation.

    A distinctive feature of monopolies is lack of accountability. Because district schools are not held directly accountable to parents, some policymakers have attempted to impose accountability through top-down government regulations. Yet decades of attempts to regulate district schools into quality have had little effect. Unfortunately, too many policymakers have still come to see centralized government regulations as synonymous with “accountability” rather than an inferior alternative to direct accountability to parents, and have therefore sought to impose similar regulations on choice programs. However, regulations designed for a monopoly system are inappropriate for a market-based system.

    In a market-based system, producers are held directly accountable to consumers for results. The government sets certain rules against fraud or health and safety standards, but the consumers ultimately decide whether a product or service meets their needs. Likewise, the government could ensure that ESA funds are spent on qualifying educational products and services, but the accountability for results should lie with parents, who are in the best position to evaluate those results. Holding education providers directly accountable to parents creates a feedback loop that does not exist in more centralized, top-down systems like the district schools. As social scientist Yuval Levin has argued, this enables the system to “channel social knowledge from the bottom up rather than…impose technical knowledge from the top down.” This channeling is accomplished “through a process of experimentation, evaluation, and evolution.”

    Markets are ideally suited to following these steps. They offer entrepreneurs and businesses a huge incentive to try new ways of doing things (experimentation); the people directly affected decide which ways they like best (evaluation); and those consumer responses inform which ways are kept and which are left behind (evolution).
    This three-step process is at work well beyond the bounds of explicitly economic activity. It is how our culture learns and evolves, how norms and habits form, and how society as a general matter “decides” what to keep and what to change. It is an exceedingly effective way to balance stability with improvement, continuity with alteration, tradition with dynamism. It involves conservation of the core with experimentation at the margins in an effort to attain the best of both.[13]

    This process builds on strengths and corrects errors more effectively than regulatory fiat. In a market-based system, high-quality education providers that attract families have a strong incentive to expand while less effective providers must either go out of business or imitate their more successful competitors. For this process to work, education providers should have the freedom to innovate and parents the freedom to choose the providers that work best for their children. Policymakers should keep this in mind when designing educational choice policies.

    Educational Choice: Equity and Access

    Eligibility Criteria: Universal vs. Targeted. When crafting an education choice policy, policymakers need to determine who should be eligible.

    As noted in the introduction, Milton Friedman advocated for making private school choice available for all children. Choice could revolutionize our education system, he explained, “only if they create a large demand for private schools to constitute a real incentive for entrepreneurs to enter the industry.” Moreover,

    [t]hat requires first that the voucher be universal, available to all who are now entitled to send their children to government schools, and second that the voucher, though less than the government now spends per pupil on education, be large enough to cover the costs of a private profit-making school offering a high-quality education. If that is achieved there will in addition be a substantial number of families that will be willing and able to supplement the voucher in order to get an even higher quality of education. As in all cases, the innovations in the “luxury” product will soon spread to the basic product.[14]

    Innovation and growth in the education sector require a critical mass of potential students. If policymakers limit ESA eligibility to a small number of students scattered around the state, they will help those students, but they will do little to encourage existing education providers to expand or new and innovative providers to enter the sector.

    Much of the argument for parental choice in education has focused on the urgent needs of disadvantaged students, who are often trapped in unsafe and failing schools because of the system of zip-code assignment. Providing these students with a lifeline to better options has been a top priority for many education reformers. While political realities may necessitate initially limiting eligibility to disadvantaged students, such as those with special needs or from low-income families, or those who were assigned to a low-performing district school, these students’ interests will be best served in the long term by a universal system of choice.

    Universal education choice maximizes the number of families of all backgrounds who can afford to pay for private education options, and as such, fosters the introduction of new private schools and other education providers that meet the needs of all families, at a variety of price points. By contrast, more limited school choice models tend not to result in the emergence of new options, but instead simply increase student enrollment in the schools that already exist.[15] A universal option could pave the way for innovation in the supply of education providers. This would potentially open more and higher quality options to disadvantaged students than they would have had merely from gaining more access to existing options. Moreover, universality would break the link between where children live and what school they attend, creating competition among all schools to catalyze improvements for all children.

    Universality can create broad public support for an education choice initiative to increase its likelihood of long-term viability. In order to remain politically viable in the long run, government policies require the broad support of the public, and surveys consistently show that Americans favor universal choice programs over targeted ones.[16]

    As Jay P. Greene of the University of Arkansas has argued, even more than broad public support, choice policies need the support of concentrated constituencies along with that of the general public to counter-balance the opposition of concentrated constituencies that want to curb or eliminate educational choice programs: “As much as reformers may be motivated to promote equity, a basic lesson about political reality is that more advantaged people tend to have more political power.”[17] In aligning the interests of the poor and upper income alike, universal choice programs are likely to become less politically vulnerable in the long run. Broad public support has tangible value when it brings the interests of the public at large against the concentrated opposition of unions.

    This section has explained why an ideal educational choice policy would offer universal eligibility. However, the political realities of the legislative process in some states have required the adoption of more limited choice policies or pilot programs. In such cases, priority should be given to those who have the greatest need, particularly low-income families and students with special needs.

    Four of the five states that have so far enacted ESAs originally limited eligibility to students with special needs, although Arizona has since expanded eligibility to include several other categories of disadvantaged students. ESAs are particularly beneficial for these students because they require educational therapies and services that are generally not present in a typical classroom and that many schools are ill-equipped to provide. In a survey of parents of students with special needs participating in Arizona’s ESA program, 100 percent of respondents reported being satisfied with the education they purchased with their ESA while only 43 percent were satisfied with the district school their child previously attended.[18] Moreover, the lowest-income families were both the most likely to have been dissatisfied or very dissatisfied with their previous district school (67 percent) and the most likely to say that they were very satisfied with the ESA (89 percent).[19]

    States contemplating a pilot program or otherwise limited choice policy should avoid limiting eligibility to students assigned to so-called “failing” district schools. Some states have tied student eligibility for educational choice programs to the state’s district school accountability system, offering vouchers or ESAs to students assigned to district schools receiving “D” or “F” grades, for example.

    Despite its good intentions, the “failing schools” model of choice is flawed. It is institution-centric rather than student-centric. The purpose of expanding educational choice is to provide every student with the opportunity to get the education that works best for him or her. However, even the highest-performing schools are not necessarily the right fit for all the students who happen to live nearby. A student assigned to a school that earns an “A” on average might be better served elsewhere. Moreover, school grades are subject to change, which creates uncertainty regarding the eligibility of students from year to year.

    One reason policymakers often consider a pilot program rather than full-scale adoption is to give the existing system time to adjust. Limiting eligibility to students who are switching out of the district or charter school system, or who are entering kindergarten or first grade for the first time, is one way to phase in ESAs in a manner that also maximizes the potential for savings.

    Tuition Caps, Parental Responsibility, and the Role of Prices. A tuition cap is another policy intended to guarantee access to education. To ensure that students are not turned away because of their parents’ inability to pay tuition, some school voucher programs forbid private schools from charging any tuition or fees beyond the amount of the voucher. However, as with admissions mandates, such price controls can undermine the intent of the policy. Such controls are particularly unsuited to ESAs, which empower families to purchase a much wider range of educational goods and services than vouchers.

    Direct parental financial responsibility is a key ingredient in a well-functioning education market. Capping tuition at the amount of the voucher or ESA allocation reduces parents’ direct financial responsibility, which is essential to maintaining parental control over “what, where, and by whom their children are taught.”[20] As Andrew J. Coulson concluded in his review of the global research literature on the effectiveness of different models of education systems: “Education markets work best when families pay directly for their own children’s education, and so the ideal education policy is one that makes it easier for parents to assume that financial responsibility themselves.”[21]

    When families can spend their own funds in addition to their ESA allocation, higher-income families can essentially subsidize innovation. Higher-income families have both the financial capital to afford pricier innovations and the social capital to afford to take risks that could be ruinous for lower-income families. As Friedrich A. Hayek explained:

    If we, in the wealthier countries, today can provide facilities and conveniences for most which not long ago would have been physically impossible to produce in such quantities, this is in large measure the direct consequence of the fact that they were first made for a few. All the conveniences of a comfortable home, of our means of transportation and communication, of entertainment and enjoyment, we could produce at first only in limited quantities; but it was in doing this that we gradually learned to make them or similar things at a much smaller outlay of resources and thus became able to supply them to the great majority. A large part of the expenditure of the rich, though not intended for that end, thus serves to defray the cost of the experimentation with the new things that, as a result, can later be made available to the poor.[22]

    For example, in 2007, the original iPhone sold for $500 to $600; in 2015, Walmart was selling smartphones with better specifications than the original iPhone for a mere $10.[23] This rapid innovation—which greatly improved the welfare of the poor—was possible because wealthier people paid a premium to be early adopters. Had the government, in a well-intentioned but misguided effort to ensure that everyone had access to a mobile phone, mandated that mobile phones could be sold for no more than $100, the iPhone might not have been developed. Likewise, innovations in education—such as online courses, blended learning, or new models yet to be developed—require the freedom of education entrepreneurs to cover the costs of innovating so that the models can be honed and eventually made affordable for a mass audience.

    At this point, a policymaker might consider a compromise: The program could cap tuition at the value of the voucher or ESA allocation for lower-income families while allowing higher-income families to “top off” tuition. However, such a compromise would be unwise because price controls have adverse consequences for low-income families as well.

    Price controls distort the allocation of resources.[24] Prices provide valuable, albeit tacit, information to producers and consumers that helps them determine what to produce or consume and in what quantities. Price controls distort that process. Shortages ensue when the government imposes a price ceiling, setting the maximum price below the market-clearing price (i.e., where the marginal cost per unit is equal to the marginal benefit). For example, when the federal government imposed price controls on gasoline in the 1970s, the result was a massive shortage that led to rationing, thereby producing long lines at gas stations nationwide. Research has shown that the time wasted in gas lines resulting from the price ceiling cost most American consumers more money than what they “saved” by paying less for price-controlled gasoline.[25]

    Capping tuition may be intended to increase access to more educational options, but the effect is to discourage private schools from expanding or participating at all. As economists Art Carden and John Merrifield have explained:

    Tuition caps create shortages and they drive down schooling quality by eroding school accountability to its clients. Setting tuition limits at the level established politically creates a gap between the legal purchase price of a schooling option and what families are willing to pay.[26]

    Louisiana’s experience with tuition caps is instructive. According to a report by the American Enterprise Institute (AEI), the private school participation rate in the Louisiana Scholarship Program (LSP), a highly regulated traditional school voucher program, is considerably lower than in other states. Whereas a majority of private schools in Florida and Indiana accept students participating in school choice programs, only about one-third of private schools in Louisiana accepted students paying with a voucher.[27] Among numerous contributing causes, Louisiana’s tuition cap stands out as a significant factor.

    The AEI report finds that 57 percent of respondents participating in Louisiana’s scholarship program stated that the scholarship amount was inadequate to cover the cost per pupil.[28] During the 2015–2016 academic year, the average voucher value in Louisiana was $5,856, which was barely more than half the statewide average expenditure per pupil in the district schools. Among non-participating private schools, 28 percent said that the inadequate scholarship amount played a role in their decision not to accept LSP students, and 43 percent expressed concerns that the voucher amount would not keep up with increasing costs.

    Price controls likely also play a role in reducing the incentive of participating private schools to expand enrollment. According to the AEI survey, most private schools participating in school choice programs in Florida (59 percent) and Indiana (57 percent) reported that they were planning to increase their enrollment, but in Louisiana, 65 percent of LSP-accepting schools planned to maintain their current level of enrollment and 13 percent planned to decrease enrollment.[29] By contrast, only 2 percent of participating private schools in Florida and Indiana planned to decrease their enrollment.

    Some low-income families may not be able to afford a $7,000 tuition bill, but many could make up the difference between that and a $5,856 scholarship, especially when combined with other sources of tuition aid. However, if the state forbids private schools from charging more than the scholarship amount and the schools respond by offering only a limited number of seats or refusing to accept scholarships at all, then the price controls have not actually expanded access to a quality education. The best way to expand access is to ensure that the tuition aid is as close to comparable with district school spending per pupil as possible while still producing a net savings for the taxpayer.

    Of course, like price controls, subsidies can also distort the market. Numerous studies have confirmed the so-called Bennett Hypothesis, put forward in 1987 by then–U.S. Secretary of Education William J. Bennett, which asserted that federal student aid policies were helping to fuel college tuition inflation.[30] For example, a 2015 study by the Federal Reserve Bank of New York found that both Pell Grants and especially federal student loans have significant inflationary effects.[31] A 2015 NBER study estimated that the Federal Student Loan Program alone could account for a 102 percent net increase in college tuition between 1987 and 2010.[32] Large-scale subsidies for K–12 education have the potential to produce similar effects. At the very least, because traditional vouchers must be spent in their entirety at one place and at one time, they create a price floor, as it would be foolish for schools to leave “free money” on the table. ESAs, however, are designed to mitigate such effects.

    Unlike vouchers, ESAs do not create a price floor because parents can spend ESA funds on multiple purposes and save unused funds from year to year. Moreover, these features give parents a greater incentive to economize and foster increased competition not only among schools, but also between schools and other education providers, such as tutors, online courses, educational software, and homeschool curricular materials. Robust competition is a more effective means of keeping prices affordable than government-imposed price controls. Moreover, it would be both impossible and undesirable to impose price controls on all of the possible educational goods and services that parents might wish to procure with their ESA funds, as well as nonsensical to single out private schools for price controls among all these options.

    In the case of Texas, the state could provide families with education savings accounts into which the state would deposit a portion of the annual average maintenance and operations (M&O) expenditure per pupil. During the 2014–2015 academic year, Texas district schools spent an average of $11,704 per pupil according to the Texas Education Agency.[33] The state’s average M&O expenditure per pupil was $9,022.[34] An ESA that provided families[35] with somewhere between 75 percent to 90 percent of the state average M&O expenditure per pupil would have been worth between $6,767 and $8,120 in 2014–2015. Although significantly less than what the district schools spend per pupil, these ESAs would put the average private school within financial reach of most families. During the 2014–2015 academic year, the average private school tuition in Texas was $7,847.98, and many private schools charged less than $5,000.[36]

    Legislators should strive to design educational choice policies that maximize parents’ direct financial responsibility while still ensuring universal access to a quality education. In this section, we have shown how capping tuition at the amount of a scholarship can undermine the intended purpose of the price controls. Tuition caps can discourage private schools from expanding enrollment or even accepting scholarship students. However, without such caps, state scholarships also risk fueling tuition inflation. Moreover, because ESA funds can be spent at multiple education vendors or saved for later, ESAs expand access to an even wider spectrum of educational options than vouchers without significantly fueling tuition inflation or causing shortages.

    Educational Choice: Accountability

    Testing Mandates. At the center of the technocratic approach to “accountability” is the standardized testing mandate. Standardized tests are intended to provide parents with the information necessary to assess the quality of a school they are considering or where their child is enrolled. Standardized test scores allow parents to compare the performance of their own child and the average performance of their child’s school against the performance of students in the same grade in other schools around the state and nationwide.

    As of 2015, 23 of the 59 educational choice programs nationwide required some form of standardized testing, including:

    • Fifteen of 25 voucher programs,
    • Six of 20 tuition tax-credit scholarship programs,
    • Two of the nine individual tax-credit and deduction options, and
    • Three of the five ESA programs.[37]

    These testing requirements can be divided into two types.

    1. Those, like Louisiana’s scholarship program, that require participating private schools to administer the same, uniform statewide assessment levied on the public school system.
    2. Those, like Nevada’s ESA, that require participating students to take their choice of several nationally norm-referenced (NNR) tests, like the Stanford Achievement Test or the Iowa Test of Basic Skills.

    The test scores are also used differently. In some states, schools must report the test scores only to parents. In other states, schools that accept scholarship students are required to publicly report their scores. As discussed in greater detail below, Louisiana goes even further in using the scores to sanction or even evict private schools from its scholarship program if their students fail to achieve adequate scores.

    Narrowing the Curriculum. Over-reliance on standardized math and reading tests has the propensity to narrow the curriculum. Standardized tests can provide parents with useful information about their child’s academic performance, but problems can arise when the tests are used as a part of a top-down “accountability” regime.

    When schools are rewarded or punished based on their students’ performance on math and reading tests, they have a strong incentive to divert their time and resources to tested subjects and away from others.[38] A study by the Center on Education Policy found that the time district schools spent on subjects besides math and reading declined considerably after Congress enacted the No Child Left Behind Act (NLCB), which mandated that states require district schools to administer the state standardized math and reading tests in grades three through eight and report the results. In the five years after NCLB was implemented, approximately 62 percent of elementary district schools reported increasing the amount of time spent on English language arts and/or math, and 44 percent reported decreasing time spent on social studies, science, art and music, physical education, lunch, or recess.[39]

    The narrowing curriculum is particularly alarming because, as Jay P. Greene has noted, recent research has found that “later success in math, reading, and science depends on early acquisition of the kind of ‘general knowledge’ and fine-motor skills learned through art and other subjects.”[40 ]In other words, a narrower curriculum not only deprives students of having a broader and more enriching education, but also negatively impacts their performance in the tested subjects. “If we narrow education to the mechanics of math and reading as captured by yearly testing,” Greene concludes, “we short-change the broader knowledge that is the key to academic success later.”[41]

    Mandating a single test exacerbates this phenomenon. Within the tested subjects, schools have a strong incentive to teach the concepts that will be on the mandated test. This incentive to “teach to the test” can result in a de facto curriculum. For example, if a school had been teaching math concepts A, B, and C in grade 7, but the new state test was going to cover concepts B, D, and E, the school would almost certainly drop concepts A and C in favor of D and E, even if the math teachers believe that the original curriculum was superior. Keeping the original curriculum would put their students at a disadvantage on the state test vis-à-vis students at other schools that had aligned their curriculum to the test. This standardization might make sense in a world in which there was one right way to teach math, or at least one right order to teach concepts, but that is not the case.

    Effects of State Test Mandates on School Participation. Uniform testing mandates can have a substantial impact on the number of private schools willing to participate in an educational choice program. Louisiana serves as a cautionary tale about the adverse effects of a uniform testing mandate.

    The Louisiana Scholarship Program (LSP) requires that participating private schools administer LSP students the same assessments required of students at district schools statewide.[42] Specifically, LSP students must take:

    • The Louisiana Educational Assessment Program (LEAP) state test in grades four and eight;
    • The iLEAP in grades three, five, six, and seven; and
    • End-of-course test in high school, along with the ACT.[43]

    Schools with more than 10 scholarship students in each grade or more than 40 students overall are assigned a Scholarship Cohort Index (SCI) score. Schools that receive an SCI below 50 in year two onward, or have less than a 25 percent proficiency rate on state assessments are barred from enrolling new scholarship students during the subsequent year.[44] If a private school’s results on state assessments do not meet the state’s expectations, they may be forbidden from enrolling LSP students “until their results align with program requirements.”[45]

    Private schools must also develop test security measures approved by the Louisiana Board of Elementary and Secondary Education (BESE) and can be required to make revisions to test security policy by BESE. Moreover, participating private schools must employ a testing coordinator and provide the agency with contact information for the individual. The testing coordinator must attend BESE testing workshops “as well as any additional training required to administer the state tests.”[46] Participating private schools must also appropriately code testing documents with associated accountability and demographic information, and provide that information to the state, which monitors overall testing implementation and conducts school visits during testing periods.[47]

    Although the LSP regulations explicitly state that they “exist to minimize the rare harmful circumstance rather than to regulate day-to-day conditions in participating schools,”[48] many schools have chosen not to participate in the scholarship program due to concerns about the impact of the regulations on school autonomy. The AEI survey of private schools found that among school leaders of non-participating private schools, the possibility of future regulations was the most frequently cited concern, followed by concerns about the effect of regulations on school character and identity.[49] One school leader explained:

    Private schools differentiate themselves by offering an alternative to state-run public schools. If we are forced into a mold already deemed by ourselves and our parents as inadequate, then we are no different than what the children come from.[50]

    As noted above, only one-third of private schools in Louisiana choose to participate in the LSP. By contrast, more than two-thirds of private schools participate in Florida’s tuition tax-credit scholarship program, which does not include a state testing mandate. Low levels of participation are likely due in part to state testing requirements. Of the surveyed private schools, 68 percent reported that “concerns about testing requirements” played a role in their decision not to participate in Louisiana’s voucher program, including 42 percent that said it “played a major role” in their decision.[51]

    To add insult to injury, this reduction in private school options for participating families stems from mandating one of the least important factors those parents consider when selecting a school. In a 2013 survey, only 10 percent of parents of students receiving tax-credit scholarships in Georgia listed “higher standardized test scores” as one of their top five reasons that they chose their child’s school.[52] Likewise, a comprehensive study of families participating in Washington, DC’s private school choice program found that “parents do not view test scores as the key metric of success in education.”[53] Instead, DC choice families ranked “safety” as their first priority. Once they were assured that their children were in a safe environment, they pursued schools that provided academic rigor, but rather than rely on test scores, the parents used “less-formal measures of student growth and development to assess progress and satisfaction.”[54]

    Indeed, testing mandates do not guarantee quality. With the example of Louisiana, the testing mandate for the LSP, in conjunction with other regulations, may have contributed to scholarship recipients being relegated to a selection of lower-quality private schools.

    In 2015 and 2016, education researchers published two empirical evaluations of the Louisiana Scholarship Program (LSP). These reports were notable as they were the first to find negative effects resulting from private school choice. In the more than two decades since the introduction of the first modern-day school choice program, every other random-assignment study had found that students who received a scholarship to attend a private school performed as well or better than students in the control group who did not receive a scholarship.[55]

    • The first study found significant negative effects on math, reading, science, and social studies scores as a result of participation in the LSP, with program participation reducing academic achievement among participants.[56]
    • The second study also found negative effects on the academic achievement of LSP students, particularly on their mathematics achievement scores, even as it found that the competitive effects from the voucher program improved the scores of district school students.[57]

    One possible explanation for the negative effects on academic achievement is that the better schools opted not to accept LSP students because the program was over-regulated. The AEI survey found that most private schools in Louisiana did not accept the scholarships due to concerns about the regulations on admissions and testing requirements.

    The first Louisiana study shows that, prior to enrolling in the LSP, participating schools were already experiencing declining enrollments, which may be an indicator of school quality.[58] More conclusive evidence is needed for certainty, but the available evidence suggests that the regulations promulgated by the state, including the state testing mandate, discouraged higher-quality private schools from participating in the scholarship program. Higher-quality private schools that already enjoyed a stable base of students, and were financially stable, may have been less willing to incur the burden of the LSP regulations. Along the same lines, private schools that were willing to participate in the LSP may have been those that were less financially stable, and were therefore willing to jump through the regulatory hoops required for participation. So-called accountability measures in Louisiana, including the statewide testing mandate, ironically enough may have discouraged high-quality providers from entering the market, thereby narrowing the options available to parents of students using the scholarships.[59]

    Accountability and Education Savings Accounts. A uniform statewide testing mandate can limit the supply of high-quality schools and education providers willing to participate in an education choice option as well as create an incentive for participating providers to teach to the test. However, parents need information in order to determine the quality of various education providers. There are ways to provide parents with the necessary information without reducing access or stifling innovation.

    Nevada’s ESA requires students to choose from a menu of NNR tests. Unlike uniform testing mandates, this is unlikely to constrict curricular diversity. While allowing for a range of NNRs to satisfy the ESA testing requirement provides information about student performance, NNRs do not drive school-level content decisions the way criterion-referenced statewide assessments can, which research suggests can dissuade private school leaders from participating in education choice programs.[60]

    However, even mandating NNR tests is not necessary to ensure that parents get the information they need, as the U.S.’s highly regarded higher education system demonstrates. Despite the availability of Pell Grants and federal subsidized loans, the government does not mandate that colleges administer standardized tests. Instead, parents and students rely on numerous other market mechanisms that channel expert knowledge and user experience in order to make an informed decision.

    One form of market accountability is private certification, similar to Underwriter’s Laboratories or the Good Housekeeping Seal of Approval. Parents and students rely on the judgment of these independent, third-party certifiers to know that a school meets a certain threshold of quality. Such accrediting organizations are already common in both K–12 and post-secondary education.

    Perhaps the most popular source of expert knowledge about college is the plethora of college ratings providers, such as U.S. News & World Report, Princeton Review, Forbes, Kiplinger’s, and Business Insider. These ratings offer prospective students a variety of information about student outcomes, expected earnings, course offerings, campus life, and so on. No rating system is perfect, but parents and students can compare multiple ratings to get a clearer picture of the strengths and weaknesses of different colleges and figure out which features are most important for them. Some outfits even rate the raters.[61]

    Parents and students are also relying in increasing numbers on user reviews to find the information they seek. Sites like College Times, Students Review, Rate My Professors, and Get Educated provide a platform for students to share information about their actual experiences at the college they attended.

    The K–12 sector has fewer expert reviewers or platforms for user reviews than does higher education because the vast majority of students attend their assigned district school and hence have little need to compare various options. Nevertheless, websites like Great Schools and Niche already provide expert ratings based on available data and give parents and students a platform to review their schools based on their personal experience. As the market for K–12 education grows, we should expect to see even more expert reviewers and platforms for user reviews to fill the growing demand for information about education providers. In the meantime, policymakers need to resist the urge to attempt to impose accountability from above, which would risk crowding out market alternatives.

    Reporting Requirements and the Paperwork Pileup. Another facet of government-centric accountability is the bureaucratic paperwork burden that results from reporting requirements levied on participating private schools. This compliance burden imposes real costs in terms of financial and manpower resources for participating private schools, diverting school leaders’ full attention from their primary work of running a school. Two states with statewide choice programs—Louisiana and Indiana—have reporting requirements that pose high and medium levels of administrative burden on participating schools. In both Louisiana and Indiana, the highest number of comments received on a survey of school leaders pertained to the burden imposed on them by state reporting requirements.[62]

    Louisiana again serves as a particularly instructive test case for the types of reporting requirements that can unduly burden school leaders. Private schools that participate in the LSP must provide to the Louisiana Department of Education the daily attendance of scholarship students and any absences.[63] Participating private schools are required to:

    • Allow scholarship students to enroll using an open admissions process;
    • Notify the Department of Education of those enrollments within 10 business days of the beginning of the school year; and
    • Alert the Department of Education of any student expulsions within two days of the expulsion.

    After obtaining the information from participating private schools, the Louisiana Department of Education reports, on an annual basis, to the Louisiana Senate Committee on Education, the House Committee on Education, the Joint Legislative Committee on the Budget, and individual legislators about various aspects of program implementation. These reports include:

    • Aggregate results on state tests for all scholarship recipients;
    • Aggregate proficiency rates for scholarship students;
    • Completion rates, by cohort, of scholarship students for the highest grade at a given private school;
    • Retention rates;
    • Parental satisfaction levels; and
    • Admittance rates at private schools of children with special needs.

    All private schools in Louisiana must publish their tuition and fees, help families apply to their schools, and provide families with all school rules, disciplinary policies, and academic policies, to which the parent must agree in writing. Private schools that provide special education services must provide information about what services will be offered, and must contrast that with the special education services provided by the local school system.[64]

    All surveyed Louisiana school leaders in the AEI survey of private schools reported that the paperwork required for program participation was either a minor or major burden on them, with 45 percent stating that it was a major concern.[65]

    In Indiana, which has a moderate level of regulations, private school leaders also expressed concerns with the amount of reporting requirements with which they have to wrestle. Reporting requirements also disproportionately burden smaller private schools that, as one school leader explained, “have just one principal, not a team of administrators.” Another school head of a small private school expressed similar concerns:

    I appreciate the need for accountability; however, the tangible costs, intangible costs, and opportunity costs of accreditation and administration of the voucher program with its current requirements are currently prohibitive for our small school.[66]

    The conclusion of one Louisiana school leader sums up the problem of government-centric accountability and attendant reporting requirements:

    I will not accept state tax money to then have the state tell me how to run our school. If I thought the state ran a school well, we would not have started this one.[67]


    The goal of public education should be to ensure that every child has access to a quality education. The district-based system, sadly, falls far short of that goal. America’s one-size-fits-all district schools have failed to compete internationally and performance has been flat for the past few decades, despite massive increases in per-student funding. Moreover, large gaps in the quality of district schools persist in high-income versus low-income areas.

    No school can best meet the needs of all the children who just happen to live nearby. A system in which children from low-income families are trapped in low-performing schools is one that fails its goal at the most basic level. It is time to shift from an institution-based model of education to a student-based model that empowers parents to choose the education that works best for their children. This shift will require redesigning the way we fund education and rethinking accountability for education providers.

    ESAs expand educational opportunity, empower parents to customize their child’s education, and make education providers directly accountable to parents. In the long run, ESAs have the potential to unleash the forces of innovation and revolutionize how children are educated.

    Of course, properly designing an ESA program is crucial. ESAs will foster innovation and improved quality only if there is a robust market in education. This market will require a critical mass of potential students, so ESAs should be made available to all families. A robust education market will also require education providers to have the freedom to innovate and parents to have the freedom to choose the providers that best meet their child’s needs. States therefore should avoid well-intentioned but misguided regulations such as open admissions requirements, price controls, state testing mandates, and excessive reporting requirements. Although intended to guarantee access and accountability, these regulations produce consequences that can reduce the effectiveness of ESAs and even undermine their goals.

    Education reform has no panaceas or one-size-fits-all solutions. The best way to foster innovation and improved quality at scale is to empower parents—through mechanisms such as ESAs—to find education options that meet the unique learning needs of their children. Universal ESAs would be a significant step toward the goal of ensuring that every child has access to a quality education.

    —Jason Bedrick is a Policy Analyst at the Cato Institute’s Center for Educational Freedom. Lindsey M. Burke is Will Skillman Fellow in Education Policy Studies in Domestic Policy Studies, of the Institute for Family, Community, and Opportunity, at The Heritage Foundation.

    [1] Milton Friedman, “Public Schools: Make Them Private,” Cato Institute Briefing Paper No. 23, June 23, 1995, (accessed October 26, 2016).

    [2] Greg Forster, “A Win-Win Solution: The Empirical Evidence on School Choice,” Fourth Edition, EdChoice, May 2016, (accessed October 26, 2016).

    [3] Brittany Corona, “Education Savings Accounts: Enabling Customized Learning,” The Daily Signal, August 28, 2016, The comment is relayed by Marc Ashton at 6:05 in the embedded video.

    [4] The Program for International Student Assessment (PISA) is an international exam of 15-year-old students’ reading, mathematics, and science literacy conducted in more than 70 nations every three years. Although the exam was administered to students in the fall of 2015, the most recent data available is from 2012.

    [5] Paul E. Peterson, Ludger Woessmann, Eric A. Hanushek, and Carlos X. Lastra-Anadón, “Globally Challeged: Are U.S. Students Ready to Compete?” Program on Education Policy and Governance, John F. Kennedy School of Government, Harvard University, August, 2011, (accessed October 26, 2016).

    [6] Andrew J. Coulson, “Census Bureau Confirms: DC Spends $29,409 / Pupil,” Cato at Liberty blog, Cato Institute, June 26, 2012, (accessed October 26, 2016).

    [7] Eric A. Hanushek, Paul E. Peterson, and Ludger Woessmann, “Is the U.S. Catching Up?” Education Next, Vol. 12, No. 4 (Fall 2012), (accessed October 26, 2016).

    [8] Andrew J. Coulson, “Comparing Public, Private, and Market Schools: The International Evidence,” Journal of School Choice, Vol. 3, Issue 1 (2009), p. 48, (accessed October 26, 2016).

    [9] Ibid., p. 33.

    [10] Ibid., pp. 45–46.

    [11] For hundreds of examples of this, see “Public Schooling Battle Map,” Cato Institute, (accessed October 26, 2016).

    [12] See Neal McCluskey, “Why We Fight: How Public Schools Cause Social Conflict,” Cato Institute Policy Analysis No. 587, January 23, 2007, (accessed October 26, 2016).

    [13] Yuval Levin, “The Roots of a Reforming Conservatism,” Modern Age, Vol. 57, No. 2 (Spring 2015), (accessed October 26, 2016).

    [14] Friedman, “Public Schools: Make Them Private.”

    [15] Eric A. Hanushek, “Milton Friedman’s Unfinished Business,” Hoover Digest, No. 1 (Winter 2007), pp. 42–49, (accessed October 26, 2016).

    [16] For example, in the 2016 edition of Education Next’s yearly survey of Americans’ opinions on education policy, 45 percent of respondents supported universal vouchers while 48 percent opposed a voucher program limited to low-income families. The 2011, 2012, and 2015 Education Next surveys also found higher levels of support for universal choice. (The 2013 and 2014 editions did not ask about vouchers for low-income students.) See Paul E. Peterson, Michael B. Henderson, Martin R. West, and Samuel Barrows, “Ten-year Trends in Public Opinion from the EdNext Poll,” Education Next, Vol. 17, No. 1 (Winter 2017), (accessed October 26, 2016).

    [17] Jay P. Greene, “Political Science for Ed Reform Dummies,” Education Next, August 15, 2016, (accessed October 26, 2016).

    [18] Jonathan Butcher and Jason Bedrick, “Schooling Satisfaction,” EdChoice, October 2013, (accessed October 26, 2016).

    [19] Ibid.

    [20] Andrew J. Coulson, “Direct Financial Responsibility for Parents,” in “Forging Consensus: Can the School Choice Community Come Together on an Explicit Goal and a Plan for Achieving It?,” Mackinac Center for Public Policy, April 30, 2004, (accessed October 26, 2016).

    [21] Andrew J. Coulson, “Markets vs. Monopolies in Education: A Review of the Global Evidence,” Cato Institute Policy Analysis No. 620, September 10, 2008, p. 10, (accessed October 26, 2016).

    [22] Friedrich A. Hayek, The Constitution of Liberty: The Definitive Edition (Chicago: University of Chicago Press, 2011), p. 97.

    [23] Nicholas Deleon, “Walmart’s $10 Smartphone Has Better Specs Than the Original iPhone,” Motherboard blog, November 15, 2015, (accessed October 26, 2016).

    [24] Fiona M. Scott Morton, “The Problems of Price Controls,” Regulation, Vol. 24, No. 1 (Spring 2001), (accessed October 26, 2016).

    [25] Robert T. Deacon and Jon Sonstelie, “The Welfare Costs of Rationing by Waiting,” Economic Inquiry Vol. 27, No. 2 (April 1989), (accessed October 26, 2016).

    [26] Art Carden and John Merrifield, “Significant Delusions Implicit in Schooling Price Controls,” School System Reform Studies, (accessed October 31, 2016).

    [27] Brian Kisida, Patrick J. Wolf, and Evan Rhinesmith, “Views from Private Schools: Attitudes about School Choice Programs in Three States,” American Enterprise Institute, January 2015, (accessed October 26, 2016).

    [28] Ibid.

    [29] Ibid.

    [30] For a discussion of the Bennett Hypothesis, see William J. Bennett, “Our Greedy Colleges,” The New York Times, February 18, 1987, (accessed October 26, 2016). For list of studies confirming the Bennett Hypothesis, see Neal McCluskey, “College Aid Not Helping Students: An Education Policy Reading List,” Cato Institute, (accessed October 26, 2016).

    [31] David O. Lucca, Taylor Nadauld, and Karen Shen, “Credit Supply and the Rise in College Tuition: Evidence from the Expansion in Federal Student Aid Programs,” Federal Reserve Bank of New York Staff Reports No. 733, July 2015, (accessed October 26, 2016).

    [32] Grey Gordon and Aaron Hedlund, “Accounting for the Rise in College Tuition,” National Bureau of Economic Research, September 2015, (accessed October 27, 2016).

    [33] Texas Education Agency, “PEIMS District Financial Actual Reports, 2014–2015,” (accessed October 26, 2016).

    [34] Texas Education Agency, “State Funding Reports and Data: Total Operating Expenditures 2006–2015 by School District,” (accessed October 26, 2016).

    [35] Defined here as having a household income of less than or equal to 200 percent of the federal poverty line, or $42,600 for a family of four in 2016.

    [36] Texas Public Policy Foundation, ”Texas School Finance: Basics and Reform,” March 2016, p. 44, (accessed October 31, 2016).

    [37] “The ABCs of School Choice: The Comprehensive Guide to Every Private School Choice Program in America,” EdChoice, 2016, (accessed October 26, 2016).

    [38] Erik Robelen, “Most Teachers See the Curriculum Narrowing, Survey Finds,” Curriculum Matters blog, December 8, 2011, (accessed October 26, 2016).

    [39] Jennifer McMurrer, “NCLB Year 5: Choices, Changes, and Challenges: Curriculum and Instruction in the NCLB Era,” Center on Education Policy, July 24, 2007, (accessed October 26, 2016).

    [40] Jay P. Greene, “The Case for a Broader Approach to Education,” Jay P. Greene’s Blog, March 19, 2016, (accessed October 26, 2016).

    [41] Ibid.

    [42] State of Louisiana, Department of Education, “Title 28, Part CLIII: Bulletin 133—Scholarship Programs,” April 2014, (accessed October 26, 2016).

    [43] State of Louisiana, Department of Education, “Louisiana Scholarship Program: Annual Report 2013–2014,” (accessed October 26, 2016).

    [44] Ibid.

    [45] Ibid.

    [46] State of Louisiana, Department of Education, “Bulletin 133—Scholarship Programs.”

    [47] Ibid.

    [48] Ibid.

    [49] Kisida et al., “Views from Private Schools.”

    [50] Ibid.

    [51] Ibid.

    [52] Benjamin Scafidi and James P. Kelly III, “More than Scores: An Analysis of Why and How Parents Choose Middle Schools,” EdChoice, November 13, 2013, (accessed October 26, 2016).

    [53] Thomas Stewart and Patrick J. Wolf, “The School Choice Journey: Parents Experiencing More than Improved Test Scores,” American Enterprise Institute, January 2015, (accessed October 26, 2016).

    [54] Ibid.

    [55] Forster, “A Win-Win Solution.”

    [56] Atila Abdulkadiroglu, Parag A. Pathak, and Christopher R. Walters, “School Vouchers and Student Achievement: Evidence from the Louisiana Scholarship Program,” National Bureau of Economic Research Working Paper No. 21839, December 2015, (accessed October 26, 2016).

    [57] Jonathan N. Mills, Anna J. Egalite, and Patrick J. Wolf, “How Has the Louisiana Scholarship Program Affected Students?” Education Research Alliance for New Orleans Policy Brief, February 22, 2016, (accessed October 26, 2016).

    [58] Abdulkadiroglu et al., “School Vouchers and Student Achievement.”

    [59] Mills et al., “How Has the Louisiana Scholarship Program Affected Students?”

    [60] Kisida et al., “Views from Private Schools.”

    [61] See, for example, the CBS News ratings of six of the most prominent college ratings systems. Lynn O’Shaughnessy, “Best Colleges: The Real Rankings,” CBS News, September 24, 2009, (accessed October 26, 2016).

    [62] Kisida et al., “Views from Private Schools.”

    [63] “Bulletin 133—Scholarship Programs,” Louisiana Department of Education, 2014.

    [64] Ibid.

    [65] Kisida et al., “Views from Private Schools.”

    [66] Ibid.

    [67] Ibid.


    Lindsey Burke
    Lindsey Burke, PhD

    Director, Center for Education Policy

    Jason Bedrick
    Jason Bedrick

    Director of Policy, EdChoice