Every gained minute comes from somewhere. The 28 extra minutes of sleep, the 8 minutes of cooking, the 17 minutes of gaming — none of these appeared from thin air. They were subtracted from other parts of American life. Reading for pleasure lost 5 minutes. Socializing lost 12. Shopping in physical stores lost 5. Social events lost 3. Together with smaller declines in a dozen minor activities, these subtractions tell the story of what Americans traded away for the new time allocation. Some of these trades are neutral. Some are alarming. This is the ledger.
The 24-hour day is a zero-sum game. When one activity grows, another must shrink. Over the past eight episodes, we’ve tracked the winners: sleep gained 28 minutes, cooking gained 8, gaming gained 17, phone calls gained 3. That’s 56 minutes of new claims on the day. Where did they come from?
The biggest donor was work, which shed 32 minutes (Episode 3). Beyond work, the losses come from a cluster of activities that share a common thread: they all involve leaving the house, engaging with other people, or focusing attention on something that isn’t a screen. Reading. Shopping. Socializing. Social events. Volunteering. Together, these “community activities” surrendered about 25 minutes. Add a few more from scattered smaller categories, and the books balance.
In 2003, the average American spent 22 minutes per day reading for personal interest. By 2024, that had fallen to 17 minutes — a 22% decline. The numbers are small in absolute terms, but the decline is culturally significant. Reading was already a minority activity in 2003; now it’s becoming residual. The sharpest drop came between 2005 and 2010, from 0.38 to 0.30 hours, coinciding precisely with the smartphone’s arrival as a mass-market device.
The loss was not evenly distributed. Retirees — historically the heaviest readers — went from 59 minutes of daily reading in 2003 to just 35 minutes in 2024, a stunning loss of 24 minutes (−40%). This single group accounts for the bulk of the national decline. When a 70-year-old picks up an iPad instead of a novel, the time-use survey records it. The shift from physical books to tablets and e-readers means that some “reading” has migrated to devices that the survey may classify differently, but the magnitude of the decline suggests a genuine loss of long-form reading, not just a format change.
Young adults (15–24) barely read in either period: 8 minutes in 2003, 8 minutes in 2024. The generation that never developed the habit couldn’t lose it. The gender gap narrowed as women’s reading fell faster: women went from 25 to 19 minutes (−24%), men from 19 to 15 (−22%). Both converged downward.
Americans spent 24 minutes per day on consumer goods purchases (non-grocery shopping) in 2003. By 2024, that had fallen to 19 minutes — a 23% decline. Grocery shopping held steady at 6–7 minutes, virtually unchanged. The decline is almost entirely in discretionary retail.
The cause is no mystery: e-commerce. When you buy on Amazon, the time cost rounds to zero in a diary survey. When you drive to the mall, park, browse, try on clothes, wait in line, and drive home, it registers as 30–60 minutes. Every purchase that migrated online erased time from the ATUS. But the data captures something beyond convenience. Shopping in stores was never just transactional — it was social. The mall was a destination, a meeting place, a form of entertainment. Its decline mirrors the broader retreat from public spaces documented in Episode 6.
The trend was steady and relentless: 0.40 hours in 2003 → 0.37 in 2010 → 0.35 in 2019 → 0.31 in 2024. Even the pandemic didn’t create a sharp break here; the decline simply continued at its existing pace. Shopping time is still falling, suggesting e-commerce hasn’t finished displacing physical retail.
| Activity | 2003 | 2019 | 2024 | Δ (min) | Δ (%) |
|---|---|---|---|---|---|
| What We Lost | |||||
| Socializing & communicating | 0.78 | 0.64 | 0.59 | −11 | −24% |
| Reading for pleasure | 0.36 | 0.27 | 0.28 | −5 | −22% |
| Consumer goods shopping | 0.40 | 0.35 | 0.31 | −5 | −23% |
| Social events | 0.10 | 0.06 | 0.05 | −3 | −50% |
| Volunteering | 0.14 | 0.13 | 0.13 | −1 | −7% |
| Work (incl. commute) | 3.68 | 3.47 | 3.43 | −15 | −7% |
| What We Gained | |||||
| Sleep | 8.57 | 8.84 | 9.04 | +28 | +5% |
| Gaming & computer leisure | 0.29 | 0.43 | 0.57 | +17 | +97% |
| Household activities | 1.83 | 1.78 | 2.01 | +11 | +10% |
| Telephone calls | 0.08 | 0.11 | 0.13 | +3 | +63% |
| What Held Steady | |||||
| Television watching | 2.58 | 2.75 | 2.60 | +1 | +1% |
| Sports & exercise | 0.29 | 0.31 | 0.31 | +1 | +7% |
| Religious activities | 0.14 | 0.14 | 0.14 | — | — |
| Grocery shopping | 0.11 | 0.10 | 0.11 | — | — |
Not everything declined. A few activities held their ground or even grew, and their survival is just as revealing as the losses.
Exercise and sports participation barely moved: 0.29 hours in 2003, 0.31 in 2024. In a landscape of collapsing social and cultural activities, this stability is remarkable. Americans aren’t exercising less despite doing less of almost everything else outside the home. The fitness industry — from boutique gyms to Peloton to running culture — has created sufficient cultural gravity to hold its share of the day. Exercise may also be one of the few activities that successfully migrated to the home (home gyms, fitness apps) without losing time-use share.
Television is the other great survivor. As we documented in Episode 5, TV watching is essentially flat at 2.60 hours. Streaming replaced cable, but the total screen time barely budged. TV has proven to be the most durable leisure activity in the American repertoire — immune to cultural shifts that cratered reading, socializing, and physical shopping.
Religious and spiritual activities held at 0.14 hours overall, though this masks turbulence underneath: a dip to 0.11 in 2021 and 2023, suggesting that church attendance is volatile year-to-year and may be in secular decline that the annual averages partially obscure. Still, religion has proved more resilient in the time data than in the polling data on religious affiliation, which shows sharper drops.
Grocery shopping — unlike discretionary retail — barely changed (0.11 to 0.11 hours). People still go to the store for food. Instacart and grocery delivery haven’t displaced the physical trip the way Amazon displaced the mall trip, perhaps because choosing produce and proteins remains a sensory activity that resists digitization.
The reading decline deserves a closer look because of who lost the most. In 2003, Americans over 65 read for 59 minutes per day — nearly an hour, and more than triple the national average. By 2024, that had fallen to 35 minutes. Retirees lost almost as much reading time as the entire national average for reading itself. The shift likely reflects the tablet revolution: an e-reader or iPad offers news, social media, puzzles, and video alongside books, and each of those alternatives chips away at dedicated reading time.
Working-age adults (25–34) read just 10 minutes in 2024, up slightly from 10 minutes in 2003 — essentially unchanged and near the floor. Young adults (15–24) read 8 minutes in both periods. For these groups, reading was never a major time use, and it still isn’t. The national decline is predominantly a story about older Americans abandoning a habit that younger generations never adopted.
| Reading by Group | 2003 | 2010 | 2019 | 2024 | Change |
|---|---|---|---|---|---|
| All persons | 22 min | 18 min | 16 min | 17 min | −22% |
| Men | 19 min | 15 min | 14 min | 15 min | −22% |
| Women | 25 min | 21 min | 19 min | 19 min | −24% |
| Age 15–24 | 8 min | 8 min | 7 min | 8 min | — |
| Age 25–34 | 10 min | 8 min | 7 min | 10 min | — |
| Age 65+ | 59 min | 47 min | 37 min | 35 min | −40% |
The 24-hour day is a closed system. Every minute gained somewhere was lost somewhere else. The American time reallocation of 2003–2024 follows a clear pattern: activities that require leaving the house, engaging with others in person, or sustaining focused attention on non-screen media all declined. Activities that can be done at home, alone, or on a screen all grew.
Reading, socializing, in-store shopping, social events, and volunteering collectively lost about 25 minutes. Sleep, gaming, cooking, and phone calls gained about 59 minutes, with the balance coming from reduced work time. The result is a population that is more rested, better fed (home-cooked), and better entertained — but more isolated, less literate, less civically engaged, and less likely to run into a neighbor at the store. Whether that trade is net positive or net negative depends on what you believe a society needs to function. The data just measures the exchange rate.