Search the Site

Episode Transcript

Ben HANDEL: I’ve loved working on this paper. The idea is very simple, yet it’s not something that — you know, I’ve been around for a long time doing economics — it’s not something that really seems to be in the consciousness.

Ben Handel is an economist, at the University of California Berkeley. The paper he loved working on was a collaboration with Rafael Jimenez, Christopher Roth, and Leonardo Bursztyn. And what was this simple but overlooked idea? Here is Bursztyn:

Leonardo BURSZTYN: It is possible that when you don’t consume a product, you face a big cost.

And what kind of product can be costly even if you don’t consume it? Well, consider the title of their paper: “When Product Markets Become Collective Traps.” And now consider one of the survey questions they asked their research subjects:

BURSZTYN: We decided to ask a very simple question, which is, “Would you prefer to live in a world with or without TikTok or Instagram?”

Today on Freakonomics Radio: is social media a trap that most of us wish we never got caught in?

*      *      *

Leonardo Bursztyn teaches at the University of Chicago.

BURSZTYN: I have been focusing my research on understanding how individuals’ choices and main economic decisions are shaped by their social environment.

As for the idea that we’re talking about today …

BURSZTYN: The first time I thought about this idea was back in the day, maybe seven — time flies — maybe almost ten years ago. I was working with some co-authors on a project about platinum credit cards. That was a series of experiments with a bank in Indonesia, where we were showing that part of the demand for these fancy credit cards is not because of the instrumental value of the product, but more for the signaling value. People want others to see that they have a fancy credit card, which sends a signal that they are rich and successful, right? And then, as part of that experiment, we realized that what these companies are doing is that they keep introducing new versions that are fancier and fancier, and creating demand. So the idea that we had back then is, what if people don’t want some new products to be created, but once they’re created, they can’t help it, they just feel like they have to buy it? I mean, an example: weddings in India. This is a bit of a rat race. People have spent more and more money on these fancy weddings. And then the idea there was like, well, what if we team up with a wedding agency and propose a new tier of fancy weddings and then get some people to vote on whether they want this or not in some villages, and then actually see if they just buy it, even though they would prefer that it didn’t exist.

DUBNER: When you started studying economics, did you think about running experiments as a major tool?

BURSZTYN: I was looking at my statement of purpose, like when I was applying for grad school — I was proposing to study monetary policy. Things change. 

DUBNER: What brought you over to this side? 

BURSZTYN: I remember that I was working on economic growth on the theory side, and I was interested in understanding education, how this affects development. I’m from Brazil originally. And I had some connections back in Brazil, figured out that it was possible to run an experiment to understand the schooling choices of low-income kids. And then basically, I got really excited about it, and I was like, well, this is a very powerful tool that allows us to both test theory but also understand real-world problems, and come up with solutions. So it was, like love at first — I wouldn’t say “regression” — but love at first sight. 

In 2017, Bursztyn co-authored a paper called “Acting Wife.”

BURSZTYN: We were interested in understanding the role of gender norms in labor-market investments by women. Because there is a lot of evidence that women are still penalized in the dating market if they are particularly successful, ambitious, if they earn more money than their husband. 

So Bursztyn and his collaborators looked at students in one elite American M.B.A. program.

BURSZTYN: These students, in the beginning of the program, have to fill out a questionnaire, telling the career center of this university what their preferences were for their internship that will lead eventually to a job. 

One important thing to consider: a lot of students in M.B.A. programs, when it’s time to get married, often marry other students from their M.B.A. program.

BURSZTYN: And we randomized whether these students thought that only the career center would see their answers, or whether they thought their classmates will see their answer. 

So what did they find? It turns out that when unmarried female students thought their classmates would see their answers on the questionnaire, the answers were substantially different than if they thought only the career center would see them.

BURSZTYN: They would say that they want a lower salary, that they’re willing to travel less for work, work fewer hours, and so on. The evidence really seems to indicate that these women, even in a setting where you select achievers, these women are trying to portray themselves as less ambitious, less professionally driven, because of the dating market incentives they’re facing. 

Most of us like to think we are consistent in our decision-making, that we have some kind of intellectual or moral or even just a pragmatic framework that shapes our behavior. But the reality is that most of us are heavily influenced by our social environment. For a researcher who is interested in that gap between our perception and the reality, you could imagine there is one ecosystem — one perfect ecosystem — to examine these decisions. It’s right there in the name: social media.

BURSZTYN: The first generation of papers were showing all these great things about social media. “Oh, it helps facilitate coordination to go against totalitarian regimes. And then we started trying to look at other aspects, to get a more nuanced picture of the effects of social media. We studied this social media platform called VK in Russia, which is basically the Facebook there. We find an effect of the social media platform on xenophobic crimes and also xenophobic attitudes in Russia. That helps build this case that social media could have these potentially very negative effects on society. 

DUBNER: So should I assume that you came into this study, the one that we’re talking about today, with a view that social media has really significant downsides? Or do you try to play it more to the center?

BURSZTYN: That’s a very good question, because the way this paper came to exist was not by focusing necessarily on social media. Late 2021, I was visiting Berkeley. I was considering a move there. And I began talking to Ben Handel, he’s one of the people that was hosting me there. We really got along well. And Ben said, “Hey, too bad you’re not coming to Berkeley, but we should definitely work on something together.” So I mentioned a little bit what I described to you, which is this idea that, well, maybe there are products that people buy but they wish these products didn’t exist. I was thinking about weddings in India or luxury products, and so on. But, you know, life happened. We stopped talking for a little while. And then eventually — there was a conference last summer on the economics of social media. And the organizers — one of them was Rafael Jimenez  — told me, “Hey, listen, I wanted you to give a keynote at this conference.” And I was like, “Well, thank you, great. However, I don’t know what I’m going to be talking about.” Then eventually something clicked and I was like, “Well, that idea, oh we could actually apply it to social media.”

HANDEL: My name is Ben Handel. I’m a professor of economics at U.C. Berkeley. I’ve been there for 13 years, teaching classes in a variety of areas, but most notably in industrial organization — which is a general area that does relate to this paper on social media — but then a lot of my applications have been in health care. 

DUBNER: You said that this topic intersects a bit with I.O., with industrial organization. How is that the case?

HANDEL: So industrial organization studies how do firms compete with one another, and studies welfare in product markets. 

DUBNER: “Welfare in product markets”: can you put that in slightly plainer English? 

HANDEL: You can actually put it in very plain English, which is: Is this product good for people? How good for people is this product? And what types of people is it good for or bad for?

DUBNER And then talk about a couple economic concepts surrounding that theme. Economists like to talk about something called W.T.P., willingness to pay. What does that mean? 

HANDEL: Willingness to pay is a measure that economists have used for a long time now when thinking about welfare. And what willingness to pay means is — you go to the store, there’s a product, it’s a certain price, $12. Are you willing to pay that much to buy this product? Let’s say you are. Then let’s say you go to a different store. And the product is $15. And you’re not willing to buy this product at $15. What that tells us, under very simple assumptions, is that your value for this product — how much as a society we should value you having this product or how much you value having this product — is between $12 and $15.

BURSZTYN: When people are trying to understand the welfare that a given product gives, people usually think about, “Okay, how much do I have to pay you to stop using a product, or how much are you willing to pay to consume the product,” right? Now, there’s an implicit assumption when people are evaluating welfare this way, that if I pay you to stop using the product, whether other people are consuming the product or not, shouldn’t matter in the way we calculate the value that this product has for you. 

HANDEL: The whole point of our paper is to move beyond these willingness-to-pay measures, and have a different framing on welfare. For most products — let’s take an example of a standard product — a refrigerator. Not exciting, no social-status concerns usually —

DUBNER: I don’t know, some refrigerators do excite me, but okay.

HANDEL: As I said it, I was like, “Well, you never know.” Um, so take a refrigerator, and say a new, exciting refrigerator enters the market. It could deliver value to some consumers; to other consumers, not deliver value. The key thing is — if you’re a consumer and you don’t buy this refrigerator, your life doesn’t change, at all, right? Nothing happens to you. As economists would say, your change in utility is zero. Now, in our paper, the difference is that that’s not true. So, take a non-user of social media. Maybe your friends at school are using it, and you feel like you’re missing out on what they’re doing. Maybe people migrate conversations to social media and now, you’re missing out on conversations. So now, as more and more people use social media, your well-being is going down, even if you’re a non-user. And so you might say, “Hey I should use social media because right now I’m getting this really negative effect from missing out on this product.” So we’re calling this a “product market trap.” What that means is that the product market exists, people are willing to buy this product, but from an overall standpoint, society might say we’d be better off if this product didn’t exist. That’s — that’s new.

So this new idea, of a product market trap — specifically, a social-media trap — this would lead to some strange and surprising insights. That’s coming up.

*      *      *

When it comes to exploring new ideas, theories are useful. But, to explore the idea of whether social media is what they call a product market trap, the economists Ben Handel and Leonardo Bursztyn first needed some data. Here’s Bursztyn:

BURSZTYN: So, we decided to look at college students. One, they’re big users of social media. And two, there is a big overlap between the social media network and their physical network of college mates. So we found a partner, this company called College Pulse, that has these very large samples of college students. The experiment starts with the basic elicitation, which is how much do we need to pay you to stop using either TikTok Instagram — we did both — for four weeks. 


And here’s Handel:

HANDEL: So what we found is that, on average, you would have to pay people $50 to deactivate their TikTok or their Instagram account for four weeks.

DUBNER: So that was step one. 

HANDEL: Yeah. That’s where most of the literature stops. Step two is, we wanted to get a sense of something that economists care about called network effects — which is, how does your value of using this product depend on the number of other users? And so there, we implemented an incentivized experiment that basically says, if people in your social network deactivate, some of them do, now how much are you willing to accept to deactivate your account? And what we found is that you’re willing to accept about one-third lower in dollars in order to deactivate your own account.

DUBNER: In other words, if I were a research subject here, I might have said, “You need to pay me $50 to not use TikTok for a month. But if I know that a certain amount of my network, friends, whatever, are also not using it for a month, then you only need to pay me, like, whatever, 33 bucks.” That’s about right?

HANDEL: Exactly. But the most exciting part of the results is the next step. The next step says — we’ve elicited from your network how much they are willing to accept to deactivate their account. Now imagine you’re in control. You’re the boss. And we’re going to ask you, how much are you willing to pay to be social media god and deactivate everybody’s account?  

BURSZTYN: Exactly. Now it’s not about you quitting alone. It’s about you and everyone else, all the other students in your school, also quitting together.

HANDEL: And now what we found is that people are actually willing to pay to deactivate everybody’s account, including their own. And on average for TikTok, for example, they’re willing to pay $30 to deactivate everybody’s account, including their own, for four weeks.

DUBNER: Which is roughly the same amount they said they would need to accept to deactivate if some of their friends were deactivating. 

HANDEL: Exactly. Exactly. 

DUBNER: It’s kind of mind-blowing, isn’t it? I mean, believable but mind-blowing. That’s the category I place it in, does that sort of fit with your thinking? 

HANDEL: That’s exactly the way I think about it, yeah. But I think this insight applies to a lot of markets, not just social media. 

DUBNER: Tell me more.

HANDEL: Status products are an obvious application here. You could think about fashion, for example. You could think about fancy cars. Anything where we really think about this as a Keeping Up With the Joneses kind of product. 

DUBNER: To be fair, you include a look at luxury goods in this paper as a sort of check against the impulses you’re measuring, yes? 

HANDEL: Exactly. And there, we ask people, do you consume luxury products — and I think the examples we give are Gucci, Versace, Rolex, etcetera. And then, “Do you wish these products didn’t exist?” And what we find there is that 44 percent of buyers of those products say they wish they didn’t exist.

DUBNER: So 44 percent of buyers of one of this list of luxury goods say that even though they are a buyer — and let me add, this is an online survey where someone says they have bought one of these luxury goods, right? — these alleged luxury-goods buyers say they would prefer that the thing itself literally not exist. Is that right? 

HANDEL: Exactly. That’s correct. And then we find that 70 percent of non-users wish these products didn’t exist. 

DUBNER: I mean — again, bonkers, but I get it. 

HANDEL: And I should say, in our social media example, the numbers are similar. Of users, roughly 50 percent indicate that they wish the product did not exist. And about 70 or 80 percent of non-users say the same thing.  

DUBNER: You claim overall that most respondents in your sample “would prefer to live in a world without TikTok and Instagram.” And yet, many of those claimants are also users of these products, even most of them. You’re arguing that people really, really, really want to do one thing, but in reality, they’re doing the other. At some point, shouldn’t we believe what people are doing rather than what they say they want to do in an experimental setting like this? 

HANDEL: That’s a great question. What we’re eliciting does not run counter to what people are doing. The reason why that’s true is the version of our experiment that corresponds to the real world is the version where people are willing to accept money. So you have to pay them to quit. And the reason that’s the case is that in the real world, people can’t play social media god, and people don’t have control over whether everybody cancels their account. Our finding that people would like to pay to have everyone quit, that’s what I would think of as a counterfactual, which is that’s something that in the real world we don’t observe. 

DUBNER: But let me ask you a foundational question here about your paper. People like you love to talk about the difference between declared preferences and revealed preferences, right? Declared is what people say they will pay for something, or how they say they will vote, or spend their time. And then revealed preferences is what they do. And we know that those are often not remotely the same. So when you set out to measure people’s preferences and appetites and willingness to pay for something like social media consumption, you did it in a big experiment by getting a bunch of, I guess, undergraduate students, were they, primarily?

HANDEL: Yes, exactly, at different college campuses.

DUBNER: So given that economists especially are skeptical of survey data, and that they’d usually rather have data from the real world, why did you choose to address these questions not by some kind of revealed preference experiment, but instead you went with essentially questions on a screen, and using that as the main data from which to draw these conclusions? In other words, why did you do a lab experiment, essentially, versus a field experiment?  

HANDEL: It’s actually kind of in between what I would think of as a lab experiment and a field experiment. And part of the reason is that the mechanism tells them that there’s some chance that the hypothetical scenario we’re talking about will actually happen.

DUBNER: Now, you told your research subjects that there was some circumstance by which their social-media network essentially would be deactivated, right? 

HANDEL: Exactly, exactly.  

DUBNER: But could you and would you have actually followed through on that threat? 

HANDEL: Potentially, yes. 

DUBNER: But here’s the sentence in the paper that made me say, “Okay, come on, no way.” You write, “Participants are truthfully” — that’s the word that caught my attention — “participants are truthfully told that this large-scale deactivation will be conducted if we recruit two-thirds of students at their university.” There’s no way you’re actually even able to think about recruiting two-thirds of the students at their university, correct? 

HANDEL: I’m not sure that’s true, simply because, um, you know, there is the money to do it.  

DUBNER: But, I mean, the fact is, in reality, at the different universities you worked with, you may have used in this study something like, I’m guessing, one percent, half of one percent, two percent of the students at that university, right?

HANDEL: Yes, I think that’s correct. Because we looked at a little more than 1,000 students total, but across different campuses.

DUBNER: So you’re saying this is kind of a useful fiction in this sort of experimentation? Is that right? 

HANDEL: Yes. You said it, not me, but yes. By the way, there’s one thing I should say here, which I think is very germane to this point. We have a lot of what’s called robustness checks, which means we’re checking different stuff to make sure that we think the answers are as believable as possible. And so one thing we did is we asked people, “What do you think the chance is that we will actually conduct this large-scale deactivation study?” And I believe the response that people thought, “Yes, we’re actually going to do it,” was about 45 to 50 percent.  

DUBNER: So I guess you could read that in one of two ways. Either they really, truly believe it, or they so deeply want it to happen — which sort of lines up with the actual experimental evidence from your studies, yeah? 

HANDEL: Yeah, absolutely. 

DUBNER: You do call it “a social media trap” in the paper. And “trap” implies that people are in a place where they would prefer not to be. And the mechanisms that might keep you in the trap could be all over the place. Could be addiction, could be laziness, could be lack of self-awareness, could be lack of discipline. So when you see these results — for social media and luxury goods — did you think about, or learn anything useful about, the motivations to consume these things that we say we wish didn’t exist? 

HANDEL: We did. We followed up and asked an open-ended question: “You indicated that you value this product conditional on everyone else using it. If you could cancel this product, you would. Why? Why are you using it?” And we coded their responses as falling into these different categories, including things like fear of missing out.

DUBNER: And that was number one I gather, yes, in your survey?

HANDEL: Exactly. Another type of answer we coded was addiction. But it was quite a bit lower than fear of missing out.

DUBNER: To be fair, no one wants to call themselves an addict. Like, isn’t that a neighbor on the spectrum — fear of missing out and addiction — in that I might be more comfortable calling myself someone who’s scared of missing out versus someone who cannot stop myself from doing it because I don’t want to miss out? 

HANDEL: I agree with that. And I agree that these concepts are somewhat fluid, right?

DUBNER: When you mentioned that this seems to be a new arena for economic study — a product that people wish did not exist — that does make me think about cigarettes, which are plainly — I think there’s plenty of evidence to show that their consumption is fueled by addiction. What do you think of that comparison, between social media use and smoking?

HANDEL: I think it’s actually a fairly apt comparison. But, obviously, with cigarettes, there’s a clear physiological component to the addiction. Now, with social media, there could be as well. I actually have another analogy, which is on my list of things I’d like to study in the future. There’s a new wave of drugs coming onto the market designed to deal with obesity. And so since I’m a health economist, this is stuff I like to think about. And that’s an area where you could imagine a similar phenomenon where someone says, “You know, I wish this drug didn’t exist, and I didn’t feel like I had to take it. But everyone in my social circle took it, they’re all skinny, that makes me feel bad. Now I want to take it.”

Coming up: we ask our economist friends how to avoid the social media trap.

*      *      *

A new research paper by Ben Handel, Leonardo Bursztyn, Rafael Jimenez, and Christopher Roth has found that a lot of people feel trapped by their social media. They don’t really want to be there, but a powerful sense of FOMO — fear of missing out — keeps them from quitting. I went back to Handel and Burstzyn to ask about the ramifications, and to ask for advice.

DUBNER: Your research maybe suggests a new product. If people are, at least in your experiment, willing to pay to avoid social media traps, then perhaps they’d be willing to pay a share of what they’d be willing to pay to avoid them for a new piece of software that prevents them from participating in the platforms that trap them, no? 

HANDEL: Yeah, I actually disagree with that. Because let’s say your product existed. In our environment, you’re not going to want that product. And you’re not going to want that product because everyone else is still going to be using social media. You’re still going to have the FOMO, you’re going to think, “I got locked out, but I don’t want to be locked out. I want everybody to be locked out, but not just me.”

DUBNER: If you didn’t do this research yourself and you just read it, whether as an economist or — I don’t know, are you a parent, do you have children? 

HANDEL: I do. I have a seven-year-old and a four-year-old.

DUBNER: Okay, so they’re getting right into the social media range now. 

HANDEL: Almost, almost.

DUBNER: So how would you interpret this research? Or beyond interpreting, what would be your next steps? What would be some things that you would choose to do or not do as a result of reading this?

HANDEL: I think in some sense, because the research is about a collective trap, rather than an individual behavior like addiction — 

DUBNER: You’d tell them to move to Antarctica because that’s the only way to escape —

HANDEL: I think what I would say is, you know, if your friends are using social media, you can and should use it. I would actually be, from an individual standpoint, more concerned about the addiction point, and try to think about time limits, things more like that.

BURSZTYN: A lot of the discussion has been on aspects of addiction or lack of self-control, which are very important. But we add another story to this, which is even if the person is fully aware of the harm that it’s causing them, they might rationally choose to use it because it’s even costlier to be the one who’s not using it when everyone else is. That’s the aspect of the trap, which is, “I know I don’t like being on Instagram, makes me feel bad, but if I’m not, I’m not going to know what’s going on, I’m going to be isolated, I’m not going to be a cool kid, and so on. So I have to be there even though I know it’s harming me.” So in a way, it could actually make the argument stronger because you don’t need to assume that users are naive, they don’t understand, they have self-control problems. They could be fully rational and they understand the harm, but the cost of being left out has become too high for them to actually not use the platform.

DUBNER: All this conversation so far has been about the demand side, the users. Can you talk about the supply side for a minute? I’m curious to know how much you know and care about how these platforms are built, and how they adjust their algorithms constantly to produce what turn out to be these outcomes that are making people not so happy.

BURSZTYN: Our work, and a lot of other people’s work, point to the benefits for the platform of having scale, having big networks. And you can see how a platform might even provide subsidies to really grow their network — because when you get so big, it becomes very hard for consumers not to join your platform, so there’s really a rationale for growing, growing, growing. That’s one thing. Now, there’s another set of questions that also relate to social media, which is how platforms can design their algorithms to reinforce these effects. People have focused a lot on how can they make things addictive, like dopamine kicks and so on, or pushing toxic content that increases engagement and outrage and so on. Now, platforms can also do things to increase the cost of being out of the market, right? They can integrate — for example, they introduced a messaging feature to Instagram. So, perhaps you were like, “I don’t want to be on Instagram because I don’t like that.” But now if all my friends are communicating on Instagram, that increases the cost for me to stay out of Instagram because I can’t even communicate with my friends. So by integrating with features that increase the cost of not being on the platform, you might be pulling more people in, even though they might not like the product.

DUBNER: Do Instagram and/or TikTok know about your paper? 

BURSZTYN: That’s a good question. I don’t know. Hopefully they do.

DUBNER: Let’s say you get a call from Instagram or TikTok and they say, “Leo, we’ve read your paper. We recognize that it provides us with an opportunity to do better. We don’t want people feeling that way about our platforms. We like people thinking that our platforms are socially beneficial, not social traps. And we’d like some advice from you, for how to dial down the negative feelings while not destroying our business model.” Would you have any good advice for them? 

BURSZTYN: Uh. Well, at least I’ll be more prepared after this, in case they call me. I think one aspect is about changing how harmful the platform is, right? Making it actually more enjoyable. So really, you get more benefit from it, even if you’re someone who right now actually prefers it didn’t exist. So it’s reducing the share of people who don’t want the platform to exist. Of course, it could be that this goes against the profit motives for these platforms, right? So one option is to reduce the share of people who wish it didn’t exist, right, by making it better. The other option is by making it easier for people to quit if they want. 

DUBNER: A lot of people who pay attention to how social media and other digital products work, and how their business models work, understand that they, the users, are the product essentially, or at least their data is the product. I’m curious if you have any sense of whether your research subjects understood or thought about that in any way.

BURSZTYN: Good question. I think that is something that unfortunately we didn’t examine. I think it’s a very important question you asked there because when there are cases such as a Cambridge Analytica, all the aspects of privacy in your data — and understanding how much people care about that. If you’re trying to defend that if you’re Meta, you could always say, “Hey, people don’t really care about it because they kept using Facebook.” And then, you could say, “Well that’s true, that’s a good point, if they really cared, they would have stopped using it.” Now our work would say, “Well, wait a minute. They might have hated that, but they’re still stuck in the trap.” If they could coordinate and all move out to a different platform, maybe they would have done it, but it’s very hard to coordinate.

DUBNER: Here’s a parallel, or another example of behavior that people were doing that they wish they weren’t doing, I guess, in retrospect. And I — I just want to know if you think this is a totally crazy parallel to make. Steve Levitt, when he was looking at the causes behind the rise and fall of crime over the past few decades — obviously, abortion was one that he looked into, but there were many others that he looked into. So one big factor for a relatively brief time was this huge market in crack cocaine sales. The markets around that tended to be pretty valuable, and therefore there was a lot of violence around them. And so he was trying to measure the size of that market, measure the violence associated with it, but then, as we know, the markets for crack cocaine really diminished. It never went away, but it fell a lot. And his theory — and this is really just a theory, I don’t think this is very empirical, but it does make me think of social media, having read your paper — is that the would-be second-generation users of crack cocaine, like the little brothers or the cousins or the children of the people who used it, were able to see the users ahead of them and how terrible it was for them, just how damaging a drug it was, and then it kind of cycled out of favor. I am curious if you have any thoughts on — you know, when you look at social media, if you have any philosophical thoughts about whether users will observe the damages and somehow maybe step out voluntarily? 

BURSZTYN: You can think about the introduction of these platforms 15 years ago, 10 years ago and the role of parents, right? I think there was less of an awareness of the negative impacts that these platforms have, especially on kids. So parents were probably very like, “Okay, just do it,” you know. And I think now, with all the change in the narrative and in the debate, we might start seeing parents being much more careful about allowing their kids, or pushing, or kind of controlling that. And even legislation might be changing as well.

DUBNER: One thing that Ben mentioned that he was so excited about was trying to think of other industries or ecosystems to do a similar kind of research. Like, if this now is a new and interesting tool to see things in a way that economists hadn’t before, what are some areas that you, with your own background, are interested in looking at next? 

BURSZTYN: One thing we’re starting to think about is to say, “Okay, let’s do these surveys of time allocation, what are people spending time on during the leisure time?” It’s going to be digital products, some of them non-digital products, and try to get at people’s views on how many of these hours are spent on things you wish didn’t exist. Imagine if, say, people spend two hours per day on something that they wish didn’t exist. That’s one month per year, a whole, full month.

DUBNER: And this is just leisure. You’re not even talking about the things you have to do.

BURSZTYN: Yes, exactly. So I think it’s interesting because people talk a lot about the sovereign consumer, right? I mean, is the consumer still sovereign right now?

DUBNER: Size is obviously really important because of network effects. But also, just, size creates leverage. And one thing that is fundamentally different about the digital revolution — we’re not even including A.I. or machine learning or anything here — but the digital revolution just makes size much easier to achieve and much cheaper, right?

BURSZTYN: Absolutely.

DUBNER: So I’m curious whether this insight that you’ve had, and this potential problem that you’re identifying — like, a lot of people, I think, think of social media as the end of the road, but to me, it sounds like you’re thinking it’s just the beginning, that the impulses that draw people to this environment are something that we really need to get used to. Because, you know, cigarettes may have been addictive and bad for us, and it’s taken a long time and trillions of dollars to address that. And that’s one kind of problem. But it strikes me that what you’re describing now is a sort of, I don’t know, potential global groupthink, where we are all roped into environments that, because those environments are so easy and cheap to build, when they’re successful, that there will be more traps, bigger traps, and harder to escape. How terribly wrong am I, or dystopian am I, in that question?

BURSZTYN: I think you’re pretty on point there, actually, unfortunately. I think the analogy with cigarettes is actually interesting, which is you might say, “Well, tobacco, of course, cigarettes are addictive, they’re harmful, and so on, and they also have an externality, right? There is secondhand smoke.” Now with cigarettes, what you can do to avoid secondhand smoke is to avoid hanging out with smokers. You can just go somewhere else, right? What is particularly tricky here, for example, social media, is that the only way to avoid secondhand smoke is by smoking. If you’re the only person who doesn’t have social media, you’re left out. The only way to undo the externalities is by consuming the product, right? So that’s a very — heavy situation right there. 

DUBNER: So, Ben, what would you say is the overall argument, if you had to make one, that your finding adds up to?  In other words, if you had to be a little bit — not prescriptive necessarily, but maybe advice-giving. Because it sounds to me a little bit like the arguments from what used to be called the New Left, like 40, 50, 60 years ago, against consumer capitalism, right? That the capitalists are too good at manipulating for the good of the public. Is that too strong a read of your finding?

HANDEL: I think a little bit too strong a read. The reason I think is that — okay, this is going to be a deeper answer, where I don’t necessarily even have, you know, the psychology expertise. The way I think about this, and in a future project something I’m hoping to work on, is what I think of as a status budget for humans. So, imagine a world — or you don’t need to imagine, it’s probably the real world. You know, humans evolved over time, and humans like status. I think that’s uncontroversial. A question — and psychologists probably know the answer to this question — is what defines our preference for status? Are those preferences innate? Do we crave status and do we need status, and are we going to find status-seeking and pursue status-seeking regardless of what’s available, right? Let’s take an example where, you know, there’s tons of luxury products — Gucci, Rolex, a fancy new car, whatever. The question I want to ask and answer in the future is, you know, “Are there diminishing marginal returns to these purchases in status?” Imagine that you have 32 status products. The next one comes out, and you say, you know, I don’t really need this. I already have these 32. Now let’s imagine a world — and I’m not advocating for this — but let’s say we cancel all social media, right? Status concerns gone. Maybe welfare goes up. I’m certainly not saying that would happen, but let’s say maybe welfare goes up. I think a really important question is, Do people then increase the status they associate with other things? I could say, “Okay, now I’m not getting this kind of status comparison in social media. I have time and energy to focus on other things. Maybe now I’m going to think about status more heavily in this other domain.” The reason that matters for the question you asked is that, say regulators — or, doesn’t have to be regulators, it could be school teachers or schools — say you want to somehow regulate the use of social media. Ban use in the school. Or say students can’t use the app in school. Or ban TikTok, which of course is a popular, you know, discussion.

DUBNER: We should say, that’s primarily because TikTok is owned by a Chinese firm, not because of its — 

HANDEL: True. True. I agree with that. Yeah. Say you want to do that. You know, our work and the work on addiction shows, “Okay, there’s a rationale for this. It doesn’t mean you want to do this, but there’s a rationale for this.” But from our research, that component of the kind of negative welfare, that’s going to go away. If you then say, okay, people are going to take this status preference and just translate it into smoking. Now it’s going to be all about whether you’re smoking instead of whether you’re on social media. And that’s an open research question that I’m really excited to answer.

That was Ben Handel, from U.C. Berkeley; and Leonardo Bursztyn, from UChicago. Their paper is called “When Product Markets Become Collective Traps: The Case of Social Media.” Their coauthors are Rafael Jimenez and Christopher Roth.

*      *      *

Freakonomics Radio is produced by Stitcher and Renbud Radio. This episode was produced by Ryan Kelley. Our staff also includes Alina Kulman, Augusta Chapman, Eleanor Osborne, Elsa Hernandez, Gabriel Roth, Greg Rippin, Jasmin Klinger, Jeremy Johnston, Julie Kanfer, Lyric Bowditch, Morgan Levey, Neal Carruth, Rebecca Lee Douglas, Sarah Lilley, and Zack Lapinski. Our theme song is “Mr. Fortune,” by the Hitchhikers; the rest of the music this week was composed by Luis Guerra. You can follow Freakonomics Radio on Apple PodcastsSpotifyStitcher, or wherever you get your podcasts.

Read full Transcript




Episode Video