The past decade may rank as one of the worst in the history of American education. It marks a stark reversal from what was once a hopeful story. At the start of the century, American students registered steady improvement in math and reading. Around 2013, this progress began to stall out, and then to backslide dramatically. What exactly went wrong? The decline began well before the pandemic, so COVID-era disruptions alone cannot explain it. Smartphones and social media probably account for some of the drop. But there’s another explanation, albeit one that progressives in particular seem reluctant to countenance: a pervasive refusal to hold children to high standards.We are now seeing what the lost decade in American education has wrought. By some measures, American students have regressed to a level not seen in 25 years or more. Test scores from NAEP, short for the National Assessment of Educational Progress, released this year show that 33 percent of eighth graders are reading at a level that is “below basic”—meaning that they struggle to follow the order of events in a passage or to even summarize its main idea. That is the highest share of students unable to meaningfully read since 1992. Among fourth graders, 40 percent are below basic in reading, the highest share since 2000. In 2024, the average score on the ACT, a popular college-admissions standardized test that is graded on a scale of 1 to 36, was 19.4—the worst average performance since the test was redesigned in 1990.[Read: The teen-disengagement crisis]American schoolchildren have given up almost all of the gains they achieved at the start of the century. These learning losses are not distributed equally. Across grades and subjects, the NAEP results show that the top tenth of students are doing roughly as well as they always have, whereas those at the bottom are doing worse. From 2000 to 2007, the bottom tenth of fourth graders in reading ability showed substantial improvement, before stagnating. But by 2024, those gains had been erased. In 49 out of the 50 states (all except Mississippi), the gap between the top tenth and the bottom tenth grew. Nat Malkus, of the American Enterprise Institute, has pointed out that this surging inequality has grown faster in America than in other developed countries. The upshot is grim: The bottom tenth of 13-year-olds, according to NAEP’s long-term-trend data, are hitting lows in reading and math scores not seen since these tests began in 1971 and 1978, respectively.A seemingly plausible culprit, and a familiar boogeyman for progressives, is insufficient spending. The problem with this tidy explanation is that it’s not tethered to reality. School spending did not decline from 2012 to 2022. In fact, it increased significantly, even after adjusting for inflation, from $14,000 a student to more than $16,000.Besides, America recently ran a very large natural experiment in dropping money on schools that, in a word, failed. During the pandemic, Congress appropriated a gargantuan sum of money, $190 billion, to ameliorate learning loss, most of it as part of the Biden administration’s American Rescue Plan. (For scale, this is roughly the sum recently given to the Trump administration to fund its border wall and immigration-enforcement agenda.) States were given latitude to spend their funds as they saw fit, which, it seems, was a mistake. Instead of funding high-quality tutoring programs or other programs that benefited students, districts spent money for professional development or on capital expenditures such as replacing HVAC systems and obtaining electric buses. “The scientific term for this is that we didn’t get jack shit out of that money,” says Michael Petrilli, the president of the Thomas B. Fordham Institute, an education-policy think tank. “There are some studies that can detect small impacts, but they’re small. I think it’s also fair to say that a lot of the money was wasted.”A more likely culprit for learning loss is smartphones. Jonathan Haidt, the social psychologist and author of The Anxious Generation, is the most prominent evangelist of this thesis. He argues that declining school performance and other worrying trends among Gen Z, such as the rise in anxiety, depression, and suicide, can be traced to the new “phone-based childhood.” And his argument matches the time trend well. Smartphone ownership rocketed upwards around the time that American educational performance crested: In 2011, just 23 percent of teenagers had smartphones. By 2013—roughly the peak of American education—37 percent did. By 2015, 73 percent had access to one. And by 2018, that figure was 95 percent, where it remains today. Nearly half of teenagers say that they use the internet almost constantly. For parents, this explanation is also intuitive. You can apply your own experience of smartphone-induced self-sabotage to children (who do not have the biological benefit of a mature prefrontal cortex) and conclude that unregulated phone use is destructive to learning and creativity.[Gail Cornwall: What many parents miss about the phones-in-schools debate]But the smartphone thesis has a few weak spots. It’s not just middle schoolers and high schoolers whose performance is declining; it’s also kids in elementary school. Phone use has certainly increased among young children, but not to the ubiquitous proportions of adolescents. And even though smartphone use is almost universal, the learning losses have not been. High-achieving kids are doing roughly as well as they always have, while those at the bottom are seeing rapid losses. The thesis needs some elaboration to explain this dispersion pattern. Perhaps kids who have higher levels of executive functioning and impulse control (or are lucky enough to have parents who do) are better able to navigate the sea of distractions. At any rate, few broad social trends—whether the decline of marriage in America or the slow rate of productivity growth in Europe—are monocausal. It would be surprising if the decline in American education were.An explanation that deserves equal consideration is what one might call the low-expectations theory. In short, schools have demanded less and less from students—who have responded, predictably, by giving less and less. The timing lines up here, too. Around the same time that smartphones were taking off, a counterrevolution was brewing against the old regime of No Child Left Behind, the George W. Bush–era law passed in 2002 that required schools to set high standards and measured school progress toward them through stringent testing requirements. Bush famously said that he wanted to tackle “the soft bigotry of low expectations,” and there’s real evidence that he did. As controversial as it was, No Child Left Behind coincided with increased school performance, especially for those at the bottom.That’s not to say the regime was perfect. The No Child Left Behind approach to struggling schools was largely punitive, including threats of mandatory restructuring for institutions that failed to meet targets. And expectations for progress rose higher and higher each year, ultimately seeding the demise of the law. Schools were supposed to have all their kids at grade level by 2014. But as this deadline approached, it became clear that schools would miss it. In 2012, the Obama administration began giving states waivers from the requirements. Then, in 2015, Congress passed the Every Student Succeeds Act, which returned responsibility for improving low-performing schools to the states. But according to Martin West, the academic dean of Harvard’s education school, “most states have not been particularly ambitious in the design of those systems.”Low-expectations theory explains other trends that the smartphone thesis, by itself, does not. If the bar for grading and graduating were constant year over year, we would expect both to decline in line with student performance. Instead, we see the opposite. An ACT study found that the share of students getting A’s in English rose from 48 percent in 2012 to 56 percent in 2022, even as their demonstrated mastery of the subject declined over that period. (The same is true of other subjects, including math, social studies, and science.) Over the same decade, high-school graduation rates improved from 80 to 87 percent despite objective declines in academic achievement.[Rose Horowitch: The perverse consequences of the easy A]If the incentives to learn decrease, children—just like adults—will respond to that. One in four students today is chronically absent, meaning that they miss more than a tenth of instructional days, a substantial increase from pre-pandemic averages. The past decade also marked a shift in concern among educators, toward equity and away from excellence. Elements of so-called equitable grading, which is supposed to be more resistant to bias than traditional grading, have taken off in American schools. Roughly 40 percent of middle-school teachers work in schools where there are no late penalties for coursework, no zeroes for missing coursework, and unlimited redos of tests.What would it take to reverse America’s educational declines? In good part because of Haidt’s arguments that smartphones are both dulling and immiserating children, states are now instituting bans on smartphone use during the school day. If districts that ban smartphones see swifter improvements in academic outcomes than those that do not, that will provide solid evidence that Haidt was correct. But getting screens out of the classroom likely won’t be enough to escape the malaise of the past decades. What lower expectations have inflicted in the past, only higher expectations in the future can remedy.The experience of a few outlier states gives reason for optimism. Matthew Chingos and Kristin Blagg, two scholars at the Urban Institute, computed “demographically adjusted NAEP scores,” examining how effective states are at educating kids after accounting for significant differences in socioeconomic status. Their analysis of the 2024 NAEP results found that Mississippi was best at educating kids in fourth-grade math, fourth-grade reading, and eighth-grade math. (In 2013, Mississippi was at the bottom of the unadjusted league table.) When I computed the correlation between these demographically adjusted scores and state spending, I found that there was none. If you’re an underprivileged kid in America, you will, on average, get the best education not in rich Massachusetts but in poor Mississippi, where per-pupil spending is half as high.This is a recent phenomenon. Some have called it the “Mississippi miracle” or—if you include relative outperformance in states such as Alabama, Louisiana, and Tennessee—the “southern surge.” From 2013 to 2024, reading performance declined among fourth graders in 46 out of 50 states. In only two states, Mississippi and Louisiana, did they meaningfully improve.A clear policy story is behind these improvements: imposing high standards while also giving schools the resources they needed to meet them. In 2013, Mississippi enacted a law requiring that third graders pass a literacy exam to be promoted to the next grade. It didn’t just issue a mandate, though; it began screening kids for reading deficiencies, training instructors in how to teach reading better (by, among other things, emphasizing phonics), and hiring literacy coaches to work in the lowest-performing schools. Louisiana’s improvements came about after a similar policy cocktail was administered, starting in 2021. And this outperformance might continue in the future: The state recently reported that the number of kindergartners reading at grade level more than doubled in the past academic year—rising from 28 percent to 61 percent.The “Mississippi miracle” should force a reckoning in less successful states and, ideally, a good deal of imitation. But for Democrats, who pride themselves on belonging to the party of education, these results may be awkward to process. Not only are the southern states that are registering the greatest improvements in learning run by Republicans, but also their teachers are among the least unionized in the country. And these red states are leaning into phonics-based, “science of reading” approaches to teaching literacy, while Democratic-run states such as New York, New Jersey, and Illinois have been painfully slow to adopt them, in some cases hanging on to other pedagogical approaches with little evidentiary basis. “The same people who are absolutely outraged about what” Robert F. Kennedy Jr. “is doing on vaccines are untroubled by just ignoring science when it comes to literacy,” Andrew Rotherham, a co-founder of the education-focused nonprofit Bellwether, told me.Some promising educational reforms, moreover, seem to brush up uncomfortably against liberal political priors. Progressive Democrats, for instance, still regard charter schools with suspicion and tend to fight to cap their number. But in a lot of places, that only hinders the equity these people profess to care about: High-performing charter networks in American cities have registered serious improvements in learning for some of the most disadvantaged children in the country. These have been verified through several lottery studies, comparing students who got into those schools with those who didn’t based on random chance alone, which is the gold standard for policy research. Another evidence-supported reform that upsets teachers’ unions, and their partners in the Democratic Party, is merit-based pay. We could “move to a system where teachers are rewarded based on their performance, not just a simple salary matrix, especially early in their careers,” says Jim Wyckoff, an education-policy professor at the University of Virginia, citing success with the policy in Washington, D.C.The economic costs already incurred by declining academic achievement are immense. Eric Hanushek, an education economist at the Hoover Institution, calculated that recent students will earn 7.7 percent less over their lifetime than they would have had they graduated at the time of peak educational performance. And because learning lost today means forgoing growth for decades in the future, Hanushek calculates that GDP will be 6 percent lower for the remainder of the century than if scores had stayed level. (This adds up to the modest sum of $90 trillion in present-day dollars).[Listen: Bring back high-stakes school testing]One optimistic theory is that artificial-intelligence tools, which will only grow more powerful over the coming decades, will correct for this economic catastrophe by letting everyone externalize their thinking to superintelligent computer programs. The once-ironclad relationship between schooling quality and earnings might break down just in time, a somewhat literal deus ex machina. Hanushek thinks that is too rosy, though. In fact, the opposite might occur: “If we look at all the inventions in the past,” he told me, “they’re complementary to the high-skilled people and substitutes for low-skilled people.”In 1983, after another sustained decline in academic performance, a government commission released a landmark report titled “A Nation at Risk.” The authors argued that “the educational foundations of our society are presently being eroded by a rising tide of mediocrity that threatens our very future as a Nation and a people,” because America had “squandered the gains in student achievement made in the wake of the Sputnik challenge.” You could make a similar argument today as great-power competition between America and China intensifies.America’s scientific and technological hegemony is being seriously challenged, and China already leads in industries such as electric-vehicle production and solar-cell manufacturing. In the industries where America still leads, much technical prowess is owed to immigration policies that have attracted the brightest and most ambitious from around the world and to the research universities that train them. The Trump administration is pursuing a policy of browbeating these universities and of restricting visas, including for high-skilled workers—turning away talent amid an international talent war. The idea is that students in America today, and not those educated elsewhere, will be the labor force holding up the economy. That bet—like America’s students—may be mathematically unsound.