This is the last issue of Education Next for which I will serve as editor-in-chief. In an era when many magazines have disappeared from newsstands, it is an honor that so many of you continue to find the journal’s material worthy of your consideration.

It has been a great ride. And the road into the future promises to be no less exciting, as leadership of the journal passes to Martin West. An associate professor at the Harvard Graduate School of Education, research fellow at the Hoover Institution, deputy director of Harvard’s Program on Education Policy and Governance, and, when Congress was drafting the Every Student Succeeds Act, a senior policy advisor to the ranking member of the U.S. Senate Committee on Health, Education, Labor, and Pensions, Marty stands tall within the education research and policy communities. I cannot think of anyone better qualified to assume this post.

Marty will collaborate with the journal’s stellar editorial team. Matthew Chingos of the Urban Institute will take Marty’s place as the executive editor with special responsibility for the journal’s research section. Especially important, Amanda Olberg, our managing editor, will continue to oversee the vicissitudes of the editorial and production processes. I look forward to working with the team in my new role as senior editor.

Education Next’s 17-year journey began in 2000 when Jay Greene, now at the University of Arkansas, insisted that K–12 education needed an independent voice. From the very beginning, the journal has been a joint venture, with multiple perspectives and diverse sources of support. The Hoover Institution at Stanford University agreed to publish the journal, and the members of its Koret Task Force on K–12 Education unanimously agreed to serve as the editorial board. Chester E. Finn, Jr., committed Fordham Foundation resources and signed on as senior editor. Fordham’s Marci Kanstoroom, also with us from the start, became an executive editor. Carol Peterson left her position at the Harvard Graduate School of Education to serve as managing editor for the new start-up. Several foundations put the journal on a strong financial footing, while the Taubman Center for State and Local Government at the Harvard Kennedy School provided space and logistical support.

There were bumps along the way, of course. Our first two issues appeared under the moniker Education Matters, a title to which a consulting firm took exception, even though it had never registered the name as a trademark. As a result, the second issue’s publication was delayed. Some of the feistier members of our team wanted to take the battle to court, but wiser heads realized that lawsuits ordinarily produce two losers, whatever the official outcome. The new title, Education Next, has served us well.

Early on, three new executive editors were brought on board: Marty West, who assumed responsibility for the journal’s research section; Rick Hess, director of education policy studies at the American Enterprise Institute, who has taken charge of the forum and book review sections; and Michael Petrilli (now president of the Thomas B. Fordham Institute), who along with Marci Kanstoroom, still serves as editor of the features section. When we realized that digital learning was coming next, Michael Horn, co-founder of the Clayton Christensen Institute, joined the editorial team.

Ten years ago we launched the first annual Education Next survey of a representative sample of the American adult population, and it has blossomed into an institution in its own right. Marty West, William Howell, and Michael Henderson have helped make this happen.

In 2009 the Education Next web site was dramatically reconceived, expanding the journal’s online audience nearly sevenfold over the next seven years. Our web team, guided by Marci Kanstoroom, ensures that the web site has fresh content daily. Our designer, manuscript editors, and media officers make certain the content is well prepared and reaches the widest possible audience.

I am in debt to the many members of our team, as you can see, but I owe the most to you, dear reader. Thank you for your support.

—Paul E. Peterson

The state of Massachusetts introduced a system of standardized testing in its public schools three years before the federal No Child Left Behind Act of 2001 mandated such practices for all 50 states. Although the tests have evolved over time, the Massachusetts Comprehensive Assessment System (MCAS) has been in place ever since. But after Massachusetts adopted the Common Core State Standards in 2010, its education leaders faced a decision: whether to stick with MCAS, which it had already revised to align with the Common Core, or switch to a “next-generation” test that was specifically designed for the Common Core—and to assess students’ readiness for college. More than 40 other states have signed on to Common Core, and many face similar decisions about their student assessment systems.

As a member of the multistate Partnership for Assessment of Readiness for College and Careers (PARCC) consortium, Massachusetts had a ready alternative in the new PARCC assessments. As of 2010, 45 states had joined either PARCC or the Smarter Balanced Assessment Consortium that was likewise developing new assessments seeking to better gauge students’ higher-level thinking skills, but the number of states participating in both consortia has since fallen.

The stated goal of the PARCC exam is to measure whether students are on track to succeed in college, while the MCAS test aims to measure students’ proficiency relative to statewide curriculum standards. But whether the PARCC test actually does a better job of measuring college preparedness was an open question prior to the fall of 2015. The Massachusetts Executive Office of Education commissioned this study in hopes of uncovering timely, rigorous evidence on how accurately the two tests assess college readiness.

This is the first study of its kind. Prior to its authorization, there was no reliable evidence that could demonstrate whether the new Common Core–aligned assessments (PARCC or Smarter Balanced) provide accurate information about which students are prepared for success in college.

Ultimately, we found that the PARCC and MCAS 10th-grade exams do equally well at predicting students’ college success, as measured by first-year grades and by the probability that a student needs remediation after entering college. Scores on both tests, in both math and English language arts (ELA), are positively correlated with students’ college outcomes, and the differences between the predictive validity of PARCC and MCAS scores are modest. However, we found one important difference between the two exams: PARCC’s cutoff scores for college-and career-readiness in math are set at a higher level than the MCAS proficiency cutoff and are better aligned with what it takes to earn “B” grades in college math. That is, while more students fail to meet the PARCC cutoff, those who do meet PARCC’s college-readiness standard have better college grades than students who meet the MCAS proficiency standard.

These results likely played a role in the November 2015 decision of the Massachusetts Board of Elementary and Secondary Education to adopt neither MCAS nor PARCC, but rather to develop a hybrid assessment that will aim to draw on the best of both tests. Our analysis cannot speak to the wisdom of that choice, which will become clear only with time. Nor should one assume that our study’s results are applicable to other states facing similar decisions: Massachusetts has been a national leader in establishing high-quality learning standards for its students, and MCAS is widely regarded as one of the country’s more sophisticated assessment systems. We do not have evidence on whether PARCC outperforms the assessment systems used in other states.

By examining rigorous evidence about the validity of both of these tests, however, Massachusetts provides a model for other states facing difficult choices about whether and how to upgrade their assessment systems.

An Experimental Test

Whether the PARCC test succeeds in measuring college preparedness better than the MCAS is an empirical question; answering it requires a rigorous, independent analysis of which test better predicts college outcomes. Our study sought to provide such an analysis. Our primary focus was the strength of association between students’ MCAS or PARCC scores and their first-year college grades. We also examined how well each test predicts whether students are assigned to remedial coursework in their freshman year.

At the end of the 2014–15 academic year, Massachusetts arranged to have a sample of 866 college freshmen take the 10th-grade MCAS and PARCC assessments. (Our final analytic sample was 847 after the scores of 19 students were removed for technical reasons—for instance, because the students did not complete the exam or their scores showed evidence of low effort.) The students were enrolled at 11 public higher-education campuses in Massachusetts. Each student was randomly assigned to complete one component of either the MCAS or the PARCC exam. This approach ensured that the students taking the PARCC assessments were not systematically different from those taking the MCAS tests.

We collected college transcript data for all students in the sample, allowing us to examine the relationship between exam scores and several different outcomes, including grade point average (GPA) and enrollment in remedial courses. By examining whether high-scoring students perform better in college than low-scoring students, we can determine whether or not the exam scores have validity in predicting college outcomes. We were also able to examine whether students who meet designated standards on the tests (“proficient” on MCAS and “college-ready” on PARCC) are likely to be prepared for college as indicated by their need for remedial coursework and by their ability to earn “C” or “B” grades in college.

This methodology has its limitations. Ideally, a study of predictive validity would be longitudinal, tracking the outcomes of students over at least three years, from the point when they complete each exam (in 10th grade) to the end of their first year in college. But Massachusetts could not wait that long before choosing its assessment. By testing college freshmen, we could immediately provide evidence regarding the college outcomes of students relative to their performance on the MCAS or PARCC exams. Our own data suggest that this approach is an acceptable proxy for a longitudinal study: the relationship between our sample’s scores on MCAS when the students took it in 10th grade and their college GPA is very similar to the relationship between their 2015 MCAS scores and their college GPA.

Our study was also limited to college students at public institutions in the state, a group that is not representative of the statewide population of public high-school students. Although our slate of participating institutions (six community colleges, three state universities, and two University of Massachusetts campuses) roughly mirrors public higher education across the state, our sample did not include students who dropped out of college before the spring semester or who enrolled in private or out-of-state institutions.

Nonetheless, the students in our sample do not differ greatly in terms of their exam performance from students statewide: students in the sample had average MCAS scores that were only slightly different than statewide averages among all 10th graders.

Predicting College Performance

We first assessed the extent to which students’ scores on the PARCC and MCAS assessments are related to their college performance (as measured by GPA) and college readiness (as measured by placement in remedial courses). We report the results of these analyses below as correlation coefficients, a statistical measure that summarizes the strength of the relationship between two variables. Correlations have a minimum possible value of -1 (indicating a perfect negative relationship) and a maximum value of 1 (indicating a perfect positive relationship). A correlation of 0 indicates that there is no linear relationship between the two variables. As a benchmark, in our study data, students’ high-school SAT scores had a correlation of 0.27 with their college grades. Given the size of the sample, our study would have been able to detect as statistically significant any differences in the correlations of the two tests as small as 0.2.

Exam scores as predictors of college performance. The primary indicator of college success that we examined is GPA. One challenge in working with this outcome is that course grades can reflect the difficulty of a subject and rigor of an institution’s grading standards. More-demanding grading standards at some institutions, for example, could lead to lower overall grades among those schools’ students, irrespective of the students’ general college preparedness. Similarly, particular subject areas might be more challenging, leading to lower GPAs for students who take more courses in those subjects. Failing to account for these differences could have biased the study’s findings. We therefore used a two-step process to establish consistency across the study sample before examining the two tests’ relationship with GPAs. First, we adjusted grades based on whether or not they were from a remedial course. Second, we adjusted grades for the institution and course subject. (The results did not change when we tested alternative methods for standardizing GPAs, such as omitting remedial course grades or accounting for students’ 10th-grade test scores.)

PARCC and MCAS scores performed about equally well in predicting college GPA: the correlations are not statistically distinguishable. In English language arts, the two correlations are identical: scores on both tests have a 0.23 correlation with grades in ELA courses. The math correlations are a bit higher for both assessments, at 0.36 for MCAS and 0.43 for PARCC; the difference between the two is not statistically significant.

Figure 1 shows the subject-specific correlations on a scale. To provide an additional benchmark, the figure also displays the correlation between students’ SAT scores and GPA in the given subject. As seen in the figure, MCAS and PARCC correlations are similar to each other, and both exams are at least as correlated with college grades as are SAT scores.

Taken together, these results allow us to conclude that the scores on the PARCC and MCAS exams are similar in their relationship to college GPA.

Exam scores as indicators of college readiness. PARCC and MCAS also do equally well at predicting which students will need remedial coursework in college, a sign that the students are not fully prepared for college-level work. In ELA, the correlation between MCAS scores and not needing remedial coursework in any subject (0.36) is very similar to the correlation between scores on PARCC’s ELA tests and not needing remediation (0.35). Likewise, in math, there is no significant difference between the MCAS (0.35) and PARCC (0.28) correlations with an indicator of which students do not enroll in remedial courses (in any subject) during their first year of college. When we examined whether students enroll in remedial courses in the tested subject specifically, we again found no statistically significant differences in the predictive ability of either test.

The SAT exam provides another measure of college preparedness. The SAT consists of three tests: reading, math, and writing. In total, 737 of the 847 students in our sample had scores in all three components of the exam, allowing us to analyze the relationships between their SAT scores and their MCAS or PARCC scores. In keeping with our other results, we found no clear pattern of differences between the MCAS and PARCC tests with respect to their relationship to SAT scores.

Comparing Performance Categories

In addition to assessing the predictive validity of the MCAS and PARCC test scores, we also evaluated the utility of the cutoff scores that define performance levels on each exam. Massachusetts has traditionally used MCAS to assign students to one of four performance categories in each subject. High school students are required to achieve, at minimum, a “needs-improvement” (level two) score in both math and ELA in order to graduate from high school. The percentage of students achieving “proficiency” (level three) also has consequences for schools under federal and state accountability regimes. In our sample of first-year college students, 75 percent of MCAS students scored as proficient or better in math, and 66 percent scored as proficient or better in ELA.

The PARCC exam has defined five different performance categories and specifies that students scoring in the two highest performance categories (level four or five) should be considered college-and-career ready in that subject. In our study data, 60 percent of PARCC students scored as college-and-career ready in math and 66 percent scored as such in ELA.

PARCC’s college-and-career readiness standard is meant to identify students who have at least a 75 percent chance of earning a “C” average in college. We examined whether the PARCC standard meets this goal by modeling the relationship between PARCC scores and the likelihood of obtaining a GPA of 2.0 (equivalent to a “C”) or better, and then calculating this likelihood at the PARCC cutoff score for college-and-career readiness.

We find that the PARCC exam’s college-ready standard not only meets but exceeds its stated target. In ELA, students at the college-ready cutoff score have an 89 percent probability of earning at least a “C” average, and in math, students at the cutoff score have an 85 percent probability of earning a “C” average or better.

In comparison, students at the MCAS cutoff score for proficiency have an 89 percent probability of earning at least a “C” average in ELA, but only a 62 percent probability of earning at least a “C” average in math. This indicates that meeting the PARCC college-ready standard in math provides a better signal that a student is indeed prepared for college-level work than does achieving proficiency on the math MCAS. At the same time, a higher share of students who were not deemed college-ready on the PARCC math test would have nonetheless been able to earn a “C” average.

More differences between the MCAS and PARCC performance levels come to light when we examine students’ average college GPAs and the percentage of students earning at least a “B” average (see Figure 2). Students in the proficient category on the MCAS ELA assessment earned an average GPA of 2.66 in their first-year college English classes. This was not statistically distinguishable from the 2.76 GPA earned by students in the college-ready category on the PARCC ELA assessment. In contrast, students who were rated proficient on the MCAS math exam had a significantly lower math GPA (2.39) than students scoring in the college-and-career ready group for PARCC in math (2.81); this margin is equivalent to the difference between a “C+” and a “B-.”

A similar pattern emerges in the percentages of students achieving a “B” average. In ELA, students in PARCC’s college-ready performance category were about 8 percentage points more likely to achieve a 3.0 GPA than students rated as proficient on MCAS, but the difference is not statistically significant. In math, however, the difference is larger: in the PARCC college-ready group, students were 24 percentage points more likely to achieve “B” grades than students rated as proficient on the MCAS math test, and the difference is statistically significant.

We also compared the validity of these performance categories by examining the percentage of students who needed remedial coursework in their freshman year despite meeting a test’s key performance threshold. This reveals a similar pattern to our results for college GPA: in math, the percentage of proficient MCAS students who enrolled in remedial courses (23.9 percent) exceeds the percentage of college-ready PARCC students who took remedial courses (12.6 percent). For the ELA performance threshold, the remediation rate for proficient MCAS students (22.5 percent) is also higher than the remediation rate for college-ready PARCC students (15.0 percent), though this difference is smaller and is not statistically significant.

It is helpful to remember that the definitions of the PARCC and MCAS performance categories are not directly comparable: the PARCC exam explicitly seeks to identify students who are prepared for college, whereas the MCAS performance levels are more narrowly targeted to measure proficiency relative to state curriculum standards.

In addition, the differences in the success rates of students meeting key performance levels on each test are not due to differences in the tests’ underlying ability to predict college outcomes. Because the underlying predictive ability of the scores is similar, performance levels could be defined in a comparable way for MCAS and PARCC, thereby making them equally predictive of college outcomes.

Implications for Massachusetts

This is the first study in any state to compare the predictive validity of one of the next-generation, consortium-developed assessments with that of the state assessment it would replace. As such, it provided timely evidence to education officials who were deciding which evaluation system to use in Massachusetts. The study’s results demonstrate that scores on PARCC and MCAS do equally well at predicting students’ success in college—an important characteristic of any state high-school assessment.

Results regarding the performance standards of each exam are also relevant to the decisionmaking process. In mathematics, the PARCC exam has defined a higher performance standard for college-and-career readiness than the current MCAS standard for proficiency, making the PARCC performance standards better aligned with college grading standards and remediation needs.

But because the underlying scores on the MCAS and PARCC assessments are equally predictive, Massachusetts policymakers had more than one option to align high-school mathematics-test standards with college readiness: one possibility would have been to adopt the PARCC exam, but another option would have been to continue using the MCAS test while simply setting a higher score threshold for college readiness. Either of these options would have achieved the goal of ensuring that the state’s high-school assessments provide better information about college readiness to students, parents, educators, and policymakers.

While our study provides valuable evidence on predictive validity, this, of course, is not the only consideration that should inform a state’s decision as to its preferred assessment. Exams may differ on a variety of other dimensions that are relevant to the state’s choice. For example, the content knowledge and problem-solving skills measured by the PARCC and MCAS tests are not identical, and the tests might differ in the extent to which they align with specific high-school curricular reform goals or teaching standards. Differences in the content of the tests could also prompt changes in curriculum or instruction that might later produce differences in college success; our study cannot assess this possibility. These additional considerations, as well as a desire to maintain independent control of its assessment program outside of the constraints of a multistate consortium, likely played into the Massachusetts state board of education’s ultimate decision to develop and adopt a hybrid test.

Broader Implications

This study provided timely evidence to decisionmakers in Massachusetts seeking to choose an examination system. For those who might be tempted to use these results to draw conclusions about the Common Core standards themselves, it’s worth repeating that the MCAS exam has in recent years been revised to align with those standards. In other words, this was a test of two Common Core–aligned exams, not a Common Core–aligned exam and a starkly different alternative.

Furthermore, over most of the past decade, the standards for student proficiency that Massachusetts has set on the MCAS exam have far exceeded those established by most other state testing programs. If the current Massachusetts proficiency standards fall somewhat short of identifying students who are fully prepared to succeed at college-level math, it is likely that the proficiency standards used in other state assessment systems under No Child Left Behind fell far short of identifying college readiness. Between 2013 and 2015, however, many states dramatically raised their proficiency standards—in some cases by adopting new assessments aligned to the Common Core (see “After Common Core, States Set Rigorous Standards,” features, Summer 2016).

Even though we cannot directly compare other states’ assessments with the PARCC test, our study provides useful evidence for any state considering adopting PARCC. In particular, it demonstrates that the PARCC exam performs at least as well as the SAT in predicting students’ outcomes in college. It also demonstrates that PARCC chose demanding thresholds for deeming a student “college-ready,” giving students good information about whether they are prepared to succeed in college courses. This is particularly important, because individual states using PARCC have the discretion to set their performance levels lower than those specified by the test developers. In Ohio, before dropping out of the PARCC consortium, the state chose to adopt a lower standard of proficiency on the PARCC exam. Ohio’s decision amounted to a state policy of grade inflation. It may have made students, parents, and educators happy in the short run, but it gave students unrealistically optimistic signals about their true readiness for college.

The bottom line is that, as many states weigh difficult choices about whether to keep or replace current statewide assessments, there is no substitute for examining rigorous evidence about the validity of the alternatives under consideration. By commissioning this study, Massachusetts has again provided a model for the nation.

Ira Nichols-Barrer is a researcher at Mathematica Policy Research, where Erin Dillon and Kate Place are analysts and Brian Gill is a senior fellow.

At the turn of the 21st century, the United States was trying to come to grips with a serious education crisis. The country was lagging behind its international peers, and a half-century effort to erode racial disparities in school achievement had made little headway. Many people expected action from the federal government.

George W. Bush and Barack Obama, the century’s first two presidents, took up the challenge. For all their differences on how best to stimulate economic growth, secure the national defense, and fix the health-care conundrum, the two presidents shared a surprisingly common approach to school reform: both preferred the regulatory strategy. In 2001, Bush persuaded Congress to pass a new law, No Child Left Behind (NCLB), which created the nation’s first reform-minded federal regulatory regime in education. When NCLB ran into trouble, Obama invented new ways of extending the top-down approach. Unfortunately, neither president came close to closing racial gaps or lifting student achievement to international levels.

The Obama administration is now packing up and heading home, leaving the regulatory machine in ruins. A new federal law, the Every Student Succeeds Act (ESSA), has unraveled most of the federal red tape. Although the mandate for student testing continues, the use of the tests is now a state and local matter. School districts and teachers unions are rubbing their hands at the prospect of reasserting local control.

With districts beset by collective bargaining agreements, organized special interests, and state requirements, choice and competition are the main levers of reform that remain. Vouchers and tax credits are slowly broadening their legal footing. Charter schools are growing in number, improving in quality, and beginning to pose genuine competition to public schools, especially within big cities. Introducing such competition is the best hope for American schools, because today’s public schools are showing little capacity to improve on their own.

From Dreary to Dismal

Two gaping gorges have been cleaved through the dreary education landscape. The black-white divide, obvious in 1966 when James Coleman analyzed equality of opportunity in the nation’s schools, remains virtually intact. Today’s racial gap is nearly one full standard deviation—approximately the difference between the performances of 4th and 8th graders. Only the South has done much to bridge racial differences between black and white high-school students, allowing the region to pride itself today on having no worse a record than the rest of the country (see Figure 1). The international gap is no less distressing. On the tests administered by the Organization for Economic Cooperation and Development, known as the Program for International Student Assessment (PISA), the math performance of U.S. 15-year-olds trails that of their peers in most other industrialized nations. The gaps between U.S. students and those in Japan, Korea, and the Netherlands are nearly as large as the ones between blacks and whites in the United States. Even students from households where a parent has been to college are not performing anywhere near the level attained abroad (Figure 2).

>U.S. 15-year-olds who scored at the 25th percentile on the PISA showed some signs of improvement between 2003 and 2012. Their scores climbed a modest 7 points in reading over this period of time, with no change in the math scores. Those at the very bottom—the lowest 10 percent—showed an even sharper gain of 17 points in reading and 12 points in math. Arguably, these results indicate that NCLB was proving helpful for the lowest-performing students, who were the focus of the regulations. But those gains were offset by a decline of 7 points in both reading and math among students scoring at the 75th percentile, with similar declines appearing among those scoring at the 90th percentile. Those who claimed that the schools were ignoring the educational needs of the better-prepared student in order to avoid leaving any child behind may have had a point. Overall, average student performance remained essentially unchanged over this time period (Figure 3). Whatever equalization was occurring did nothing for the overall quality of American schools.

Bully Pulpit

Regulation is not the only reform tool available to presidents. There is also persuasion. Since the days of Theodore Roosevelt, presidents have used the “bully pulpit” to bend the recalcitrant to their desires. Two years into the presidency of Ronald Reagan the National Commission on Excellence in Education issued a report, “A Nation at Risk,” to highlight the low, declining performance of U.S. schools. Though lacking any regulatory components, the report proved to be a blockbuster, mobilizing reform efforts in states and school districts across the country. SAT scores that had trended downward now reversed themselves. The reading scores of African American 17-year-olds on the National Assessment of Educational Progress (NAEP) leaped dramatically by no fewer than 31 points, a gain equivalent to roughly two to three years of learning (Figure 5). However, it was during this period—and only this period—that the racial gap in reading narrowed significantly.

Unlike Reagan, Bush was no rhetorician. His “compassionate conservative” messages asking that no child be “left behind” inspired little enthusiasm among either educators or students, perhaps because the close presidential election in 2000 had left the country hopelessly divided.

Eight years later, expectations for Obama were greater. The young had flocked to the Obama bandwagon, and the enthusiasm in the African American community was particularly dramatic. But the president saved his bully pulpit for Wall Street profiteers and health-care redesign, leaving him with little leverage for K–12 education reform. Early in his tenure, Obama pointed out that “leadership tomorrow depends on how we educate our students today, especially in math, science, technology, and engineering.” But even that restrained language disappeared as Obama’s term wore on. By the time of his final State of the Union address in January 2016, he had nothing to say about K–12 other than to mislead the public into a false sense of well-being: “Today, our younger students have earned the highest math and reading scores on record. Our high-school graduation rate has hit an all-time high,” he said. The rosy proclamation obscured the fact that racial achievement gaps were nearly as wide as when he entered the White House. Obama’s main education issue had become student loans and the rising cost of a university education, a topic far distant from school reform but one of great interest to present and future members of the upper middle class.

The Money Pit

Just as the bully pulpit was left vacant, the billfold was barely opened. The Bush administration was reluctant to “throw more money at the problem” of educational disparities, though it agreed to some additional spending as the price for securing NCLB’s enactment. During Bush’s administration, expenditures from federal coffers edged upward from 10 percent to 11 percent of total spending on K–12 education (with the remainder of the costs shared about equally by state and local governments). When President Obama took office, it seemed at first he would dramatically alter the federal fiscal role. With an overwhelming Democratic majority in Congress, he secured passage of a trillion-dollar tax-and-expenditure package that included more than $100 billion for K–12 education. The new money was to be spent over a two-year period, with some of it devoted to compensatory education or special education, the rest to district priorities. Federal aid to K–12 and preschool education jumped from $39 billion in 2008 to a high of $73 billion in fiscal year 2010 (0.49 percent of GDP). The following year, $66 billion in federal funding continued to flow. Much of the aid targeted urban districts with heavy concentrations of low-income and special-education populations. For two years, local school districts enjoyed a generous flow of federal cash.

<Ironically, the federal dollars arrived before the recession-induced fiscal crunch hit local revenues from local property taxes, as it takes a year or two, sometimes longer, for depressed property to be assessed at its new, lower value. But the federal dollars had to be spent immediately, in the administration’s view, and local expenditures could not be reduced. As a result, total per-pupil expenditures on education reached a near all-time high in the recession school year of 2009–10, climbing (in constant dollars) to $13,154 from a $12,520 level in 2005–06.

The education industry hoped and expected the stimulus package would set a new floor for federal expenditure. But the 2010 election knocked that plan into a cocked hat, as the newly elected Republican majority in the House of Representatives pushed federal expenditures downward. Federal aid to preschool and K–12 education dropped steadily—to just $41 billion in 2014, 0.24 percent of GDP, less than in the last year of the Bush administration. State and local governments could not—or would not—make up the difference. Expenditures per pupil (in constant dollars) slid to $11,012 in 2011–12, the latest school year for which data are available, a 4.5 percent decline. For big cities, the cuts were much larger. In Philadelphia, for example, revenue per pupil (in constant dollars) dropped from more than $15,400 at the height of the stimulus package to just $13,660 in 2013, a free fall of 12 percent, which forced deficit financing, personnel cuts, and shortened school years (see “The Philadelphia School District’s Ongoing Financial Crisis,” features, Fall 2014).

Regulation to the Rescue

Short on both rhetoric and ready cash, the presidents turned to regulation. But instead of trying to knock schools into shape with a regulatory hammer, they employed measuring sticks and magnifying glasses: each year states were asked to release information for every school on the performance of its students in math, reading, and science. This strategy rested on the assumption that the public would notice deficiencies and demand corrective actions, giving regulators the power to impose solutions.

Admittedly, regulatory reform was not invented in Washington. Calls for higher standards, minimum competency tests, and school accountability had surfaced at the state level as early as the 1970s. Southern governors—James Hunt in North Carolina, Bill Clinton in Arkansas, Jeb Bush in Florida, Ann Richardson in Texas, and others—played major roles. Outside the South, Massachusetts took the lead.

That set the stage for a strange alliance between Texas and Massachusetts in the months following the fall of the twin towers in 2001. Putting aside partisan feuding sparked by the nail-biting presidential contest in 2000, Senator Edward Kennedy and President Bush worked together to persuade Democrats and Republicans to pass NCLB, which was signed into law in January 2002. Every state was henceforth expected to set proficiency standards toward which students had to make adequate progress each year until all students had crossed that bar in 2014. The law also required annual statewide tests in grades 3 through 8, and again in high school, and states had to publish the performances of students on these tests for every school, breaking out the results by ethnicity, eligibility for a subsidized lunch, and a variety of other categories. If students were not making the requisite progress, families would have the option of picking another public school within the district. If that didn’t work, students were to have access to afterschool study programs. And if that failed, schools were to be reconstituted under new leadership.

All these steps required a vast number of regulations. But school districts still found ways of undermining federal objectives. They instituted byzantine procedures that parents had to navigate before they could exercise choice. Afterschool programs offered by private providers were frequently denied space at local schools. Reconstitution of low-performing schools often consisted mostly of window dressing.

Nonetheless, NCLB did shine a spotlight on the public schools. Every school had to report to the public the percentage of students at each grade level who performed at “proficient” or above in reading, math, and, later, science. If schools failed to make adequate progress, officials had to explain themselves to reporters, parents, and the public at large. As the goal was to make all students proficient by 2014, the explanations proliferated with each passing year.

The utopian goal set for 2014 was never meant to be taken seriously. After all, NCLB, like many other federal laws, had a five-year expiration date, and it was generally assumed that a new piece of legislation would be on the books by 2007, long before the full-proficiency deadline was reached. But when 2007 arrived, the two houses of Congress were caught in a deadlock and could not agree on new reform legislation. Instead, Congress simply extended NCLB from one year to the next (a necessary step if federal funds were to continue flowing to the states). Not until December 2015—eight years past the deadline for new legislation—did the legislators replace NCLB with ESSA.

In the meantime, the absurdities in NCLB were becoming increasingly apparent. With nearly every school failing to bring all of its students up to full proficiency, nearly every school was theoretically at risk of reconstitution. Criticisms of NCLB escalated, and many were justified. For instance, the definition of “failing schools” unfairly picked on those serving disadvantaged students. But the critiques of NCLB quickly degenerated into blanket attacks on all standardized tests: “The tide on testing is turning,” said Randi Weingarten, president of the American Federation of Teachers, who then called for NCLB revisions that would “address the root cause of test fixation.” Secretary of Education Arne Duncan, averring that testing was “sucking the oxygen out of the room,” promised to do something about it.

Race to the Top

Even before criticisms had escalated to this peak, the Obama administration, anticipating the growing opposition to NCLB, had invented an alternative way of perpetuating regulatory reform. Duncan announced Race to the Top (RttT), a competitive grants program that had been authorized and funded by the education stimulus package. At $4 billion, the money for RttT was but a minor component of the stimulus package and, overall, it amounted to less than two-tenths of 1 percent of school expenditures in the United States. Yet the idea of a competition among states for a fixed sum of money captured media attention. RttT’s purpose, the president said, was to “incentivize excellence and spur reform and launch a race to the top in America’s public schools.” In an analysis of the program, political scientist William Howell wrote that RttT encouraged applicants to develop “common core state standards,” design a teacher evaluation plan based in part on the performance of their students, ensure “successful conditions for high-performing charter schools,” and numerous other reforms (see “Results of President Obama’s Race to the Top,” research, Fall 2015). Eighteen states and the District of Columbia won RttT awards that ranged between $17 million and $700 million.

Frederick Hess at the American Enterprise Institute attacked the grant program for “its emphasis on promises rather than accomplishments,” but Joanne Weiss, who helped direct the program, later defended it by claiming that “it moved away from the notion that federal policy is designed chiefly to prevent bad actors from doing harm, and it set its sights on excellence. It urged idea-rich, capable states to define and navigate paths to educational excellence, and in so doing, to blaze trails that could show the way for other states” (see “What Did Race to the Top Accomplish?” forum, Fall 2015). Howell found that most states made efforts to follow the federal directives to which they had agreed, though he was not able to determine whether plans were effectively implemented at the local level.

The RttT competition proved so successful the Department of Education relied upon its framework for an even bolder policy: states could seek a waiver of the most onerous NCLB requirements by submitting alternative reform plans broadly similar to ones encouraged by RttT. The list included two significant proposals—the Common Core State Standards and test-based teacher evaluations. Eventually, 43 states, the District of Columbia, and Puerto Rico were granted waivers from NCLB, in effect gutting the federal law.

RttT and the waiver policies it engendered must therefore be counted as extraordinary political successes if only because they allowed the Obama administration to substitute its priorities for those of its predecessor. Although nothing in either program specifically ordered states to adopt Common Core, the standards were enacted into law by 46 states and the District of Columbia, and state expectations for student performance shifted upward (see “After Common Core, States Set Rigorous Standards,” features, Summer 2016). Teacher evaluations were instituted in the District of Columbia, and, in the following few years, student performance in the district improved by a surprisingly large margin. Obama had burnished his reform credentials every bit as bright as Bush had.

But as Martha Derthick wrote at the time, waivers “undermine the rule of law,” raising “a concern that extends well beyond the field of education.” Secretary Duncan had left himself badly exposed by constructing an education policy on a series of questionable administrative maneuvers rather than a solid piece of congressional legislation. Political opposition began to arise to two of the waivers’ key recommendations—establishing higher state standards and tightening teacher evaluations. Tea Party activists attacked Common Core, objecting to what the Heritage Foundation called the Obama administration’s intent to nationalize “the content taught in every public school across America.” And teachers unions tightened the screws by balking at unfair evaluations of teacher performance. “Old tests are being given, but new and different standards are being taught,” National Education Association president Dennis Van Roekel declared. “This is not ‘accountability’—it’s malpractice.” Meanwhile, student progress on NAEP tests came to a virtual standstill (Figure 4).

Caught in the maelstrom, the Obama administration was unable to defend against a bipartisan move on Capitol Hill to end waivers altogether by enacting, for the first time since 2002, a new federal education law, ESSA. The law requires annual testing but leaves it to the states to decide how the results will be used. Most of the other regulations have been removed, shifting authority over schools back to states and localities. As an education reform strategy, federal regulation is dead. The regulations had little long-term effect, and the political opposition crescendoed. The regulated captured the regulators. Nor is there much appetite for new accountability rules at the state level. If reform is to take place as the rest of the 21st century unfolds, it will happen because more competition is being introduced into the American education system.

Competition

Introducing competition is slow, arduous, disruptive, upsetting, and politically divisive. Benefits come slowly. Losses are painful. The politics is messy at best, disastrous at worst. Winners are ingrates who feel they deserve any benefits they enjoy. Losers blame not themselves but changes in the rules of the game. But the long-term consequences of competition for consumers and society as a whole can be amazingly beneficial, as deregulation of the airlines and telecommunications industries has shown. Comparable gains have yet to appear throughout American K–12 education, but to see how it might happen, let us reflect on the slow growth of choice and competition via vouchers and charters that has taken place over the past quarter of a century.

Vouchers. Milton Friedman made the case for school choice in his seminal 1955 article on school vouchers, writing:

[School choice] would bring a healthy increase in the variety of educational institutions available and in competition among them. Private initiative and enterprise would quicken the pace of progress in this area as it has in so many others. Government would serve its proper function of improving the operation of the invisible hand without substituting the dead hand of bureaucracy.

Thirty-five years after these words were penned, Wisconsin enacted a voucher program for the city of Milwaukee. Since then, another 28 state legislatures have passed some kind of voucher program, tax credit, education savings account, or other intervention that provides government aid to students attending private schools. None of these programs are at scale, however. Nationwide, less than 1 percent of the school-age population is participating. But studies show that voucher students of minority background, even if they do not perform much better on standardized tests than their peers in public school, are more likely to graduate from high school and go on to college (see “The Impact of School Vouchers on College Enrollment,” research, Summer 2013). Apparently, private schools seem to do better at fostering character and grit than at academic instruction per se (see “Schools of Choice,” features, Spring 2016).

Perhaps this is why vouchers are popular within the minority community. According to Education Next’s annual public-opinion poll (see “The 2015 EdNext Poll on School Reform,” features, Winter 2016), 48 percent of African Americans support universal vouchers that any family could access, and 65 percent favor a voucher plan limited to those of low income, a feature of most current voucher plans. Support among Hispanic adults is comparable, but the public as a whole is less enthusiastic. Only a third of the public supports targeted vouchers, though opinion with respect to universal vouchers is more or less evenly divided between supporters and opponents.

The strongest opposition comes from teachers unions. Al Shanker, the brilliant (if controversial) leader who led union efforts to win collective bargaining rights in New York City, denounced the idea of vouchers: “Without public education, there would be no America as we know it,” he cried. Vouchers for the poor would be “merely the nose of the camel in the tent.” School boards and teachers themselves could not have agreed more.

Charters. Union opposition to vouchers was so intense it opened the way for another choice reform—charter schools. When first enacted in Minnesota in 1989, charters appeared to be nothing more than a safe place for teachers to try out new ideas that public schools could adopt. Shanker himself initially endorsed charters, making it difficult for subsequent union leaders to express unconditional opposition. Unions nonetheless resist charter growth because the schools are run by nonprofit organizations rather than the government; they are free of many state regulations; and they are usually not subject to collective bargaining agreements. Yet charters can claim that they are in fact public schools. They are authorized by a government agency (a state department of education, state university, mayor’s office, or local school district). Their operating funds come primarily from government sources. Their educational mission is secular. When parental demand for a charter school exceeds available space, the school typically holds a lottery in order to choose impartially among the applicants. Because of these characteristics, charters regularly win endorsements from Democratic and Republican leaders alike, and the movement has enjoyed steady, if unremarkable, growth. Forty-three states allow the authorization of charters, more than 6,000 charter schools have been established, and nearly three million children now attend them (Figure 6).

As their numbers grow, charters are beginning to disrupt the status quo more than vouchers are. In places where these schools of choice are allowed, students are no longer limited to attending their neighborhood public school but can apply to a charter school elsewhere in the community. In other words, charter schools compete with district schools for students. The greater the number of charter schools, the more intense the competition.

Admittedly, charter schools have had difficulty penetrating rural communities. There, a public school, no matter its quality, is perceived as a valuable community institution. Sports events enliven Thanksgiving mornings and Friday nights, school auditoriums are a favorite spot for community events and town meetings, and successful high-school graduates are celebrated in hometown newspapers. The school district is also one of a community’s major employers.

In suburban communities, charters appeal to families who reject the rigid, routinized forms of instruction thought to be prevalent in district schools. The pedagogy often emphasizes project learning, experiential learning, and other student-determined exercises. However beneficial such approaches may be in some respects, they do not seem to translate into higher performances on state-required exams. Even more importantly, school choice among traditional public schools already exists for those who have the resources and resourcefulness to purchase homes or rent apartments in neighborhoods that offer the best educational opportunities. The affluent already have the options they need.

The story is different within central cities. Big-city public schools are in big-time trouble, and many families send their children to their local school more out of necessity than choice. For these families, the charter school option often holds strong appeal. Such schools are generally perceived to be smaller, safer, friendlier, and, more often than not, a better place to learn than district schools. In contrast to progressive charters in suburban areas, central-city charters typically embrace the “no-excuses” model of teaching and learning, emphasizing strict dress codes, rigorous discipline, extended school days and school years, and high expectations for performance on standardized tests. In general, urban charters are outperforming their traditional public-school counterparts. The charter advantage seems to be particularly striking for African American students from low-income families.

The charter school movement has benefited from the spectacular results achieved by the Harlem Children’s Zone Promise Academies, Success Academy, BASIS Schools, KIPP Schools, Uncommon Schools, and others in New York City, Boston, Los Angeles, New Orleans, Washington, D.C., and other prominent cities. Led by strong entrepreneurs, staffed by high-quality teachers from selective colleges, financed by local donors and major foundations, these institutions are providing rigorous instruction over an extended school day and school year. Their ability to lift students who come from low-income, single-parent families to a high level of performance that prepares them for college has shed a warm glow over the entire charter-school undertaking. According to Education Next’s 2015 poll, supporters of charters outnumber opponents by a two-to-one margin, both among the public at large and in minority communities. Partisan differences remain, however, as nearly a third of all Democrats express opposition to charters, as compared to less than one in five Republicans.

Charters serve about 5 percent of the public-school population, less than the 11 percent attending private schools. But the market share of private schools has been stable for decades, while charter school enrollments are growing. In 16 cities, more than 25 percent of public school students are enrolled in charter schools. In New Orleans, the percentage is no less than 79 percent; in Detroit, 51 percent; in the District of Columbia, 43 percent; and in Philadelphia, 28 percent. In many other cities—Los Angeles, Boston, New York City, Chicago, and elsewhere—charter enrollments would be much higher if supply were not artificially constrained by state laws limiting charter growth.

Will the competition between charters and standard, district-operated public schools intensify over the next decade? Is this competition the new reform wave that will sweep over American education? Is there a tipping point at which the demand for charters will force a reconstruction of the educational system more generally? Several factors point in that direction:

· Many charters in urban areas are oversubscribed.

· Big-city school districts must spend a large share of their budget for employee health-care benefits and pensions, a problem charters have escaped thus far.

· Charter school parents can be mobilized in numbers when political confrontations occur.

· Student performance at charter schools is showing signs of improvement over time (mainly because of the closing of weak charter schools).

· Some charters are using new technologies that personalize and customize the learning experience.

· Competition, once introduced, is difficult to reverse.

· Big-city school systems are fighting charters by giving parents a wider array of choices among their public schools, suggesting that the choice genie has escaped from the bottle.

These are straws in the wind, but it is still too soon to predict confidently the degree to which choice will be introduced into American education over the next decade. Teachers unions are mobilizing to block charter expansion in state legislatures and through collective bargaining agreements with local districts. Also, one finds little support for charters in suburbia, small towns, or rural America. If charters achieve a breakthrough, it will be in the country’s largest cities. Spreading out from that base will be a slow, arduous process, achieved only if charters demonstrate that they can deliver a superior educational experience.

If the future of charter schools remains uncertain, the same cannot be said for top-down regulation. Unless teachers surprise us all by embracing a new curriculum generated by Common Core standards, and that curriculum motivates students to make a greater commitment to their learning, reforming the system from within is unlikely to succeed in the years ahead. If school reform is to move forward, it will occur via new forms of competition—whether they be vouchers, charters, home schooling, digital learning, or the transformation of district schools into decentralized, autonomous units. And if student testing has an impact on reform, it will be due to the better information parents receive about the amount of learning taking place at each school. The Bush-Obama era of reform via federal regulation has come to an end.

The essay was the Friedman Lecture presented in January 2016 before the Centre for the Study of Market Reform of Education (CMRE), London, U. K.

Paul E. Peterson, editor-in-chief of Education Next, is professor of government and director of the Program on Education Policy and Governance at the Harvard Kennedy School.

There are no bells at Salt Lake City’s Innovations Early College High School, and no traditional “classes.” Students show up when they like, putting in six and a half hours at school between 7 a.m. and 5 p.m. Working with a mentor teacher, students set their own goals and move through self-paced online lessons. They can take more time when they need it or move ahead quickly when they show mastery.

This model would feel cutting edge almost anywhere, but it is all the more so at Innovations, a public high school that was founded by the Salt Lake City School District four years ago. The district designed Innovations to capitalize on “blended learning,” a mix of online and teacher-led instruction. Ken Grover, the school’s founder and principal, proposed a school that would put teenagers in charge of their own education. Innovations draws high, average and low achievers who crave that freedom.

The school attracts “students who the existing system has failed,” said Michael B. Horn, a blended-learning consultant. “It’s really ‘bleeding edge.’ They don’t really have classrooms.” While individualized online instruction has become prevalent in many high schools, it is mostly used as an add-on, to offer special classes like foreign languages or credit-recovery courses. As a major aspect of a school model, it is more readily associated with charter schools, such as Summit Public Schools and Rocketship in Silicon Valley.

Innovations is among a small group of district-based trailblazers to use the “flex model” of blended learning as its foundation. Last year, the Christensen Institute, co-founded by Horn, cited Innovations as one of a dozen “proof points” nationwide for blended learning.

With help from the new federal Every Student Succeeds Act (ESSA), blended learning could move from the margins to the center at more district schools. The law supports state-led innovation broadly, and provides funding for states and districts to implement personalized, blended, and online learning.
Indeed, Horn believes that the Innovations model is “where high school is going in the long run.” But, he cautioned, “if you don’t have the right culture in place and teachers haven’t bought in, it can go horribly wrong.”

Students Set the Pace

Innovations is growing in popularity. Now in its third year, the school serves 360 students, with another 150 late applicants on a waiting list.
In 2012, when Innovations was under development, the district was looking for ways to raise its graduation rate. Grover, who’d led a traditional Salt Lake City high school as well as the district’s career and technical programs, asked students what they liked and disliked about high school. What did they want?

They said they valued the relationships they’d made, and “even the learning,” but felt the school day was “structured to the needs of teachers and sports teams,” he said.
Plus, they reported, school was “boring.”

What did that mean?

Students said that if they didn’t understand something, they couldn’t get their questions answered right away. Teachers with a whole class to manage couldn’t slow the pace to deal with one student’s confusion.

For other students, the pace was too slow.

For someone who’s confused, or already knows the material, those 90-minute classes could be “painful,” Grover said. “I think it’s one of the primary reasons students walk away from school.”

Students wanted to “come when they wanted, work at their own pace, and be treated like adults,” he said.

Today, Innovations students do just that while also taking advantage of the early-college and career-tech opportunities that the school’s location encourages.

Innovations occupies the corner of an airy new building on the South City campus of Salt Lake Community College (SLCC), along with the district’s Career and Technical Center (CTC). The state reported that last year about 1 in 10 Innovations juniors and seniors took community-college classes and more than half took courses through the CTC, whose offerings range from computer programming and CD graphics to barbering and nail design. In addition, students may ride a shuttle bus to their local high school to participate in classes or extracurriculars, such as choir, band, foreign language, or sports.

This flexible model appeals to students with diverse goals.

Sophomore Jenny Le is looking to turbocharge her high-school years. She hopes early college classes will enable her to earn a university degree in two years, getting a head start on medical school and saving money, and she is already thinking about how old she’ll be when she completes her medical residency.

Career tech and a small-school environment drew Jacquie Robb, also a sophomore, to Innovations. She’d heard that animal science students qualify to train a guide dog. She put that off to take a law enforcement class—“taught by police officers,” she pointed out—and then fire science, “taught by firefighters.” She now wants to become a police officer and “catch the bad guys,” she said.

In addition, both girls took an emergency medicine class taught by an EMT, which can lead to employment-ready certification when they turn 18. The class appealed to Robb’s interest in first-responder careers, and Le’s plan to work her way through college, on her way to becoming a doctor.

Last year, Innovations had its first student—a Latino male—graduate with an associate degree as well as a high school diploma. This year, five to seven 12th graders are on that track, Grover said.

An Early Exercise in Adulthood

At first glance, it all seems loosey-goosey.

Innovations has bare-bones “classrooms”—with tables, chairs, and computers—but no classes in the traditional sense. At any given time, there are likely to be some students working independently or with a partner and a teacher talking with one to four students.

A few early birds show up at 7 in the morning, but most students start their day between 9 and 9:30, and a few don’t arrive till midmorning. After logging on to the learning management system, which shows personal goals and progress data, a student decides what to study and when to move on. Bored with bio? Pop into the English room to watch The Taming of the Shrew or polish an essay. Students can take more time on an assignment when they need to, or move ahead quickly to the next thing when they demonstrate mastery.

Throughout the day, small groups of students form to collaborate on projects or labs. They may choose the same book to read for their English course so they can discuss it together. Other students opt to work alone. Meanwhile, tutors and teachers walk around looking for students who need help, or meet by appointment to work with individuals or small groups.

The school follows the district curriculum, which is aligned to Common Core State Standards. Everyone takes English language arts, math, and two other classes at a time, unless a counselor agrees to more.

Despite the do-your-own-thing atmosphere, students are expected to complete at least eight classes a year. Teachers and mentors track each student’s progress and step in when a student isn’t moving fast enough to stay on track for graduation.

Career tech and community college classes are taught face to face. Innovations students are expected to figure out where they need to be and when they need to be there.

It’s a combination of self-directed study and responsibility that can help students get ready for life after graduation—on the job or at college—when it’s executed well.

“The right personalized-learning model” can engage students who haven’t done well in traditional schools, said Don Soifer, executive vice president of the Lexington Institute, who’s researched blended learning. Teachers “are able to spend less time ‘teaching to the middle’ and more time connecting with students as individual learners.”

But not everyone thrives in a self-directed environment, at least not at first, said Andrew Calkins, deputy director of the Next Generation Learning Challenges initiative at Educause, which supports learning-technology innovations. (Innovations is not a grantee.) “We’re finding with our grantees that a third of students do very well in this model, a third will adapt in a semester, and the final third need a lot of help to make the transition,” he said.

Diana Senechal, a high-school philosophy teacher who warns of the distractions of technology in her book Republic of Noise, sees a downside in letting students “tailor their learning to their own needs and preferences.” She asked, “How will they ever be challenged?”

“What Are You Supposed to Do All Day?”

Innovations students are partnered with a mentor teacher who will guide them through all four years of high school. Mentor teachers meet at least once a week with students, and four times each year with students’ families. They also e-mail and text families, and send home paper transcripts each month for easy display on the refrigerator door. The communication and transcripts are designed to explain where each student is, and where he or she should be.

“You can’t hide here,” said Grover, who attributes the success of Innovations to a strong culture and cadre of teachers. “Some of our students don’t like that.”

Heather Bauer is in her third year as an English teacher at Innovations. Previously, she taught in a traditional classroom setting at a charter school.

“At first, I thought, what are you supposed to do all day?” said Bauer, who noted she now loves her role. “I know my students so much more, and I can teach at a different level. I can delve deep.”

Bauer starts her day at 7 a.m. Not many students arrive that early, so she has time to send e-mails, grade papers, tweak units, and hold mentor meetings. Then, from 10 a.m. to noon, she gathers small groups of students for class meetings or individual coaching—discussing strategies with a student who needs backup for his thesis statement in an essay, for example.

Throughout the day, she will gather students—who may be in different rooms—for a “class meeting” to discuss a concept. Students who’ve already mastered it need not participate. Usually, students are working independently, or with a friend, while the teacher coaches one student at a time. Teachers stagger their work schedules so there’s always a math teacher and an English teacher available onsite.

It can be frustrating to teach a lesson to two students, and then do it again with another two who weren’t available before, said Bauer. But while she was teaching to a classroom of students at one time in a conventional school, she said, “I felt I was just talking to myself.”

The opportunity to revisit lessons is baked into the Innovations model. Since classes don’t follow any particular schedule, students can complete them as slowly or as quickly as time and preparedness allow. To complete a class and earn credit, students must show 70 percent mastery of 100 percent of the content, and cannot progress to the next class until they do. In addition, they can return to classes they’ve finished if they want to improve their grades. A teacher will “reactivate” the subject and show what work students can do to raise a “C” to a “B,” or a “B” to an “A.”

In other words, “you can’t get an F,” said Grover, the principal. You only can get more time.

Accordingly, teachers play multiple roles to keep students on their many tracks. Bauer mentors individual students in all four grades, and is also responsible for the English language arts progress of 200 students in the 11th and 12th grades.

Teaching at Innovations can be labor-intensive, she said, especially among 9th-grade students, who come in expecting to be “led from one class to another” and earn credit by sitting quietly.

They have to learn to handle freedom, she said. When they do, “the fact that students take ownership of their education” makes all the difference.

Unique Goals, and a Unique Culture

Despite Salt Lake City’s image—white and Mormon—city schools serve an ethnically diverse student body. At Innovations this year, 42 percent of students are white and 30 percent are Hispanic, with multiracial, Pacific Islander, African American, Asian, and Native American students attending in smaller numbers. More than half of students qualify for free or reduced-price school lunch. Some 17 percent are English language learners, and 7 percent receive special education services.

Innovations also enrolls students with a range of historic academic performance. Among entering students, about 15 to 20 percent perform above grade level, about 15 to 20 percent below; the rest are in the middle, said Grover. Each student has an individualized education plan.

Innovations students take state exams, just like students at Salt Lake City’s four other district high schools. Teachers analyze results to see how well students are meeting learning objectives.

After a shaky first year, the school’s graduation rate rose to 89 percent in 2014, hit 93 percent in 2015, and is projected to reach 95 percent in 2016. That’s significantly higher than the district and state average. In addition, Innovations students earn more college credits than those at the city’s other high schools combined.  
Many Innovations graduates go on to SLCC and the University of Utah and other public institutions, Grover said. Achievement among Latino students is a particular point of pride.

Students and teachers also praised their school for offering an encouraging, intimate atmosphere.

For all the computer time, Innovations is a high-touch school, said Jenny Le, the aspiring doctor. “You get moral support here. The teachers all learn your names. If I reach a goal, my mentor says, ‘Jenny did it!’ in a really loud voice so everybody knows.”

Craig Ellis, who is working as a math tutor while he completes his teaching degree, said he loves the one-on-one time he gets with students and the opportunity to build long-term relationships over their high-school careers. At the graduation ceremony, the mentor hands the diploma to the graduate. “It’s a big moment, very emotional,” he said. “I’m looking forward to it.”

Jacquie Robb, the aspiring police officer, said she was bullied in middle school, where cliques and “mean girls” ruled the day. She arrived at Innovations too shy to ask teachers or tutors for help in math, her weakest subject. Then a classmate—a girl who’s “going to be Einstein,” Robb said—helped her in math. “I was so quiet when I came here. They opened up my shell. I feel loved here.”

At Innovations, people “cheer you on,” she said. “You don’t have to look perfect every day.”

Unlike other Salt Lake City schools, Innovations has no behavior management plan, because it’s not needed, said Grover. As principal of a traditional high school, he dealt with fights every day; at Innovations, there have been none. He has not suspended a student, ever. “When a child isn’t frustrated, and is having their needs met, they don’t lash out,” he said. “We build support structures here.”

The Price of Flexibility

Innovations had few obvious peers when it opened in 2012. The closest parallels were the private micro-schools, such as AltSchool and Acton Academy, and Summit Public Schools, a charter network in Silicon Valley that incorporates a variety of learning software into its programs.

Now, Innovations-inspired district schools have opened in Idaho and Indiana, and more are planned in Arkansas, Oklahoma, Virginia, and Atlanta, Georgia, said Horn of the Christensen Institute. At least a thousand visitors have toured Innovations since it opened, said Grover, who frequently speaks about his school at conferences.

As in Salt Lake City, “districts are starting to create lab schools to try personalized, student-centered strategies with hopes of finding what will transfer to traditional schools,” said Calkins of Educause. ESSA, he said, “hangs a big welcome banner for this kind of model . . . Thanks to ESSA, the pace of experimentation will pick up.”

ESSA includes the Student Support and Academic Enrichment Grant program, which is authorized at about $1.6 billion annually and can support blended learning. Probably the most critical piece is funding training for teachers so they can figure out how to use digital tools effectively.

Innovations is doing blended learning on a shoestring. Utah ranks last in the nation in per-pupil spending, and Innovations gets the same relative pittance as Salt Lake City’s conventional high schools: $3,100 per student for operating costs and $3,200 for facilities-related costs.

It saves money by not offering electives or extracurriculars; students visit their neighborhood high schools for those. As at other Salt Lake City high schools, the student-teacher ratio is 29:1.

However, Innovations pays tuition for students who take community college classes and pays a fee for CTC enrollees.

Students use iMacs or MacBook Air laptops, or bring their own. The learning management system and the 
digital content, curriculum, and grading are provided by 
Spark Education (formerly iGo), and PowerSchool tracks student information. Teachers edit student work and add questions during their review using customizable software.

A student who’s taking a full load at SLCC is costly. “Our relationship to the community college is like any marriage that’s struggling,” said Grover. “You fight over children and money.”

Still, Grover—whose own daughter chose to enroll in Innovations—wants that marriage to grow. Someday, he hopes, half of Innovations students will graduate high school having also earned an associate degree or certificate.

The secret to its current and continued success is simple, he said. “We create a culture, curate content, and hire the right teachers.”

Joanne Jacobs, the author of Our School, is a freelance writer and education blogger at joannejacobs.com.

Education Savings Accounts (ESAs) apply the logic of school choice to the ever-expanding realm of education offerings. Rather than simply empowering families to select the school of their choice, ESAs provide families with most or all of the funds that the state would have spent on their child’s education. Funds can be spent on private school tuition, tutors, and online courses, for example, or be saved for future use.

In this forum, Matthew Ladner, senior advisor for policy and research at the Foundation for Excellence in Education, argues that ESAs offer a grand advance over charter schooling and deserve the support of reformers. Nelson Smith, education policy consultant and senior advisor to the National Association of Charter School Authorizers, argues that the latest incarnation of ESAs in Nevada poses substantial risks and threatens to disrupt an increasingly successful charter school movement.

• “The Next Step in School Choice,” by Matthew Ladner

• “Expand Choice, but Keep the Public Interest in Mind,” by Nelson Smith

We know little about how to improve the cost-effectiveness of K‒12 spending, and our ignorance is not likely to diminish under the status quo. Paul Hill, of the University of Washington’s Center on Reinventing Public Education, succinctly summarizes our situation:

Money is used so loosely in public education—in ways that few understand and that lack plausible connections to student learning—that no one can say how much money, if used optimally, would be enough. Accounting systems make it impossible to track how much is spent on a particular child or school, and hide the costs of programs and teacher contracts. Districts can’t choose the most cost-effective programs because they lack evidence on costs and results.

In a few states, policymakers have taken a creative step toward solving this problem by giving public-education money directly to K–12 parents in a liberal but use-restricted and monitored account that encourages families to treat the money as their own. Known as Education Savings Accounts, these young programs represent a possible path out of our ignorance.

The founders of American public education promised fantastic returns on investment from Prussian-inspired schools run by local governments. This system has worked well for many children, but for others, not so much. It’s a safe bet, however, that public education is here to stay. Every state constitution guarantees state funding for K‒12 education, and the public broadly and deeply supports it. Sadly, though, our current public-school practices fall far short on equity, efficiency, and overall effectiveness. A 19th-century institution cannot fulfill the needs of 21st-century America.

Policymakers have engaged in a series of decentralized efforts to improve education outcomes by increasing choice for families. “First-generation” choice programs such as open enrollment, magnet and charter schools, and voucher plans have indeed increased the number of schooling options available. However, these models provide no incentive for parents to consider the issue that Hill rightly spotlights: cost-effectiveness. Building on the success of charters and vouchers, supporters of Education Savings Accounts hope to overcome the inherent limitation of those choice strategies. The next generation of choice programs must not only give parents and guardians the freedom to try different methods but must also incorporate incentives for them to consider academic benefits and financial costs.

A small but growing number of states (Arizona, Florida, Mississippi, Nevada, and Tennessee) have created ESA programs, which enable parents to choose educational services for their children from among a wide range of offerings. Parents enroll in the program by agreeing to provide their children with an alternative education that replaces the one the public school would provide. In return, the parent receives a state-funded account that can be put toward multiple but limited uses: private-school tuition, tutoring from certified tutors, individual public-school courses, online programs, community college and university tuition, standardized testing fees, curriculum costs, and saving for future higher-education expenses in a tax-advantaged federal Coverdell Account. ESA program details vary by state, but the core features generally include parent-managed accounts, a range of allowed uses, and the option to save funds for future educational use.

The first ESA program in the nation, Arizona’s Empowerment Scholarship Accounts, aims for a system of “ordered liberty.” Parents can freely mix education methods and providers, but while their options are many, they are not infinite. Accounts operate under state supervision, with safeguards in place to ensure the appropriate use of funds. Arizona’s lawmakers initially crafted the ESA program for families of children with disabilities and have since expanded it to other groups, including children in foster care, military dependents, children who attend low-rated schools and districts, children living on Native American reservations, and the siblings of otherwise eligible students. Lawmakers in Florida, Mississippi, and Tennessee chose to limit eligibility to students with special needs, while Nevada lawmakers, facing extreme overcrowding in public schools, made all public school students (if previously enrolled for at least 100 days) eligible to participate in an ESA.

A Better Mousetrap and Why We Need It

The British coined the phrase “quasi-market mechanism” to describe policies that replace state service monopolies with a roster of independent providers who compete for business. The American experiments with charter schools, school vouchers, and now ESAs fit comfortably within this definition. High-quality evaluations of charter and voucher programs demonstrate greater parental satisfaction, along with higher graduation rates, often at lower overall taxpayer cost. Account-based programs, the most recent of these innovations, may offer the greatest potential benefits yet for students and taxpayers. Assuming that ESA administrators can successfully develop oversight techniques, ESAs could become the most powerful and flexible mechanism we have for customizing education to individual student needs.

Charters, vouchers, and tax credits have had some impact on cost-effectiveness. The available evidence suggests they provide a bigger bang for the education buck and have contributed to the evolution of our K–12 delivery and governance models. But these market-based reforms lack an incentive for consumers themselves to consider cost—a crucial element for market efficiency and effectiveness. Value seeking in the process of voluntary exchange has driven human material progress, but until now it has been all but absent in public education. An ESA system incorporates competition and thus encourages service providers to create and offer the best possible product at the lowest possible price.

Despite the success of the charter school enterprise, such schools have not faced pressure to improve their product while continuously lowering costs. Applying such a principle to the education sphere may sound Darwinian, but in the rest of the world—whether in manufacturing, technology, energy, or agriculture—it is simply the way things are done. Innovators have developed amazing devices and services whose prices continue to fall. Meanwhile, in both voucher and charter models, schools simply spend the maximum amount the state provides them.

The Future and Her Enemies

My forum partner Nelson Smith has leveled a number of critiques against the Nevada legislation—by far the boldest of the ESA experiments. It is true that ESAs remain a learn-by-doing experience. No legislation (ESA or otherwise) passes in perfect final form, and an undertaking as ambitious as Nevada’s will require further refinement.

Smith’s concerns range from serious to poorly considered. Let’s start with one of the latter. Smith expresses concern that the Nevada ESA law does not employ a means test. This is an odd objection, given that nothing in American K–12 education (or higher education for that matter) employs a means-test. A student is never told, “Sorry, Susie, but your parents are too wealthy so you aren’t allowed to attend this school.” Everyone pays the taxes that support public education, and everyone is eligible to participate. Polling by Education Next and others continues to find that the public prefers universal programs to means-tested approaches—responding more positively, for instance, to the notion of vouchers for all than to vouchers for low-income families only (see “The 2015 EdNext Poll on School Reform,” features, Winter 2016).

Smith notes that the education reformer Howard Fuller opposes Nevada’s ESA program because it could reinforce the advantage of the rich over the poor in gaining access to a scarce supply of private school seats. In states where private schools are a major presence, this could be a legitimate concern. However, a phase-in starting with lower-income families would be preferable to a marginalizing means test. For instance, the Cleveland Scholarship Program gives enrollment preference to low-income students, but includes students of all income levels. Large Midwestern cities in the 1990s had a significant number of inner-city private schools with available seats, making for a valid concern in terms of their distribution. In Nevada’s case, the preexisting stock of private schools is simply too small to cause much concern.

Nevada public schools are badly overcrowded, and the school-age population projects to grow further still by the hundreds of thousands. The New York Times quoted the Clark County (Las Vegas area) superintendent as saying that he could build 23 new elementary schools and they would quickly become overcrowded. Las Vegas schools are surrounded by portable buildings manned by substitute teachers. The current Nevada private schools, however, will not be riding to the rescue for many. Few private schools exist in Nevada, and those that do mostly (and predictably) operate in areas with enough high-income families to support them. In 2010, the Nevada Department of Education estimated K–12 private-school enrollment in the state at a whopping 3 percent of the total. Let’s assume those schools could clear out their basements and make room for an additional 1 percent of the students. This limited number of new seats would barely move the needle on coping with overcrowding, requiring us to look elsewhere.

Success or failure for the Nevada ESA program lies in innovation—new private schools, new micro-schools, new Cristo Rey-model schools (students sharing office jobs to generate revenue), blended schools, co-op arrangements drawing from the home schooling experience, combinations of the above, and who know what else? As one of many reforms promoted by the Nevada legislature, the ESA program is not a magic cure-all, nor does it represent a “fire-it-and-forget-it” missile. The program will need considerable philanthropic investment if it is to realize its full potential—especially for low-income children.

Nevada’s ESA program does, however, provide every low-income student with schooling options they previously lacked—professional tutoring, textbooks, therapies, university, college or community college tuition, transportation, curriculum, distance education, testing, individual public school courses, and extracurricular activities, in addition to private school tuition. What uses will Nevada parents make of this new freedom? There is only one way to find out.

Nevada lawmakers chose to reflect equity concerns by providing additional resources to disadvantaged students. The law provides $5,700 to participants whose families qualify for free or reduced-price lunch or who have an Individual Education Plan, and $5,100 to all other students. Critics have been quick to criticize both the relatively low level of funding overall and the amount of the additional assistance (about 12 percent more) given to low-income children and children with disabilities.

These relatively low allotments would have been higher, something close to the full amount of per-pupil spending, if not for a late amendment to the enabling legislation. The amendment grew from the lawmakers’ desire to protect the interests of school districts rather than an indifference to equity concerns on the part of school choice supporters.

A tendency to view locally generated funds as the entitlement of districts rather than the entitlement of the child, however, is a problem that afflicts both charter and private choice programs. Many states, for instance, attempt to make up for the lack of local funding for charters with additional state assistance on a per-pupil basis. Such assistance is very much needed, but it also creates a growing strain on state general funds. This is an issue that both charter and private choice supporters must soon confront; but it is hardly unique to the Nevada ESA program.

The additional funds provided to special-needs children stand as clearly inadequate, but that is because Nevada overall does not use a weighted funding formula that would provide extra money for special-needs children. The state should develop such a system for funding the public schools and apply it to the ESAs as well.

Additional assistance to low-income children would also be most welcome, but much of the criticism on this front sorely lacks context. The Nevada public school system routinely gives the most to the kids born on third base. If you attend Incline High School in the upscale town of Incline Village, for instance, you in effect “receive” more than $13,248 in public funds—that is, the per-pupil expenditure in that community, which is far above the state average of $8,274 per pupil. I agree that giving low-income students just 12 percent more funding in an ESA program is not enough, but it’s important to note that few people balk at students in communities such as Incline Village receiving some 60 percent more funding than average. Compared to the general Nevada funding formula or the formulas that govern most district and charter schools nationally, the Nevada ESA program looks positively progressive in giving more money to kids starting off with less.

Investments and Institutions Needed to Make ESAs Work

Fortunately, governments and private enterprise have been developing techniques applicable for ESA account oversight for decades. ESA implementation efforts must adapt and customize techniques and lessons learned from programs such as food assistance (which transitioned from a voucher system to an account mechanism) and health savings accounts. A system of public oversight and controlled reimbursement for expenses can ensure public confidence in proper use of funds.

Many will rightly worry about the possibility of charlatans duping parents with education snake oil. Smith correctly notes that start-up enterprises are of uneven quality. Once again, practices outside of education suggest a way forward. Online rating systems such as Yelp that aggregate customer reviews could easily be adapted to ESAs. If I were an ESA parent, I would have zero interest in what my state education officials had to say about the quality of, say, the online courses offered by a given university. On the other hand, I would be very interested in whether other ESA parents and students found such courses useful, appropriate, and worthwhile. With an online consumer rating system, parents themselves could evaluate the programs and services of the vendors, such as tutors, universities, schools, and community colleges. Such a system could be a valuable resource for ESA parents and might help them avoid the rip-off artists and sub-standard providers.

Account mechanisms such as ESAs could well become our most powerful tool in reengineering the way we provide public services, not only in education but also in health care. There is much to be gained from incorporating voluntary exchange as a core principle of public education. Those of us who support ESAs recognize how little we know. We don’t have the answer to Paul Hill’s cost-effectiveness puzzle, but we do have an idea about how to empower parents to figure it out themselves. Experience is the best teacher, so let’s get on with it. We just might learn something.

Matthew Ladner is a senior advisor for policy and research at the Foundation for Excellence in Education.

This is part of a forum on education savings accounts. For an alternate take, please see “Expand Choice, but Keep the Public Interest in Mind,” by Nelson Smith.

There are about a quarter million low-income kids enrolled in Nevada schools. Some of these children live in small rural communities, but most reside in the Clark County school district surrounding Las Vegas. Their numbers are concentrated in the state’s worst-performing schools, and 49 of the 78 schools identified by the state as chronically failing are in Clark County.

The state’s overall academic performance lags national averages, with its students scoring in the bottom quartile in both reading and math on the 2015 National Assessment of Educational Progress results. But the state’s urgent task is to provide new, high-quality seats for the 57,000 students languishing in its worst schools, those occupying the bottom 10 percent of academic performance for at least three years. A new Achievement School District (ASD),  modeled after those in Louisiana and Tennessee, is charged with transforming these very low-performing schools by taking in a small number (no more than six per year) and pairing them with successful charter networks, with the hope that they will return to district supervision once they’re humming. The ASD has a distinct, limited mission, its resources aimed squarely at saving students who are stuck in intolerably lousy schools.

That work could be supplemented by a more general expansion of charter schools. The state has no cap on charters, and there is ample room to grow. Clark County has more than 19,000 students in charter schools, but they account for just 6 percent of the district’s more than 336,000 public-school students. Furthermore, the sector’s performance is far from exemplary at this point, and aggressive efforts by state charter officials to recruit top operators from around the country have been hampered by Nevada’s abysmally low per-pupil funding.

However, the state has taken major strides to position its charter sector for increased success. New state revenues from last summer’s tax hike will help attract strong operators to the state; the federal Department of Education has just awarded a $16 million grant for new charter start-ups; and reforms passed in 2013 and 2015 persuaded the National Association of Charter School Authorizers to declare Nevada’s charter law the strongest in the nation.

Instead of keeping an unrelenting focus on students facing dire needs, however, Nevada is rolling the dice on another marquee program, a custom version of Education Savings Accounts (ESAs). Typically, ESAs are designed to provide education options to families who otherwise could not afford them. The Nevada program is explicitly not targeting low-performing schools or low-income families but rather is being made available to all, including affluent families who can already exercise choice by locating in a good school district or paying tuition for a private school.

That feature is what lost the support of longtime social-justice warrior (and founder of the pro‒school choice Black Alliance for Educational Options) Howard Fuller, who in July shocked many allies by stating his opposition to the Nevada plan: “Parental choice should be used principally as a tool to empower communities that face systemic barriers to greater educational and economic opportunities … I could never approve of a plan that would give those with existing advantages even greater means to leverage the limited number of private school options, to the detriment of low-income families.”

That is not only a sound moral argument but also a good synopsis of the program’s flawed economics. An ESA provides $5,100 per pupil ($5,700 for a low-income student or one with disabilities), which is supposed to secure a private-school spot. But average tuition for private elementary schools in Nevada is $8,558, and at the high school level, $10,322. That leaves a mighty big gap for a low-income parent to fill, but it’s a much lesser lift for folks who bring home a generous paycheck.

Supporters argue that upscale families won’t be tempted away from their public schools by the lure of ESAs, but indications are to the contrary. According to data from the state treasurer’s office, early enrollment is coming mostly from well-off neighborhoods. “Overall, half of the nearly 3,100 applications submitted as of Oct. 28 list an address in a ZIP Code among the top 40 percent of median households in Nevada. That’s in contrast to just 10.7 percent of applications from households with median incomes in the bottom 40 percent.”

Now, my forum partner Matt Ladner dismisses this as a “poorly considered” argument because “nothing in American K–12 education (or higher education for that matter) employs a means test.” Come again? Buying a home in Greenwich or Grosse Point or Chevy Chase so your child can attend a great public school certainly involves a hefty means test—and if you don’t agree, ask the low-income folks in the urban centers down the road. Moreover, we apply means tests every day in deciding which students and schools will benefit from Title I and other programs that are federally-resourced but administered at the district level. True, no child would ever be told, “Your parents are too wealthy so you aren’t allowed to attend this school.” But that point misstates the issue. Susie’s wealthy parents are already taking good care of her schooling and don’t need an additional state subsidy. We ought to spend scarce public dollars on those who need the help. To use an analogy: Everyone pays taxes that fund the fire department, so everyone’s entitled to protection. But that doesn’t mean you send trucks to spray every house in town. You deploy them to the fires.

Here, let me note what I am not worried about: the set of church-state issues raised by the ACLU, which is currently suing Nevada state treasurer Dan Schwartz to halt implementation of the accounts because they will steer tax dollars to sectarian institutions. While Nevada’s constitution is unusually direct in forbidding such funding—and about half of the state’s private schools are sectarian—the ESA program has been carefully crafted to fit the contours of the U.S. Supreme Court’s landmark Zelman v. Simmons-Harris decision (2002), which held that an Ohio voucher program allowed parents “to exercise genuine choice among options public and private, secular and religious.”

Rather, my concerns are secular and pragmatic.

The Supply Side Is Weak 

Right now there are 186 private schools in Nevada serving just over 29,000 students in a state with more than 450,000 students in public schools. According to a separate lawsuit that concentrates on funding issues, “very few of Nevada’s private schools are in the urban core of Nevada’s two largest cities, accessible to students in those neighborhoods.” This sounds plausible, since just 23 percent of those enrolled in the state’s private schools are minority students.

The scarcity of private options is conceded by ESA advocates, including Ladner. As he notes: “Few private schools exist in Nevada, and those that do mostly (and predictably) operate in areas with enough high-income families to support them”—a striking admission. He recommends focusing on the creation of new private school seats and on options outside of private schools.

Some of those nonschool options are appealing. I have a hard time objecting  to a program that gives parents funding for tutoring and technology, so long as the available resources are of sufficiently high quality (an especially pertinent question in Nevada’s vast rural areas). But parents may also need some help in sifting through vendor claims and judging the type and depth of services to purchase. For example, in a 2012 American Enterprise Institute paper examining No Child Left Behind’s “supplemental educational services” (SES) provisions, Carolyn J. Heinrich and Patricia Burch noted a “critical threshold” in terms of how much time a student spends in tutoring: “Below 40 hours [of total tutoring time] we do not identify any statistically significant effects of SES on students’ math and reading gains.” They also found that online providers (who charged more for their services) were less likely to produce learning gains, a finding that regrettably parallels research showing dismal performance of virtual schools in the tuition-free charter sector. Findings like these suggest a strong need for robust, accurate information, provided by a disinterested third party.

As for new private schools, they won’t just materialize automatically and may be of dubious quality. Hard experience in the charter sector teaches that you can’t hothouse good schools, and that even replicating successful ones takes skill. We’ve learned that you need serious review of operator applications, plenty of due diligence about their track record, and a good long on-ramp to ensure a successful opening.

Nevada’s ESA program provides none of these guardrails, relying instead on two assurances. One is found in Section 11 of the enabling legislation, which requires that schools and other vendors likely to receive more than $50,000 in ESA funds annually obtain a surety bond. This requirement will help the state avoid financial losses if a school goes belly-up, but it is no defense against shoddy operators, since such bonds are not exactly hard to come by. (Check surety company websites and you’ll see repeated variations on “Bad Credit? No Problem!”)

The act also requires that participating schools be accredited. But the provision only applies to private schools that are licensed by the state, and religious schools are exempt, creating a rather considerable loophole.

Of course, quality will depend mainly on parents making sound choices. But where one would hope for plentiful public information to help parents understand the performance of participating schools, Section 12 of the act requires only that the department publish aggregated results, sliced by grades and income levels, and that it conduct a survey
of parent satisfaction with the ESA program, not the schools.

Finally, the act allows the state treasurer to deny participation in the program to any entity that routinely fails to comply with the law, or fails “to provide any educational services required by law to a child receiving instruction.” Stern-sounding but vague, these provisions don’t actually provide a clear course of action when a school is doing a lousy job. While the ASD and other authorizers like Nevada’s State Public Charter Schools Authority create contracts with clear performance expectations, the ESA program provides no apparent standards for judging whether public funds are buying strong outcomes.

Are Student Interests Protected?

Schools taken under the ASD umbrella will be open to all students currently enrolled and any others who satisfy geographic requirements. The state’s charter schools are similarly open to all. Private schools, on the other hand, do not share this obligation to openness, and nothing in the program’s enabling legislation directly addresses discrimination, further narrowing the chances that choice will be realized for those who need it most.

While the U.S. Supreme Court has ruled since 1976 that private schools cannot deny admission based on race, some states have upheld the right of private schools to expel students because of their own or their parents’ sexual orientation. Nor are private schools required to enroll students with disabilities. According to the U.S. Department of Education, students with disabilities “do not have an individual entitlement to services they would receive if they were enrolled in a public school. Instead, the [local school district] is required to spend a proportionate amount of IDEA [Individuals with Disabilities Education Act] federal funds to provide equitable services to this group of children. Therefore, it is possible that some parentally placed children with disabilities will not receive any services while others will.” A similar approach is taken for English language learners; district help is available if a private school chooses to enroll the student in the first place.

It must be possible to navigate the trade-offs between the public and private spheres when seeking to provide benefits of private schooling through the use of public funds. The Nevada program doesn’t make much of an attempt.

I wish it were possible to applaud Nevada’s Education Savings Accounts without reservation. The program is a bold stroke, and it has broken school choice out of the tiny, marginal voucher programs seen in other states. With some modifications, the program could have a vast and important impact without the unintended side effects noted here. Taking greater cognizance of income would not only direct the benefits of ESAs to where they are truly needed, but would also reduce the risks of overburdening existing private schools and of encouraging charlatans to open new ones simply to collect public monies. If the state is successful in its current court appeal and the ESA program moves forward, I would love to see the kind of robust Yelp-like parent evaluations Ladner envisions—so long as they’re paired with strong oversight representing the public interest of all the taxpayers whose kids aren’t attending ESA-financed schools.

Until such changes are made, the best way of getting help to high-need students in Nevada is through the two charter-based channels: a modest quickening of the pace at which the ASD takes in troubled schools for turnaround and a concentrated effort to attract high-caliber talent that can expand capacity and enhance performance in the state’s public-charter sector.

Nelson Smith is an education policy consultant and senior advisor to the National Association of Charter School Authorizers.

This is part of a forum on education savings accounts. For an alternate take, please see “The Next Step in School Choice,” by Matthew Ladner.

Projections and additional analysis for each state are available here.

---

Last year, Congress passed the Every Student Succeeds Act, supplanting No Child Left Behind and placing responsibility for public school improvement squarely upon each of the 50 states. With the federal government’s role in school accountability sharply diminished, it now falls to state and local governments to take decisive action.

Large economic benefits should accrue to states that take advantage of this new flexibility. When students learn more in school, they remain in the educational system longer and become more-skilled and -effective participants in the state’s workforce. While some graduates will migrate to other parts of the country, a majority will join the labor market in their own states, thus contributing directly to their economic strength. Over the long run, each state stands to receive an extraordinary rate of return on successful efforts to improve school quality.

Even though most education policy debates have focused on school quality and student achievement, most research on the economic impact of schooling has focused narrowly on the number of years students remain in the educational system. This metric is not an adequate measure of student achievement and thus not a reliable indicator of economic impacts: it hardly matters how long one sits at a school desk if one learns little while occupying that seat. Recently, mounting evidence has suggested  that measures of individual cognitive skills that incorporate dimensions of test-score performance provide much better indicators of economic outcomes—while also aligning the research with the policy deliberations. The importance of including direct measures of achievement is especially apparent when looking at differences in economic growth across states.

In this essay, we document the long-term economic impact of a state’s student-achievement levels, which in turn permits us to calculate the economic returns from school improvement. First, we show that in the 40 years between 1970 and 2010, the spread among the states in their per-capita gross domestic product (GDP) widened considerably. Next, we show that the level of student achievement is a strong predictor of the state’s growth rate in GDP per capita over that time period, even after accounting for both the standard measure of school attainment and other economic factors. Finally, we project for each state the large positive impact that improvements in student achievement would have on a state’s GDP (See Figure 1).

Any state political leader of vision would do well to make school quality a high priority.

The Wealth of States

States vary sharply in the size of their per-capita GDP, that is, the total value of goods and services produced within a given year divided by the number of residents. In 2010, the wealthiest state, Delaware, enjoyed a per-capita GDP that was twice that of the poorest state, Mississippi (see Figure 2). Geographical and historical factors account for some of this variation among the states, but the discrepancies have grown over time. Importantly, the rate at which state GDPs have increased differs widely: for example, the per-capita GDP of North Dakota, the most rapidly growing state between 1970 and 2010, increased annually at a rate of 3.0 percent, while Nevada’s rate of increase was just 1.2 percent, the least of any state during this time period.

The spread among the states has remained wide and grown in absolute terms over the past several decades. In 1970 the spread between the 10th and the 40th state in the distribution was about $5,000 per capita in 2005 dollars, but by 2010 it had increased to nearly $12,000 per capita.

Knowledge Capital vs. School Attainment

Economists have long used the term human capital to refer to the skills individuals possess that have economic value and that pay off in the labor market. But their near-ubiquitous reliance on school attainment to measure individual skill differences has made years of schooling virtually synonymous with human capital. That measure of human capital, however, implicitly assumes that each additional year of schooling translates into a comparable increment in the stock of relevant skills, totally ignoring any variations in the quality of the student’s home, community, school, teachers, and other factors.

Here, we combine the quantity of schooling with a measure of cognitive skills in order to develop a more complete understanding of differences in individuals’ labor-market skills. In the aggregate, we call this broader measure the knowledge capital of states in order to distinguish it sharply from school attainment, or conventionally measured human capital. We rely upon math test scores from the National Assessment of Educational Progress (NAEP) and various international tests to provide data on the cognitive skills of each state’s adult workers.

Using these estimates, we then consider the impact of knowledge capital on the growth in a state’s GDP. In that way, we can estimate the impact of knowledge capital on a state’s wealth and can explain, at least in part, the divergent growth rates in GDP per capita among the states between 1970 and 2010.

Developing state-by-state measures of knowledge capital requires some effort. If we are to obtain an unbiased estimate of the achievement levels of a state’s adult workers, we cannot simply calculate the test scores of students currently attending the state’s schools. Many workers have migrated from a different state, and still others have immigrated to the United States from abroad; both of these groups will tend to differ in their cognitive skills from those who remain in a state after finishing their education. The degree to which state workforces consist of migrants from other parts of the United States is illustrated in Figure 3a, and the impact of foreign immigration is shown in Figure 3b. In 2010, less than 60 percent of adults living in the median state were also born in that state. The range across states varies from less than 20 percent (Nevada) to almost 80 percent (Louisiana). The share of adults not born in the United States in 2010 ranges from just 1 percent (West Virginia) to almost 30 percent (California).

To obtain our estimate of the student achievement component of the knowledge capital of a state at any point in time, we use census data to trace workers back to the place in which they were born. With that information, we can obtain a good estimate of the achievement of migrants from various states, because, on average, 86 percent of children age 14 or younger attend school in their state of birth. To estimate the achievement of workers born in the United States, we use mathematics test scores on the NAEP for 8th graders by birth state between 1990 and 2011. For workers born and educated outside of the United States, we use mathematics scores from the Program for International Student Assessment (PISA) and the Trends in International Mathematics and Science Study (TIMSS) conducted between 1995 and 2011.

We know, however, that both state migration patterns and the skills of interstate migrants are likely to differ depending on people’s educational background, so we estimate NAEP scores separately for workers with different levels of educational attainment. For example, we assume that we can assign to a college-educated individual born in Massachusetts (but possibly living elsewhere) the average test score of students with college-educated parents in Massachusetts. The achievement levels of international immigrants educated abroad are assumed to be the same as those of students performing at the 90th percentile of the distribution in their home country. We make this assumption because studies have shown that a country’s emigrants to the United States tend to be among its most talented people. (In a separate analysis, we modify this assumption to account for the less-selective nature of Mexican immigration into the United States; these results differ little from the ones reported here.)

Knowledge Capital and Economic Growth

To estimate how knowledge capital relates to the growth in a state’s GDP, we correlate the rate of GDP growth from 1970 to 2010 with our measures of the average knowledge capital of the state’s workers (based on the state’s workforce in 1970, the beginning of our growth period). Simultaneously, we adjust for the influence of three other factors that are usually hypothesized to be related to growth rates: the initial (1970) values of the level of GDP per capita, of physical capital per worker, and of the average number of years of schooling.

Figure 4 reveals a strong relationship between the achievement component of the knowledge capital of a state’s adult workers and economic growth in that state. The cluster of states in the lower left-hand corner of the graph—Alabama, Mississippi, Utah, Nevada—have suffered from both low math achievement levels on the part of their workforces and disappointing rates of economic growth. Those in the upper right-hand corner—North Dakota, South Dakota, Minnesota, Texas, Massachusetts, and Virginia—have enjoyed both significantly higher levels of math achievement and higher rates of economic growth.

The connection between the two variables—achievement levels and economic growth—is not perfect, of course. Given the levels of achievement of workers in Kentucky, Maine, Vermont, and Montana, these states should have enjoyed higher rates of economic growth. Conversely, the economies of Connecticut, Maryland, Virginia, and Louisiana have performed better than expected, given achievement levels. But, overall, our results suggest that achievement levels that are 1 standard deviation higher—for example, having the average worker in a state achieve at the 69th percentile rather than at the 31st percentile of the overall distribution of cognitive skills—yield an average annual growth rate that is 1.4 percentage points higher.

Some may question whether this correlation actually reflects a causal relationship. One could argue that students simply learn more when their state is performing well economically, perhaps because growth generates additional resources that can be spent on education or because students are more motivated to learn when prosperity is close at hand. We are not persuaded by these arguments, in part because of the very weak correlation between increased spending on schools and higher levels of student achievement. Furthermore, the cross-state results are virtually identical to previous results from international research, and extensive analysis of the cross-country evidence has shown that a causal interpretation of the relationships is credible.

To test the credibility of our results further, we also undertook a standard accounting exercise used by economists to determine how much of the total variation in economic performance among states at any point in time can be attributed to differences in a specific factor. In particular, we use existing research about how much a high level of achievement boosts the earnings of an individual worker, combined with our new measures of the average achievement levels of workers in each state, to gauge the contribution of differences in achievement to differences in income levels across states. And we perform a parallel analysis to shed light on the role played by differences in average years of schooling.

The results of this exercise again suggest the importance of knowledge capital for state economic prosperity. We find that differences in achievement and attainment account for 20 to 35 percent of the current variation in per-capita GDP among states, with average years of schooling and achievement levels making roughly even contributions. In a sense, this estimate is surprisingly large, because both labor and capital are free to move across states—and thus tend to equalize rewards to workers with different skills. But our results are quite consistent with those obtained from similar analyses of the role of student-achievement levels in explaining differences in economic performance across countries (see “Education and Economic Growth,” research, Spring 2008).

Gains from Educational Improvement

The fact that the achievement level of a state’s workers is a key driver of its economic performance suggests that the gains from improved school quality could be substantial. Just how large would they be? We consider a range of improvements in student achievement and estimate the economic impact for each of the 50 states and for the nation as a whole. The various scenarios include:

1) Bringing every state up to the best state in the United States (Minnesota)
2) Bringing every state up to the best state in its region
3) Bringing all students in each state up to the NAEP basic achievement level
4) Bringing each state up to the best state, but assuming others do not make any gains at all, thereby isolating  the direct impact of a state’s efforts.

The calculations of the economic impact are straightforward. First, we estimate the expected growth of a state’s economy if the current skill level of workers were to remain unchanged. Then we compare this growth path to the one that would be achieved with better schools (and subsequently improved skill levels). The gains in GDP are discounted (at 3 percent per year), so that near-term gains are given more weight than gains in the more distant future. The resulting present values of income gains can be compared directly to current state GDP levels.

Our projections account for the fact that improvement in worker skills is not instantaneous. First, we assume that education reforms take 10 years to be fully effective, with student skills improving steadily over that time. Second, the labor force improves only as new, more-skilled students replace retiring less-skilled workers. We assume that 2.5 percent of the labor force retires each year and that these workers are replaced by better-educated ones, implying that the labor force does not fully reach its ultimate new skill level for 50 years 
(10 years of reform followed by 40 years of retirements).

Figure 5 displays the economic gains from each reform scenario for the United States as a whole over the expected lifetime of a person born today (80 years), expressed in trillions of 2015 dollars. If all states improved their schools to the point where average student achievement matched that of the top state, Minnesota, the overall gains would be $76 trillion, or more than four times the current GDP of the United States. An alternative way to view this is that the nation would, on average, see a 9 percent higher level of GDP across the next 80 years. Such an increase is easily large enough to allow even the most cash-strapped state to meet current demands for public services while maintaining a balanced budget. In 2095, the GDP would be more than 36 percent larger than would be seen without school-quality improvements.

The projected economic impact of the school-improvement reforms varies considerably across the country, according to differences in the current economic position and knowledge-capital stock of each state. For example, Figure 6 shows the gains in economic outcomes that result when all states are brought up to the skill level of top-performing Minnesota (Scenario 1). This improvement means the least in North Dakota and Massachusetts, whose students are currently very close to that level, and the most in Alabama and Mississippi, where achievement levels are lowest. If California’s students could perform at the same level as Minnesota’s, the benefits to the state would exceed $16 trillion, assuming other states reached the Minnesota level. Even in North Dakota and Massachusetts, the current value of gains over the next 80 years would amount to 70 percent of current state GDP.

We find somewhat smaller gains from having each state meet the achievement level of the best state in its region (Scenario 2). This growth is necessarily less than that of the first scenario, because the achievement levels of the regional leaders vary widely. Nonetheless, the aggregate gains from Scenario 2 still have a present value of more than $35 trillion, almost twice the nation’s current GDP.

Scenario 3 essentially projects the results of realizing the achievement goals of NCLB—getting all students to a basic level of academic proficiency—but by the year 2025. The gains from all states getting students to the NAEP basic achievement level are roughly twice current GDP, or about the same as for Scenario 2.

The results for Scenario 4 represent what happens if one state acts on its own to improve school quality while all other states do not. This is an important perspective to consider, since no state that commits to a path of reform can necessarily expect others to join in, even though that would be desirable. In any given state, some of the students who profit from the improved quality of its schools will move out of the state. While the better-educated out-migrants will boost the economy of their new states, their native states will experience a brain drain.

So, what if a single state improves but others do not? Will it still benefit? Figure 5 shows that the single-state improvement strategy (summed up across all states) yields a gain of $46 trillion. When we compare this present value to that of Scenario 1, where all states move to perform at the level of the best state, we see that joint action yields gains that are 65 percent larger than the gains that would accrue to each state acting on its own. That is, aggregate rewards are smaller if any state acts without comparable efforts by others; at the same time, even the gains of acting independently are substantial.

Summing Up

Clearly, the United States stands to reap enormous economic gains from improving its schools. The goals for boosting student achievement considered in the separate scenarios of this paper are within the feasible range for most states. The largest gains would come from a coordinated improvement in performance, since states are all linked by flows of people over time. But even if states act individually, they can promote a better economic future for their residents through education reform. The gains projected here not only make the residents of each state better-off but also show how states’ fiscal problems can be tackled when knowledge capital increases.

A key feature of this analysis is that we built in realistic patterns of movements of the labor force across U.S. states and of the dynamics of school improvement. Simply put, raising the achievement of today’s students has no immediate impact on a state’s economy, because these students are not yet in the labor force. But as the skills of today’s students improve, the skills of tomorrow’s workers advance as well. Realizing these gains does require a sustained commitment on the part of a state’s political leaders. But such commitment to better schools has already given rise to dramatic gains in the United States (for instance, in Massachusetts) and abroad (as in South Korea). If we are to achieve prolonged economic growth in our nation, we have little choice but to strengthen the skills of our people.

Eric A. Hanushek is senior fellow at the Hoover Institution at Stanford University and research associate at the National Bureau of Economic Research. Jens Ruhose is an economist at Leibniz University Hannover. Ludger Woessmann is professor of economics at the University of Munich and director of the Ifo Center for the Economics of Education.

Predicting how the U.S. Supreme Court will rule based on oral arguments is a risky business, and that reality came into high relief after Justice Antonin Scalia’s unexpected passing in February. After the January 11 arguments in Friedrichs v. California Teachers Association, a majority of the justices were clearly poised to overturn a 38-year-old mistake and eliminate one of the most cherished powers of teachers unions—the authority to confiscate money from nonmembers. But Scalia’s death led to a split decision, leaving the union’s power intact, at least for now, and raising the stakes in this year’s presidential election.

In 1977, the court held in Abood v. Detroit Board of Education that public employees could not be compelled to join a union but could be forced to pay “agency” fees, AKA “fair-share” fees, to help cover costs associated with collective bargaining. Unless all workers were required to pay, union leaders argued, many nonmembers would become “free riders” who didn’t contribute their fair share.

For teachers unions, Abood has been
 a financial windfall. In states that allow
agency fees, more than 90 percent of teachers join a union, while only 68 percent join in states that don’t. Since agency fees cost a teacher nearly as much as union dues, many see little reason not to join the union and get full membership benefits. As well, unions impose opt-out policies requiring teachers to “affirmatively decline” every year to support the political activities of the union, which allows them to request a partial refund of their dues.

In 2013, California teacher Rebecca Friedrichs and eight others filed suit contending that the mandatory fees violated their rights to freedom of speech and association. To expedite the case, they asked the Ninth Circuit Court to rule in favor of the union in hopes of sending the case straight to the Supreme Court. Under existing precedents, the plaintiffs pointed out, the union’s case would probably hold, but new rulings had called those precedents into question, making the supreme Court the only one that could adjudicate the teachers’ claims. The Ninth Circuit agreed, setting up a January showdown before the Supreme Court.

At oral argument, the plaintiffs seemed to carry the day. Roberts, Scalia, and Alito pointed out that the union’s position rests on a false assumption: that one can draw a clear line between the political and nonpolitical activities of public-sector unions. Scalia contended that one cannot, that “everything that is collectively bargained with the government is within the political sphere.” The only way for unions to credibly claim they represent the interests of all teachers is to assume that all teachers have the same preferences. But the very existence of Friedrichs shows that to be false. Justice Kennedy said that the “union is basically making these teachers compelled riders for issues on which they strongly disagree.” In short, it’s not a free ride if you never wanted the ride. It’s more like being clubbed in the head, tied up, and thrown in the union trunk.

Notably, the best evidence for the weakness of the union’s position came from the court’s liberal bloc of Breyer, Sotomayor, Kagan, and Ginsburg, who made little effort to dispute the plaintiff’s First Amendment arguments. Instead, those justices mainly argued that Abood should be upheld because the case “was [decided] forty years ago” and overturning such a “deeply entrenched” precedent would be unsettling. Fundamental rights, however, do not normally yield to that kind of utilitarian calculus, and hiding behind stare decisis looked like weakness, not strength.

After oral argument, Friedrichs was set to become one of the most significant cases of this term, but Scalia’s death threw the outcome into doubt and gave the union hope. That hope was confirmed in March, when the court issued a per curiam decision, saying, “The judgment is affirmed by an equally divided Court.”

Despite this decision, the future of Friedrichs remains uncertain. The court’s judgment leaves the Ninth Circuit ruling in place, but it has no precedential value. Once the court has a full complement of members, the plaintiffs will ask the court to rehear the case. Since Scalia’s replacement is unlikely to be confirmed until after the election, it looks like the new president will be the one who determines the final fate of agency fees and teachers’ First Amendment rights.

Joshua Dunn is associate professor of political science at the University of Colorado-Colorado Springs.

Some of the most dramatic gains in urban education have come from school districts using what’s known as a “portfolio strategy.” Under this approach, districts negotiate performance agreements with public schools—traditional, charter, and hybrid models. The arrangement affords school leaders substantial autonomy to handcraft their schools to fit the needs of their students. Districts give parents choices among the schools while working to replicate successful schools and replace failing ones.

Many doubt such a strategy is possible with an elected board, because closing schools and laying off teachers triggers fierce resistance. Most cities pursuing the portfolio strategy, including New Orleans, Washington, D.C., and Camden, New Jersey, have done so with insulation from local electoral politics. In New Orleans, the state board of education and its Recovery School District (RSD) oversee most of the schools; Congress created the appointed D.C. Public Charter School Board; and in Camden the state is in charge.

All of which explains why reformers are paying close attention to Denver, Colorado. With an elected board, Denver Public Schools (DPS) has embraced charter schools and created innovation schools, which it treats somewhat like charters. Since 2005 it has closed or replaced 48 schools and opened more than 70, the majority of them charters. In 2010 it signed a Collaboration Compact committing to equitable funding and a common enrollment system for charters and traditional schools, plus replication of the most effective schools, whether charter or traditional.

Of the 223 DPS schools today, 55 are charters, which educate 18 percent of its students, and 38 are innovation schools, which educate 19 percent (see Figure 1). Soon DPS will take the next step, creating an Innovation Zone with an independent, nonprofit board, which will negotiate a performance contract with the district. Beginning with four innovation schools but able to expand, the zone could for the first time give district schools the autonomy charters enjoy.

For years, Denver’s reforms stirred controversy. When the board closed or replaced failing schools, protests erupted and board meetings dragged into the wee hours. During most of current superintendent Tom Boasberg’s first five years, he had only a 4–3 majority on the board. But the strategy has produced steady results: a decade ago, Denver had the lowest rates of academic growth among Colorado’s medium and large districts; for the last three years it has ranked at the top. Voters have responded by electing a board with a 7‒0 majority for reform.

Denver’s U-Turn

In 2005, DPS was floundering. Out of 98,000 seats, 31,000 were empty, and many school buildings were half full. Almost 16,000 Denver students had left DPS for private or suburban schools. A financial crisis loomed, in the form of pension contributions the district could not afford. When Superintendent Jerry Wartgow retired in 2005, the Denver Board of Education chose Michael Bennet, chief of staff for then mayor John Hickenlooper, to replace him. Bennet had no background in public education, but he had spent time turning around failing companies for a local investment firm.

A few reforms were already underway: Wartgow had negotiated a pay-for-performance system, called ProComp, with the teachers union, and he was reconstituting 13 elementary and middle schools. He had built support for DPS among business and community leaders. Several foundations were pushing for reform; African American and Latino leaders were engaged; and a 27-member Commission on Secondary School Reform, appointed by the school board, had submitted reform recommendations.

“There was a consensus that we had to do something,” says David Greenberg, who founded the Denver School of Science and Technology (DSST), which has since grown into the city’s most successful charter network. “But there was no consensus about what.”

Bennet knew he had to lure students back from other districts to stave off financial ruin. He considered the charter sector, which in 2005 had only 17 schools and served just 7 percent of the district’s students, too small to make a difference. He and the board initially chose to centralize control over curriculum, budgets, hiring, and almost everything else.

DPS was so dysfunctional, Bennet concluded, that he could not fix it without significant outside pressure. So he asked several foundation leaders to create an organization of civic leaders, chaired by two former mayors, to push for change and support the board when it promoted reform. They called the initiative A+ Denver, and it has championed the portfolio strategy, along with the Piton, Donnell-Kay, and Gates Family foundations.

In April 2007, a study by the Piton Foundation and the Rocky Mountain News revealed just how many students were leaving Denver for private schools and other districts. Bennet and the board responded with a call for dramatic change. “It is hard to admit,” they wrote, “but it is abundantly clear that we will fail the vast majority of children in Denver if we try to run our schools the same old way.” The district should “no longer function as a one-size-fits-all, centralized, industrial age enterprise making choices that schools, principals, teachers, and most, most important, parents are in a much better position to make for themselves.” Instead, it should “function more like a partner, building capacity and leadership at the school level and serving as an incubator for innovation.”

To help that process along, Bennet in 2008 shifted from traditional budgeting to a weighted, student-based budgeting system, under which about 56 percent of operating money follows students to their chosen schools (see sidebar). This increased competitive pressures on schools, because losing students meant losing money.

DPS also unveiled a School Performance Framework (SPF) that measured test scores, academic growth, student engagement, enrollment rates, and parental satisfaction. Using the Colorado Growth Model, it gave far more weight to academic growth than to current proficiency levels—triple the weight at the elementary level, double in secondary schools. Every school wound up with a score that summarized its performance, and charter schools quickly dominated the top-ten lists.

That spring the Rocky Mountain News splashed a full-page photo across its cover of the first graduating class from DSST, a charter school. Every one of its graduates had gained admission to college—the first time that had happened in a Denver school with many low-income students. Bill Kurtz, founding head of the school, believes that was a turning point. “When I came to Denver,” he says, “there was a mindset that not all kids can go to college, that your income and race would determine that.” But DSST’s accomplishments gave “the leadership of the district an understanding that what was thought impossible was possible.”

With the SPF in place to ensure accountability and charters proving that autonomy worked, Bennet and the board switched to a strategy they called “Performance Empowerment.” It endorsed moving more decisionmaking to the school level as principals proved their schools could perform.

They also decided to replicate DSST and other strong charters as fast as they could. Given the teachers union’s opposition, however, they didn’t trumpet the strategy. They kept their message simple: they would replace failing schools with better schools, regardless of their type.

An astute politician, Bennet also worked with two community organizations, soliciting their views and support. The group Together Colorado is a multiracial, multifaith coalition of more than 60 congregations and clergy, schools, and youth committees, affiliated with the national PICO network. Padres & Jóvenes Unidos is an organization of Latino activists that dates to the 1970s. Both work on a variety of issues but were instrumental in supporting Bennet’s reforms.

“They really inoculated the district from having the kind of blowback that other districts have had from low-income communities of color,” says Van Schoales, CEO of A+ Denver. “It made it harder for the traditional factions—they lost some of the potential opposition to a lot of these reforms.”

It is easier to open and close schools when your student population is rising, and Denver’s was growing rapidly. DPS enrollment has increased 25 percent since 2005, driven by population growth, residential development on a closed airport and military base, expanded preschool programs (available to most four-year-olds and a few three-year-olds), and students returning from neighboring districts. In October 2015, DPS reported 91,429 students, up from 73,018 in 2005. (Note that, unlike those in Figure 1, these numbers include pre-kindergarten students.) Just over 56 percent are Hispanic; 23 percent are white; 14 percent are African American; 70 percent qualify for free or reduced-price lunch; and 38 percent are English language learners.

In early 2009, Governor Bill Ritter appointed Bennet to fill the U.S. Senate seat of Ken Salazar, the new secretary of the interior. Bennet urged board members to appoint his deputy and lifelong friend, Tom Boasberg, as superintendent, and they quickly agreed. Though Boasberg embraced the portfolio strategy, he also eschewed the words, preferring “an intentional strategy to say we are going to focus on great schools as opposed to political arguments about governance structures.”

Winning the Political Battle

In 2009, the district opened eight new schools and planned to open seven more for 2010. By this time the Denver Classroom Teachers Association was on alert. It backed a slate of board candidates that fall and won a majority of open seats, and one of its supporters prepared to take the board presidency. But the union had been a bit careless in vetting Nate Easley, an African American who grew up in Denver but had recently returned from Washington, D.C., to help lead the Denver Scholarship Foundation. Easley surprised everyone by embracing reform, and—being the swing vote—he was elected board president. Suddenly the union’s 4‒3 board majority had reversed, triggering a bitter divide that lasted for four years.

Tensions came to a head in the fall of 2010, when the board decided to replace a group of schools in the far northeast area of the city, including Montbello High School, with 10 innovation schools and a handful of charters. “That was when the controversy really got enormous,” says Mary Seawell, a board member at the time. “The scale and scope was like nothing the district had ever done before, and there were so many schools impacted. It was a highly charged, emotional political process,” with people screaming at community and board meetings, which often lasted until well past midnight.

The antireform block warned Easley that if he voted for the replacement strategy, they would recall him. He had been a straight-A student at Montbello High, but at Colorado State College he had tested into remedial classes—a devastating blow, as he describes it. So he understood exactly how Montbello was failing its students. He voted for the changes, and his opponents launched a recall effort—but failed to secure the required number of signatures.

In the fall 2011 elections, reformers managed to preserve their 4‒3 majority. Two years later, Democrats for Education Reform and its allies raised significant money and recruited a former lieutenant governor, a former city council president, and a former chairman of the Denver Democratic Party to run. Finally, they broke the logjam. With six reformers, the new board initiated a turnaround strategy in southwest Denver and approved another major expansion of DSST, which will educate one-quarter of all middle and high schoolers by 2025. In 2015, a reformer won the final seat.

Delivering Results

The reformers won in part because they had more money and better-known candidates, and in part because their approach has yielded results. In 2005‒06, 11.1 percent of DPS students dropped out each year, and in 2006‒07 less than 39 percent graduated on time. By 2014‒15, only 4.5 percent dropped out each year, while 65 percent graduated on time, including 72 percent of those who entered DPS high schools and stayed for four years. (All data include charter schools.)

In the decade ending in 2014, the percentage of students scoring at or above grade level in reading, writing, and math increased from 33 to 48, far faster than the state average (see Figure 2). In 2015, Colorado switched to the PARCC tests, so comparisons to previous years are no longer possible. But Denver schools appear to have adjusted far better to the more demanding, Common Core‒aligned PARCC tests than schools in the rest of the state. In 2014, Denver students outperformed only 16.7 percent of Colorado students on the elementary English language arts test, but in 2015 they outperformed 42.4 percent of their peers statewide. In elementary math, Denver jumped from the 19th percentile to the 49th, almost reaching the state median. Middle schools were even stronger: in English, they jumped from outperforming 17.5 to 51.4 percent of their peers, and in math from the 39th to the 65th percentile, far above the state median. These results should be interpreted cautiously, however, as roughly 1 in 10 Colorado students opted out of the spring 2015 PARCC test.

DPS has more than doubled the number of students taking and passing Advanced Placement courses, and African American students now take advanced math classes at the same rate as whites, while Hispanics lag only 1 percentage point behind. Only 48 percent of DPS graduates enrolled in college in 2014, but 1 in 7 low-income students in Denver did so compared to 1 in 20 in the rest of the state.

On the other hand, Denver’s steady improvement has widened the achievement gap, something that happens in many urban districts that improve, as white and middle-class students raise their scores faster than poor and minority students. In 2014, the gap between the percentage of low-income and non-low-income students who tested at grade level was almost 40 points across all subjects, and the gap between African Americans and Latinos, on the one hand, and whites, on the other, was 42 points.

Charter Schools Lead the Way

Because Denver’s two largest and most successful charter networks, DSST and STRIVE Prep, started with high schools and middle schools, respectively, the city’s charters are unusually concentrated at the secondary level. And though DPS has clearly improved its elementary schools, at the secondary level charters account for almost all the academic growth. In a 2014 study published by the Donnell-Kay Foundation, which used SPF data through 2013, author Alexander Ooms concluded,

“The decision to close poorly-performing schools of all types appears to be paying dividends and is especially encouraging for low-income students. Likewise the decision to encourage replication of the best charter schools has clearly led to positive results. But the district’s attempts to open its own new schools, and particularly to improve its continuing schools serving secondary grades, have yielded remarkably little.”

A year later, an analysis of 2014 SPF scores revealed little change. Six of the top eight schools were charters. And though only 25 percent of DPS schools are charters, they account for 40 percent of the most sought-after schools for Kindergarten, 6th grade, and 9th grade. Of the 30 schools on the top-ten lists for these transition grades, 12 are charter schools.

A study of test scores from 2010 through 2014, by economists at the Massachusetts Institute of Technology and Duke University, found that Denver’s charters produced “remarkably large gains in math,” large gains in writing, and smaller but statistically significant gains in reading, compared to DPS-operated schools. The gains in math were the equivalent of closing almost half the yawning gap between white and black students in the U.S.

Do charters perform better because they attract better students? They do have advantages: 100 percent of their families make an active choice to enroll; their students arrive with slightly higher test scores; and they don’t have to accept new students after the school year begins. On the other hand, charters get 19 percent less money per student than district-operated schools, according to one analysis. Though the district strives for equity, charter teachers are not eligible for ProComp bonuses, which average $7,396 for a second-year teacher. Nor do most Denver charters get district-funded transportation for students.

Charters’ success does not appear to stem from an easier mix of students. Charters enroll almost as high a percentage of special education students as DPS-operated schools do—10 vs. 11 percent. But charters serve 3 percentage points more low-income students (those who qualify for free and reduced-price lunch) and 10 percentage points more English language learners.

Perhaps the most reasonable way to compare charters and DPS-operated schools is to analyze school test scores and percentages of low-income students together, on the same scatter plot. Using 2015 test-score data and comparing schools with similar percentages of low-income kids, charters outperform DPS-operated schools at the middle and high school level but not at the elementary level, where there are only 10 charters.

Innovation Schools Struggle for Autonomy

One of Michael Bennet’s first moves, back in 2005, was to recruit Brad Jupp, a union leader who had helped negotiate performance pay, to be his senior policy advisor. A former teacher, Jupp was convinced that DPS principals needed more autonomy to improve their schools. In the fall of 2006, he and Bennet decided to create something like Boston’s Pilot Schools—in-district “Beacon Schools.” They negotiated a memo of understanding with the teachers union, then asked teachers and principals to make proposals. We offered “greater resources, the opportunity to have a new school design, and a bit of autonomy,” Jupp says. Their offer generated considerable excitement and 24 proposals.

But Beacon Schools quickly bumped into the limits imposed by the district and the teachers’ contract. Frustrated leaders at one school proposed a novel use of a waiver clause in the contract, to waive everything but the provisions that permitted union membership and representation. The board approved the waiver—plus two more, for other schools.

State Senate president Peter Groff, from northeast Denver, seized on the idea, and with help from Bennet’s staff and other reformers, drafted a statewide Innovation Schools Act. Passed in 2008, it invited proposals for creating innovation schools, which could request waivers to district policies, state statutes, and union contracts—including tenure for new teachers—if 60 percent of the teachers voted for the Innovation Plan.

In the early years, the district rushed innovation schools into place without much attention to their design. Not surprisingly, the results were disappointing. But for the past few years, DPS has treated the innovation school authorization process much like the charter authorization process, and new innovation schools have looked far more like charters—with a year to plan, clear visions and strategies, and careful hiring of teachers.

Some innovation schools have made remarkable progress, but as a group, they have not performed nearly as well as charter schools on standardized tests, according to two separate studies. Kelly Kovacic, the DPS executive director of portfolio management, acknowledges that innovation schools have not bent the curve on performance.

Three important differences between innovation schools and charters may contribute to their different success rates. First, charters are often closed if they fail to meet their performance targets. So far no failing innovation school has been closed. The board adopted a policy last December, however, to apply exactly the same standards and process to closing all schools—charter, traditional, and innovation.

Second, successful charters usually replicate, while innovation schools have yet to do so. That, too, is about to change: two successful innovation schools will open new campuses next fall. Unless they fall flat, Boasberg told me, he intends to continue the practice.

Third, innovation schools have more autonomy from district mandates than traditional DPS schools but far less than charters. Some of the innovation school principals I interviewed were happy with the degree of autonomy they enjoyed, but others were quite frustrated. In part, it depended on whom they dealt with at the district level: their instructional superintendent, their HR partner, their budget partner, and so on.

“It’s infuriating to innovation school principals, because they feel like they have the blessing of the top leadership, but it’s like cutting through frozen molasses,” says Alan Gottlieb, who has been following education in Denver for more than 20 years, as a journalist, a foundation executive, and the founder of an online education magazine. “Any little thing—hiring somebody, getting a school bus, ordering new furniture—it’s all impossible. And it is because everybody below the top level is operating as though they’re still just working for a traditional school district.”

The issue is not so much outright restrictions as the constant bureaucratic battles principals must endure. Zach Rahn, who runs Ashley Elementary, offers the example of a district initiative to create teacher leaders, who teach half time and coach other teachers half time. It has been an “unbelievably huge benefit to our school communities,” he says. “But now as they seek to bring it to the whole district, they’ve put all these strings attached to it that actually take away from it. They sent us a 42-slide PowerPoint on how we need to organize the ecosystem in our school.” It included a rubric to rate the school’s readiness, a survey to fill out, and essays to write on why the teachers they chose were the most qualified—though Ashley was already using teacher leaders.

Rahn and some of his peers routinely ignore such mandates, then have to waste precious hours fighting the resulting battles. The time they lose troubles him, Rahn says, but “I also think about my colleagues around the district, who are just as capable as I am, and they aren’t getting that option” to refuse.

This is the biggest reason innovation schools have not performed as well as charters, he believes. “Hold me accountable to whatever levels you want, but I need to be able to lead, and do my job, and not be stuck in the weeds down here.”

Rahn and three other innovation school principals proposed the Innovation Zone. In December 2015, the Board of Education endorsed the idea, though details are still being worked out. If done well, this initiative could be a real breakthrough, giving some innovation schools the same autonomy that charters enjoy, along with a board to shield them from district mandates and politics.

The struggle over autonomy is part of an overall lack of alignment within the DPS bureaucracy. The district has had a “strategic plan” since 2005, now called The Denver Plan 2020. But it is primarily a set of goals, with less emphasis on strategies. Because they did not want to give the opposition a big target, Bennet, Boasberg, and their allies on the board chose never to use the phrase “portfolio strategy” and never to broadcast their intention to increase the number of charter schools.

This silence has been successful, politically. In contrast, when Newark superintendent Cami Anderson presented her strategies in her “One Newark” plan, it gave her opponents a big, fat target, and she was gone within two years. But the price of Denver’s success has been frustration on the part of principals, who have to deal with central office staffers who don’t all share Boasberg’s vision.

The lack of alignment has also led to inconsistencies in hiring central office and school leaders. Finding leaders who can turn around schools full of low-income kids is usually the toughest challenge, and in Denver those making the choices have not always been on the same page. If DPS leaders more clearly articulated their overall strategies, their employees might better understand their priorities.

Is Denver’s Strategy Sustainable?

Boasberg and the board deserve credit for putting in place many of the elements of a portfolio strategy. They have embraced charter schools and learned a great deal from their success, bringing many charter practices into district schools. The SchoolChoice enrollment process has increased access to good schools and made it more difficult for schools to cream the best students. The district is moving more special-education centers for extremely disabled students into charters, correcting an imbalance.

DPS staff are also working to fix the SPF’s big flaw, its overreliance on academic growth. Since the Colorado Growth Model compares students only to those who had similar test scores in the past, a student can show “high growth” by gaining five months of learning a year if the comparison group is only gaining four months. So schools can look successful, even while their students are falling further behind grade level every year. In the future, growth will only outweigh proficiency by 3:2 at the elementary level, though the ratio will remain 2:1 at the secondary level.

Boasberg is on a six-month sabbatical, but when he returns in the summer of 2016, he is likely to enjoy support from a majority of the board for quite a few years. The opposition is weak and disorganized, and all the momentum is on the side of the reformers. If anything, some on the board are frustrated that Boasberg is not moving faster.

It is hard to see what might derail the portfolio strategy, even if the bureaucracy continues to slow it down. Denver has proven, for a decade now, that charter schools offer a more effective model of urban education. It is about to launch an Innovation Zone, which—if done properly—will give some district schools the autonomy and accountability that make charters so effective.

Within a decade, the district could reach a tipping point, where a majority of public school families choose charters or innovation schools. When that happens, the reforms will be difficult to undo. And Denver will be well on the road to proving that an elected board can transform a 20th-century system organized on the principles of bureaucracy into a 21st-century system built to deliver continuous improvement.

David Osborne, co-author of Reinventing Government and other books on public sector reform, is director of the project on Reinventing America’s Schools at the Progressive Policy Institute. A longer treatment of this subject will be available at www.ppionline.org.