Blaming Tech for Teen Troubles

Jonathan Haidt’s clever, insufficient case against smartphones.


The Anxious Generation: How the Great Rewiring of Childhood Is Causing an Epidemic of Mental Illness, by Jonathan Haidt, Penguin Press, 400 pages, $30

Jonathan Haidt opens The Anxious Generation with what is supposed to be an analogy for kids' use of smartphones and social media: Would you let your child travel to Mars, he asks, if some Silicon Valley CEO said it was safe? It's an absurd comparison: Whatever harms may or may not befall minors with iPhones, they're light-years less apparent, substantial, or universal than those facing kids shuttling through outer space to a desert planet with an atmosphere mainly made of carbon dioxide.

Happily, most of this volume is far less hysterical than that opening might lead you to believe. Yes, this book is filled with unwarranted pessimism, unjustified conclusions, and unsavory solutions. But as he lays out his case that a "phone-based childhood" is replacing a "play-based childhood," Haidt makes many points that even the most ardent opponents of tech panic and state intervention should be able to appreciate. Unfortunately, he can't keep the spirit of that opening analogy from periodically seeping back in.


Haidt, a social psychologist at New York University, believes that young people's rising use of screens and rising rates of emotional fragility both stem from our overprotection of kids in nondigital spaces. He rails against policies that punish parents for letting children have some independence, and against the mindset that tries to shield the young from every possible emotional harm. Parts of the book were even written with Free-Range Kids author (and regular Reason contributor) Lenore Skenazy, with whom Haidt helped found Let Grow, a nonprofit that pushes back against helicopter parenting.

Haidt also stresses that many of his ideas for curbing teen tech use shouldn't be legislated, though he does favor a number of new regulations too. He recognizes the guidelines he offers parents are not one-size-fits-all—that activities OK for most minors might be problematic for some and that things that are generally bad at a certain stage of development may be OK for individual kids.

He also marshalls a lot of data as he tries to tie the recent rises in youthful depression, anxiety, suicide, and learning loss to phone-based childhoods. But here we run into trouble. Haidt seems convinced there is one right way to interpret this data: the way that implicates technology.

Pointing to Jean Twenge's research, for example, Haidt reports that "teens who spend more time using social media are more likely to suffer from depression, anxiety, and other disorders, while teens who spend more time with groups of young people (such as playing team sports or participating in religious activities) have better mental health." But what does this really tell us? Not that these platforms are causing these problems. That's possible. But it's also possible that teenagers use more social media and avoid more group activities because of their depression or social anxiety—or that some third factor triggers both mental health difficulties and problematic internet use.

Haidt points to data showing rates of depression and anxiety in young people rose at the same time that smartphone and social media adoption skyrocketed among the same age group. But this was also a time period in which awareness of those conditions increased and the stigma around them decreased. And it was a period in which identity politics and a sort of victimhood cache began to permeate places where many young people gravitated online. If phones and social media play a role here, it might not be as a direct trigger—perhaps the veneration of various psychological diagnoses in certain online spaces led more kids to embrace those labels.

Haidt responds to the awareness/destigmatization theories by pointing out that we have been seeing more suicides and self harm in young people. Kids aren't just saying they're struggling, he argues; they're taking action that shows it. But the suicide picture is complicated. Yes, U.S. suicides have been rising. But the rate among older adults—the group least likely to be heavy social media users—has also increased sharply and in fact is much higher than for the youngest cohort. (In 2021, the rate for Americans ages 65 and up was 17.3 suicides for every 100,000 people. For 12- to 17-year-olds, it was 6.5 per 100,000.) In 2022, suicide rates actually went down among younger Americans while increasing among men over 34 and women over 24. (Depression, too, is up across American age groups, even preceding the pandemic.)

It's also notable that youth suicide rates in the U.S. started rising before the advent of smartphones and social media (something Haidt acknowledges) and that their rise over time has not been linear. (The same goes for the general population, with the 2022 suicide death toll about equal to what it was in 1950.) It's also much higher in some states than others, something we'd be less likely to see if a nationwide phenomenon like social media was a leading cause. The rise has not been consistent around the world, even in countries with similar phone and social media adoption patterns. In many places with widespread phone use—France and Russia, for example—teen and young adult suicide rates have fallen, sometimes precipitously.


Haidt argues that no single thing other than "the great rewiring of childhood" can explain the breadth of data suggesting young people are faring worse than before on an array of measures. Maybe no single thing can do this. Yet for each data point, a variety of nontech explanations could work.

For instance: Haidt cites a 2023 survey in which 68 percent of college students said they felt anxious at least half of the time. Social media could be a culprit, but so could doom coming from professional media and gloom coming from politicians. So could the anxiety emanating from adults around them. So could shifting sexual norms, fears of another pandemic, or a lifetime of what Haidt calls "modern overprotective parenting."

Or take the rise in 12- to 17-year-olds who answered national survey questions in ways suggesting they've suffered a major depressive episode. This rise started in the early 2010s and continued throughout the decade (from 8 percent in 2010 to 11.4 percent in 2014, 14.4 percent in 2018, and 19.5 percent in 2022)—a time period that corresponds with increasing smartphone adoption and the introduction of platforms like Instagram, Snapchat, and TikTok. But this was also a time of economic downturn, a pandemic, unrest in the streets, and extreme turbulence in U.S. politics. Surely some of this could also explain self-reports of hopelessness, sadness, etc. Maybe other health issues, such as rising obesity rates, had a role. Maybe shifts in academic expectations played a part. Or maybe the depression isn't related to events that took place during this period at all; it could stem from something in the way kids coming of age during this period were raised in the preceding decade, or even from genetic factors. Plenty of counterevidence and rival interpretations are available.

That doesn't mean we should simply throw up our hands and declare unlimited screen time to be fine at any age. Parents who look at the evidence may deem it wise to adopt some or all of Haidt's agenda—measures like limiting phone or laptop use to certain times of day, or waiting until your kids are a certain age before letting them own a phone or participate on social media. Tech companies could offer more options—including device-based options—for parents who want to prevent their kids from creating social media accounts or visiting certain sites. Individual schools or school districts could experiment with tough-on-phone policies. Of course, there's the other half of Haidt's prescription: more time for unsupervised play, more play at school, and more autonomy in general for kids.

But the data just aren't strong enough to unequivocally back up Haidt's claims about what smartphones and social media are doing to kids, which seems like a good reason to hold off on government mandates, and not just out of first-principles libertarianism.

Even if we could somehow overcome the privacy and First Amendment concerns that something like a minimum age for social media would pose for adult users (and that's a big if), we still don't know what sort of unintended consequences it might have for young people. Could a lack of online community lead to worse mental health outcomes? Would kids turn to more private methods of communication that make it harder to prevent harm or punish perpetrators? Separated from their digital distractions, would teens start drinking more and having more unintended pregnancies—both of which have decreased in recent decades?

Haidt says that if we wouldn't send our kids to Mars without having all the evidence and knowing all the risks, we shouldn't give them unfettered access to the digital world either. That goes for cutting off their connections to the digital world too.