Why the Sharing Economy Is Making All of Us More Lonely

"This is the tension of modern life in the United States: You have a lot of connections, but do they mean what they used to?"
Image may contain Symbol
Illustration by Alicia Tatone

The era of stranger danger is long over: we hop in their cars, we invite them over to assemble our dining room tables, and we even sleep in their beds. You can thank reputation for this new share economy. We know we aren't willfully entering into a transaction with an axe murderer because when we book Ben's "Brand New Mini Loft!!" there's no one-star review that says, "Ben's freezer was full of dead bodies."

Though there are obvious benefits, there are drawbacks, too: it's not hard to imagine (probably because it already exists in China, and in the most recent season of Black Mirror) a slippery slope into a dystopia, where our humanity and character is reduced to a 1-5 rating—an idea we explored in our latest issue.

As part of the resulting piece, I called up Paolo Parigi, a Stanford professor who has worked at Airbnb and Uber studying how the peer-to-peer marketplace is affecting reputation, trust, and our ability to connect with one another. Here, he tells why increased trust in one another may actually lead to increased loneliness.


GQ: On your website, you write,“On the one hand, the accumulations of reputations increases the ease with which we form friendships around common cultural interests. On the other hand, the accumulation of reputation diminishes the bonding power of the experiences by reducing the serendipity of discovering something new." Could you unpack that a little bit?

Paulo Parigi: Because more and more of our lives went online, and more and more information about us is available to others, it allows you to form friendships faster. However, at the same time, there used to be a process through which you would discover these commonalities. And this process of discovering was the process that made the friendship. Now, this process has either been accelerated or removed. The information is available to you right upfront. And so there's no discovery process.

Because of that—and this actually is based on some research I did a few years back, using CouchSurfing—what we found is that when the couch surfer had more information about that other person he or she was going to meet, the resulting friendship after they met was weaker than in a world where the couch surfers did not have that amount of information about the unknown other. So that's exactly where I started thinking about this kind of trade off: it's easy to form relationships, but they're potentially less binding. The friendship that results was weaker.

If you think about what this suggests, it's that we are less likely to be isolated in the world, because, even if you have very unique tastes, you can find someone who has your unique tastes. But, potentially you feel more lonely. This is to me the tension of modern or contemporary life in the United States, and maybe in the industrialized Western world, maybe even China: You have a lot of connections, but do they mean what they used to? Technology has played a big role into this process, altering the meaning of relationships.

Based on your work at Airbnb and Uber, how is trust being affected by the peer-to-peer marketplace?

Let's call reviews and ratings “reputation.” The reason why it's easier to place trust in strangers is because we have all acquired a reputation. These marketplaces will not sustain themselves at the scale they are if you were to remove the reputations of the users.

The other thing that I found—[from] my experience at Uber and my relatively new experience at Airbnb—is that these reputations are very, very specific. And this is potentially a change in the meaning of the word trust. Because it used to be that [the] extension of trust would cover different aspects of life. I trust this person with my house key. I trust this person with my car. I trust this person with my dog. In the online markets—gig economy, share economy, whatever label you want to use— they're very specific. You would trust an Uber driver to take you from A to B. You wouldn't trust him with the key to your house. And with Airbnb, you would trust the guest to stay at your house but not to babysit your kids. So it's a very thin definition of trust that is operating in this market.

If you think about reputation and how it used to work in the past—go back 300, 400 years ago when you were in small villages and you had reputation. And you would say, “Oh, Paolo is trustworthy person.” Period. They wouldn't say, “Paolo is a trustworthy person on X thing. “ That would be absurd.

This is very Black Mirror, you ever see a world in which we would have a centralized “global reputation score”?

I think that technically it's possible. This obviously connects very directly with China and the Chinese government. The technology is there—it’s possible to connect all of these things. The real question to me is: should we do that? And here I'm stepping outside what is normally my role as a researcher. Normally, I talk about things I’ve studied. “Should” is a different direction.

I'm kind of skeptical, so I'd like to say we shouldn't do it. Because I am afraid of the byproduct: creating a reputation that extends through different aspects of life. While I understand it can solve some problems for actors that want to enter into any of these platforms, because it gives them a reputation and a way to enter faster, I am quite concerned that it could extend to things we don't necessarily want [to be] scored on. At least in a public way. Look, I'm a parent, I have two kids. Let me show you a dystopian scenario where somehow there's some way to rate me as a parent, I don't think that's the right direction we want to follow. That could lead to dangerous changes.

I want to go back to the example of 300-400 years ago: people had reputations, the reputations tend to be stable, but there were people who were the outcasts, that were rejected—we don't want that. To me, that's one of the positives of modernity. Sure, we may not have the same sense of belonging of a little town but people accept more diversity because of that. I understand there are constraints and difficulties, but in general, there is a greater acceptance of diversity potentially now than 400 years ago. My fear is that if we aggregate too much reputation from different domains, we are threatening acceptance of diversity. We are making homogeneity again. Accepting of diverse others is already very difficult, clearly. I don't want anything that undermines this.

How do expectations affect our interactions with other people?

When I was doing the research on CouchSurfer, some of the hosts had been hosts for many years and they had accumulated quite and extensive reputation. If you were traveling, you could scan through the reviews and say, “I want to be with someone that like this type of music, doesn't wake up super early, likes this type of food for breakfast.” It could be very specific. And when you found that person, it's kind of like you bought a good. You purified that potential interaction to the bones and when that happens, that's what you were expecting is going to happen. It's very different than if you go into the house of someone you don't know and you can't tell he likes the same music that you like.

Now I have to say, all of this is on the negative side, but I want to point out a positive effect of this extra information. Because it is the case that it also allows—it makes it easier to interact. In another piece of research I did with Airbnb, we found that reputation had positive effect in overcoming certain types of biases. We tend to trust, in general, people that look like we look. Most times, it’s just bias, unconscious bias. What we found was that if you were interacting with someone who was different from you, but had a good reputation, using Airbnb style of reputation (ratings and reviews), you were more likely to trust that person than if you had someone that was very like you but had a lower reputation. And so that's positive. If we are able to deploy technology to overcomes biases? That's great.

To go back to what you were talking about earlier: I think it’s interesting that if you did have a global reputation score and it was 9, someone would see that and go into meeting you with the expectation of “Oh, he’s going to be great,” and it takes the magic out of discovery.

Yes, I think it would do that. And again, maybe it will make all of us feel more lonely. It would create a strange feedback if everybody is scoring you on your interactions. Then you have incentives to behave positive right? And that's weird. Not because you want to behave negatively, but [because] everybody becomes a spy, in a sense. There have been, in history, examples where certain dictators or totalitarian regimes have turned an entire population on spying on each other. There's the case of the communists of Eastern Germany. Is the social score creating something like that? Is it going to operate like that in China, too? Usually it's some sort of totalitarian regime that comes up with this.

Also, if we’re all behaving better but only doing it for external reasons—to get a better score—is that really better?

If you start tainting the altruists with "I also get a benefit out of it," people react differently, right? It’s not truly altruistic. If you help a stranger versus you help a stranger because then the stranger would give you a positive review, that's different. And also you—the person receiving the help—would interpret things differently. I think it makes that experience less meaningful.

This interview has been edited and condensed.