# Social Betworks: nothing but online pubs (Part II of II)

## We’re looking into the high priced bets being made in the business sector of social networks, and the economic theories underlying those bets. Part I looked at Facebook Inc. (NASDAQ:FB) and its recent purchase of WhatsApp Inc. for a hefty price. Part II will examine whether optimistic assumptions about the “network effect”, and the monopolies they supposedly confer on the winners, really hold up to scrutiny.

The idea of the network effect is the simple idea that, as the number of users in a network expands, the usefulness of the network expands at a much faster rate. This is because of the number of possible connections that can be made between all the users.

I’ll have to do a bit of simple maths here, but bear with me as it’s important if you want to understand the investment case for social networks made by the optimists. And also why I think those optimists are probably wrong in the real world. Plus later I’ll explain what I think really happens using the far more enjoyable analogy of English pubs, those purveyors or warm beer and stale peanuts that act as hot beds for English social intercourse.

Let’s begin.

If there is 1 lonely user of a network – let’s call him Ajit – there are no possible connections. He’s on his own. Then Bob joins and we have two users and one possible connection. Then along comes Chan, a third user, making three possible connections: Ajit and Bob, Ajit and Chan and Bob and Chan. And now Diego joins in, and we’re up to six possible connections: the three connections from before plus one between Diego and each of Ajit, Bob and Chan.

You can see what happened. As each user is added the range of possible connections goes up at a much faster rate than the rate at which users were added. Adding Diego increased the number of users by a third, from four to three. But the number of possible connections doubled, from three to six.

The formula needed to work this out for bigger networks is pretty simple. If n is the number of users, then the number of possible connections is n times n-1 divided by 2. That’s n * (n-1) / 2.

Below is a chart which shows the relationship up to 30 network users, which gives 435 possible connections (30 times 29 divided by 2).

Ok, so now we understand the basic principle of the network effect we can look at how it applies to social networks. The optimists would say this means that, as the numbers of users of social networks go up, then their usefulness goes up much more quickly due to the faster growth of possible connections.

WhatsApp has roughly 500 million users, which using this formula gives it almost 125 quadrillion possible connections between users (that’s 125 with 15 zeros after it, or 125,000,000,000,000,000). Facebook has 1.35 billion users, which gives it just over 911 quadrillion possible connections.

Notice how Facebook has 2.7 times as many users as WhatsApp but the number of potential connections is 7.3 times more. This is the power of networks. Utility to the users goes up much faster than the rate at which users are added.

At least it would in a true network, where everyone was a potential contact of everyone else. In fact, when it comes to usefulness for individual users, the network effect is severely limited in many of these businesses, including both Facebook and WhatsApp.

Say you sign up to Facebook. You have 30 good friends in the real world. Say 10 from school days, 10 from university, and 10 from your current life.

For each group of 10 you’d like as many of them as possible to be on Facebook so you can be in contact with each other. If only one school friend is also a user you may as well use email or the phone, and forget Facebook altogether.

But if all 10 school friends are there you can interact with each other. It’s like having a permanent school reunion, with 45 possible connections between the group (10 times 9 divided by 2). People can pass on their latest news to the group, post photos, share jokes, and you can all chat about it.

So far so good. The more of you on the network the better it becomes within your group. And the more of your groups that are on there the better also: friends from school, university and your current situation can all enjoy themselves.

And then you can have unlimited numbers of individual contacts that aren’t related to one another. Someone you met at a party and promised to stay in contact with. Maybe a prospective date. Or a distant relative that you hardly ever see in person.

But although you could have hundreds or even thousands of such one-to-one contacts, the more you have the less each one is worth. In economic terms there is “declining marginal utility”. More volume of contacts equals lower quality of contacts. You’re spread thinner and thinner.

Years ago I read a study claiming that humans have evolved to be capable of having around just 80 close acquaintances, including friends, family and business contacts or colleagues. This is because we evolved living in groups of about that size or smaller – first as troops of monkeys and later in villages. You had to know your friends and also your foes –  from other troops of monkeys, or neighbours from other villages.

You can know a lot more people than 80, but you can only know a certain number well.

Based on personal experience that feels about right. You can know a lot more people than 80, but you can only know a certain number well.

So if all your different groups of friends are on Facebook then there is value in it to each group. But not between the groups that don’t know each other. Or between the individual contacts you have that don’t know each other. They’re all too busy with their own close friends.

Much less is there value to you when Facebook continues to grow its business further. If most of my friends are already present then it makes no difference to me at all if more users join on the other side of the world. That’s just as true whether it’s a few dozen or a few hundred million more.

Let me explain what I mean by using the analogy of an English pub. Let’s say you live in a village and at the weekends you meet up with your friends for drinks and conversation. You catch up on the latest gossip. You share jokes. You play games.

Now let’s say that pub is part of a growing chain of pubs, with establishments in other villages and towns. Does it matter to you how big that chain becomes? If there are ten pubs run by the same company, or 100, or 1,000 will it change you experience in your own local one? Of course not.

Each pub is a different sub-group. It’s own miniature social network. It may sell the same beer, play the same music and serve the same food as other pubs. But the people in each pub vary, mainly coming from the local area.

I believe social networks like Facebook are more or less the same as chains of pubs in network terms. It doesn’t matter how many pubs there are in the chain, or how many users there are on Facebook. You still hang out at your “local”, with your own friends.

I believe social networks like Facebook are more or less the same as chains of pubs in network terms. It doesn’t matter how many pubs there are in the chain, or how many users there are on Facebook. You still hang out at your “local”, with your own friends. The value of your time at the pub is specific to the one you frequent, not on how many other pubs are available in theory.

Now let’s say you have kids, and so do your local friends. You all take them along to your pub for lunch or dinner, and they play in the playground while you do stuff for grown ups. But eventually the kids become teenagers and then young adults. Then they no longer want to hang out all the time with Mum & Dad, and their ooooh, so boring old friends.

So what do they do? They move to another village pub, which is new and isn’t part of the pub chain. It has an edgier, cooler, younger atmosphere. The live jazz piano that you like so much is replaced by the latest hit songs, played loud. Beer and wine is replaced by cheap and potent cocktails. There are no screaming young kids, and more importantly no oldies. There are unisex facilities for relieving yourself, instead of the traditional segregation between the ladies and the gentlemen.

I reckon this is the threat to Facebook, however big it gets: new and cooler competition will emerge. I’d be amazed if my kids use it in five years time, let alone ten years. Who wants to hang out with their parents all the time? So, as the years pass by, groups of friends will easily be able to set themselves up in new places, to use new networks, to move to a “new pub”. (Maybe I’ll have got bored of it as well…)

The theory is they have a monopoly due to first mover advantage, and that this will be preserved due to an entrenched network effect. And current users are assumed to remain active, and indeed become more valuable prospects for advertisers over time as new features are added to the base product.

In other words I don’t accept the theory that the likes of Facebook have true monopoly status or even quasi-monopoly status. The theory is they have a monopoly due to first mover advantage, and that this will be preserved due to an entrenched network effect. And current users are assumed to remain active, and indeed become more valuable prospects for advertisers over time as new features are added to the base product.

As with so many theories, they don’t stand up to hard scrutiny. Just by applying a little thought about how people behave in the real world we can quickly find holes opening up.

Little groups will be able to break away from Facebook and other networks all the time, as new platforms keep emerging. And existing users may get bored and less active. So, in my view at least, ongoing growth and market dominance are far from certain for today’s market leaders in these spaces.

The same isn’t true of some other internet businesses or tools, which do have the ability to create or preserve virtual monopolies. For example, Google Inc. (NASDAQ:GOOG) has achieved clear leadership over competitors, originally by simply having the best search engine. Many of Google’s original competitors died an early death, although some still limp along.

But even Castle Google may not be impregnable. Who would have thought 15 or 20 years ago that the volumes of Encyclopedia Britannica would become obsolete as the main trusted sources in the world for reference information? Or that Yellow Pages, the ubiquitous, fat directory of local providers of goods and services, would no longer be propping open our doors? Just because we don’t know what’s around the corner doesn’t mean there isn’t something lurking there.

But unlike Facebook, even certain social networks could maintain dominance because of the nature of what they do. For example LinkedIn Corporation (NYSE:LNKD) is a business networking site. People use it to look for jobs and business contacts and share business ideas. This system has genuine network effects, where it’s value to each user increases substantially as the user base grows, although it still has its limits. For example maybe it’s just too generalised, and niche sites, tailored to specific sectors or countries would be better. After all, few people living in Moscow are looking for a job in Johannesburg. However, in general, if you want a great job you’d best go to a huge convention in a big local city, not a sad little affair at the village hall.

Not all social networks or other internet based market leaders are created equal when it comes to network effects, or their ability to monopolise their sectors as a result. But social networks like Facebook have their limitations which makes me doubt that they will prove to be good investments at current prices, however much they grow in the immediate future.

Remember, Facebook shares are priced at 72 times last year’s earnings and nearly 17 times sales. As I explained in part I there is a huge technology profit mountain to climb to justify that kind of valuation and still give investors a reasonable return for the risk involved. Not impossible, just highly challenging and uncertain.

Of course Facebook could continue to buy growth as well. In the same way as it bought WhatsApp it could buy other new user networks (that are loss making…).

But then it just becomes a conglomerate of individual networks, bought at excessively high prices. It doesn’t solve the underlying problem of how to prevent people switching or how to justify the exorbitant prices paid for unproved businesses.

Today Facebook has a highly successful business and is fun to use, even useful. But as with many areas of technology that doesn’t automatically make it a good investment. The share price could go to the moon and stay there. But given all the uncertainties I don’t rate its chances.

Buying Facebook shares, and the like, is speculation and nothing more. A gamble. A bet. And making bets can be fun, if done in small doses. Although, for my own part, I’d rather get my kicks with occasional trips to the horse races. So, just make sure you don’t confuse buying shares in this kind of speculative business with real investment.

Investment is a totally different concept, in my book at least. You don’t invest for thrills or excitement. You do it to make a reasonable profit for a reasonable risk, assuming you approach it correctly. True investors wouldn’t go near the social betworks.

Stay tuned OfWealthers,

Rob Marstrand
robmarstrand@ofwealth.com

Previous ArticleNext Article