Is School Choice Enough?
As with so many issues — from trade and immigration to Russia and taxes — the Trump presidency has exposed a schism within the conservative movement when it comes to education policy. While expanding parental choice is a paramount objective on the right, a key question is whether choice alone is enough, or if results-based accountability ought to be sustained and strengthened, too. How this question is resolved will have wide-ranging consequences — for education reform in general and for the design of school-choice initiatives in particular.
Let's start in the realm of broad agreement: Conservatives believe that parents should be able to choose schools for their children that match their educational priorities and moral values. This principle stems from our deep respect for the family as the building block of a free society. A child is not a "mere creature of the state," and thus the state should not get to dictate where the child attends school.
Conservatives also value religious freedom. At a time when so much of the culture seems to be warring against traditional values, faith, and personal liberty, it's unconscionable to force children to learn in institutions that are at odds with their families' core beliefs. Thus, conservatives don't just support public-school choice, including charter schools; most also insist that choice extend to private and religious schools. Otherwise, only those wealthy enough to afford private-school tuition would be able to exercise religious liberty on behalf of their kids.
Many conservatives are particularly enthusiastic about school choice for low-income children, who need great schools the most and who often get shortchanged at their government-assigned institutions. Conservatives view upward mobility as a key objective of social policy, and want to empower poor families to choose schools that can catapult their children into the middle class — schools that aren't afraid to ask students to work hard, behave responsibly, and serve a purpose higher than themselves. Conservatives understand that religious schools can be particularly powerful institutions in the lives of the poor, especially those from communities where social capital, authority, and hope are in short supply.
Those on the right are also deeply skeptical that today's version of the traditional school system can deliver good results at a reasonable price. Like the Department of Motor Vehicles, it is plagued by bureaucracy, conflicting and self-defeating policies, a civil-servant workforce that enjoys effectively lifetime tenure, and a susceptibility to fads and nostrums that range from silly to dangerous. That many public schools remain respectable places of learning attests to the heroism of individual teachers and principals, as well as the presence and pressure of demanding, engaged families in the communities where public schools tend to be the best. Conservatives adhere to the teachings of Milton Friedman, namely that in education, as in so many other domains, the marketplace will yield greater quality, efficiency, and happiness than anything devised by politicians, bureaucrats, and "experts."
There's a consensus among conservatives, then, that expanding parental choice, especially for poor kids, is a moral good, a societal need, and a worthy end in itself. But is parental choice enough? There are two schools of thought within the school-choice movement — those who advocate pursuing a strategy that focuses only on parental choice, and those who prefer a two-pronged approach of choice and broad accountability. The debate between these two factions can help inform the way forward for conservative education reformers.
THE STATE OF SCHOOL CHOICE
While school choice is certainly the main focus of the conservative education-reform movement, there is disagreement on the question of whether school choice is sufficient, or whether reformers should set their sights on more ambitious goals. This split stems in part from honest disagreement about whether there's anything to be gained from trying to fix the public-school system. Some on the right see the public schools as so dysfunctional, so hopeless, and so impervious to change that any such effort would be a waste of energy if not downright counterproductive.
This is almost certainly true of many big-city school systems. There does not appear to be much evidence that mayors, school-board members, or teachers' unions in places like Detroit, Kansas City, or Los Angeles are willing or politically able to make the hard choices necessary to improve their schools. Though there are occasional exceptions — New York City under Mayor Michael Bloomberg and Chancellor Joel Klein; Washington, D.C., under Michelle Rhee and Kaya Henderson — they are rare and often ephemeral. So it makes sense in much of urban America for reformers to focus on choice alone — to increase the number of charter schools and help more kids get into private schools.
But everywhere else? The United States has an enormous K-12 enterprise with over 100,000 schools serving 55 million students. Many if not most of these children will continue to attend traditional public schools. Shouldn't we do what we can to make their schools even a little better? Considering that taxpayers spend more than $600 billion a year on those schools — and nobody realistically thinks the money is going to stop flowing — it seems irresponsible to just give them a pass.
For more than three decades, the conservative approach has been to pursue a two-track strategy: Push for more school choice, but also demand greater accountability from traditional public schools. In 1988, on the fifth anniversary of the publication of A Nation at Risk, the landmark report by the National Commission on Excellence in Education, President Ronald Reagan gave a speech that extolled the virtues of school choice, arguing that the way to "release the creative energies of our people" was "[b]y giving parents choice, by allowing them to select the schools that best meet the unique needs of their children, by fostering a healthy rivalry among schools to serve our young people." But that wasn't all he said. He also declared that "all of American education needs to be accountable for the only result that matters: student learning."
This mantra was picked up by conservative governors in the 1990s, including Michigan's John Engler, Florida's Jeb Bush, and, most famously, Texas's George W. Bush. For these conservative leaders, results-based accountability meant setting high standards in core subjects, testing students regularly, publishing the results, hailing success where it occurred, and using corrective tactics to get failing schools to change their ways. Bush's education successes in the Lone Star State, along with his calls to end the "soft bigotry of low expectations" and "leave no child behind," helped to cast him as a different kind of conservative — and helped him win the White House. His administration oversaw what might be considered the apex, but also the breaking point, of conservative education reform: the 2001 No Child Left Behind Act.
Much ink has been spilled in efforts to catalog the pros and cons of NCLB, but at the time it was seen by many (including me, while working in the Department of Education under Secretary Rod Paige) as a triumph of the choice-plus-accountability model. States would set academic standards against which public schools would be measured via test scores; students in chronically low-performing schools would get access to public-school options and privately provided tutoring. Charter schools got a boost, too. Private-school choice did not survive the congressional meat-grinder — the final form of NCLB had bipartisan support in both the House and Senate — but two important principles were enshrined in federal law: Achievement matters, and no child should be forced to attend a failing school.
In retrospect, it's evident that NCLB was a bridge too far. For many on the right, it came to be seen as a prime example of Bush's "big government conservatism." Myriad implementation problems drove educators and parents to distraction, while its school-choice programs fizzled and its popularity plummeted. The Obama administration made things even worse when it encouraged states to adopt the controversial Common Core standards in exchange for Race to the Top funds, then forced new mandates on schools in return for flexibility with some of No Child Left Behind's most onerous provisions. In 2015, Congress reduced NCLB to a fraction of its former self: The reauthorized version, the Every Student Succeeds Act, maintains NCLB's testing requirements, but sends most key decisions back to the states.
At the time, even many of NCLB's original supporters agreed that it was time to clip Washington's wings in education. For others, though, NCLB's political and substantive failures weren't just about Washington run amok; some prominent conservatives concluded that NCLB's animating feature — a results-based accountability system for all public schools — was not worth defending, even if it were henceforth to be state-led. They argue that conservative education reformers should jettison standards and accountability and instead focus on choice alone. And they may have an ally in Secretary of Education Betsy DeVos.
A PRIVATE OR PUBLIC GOOD
Though it's easy to lump conservative school-choice advocates into a single category, there are some major disagreements between those who argue for school choice alone and those who wish to combine it with other reform strategies. The first and perhaps the most fundamental of these disagreements can be reduced to one question: Is parental satisfaction enough?
To better understand this question, a simple thought experiment is clarifying: Imagine that conservatives are wildly successful in expanding school choice. Every parent in America, however poor or rich, gains access to several educational options, including religious schools. In a thriving marketplace, these schools compete to attract market share, innovate in ways currently unimaginable, and provide unprecedented levels of customer satisfaction. School choice turns out to be everything conservatives promised it would be.
But what if, with all of that, America's international rankings in reading, math, and science are still mediocre, or even decline? What if almost 20% of our young people still drop out of high school, well over half of those who graduate are still ill-prepared for college-level work, and half of all college matriculants still leave without a degree or credential? What if upward mobility for poor students remains stuck at low levels, and intergenerational poverty remains widespread? Could we really declare victory?
To be sure, plenty of choice advocates contend that it will indeed boost outcomes for students — both those who participate in choice initiatives and those who remain in traditional public schools. But what if that doesn't happen? Are we really content with customer satisfaction as the measure of our efforts?
If parental satisfaction is all we're after, it shouldn't be terribly hard to achieve. Polls find that most parents are already happy with their schools. The most recent survey by Education Next, for instance, finds that 62% of parents would give their own kid's present school a grade of A or B.
Yet a closer examination of America's educational outcomes indicates that our schools are mediocre almost across the board. The math and science skills of our affluent students are middling compared to affluent students in other countries; our children of college graduates do so-so compared to the children of college graduates elsewhere. Everyone talks about the achievement gap in America — between rich and poor students, or white, black, and Latino kids. But there's another gap, between the lofty perceptions Americans have of their own children's schools and the objective truth, at least as measured by results.
Why so many parents, including affluent and well-educated ones, are satisfied with schools that provide mediocre outcomes is something of a mystery. Perhaps they want to believe their kids' schools are better than they actually are to avoid feeling guilty about paying for that kitchen renovation or new SUV instead of private-school tuition. Or perhaps they appreciate other forms of educational excellence — strong sports programs, caring teachers, or engaging extracurricular activities — that don't translate into improved test scores, higher graduation rates, and college success. As education reformer Jason Bedrick recently wrote, "Parents are interested in more than scores. Parents consider a school's course offerings, teacher skills, school discipline, safety, student respect for teachers, the inculcation of moral values and religious traditions, class size, teacher-parent relations, college acceptance rates, and more."
So maybe parents don't prioritize measurable learning outcomes. But conservatives — indeed all policymakers — should prioritize them. It may be a cliché to say that "children are the future," but plenty of empirical evidence demonstrates that a well-educated citizenry is more likely to be wealthy and upwardly mobile. Studies by Hoover economist Eric Hanushek, for instance, demonstrate a strong relationship between the cognitive ability of a nation's population and economic growth. He and his colleagues have estimated that bringing all states' performance up to the level of Minnesota and other top-performing states would result in an average increase of 9% in GDP over the next eight decades, creating trillions of dollars and easily solving America's coming fiscal challenges.
School quality also has an impact on upward mobility. When Raj Chetty and his colleagues examined the likelihood that individuals growing up in poverty would make it to the middle class or beyond, the quality of area schools was one key factor. But what denoted quality wasn't spending or class size, but test scores.
New studies by Hanushek and his colleagues might explain why. In the United States, the economic "return to skills" — the benefit to individuals in terms of higher pay based on what they know or can do — is higher than in most other advanced nations. Because of our freer economy and lighter labor-market regulations, education really is what helps people get ahead. Assuming that conservatives don't want to embrace European-style economic models, developing high-level skills is essential for our children. At this point, it should be clear to everyone that Americans with low skill levels are at grave risk of being left behind.
Finally, great schools can even have an impact on family formation. A random-assignment study of career academies by MDRC (an education-research organization) found that young men who went through high-quality career and technical programs in high school were more likely later to be married and to have custody of their children.
If we care about economic growth, upward mobility, and strong families, we should make improving America's educational outcomes a priority. Education is both a private good and a public good, and a society has a legitimate interest in the education of its next generation — the more so when public dollars pay for it.
OUTCOMES AND ACCOUNTABILITY
Most choice-alone advocates concede that improving America's overall education outcomes is a worthy goal, but they are convinced that a purely market-driven system is likelier to get us there than one that combines choice and results-based accountability. This leads us to a second defining question: Should schools be held accountable for results?
It's useful to separate the question into two parts. First, does accounting for results make sense for traditional public schools? Second, should charter schools, voucher schools, and other school-choice programs be accountable for measurable success in addition to customer satisfaction? Let's take each one in turn.
The question of whether public schools ought to be held accountable for results seems like a no-brainer: Of course they should. While it's true that "competitive effects" can spur public schools to improve, the impacts to date are mild, limited to the relatively few schools that actually face competition, and dwarfed by the good that comes from results-based accountability.
Of course, there is some evidence that competition can in fact "lift all boats" — at least, all boats in reasonable proximity to one another. One well-known early study by Stanford economist Caroline Hoxby demonstrated that there were greater academic improvements in a metropolitan area with many smaller, competing school districts than in one with fewer, larger districts. More recently, scholars have found that both charter schools and voucher programs can encourage nearby public schools to improve, at least as measured by student test scores. A recent study in Washington, D.C. — where charters now serve almost half the students — found positive effects, particularly in district schools located near high-performing charters. Other studies have found negative effects, though, as traditional public schools lose some of their best students to charters.
The best way to describe the evidence is "mixed": The effects are relatively small and mainly confined to schools that actually face competition. As Dennis Epple, Richard Romano, and Ronald Zimmer wrote in a review of relevant research, "Where charters have small shares, it may be unrealistic to expect charter schools to exert much of a competitive effect." In most states, charter schools and private-school-choice programs tend to be concentrated in cities. Barring a dramatic expansion of school options into suburbs and rural communities — the places where most American kids grow up today — choice is unlikely to move the needle much via competition alone.
Results-based accountability, on the other hand, has been shown to boost outcomes significantly across a broad set of schools. Several rigorous studies have shown that the simple act of labeling schools as underperforming can lead to improvement, especially in math, and particularly for the lowest-scoring students. In one well-regarded study, Thomas Dee and Brian Jacob found that accountability's impact on math performance was significant, especially for younger, disadvantaged students.
Of course, not everyone agrees on the implications of these test scores. Jay Greene, a leading school-choice proponent and head of the Department of Education Reform at the University of Arkansas, has lately argued that test scores are weak predictors of long-term outcomes like high-school or college graduation. "[T]ests are much less reliable indicators of quality than most policy analysts, regulators, and policy makers imagine," he wrote in a recent online debate.
Yet Greene waves off several high-quality studies that demonstrate precisely this link. One, by Chetty and colleagues, linked test-score growth with later earnings gains. The Hanushek studies connect test scores to economic growth. And long-standing research on other education interventions indicates that short-term test-score gains predict important long-term outcomes, from lower rates of incarceration to higher rates of graduation. Plus, a National Bureau of Economic Research study found that the threat of sanctions against low-performing schools in Texas led to increases in their pupils' college enrollment, college completion, and eventual incomes.
WORTH THE COST
Strong evidence indicates that accountability, done right, can spur public schools to improve. Choice-only advocates may protest, though, asking yet another question: Are the benefits of accountability worth the costs? It's no secret that testing is increasingly disliked by both parents and teachers. They tell pollsters that high-stakes tests force teachers to "teach to the test," narrow the curriculum to tested subjects, encourage schools to focus solely on students who are at risk of failing, and dry up any joy and inspiration that children might have otherwise found in learning.
It's true that some studies have found evidence of narrowed curriculums and aggressive test-prep practices in certain schools. There's a solution to these problems, though: improving the tests, so that only good teaching can prepare students to do well. This is not a new idea. The College Board's Advanced Placement exams have long been celebrated for encouraging rich, rigorous instruction in their subjects. State assessments, properly designed, can do the same. Thankfully, such state tests have been dramatically improved in recent years. The Thomas B. Fordham Institute, the think tank I lead, commissioned an independent review of new tests associated with the Common Core standards and found them to be much more rigorous and encouraging of good instructional practices than the previous generation of assessments.
Furthermore, under ESSA states have the freedom to move past the hackneyed accountability regimes of the NCLB era. They can, for instance, focus more on student growth from year to year, rather than on test scores at one point in time. They can look at subjects beyond reading and math, and introduce measures of school quality that go far beyond test scores (such as parent or student surveys). They can design systems that encourage schools to raise the performance of all their students, not just the "bubble kids" at risk of failing. It is early days, but it appears that states are in fact doing many of these things.
The appropriate alternative to accountability done wrong is accountability done right. Thankfully, we appear to be moving in that direction.
ACCOUNTABILITY IN SCHOOL CHOICE
A final dispute between choice-only conservatives and the choice-plus-accountability crowd concerns school-choice programs themselves: Since charter schools, voucher schools, and other choice programs are already accountable to parents, must they be accountable to society as well? Might accountability for results be not only unnecessary, but counterproductive?
This debate has significant real-world consequences. Charter supporters have long argued that the very idea of a chartered school rests on the principle that a temporary operating license can be renewed if a school's performance is satisfactory and revoked if it isn't. Yet some libertarians, along with defenders of low-performing charters, have accused regulators of paternalism or elitism when they try to close schools that parents have chosen for their children. As Bedrick has put it, "[B]ureaucrats pay no price for being wrong....Better to leave such important decisions in the hands of people who do bear the consequences of their decisions: parents."
Closing a beloved school can be devastating; my own organization acts as a charter authorizer in Ohio, and we have learned this hard lesson from direct experience. That's why closure due to low performance is rare. Only around 3% of charter schools have been forced to close in this way over the past quarter-century. Why second-guess the decisions of moms and dads, when they know their kids best?
The flip side is that allowing schools to operate without regard to performance may harm students — and the charter sector as a whole. Charter opponents regularly use well-publicized stories of lackluster outcomes and financial (or other) scandals to paint all charters with a broad brush. States that have struggled with slipshod charter sectors, such as Ohio, have struggled to convince legislators to fund them adequately, in part because of their bad reputation. This creates a vicious cycle whereby charters struggle to improve but lack the resources to do so — in large part because their weak performance has left policymakers disinclined to help them. Meanwhile, in higher-performing places such as the District of Columbia, where the authorizer is willing to shutter failing schools, charters enjoy a positive reputation and ample resources.
The best solution, unsurprisingly, may lie between the extremes. Charter-school closure should be an option of last resort, since it is true that parents deserve deference when deciding whether a school is a good fit for their child. Parents have important information and insights about the best placement for their kids. Still, in instances of extremely low performance, authorizers should move to close schools in spite of parental objections, after looking at a wide range of indicators that go far beyond test scores, and after giving schools a chance to turn things around. Happily, more authorizers nationwide have been willing to adopt this kind of tough-love approach, which may be why charter performance has improved markedly in recent years.
A similar approach can work for private-school choice, though there are important nuances. In most states, schools participating in voucher, tax-credit-scholarship, or education-savings-account programs face few demands in terms of public accountability. Sometimes, students who are aided by such programs must take a standardized test, the results of which are used to evaluate each program as a whole. In a few states, these outcomes are published at the school level. But in Indiana, private schools accepting voucher students must administer the state test to all enrolled students, and the state applies the same performance measure to public and private schools. A participating school that doesn't perform well is not permitted to accept new voucher students until it shows improvement. (Imagine if we applied that rule to traditional public schools!)
Many conservatives understandably balk at applying this sort of top-down accountability to private schools. Besides second-guessing parents and putting too much weight on test scores, they fear this kind of results-based accountability will force private schools to become more like public schools — and less distinguishable from one another. An even bigger concern is that requiring testing, transparency, and accountability will discourage private schools from accepting scholarship students. If true, this would be a big problem, because without the participation of private schools there wouldn't be much of a private-school-choice program. And if the only private schools willing to take part are those desperate for pupils, the choice program is almost destined to be low-performing.
Some choice supporters believe this is precisely what has happened in Louisiana. Whereas the majority of rigorous, random-assignment voucher studies have shown positive or neutral effects for participants, recent research out of the state found negative outcomes for students using vouchers to move to private schools. Some analysts, including Pat Wolf at the University of Arkansas, have speculated that it may be because the best private schools aren't participating in the program. (The private-school participation rate is about one-third, which is much lower than in other states with voucher or tax-credit scholarship programs.) A survey of Louisiana private-school leaders indicated that they are, indeed, concerned about government regulations. Whether this refers to the accountability system or to something else — like limits on their ability to use admissions standards — remains unclear.
An earlier survey that my organization sponsored indicated that private-school leaders, by and large, weren't afraid of testing and accountability. However, they insisted on protecting their schools' missions, cultures, and religious identities, which would entail using their own admissions processes rather than being forced to take every student who applies. Perhaps that's why Indiana enjoys much higher school-participation rates despite its significant accountability demands; private schools there can still practice selective admissions, even if they accept voucher students. An Indiana-style system, with high accountability standards but leeway around admissions, would be a good model for other states.
THE FUTURE OF ACCOUNTABILITY AND CHOICE
Conservatives agree that school-choice programs, broadly defined to include private and religious schools, should be widely available. But many conservatives have long argued — correctly in my view — that school choice is not enough, because parental satisfaction is not enough. In order to boost economic growth, spur upward mobility, and interrupt intergenerational poverty, we need schools with higher standards. School choice is necessary for developing such schools, but results-based accountability plays an essential role as well. Many public schools remain untouched by competition from charters or private-school-choice programs; these schools in particular need to face the external pressure that comes from testing and accountability.
Ever since Donald Trump nominated Betsy DeVos to be education secretary, policy analysts and journalists have tried to pin down exactly where she stands on the question of accountability — both for the public-school system and for school-choice programs. It hasn't been easy. At times, she has expressed support for testing and accountability. At her confirmation hearing, for example, she stated that "all schools that receive public funding should be accountable." On other occasions, however, she has appeared to walk this back, even saying at one point that she's "not a numbers person."
In practice, it may not matter much where DeVos lands on this question, now that most key education decisions have shifted back to the states. There's talk of a big new federal push for school choice, but most analysts doubt it will make it past today's congressional dysfunction.
Still, as the nation's most visible conservative education reformer and most prominent school-choice cheerleader, DeVos's bully pulpit could have significant influence where it matters — in the states. In virtually every legislative session, there are fierce debates about accountability for public schools, with teachers' unions and their allies eager to return to an age when outcomes were not so well publicized. And there are fierce arguments about the appropriate amount of oversight and accountability for charter schools and voucher programs, especially in states with a history of well-documented low performance and scandal.
DeVos could do a lot of good in signaling to her fellow conservatives that they should stick to choice and accountability, rather than school choice alone. In doing so, she would be continuing a proud, decades-long tradition — one that is slowly but surely making schools better for all of our kids.