How America Stopped Reading

In 1982, the heaviest readers in America were its twenty-somethings. Four decades later, they were its retirees. Somewhere in between, the young and the old traded places, and the federal survey that caught it barely made the news.

That reversal is the loose thread in a story we think we already know. The story goes: phones killed reading, kids don't read anymore, the end. It isn't wrong, exactly. But it's lazy, and when you actually pull the receipts (a century of census literacy counts, four decades of federal arts surveys, the longest-running study of American teenagers on record), the real picture turns out to be stranger, and more damning, than "kids these days."

So we spent a while with the primary sources and put the numbers next to each other. What follows is what they actually say, chart by chart, with every figure linked back to where it came from. Fair warning: the most comfortable version of this story, the one where reading is simply dying, is the version the data supports least. What's really happening is, in a way, worse. It's also more fixable.

01 - The triumphFirst, we taught everyone to read

Start with the good news, because it's genuinely spectacular. For most of American history, a large share of the country simply could not read. In 1870 the Census counted one in five adults as illiterate1, and that single number hides a brutal fact: among Black Americans, most only a few years out of slavery, roughly four in five could not read or write.1

Then the country did something remarkable. It built schools, and it filled them. By 1900 adult illiteracy had fallen to about 11%. By 1940, under 3%. By 1979 it was six-tenths of one percent, so close to zero that the Census Bureau essentially stopped asking the question.1 The color line ran straight through the achievement (in 1900, 94% of white Americans were literate against 56% of everyone else),1 but the trajectory was unmistakable, and by the late twentieth century the gap in basic literacy had all but closed.

Chart 1 · The century we won

America learned to read

Share of Americans age 14+ able to read and write, 1870–1979. The dashed marker shows the 1900 racial gap the national average conceals.

Source: National Center for Education Statistics, 120 Years of Literacy. Basic literacy, self-reported; the series ends in 1979 because illiteracy fell below 1%.

Hold onto that shape, because it reframes everything that follows. The thing we spend the most breath worrying about (can people read?) was basically settled by the time the Beatles broke up. Whatever went wrong next is not a story about ability. It's a story about choice: not whether Americans can read, but whether they do.

02 - The quiet turnThen we stopped opening the books

In 1982 the federal government began asking a different question. Not "can you read?" but "did you?" The National Endowment for the Arts folded a reading module into its Survey of Public Participation in the Arts, and it has repeated it, more or less the same way, ever since. It is the closest thing America has to a forty-year electrocardiogram of the reading habit.

The line trends down. The share of adults who read any literature (a novel, a short story, a poem, a play) in the prior year fell from 56.9% in 1982 to 46.7% in 2002.4 In 2004 the NEA published the results under the deadpan-alarming title Reading at Risk, and for once a government report read like a warning flare.4 There was one genuine reprieve: a rebound to 50.2% in 2008, right before the smartphone arrived, which nobody has fully explained (the late-Harry-Potter, Oprah's-Book-Club years get some of the credit).6 Then the slide resumed: 47% in 2012,7 and by 2020 it was down near 40%.10

Widen the lens from "literature" to "any book at all, not for work or school," and the news gets blunter. That number sat comfortably above half for decades, and then, in 2022, it didn't: 48.5%.9 For the first time in the survey's history, fewer than half of American adults had read a single book in a year. Fiction specifically hit 37.6%, the lowest share ever recorded.9

Chart 2 · The forty-year slide

Reading participation, 1982–2022

Two measures from the same federal survey. Both drift down; in 2022 "read any book" slipped below half for the first time. Note the 2008 rebound: the last one before the smartphone.

Read literature (novel, story, poem, play) Read any book (not for work or school)

Source: National Endowment for the Arts, Survey of Public Participation in the Arts: Reading at Risk (2004), Reading on the Rise (2009), and the 2022 highlights. The two lines use different definitions and shouldn't be read as one continuous series.

We did not forget how to read. We just, increasingly, don't.

03 - The steelman"But reading's fine" - the case for calm

Before going any further, the case for the defense deserves its turn, because it's real, it's well-evidenced, and if you only read the doom-takes you'll never hear it.

Since 2011, Pew Research has asked Americans a plain question: did you read at least one book in the past year, in any format? The answer has barely moved: about three in four say yes, 79% then, 75% now.11 By that measure the patient is stable.

The book itself, meanwhile, has been changing shape rather than dying. Audiobook listening has more than doubled, from 11% of adults in 2011 to 26% in 2025, and e-books climbed from 17% to 31%.11 Print didn't collapse either. It's still how most people read.11 Among the people who do sit down with a book, the government's time-use data shows the average reading session actually got longer over the last twenty years, not shorter.14 Even among fifteen-year-olds, a third still call reading one of their favorite hobbies.15

Chart 3 · The book didn't die; the paper loosened its grip

How Americans read, by format

Share of US adults reading in each format in the past year. Print holds; audiobooks double. The container is changing faster than the contents.

Print E-book Audiobook

Source: Pew Research Center book-reading surveys, 2016 and 2026. A 2021→2025 change in survey method means the last step should be read gently.

So if you want to believe reading is fine, the data will hand you the evidence. We don't believe it, and the reason is that these reassuring averages hide two things they were never designed to show. One is frequency: whether reading is a daily habit or an annual event. The other is who. Split the population by age, and the calm numbers come apart in your hands.

04 - The tellThe young and the old traded places

This is the chart that changed how we think about the whole question. Go back to 1982. Reading was, unmistakably, a young person's habit: it peaked among 25-to-34-year-olds (62% read literature) and 18-to-24-year-olds (60%), then declined with every decade of age, bottoming out at 41% among the over-75s.4 The stereotype of the bookish kid and the TV-watching grandparent was, statistically, true.

By 2012 the curve had turned upside down. The oldest readers now read the most; the youngest adults had sunk toward the bottom of the pack.7 By the 2017 survey the inversion was total: 18-to-24-year-olds, at 47%, were flatly the least likely age group in America to have read a book, and the 65-to-74s, at 57%, the most.8 The NEA put it in the driest possible terms: young adults had gone "from the group most likely to read literature" to "the group least likely."5

Chart 4 · The reversal

Reading used to fall with age. Now it rises with it.

Share reading literature in the past year, by age group, 1982 vs 2012. Watch the lines cross: the youngest cohorts start at the top and end near the bottom.

Source: NEA SPPA, Reading at Risk (1982) and the 2012 highlights. Both use the same "literary reading" definition, so the comparison is like-for-like.

Why does one crossing pair of lines matter so much? Because reading is a habit set young. It's built in childhood bedrooms and teenage summers, and people mostly carry the setting they leave adolescence with for the rest of their lives. When each new cohort ages into adulthood reading less than the one before, you're not looking at a dip that will bounce back. You're looking at a decline with a fifty-year fuse already lit. The grandparents out-reading the grandchildren aren't a paradox: they're the last generation whose habit was formed before.

Reading didn't get old. Its readers did.

05 - The cliffWhat happened to the teenagers

To see what "before" and "after" mean, you need the teenagers, and here America has an unusually good witness. Since 1976, University of Michigan researchers have asked a nationally representative slice of high-school seniors, every single year, how often they read. The study is called Monitoring the Future, and the trend it has captured is one of the starkest in all of social science.

In the late 1970s, 60% of twelfth-graders said they read a book or magazine almost every day. By 2016 that was 16%.12 The mirror image is just as grim: the share of seniors who read zero books for pleasure in a year roughly tripled, from about one in nine in 1976 to one in three by 2016, then kept climbing to 41% by 2021.12

Now the part that keeps this from being another phone panic. This decline did not begin with the iPhone. Look closely and the twelfth-grade line is already falling in the early 1980s, long before anyone owned a smartphone.12 Reading has been losing an attention war for forty years, and its earlier conquerors have names too: broadcast television, then cable, then the shopping mall, then the open internet. The phone didn't start the fire.

But watch what the phone does to the slope. Trace the reading line past 2011, as smartphone ownership climbs from a third of adults to three-quarters in five years,13 and the decline doesn't level off. It steepens. The psychologist Jean Twenge, who has spent a career in this data, dates the sharpest break to around 2012, the year US smartphone ownership crossed 50%.12 By 2018 the OECD's global test of fifteen-year-olds found that half of them (and 55% of American teens) agreed with the sentence "I read only if I have to." More than a quarter said reading was a waste of time.15

Chart 5 · The teenage cliff

Daily teen reading fell as the smartphone rose

Share of US 12th-graders who read a book or magazine almost every day (orange), against the share of US adults who own a smartphone (blue). The habit was already sliding. Then it fell off a ledge.

Teens reading almost daily US adults owning a smartphone

Sources: Twenge, Martin & Spitzberg (2019), analyzing Monitoring the Future; smartphone ownership from the Pew Mobile Fact Sheet. Pre-2010 teen points are multi-year averages; the survey item counts magazines and newspapers alongside books.

06 - The mechanismNot fewer readers. Fewer minutes.

Now we can resolve the paradox from Chart 3: how "three in four still read a book" and "reading is collapsing" can both be true at once. The answer is that what's vanishing isn't readers. It's occasions.

The Bureau of Labor Statistics has timed how Americans actually spend their days, every year since 2003. Average time spent reading for pleasure has fallen from about 23 minutes a day in 2004 to 16 minutes in 2023, down roughly a third.14 More telling still: the share of Americans who read at all on a given day fell from 28% to 16%.14 Fewer people are reading on any given day, and the count of books bears it out: Gallup's running tally of books read per person hit a record low of 12.6 in 2021, the lowest in the three decades it has asked.3

And the age gap in raw minutes is almost comic. Americans aged 75 and over read about 46 minutes a day. Teenagers and young adults? Around 8.14

Chart 6 · The vanishing minutes

Time spent reading for pleasure is draining away

Average minutes per day Americans spend reading for personal interest, 2004–2025. Down about a third. (No 2020 figure: the pandemic disrupted collection.)

Source: Bureau of Labor Statistics, American Time Use Survey; trend corroborated by an ATUS study in iScience (2025). Population average across everyone, readers and non-readers alike.

Chart 7 · The generational gap, in minutes

The oldest Americans read five times more than the youngest

Average minutes per day reading for pleasure, by age, 2024.

Source: BLS American Time Use Survey, Table 11A (2024).

One caution, in the spirit of not overselling: the young were always the floor here. Even in 2004 they read only about 7 or 8 minutes a day, and the old-versus-young gap has, if anything, narrowed slightly, because older readers' minutes fell too.14 This isn't a Gen Z morality tale. The erosion is broad, it's across every age, and it's mostly about frequency: the number of days on which reading happens at all. The minutes reading used to own (the commute, the checkout queue, the doctor's waiting room, the last hour before sleep) got repriced and sold to the feed.

07 - The argumentSo what actually happened

Put the seven charts in a row and a single explanation does most of the work, and it isn't the one we reach for first.

It was never literacy. That war was won by 1979; essentially every American adult can read. It was never even desire, not really: three-quarters still read a book each year, a third of teenagers still call it a favorite thing, and the committed readers who remain read more deeply than ever. What collapsed is something more specific and less flattering to our self-image: attention. Reading is one of the most attention-hungry things a person can do, and for the first time in history it has to win that attention in open competition against products engineered by some of the best-funded talent on Earth, tuned against quarterly revenue targets, to capture exactly the idle minutes reading used to live in.

For most of the twentieth century the book had a structural advantage we never noticed: it was what you did when you were bored and stuck. Boredom was reading's distribution channel. The paperback rack sat by the checkout because the checkout was dead time, and a book was the best thing on offer. The feed's genius wasn't to out-write the book. It was to be there (in the pocket, always, asking nothing, rewarding instantly) in every one of those dead minutes, and to be just a little more immediately gratifying than a paragraph that makes you wait.

The feed didn't outwrite the book. It outwaited it.

The age reversal is the fingerprint on the weapon. The first generation to grow up entirely inside the feed is the first in the recorded history of these surveys to read less than its own grandparents. That is not a coincidence, and it is not a story about young people being lazier or dumber than their elders. Hand a generation an infinite, frictionless, always-available alternative during precisely the years when the reading habit is supposed to set, and the habit doesn't set. That is the whole mechanism.

08 - The conclusionThe good news is also the point

The doom version gets one thing exactly backwards. If this were a literacy problem, or a problem of desire, it would be close to hopeless: you cannot quickly re-educate or re-inspire a whole country. But it isn't either of those. It's an environment problem. And the environment is the one thing you can actually change.

The wanting is still there: the surveys are unanimous on that. What's missing is the moment: the small, defended pocket of time in which reading used to happen by default and now doesn't, because something else got there first and never leaves. You don't fix that by scolding people into virtue, or by pretending the last forty years didn't happen. You fix it the way you'd fix any habit that lost its cue: you give the cue back. Reclaim the queue, the commute, the last hour of the night, and the reading tends to return, because the desire never actually left the building.

We spent a century learning to read. We can spend a decade remembering to.