Paid to Quit Facebook: What a Month Away Changed

Ask people to imagine giving up Facebook for a month and most can guess how it would go: a little relief, maybe some boredom, nothing dramatic enough to try. A team of economists tested the real thing instead, paying Facebook users real money to deactivate for four weeks and tracking what actually changed.

What the researchers did

The study, by Hunt Allcott, Luca Braghieri, Sarah Eichmeyer and Matthew Gentzkow, was published in the American Economic Review in 2020. The team recruited participants with ads on Facebook itself, offering $30 in gift cards for a study on internet browsing. Nearly 1.9 million people saw the ads. After screening, which kept only people who used Facebook for 15 to 600 minutes a day, and a baseline survey, 2,743 US users were asked how much they would need to be paid to deactivate Facebook for four weeks, running up to and just past the November 2018 midterm elections.

The 61 percent whose price came in under $102 were randomly split into two groups: a Treatment group paid $102 to deactivate, and a Control group that kept using Facebook as usual. Everyone also earned $5 per survey. That left 1,661 people in the experiment itself. Compliance was checked by pinging each person's public Facebook page, which goes dead once an account is deactivated. Fewer than 2 percent dropped out before the final survey, and more than 90 percent of the Treatment group complied.

What they found

• Deactivating freed up about 60 minutes a day. The Treatment group didn't move that time to other social media or other online activity, both of which fell too. Instead they watched more television alone and spent more time with friends and family in person.
• They knew less about the news. News consumption fell 15 percent overall, and an index of news knowledge dropped 0.19 standard deviations, with fewer correct answers to factual questions about recent events. Voter turnout and other political engagement showed no detectable change.
• Political polarization fell. An index of seven measures of polarization, from views on policy issues to exposure to news that favors one's own side, dropped 0.16 standard deviations, although how people felt about the opposing party didn't change significantly on its own.
• Well-being improved, modestly. An index of ten well-being measures, including happiness, life satisfaction, loneliness, depression and anxiety, rose 0.09 standard deviations. The authors put that at about a quarter to two-fifths of the effect of psychological interventions such as self-help therapy, group training and individual therapy. Daily text-message check-ins on mood moved in the same direction but weren't statistically significant on their own.
• They kept using Facebook less. Weeks after the experiment, the Treatment group's Facebook use was still 11 minutes a day (22 percent) lower than the Control group's, and 5 percent were still deactivated nine weeks on. About 80 percent said deactivating had been good for them.
• Facebook still had real value to them. Weighted to resemble the average US Facebook user, the price to give up Facebook for four weeks had a median of $100 and a mean of $180, which across all US users implies about $31 billion for a single month. But after going without it, people lowered their asking price for another four weeks by up to 14 percent, which the authors read as a sign that standard valuations may overstate what Facebook is worth to its users.

What the study can't tell you

The sample wasn't representative of the typical Facebook user. Participants were younger, more educated and more politically left-leaning on average, and the study only included people who already used Facebook between 15 and 600 minutes a day. People willing to sign up for a study that might pay them to quit Facebook may differ in other ways too.

The setting was specific: one platform, one country, a closely watched midterm election. The authors note that a longer deactivation, a quieter political season, or many people deactivating at the same time could change the results. Most outcomes came from self-reports, which opens the door to participants telling researchers what they think they want to hear. The authors ran several tests for this and concluded it was unlikely to be a major source of bias.

What it means for your phone

Even a short, defined break from one platform freed up real time, and most of it went to people in the room, not another screen. If you get your news through social media, the trade-off matters: a month off meant knowing less about current events, not just scrolling less. Lining up another news source before you go, a couple of outlets you read directly, keeps you informed without the feed.

If a full deactivation feels like too much, a daily limit in Unrotten offers a gentler version of the same idea: the Less scrolling goal locks the apps you choose with Apple's Screen Time shield once your time for the day is used up, until the day resets. For a longer-term plan, read our guide to digital minimalism, or go further with a full social media detox.