AI transcript
0:00:02 (dramatic music)
0:00:08 I am going to say a dirty word, screen time.
0:00:13 That simple and suddenly terrifying word
0:00:15 includes all the hours that you spend streaming movies,
0:00:18 playing video games, especially the hours you spend
0:00:21 caught in the vortex of social media and news
0:00:25 and communication that gushes from your phone.
0:00:28 Everyone seems to have something to say about screen time
0:00:30 from the American Academy of Pediatrics
0:00:33 to religious and educational leaders
0:00:35 to the New York Times opinion section,
0:00:38 especially the New York Times opinion section.
0:00:42 Our own devices scold us with reminders to get off Instagram,
0:00:46 to hold the phone further away, to avoid eye strain.
0:00:48 If you have an iPhone, you know that it tracks
0:00:50 the hours you spend on it every day
0:00:53 and then makes a point to tell you that number,
0:00:57 that always surprisingly large number.
0:00:58 The people we are most worried about
0:01:02 when it comes to screen time are children and teenagers.
0:01:05 Parents around the world are in a full-on panic
0:01:07 about the relationship between screen time
0:01:10 and mental health, in part because there is
0:01:13 new scientific evidence on this relationship.
0:01:18 What we’ve observed from around 2014 was high school kids,
0:01:20 we were seeing a rise in their use of internet
0:01:23 and smartphones and so on, and a big rise
0:01:25 in their anxiety levels, depression levels,
0:01:26 all kinds of things.
0:01:30 That sounds terrible, but is the case
0:01:33 against screen time really that clear cut?
0:01:37 Today on Freakonomics Radio, we will hear the evidence
0:01:40 and some challenges to the evidence.
0:01:43 It’s very easy to fool yourself as an analyst.
0:01:46 We’ll also discuss the incentives at play.
0:01:49 That’s how op-eds work, like people are desperate
0:01:51 for one weird trick to save your life.
0:01:54 Still, you’d have to be a cold-hearted person
0:01:56 to think that nothing should change.
0:01:59 The question is, well, here’s the question.
0:02:02 – What the hell are we gonna do about it?
0:02:04 (upbeat music)
0:02:15 – This is Freakonomics Radio,
0:02:19 the podcast that explores the hidden side of everything
0:02:21 with your host, Stephen Dubner.
0:02:24 (upbeat music)
0:02:27 (upbeat music)
0:02:32 – David Blanchflower, his friends call him Danny
0:02:34 after the much-loved British footballer,
0:02:37 is himself British, but he has spent the past few decades
0:02:39 at Dartmouth College in New Hampshire.
0:02:42 He is an economics professor and researcher
0:02:45 who has also been involved in economic policy making.
0:02:47 – The way I would describe it is I’m a data guy,
0:02:49 and I look at patterns in the data, and I say,
0:02:52 let’s kick and poke and punch the data as hard as we can,
0:02:53 and let’s see if we get an answer.
0:02:55 – Blanchflower is a pioneer of what some people
0:02:57 call happiness economics.
0:03:01 He is particularly well-known for one key finding.
0:03:03 – In any well-being equation I ever looked at,
0:03:05 there was a midlife crisis.
0:03:08 There was a U-shape in age in the data.
0:03:12 – This U-shape is known as the happiness curve.
0:03:14 It is derived from many studies over the years
0:03:17 using survey data from thousands of people
0:03:19 of all ages from around the world.
0:03:22 Now, you may know how we feel about survey data
0:03:26 around here that it’s not the most reliable form of data.
0:03:28 On the other hand, it’s hard to find a better way
0:03:32 to measure something as intangible as happiness.
0:03:35 Anyway, here’s what Danny Blanchflower has found.
0:03:38 – Basically, the happiest people of the young,
0:03:40 it takes a U-shape, it diminishes down
0:03:43 to around 50 years of age, and then it rises again.
0:03:45 – There are a number of stories you could tell
0:03:47 about why happiness has a U-shape
0:03:49 over the course of a lifetime.
0:03:54 – For many young people, life is fun and relatively easy.
0:03:57 Adulthood, meanwhile, brings challenges,
0:03:59 obligations, quite possibly offspring,
0:04:02 and those offspring in turn bring more challenges
0:04:04 and obligations.
0:04:09 But then, happiness begins to rise again at around age 50,
0:04:12 an age at which many parents are no longer
0:04:14 so obligated to their children.
0:04:17 Is that a coincidence?
0:04:19 Maybe, maybe not.
0:04:21 There are other possible drivers of that upswing
0:04:23 and happiness in middle age.
0:04:27 You might just become more mature over time, more satisfied.
0:04:30 You might become more comfortable with who you are,
0:04:33 or at least better at managing your expectations.
0:04:37 Danny Blanchflower has written more than 30 papers
0:04:38 about this happiness curve,
0:04:42 and his findings have been replicated over 600 times,
0:04:43 an impressive record when you’re trying
0:04:46 to measure an emotional state.
0:04:49 The scientific consensus was so strong
0:04:51 that many researchers had come to see the happiness curve
0:04:56 as a natural part of human life, at least Blanchflower had.
0:05:00 But then, one day, he came across the work of Jean Twenge,
0:05:03 a psychology professor at San Diego State University.
0:05:05 – She was interviewed in the New York Times,
0:05:09 and read that interview, and I started to read the papers,
0:05:12 and the papers were about things that I had written about.
0:05:15 She’d used some of the data that I’d used.
0:05:18 But Twenge’s work didn’t fit the happiness curve
0:05:20 that Blanchflower had made famous.
0:05:23 Her work showed that happiness starts to decline
0:05:25 before adulthood.
0:05:27 She showed that teenagers in particular
0:05:30 were reporting much higher rates of unhappiness
0:05:31 than in the past.
0:05:33 – I started to look at the data,
0:05:35 and I realized that I had missed something.
0:05:38 I mean, I wrote 30 papers saying
0:05:40 there was a hump shape in age, or a U shape.
0:05:43 I literally said this is one of the most phenomenal facts
0:05:46 in social science, and then I wrote until it wasn’t.
0:05:48 – So what did Blanchflower miss?
0:05:51 – So now, instead of the hump shape, we have a decline.
0:05:54 – Meaning more people reported being unhappy
0:05:56 at an earlier age than he had seen before.
0:05:59 – That’s only really started since 2015,
0:06:01 and I missed it, and the world missed it.
0:06:03 Do you know why we missed it?
0:06:05 Because a bit of data took a while to come in,
0:06:08 so we didn’t really get data until about 2018, 2019,
0:06:10 and then COVID came.
0:06:12 And everybody said, “Oh, look at the effects of COVID.”
0:06:15 Well, it turns out now we have to rethink it,
0:06:17 because a lot of it was actually occurring prior to COVID,
0:06:21 and COVID merely extended pre-existing trends,
0:06:22 and we didn’t know it.
0:06:24 – Blanchflower, to his credit,
0:06:28 became convinced that his famous old happiness curve
0:06:30 no longer fit the current reality.
0:06:33 So now he began to, as he would say,
0:06:35 kick and poke and punch the data.
0:06:39 – The data that I like is from the CDC.
0:06:41 People are asked the following question.
0:06:43 Over the last 30 days,
0:06:46 how many of those 30 were bad mental health days?
0:06:48 Most people just say no days.
0:06:49 – But that’s most people.
0:06:52 One group of people gave the opposite answer.
0:06:55 They said that every day of the previous 30
0:06:57 was a bad mental health day.
0:06:59 – So what we’ve seen over time in America
0:07:01 is that that has increased by quite a lot.
0:07:05 The group that has increased the most are young women.
0:07:09 So today, 10% of young women report,
0:07:12 and 7% of young men report that every day of their lives
0:07:14 is a bad mental health day.
0:07:16 We basically see this problem of a collapse
0:07:18 in the well-being of the young in America,
0:07:19 and then we see it in the UK.
0:07:21 So the UN says to me, Danny, we gotta look at the world.
0:07:23 Let’s go look at the world.
0:07:23 – That’s right.
0:07:25 Blanchflower’s happiness research
0:07:28 had caught the attention of the United Nations.
0:07:31 He and other researchers were asked to find out more.
0:07:34 – So off we go, and now basically we find
0:07:36 it’s true everywhere.
0:07:38 And so we’re talking about what I think
0:07:40 is a huge global crisis.
0:07:42 But the genie has been released from the bottle,
0:07:44 and it was released from the bottle a decade ago
0:07:46 before we realized what was going on.
0:07:49 – So what happened a decade ago
0:07:53 that caused this decline in mental health among young people?
0:07:56 For Gene Twenge, the San Diego state psychologist,
0:07:58 there is an obvious and easy answer.
0:07:59 The smartphone.
0:08:02 Apple introduced the iPhone in 2007
0:08:06 and sold around 1.4 million units that year.
0:08:10 These days, Apple sells more than 200 million iPhones a year,
0:08:12 and that’s just Apple.
0:08:15 For every iPhone, four or five non-Apple smartphones
0:08:19 are sold around the world a billion a year.
0:08:24 In 2017, Twenge published a book called “Deep Breath” here.
0:08:26 I, Gen, why today’s super connected kids
0:08:30 are growing up less rebellious, more tolerant, less happy,
0:08:34 and completely unprepared for adulthood,
0:08:37 and what that means for the rest of us.
0:08:39 Twenge also published some related pieces
0:08:41 in “The Atlantic” and “The New York Times,”
0:08:43 which amplified her argument.
0:08:47 In early 2024, the influential social psychologist
0:08:50 Jonathan Haidt, who teaches at New York University,
0:08:54 joined this argument with a book called “The Anxious Generation,”
0:08:56 how the great rewiring of childhood
0:08:59 is causing an epidemic of mental illness.
0:09:03 Here is Haidt on MSNBC discussing this epidemic.
0:09:06 Everyone has a theory about what causes it.
0:09:07 There is only one explanation.
0:09:09 There is no other theory that can make sense
0:09:11 of a synchronized global collapse in mental health,
0:09:14 other than the fact that in 2010,
0:09:16 the great majority of kids had a flip phone,
0:09:20 no high-speed internet, no unlimited data, no Instagram.
0:09:23 And by 2015, we all have a smartphone,
0:09:26 high-speed internet, unlimited data,
0:09:28 Instagram, front-facing camera.
0:09:31 Haidt makes a compelling argument, as does Gene Twenge,
0:09:34 and their argument has resonated with many people,
0:09:36 parents especially.
0:09:38 There have been suggestions that social media
0:09:42 and photo-sharing apps can be particularly damaging
0:09:43 to adolescent girls and young women
0:09:48 by amplifying pressure and anxiety around their appearance.
0:09:52 But is the smartphone really the one explanation,
0:09:54 as Haidt puts it?
0:09:56 As we often preach around here,
0:09:59 correlation does not equal causation.
0:10:02 Maybe there are some other factors contributing
0:10:04 to the anxiety and unhappiness
0:10:07 that young people say they are feeling.
0:10:09 For instance, the last couple decades have brought
0:10:13 wave after wave of political and economic turmoil,
0:10:16 sometimes tipping into chaos.
0:10:18 If you have parents or grandparents
0:10:20 who grew up during the Great Depression,
0:10:23 you know how much they were shaped by that experience.
0:10:25 So have young people today maybe been shaped
0:10:27 by all that chaos?
0:10:30 And let’s not forget the widespread anxiety
0:10:31 over climate change.
0:10:35 And for kids who grew up in the US, especially New York,
0:10:40 the 9/11 attacks set a tone of fear and anger.
0:10:41 There’s also this.
0:10:45 What about the benefits of the smartphone?
0:10:47 Anti-phone advocates have done a good job
0:10:50 pointing out the costs, but let’s not forget the benefits.
0:10:52 The smartphone provides connection,
0:10:55 it facilitates the sharing of interests,
0:10:58 helps you navigate to pretty much anywhere,
0:11:02 and gives access to just about any piece of information
0:11:05 or music or whatever else you might want.
0:11:07 Of course, this can be overwhelming
0:11:10 and if you consume too much of it, you may get sick,
0:11:12 just as you’ll get sick consuming too much of anything
0:11:15 that humans consume, like food.
0:11:19 But just as food is pretty important for humans,
0:11:21 so is connectivity.
0:11:24 And to discount the benefits, especially for young people,
0:11:26 may be short-sighted.
0:11:30 So I went back to the economist Danny Blanchflower
0:11:33 to talk through the causality piece of this.
0:11:36 If I asked you to give the strongest piece of evidence
0:11:39 that you’re convinced that this is a causal relationship
0:11:41 between the rise of, let’s call it digital tech use,
0:11:44 and the rise of anxiety or depression,
0:11:47 how do you know that that instrumental variable
0:11:50 is the one that’s driving the bulk of that problem?
0:11:52 Because I could posit a variety of other things.
0:11:52 – Yeah, of course.
0:11:55 If we throw one at you just off the top of my head,
0:11:58 your Dartmouth colleague, Bruce Sassardot,
0:12:01 has done research on the rise of negativity in media,
0:12:03 especially in the U.S.
0:12:07 And we know the effect of media on the psyche.
0:12:10 So could it be that what we’re thinking of
0:12:14 as, quote, “digital tech” being the driver is in fact,
0:12:17 digital tech is in this case more of a delivery system
0:12:20 for a massive wave of negativity,
0:12:24 also a massive wave of information about mental health
0:12:27 that may cause more people to self-diagnose and so on?
0:12:28 – What a great question.
0:12:30 Actually, Bruce and I were working on this together.
0:12:32 I was just talking to him 20 minutes ago.
0:12:35 So obviously there’s a set of questions, what is it?
0:12:39 And I sort of posit as a policymaker,
0:12:41 well, maybe it is something else,
0:12:44 but this thing appears to disproportionately impact the young
0:12:47 and above that disproportionately,
0:12:49 it impacts young women.
0:12:52 This trend started prior to COVID.
0:12:55 Every piece of evidence we have is that it doesn’t appear
0:12:57 to have been caused by the Great Recession.
0:13:00 So all the things that you’ve talked about, that is true,
0:13:03 but why would it especially be true of women?
0:13:04 And you have to get the timing right.
0:13:07 So the timing fits, the rise we observe
0:13:11 in the ill-being of the young starts around 2014.
0:13:14 At exactly that time, you see the explosion of digital usage,
0:13:17 internet usage, smartphone usage, and so on.
0:13:21 And I think in the end, the causal question is sort of irrelevant
0:13:24 in the sense that here we have a problem, right?
0:13:26 The worry is what do you do about it, right?
0:13:28 I mean, I’m obviously concerned
0:13:30 that this declining wellbeing of the young
0:13:33 will translate itself into something bad.
0:13:35 We’ve seen things like rising self-harm,
0:13:38 evidence of rising thoughts of suicide,
0:13:40 but to this point, thank God,
0:13:43 we haven’t yet seen lots of bad outcomes.
0:13:45 The people who have the least incidence of deaths
0:13:47 by a drug overdose are the young.
0:13:49 I don’t see much evidence of an increase there.
0:13:52 And the evidence around the world is that in the US,
0:13:54 suicides have risen slightly for young men,
0:13:55 not in other places.
0:13:58 – Wait a minute, this seems like a pretty big deal here.
0:14:00 You’re saying this engagement with smartphones
0:14:02 is leading to, let’s just call it generally,
0:14:05 declining mental health among the young.
0:14:07 Could it be that this is a kind of gantlet
0:14:09 that young people now go through
0:14:13 and emerge relatively undamaged?
0:14:15 And who knows, maybe even stronger?
0:14:17 – Supposing the answer to that is no,
0:14:18 and we don’t do anything.
0:14:21 We have to err on the side of caution, right?
0:14:23 I mean, I agree with you, it may well be that.
0:14:24 You intervene and you say,
0:14:27 “Well, we were mistakenly intervening
0:14:29 to make sure your child wasn’t in deep trouble.
0:14:31 Who’s your ever rejected that?”
0:14:34 We were wrong, okay, but it’s much better to be wrong
0:14:36 trying to protect them than to do nothing.
0:14:38 And then suddenly we’re overtaken
0:14:41 by a huge splurge in deaths from overdoses.
0:14:43 In a way, the experience that I had
0:14:46 at the Bank of England was interesting.
0:14:47 – Blanche Flower is talking here
0:14:50 about when he served as an external member
0:14:53 on a Bank of England committee that sets interest rates.
0:14:56 In October of 2007, the global economy
0:14:58 suddenly appeared fragile,
0:15:00 perhaps on the brink of a deep recession.
0:15:02 – So I sit there and I make a decision
0:15:04 on interest rates every month.
0:15:07 – Blanche Flower was the first and, at the time,
0:15:09 only person on the committee
0:15:11 to vote in favor of cutting interest rates.
0:15:12 – The hardest thing in the world
0:15:14 is to know where you are,
0:15:15 because you’ve got no clothes dead,
0:15:16 it doesn’t come in for a year.
0:15:18 So you sit and you try and make a decision
0:15:19 on limited information.
0:15:21 – Being out front on interest rates,
0:15:23 Blanche Flower would later say,
0:15:26 was, quote, “not a comfortable place to be,
0:15:28 they called me bunkers.”
0:15:29 – And always you care about,
0:15:31 have I made the right decision?
0:15:35 I’m used to being in a world of incomplete information,
0:15:36 where you have to make a decision.
0:15:38 You’ve got to do something.
0:15:41 – A year later, the Bank of England
0:15:43 finally did cut interest rates.
0:15:46 And in some circles, Danny Blanche Flower
0:15:49 was praised for his early call.
0:15:51 So you could say that Blanche Flower
0:15:53 has a proactive disposition,
0:15:57 which, in the case of interest rates, served him well.
0:15:59 How about in the case of what he calls
0:16:04 the huge global crisis caused by the smartphone?
0:16:05 I’m sure you’ve read about the growing number
0:16:08 of smartphone bans or other restrictions
0:16:11 in schools and elsewhere.
0:16:14 But some people think we shouldn’t rush to judgment.
0:16:18 They argue that the research behind the smartphone panic
0:16:19 isn’t very solid.
0:16:22 After the break, we will hear from one such critic.
0:16:25 – If you can’t properly diagnose the problem,
0:16:26 you can’t possibly solve it.
0:16:27 – I’m Stephen Dubner.
0:16:29 This is Freakonomics Radio.
0:16:30 We’ll be right back.
0:16:33 (gentle music)
0:16:39 (dramatic music)
0:16:43 – The smartphone is a technology
0:16:45 that has changed our society dramatically.
0:16:49 And many people, including the economist Danny Blanche Flower
0:16:52 and the psychologist Jean Twenge,
0:16:54 argue that the smartphone is causing harm,
0:16:56 especially to young people.
0:16:57 This puts the smartphone in the company
0:17:00 of earlier inventions like the telephone,
0:17:03 the television, the bicycle, even electricity.
0:17:05 – It’s really important to acknowledge
0:17:07 that new things are scary.
0:17:09 – That is Andrew Shibilsky,
0:17:13 a professor of human behavior and technology at Oxford.
0:17:15 – New things should cause anxiety.
0:17:17 If someone or something turns up in your tribe,
0:17:19 they try to feed you a novel food.
0:17:21 They wanna take care of your kids.
0:17:23 It makes a lot of sense for your first reaction
0:17:25 to be aversion and to be skeptical.
0:17:27 So it’s absolutely all right to be skeptical
0:17:28 about new technologies.
0:17:31 But what will often happen is that some people will come along
0:17:34 and they’ll kind of give the panic its skeleton.
0:17:37 There’s a class of person called a moral entrepreneur.
0:17:39 What a moral entrepreneur will do
0:17:43 is they’ll identify a at-risk group for a novel technology.
0:17:44 And then you need a mechanism.
0:17:47 You need a reason to say this time it’s different.
0:17:49 Let’s say it’s Dungeons and Dragons
0:17:51 and Teenagers and Satanism.
0:17:55 Then you have a mechanism and that would be role-playing
0:17:59 or the fact that it’s animated or that in radio serials,
0:18:02 there’s actually somebody talking through a crime.
0:18:04 And so when you combine those three things,
0:18:06 moral entrepreneurship thrives.
0:18:08 There’s congressional inquiries,
0:18:09 Senate inquiries about comic books
0:18:12 or Lieberman holds up a Nintendo blaster
0:18:15 and everyone pats themselves on the back.
0:18:16 – Shabilsky is talking about the time
0:18:19 in the early 1990s when US Senator Joe Lieberman
0:18:22 went on a crusade against violent video games.
0:18:25 That may seem like a distant era.
0:18:27 – People get older, they forget.
0:18:29 New technology comes along
0:18:33 and you have the person who is whipping up today’s panic
0:18:35 recalling the good old days
0:18:37 when people would play video games in their basement.
0:18:39 – In one paper that you’ve co-authored,
0:18:41 this is called global wellbeing and mental health
0:18:42 in the internet age.
0:18:45 The abstract reads, in the last two decades,
0:18:47 the widespread adoption of internet technologies
0:18:50 has inspired concern that they have negatively affected
0:18:52 mental health and psychological wellbeing.
0:18:56 However, research on the topic is contested and hampered
0:18:58 by methodological shortcomings
0:18:59 leaving the broader consequences
0:19:02 of internet adoption unknown.
0:19:04 We show that the past two decades
0:19:07 have seen only small and inconsistent changes
0:19:10 in global wellbeing and mental health
0:19:11 that are not suggestive of the idea
0:19:14 that the adoption of internet and mobile broadband
0:19:18 is consistently linked to negative psychological outcomes.
0:19:20 Anyone reading that who’s also been paying attention
0:19:24 to the news would say, wait a minute, Andy,
0:19:27 this is exactly backwards.
0:19:29 This is the opposite of everything I’ve been told.
0:19:33 So tell me why you think you’re right
0:19:35 and what the argument that has gotten
0:19:38 so much heat lately has gotten wrong.
0:19:40 – I think the reason why your listeners
0:19:43 would be confused and say, wait, why?
0:19:44 Is a factor of two things.
0:19:47 The first being, like me, they’re intensely skeptical
0:19:49 about the role of technology in our lives
0:19:51 and in our children’s lives.
0:19:54 That’s natural and that needs to be listened to very closely.
0:19:57 And then two, that academic paper that you just talked about,
0:20:00 that wasn’t accompanied by a massive press campaign,
0:20:02 two massive popular press book campaigns.
0:20:06 And so it doesn’t get you in the opinion section of the times.
0:20:08 And so I think that part of that double take
0:20:10 is in part manufactured.
0:20:12 But I think that actually we’re at a local maximum
0:20:15 in terms of what many people call the tech lash
0:20:17 or the backlash against technology.
0:20:20 There was kind of a high watermark in the other direction
0:20:23 in 2011 with Terrier Square and the Arab Spring.
0:20:25 And then through this node-in period,
0:20:26 through Cambridge Analytica,
0:20:28 there was a fairly rapid swing
0:20:31 that brings us all the way to the kind of scholarship
0:20:32 that you were asking about.
0:20:35 But to get to the why I think that we’re quote, unquote,
0:20:38 right here is that the work that I do and others,
0:20:40 it is work that is not as glamorous.
0:20:43 And I would argue that there’s an inverse relationship
0:20:45 between how well a study is done in this field
0:20:49 and how shocking the results might seem
0:20:51 when you cross your T’s and dot your I’s,
0:20:54 when you share your code and you share your data,
0:20:56 you don’t find the kinds of things
0:20:57 that you or I would have
0:21:00 as kind of a preexisting bias about tech.
0:21:01 – Can you give me an example
0:21:03 that a lay person could understand
0:21:05 of the methodological shortcomings
0:21:08 or failures of the research you’re talking about?
0:21:11 – Phones, screen time, social media,
0:21:13 these are all catch all categories.
0:21:16 And so unless you’re epistemically humble
0:21:19 about what you’re measuring and what you’re not measuring,
0:21:22 it can be very tempting to claim
0:21:27 that when a parent or a kid is filling out a questionnaire
0:21:30 that asks, think back on the last year of your life,
0:21:32 on average, how many hours a day do you spend
0:21:36 with a computer, a smartphone, a game console,
0:21:40 and a cell phone, and then call that screen time
0:21:42 in the method section of your paper,
0:21:45 and then call the title of your paper,
0:21:49 “Social Media and Its Impact on Teenage Girls.”
0:21:51 And then you also have the measurement problem,
0:21:54 and this pains me very deeply as a psychologist,
0:21:57 of what the heck is mental health or wellbeing,
0:21:59 because saying you’re sad,
0:22:01 or saying you’re satisfied with your life,
0:22:03 is very, very different than turning up
0:22:06 in a therapist’s office and being diagnosed
0:22:09 as having major depressive disorder or anxiety disorder.
0:22:11 And all of those little decisions
0:22:13 of how you deal with measurement,
0:22:16 those actually can be quite consequential
0:22:18 if all of these arbitrary choices
0:22:20 that are made by a research analyst,
0:22:22 if they’re consistent with the researchers’
0:22:25 preexisting biases or they’re consistent
0:22:27 with the topic of their popular book,
0:22:30 it’s a bit of a red flag or at least a yellow flag.
0:22:32 When you think of everyday behaviors
0:22:35 as being potentially pathological or potentially addictive,
0:22:40 it’s a fairly slippery slope to pathologize everything.
0:22:43 I mean, video gaming is as damaging to mental health
0:22:45 as beds are to mental health.
0:22:48 You could create a disorder called bed addiction disorder,
0:22:52 where people have low affect and low mobility
0:22:54 and get bed sores.
0:22:56 And you could try to regulate the sale of beds
0:22:59 and haul bed companies in front of Congress.
0:23:00 A lot of the evidence you’d have
0:23:02 that something’s bad for you.
0:23:03 It’s not built on a basis of chemistry
0:23:06 or biochemistry or biology.
0:23:08 – You’ve made a similar argument in the past
0:23:12 saying that the effects of digital technology on teenagers
0:23:15 are about as big as the effects of eating potatoes.
0:23:18 – Well, yeah, social scientists will often try to draw
0:23:21 inferences about the population level effects
0:23:24 or associations that might link any given activity
0:23:25 to a health outcome.
0:23:28 And the problem with that is that if you
0:23:30 aren’t crossing your T’s and dotting your I’s,
0:23:33 you can interpret the noise in a data set
0:23:35 in a way that’s consistent with your preexisting
0:23:36 beliefs or biases.
0:23:38 And that’s what happens with technology.
0:23:40 We were trying to make a very simple point,
0:23:42 which is that if you don’t have your hypotheses
0:23:44 before you look at your data,
0:23:47 you can be led astray by very small effects,
0:23:50 whether it’s left-handedness, wearing glasses,
0:23:52 enjoying bicycling.
0:23:55 These are all things that have the same quote-unquote effect,
0:23:58 which is really just a correlation in a large data set.
0:24:00 – I think one interesting thing about this topic,
0:24:03 which is why it’s got so many people riled up,
0:24:05 is that this is a technology
0:24:08 that just about everybody has experience with.
0:24:10 So they can sort of fill in the blank
0:24:12 for the causal mechanisms, right?
0:24:16 They can say, oh, I’ve had the experience on my phone
0:24:19 where rather than going to that event tonight
0:24:20 at my place of worship,
0:24:23 or rather than going to play soccer with my friends,
0:24:27 I get caught up in my silo of anxiety and depression
0:24:28 on the phone.
0:24:32 And this lets people layer their own experiences
0:24:34 onto the moral panic argument.
0:24:37 But aren’t personal experiences in general
0:24:41 a good starting point for a lot of social science research?
0:24:43 – I think that they’re a really important starting point,
0:24:46 but I think they’re a highly invalid ending point,
0:24:49 and they shouldn’t be where people’s thinking stops.
0:24:50 It’s very easy to think
0:24:53 that there’s a digital world and an analog world,
0:24:55 and these worlds don’t connect.
0:24:58 But the problem isn’t necessarily that one is good or bad.
0:25:01 The problem is that we’re putting this wedge between them.
0:25:02 And when we put that wedge between them,
0:25:05 what we do is we cut ourselves off from
0:25:07 being able to investigate really interesting questions
0:25:09 about actually what is happening to someone
0:25:11 if they’re depressed and they’re using social media.
0:25:14 There’s a gigantic difference between feeling unhappy
0:25:16 with how you spent your evening
0:25:18 and suffering from agoraphobia,
0:25:20 having major depressive disorder,
0:25:22 or some form of crippling social anxiety.
0:25:24 I’m sure you know people who have suffered,
0:25:26 and many people who are listening,
0:25:29 know people who suffered with opioid addiction
0:25:30 or depression or anxiety.
0:25:32 And claiming that the thing that happens
0:25:35 when you play a video game is like what happens
0:25:37 when you can’t stop taking opioids,
0:25:40 or saying, I don’t feel like going outside,
0:25:43 I have had a bad Twitter argument, I feel so anxious.
0:25:46 Saying that’s the same thing as crippling agoraphobia,
0:25:48 where people won’t leave their home for years.
0:25:50 Frankly, that’s insulting,
0:25:53 because there’s no reason to think that’s the same mechanism.
0:25:59 – Over the last seven or eight years,
0:26:02 I have struggled with what I am quite comfortable
0:26:05 calling an internet addiction.
0:26:07 – That is Lauren Euler,
0:26:10 a novelist and cultural critic in her early 30s.
0:26:14 She grew up in West Virginia and now lives in Berlin.
0:26:17 Euler has spent a lot of time thinking and writing
0:26:19 about life online and off.
0:26:22 – I have met great friends on social media,
0:26:23 I’ve gotten boyfriends on social media,
0:26:26 but at the same time, I do strongly feel
0:26:28 that it can produce an actual addiction,
0:26:30 like a straightforward addiction, like a drug addiction.
0:26:33 And I know that because since I divested
0:26:35 from social media, I’ve picked up smoking
0:26:38 and it’s the same sort of thought process
0:26:41 about like compulsively wanting to either look at Twitter
0:26:43 or like smoke a cigarette.
0:26:45 Like you train yourself to think,
0:26:46 something could be happening on my phone.
0:26:47 Someone could have written me,
0:26:50 some news could have broken and I need to see it.
0:26:52 – Euler recently published a collection of essays
0:26:54 called “No Judgment”.
0:26:57 One of the pieces is titled “My Anxiety”.
0:26:59 – I think anxiety does create a barrier
0:27:02 around actually doing things that you’re sitting
0:27:04 and worrying about and sitting and looking at your phone
0:27:09 is just kind of externalizing this worrying feeling, right?
0:27:11 Like your phone is doing the worrying for you
0:27:14 because it has this short attention span.
0:27:15 My thoughts are like bouncing around
0:27:16 and I might be thinking about taxes
0:27:17 and then I’m thinking about some guy
0:27:19 and then I’m thinking about my late article
0:27:21 that I need to turn in and I’m thinking about,
0:27:22 I embarrass myself at a party
0:27:25 and all this happens in the span of two minutes,
0:27:27 which if you’ve been on social media,
0:27:29 that is exactly what it’s like, right?
0:27:32 – So Lauren Euler does see a deep connection
0:27:33 between her smartphone
0:27:35 and what researchers like Danny Blanchflower
0:27:37 call mental ill-being.
0:27:42 But again, is the one necessarily causing the other?
0:27:44 – I get anxious that I’ll become depressed
0:27:46 and depressed that I’m so anxious.
0:27:49 And if the phone is a kind of conduit for feeling
0:27:52 or if it’s an extension of your brain,
0:27:54 then of course like the things on your phone
0:27:55 might make you depressed
0:27:57 just as they might make you anxious.
0:28:00 So I think on one hand, of course it’s causal
0:28:03 but I do think it’s sort of self-defeating
0:28:07 and short-sighted to suggest that the phone
0:28:12 is the only cause of this mental health crisis
0:28:14 among teenagers or among anyone.
0:28:17 Because that’s just not how the world works.
0:28:19 That’s how op-eds work.
0:28:20 Like people who are desperate
0:28:22 for one weird trick to save your life,
0:28:23 it’s like I just needed to throw my phone in the river
0:28:25 and then everything would end up fine.
0:28:27 It’s just not how the world works.
0:28:32 – Okay, so if there is rising anxiety among young people
0:28:37 and if throwing your phone in the river isn’t a solution,
0:28:38 what is?
0:28:40 That’s coming up after the break.
0:28:43 I’m Steven Dubner and this is Freakonomics Radio.
0:28:52 Andy Shibilsky at Oxford
0:28:57 argues that the panic over smartphones is overblown.
0:29:01 But that doesn’t mean he thinks everything is fine.
0:29:04 In the U.S., there are some really worrying trend lines
0:29:06 for people across the population.
0:29:08 And with the most worrying trends,
0:29:11 things like self-harm and death by suicide,
0:29:16 I would be much more worried as a middle-aged man
0:29:18 in Appalachia with access to a gun
0:29:21 and with access to West Virginia’s social safety nut
0:29:25 than I would be as a white teenage girl anywhere in the U.S.
0:29:27 And so I would say that there probably
0:29:29 is something broken in America.
0:29:31 Here’s something I also wonder about.
0:29:35 There is so much more awareness around mental health now
0:29:37 than there was 20, 30 years ago
0:29:39 and so much less stigma.
0:29:41 Some people have argued that at least.
0:29:44 Do you think that that acceptance might be showing up
0:29:47 in the data as more young people today saying,
0:29:50 yes, I’ve experienced mental health problems
0:29:51 than they used to?
0:29:53 – There’s a concept called measurement variance.
0:29:56 This is the basic idea that if I asked you
0:29:58 how I’m feeling today and then I compare
0:30:01 how I’m feeling today to how I said I felt 10 years ago,
0:30:02 the way that people answer that question
0:30:04 actually changes over time.
0:30:07 And so there’s a bit of that, absolutely.
0:30:09 And then the way that we actually track
0:30:10 things like mental health problems
0:30:12 in the U.S. in particular,
0:30:14 it’s very, very susceptible to change
0:30:18 as a function of changes in insurance
0:30:21 and changes in way that diseases can be classified.
0:30:25 – On this note, Shabilsky points to a study
0:30:27 by two health economists,
0:30:30 Adriana Cortador Waldron and Janet Curry.
0:30:31 Their paper is called,
0:30:34 to what extent are trends in teen mental health
0:30:37 driven by changes in reporting?
0:30:39 – It’s a study of clinical intakes
0:30:43 of young women in New Jersey across the last 10 years.
0:30:45 There’s a series of spikes and hospital admissions
0:30:47 for different types of serious disorders
0:30:49 when a young person shows up in crisis.
0:30:52 And one way that you could have interpreted the data
0:30:54 is to say that there is a mental health epidemic
0:30:56 in New Jersey for teenage girls in particular,
0:30:59 but the researchers went back and looked at changes
0:31:02 to the best practices for clinical intake.
0:31:05 And the peaks occurred in response to two changes.
0:31:09 The first is a special awareness campaign for clinicians,
0:31:11 for focusing on young women
0:31:13 as a target area for intervention.
0:31:15 And then the second was allowing there
0:31:18 to be a second reason for clinical intake.
0:31:19 So it used to be when you came in,
0:31:23 you would just say the person is schizotypal or depressed,
0:31:26 but then you could say schizotypal and depressed.
0:31:29 And so when that change went through, the amounts went up.
0:31:30 But, and here’s the rub,
0:31:35 it only went up for young people who were on insurance
0:31:38 because it created a new billable category.
0:31:40 It didn’t go up for those who didn’t.
0:31:42 So I would say when you’re a hammer,
0:31:44 the problems of the world are nails.
0:31:47 I think it’s very tempting to think that the atomization
0:31:50 of certain Western cultures is a result of technology
0:31:53 and not other things like a decline in religiosity
0:31:55 or these other kinds of things.
0:31:57 There aren’t gonna be quick fix solutions.
0:31:58 – Like throw away your phones.
0:32:01 – Throwing away your phones or putting them a locker.
0:32:03 What you’re going to have to do
0:32:05 is actually rebuild communities.
0:32:08 You’re gonna have to hire psychotherapists.
0:32:11 You’re gonna have to build civil and economic infrastructure
0:32:12 to support those in your society
0:32:14 who are falling through the cracks.
0:32:16 – What if I say to you, I agree with you
0:32:19 and that sounds smart and sane,
0:32:22 but it also is going to take a long time
0:32:24 and a lot of money, a lot of resources.
0:32:27 And in the meantime, I believe there are millions,
0:32:28 maybe billions of people suffering
0:32:30 and so I wanna do something quick and dirty.
0:32:32 What would you say to that?
0:32:34 – Law of unintended consequences.
0:32:37 I would say that if you can’t properly diagnose the problem,
0:32:39 you can’t possibly solve it.
0:32:42 I would say that if you take this topic seriously,
0:32:45 you would take slow and incremental steps
0:32:48 to ascertain exactly what problem you’re trying to solve.
0:32:51 And so yeah, I would say that you should not waste energy
0:32:52 on patting yourself on the back.
0:32:55 – What if I ask you to make an argument
0:32:58 for the benefits of digital tech that are overlooked
0:33:02 in the rush to proclaim all of digital technology
0:33:03 damaging to young people especially?
0:33:06 – One of the things that I think everyone can agree
0:33:08 is really worrying is content online
0:33:11 around self-harm or anorexia.
0:33:15 And it might seem like a very obvious thing
0:33:18 that platforms should be doing absolutely all they can do
0:33:22 to take down this content and to scuttle these communities
0:33:25 and make sure that it’s not on any of the big platforms.
0:33:27 That might seem like common sense.
0:33:30 That might feel like the right thing to do.
0:33:33 But you can wind up causing so much more harm than good.
0:33:35 If you don’t pay attention to actually how people
0:33:39 who struggle with these disorders use social media platforms,
0:33:41 there are really good support groups
0:33:45 for things like self-harm, suicidal thoughts,
0:33:47 and things like anorexia and bulimia.
0:33:48 One of the things that social media platforms
0:33:51 can be really good for is connecting communities
0:33:53 in a positive way that self-moderates
0:33:56 and they provide resources for people in crisis.
0:34:00 And what happens if you think that it’s the responsibility
0:34:02 of these companies to just crush it
0:34:04 so that it’s not there anymore?
0:34:05 Two things will always happen
0:34:07 and they’re both very worrying.
0:34:09 The first is people who are already struggling,
0:34:12 they will begin to stop disclosing.
0:34:13 They’ll button themselves up.
0:34:16 They won’t reach out and they’ll lose access
0:34:18 to the online social environments
0:34:20 that have been supported for them.
0:34:22 And then the second thing is the community moves
0:34:24 to other parts of the internet
0:34:27 that are more poorly moderated and are far less safe.
0:34:29 And so you wind up in a telegram group
0:34:31 where there’s no accountability
0:34:34 instead of being in a Facebook or Instagram group.
0:34:36 And so I think thinking about tools
0:34:38 to meet young people where they are
0:34:39 is a much smarter idea
0:34:42 than just burning everything down and salting the earth.
0:34:47 – As you just heard, Andy Shibilsky is more positive
0:34:49 toward the big social platforms
0:34:51 than many people in this arena.
0:34:54 In fact, he recently announced a research collaboration
0:34:58 with the Facebook and Instagram parent company, Meta.
0:35:00 – We’re going to solicit proposals
0:35:01 from researchers all around the globe.
0:35:03 They can get data on young people
0:35:05 in 40 different countries
0:35:06 and combine that with information
0:35:09 about their mental health and their wellbeing.
0:35:12 But at no point will they have a direct relationship
0:35:14 in terms of the selection of their projects
0:35:16 or the decision for the projects to go forward
0:35:18 with the researchers at Meta.
0:35:20 We’ve put in a series of firewalls
0:35:23 between the data scientists at Meta
0:35:25 and researchers who would be very interested
0:35:28 in investigating the idea that Instagram
0:35:31 relates to the mental health and the wellbeing of teenagers.
0:35:34 Myself and my co-editors will be selecting projects
0:35:37 on the basis of their scientific merit
0:35:39 and their proposals will be peer reviewed
0:35:41 in advance of data collection
0:35:43 by researchers who are specialists
0:35:44 in topics like mental health,
0:35:47 topics like online engagement,
0:35:50 but also generalists who have a sense
0:35:52 of what is good methodology.
0:35:55 And only when that peer review process finishes,
0:35:57 the data request goes directly to Meta.
0:36:00 The idea here is you have these firewalls in place.
0:36:04 You don’t have big tech negatively affecting the process
0:36:05 and you don’t have researchers
0:36:07 potentially moving the goalposts
0:36:10 because they think they’ve found something interesting
0:36:12 after having a rummage through the data.
0:36:15 – It might surprise you
0:36:17 that the rigorous analysis Shabilsky is talking about
0:36:20 hasn’t already been done on Instagram and Facebook
0:36:22 and other social media platforms,
0:36:27 but it’s worth remembering how new this ecosystem is.
0:36:30 Here again is the writer Lauren Euler.
0:36:33 – The social media era has been really short, right?
0:36:35 Like when was MySpace founded?
0:36:38 When was Facebook founded in the early 2000s, right?
0:36:39 That’s not very long.
0:36:41 And already Facebook, nobody uses that.
0:36:43 They had to buy Instagram and WhatsApp
0:36:44 and we’re gonna say relevant.
0:36:47 And Instagram is actually quite young, right?
0:36:49 I think it’s like 20, what?
0:36:51 2010, yeah, that’s really young.
0:36:54 So I look forward to a day when Instagram doesn’t exist,
0:36:57 but I’m sure everything will take its place.
0:36:58 – I’m guessing Euler is right,
0:37:01 that the platforms will change
0:37:04 and the way we interact with our phones will change.
0:37:07 Some people, like Andy Shabilsky,
0:37:10 believe that if you can’t properly diagnose the problem,
0:37:12 you can’t possibly solve it
0:37:16 and therefore to rush into a hard anti-phone position
0:37:18 isn’t the right move.
0:37:22 Other people, like Danny Blanchflower,
0:37:23 say that we can’t afford to wait
0:37:26 because if the rise in anxiety and unhappiness
0:37:30 among young people is as significant as he thinks,
0:37:33 the consequences will be significant too,
0:37:34 especially the kind of consequences
0:37:37 that economists like to consider.
0:37:40 – I worry about what will happen to these young people
0:37:41 as they come to the labor market.
0:37:44 And then later on, we’re gonna worry about their mortality
0:37:45 and we’re gonna worry about their ability
0:37:48 to generate savings and investment
0:37:51 and buy a house and buy themselves a retirement package.
0:37:55 So down the road, these are issues that we care about.
0:37:57 – You are never going to run out of problems
0:37:58 to address, are you?
0:38:00 – In a way, that’s what I’ve tried to do.
0:38:02 Think about how we address the problems
0:38:04 and try and find solutions.
0:38:05 I don’t know what works.
0:38:07 I’ve been asked that question, we go,
0:38:09 “We don’t know, but we’re gonna try and find out for you.”
0:38:10 That’s what we do.
0:38:14 Do me a favor, when you find something, call me
0:38:16 and we’ll talk about it again, okay?
0:38:18 – Love to, love to.
0:38:22 – Thanks to Danny Blanchflower, Andy Shibilski
0:38:25 and Lauren Euler for their insights today.
0:38:27 And I’m curious to know what you think.
0:38:31 Our email is radio@freakonomics.com.
0:38:33 Coming up next time on the show,
0:38:37 do you ever watch an NFL game and wonder why
0:38:39 when so many of the players are black,
0:38:42 so many of the coaches are white?
0:38:44 The NFL itself wondered that,
0:38:47 and two decades ago, they put in a rule to help.
0:38:49 – The Rooney Rule is really about making sure
0:38:51 that you have a diverse slate
0:38:54 when you’re selecting and hiring people.
0:38:55 – Did it help?
0:38:57 We tell the history of the Rooney Rule
0:39:00 with the help of an actual Rooney.
0:39:02 – You know, if he was in the Senate,
0:39:04 we’d call him Majority Whip.
0:39:06 – We look at the successes and failures,
0:39:08 the lawsuits and the sham interviews,
0:39:12 and we ask how the Rooney Rule works outside of football.
0:39:13 That’s next time on the show.
0:39:15 Until then, take care of yourself.
0:39:18 And if you can, someone else too.
0:39:21 Freakonomics Radio is produced by Stitcher and Renbud Radio.
0:39:24 You can find our entire archive on any podcast app
0:39:26 also at freakonomics.com
0:39:29 where we publish transcripts and show notes.
0:39:31 This episode was produced by Teo Jacobs.
0:39:34 Our staff also includes Alina Coleman, Augusta Chapman,
0:39:38 Delvin Appalachie, Eleanor Osborn, Ellen Frankman,
0:39:40 Elsa Hernandez, Gabriel Roth, Greg Rippin,
0:39:43 Jasmine Klinger, Jeremy Johnston, John Schnarrers,
0:39:45 Julie Canfer, Lyork Bowditch, Morgan Levy,
0:39:48 Neil Coruth, Rebecca Lee Douglas, Sarah Lilly,
0:39:49 and Zach Lipinski.
0:39:52 Our theme song is “Mr. Fortune” by the Hitchhikers.
0:39:55 Our composer is Luis Guerra.
0:39:57 As always, thanks for listening.
0:40:01 – I’m Dr. Andy Shibilski.
0:40:04 I’m the University of Oxford’s professor of,
0:40:06 oh my God, what am I the professor of?
0:40:08 – Human behavior and technology?
0:40:10 – Yeah, exactly, but I was so hung up
0:40:12 on if I’m Andy or Andrew.
0:40:18 – The Freakonomics Radio Network.
0:40:20 The hidden side of everything.
0:40:24 Stitcher.
0:40:27 (gentle music)
0:40:29 you
0:00:08 I am going to say a dirty word, screen time.
0:00:13 That simple and suddenly terrifying word
0:00:15 includes all the hours that you spend streaming movies,
0:00:18 playing video games, especially the hours you spend
0:00:21 caught in the vortex of social media and news
0:00:25 and communication that gushes from your phone.
0:00:28 Everyone seems to have something to say about screen time
0:00:30 from the American Academy of Pediatrics
0:00:33 to religious and educational leaders
0:00:35 to the New York Times opinion section,
0:00:38 especially the New York Times opinion section.
0:00:42 Our own devices scold us with reminders to get off Instagram,
0:00:46 to hold the phone further away, to avoid eye strain.
0:00:48 If you have an iPhone, you know that it tracks
0:00:50 the hours you spend on it every day
0:00:53 and then makes a point to tell you that number,
0:00:57 that always surprisingly large number.
0:00:58 The people we are most worried about
0:01:02 when it comes to screen time are children and teenagers.
0:01:05 Parents around the world are in a full-on panic
0:01:07 about the relationship between screen time
0:01:10 and mental health, in part because there is
0:01:13 new scientific evidence on this relationship.
0:01:18 What we’ve observed from around 2014 was high school kids,
0:01:20 we were seeing a rise in their use of internet
0:01:23 and smartphones and so on, and a big rise
0:01:25 in their anxiety levels, depression levels,
0:01:26 all kinds of things.
0:01:30 That sounds terrible, but is the case
0:01:33 against screen time really that clear cut?
0:01:37 Today on Freakonomics Radio, we will hear the evidence
0:01:40 and some challenges to the evidence.
0:01:43 It’s very easy to fool yourself as an analyst.
0:01:46 We’ll also discuss the incentives at play.
0:01:49 That’s how op-eds work, like people are desperate
0:01:51 for one weird trick to save your life.
0:01:54 Still, you’d have to be a cold-hearted person
0:01:56 to think that nothing should change.
0:01:59 The question is, well, here’s the question.
0:02:02 – What the hell are we gonna do about it?
0:02:04 (upbeat music)
0:02:15 – This is Freakonomics Radio,
0:02:19 the podcast that explores the hidden side of everything
0:02:21 with your host, Stephen Dubner.
0:02:24 (upbeat music)
0:02:27 (upbeat music)
0:02:32 – David Blanchflower, his friends call him Danny
0:02:34 after the much-loved British footballer,
0:02:37 is himself British, but he has spent the past few decades
0:02:39 at Dartmouth College in New Hampshire.
0:02:42 He is an economics professor and researcher
0:02:45 who has also been involved in economic policy making.
0:02:47 – The way I would describe it is I’m a data guy,
0:02:49 and I look at patterns in the data, and I say,
0:02:52 let’s kick and poke and punch the data as hard as we can,
0:02:53 and let’s see if we get an answer.
0:02:55 – Blanchflower is a pioneer of what some people
0:02:57 call happiness economics.
0:03:01 He is particularly well-known for one key finding.
0:03:03 – In any well-being equation I ever looked at,
0:03:05 there was a midlife crisis.
0:03:08 There was a U-shape in age in the data.
0:03:12 – This U-shape is known as the happiness curve.
0:03:14 It is derived from many studies over the years
0:03:17 using survey data from thousands of people
0:03:19 of all ages from around the world.
0:03:22 Now, you may know how we feel about survey data
0:03:26 around here that it’s not the most reliable form of data.
0:03:28 On the other hand, it’s hard to find a better way
0:03:32 to measure something as intangible as happiness.
0:03:35 Anyway, here’s what Danny Blanchflower has found.
0:03:38 – Basically, the happiest people of the young,
0:03:40 it takes a U-shape, it diminishes down
0:03:43 to around 50 years of age, and then it rises again.
0:03:45 – There are a number of stories you could tell
0:03:47 about why happiness has a U-shape
0:03:49 over the course of a lifetime.
0:03:54 – For many young people, life is fun and relatively easy.
0:03:57 Adulthood, meanwhile, brings challenges,
0:03:59 obligations, quite possibly offspring,
0:04:02 and those offspring in turn bring more challenges
0:04:04 and obligations.
0:04:09 But then, happiness begins to rise again at around age 50,
0:04:12 an age at which many parents are no longer
0:04:14 so obligated to their children.
0:04:17 Is that a coincidence?
0:04:19 Maybe, maybe not.
0:04:21 There are other possible drivers of that upswing
0:04:23 and happiness in middle age.
0:04:27 You might just become more mature over time, more satisfied.
0:04:30 You might become more comfortable with who you are,
0:04:33 or at least better at managing your expectations.
0:04:37 Danny Blanchflower has written more than 30 papers
0:04:38 about this happiness curve,
0:04:42 and his findings have been replicated over 600 times,
0:04:43 an impressive record when you’re trying
0:04:46 to measure an emotional state.
0:04:49 The scientific consensus was so strong
0:04:51 that many researchers had come to see the happiness curve
0:04:56 as a natural part of human life, at least Blanchflower had.
0:05:00 But then, one day, he came across the work of Jean Twenge,
0:05:03 a psychology professor at San Diego State University.
0:05:05 – She was interviewed in the New York Times,
0:05:09 and read that interview, and I started to read the papers,
0:05:12 and the papers were about things that I had written about.
0:05:15 She’d used some of the data that I’d used.
0:05:18 But Twenge’s work didn’t fit the happiness curve
0:05:20 that Blanchflower had made famous.
0:05:23 Her work showed that happiness starts to decline
0:05:25 before adulthood.
0:05:27 She showed that teenagers in particular
0:05:30 were reporting much higher rates of unhappiness
0:05:31 than in the past.
0:05:33 – I started to look at the data,
0:05:35 and I realized that I had missed something.
0:05:38 I mean, I wrote 30 papers saying
0:05:40 there was a hump shape in age, or a U shape.
0:05:43 I literally said this is one of the most phenomenal facts
0:05:46 in social science, and then I wrote until it wasn’t.
0:05:48 – So what did Blanchflower miss?
0:05:51 – So now, instead of the hump shape, we have a decline.
0:05:54 – Meaning more people reported being unhappy
0:05:56 at an earlier age than he had seen before.
0:05:59 – That’s only really started since 2015,
0:06:01 and I missed it, and the world missed it.
0:06:03 Do you know why we missed it?
0:06:05 Because a bit of data took a while to come in,
0:06:08 so we didn’t really get data until about 2018, 2019,
0:06:10 and then COVID came.
0:06:12 And everybody said, “Oh, look at the effects of COVID.”
0:06:15 Well, it turns out now we have to rethink it,
0:06:17 because a lot of it was actually occurring prior to COVID,
0:06:21 and COVID merely extended pre-existing trends,
0:06:22 and we didn’t know it.
0:06:24 – Blanchflower, to his credit,
0:06:28 became convinced that his famous old happiness curve
0:06:30 no longer fit the current reality.
0:06:33 So now he began to, as he would say,
0:06:35 kick and poke and punch the data.
0:06:39 – The data that I like is from the CDC.
0:06:41 People are asked the following question.
0:06:43 Over the last 30 days,
0:06:46 how many of those 30 were bad mental health days?
0:06:48 Most people just say no days.
0:06:49 – But that’s most people.
0:06:52 One group of people gave the opposite answer.
0:06:55 They said that every day of the previous 30
0:06:57 was a bad mental health day.
0:06:59 – So what we’ve seen over time in America
0:07:01 is that that has increased by quite a lot.
0:07:05 The group that has increased the most are young women.
0:07:09 So today, 10% of young women report,
0:07:12 and 7% of young men report that every day of their lives
0:07:14 is a bad mental health day.
0:07:16 We basically see this problem of a collapse
0:07:18 in the well-being of the young in America,
0:07:19 and then we see it in the UK.
0:07:21 So the UN says to me, Danny, we gotta look at the world.
0:07:23 Let’s go look at the world.
0:07:23 – That’s right.
0:07:25 Blanchflower’s happiness research
0:07:28 had caught the attention of the United Nations.
0:07:31 He and other researchers were asked to find out more.
0:07:34 – So off we go, and now basically we find
0:07:36 it’s true everywhere.
0:07:38 And so we’re talking about what I think
0:07:40 is a huge global crisis.
0:07:42 But the genie has been released from the bottle,
0:07:44 and it was released from the bottle a decade ago
0:07:46 before we realized what was going on.
0:07:49 – So what happened a decade ago
0:07:53 that caused this decline in mental health among young people?
0:07:56 For Gene Twenge, the San Diego state psychologist,
0:07:58 there is an obvious and easy answer.
0:07:59 The smartphone.
0:08:02 Apple introduced the iPhone in 2007
0:08:06 and sold around 1.4 million units that year.
0:08:10 These days, Apple sells more than 200 million iPhones a year,
0:08:12 and that’s just Apple.
0:08:15 For every iPhone, four or five non-Apple smartphones
0:08:19 are sold around the world a billion a year.
0:08:24 In 2017, Twenge published a book called “Deep Breath” here.
0:08:26 I, Gen, why today’s super connected kids
0:08:30 are growing up less rebellious, more tolerant, less happy,
0:08:34 and completely unprepared for adulthood,
0:08:37 and what that means for the rest of us.
0:08:39 Twenge also published some related pieces
0:08:41 in “The Atlantic” and “The New York Times,”
0:08:43 which amplified her argument.
0:08:47 In early 2024, the influential social psychologist
0:08:50 Jonathan Haidt, who teaches at New York University,
0:08:54 joined this argument with a book called “The Anxious Generation,”
0:08:56 how the great rewiring of childhood
0:08:59 is causing an epidemic of mental illness.
0:09:03 Here is Haidt on MSNBC discussing this epidemic.
0:09:06 Everyone has a theory about what causes it.
0:09:07 There is only one explanation.
0:09:09 There is no other theory that can make sense
0:09:11 of a synchronized global collapse in mental health,
0:09:14 other than the fact that in 2010,
0:09:16 the great majority of kids had a flip phone,
0:09:20 no high-speed internet, no unlimited data, no Instagram.
0:09:23 And by 2015, we all have a smartphone,
0:09:26 high-speed internet, unlimited data,
0:09:28 Instagram, front-facing camera.
0:09:31 Haidt makes a compelling argument, as does Gene Twenge,
0:09:34 and their argument has resonated with many people,
0:09:36 parents especially.
0:09:38 There have been suggestions that social media
0:09:42 and photo-sharing apps can be particularly damaging
0:09:43 to adolescent girls and young women
0:09:48 by amplifying pressure and anxiety around their appearance.
0:09:52 But is the smartphone really the one explanation,
0:09:54 as Haidt puts it?
0:09:56 As we often preach around here,
0:09:59 correlation does not equal causation.
0:10:02 Maybe there are some other factors contributing
0:10:04 to the anxiety and unhappiness
0:10:07 that young people say they are feeling.
0:10:09 For instance, the last couple decades have brought
0:10:13 wave after wave of political and economic turmoil,
0:10:16 sometimes tipping into chaos.
0:10:18 If you have parents or grandparents
0:10:20 who grew up during the Great Depression,
0:10:23 you know how much they were shaped by that experience.
0:10:25 So have young people today maybe been shaped
0:10:27 by all that chaos?
0:10:30 And let’s not forget the widespread anxiety
0:10:31 over climate change.
0:10:35 And for kids who grew up in the US, especially New York,
0:10:40 the 9/11 attacks set a tone of fear and anger.
0:10:41 There’s also this.
0:10:45 What about the benefits of the smartphone?
0:10:47 Anti-phone advocates have done a good job
0:10:50 pointing out the costs, but let’s not forget the benefits.
0:10:52 The smartphone provides connection,
0:10:55 it facilitates the sharing of interests,
0:10:58 helps you navigate to pretty much anywhere,
0:11:02 and gives access to just about any piece of information
0:11:05 or music or whatever else you might want.
0:11:07 Of course, this can be overwhelming
0:11:10 and if you consume too much of it, you may get sick,
0:11:12 just as you’ll get sick consuming too much of anything
0:11:15 that humans consume, like food.
0:11:19 But just as food is pretty important for humans,
0:11:21 so is connectivity.
0:11:24 And to discount the benefits, especially for young people,
0:11:26 may be short-sighted.
0:11:30 So I went back to the economist Danny Blanchflower
0:11:33 to talk through the causality piece of this.
0:11:36 If I asked you to give the strongest piece of evidence
0:11:39 that you’re convinced that this is a causal relationship
0:11:41 between the rise of, let’s call it digital tech use,
0:11:44 and the rise of anxiety or depression,
0:11:47 how do you know that that instrumental variable
0:11:50 is the one that’s driving the bulk of that problem?
0:11:52 Because I could posit a variety of other things.
0:11:52 – Yeah, of course.
0:11:55 If we throw one at you just off the top of my head,
0:11:58 your Dartmouth colleague, Bruce Sassardot,
0:12:01 has done research on the rise of negativity in media,
0:12:03 especially in the U.S.
0:12:07 And we know the effect of media on the psyche.
0:12:10 So could it be that what we’re thinking of
0:12:14 as, quote, “digital tech” being the driver is in fact,
0:12:17 digital tech is in this case more of a delivery system
0:12:20 for a massive wave of negativity,
0:12:24 also a massive wave of information about mental health
0:12:27 that may cause more people to self-diagnose and so on?
0:12:28 – What a great question.
0:12:30 Actually, Bruce and I were working on this together.
0:12:32 I was just talking to him 20 minutes ago.
0:12:35 So obviously there’s a set of questions, what is it?
0:12:39 And I sort of posit as a policymaker,
0:12:41 well, maybe it is something else,
0:12:44 but this thing appears to disproportionately impact the young
0:12:47 and above that disproportionately,
0:12:49 it impacts young women.
0:12:52 This trend started prior to COVID.
0:12:55 Every piece of evidence we have is that it doesn’t appear
0:12:57 to have been caused by the Great Recession.
0:13:00 So all the things that you’ve talked about, that is true,
0:13:03 but why would it especially be true of women?
0:13:04 And you have to get the timing right.
0:13:07 So the timing fits, the rise we observe
0:13:11 in the ill-being of the young starts around 2014.
0:13:14 At exactly that time, you see the explosion of digital usage,
0:13:17 internet usage, smartphone usage, and so on.
0:13:21 And I think in the end, the causal question is sort of irrelevant
0:13:24 in the sense that here we have a problem, right?
0:13:26 The worry is what do you do about it, right?
0:13:28 I mean, I’m obviously concerned
0:13:30 that this declining wellbeing of the young
0:13:33 will translate itself into something bad.
0:13:35 We’ve seen things like rising self-harm,
0:13:38 evidence of rising thoughts of suicide,
0:13:40 but to this point, thank God,
0:13:43 we haven’t yet seen lots of bad outcomes.
0:13:45 The people who have the least incidence of deaths
0:13:47 by a drug overdose are the young.
0:13:49 I don’t see much evidence of an increase there.
0:13:52 And the evidence around the world is that in the US,
0:13:54 suicides have risen slightly for young men,
0:13:55 not in other places.
0:13:58 – Wait a minute, this seems like a pretty big deal here.
0:14:00 You’re saying this engagement with smartphones
0:14:02 is leading to, let’s just call it generally,
0:14:05 declining mental health among the young.
0:14:07 Could it be that this is a kind of gantlet
0:14:09 that young people now go through
0:14:13 and emerge relatively undamaged?
0:14:15 And who knows, maybe even stronger?
0:14:17 – Supposing the answer to that is no,
0:14:18 and we don’t do anything.
0:14:21 We have to err on the side of caution, right?
0:14:23 I mean, I agree with you, it may well be that.
0:14:24 You intervene and you say,
0:14:27 “Well, we were mistakenly intervening
0:14:29 to make sure your child wasn’t in deep trouble.
0:14:31 Who’s your ever rejected that?”
0:14:34 We were wrong, okay, but it’s much better to be wrong
0:14:36 trying to protect them than to do nothing.
0:14:38 And then suddenly we’re overtaken
0:14:41 by a huge splurge in deaths from overdoses.
0:14:43 In a way, the experience that I had
0:14:46 at the Bank of England was interesting.
0:14:47 – Blanche Flower is talking here
0:14:50 about when he served as an external member
0:14:53 on a Bank of England committee that sets interest rates.
0:14:56 In October of 2007, the global economy
0:14:58 suddenly appeared fragile,
0:15:00 perhaps on the brink of a deep recession.
0:15:02 – So I sit there and I make a decision
0:15:04 on interest rates every month.
0:15:07 – Blanche Flower was the first and, at the time,
0:15:09 only person on the committee
0:15:11 to vote in favor of cutting interest rates.
0:15:12 – The hardest thing in the world
0:15:14 is to know where you are,
0:15:15 because you’ve got no clothes dead,
0:15:16 it doesn’t come in for a year.
0:15:18 So you sit and you try and make a decision
0:15:19 on limited information.
0:15:21 – Being out front on interest rates,
0:15:23 Blanche Flower would later say,
0:15:26 was, quote, “not a comfortable place to be,
0:15:28 they called me bunkers.”
0:15:29 – And always you care about,
0:15:31 have I made the right decision?
0:15:35 I’m used to being in a world of incomplete information,
0:15:36 where you have to make a decision.
0:15:38 You’ve got to do something.
0:15:41 – A year later, the Bank of England
0:15:43 finally did cut interest rates.
0:15:46 And in some circles, Danny Blanche Flower
0:15:49 was praised for his early call.
0:15:51 So you could say that Blanche Flower
0:15:53 has a proactive disposition,
0:15:57 which, in the case of interest rates, served him well.
0:15:59 How about in the case of what he calls
0:16:04 the huge global crisis caused by the smartphone?
0:16:05 I’m sure you’ve read about the growing number
0:16:08 of smartphone bans or other restrictions
0:16:11 in schools and elsewhere.
0:16:14 But some people think we shouldn’t rush to judgment.
0:16:18 They argue that the research behind the smartphone panic
0:16:19 isn’t very solid.
0:16:22 After the break, we will hear from one such critic.
0:16:25 – If you can’t properly diagnose the problem,
0:16:26 you can’t possibly solve it.
0:16:27 – I’m Stephen Dubner.
0:16:29 This is Freakonomics Radio.
0:16:30 We’ll be right back.
0:16:33 (gentle music)
0:16:39 (dramatic music)
0:16:43 – The smartphone is a technology
0:16:45 that has changed our society dramatically.
0:16:49 And many people, including the economist Danny Blanche Flower
0:16:52 and the psychologist Jean Twenge,
0:16:54 argue that the smartphone is causing harm,
0:16:56 especially to young people.
0:16:57 This puts the smartphone in the company
0:17:00 of earlier inventions like the telephone,
0:17:03 the television, the bicycle, even electricity.
0:17:05 – It’s really important to acknowledge
0:17:07 that new things are scary.
0:17:09 – That is Andrew Shibilsky,
0:17:13 a professor of human behavior and technology at Oxford.
0:17:15 – New things should cause anxiety.
0:17:17 If someone or something turns up in your tribe,
0:17:19 they try to feed you a novel food.
0:17:21 They wanna take care of your kids.
0:17:23 It makes a lot of sense for your first reaction
0:17:25 to be aversion and to be skeptical.
0:17:27 So it’s absolutely all right to be skeptical
0:17:28 about new technologies.
0:17:31 But what will often happen is that some people will come along
0:17:34 and they’ll kind of give the panic its skeleton.
0:17:37 There’s a class of person called a moral entrepreneur.
0:17:39 What a moral entrepreneur will do
0:17:43 is they’ll identify a at-risk group for a novel technology.
0:17:44 And then you need a mechanism.
0:17:47 You need a reason to say this time it’s different.
0:17:49 Let’s say it’s Dungeons and Dragons
0:17:51 and Teenagers and Satanism.
0:17:55 Then you have a mechanism and that would be role-playing
0:17:59 or the fact that it’s animated or that in radio serials,
0:18:02 there’s actually somebody talking through a crime.
0:18:04 And so when you combine those three things,
0:18:06 moral entrepreneurship thrives.
0:18:08 There’s congressional inquiries,
0:18:09 Senate inquiries about comic books
0:18:12 or Lieberman holds up a Nintendo blaster
0:18:15 and everyone pats themselves on the back.
0:18:16 – Shabilsky is talking about the time
0:18:19 in the early 1990s when US Senator Joe Lieberman
0:18:22 went on a crusade against violent video games.
0:18:25 That may seem like a distant era.
0:18:27 – People get older, they forget.
0:18:29 New technology comes along
0:18:33 and you have the person who is whipping up today’s panic
0:18:35 recalling the good old days
0:18:37 when people would play video games in their basement.
0:18:39 – In one paper that you’ve co-authored,
0:18:41 this is called global wellbeing and mental health
0:18:42 in the internet age.
0:18:45 The abstract reads, in the last two decades,
0:18:47 the widespread adoption of internet technologies
0:18:50 has inspired concern that they have negatively affected
0:18:52 mental health and psychological wellbeing.
0:18:56 However, research on the topic is contested and hampered
0:18:58 by methodological shortcomings
0:18:59 leaving the broader consequences
0:19:02 of internet adoption unknown.
0:19:04 We show that the past two decades
0:19:07 have seen only small and inconsistent changes
0:19:10 in global wellbeing and mental health
0:19:11 that are not suggestive of the idea
0:19:14 that the adoption of internet and mobile broadband
0:19:18 is consistently linked to negative psychological outcomes.
0:19:20 Anyone reading that who’s also been paying attention
0:19:24 to the news would say, wait a minute, Andy,
0:19:27 this is exactly backwards.
0:19:29 This is the opposite of everything I’ve been told.
0:19:33 So tell me why you think you’re right
0:19:35 and what the argument that has gotten
0:19:38 so much heat lately has gotten wrong.
0:19:40 – I think the reason why your listeners
0:19:43 would be confused and say, wait, why?
0:19:44 Is a factor of two things.
0:19:47 The first being, like me, they’re intensely skeptical
0:19:49 about the role of technology in our lives
0:19:51 and in our children’s lives.
0:19:54 That’s natural and that needs to be listened to very closely.
0:19:57 And then two, that academic paper that you just talked about,
0:20:00 that wasn’t accompanied by a massive press campaign,
0:20:02 two massive popular press book campaigns.
0:20:06 And so it doesn’t get you in the opinion section of the times.
0:20:08 And so I think that part of that double take
0:20:10 is in part manufactured.
0:20:12 But I think that actually we’re at a local maximum
0:20:15 in terms of what many people call the tech lash
0:20:17 or the backlash against technology.
0:20:20 There was kind of a high watermark in the other direction
0:20:23 in 2011 with Terrier Square and the Arab Spring.
0:20:25 And then through this node-in period,
0:20:26 through Cambridge Analytica,
0:20:28 there was a fairly rapid swing
0:20:31 that brings us all the way to the kind of scholarship
0:20:32 that you were asking about.
0:20:35 But to get to the why I think that we’re quote, unquote,
0:20:38 right here is that the work that I do and others,
0:20:40 it is work that is not as glamorous.
0:20:43 And I would argue that there’s an inverse relationship
0:20:45 between how well a study is done in this field
0:20:49 and how shocking the results might seem
0:20:51 when you cross your T’s and dot your I’s,
0:20:54 when you share your code and you share your data,
0:20:56 you don’t find the kinds of things
0:20:57 that you or I would have
0:21:00 as kind of a preexisting bias about tech.
0:21:01 – Can you give me an example
0:21:03 that a lay person could understand
0:21:05 of the methodological shortcomings
0:21:08 or failures of the research you’re talking about?
0:21:11 – Phones, screen time, social media,
0:21:13 these are all catch all categories.
0:21:16 And so unless you’re epistemically humble
0:21:19 about what you’re measuring and what you’re not measuring,
0:21:22 it can be very tempting to claim
0:21:27 that when a parent or a kid is filling out a questionnaire
0:21:30 that asks, think back on the last year of your life,
0:21:32 on average, how many hours a day do you spend
0:21:36 with a computer, a smartphone, a game console,
0:21:40 and a cell phone, and then call that screen time
0:21:42 in the method section of your paper,
0:21:45 and then call the title of your paper,
0:21:49 “Social Media and Its Impact on Teenage Girls.”
0:21:51 And then you also have the measurement problem,
0:21:54 and this pains me very deeply as a psychologist,
0:21:57 of what the heck is mental health or wellbeing,
0:21:59 because saying you’re sad,
0:22:01 or saying you’re satisfied with your life,
0:22:03 is very, very different than turning up
0:22:06 in a therapist’s office and being diagnosed
0:22:09 as having major depressive disorder or anxiety disorder.
0:22:11 And all of those little decisions
0:22:13 of how you deal with measurement,
0:22:16 those actually can be quite consequential
0:22:18 if all of these arbitrary choices
0:22:20 that are made by a research analyst,
0:22:22 if they’re consistent with the researchers’
0:22:25 preexisting biases or they’re consistent
0:22:27 with the topic of their popular book,
0:22:30 it’s a bit of a red flag or at least a yellow flag.
0:22:32 When you think of everyday behaviors
0:22:35 as being potentially pathological or potentially addictive,
0:22:40 it’s a fairly slippery slope to pathologize everything.
0:22:43 I mean, video gaming is as damaging to mental health
0:22:45 as beds are to mental health.
0:22:48 You could create a disorder called bed addiction disorder,
0:22:52 where people have low affect and low mobility
0:22:54 and get bed sores.
0:22:56 And you could try to regulate the sale of beds
0:22:59 and haul bed companies in front of Congress.
0:23:00 A lot of the evidence you’d have
0:23:02 that something’s bad for you.
0:23:03 It’s not built on a basis of chemistry
0:23:06 or biochemistry or biology.
0:23:08 – You’ve made a similar argument in the past
0:23:12 saying that the effects of digital technology on teenagers
0:23:15 are about as big as the effects of eating potatoes.
0:23:18 – Well, yeah, social scientists will often try to draw
0:23:21 inferences about the population level effects
0:23:24 or associations that might link any given activity
0:23:25 to a health outcome.
0:23:28 And the problem with that is that if you
0:23:30 aren’t crossing your T’s and dotting your I’s,
0:23:33 you can interpret the noise in a data set
0:23:35 in a way that’s consistent with your preexisting
0:23:36 beliefs or biases.
0:23:38 And that’s what happens with technology.
0:23:40 We were trying to make a very simple point,
0:23:42 which is that if you don’t have your hypotheses
0:23:44 before you look at your data,
0:23:47 you can be led astray by very small effects,
0:23:50 whether it’s left-handedness, wearing glasses,
0:23:52 enjoying bicycling.
0:23:55 These are all things that have the same quote-unquote effect,
0:23:58 which is really just a correlation in a large data set.
0:24:00 – I think one interesting thing about this topic,
0:24:03 which is why it’s got so many people riled up,
0:24:05 is that this is a technology
0:24:08 that just about everybody has experience with.
0:24:10 So they can sort of fill in the blank
0:24:12 for the causal mechanisms, right?
0:24:16 They can say, oh, I’ve had the experience on my phone
0:24:19 where rather than going to that event tonight
0:24:20 at my place of worship,
0:24:23 or rather than going to play soccer with my friends,
0:24:27 I get caught up in my silo of anxiety and depression
0:24:28 on the phone.
0:24:32 And this lets people layer their own experiences
0:24:34 onto the moral panic argument.
0:24:37 But aren’t personal experiences in general
0:24:41 a good starting point for a lot of social science research?
0:24:43 – I think that they’re a really important starting point,
0:24:46 but I think they’re a highly invalid ending point,
0:24:49 and they shouldn’t be where people’s thinking stops.
0:24:50 It’s very easy to think
0:24:53 that there’s a digital world and an analog world,
0:24:55 and these worlds don’t connect.
0:24:58 But the problem isn’t necessarily that one is good or bad.
0:25:01 The problem is that we’re putting this wedge between them.
0:25:02 And when we put that wedge between them,
0:25:05 what we do is we cut ourselves off from
0:25:07 being able to investigate really interesting questions
0:25:09 about actually what is happening to someone
0:25:11 if they’re depressed and they’re using social media.
0:25:14 There’s a gigantic difference between feeling unhappy
0:25:16 with how you spent your evening
0:25:18 and suffering from agoraphobia,
0:25:20 having major depressive disorder,
0:25:22 or some form of crippling social anxiety.
0:25:24 I’m sure you know people who have suffered,
0:25:26 and many people who are listening,
0:25:29 know people who suffered with opioid addiction
0:25:30 or depression or anxiety.
0:25:32 And claiming that the thing that happens
0:25:35 when you play a video game is like what happens
0:25:37 when you can’t stop taking opioids,
0:25:40 or saying, I don’t feel like going outside,
0:25:43 I have had a bad Twitter argument, I feel so anxious.
0:25:46 Saying that’s the same thing as crippling agoraphobia,
0:25:48 where people won’t leave their home for years.
0:25:50 Frankly, that’s insulting,
0:25:53 because there’s no reason to think that’s the same mechanism.
0:25:59 – Over the last seven or eight years,
0:26:02 I have struggled with what I am quite comfortable
0:26:05 calling an internet addiction.
0:26:07 – That is Lauren Euler,
0:26:10 a novelist and cultural critic in her early 30s.
0:26:14 She grew up in West Virginia and now lives in Berlin.
0:26:17 Euler has spent a lot of time thinking and writing
0:26:19 about life online and off.
0:26:22 – I have met great friends on social media,
0:26:23 I’ve gotten boyfriends on social media,
0:26:26 but at the same time, I do strongly feel
0:26:28 that it can produce an actual addiction,
0:26:30 like a straightforward addiction, like a drug addiction.
0:26:33 And I know that because since I divested
0:26:35 from social media, I’ve picked up smoking
0:26:38 and it’s the same sort of thought process
0:26:41 about like compulsively wanting to either look at Twitter
0:26:43 or like smoke a cigarette.
0:26:45 Like you train yourself to think,
0:26:46 something could be happening on my phone.
0:26:47 Someone could have written me,
0:26:50 some news could have broken and I need to see it.
0:26:52 – Euler recently published a collection of essays
0:26:54 called “No Judgment”.
0:26:57 One of the pieces is titled “My Anxiety”.
0:26:59 – I think anxiety does create a barrier
0:27:02 around actually doing things that you’re sitting
0:27:04 and worrying about and sitting and looking at your phone
0:27:09 is just kind of externalizing this worrying feeling, right?
0:27:11 Like your phone is doing the worrying for you
0:27:14 because it has this short attention span.
0:27:15 My thoughts are like bouncing around
0:27:16 and I might be thinking about taxes
0:27:17 and then I’m thinking about some guy
0:27:19 and then I’m thinking about my late article
0:27:21 that I need to turn in and I’m thinking about,
0:27:22 I embarrass myself at a party
0:27:25 and all this happens in the span of two minutes,
0:27:27 which if you’ve been on social media,
0:27:29 that is exactly what it’s like, right?
0:27:32 – So Lauren Euler does see a deep connection
0:27:33 between her smartphone
0:27:35 and what researchers like Danny Blanchflower
0:27:37 call mental ill-being.
0:27:42 But again, is the one necessarily causing the other?
0:27:44 – I get anxious that I’ll become depressed
0:27:46 and depressed that I’m so anxious.
0:27:49 And if the phone is a kind of conduit for feeling
0:27:52 or if it’s an extension of your brain,
0:27:54 then of course like the things on your phone
0:27:55 might make you depressed
0:27:57 just as they might make you anxious.
0:28:00 So I think on one hand, of course it’s causal
0:28:03 but I do think it’s sort of self-defeating
0:28:07 and short-sighted to suggest that the phone
0:28:12 is the only cause of this mental health crisis
0:28:14 among teenagers or among anyone.
0:28:17 Because that’s just not how the world works.
0:28:19 That’s how op-eds work.
0:28:20 Like people who are desperate
0:28:22 for one weird trick to save your life,
0:28:23 it’s like I just needed to throw my phone in the river
0:28:25 and then everything would end up fine.
0:28:27 It’s just not how the world works.
0:28:32 – Okay, so if there is rising anxiety among young people
0:28:37 and if throwing your phone in the river isn’t a solution,
0:28:38 what is?
0:28:40 That’s coming up after the break.
0:28:43 I’m Steven Dubner and this is Freakonomics Radio.
0:28:52 Andy Shibilsky at Oxford
0:28:57 argues that the panic over smartphones is overblown.
0:29:01 But that doesn’t mean he thinks everything is fine.
0:29:04 In the U.S., there are some really worrying trend lines
0:29:06 for people across the population.
0:29:08 And with the most worrying trends,
0:29:11 things like self-harm and death by suicide,
0:29:16 I would be much more worried as a middle-aged man
0:29:18 in Appalachia with access to a gun
0:29:21 and with access to West Virginia’s social safety nut
0:29:25 than I would be as a white teenage girl anywhere in the U.S.
0:29:27 And so I would say that there probably
0:29:29 is something broken in America.
0:29:31 Here’s something I also wonder about.
0:29:35 There is so much more awareness around mental health now
0:29:37 than there was 20, 30 years ago
0:29:39 and so much less stigma.
0:29:41 Some people have argued that at least.
0:29:44 Do you think that that acceptance might be showing up
0:29:47 in the data as more young people today saying,
0:29:50 yes, I’ve experienced mental health problems
0:29:51 than they used to?
0:29:53 – There’s a concept called measurement variance.
0:29:56 This is the basic idea that if I asked you
0:29:58 how I’m feeling today and then I compare
0:30:01 how I’m feeling today to how I said I felt 10 years ago,
0:30:02 the way that people answer that question
0:30:04 actually changes over time.
0:30:07 And so there’s a bit of that, absolutely.
0:30:09 And then the way that we actually track
0:30:10 things like mental health problems
0:30:12 in the U.S. in particular,
0:30:14 it’s very, very susceptible to change
0:30:18 as a function of changes in insurance
0:30:21 and changes in way that diseases can be classified.
0:30:25 – On this note, Shabilsky points to a study
0:30:27 by two health economists,
0:30:30 Adriana Cortador Waldron and Janet Curry.
0:30:31 Their paper is called,
0:30:34 to what extent are trends in teen mental health
0:30:37 driven by changes in reporting?
0:30:39 – It’s a study of clinical intakes
0:30:43 of young women in New Jersey across the last 10 years.
0:30:45 There’s a series of spikes and hospital admissions
0:30:47 for different types of serious disorders
0:30:49 when a young person shows up in crisis.
0:30:52 And one way that you could have interpreted the data
0:30:54 is to say that there is a mental health epidemic
0:30:56 in New Jersey for teenage girls in particular,
0:30:59 but the researchers went back and looked at changes
0:31:02 to the best practices for clinical intake.
0:31:05 And the peaks occurred in response to two changes.
0:31:09 The first is a special awareness campaign for clinicians,
0:31:11 for focusing on young women
0:31:13 as a target area for intervention.
0:31:15 And then the second was allowing there
0:31:18 to be a second reason for clinical intake.
0:31:19 So it used to be when you came in,
0:31:23 you would just say the person is schizotypal or depressed,
0:31:26 but then you could say schizotypal and depressed.
0:31:29 And so when that change went through, the amounts went up.
0:31:30 But, and here’s the rub,
0:31:35 it only went up for young people who were on insurance
0:31:38 because it created a new billable category.
0:31:40 It didn’t go up for those who didn’t.
0:31:42 So I would say when you’re a hammer,
0:31:44 the problems of the world are nails.
0:31:47 I think it’s very tempting to think that the atomization
0:31:50 of certain Western cultures is a result of technology
0:31:53 and not other things like a decline in religiosity
0:31:55 or these other kinds of things.
0:31:57 There aren’t gonna be quick fix solutions.
0:31:58 – Like throw away your phones.
0:32:01 – Throwing away your phones or putting them a locker.
0:32:03 What you’re going to have to do
0:32:05 is actually rebuild communities.
0:32:08 You’re gonna have to hire psychotherapists.
0:32:11 You’re gonna have to build civil and economic infrastructure
0:32:12 to support those in your society
0:32:14 who are falling through the cracks.
0:32:16 – What if I say to you, I agree with you
0:32:19 and that sounds smart and sane,
0:32:22 but it also is going to take a long time
0:32:24 and a lot of money, a lot of resources.
0:32:27 And in the meantime, I believe there are millions,
0:32:28 maybe billions of people suffering
0:32:30 and so I wanna do something quick and dirty.
0:32:32 What would you say to that?
0:32:34 – Law of unintended consequences.
0:32:37 I would say that if you can’t properly diagnose the problem,
0:32:39 you can’t possibly solve it.
0:32:42 I would say that if you take this topic seriously,
0:32:45 you would take slow and incremental steps
0:32:48 to ascertain exactly what problem you’re trying to solve.
0:32:51 And so yeah, I would say that you should not waste energy
0:32:52 on patting yourself on the back.
0:32:55 – What if I ask you to make an argument
0:32:58 for the benefits of digital tech that are overlooked
0:33:02 in the rush to proclaim all of digital technology
0:33:03 damaging to young people especially?
0:33:06 – One of the things that I think everyone can agree
0:33:08 is really worrying is content online
0:33:11 around self-harm or anorexia.
0:33:15 And it might seem like a very obvious thing
0:33:18 that platforms should be doing absolutely all they can do
0:33:22 to take down this content and to scuttle these communities
0:33:25 and make sure that it’s not on any of the big platforms.
0:33:27 That might seem like common sense.
0:33:30 That might feel like the right thing to do.
0:33:33 But you can wind up causing so much more harm than good.
0:33:35 If you don’t pay attention to actually how people
0:33:39 who struggle with these disorders use social media platforms,
0:33:41 there are really good support groups
0:33:45 for things like self-harm, suicidal thoughts,
0:33:47 and things like anorexia and bulimia.
0:33:48 One of the things that social media platforms
0:33:51 can be really good for is connecting communities
0:33:53 in a positive way that self-moderates
0:33:56 and they provide resources for people in crisis.
0:34:00 And what happens if you think that it’s the responsibility
0:34:02 of these companies to just crush it
0:34:04 so that it’s not there anymore?
0:34:05 Two things will always happen
0:34:07 and they’re both very worrying.
0:34:09 The first is people who are already struggling,
0:34:12 they will begin to stop disclosing.
0:34:13 They’ll button themselves up.
0:34:16 They won’t reach out and they’ll lose access
0:34:18 to the online social environments
0:34:20 that have been supported for them.
0:34:22 And then the second thing is the community moves
0:34:24 to other parts of the internet
0:34:27 that are more poorly moderated and are far less safe.
0:34:29 And so you wind up in a telegram group
0:34:31 where there’s no accountability
0:34:34 instead of being in a Facebook or Instagram group.
0:34:36 And so I think thinking about tools
0:34:38 to meet young people where they are
0:34:39 is a much smarter idea
0:34:42 than just burning everything down and salting the earth.
0:34:47 – As you just heard, Andy Shibilsky is more positive
0:34:49 toward the big social platforms
0:34:51 than many people in this arena.
0:34:54 In fact, he recently announced a research collaboration
0:34:58 with the Facebook and Instagram parent company, Meta.
0:35:00 – We’re going to solicit proposals
0:35:01 from researchers all around the globe.
0:35:03 They can get data on young people
0:35:05 in 40 different countries
0:35:06 and combine that with information
0:35:09 about their mental health and their wellbeing.
0:35:12 But at no point will they have a direct relationship
0:35:14 in terms of the selection of their projects
0:35:16 or the decision for the projects to go forward
0:35:18 with the researchers at Meta.
0:35:20 We’ve put in a series of firewalls
0:35:23 between the data scientists at Meta
0:35:25 and researchers who would be very interested
0:35:28 in investigating the idea that Instagram
0:35:31 relates to the mental health and the wellbeing of teenagers.
0:35:34 Myself and my co-editors will be selecting projects
0:35:37 on the basis of their scientific merit
0:35:39 and their proposals will be peer reviewed
0:35:41 in advance of data collection
0:35:43 by researchers who are specialists
0:35:44 in topics like mental health,
0:35:47 topics like online engagement,
0:35:50 but also generalists who have a sense
0:35:52 of what is good methodology.
0:35:55 And only when that peer review process finishes,
0:35:57 the data request goes directly to Meta.
0:36:00 The idea here is you have these firewalls in place.
0:36:04 You don’t have big tech negatively affecting the process
0:36:05 and you don’t have researchers
0:36:07 potentially moving the goalposts
0:36:10 because they think they’ve found something interesting
0:36:12 after having a rummage through the data.
0:36:15 – It might surprise you
0:36:17 that the rigorous analysis Shabilsky is talking about
0:36:20 hasn’t already been done on Instagram and Facebook
0:36:22 and other social media platforms,
0:36:27 but it’s worth remembering how new this ecosystem is.
0:36:30 Here again is the writer Lauren Euler.
0:36:33 – The social media era has been really short, right?
0:36:35 Like when was MySpace founded?
0:36:38 When was Facebook founded in the early 2000s, right?
0:36:39 That’s not very long.
0:36:41 And already Facebook, nobody uses that.
0:36:43 They had to buy Instagram and WhatsApp
0:36:44 and we’re gonna say relevant.
0:36:47 And Instagram is actually quite young, right?
0:36:49 I think it’s like 20, what?
0:36:51 2010, yeah, that’s really young.
0:36:54 So I look forward to a day when Instagram doesn’t exist,
0:36:57 but I’m sure everything will take its place.
0:36:58 – I’m guessing Euler is right,
0:37:01 that the platforms will change
0:37:04 and the way we interact with our phones will change.
0:37:07 Some people, like Andy Shabilsky,
0:37:10 believe that if you can’t properly diagnose the problem,
0:37:12 you can’t possibly solve it
0:37:16 and therefore to rush into a hard anti-phone position
0:37:18 isn’t the right move.
0:37:22 Other people, like Danny Blanchflower,
0:37:23 say that we can’t afford to wait
0:37:26 because if the rise in anxiety and unhappiness
0:37:30 among young people is as significant as he thinks,
0:37:33 the consequences will be significant too,
0:37:34 especially the kind of consequences
0:37:37 that economists like to consider.
0:37:40 – I worry about what will happen to these young people
0:37:41 as they come to the labor market.
0:37:44 And then later on, we’re gonna worry about their mortality
0:37:45 and we’re gonna worry about their ability
0:37:48 to generate savings and investment
0:37:51 and buy a house and buy themselves a retirement package.
0:37:55 So down the road, these are issues that we care about.
0:37:57 – You are never going to run out of problems
0:37:58 to address, are you?
0:38:00 – In a way, that’s what I’ve tried to do.
0:38:02 Think about how we address the problems
0:38:04 and try and find solutions.
0:38:05 I don’t know what works.
0:38:07 I’ve been asked that question, we go,
0:38:09 “We don’t know, but we’re gonna try and find out for you.”
0:38:10 That’s what we do.
0:38:14 Do me a favor, when you find something, call me
0:38:16 and we’ll talk about it again, okay?
0:38:18 – Love to, love to.
0:38:22 – Thanks to Danny Blanchflower, Andy Shibilski
0:38:25 and Lauren Euler for their insights today.
0:38:27 And I’m curious to know what you think.
0:38:31 Our email is radio@freakonomics.com.
0:38:33 Coming up next time on the show,
0:38:37 do you ever watch an NFL game and wonder why
0:38:39 when so many of the players are black,
0:38:42 so many of the coaches are white?
0:38:44 The NFL itself wondered that,
0:38:47 and two decades ago, they put in a rule to help.
0:38:49 – The Rooney Rule is really about making sure
0:38:51 that you have a diverse slate
0:38:54 when you’re selecting and hiring people.
0:38:55 – Did it help?
0:38:57 We tell the history of the Rooney Rule
0:39:00 with the help of an actual Rooney.
0:39:02 – You know, if he was in the Senate,
0:39:04 we’d call him Majority Whip.
0:39:06 – We look at the successes and failures,
0:39:08 the lawsuits and the sham interviews,
0:39:12 and we ask how the Rooney Rule works outside of football.
0:39:13 That’s next time on the show.
0:39:15 Until then, take care of yourself.
0:39:18 And if you can, someone else too.
0:39:21 Freakonomics Radio is produced by Stitcher and Renbud Radio.
0:39:24 You can find our entire archive on any podcast app
0:39:26 also at freakonomics.com
0:39:29 where we publish transcripts and show notes.
0:39:31 This episode was produced by Teo Jacobs.
0:39:34 Our staff also includes Alina Coleman, Augusta Chapman,
0:39:38 Delvin Appalachie, Eleanor Osborn, Ellen Frankman,
0:39:40 Elsa Hernandez, Gabriel Roth, Greg Rippin,
0:39:43 Jasmine Klinger, Jeremy Johnston, John Schnarrers,
0:39:45 Julie Canfer, Lyork Bowditch, Morgan Levy,
0:39:48 Neil Coruth, Rebecca Lee Douglas, Sarah Lilly,
0:39:49 and Zach Lipinski.
0:39:52 Our theme song is “Mr. Fortune” by the Hitchhikers.
0:39:55 Our composer is Luis Guerra.
0:39:57 As always, thanks for listening.
0:40:01 – I’m Dr. Andy Shibilski.
0:40:04 I’m the University of Oxford’s professor of,
0:40:06 oh my God, what am I the professor of?
0:40:08 – Human behavior and technology?
0:40:10 – Yeah, exactly, but I was so hung up
0:40:12 on if I’m Andy or Andrew.
0:40:18 – The Freakonomics Radio Network.
0:40:20 The hidden side of everything.
0:40:24 Stitcher.
0:40:27 (gentle music)
0:40:29 you
Young people have been reporting a sharp rise in anxiety and depression. This maps neatly onto the global rise of the smartphone. Some researchers are convinced that one is causing the other. But how strong is the evidence?
- SOURCES:
- David Blanchflower, professor of economics at Dartmouth College.
- Lauren Oyler, novelist and cultural critic.
- Andrew Przybylski, professor of human behavior and technology at the University of Oxford.
- RESOURCES:
- “The Declining Mental Health Of The Young And The Global Disappearance Of The Hump Shape In Age In Unhappiness,” by David G. Blanchflower, Alex Bryson, and Xiaowei Xu (NBER Working Paper, 2024).
- “Further Evidence on the Global Decline in the Mental Health of the Young,” by David G. Blanchflower, Alex Bryson, Anthony Lepinteur, and Alan Piper (NBER Working Paper, 2024).
- No Judgment: Essays, by Lauren Oyler (2024).
- “To What Extent are Trends in Teen Mental Health Driven by Changes in Reporting?” by Adriana Corredor-Waldron and Janet Currie (Journal of Human Resources, 2024).
- The Anxious Generation: How the Great Rewiring of Childhood Is Causing an Epidemic of Mental Illness, by Jonathan Haidt (2024).
- “Global Well-Being and Mental Health in the Internet Age,” by Matti Vuorre and Andrew K. Przybylski (Clinical Psychological Science, 2023).
- “Are Mental Health Awareness Efforts Contributing to the Rise in Reported Mental Health Problems? A Call to Test the Prevalence Inflation Hypothesis,” by Lucy Foulkes and Jack L. Andrews (New Ideas in Psychology, 2023).
- “The Association Between Adolescent Well-Being and Digital Technology Use,” by Amy Orben and Andrew K. Przybylski (Nature Human Behaviour, 2019).
- 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, by Jean M. Twenge (2017).
- EXTRAS:
- “Are You Caught in a Social Media Trap?” by Freakonomics Radio (2024).
- “Are We Getting Lonelier?” by No Stupid Questions (2023).
- “Is Facebook Bad for Your Mental Health?” by Freakonomics, M.D. (2022).
- “Why Is U.S. Media So Negative? (Replay),” by Freakonomics Radio (2022).