The efficient level of RMT in MMORPGs

The efficient level of RMT

I am not intensely interested in MMORPG secondary markets that exist despite the operator’s wishes, and I certainly don’t have any agenda of legitimising them. But after the last SoP V workshops yesterday there was a rather lively coffee table debate on RMT in MMORPGs that resulted in a neatly economistic re-telling of that story. I’ll try to convey the tale without misrepresenting the particpants too much.

Already during one of the panels Joshua Fairfield reminded us of the positive aspect of RMT: trade creates welfare gains (“Every time you stop someone from trading, God kills a kitten”, as he politely put it). Richard Bartle obviously wouldn’t have any of that and brought up the negative aspects, centering on the violation of the achievement hierarchy associated with character levels.

As far as I know, whatever negative effects RMT has on other players can be modeled as a negative externality (as Castronova originally did). The net welfare gain from RMT is then the welfare gain from trading minus the externality. The result is often believed to be negative.

So far the usual story. But an important point that came up in the discussion is that both of these variables are a function of some measure of the prevalence of RMT in the game (the negative externality might also be a function of simply how visible RMT is).

In conditions of high prevalence/visibility, the negative externalities are usually assumed to be greater than any possible trade gains. But in lower degrees of prevalence, trade gains may well be greater than the externalities. A point where a marginal increase in prevalence results in more negative externalities than trade gains is the efficient level of RMT (Figure 1) (i.e. Pareto optimal, given a few assumptions). Where this point is depends on the preferences of the player base.

Can we link this neat construction with reality in any way? MMORPG operators have an interest in pleasing their customers. Assuming efficient pricing, player aggregate welfare correlates with the operator’s revenues. Joshua and others pointed out that MMORPG operators are not doing everything in their power to stop RMT. For example, they are not commonly banning players for buying, only for selling. A certain level of RMT is tolerated as long as it does not become too overt. This could indicate that there is indeed an above-zero efficient level of RMT towards which operators are gravitating.

Bartle world

A special case, a game where the efficient level of RMT is zero, I will call a “Bartle world” (Figure 2) (the term was used in the discussion to connote a slightly different concept, a world where the prevalence of RMT is zero). Joshua asked, do Bartle worlds exist at all?

If you enforced a ban on RMT with such zeal that every single trader would get caught and have their limbs removed (as per EULA), wouldn’t you be left with a player base that consist of 100% RMT haters, thus creating a Bartle world?

I think it was Dominic Zou who pointed out that it’s not like there are two populations out there, RMT lovers and RMT haters. Instead, people act with various degrees of schizophrenia, applying a different standard to themselves and to others. Gains are reaped by traders and costs borne by others. The same person can simultaneously both appreciate the gains and resent the costs.

Therefore, even if you were able to build a world with zero RMT, that might not be the efficient level of RMT. The player base might still be better off with a teensy bit of RMT. This is not to say that Bartle worlds are impossible, just that they seem unlikely.

Does this mean that designers should not attempt to build MMORPGs with zero RMT? Not necessarily. Operator’s costs like the cost of customer service time spent on RMT fraud and the cost of enforcing an RMT ban were not considered. The most important factor not included in the model was revealed by Richard Bartle when he said, “RMT is not part of the message I want to convey” or something to that effect. If you are creating a work of art instead of a commercial product, economic performance may not be a design criterion.

You may also be interested in a paper by Jun-Sok Huhh from last year that uses the same economic concepts of externalities and welfare in a completely different model involving RMT.


8 thoughts on “The efficient level of RMT in MMORPGs

  1. >This is not to say that Bartle worlds are impossible, just that they seem unlikely.

    I’m glad you didn’t say “impossible”, because we have indeed had these virtual worlds in the past. Neither MUD1 nor MUD2 had RMT, and no kittens were harmed in the making of either production.


  2. You quoted me right. I wanted (but failed) to make the point that there is also some elasticity in the personal stance of players, and this elasticity is itself elastic since (as you pointed out) people can be hypocritical.

  3. Thanks for your comments.

    “Neither MUD1 nor MUD2 had RMT, and no kittens were harmed in the making of either production.”

    Agreed. But the point is that even if you have a game with zero RMT, the theoretical efficient level of RMT could still be non-zero. Players might have been better off with a little bit of RMT taking place.

    Assuming so, why was there no RMT in practice? One thing the model does not consider is transaction costs such as search and negotiation, which in the absence of liquid markets can be high. “A little bit of RMT” may be hard to realise.

  4. Perhaps that’s a question of relating purpose to moral imperatives? If RMT is valued as morally bad, “a little bit of RMT” might be like being “a little bit pregnant” 😉

  5. I thought this conversation was the best one at the conference. Richard, we should take this one on the road, with Vili to keep us in check.

    I want to say at the outset that the positions I present are not “right” nor are others “wrong.” This is a point of view that I hope advances the discussion.

    A few points: Richard and I agree that there are low externalities models in the following states:

    1. People who want to trade / The world permits trade.
    2. People don’t want to trade / The world prohibits trade.

    The obvious problem: the third area, where all the juicy bits are. There, some people want to trade, some of the time. And some of the time, some people don’t want other people to trade. (The “keeping up with the Joneses” phenomenon).

    It seems to me that cases (1) and (2) aren’t that interesting to talk about. The third is hard. I really like Vili’s idea that this is susceptible to empirics, but I’d be cautious. For example:

    1. How seriously do we take moralisms involving other people’s trades? Quite seriously in the case of cocaine; of limited seriousness in one’s purchase of meat over vegetarian fare (as yet).

    2. How intensely are those moralisms felt? This idea of the “achievement hierarchy” doesn’t touch me. Yes, it subverts the value of the hard work that I put in to be able to afford my house if Paris Hilton can sashay up and buy it.

    But do we really want to treat other people’s wealth or success — in this world or the next — as a real externality? I don’t. It may exist (e.g., I’m jealous of my neighbor’s car) but it’s an externality we’ve always ignored, except in revolutionary societies.

    3. How do we value conflicting moralisms? Perhaps I resent the time aristocracy as much as others resent the moneyed elites. Why does the one set of externalities get sternly modeled, and the other ignored?

    And finally, then, is Vili’s point: how prevalent is RMT in virtual worlds. I don’t know — I have some guesses,based on Sony’s releases, but nobody knows for sure in worlds that don’t permit it. Certainly the game companies are not about to tell.

    So: I think we should survey people and ask them questions about RMT. But without documents that I am sure Blizzard will not release, we will have a hard time finding much out about case three worlds. And, even if we did know, it wouldn’t change much — since we have no idea how to value these moralisms. And, finally, it seems to me that the weight that has thus far been put on these moralisms is inappropriate, given that the negative externalities of “keeping up with the Joneses,” have, appropriately or not, been downplayed in neo-classical economics.

  6. A couple of comments:

    1. There are numerous ways to make in-game RMT impossible or impractical through game design techniques – to quote Joseph McCarthy “I have a list”. Whether they would make the game more fun or successful is a completely different question.

    2. RMT should probably be considered a separate problem from account sales or “power-leveling” services, though both are interesting.

    3. Dr. Bartle has actually captured the “RMT Problem” problem quite well with the statement:

    ‘The most important factor not included in the model was revealed by Richard Bartle when he said, “RMT is not part of the message I want to convey”‘

    The problem with this statement is that not all players in a game have the same value system for their participation as Dr. Bartle (or the specific game designer(s)) and that the game itself does not inherently or strongly enforce its value system through its implementation – the Achievement hierarchy.. or whatever.

    To abuse Dr. Bartle’s categorization of players – their are those who want to earn Achievement achievers and their is the class of Socializers and Griefers and Explorers who want the status or capabilities of having Achieved without the nuisance of actually playing the Achievement game.

    So, because the “cool kids” are Level 70 with certain equipment, Socializers who want in will buy their way in.

    Because you can kick a** with “kewl loot”, Griefers want the best car on the block.

    Explorers want to go everywhere and experience everything in comfort… so they buy themselves into “Safe Safari” mode.

    If this reiterates or misstates Dr. Bartle’s discussion of Player types, I apologize.

  7. /quote
    There are numerous ways to make in-game RMT impossible

    You mean by completely banning ANY TRANSFER of objects and/or value? Yep.

    You are missing one very point. There is NO WAY to get sufficient proof of money transaction itself. There are only ways to detect an item/value transfer in-world. So, no matter what publishers are saying they are banning RMT participants WITHOUT SUFFICIENT EVIDENCE. ALWAYS!
    How does that correlate wiht a ‘moral stance’ they are talking about, huh?

    There are two models in ANY society – free enterprize and socializm/voluntaristic (may be called
    ‘Achievement Based’ 🙂 ) distribution of wealth (on a level of political model it becomes a totalitarianism RIGHT AWAY).

    UNREALISTIC. People will do what they want to do, you can’t control them no matter how much you would love to. You become a Hitler – they leave, that’s it.

    P.S. Just for your information, Hitler WAS ELECTED by German people (probably they were ‘against RMT jews’ at the time, see ‘Crystal Nacht’). Think about it some day, pleeeease. 🙂

  8. Godwinning aside, perhaps the all-powerful coders may have a way to carve RMTs out of, well, Ts generally. This may require practically all items to be available for free, or making scamming or robbery (in terms of in-game currency) part of the game.

    Of course, the plea about there not being sufficient evidence presupposes there being some sort of right to not have certain things done to you without sufficient evidence. And this ties to the concept of property. If perhaps the in-game items and attributes and so on were less important than, say, contributing to the development of a game (or finding out ways to solve certain game puzzles, etc., …) perhaps that might undercut the traditional notions of property that maybe, just maybe, should never have crept into certain games in the first place.

    Of course, I do not consider WoW to be one of those games. Where the games are created for a profit motive, I would suggest that all gloves be off.

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