Is Screen Time as Poisonous as We Think?

I am going to say a dirty word: screentime. That simple and suddenly terrifying word includes all the hours that you spend streaming movies, playing video games; especially the hours you spend caught in the vortex of social media and news and communication that gushes from your phone. Everyone seems to have something to say about screentime, from the American Academy of Pediatrics to religious and educational leaders to the New York Times opinion section — especially the New York Times opinion section. Our own devices scold us, with reminders to get off Instagram, to hold the phone further away to avoid eye strain; if you have an iPhone, you know that it tracks the hours you spend on it every day — and then makes a point to tell you that number, that always surprisingly large number. The people we are most worried about when it comes to screentime are children and teenagers. Parents around the world are in a full-on panic about the relationship between screentime and mental health — in part because there is new scientific evidence on this relationship.

Danny BLANCHFLOWER: What we’ve observed from around 2014 was, high school kids, we were seeing a rise in their use of internet and smartphones and so on, and a big rise in their anxiety levels, depression levels, all kinds of things.

That sounds terrible. But is the case against screentime really that clear-cut? Today on Freakonomics Radio, we will hear the evidence, and some challenges to the evidence:

Andrew PRZYBYLSKI: It’s very easy to fool yourself as an analyst. 

We also discuss the incentives at play.

Lauren OYLER: That’s how op-eds work — like people are desperate for one weird trick to save your life.

Still, you’d have to be a cold-hearted person to think that nothing should change. The question is — well, here’s the question:

BLANCHFLOWER: What the hell are we going to do about it?

*      *      *

David Blanchflower — his friends call him Danny, after the much-loved British footballer — is himself British, but he has spent the past few decades at Dartmouth College, in New Hampshire. He is an economics professor and researcher who has also been involved in economic policymaking.

BLANCHFLOWER: The way I would describe it is, I’m a data guy, I’m going to look for patterns in the data and I say, “Let’s kick and poke and punch the data as hard as we can, and let’s see if we get an answer.” 

Blanchflower is a pioneer of what some people call “happiness economics.” He is particularly well-known for one key finding.

BLANCHFLOWER: In any well-being equation I ever looked at, there was a midlife crisis. There was a U-shape in age in the data. 

This U-shape is known as the happiness curve. It is derived from many studies over the years, using survey data from thousands of people, of all ages, from around the world. Now, you may know how we feel about survey data around here — that it’s not the most reliable form of data. On the other hand: it’s hard to find a better way to measure something as intangible as happiness. Anyway: here’s what Danny Blanchflower has found.

BLANCHFLOWER: Basically, the happiest people are the young. It takes a U-shape. It diminishes down to around 50 years of age, and then it rises again.  

There are a number of stories you could tell about why happiness has a U-shape over the course of a lifetime. For many young people, life is fun and relatively easy. Adulthood, meanwhile, brings challenges, obligations, quite possibly offspring — and those offspring in turn bring more challenges and obligations. But then, happiness begins to rise again at around age 50 — an age at which many parents are no longer so obligated to their children. Is that a coincidence? Maybe, maybe not. There are other possible drivers of that upswing in happiness in middle age: you might just become more mature over time, more satisfied; you might become more comfortable with who you are, or at least better at managing your expectations. Danny Blanchflower has written more than 30 papers about this happiness curve, and his findings have been replicated over 600 times — an impressive record when you’re trying to measure an emotional state. The scientific consensus was so strong that many researchers had come to see the happiness curve as a natural part of human life. At least Blanchflower had. But then, one day, he came across the work of Jean Twenge, a psychology professor at San Diego State University.

BLANCHFLOWER: She was interviewed in the New York Times, I read that interview, and I started to read the papers, and the papers were about things that I had written about. She had used some of the data that I’d used.

But Twenge’s work didn’t fit the happiness curve that Blanchflower had made famous. Her work showed that happiness starts to decline before adulthood; she showed that teenagers in particular were reporting much higher rates of unhappiness than in the past.

BLANCHFLOWER: I started to look at the data, and I realized that I missed something. I mean, I wrote 30 papers saying there was a hump shape in age, or a U-shape. I literally said, “This is one of the most phenomenal facts in social science.” And then I wrote, “Until it wasn’t.” 

So what did Blanchflower miss?  

BLANCHFLOWER: So, now instead of the hump shape, we have a decline.

Meaning: more people reported being unhappy at an earlier age than he’d seen before.

BLANCHFLOWER: That’s only really started since 2015. And I missed it, and the world missed it. You know why we missed it? Because a bit of data took a while to come in, so we didn’t really get data until about 2018, 2019. And then Covid came, and everybody said, “Oh, look at the effects of Covid.” Well, it turns out now we have to rethink it, because a lot of it was actually occurring prior to Covid, and Covid merely extended pre-existing trends, and we didn’t know it. 

Blanchflower, to his credit, became convinced that his famous old happiness curve no longer fit the current reality. So now he began to — as he would say — kick and poke and punch the data.

BLANCHFLOWER: The data that I like is from the C.D.C. People are asked the following question: “Over the last 30 days, how many of those 30 were bad mental health days?” Most people just say, “No days.”

But that’s “most people.” One group of people gave the opposite answer — they said that every day of the previous 30 was a bad mental-health day.

BLANCHFLOWER: So what we’ve seen over time in America is that that is increased by quite a lot. The group that it’s increased the most are young women. So today, 10 percent of young women report, and 7 percent of young men report, that every day of their lives is a bad mental health day. We basically see this problem of a collapse in the well-being of the young in America. And then we see it in the U.K. So the U.N. says to me, “Danny, we got to look at the world. Let’s go look at the world.” 

That’s right: Blanchflower’s happiness research had caught the attention of the United Nations. He and other researchers were asked to find out more.

BLANCHFLOWER: So off we go and now basically we find it’s true everywhere. And so we’re talking about what I think is a huge global crisis. But the genie has been released from the bottle, and it was released from the bottle a decade ago, before we realized what was going on. 

So what happened a decade ago that might have caused this decline in mental health among young people? For Jean Twenge, the San Diego State psychologist, there is an obvious and easy answer: the smartphone. Apple introduced the iPhone in 2007 and sold around 1.4 million units that year. These days, Apple sells more than 200 million iPhones a year — and that’s just Apple: for every iPhone, four or five non-Apple smartphones are sold around the world, a billion a year. In 2017, Twenge published a book called — deep breath here — iGen: Why Today’s Super-Connected Kids Are Growing Up Less Rebellious, More Tolerant, Less Happy — and Completely Unprepared for Adulthood — and What That Means for the Rest of Us. Twenge also published some related pieces in The Atlantic and The New York Times, which amplified her argument. In early 2024, the influential social psychologist Jonathan Haidt, who teaches at New York University, joined this argument with a book called The Anxious Generation: How the Great Rewiring of Childhood Is Causing an Epidemic of Mental Illness. Here is Haidt, on MSNBC, discussing this epidemic.

Jonathan HAIDT: Everyone has a theory about what causes it. There is only one explanation. There is no other theory that can make sense of a synchronized global collapse in mental health — other than the fact that in 2010, the great majority of kids had a flip phone, no high-speed internet, no unlimited data, no Instagram. And by 2015, they all have a smartphone, high-speed internet, unlimited data, Instagram, front-facing camera.  

Haidt makes a compelling argument, as does Jean Twenge. And their argument has resonated with many people, parents especially. There have been suggestions that social media and photo-sharing apps can be particularly damaging to adolescent girls and young women, by amplifying pressure and anxiety around their appearance. But is the smartphone really “the one explanation,” as Haidt puts it? As we often preach around here, correlation does not equal causation. Maybe there are some other factors contributing to the anxiety and unhappiness that young people say they are feeling.

For instance, the last couple decades have brought wave after wave of political and economic turmoil, sometimes tipping into chaos. If you have parents or grandparents who grew up during the Great Depression, you know how much they were shaped by that experience; so have young people today maybe been shaped by all that chaos? And let’s not forget the widespread anxiety over climate change — and, for kids who grew up in the U.S., especially New York, the 9/11 attacks set a tone of fear and anger.

There’s also this: how about the benefits of the smartphone? Anti-phone advocates have done a good job pointing out the costs, but let’s not forget the benefits: a smartphone provides connection; it facilitates the sharing of interests; it helps you navigate to pretty much anywhere, and gives access to just about any piece of information, or music, or whatever else you might want. Of course, this can be overwhelming; and if you consume too much of it, you may get sick, just as you’ll get sick consuming too much of anything that humans consume. Like food. But just as food is pretty important for humans, so is connectivity — and to discount the benefits, especially for young people, may be short-sighted. So I went back to the economist Danny Blanchflower to talk through the causality piece of this.

DUBNER: If I asked you to give the strongest piece of evidence that you’re convinced that this is a causal relationship between the rise of, let’s call it digital tech use, and the rise of anxiety or depression, how do you know that that instrumental variable is the one that’s driving the bulk of that problem? Because I could posit a variety of other things. And let me throw one at you just off the top of my head. Your Dartmouth colleague Bruce Sacerdote has done research on the rise of negativity in media, especially in the U.S., and we know the effect of media on the psyche. So could it be that what we’re thinking of as, quote, digital tech being the driver is, in fact, you know, digital tech is in this case more of a delivery system for a massive wave of negativity? Also, a massive wave of information about mental health that may cause more people to self-diagnose, and so on.

BLANCHFLOWER: What a great question. Actually, Bruce and I are working on this together, I was just talking to him 20 minutes ago. So obviously, there’s a set of questions, “What is it?” And I sort of posit, as a policymaker — well, maybe it is something else, but this thing appears to disproportionately impact the young, and above that disproportionately it impacts young women. This trend started prior to Covid. Every piece of evidence we have is that it doesn’t appear to have been caused by the Great Recession. So all the things that you’ve talked about, that is true, but why would it especially be true of women? And you have to get the timing right — so, the timing fits, the rise we observe in the ill-being of the young starts around 2014. At exactly that time, you see the explosion of digital usage, internet usage, smartphone usage, and so on. And I think in the end, the causal question is sort of irrelevant, in the sense that, here we have a problem. The worry is what do you do about it, right? I mean, I’m obviously  concerned that this declining well-being of the young will translate itself into something bad. We’ve seen things like rising self-harm, evidence of rising thoughts of suicide, but to this point, thank God, we haven’t yet seen lots of bad outcomes. The people who have the least incidence of deaths by drug overdoses are the young. I don’t see much evidence of an increase there. And the evidence around the world is that suicides — in the U.S., suicides have risen slightly for young men, not in other places. 

DUBNER: Wait a minute, this seems like a pretty big deal here. You’re saying this engagement with smartphones is leading to, let’s just call it generally, declining mental health among the young. Could it be that this is a kind of gantlet that young people now go through and emerge relatively undamaged and, who knows, maybe even stronger?

BLANCHFLOWER: Supposing the answer to that is no, and we don’t do anything. We have to err on the side of caution, right? I mean, I agree with you, it may well be that, you intervene and you say, well, we were mistakenly intervening to make sure your child wasn’t in deep trouble. Who’s going to ever object to that? We were wrong. Okay. But it’s much better to be wrong trying to protect them than to do nothing and then suddenly we’re overtaken by a huge splurge in deaths from overdoses. In a way, the experience that I had at the Bank of England was interesting.

Blanchflower is talking here about when he served as an external member on a Bank of England committee that sets interest rates. In October of 2007, the global economy suddenly appeared fragile, perhaps on the brink of a deep recession.

BLANCHFLOWER: So I sit there and I make a decision on interest rates every month.

Blanchflower was the first and, at the time, only person on the committee to vote in favor of cutting interest rates.

BLANCHFLOWER: The hardest thing in the world is to know where you are. Because you’ve got no clue; data doesn’t come in for a year. So you sit and you try and make a decision on limited information. 

Being out front on interest rates, Blanchflower would later say, was, quote, “not a comfortable place to be … They called me bonkers.”

BLANCHFLOWER: And always you care about, have I made the right decision? I’m used to being in a world of incomplete information, where you have to make a decision. You’ve got to do something.

A year later, the Bank of England did finally cut rates, and in some circles, Danny Blanchflower was praised for his early call. So you could say that Blanchflower has a proactive disposition — which, in the case of interest rates, served him well. How about in the case of what he calls the “huge global crisis” caused by the smartphone? I’m sure you’ve read about the growing number of smartphone bans, or other restrictions, in schools and elsewhere. But some people think we shouldn’t rush to judgment; they argue that the research behind the smartphone panic isn’t very solid. Coming up, we’ll hear from one such critic.

*      *      *

The smartphone is a technology that has changed our society dramatically, and many people — including the economist Danny Blanchflower and the psychologist Jean Twenge — argue that the smartphone is causing harm, especially to young people. This puts the smartphone in the company of earlier inventions like the telephone, the television, the bicycle — even electricity.

PRZYBYLSKI: It’s really important to acknowledge that new things are scary.

That is Andrew Przybylski, a professor of human behavior and technology at Oxford.

PRZYBYLSKI: New things should cause anxiety. If someone or something turns up in your tribe — they try to feed you a novel food, they want to take care of your kids — it makes a lot of sense for your first reaction to be aversion, and to be skeptical. So it’s absolutely all right to be skeptical about new technologies. But what will often happen is that some people will come along, and they’ll kind of give the panic its skeleton. There’s a class of person called a moral entrepreneur. What a moral entrepreneur will do is, they’ll identify a at-risk group for a novel technology, and then you need a mechanism. You need a reason to say, “This time, it’s different.” Let’s say it’s Dungeons and Dragons and teenagers and Satanism. Then you have a mechanism, and, you know, that would be role playing or the fact that it’s animated or that in radio serials, there’s actually somebody talking through a crime. And so when you combine those three things, moral entrepreneurship thrives. There’s congressional inquiries, Senate inquiries about comic books or, you know, Lieberman holds up a Nintendo blaster, and everyone pats themselves on the back.

Przybylski is talking about the time in the early 1990s when U.S. Senator Joe Lieberman went on a crusade against violent video games. That may seem like a distant era.

PRZYBYLSKI: People get older, they forget. New technology comes along, and you have the person who is whipping up today’s panic recalling the good old days, when people would play video games in their basement.

DUBNER: In one paper that you’ve coauthored — this is called “Global Wellbeing and Mental Health in the Internet Age” — the abstract reads, “In the last two decades, the widespread adoption of internet technologies has inspired concern that they have negatively affected mental health and psychological well-being. However, research on the topic is contested, and hampered by methodological shortcomings, leaving the broader consequences of internet adoption unknown. We show that the past two decades have seen only small and inconsistent changes in global well-being and mental health that are not suggestive of the idea that the adoption of internet and mobile broadband is consistently linked to negative psychological outcomes.” Anyone reading that who has also been paying any attention to the news would say, “Wait a minute! Andy, this is exactly backwards, this is the opposite of everything I’ve been told!” So, tell me why you think you’re right, and what the argument that’s gotten so much heat lately has gotten wrong.

PRZYBYLSKI: I think the reason why your listeners would be confused and say, “Wait, why?” is a factor of two things. The first being, like me, they’re intensely skeptical about the role of technology in our lives and in our children’s lives. That’s natural, and that needs to be listened to very closely. And then, two, that academic paper that you just talked about, that wasn’t accompanied by a massive press campaign, two massive popular-press book campaigns, and so it doesn’t get you in the Opinion section of the Times. And so, I think that part of that double take is in part manufactured. But I think that actually we’re at a local maximum in terms of what many people call the “techlash,” or the backlash against technology. There was kind of a high-water mark in the other direction, in 2011, with Tahrir Square and the Arab Spring, and then through the Snowden period, through Cambridge Analytica, there was a fairly rapid swing that brings us all the way to the kind of scholarship that you were asking about. But, to get to the why I think that we’re, “right,” here is that the work that I do, and others, it is work that is not as glamorous. And I would argue that there’s an inverse relationship between how well a study is done in this field and how shocking the results might seem. When you cross your T’s and dot your I’s, when you share your code and you share your data, you don’t find the kinds of things that you or I would have as kind of a preexisting bias about tech. 

DUBNER: Can you give me an example — that a layperson could understand — of the methodological shortcomings or failures of the research you’re talking about?

PRZYBYLSKI: Phones, screentime, social media — these are all catch-all categories. And so unless you’re epistemically humble about what you’re measuring and what you’re not measuring, it can be very tempting to claim that when a parent or a kid is filling out a questionnaire that asks, “Think back on the last year of your life. On average, how many hours a day do you spend with a computer, a smartphone, a game console, and a cell phone.” And then call that “screentime” in the method section of your paper, and then call the title of your paper “Social Media and Its Impact on Teenage Girls.” And then you also have the measurement problem, and this pains me very deeply as a psychologist, of what the heck is mental health or well-being? Because saying you’re sad or saying you’re satisfied with your life is very, very different than turning up in a therapist’s office and being diagnosed as having major depressive disorder or anxiety disorder. And all of those little decisions of how you deal with measurement, those actually can be quite consequential. If all of these arbitrary choices that are made by a research analyst — if they’re consistent with the researcher’s preexisting biases or they’re consistent with the topic of their popular book, it’s a bit of a red flag, or at least a yellow flag. When you think of everyday behaviors as being potentially pathological or potentially addictive, it’s a fairly slippery slope to pathologize everything. I mean, video gaming is as damaging to mental health as beds are to mental health. You could create a disorder called bed addiction disorder where people have, you know, low affect and low mobility and get bed sores. And you could try to regulate the sale of beds and haul bed companies in front of Congress. A lot of the evidence you would have that something is bad for you, it’s not built on the basis of chemistry or biochemistry or biology.

DUBNER: You’ve made a similar argument in the past, saying that the effects of digital technology on teenagers are about as big as the effects of eating potatoes.

PRZYBYLSKI: Well, yeah, social scientists will often try to draw inferences about the population-level effects or associations that might link any given activity to a health outcome. And the problem with that is that if you aren’t crossing your T’s and dotting your I’s, you can interpret the noise in a data set in a way that’s consistent with your preexisting beliefs or biases. And that’s what happens with technology. We were trying to make a very simple point, which is that if you don’t have your hypotheses before you look at your data, you can be led astray by very small effects. Whether it’s left-handedness, wearing glasses, enjoying bicycling — these are all things that have the same “effect,” which is really just a correlation, in a large data set. 

DUBNER: I think one interesting thing about this topic — which is why it’s gotten so many people riled up — is that this is a technology that just about everybody has experience with. So they can sort of fill in the blank for the causal mechanisms, right? They can say, “Oh, I’ve had the experience on my phone where rather than going to that event tonight at my place of worship or rather than going to play soccer with my friends, I get caught up in my silo of anxiety and depression on the phone.” And this lets people layer their own experiences onto the moral-panic argument. But aren’t personal experiences in general a good starting point for a lot of social-science research? 

PRZYBYLSKI: I think that they’re a really important starting point, but I think they’re a highly invalid ending point. And they shouldn’t be where people’s thinking stops. It’s very easy to think that there’s a digital world and an analog world, and these worlds don’t connect. But the problem isn’t necessarily that one is good or bad. The problem is that we’re putting this wedge between them, and when we put that wedge between them, what we do is we cut ourselves off from being able to investigate really interesting questions about, actually what is happening to someone if they’re depressed, and they’re using social media? There’s a gigantic difference between feeling unhappy with how you spent your evening and suffering from agoraphobia, having major depressive disorder or some form of crippling social anxiety. I’m sure you know people who have suffered, and many people who are listening know people who suffered, with opioid addiction or depression or anxiety, and claiming that the thing that happens when you play a video game is like what happens when you can’t stop taking opioids, or saying, you know, I don’t feel like going outside, or I’ve had a bad Twitter argument, I feel so anxious, saying that’s the same thing as crippling agoraphobia, where people won’t leave their home for years — frankly, that’s insulting. Because there’s no reason to think that’s the same mechanism. 

Lauren OYLER: Over the last, you know, seven or eight years, I have struggled with what I am quite comfortable calling an internet addiction. 

That is Lauren Oyler, a novelist and cultural critic in her early 30s. She grew up in West Virginia, and now lives in Berlin. Oyler has spent a lot of time thinking and writing about life online and off.

OYLER: I have met great friends on social media, I’ve gotten boyfriends on social media, but at the same time, I do strongly feel that it can produce an actual addiction, like a straightforward addiction, like a drug addiction. And I know that because since I divested from social media, I’ve picked up smoking, and it’s the same. It’s the same. It’s the same sort of thought process about, like, compulsively wanting to either look at Twitter or smoke a cigarette. You train yourself to think, “Something could be happening on my phone. Someone could have written me, some news could have broken, and I need to see it.”

Oyler recently published a collection of essays called No Judgment. One of the pieces is titled “My Anxiety.”

OYLER: I think anxiety does create a barrier around actually doing things that you’re sitting and worrying about, and sitting and looking at your phone is just kind of externalizing this worrying feeling, right? Like your phone is doing the worrying for you, because it has this short attention span. My thoughts are like bouncing around, and I might be thinking about taxes, and then I’m thinking about some guy, and then I’m thinking about my late article that I need to turn in, and I’m thinking about I embarrassed myself at a party. And all this happens in a span of two minutes, which, if you’ve been on social media, that is exactly what it’s like, right? 

So Lauren Oyler does see a deep connection between her smartphone and what researchers like Danny Blanchflower call mental “ill-being.” But again, is the one necessarily causing the other?

OYLER: I get anxious that I’ll become depressed, and depressed that I’m so anxious. And if the phone is a kind of conduit for feeling, or if it’s an extension of your brain, then, of course, the things on your phone might make you depressed, just as they might make you anxious. So I think, on one hand, of course it’s causal, but I do think it’s sort of self-defeating and shortsighted to suggest that the phone is the only cause of this mental health crisis — among teenagers or among anyone. Because that’s just not how the world works. That’s how op-eds work — like, people who are desperate for one weird trick to save your life. It’s like “I just needed to throw my phone in the river,” and then everything would end up fine. That’s just not how the world works.

Okay, so if there is rising anxiety among young people, and if throwing your phone in the river isn’t a solution, what is?

*      *      *

Andy Przybylski, at Oxford, argues that the panic over smartphones is overblown. But that doesn’t mean he thinks everything is fine.

PRZYBYLSKI: In the U.S., there are some really worrying trend lines for people across the population, and the most worrying trends — things like self-harm, and death by suicide — I would be much more worried as a middle-aged man in Appalachia with access to a gun and with access to West Virginia’s social safety net than I would be as a white teenage girl anywhere in the U.S. And so I would say that there probably is something broken in America. 

DUBNER: Here’s something I also wonder about: there is so much more awareness around mental health now than there was 20, 30 years ago and so much less stigma — some have argued that, at least. Do you think that acceptance might be showing up in the data as more young people today saying, “Yes, I’ve experienced mental health problems” than they used to?

PRZYBYLSKI: There’s a concept called measurement variance. This is the basic idea that if I’d ask you how I’m feeling today, and then I compare how I’m feeling today to how I said I felt 10 years ago, the way that people answer that question actually changes over time. And so there’s a bit of that, absolutely. And then, the way that we actually track things like mental health problems, in the U.S. in particular, it’s very, very susceptible to change as a function of changes in insurance, and changes in the way that diseases can be classified. 

On this note, Przybylski, points to a study by two health economists, Adriana Corredor-Waldron and Janet Currie. Their paper is called “To What Extent Are Trends in Teen Mental Health Driven by Changes in Reporting?”

PRZYBYLSKI: It’s a study of clinical intakes of young women in New Jersey across the last 10 years. There’s a series of spikes in hospital admissions for different types of serious disorders, when a young person shows up in crisis. And one way that you could have interpreted the data is to say that there is a mental health epidemic in New Jersey, for teenage girls in particular. But the researchers went back and looked at changes to the best practices for clinical intake. And the peaks occurred in response to two changes. The first is a special awareness campaign for clinicians, for focusing on young women as a target area for intervention. And then the second was allowing there to be a second reason for clinical intake. So it used to be when you came in, you would just say, “The person is schizotypal” or “depressed.” But then you could say “schizotypal and depressed.” And so when that change went through, the amounts went up. But — and here’s the rub — it only went up for young people who were on insurance, because it created a new billable category. It didn’t go up for those who didn’t. So I would say, when you’re a hammer, the problems of the world are nails. I think it’s very tempting to think that the atomization of certain Western cultures is a result of technology and not other things, like a decline in religiosity, or these other kinds of things. There aren’t going to be quick-fix solutions.

DUBNER: Like, throw away your phones. 

PRZYBYLSKI: Throwing away your phones or putting them in a locker. What you’re going to have to do is actually rebuild communities. You’re going to have to hire psychotherapists. You’re going to have to build civil and economic infrastructure to support those in your society who are falling through the cracks.

DUBNER: What if I say to you, I agree with you, and that sounds smart and sane, but it also is going to take a long time and a lot of money, a lot of resources. And in the meantime, I believe there are millions, maybe billions of people suffering. And so I do want to do something quick and dirty. What would you say to that? 

PRZYBYLSKI: The law of unintended consequences. I would say that if you can’t properly diagnose the problem, you can’t possibly solve it. I would say that if you take this topic seriously, you would take slow and incremental steps to ascertain exactly what problem you’re trying to solve. And so, yeah, I would say that you should not waste energy on patting yourself on the back.

DUBNER: What if I ask you to make an argument for the benefits of digital tech that are overlooked in the rush to proclaim all of digital technology damaging to young people especially?

PRZYBYLSKI: One of the things that I think everyone can agree is really worrying is content online around self-harm or anorexia. And it might seem like a very obvious thing that platforms should be doing absolutely all they can do to take down this content and to scuttle these communities and make sure that it’s not on any of the big platforms. That might seem like common sense, that might feel like the right thing to do. But you can wind up causing so much more harm than good if you don’t pay attention to actually how people who struggle with these disorders use social media platforms. There are really good support groups for things like self-harm, suicidal thoughts, and things like anorexia and bulimia. One of the things that social media platforms can be really good for is connecting communities in a positive way that self-moderates, and they provide resources for people in crisis. And what happens if you think that it’s the responsibility of these companies to just crush it so that it’s not there anymore? Two things will always happen and they’re both very worrying. The first is, people who are already struggling, they will begin to stop disclosing. They’ll button themselves up. They won’t reach out. And they’ll lose access to the online social environments that have been supportive for them. And then the second thing is, the community moves to other parts of the internet that are more poorly moderated, and are far less safe. And so you wind up in a Telegram group, where there’s no accountability, instead of being in a Facebook or Instagram group. And so I think thinking about tools to meet young people where they are is a much smarter idea than just burning everything down and salting the earth. 

As you just heard, Andy Przybylski is more positive toward the big social platforms than many people in this arena. In fact, he recently announced a research collaboration with the Facebook and Instagram parent company Meta.

PRZYBYLSKI: We’re going to solicit proposals from researchers all around the globe. They can get data on young people in 40 different countries, and combine that with information about their mental health and their well-being. But at no point will they have a direct relationship, in terms of the selection of their projects or the decision for the projects to go forward, with the researchers at Meta. We’ve put in a series of firewalls between the data scientists at Meta and researchers who would be very interested in investigating the idea that Instagram relates to the mental health and the well-being of teenagers. Myself and my coeditors will be selecting projects on the basis of their scientific merit. And their proposals will be peer-reviewed in advance of data collection by researchers who are specialists in topics like mental health, topics like online engagement, but also generalists who have a sense of what is good methodology. And only when that peer review process finishes, the data request goes directly to Meta. The idea here is you have these firewalls in place, you don’t have Big Tech negatively affecting the process, and you don’t have researchers potentially moving the goalposts because they think they’ve found something interesting after having a rummage through the data. 

It might surprise you that the rigorous analysis Przybylski is talking about hasn’t already been done on Instagram and Facebook and other social-media platforms. But it’s worth remembering how new this ecosystem is. Here, again, is the writer Lauren Oyler.

OYLER: The social media era has been really short. Like, when was Myspace founded? When was Facebook founded? In the early 2000s, right? It’s not very long. And already Facebook — nobody uses that, they had to buy Instagram and WhatsApp in order to stay relevant. And Instagram is actually quite young. I think it’s like 2010? Yeah, that’s really young. So I look forward to a day when Instagram doesn’t exist, but I’m sure other things will take its place. 

I’m guessing Oyler is right, that the platforms will change — and the way we interact with our phones will change. Some people, like Andy Przybylski believe that “if you can’t properly diagnose the problem, you can’t possibly solve it” — and therefore, to rush into a hard anti-phone position isn’t the right move. Other people, like Danny Blanchflower, say that we can’t afford to wait, because if the rise in anxiety and unhappiness among young people is as significant as he thinks, the consequences will be significant too. Especially the kind of consequences that economists like to consider.

BLANCHFLOWER:  I worry about what will happen to, you know, these young people as they come to the labor market, and then later on we’re going to worry about their mortality, and we’re going to worry about their ability to generate savings and investment and, you know, buy a house and buy themselves a retirement package. So, down the road, these are issues that we care about. 

DUBNER: You are never going to run out of problems to address, are you? 

BLANCHFLOWER: In a way, that’s what I’ve tried to do. Think about how we address the problems, and try and find solutions. I don’t know what works. I have been asked that question. We go, “We don’t know, but we’re going to try and find out for you.” That’s what we do. 

DUBNER: Do me a favor. When you find something, call me and we’ll talk about it again, okay?  

BLANCHFLOWER: Love to. Love to.  

Thanks to Danny Blanchflower, Andy Przybylski, and Lauren Oyler for their insights today. And I’m curious to know what you think; our email is radio@freakonomics.com.

*      *      *

Freakonomics Radio is produced by Stitcher and Renbud Radio. This episode was produced by Theo Jacobs. Our staff also includes Alina Kulman, Augusta Chapman, Dalvin Aboagye, Eleanor Osborne, Ellen Frankman, Elsa Hernandez, Gabriel Roth, Greg Rippin, Jasmin Klinger, Jeremy Johnston, Jon Schnaars, Julie Kanfer, Lyric Bowditch, Morgan Levey, Neal Carruth, Rebecca Lee Douglas, Sarah Lilley, and Zack Lapinski. Our theme song is “Mr. Fortune,” by the Hitchhikers; our composer is Luis Guerra.