Network Externalities Do Not Create Permanent Monopolies

By Graham Peterson

Elizabeth Berman has another smart post up on Orgtheory about the Facebook privacy debate.  She says people are nervous about Facebook because of its network externality, because your friend’s friends and their friends use Facebook, which means you have to use Facebook, and Facebook has a de facto monopoly.

People absolutely do get worried about monopolies, justifiably, and there is demonstrated evidence that networked goods do enjoy them.  The network externality is what has variously bothered people about electrical utilities, fax machines, financial markets, Microsoft Office, — and if you drive far enough down the network externalities literature in political theory and economics — things like languages and currencies.

Here’s the problem with network goods.  Microsoft Word is useless to me if nobody else uses it; it’s bad enough that some of us use different versions of it and email attachments don’t always open.  Same with my telephone, and the English language.  Buying a telephone and learning English is a waste of time if everyone speaks sign language.  Network externalities seem to imply a natural monopoly where users of a network good cannot switch platforms.

Except that’s exactly what happens, constantly.  One of the textbook examples of a network externality is a fax machine.  When was the last time you sent a fax?  Or more germane to the Facebook debate — when was the last time you used MySpace?

What grips people’s intuition about networked goods, is that for some incredibly large scale networks like financial markets or the electrical grid or Facebook, if the thing shut down entirely (which financial markets do every couple few decades for a couple days), the entire social world would shut down.  Moreover, the people who run the pipelines for these networks have a lot of data on people.

That seems to confer quite a bit of power on whoever manages such a network.  Except massive network failures like financial meltdowns veritably never happens, and in reality, it’s much more valuable to network operators to provide reliable networks to people than it is to shut them down periodically in order to extract hold-out rents (Enron found that out the hard way).  It’s also much more valuable to Facebook et al. to manage people’s data in a manner that they’re comfortable with, and to use those data in a way that benefits consumers.

Now, people’s fears of monopolies in communications and media industries aren’t just unfounded economic fallacies.  The world used to be full of them, and still is (we’re lookin’ at you, Charter).

Critical theory came out of the nascent communications industry of the early 20th century, which created very crappy and expensive national networks to broadcast content on.  Given high fixed costs to build cludgy networks, and given the government’s interest in telling people what is news worthy and appropriate for their children, monopolies emerged in culture industries out of a desire to reach as many people as possible, creating lowest-common-denominator content, and out of anti-competitive privileges granted by the state.

The big wigs in Frankfurt looked right down their nose at this development, and decided that we were all being brainwashed and homogenized, and ultimately controlled.

Well, that hasn’t happened, and there is no evidence that it is about to start.  Network technologies have gotten cheaper, exploding monopolies in culture industries, improving the variety and quality of content, and making the collecting of user data, and more importantly the unbelievable computational problem of interpreting it, infinitely more complex.

The technologies that networks are built with get cheaper because of innovation; that opens up competition and entry; and more people have access to the consumption and creation of content to send over those networks.  The content gets smarter, richer, and more variegated with the expanding division of social labor.  Opportunities for centrally-directed brainwashing campaigns decrease.

Elizabeth herself acknowledges this and contradicts herself in her own post.  She says first that people are scared that Facebook leaves them with no choices, but then also that, “the world of technology keeps changing pretty rapidly, and today’s Facebook is tomorrow’s Friendster.”  That’s right.  Network technologies constantly turn over.  Facebook won’t last 20 years, at least not in anything near the form it is now.  

Fortunes are made and destroyed, just like Facebook friends, and the more social we get, the more we volunteer who we are to one another and to the companies who help us talk, the harder it gets for them to manipulate us and tell us who to be, what to buy, and what to think.


2 thoughts on “Network Externalities Do Not Create Permanent Monopolies”

  1. Nice post. And I don’t disagree with your point about the temporary nature of network externalities. At the same time, though, “temporary” can last quite a while. And while turning Facebook into a public utility is not the answer, given that things -like- Facebook seem likely to remain important for quite a while, I don’t think that imposing some standards of expectation on how companies that control such networks behave for the duration of temporary is unreasonable. Of course, what those should be and how they would be implemented is a separate, and difficult, question.

    I also think you’re onto something about the similarity between social networks and institutions more generally, and that this drives part of people’s discomfort. We are vaguely aware that we are creating, via social media, algorithms, etc., new institutions that will be both constraining and enabling. But it seems to be happening without a lot of conversation about the process — it just unfolds as “natural” and “inevitable” and then suddenly we’re locked into a new set of rules. (Again, not forever, but for long enough to affect people — durability is what makes them institutions, right?)

    So when people start to think about the fact that a system is evolving that infringes on ways of acting that they value and thought that they could assume — like being able to act on the internet without always being followed, or seeing a newsfeed that follows some “fair” rules regarding display of statuses — even if their assumptions were already wrong, they are likely to get upset.

    Or something like that.


    1. Hello Professor Berman! Thank you for stopping by. The first unintended consequence of regulation is that it opens a revolving door for rent seeking from Facebook and others. Second, regulation will actually slow the pace at which new entrants come online because they’ll need Internet Administration approval clearance. Third, once the regulators are poring over the information Facebook has about you making sure they don’t use it improperly, the government will find a way to justify itself using your information improperly (and they have guns and jails, not just Flash apps).

      As far as being followed, I sympathize. I used a gate that some people at MIT built for a while to keep Facebook from tracking my other web moves. But I think the real fear people have is of Big Brother following them around. We ought to rail against the NSA and any channel through which government can co-opt information technology in order to exploit people (because that’s what governments do best – exploit people). Conflating private companies who make money by providing people with goods and services they want, and the state, is an old mistake that most often results in the promotion of policies that actually widen the pipeline between government and aspiring monopolists.


Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s