I stopped following Justin Wolfers on Twitter. Not because I don’t want his tweets, they are great, but because everyone I follow also follows Justin. They all retweet his best tweets and I see those so I am not losing anything.
Which made me wonder how increasing density of the social network affects how informed people are. Suppose you are on a desert island but a special desert island which receives postal deliveries. You can get informed by subscribing to newspapers but you can’t talk to anybody. As long as the value v of being informed exceeds the cost c you will subscribe.
Compare that to an individual in a dense social network who can either pay for a subscription or wait around for his friends to get informed and find out from them. It won’t be an equilibrium for everybody to subscribe. You would do better by saving the cost and learning from your friends. Likewise it can’t be that nobody subscribes.
Instead in equilibrium everybody will subscribe with some probability between 0 and 1. And there is a simple way to compute that probability. In such an equilibrium you must be indifferent between subscribing and not subscribing. So the total probability that at least one of your friends subscribes must be the q that satisfies vq = v – c. The probability of any one individual subscribing must of course be lower than q since q is the total probability that at least one subscribes. So if you have n friends, then they each subscribe with the probability p(n) satisfiying 1 – [1 – p(n)]^n = q.
(Let’s pause while the network theorists all rush out of the room to their whiteboards to solve the combinatorial problem of making these balance out when you have an arbitrary network with different nodes having a different number of neighbors.)
This has some interesting implications. Suppose that the network is very dense so that everybody has many friends. Then everyone is less likely to subscribe. We only need a few people to be Justin Wolfers’ followers and retweet all of his best tweets. Formally, p(n) is decreasing in n.
That by itself is not such a bad thing. Even though each of your friends subscribes with a lower probability, on the positive side you have more friends from whom you can indirectly get informed. The net effect could be that you are more likely to be informed.
But in fact the net effect is that a denser network means that people are on average less informed, not more. Because if the network density is such that everyone has (on average) n friends, then everybody subscribes with probability p(n) and then the probability that you learn the information is q + (1-q)p(n). (With probability q one of your friends subscribes and you learn from them, and if you don’t learn from a friend then you become informed only if you have subscribed yourself which you do with probability p(n).) Since p(n) gets smaller with n, so does the total probability that you are informed.
Another way of saying this is that, contrary to intuition, if you compare two otherwise similar people, those who are well connected within the network have a tendency to be less informed than those who are in a relatively isolated part of the network.
All of this is based on a symmetric equilibrium. So one way to think about this is as a theory for why we see hierachies in information transmission, as represented by an asymmetric equilibrium in which some people subscribe for sure and others are certain not to. At the top of the hierarchy there is Justin Wolfers. Just below him we have a few people who follow him. They have a strict incentive to follow him because so few others follow him that the only way to be sure to get his tweets is to follow him directly. Below them is a mass of people who follow these “retailers.”
14 comments
Comments feed for this article
November 15, 2011 at 2:16 am
Economist's View: Links for 2011-11-15
[…] Ignorance of Crowds – Cheap Talk […]
November 15, 2011 at 3:36 am
tiskagent 1
Once enough of your friends reject subscriptions to depend on the stream of friends posts, the media services would collapse. That, in turn, would create a vacuum of information that would require your best informed friends to seek other ways to find information. Presumably, this would take most of their time, requiring remuneration. An answer to that would be to charge a small fee for their services, and as time passed and demand rose, they would then hire more information gatherers, streamline their packaging and turn into a consolidator of media flow. That’s what’s happening now. While mainstream global media outlets are suffering, other less mainstream outlets — economic blogs, or consolidators like Huffington Post and the Daily Beast, for example — are turning into the same things they drove out of business. So the point is, you can cancel your subscriptions now, but you will only create new conditions that will drive you to different, but not necessarily better, subscriptions later.
November 15, 2011 at 5:58 am
joshgans (@joshgans)
I am the only person I know who re-tweets your tweets. Hmm.
November 15, 2011 at 6:54 am
mbiel
This is what Granovetter calls “strength of weak ties”. No need for econ toy models.
November 15, 2011 at 7:42 am
Marc Gawley
Here are some stats on switching off from reading the news:
http://marcgawley.com/2011/10/30/stop-reading-the-press/
November 15, 2011 at 11:41 am
Assorted links — Marginal Revolution
[…] 4. Nonetheless, you still should follow Justin Wolfers. […]
November 15, 2011 at 12:05 pm
KevinH
I’m not sure people become less well informed on average. It seems like this mirrors classic specialization. If you have a finite number of posts you can see and curate during a day, having your friends select only the best Wolfer’s posts frees up your time to go and curate some perhaps more obscure poster and share those ideas with your friends, enriching everybody.
November 15, 2011 at 4:01 pm
Graham Stanton
It actually gets more complicated if you recognize that non-followers can still retweet a retweet. In the extreme case, all worthwhile news is always retweeted, so then in a dense network, you’re virtually guaranteed to get the important news (as long as there’s a path to you from Justin Wolfers). But I wonder how it changes if likelihood to retweet is a constant r less than 1 but independent of prior number of hops.
November 16, 2011 at 2:20 am
koenfucius
Are you not assuming too much rationality on the part of the subscribers by saying that they will do so provided the value v exceeds the cost c? In other words, are the individuals in your model rational agents – or are they instead real people whose behaviour is not easily captured in a simple formula like v – c > 0?
November 16, 2011 at 5:44 am
Jeff Kish
If you rely on reteeets- you miss things others seem no value on. There is a subset of those which YOU extract value from. That’s a problem with following percentages, and the price you pay. This price will be to some degree homogeneity of the idea-space.
November 16, 2011 at 9:51 am
Clustering and the Ignorance of Crowds | Social Dynamics
[…] on the Cheap Talk blog (@CheapTalkBlog), Jeff Ely (@jeffely) has an interesting post about the “Ignorance of Crowds.” The basic idea is that when there are lots of connections among people, each individual has less […]
November 16, 2011 at 10:35 am
Gerald Biehle
This sounds related to groupthink and confirmation bias in that the more connected, or insiders, get less outside information.
November 16, 2011 at 3:45 pm
John Hyland
It seems like you are leaving out some knowledge that a user has when deciding whether to follow a particular stream. To wit, a user knows whether any of their friends are already retweeting “all the good stuff.” This removes the probabilistic angle completely: if you want Justin’s tweets, you will follow him with 100% probability if none of your friends do, and with 0% probability if at least one friend is retweeting everything you want, and you can (un)subscribe later if circumstances change.
That said, a somewhat similar approach might apply if you consider differences between your tastes and your friends’ tastes – maybe they’ll pass over a tweet you would have wanted to read. I think the effect would be much lessened in that model, however.
December 4, 2011 at 8:56 am
Beth
You are assuming that more subscriptions always means “more informed”. However, not all subscriptions are identically informative and many are repetitive. One person who reads a neutral and trustworthy but widely read aggregator is likely to be more informed than someone who reads many ideologically-identical new sources, or many news sources of limited scope.
It would be interesting to incorporate the networked quantities of the information sources in question. I don’t bother subscribing to news papers, because as unconnected news sources I find them less valuable than those that are participating in a networked discussion.