The effect of social networks on RMT

Back in January 2012, this paper was published in the journal Information Systems Frontiers. The authors explore several hypotheses of real-life demand for virtual goods, one of which I thought was worth a blog entry, all of which are worth reading the article for yourself.

By way of introduction, if you've done enough background reading on the real money trade and its pros and cons, you probably came across a paper or a blog post that made the following claim: Demand for virtual goods is positively related with social network density because of the so-called "bandwagon effect". The idea is that players with less time and more money can't "keep up with the Joneses" through play time, so they make up for it by buying virtual goods. This is one of the fundamental hypotheses underlying Castronova's cost-benefit analysis of RMT which came out in 2006, but the idea was being bounced around earlier than that.

Anyway, here come three gentleman who say "Actually, RMT demand is decreasing in the density of social networks." More specifically, they hypothesize that as the importance of social networks to game achievement increases, the value of a virtual good to an individual player falls. So suppose you had two games that were exactly the same, except that one requires you to work by yourself and the other requires you to work with a partner. Then, holding all other things constant, increasing your individual level of effort has less of a reward in the second game than it does in the first game. Therefore, any player of the second game will have less demand for virtual goods purchased with real money. It is more advantageous for them to spend their time working with others instead of buying goods to improve their avatars. (I know that we can dissect this hypothesis quite a bit, but I prefer to leave it intact. Take it for what it is.)

If you read the article, you may be skeptical of how they proxy for social network density. You may also claim that the measured effect of social networking is biased because it is endogenous, i.e. players who choose more social games may already value virtual goods less than players who choose less social games. Nevertheless, the authors find that their measure of the importance of socializing is negatively related with the price of virtual goods. Even if the numbers are wrong, the idea is theoretically sound from the perspective of economics.

What are the implications of this? Well, first of all, it doesn't negate the possibility that there is still a "bandwagon effect" that increases the value of virtual goods. But it does suggest that the bandwagon effect can be overcome by a social network density effect. Second (and more tantalizing), it means that all that time F2P companies spend trying to get your friends into the game may actually be counter-productive. Why? Because if a mostly single-player game that's about, say, simulating a farm becomes more of a multiplayer game about, say, hanging out with your buds, then your incentive to buy things from the game company diminishes!

Say what?! Yep, those social game companies might actually be doing themselves a disservice by creating... social games.

The paper is a good read for other reasons, too. For example, the ability of players to trade items in-game is found to be negatively associated with their real world value. But even this is related to the social density of the game, I think. After all, an economy's vibrance depends on a certain level of density of market participants. (I'll try to address this latter point in a future post.) In the mean time, some other follow-up questions present themselves: What kinds of social networks are more/less likely to incite demand for virtual goods? Are goods that support the group more valuable in games with denser social networks? Does that suggest a way for social game companies to progress through various product phases? I'm curious to hear your thoughts!


The effect of social networks on RMT by Isaac Knowles, unless otherwise expressly stated, is licensed under a Creative Commons Attribution-ShareAlike 3.0 Unported License.

Leave a Reply

Your email address will not be published. Required fields are marked *

You may use these HTML tags and attributes: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <strike> <strong>