Tuesday, January 6, 2015

The Rabbit Hole: Why Is RMT Bad?

Over the past 2-3 years, I wrote thousands of words concerning the subject of real money trading in MMORPGs, mainly in EVE Online.  I attempted to write a comprehensive post on the subject of RMT on a few occasions in the past, but much like Lewis Carroll's Alice, I always find myself distracted as I venture down the rabbit hole into the world inhabited by botters, hackers, and gold sellers.  So, instead of a comprehensive work, I decided to write about the subject on Tuesdays.

This initial post in the series will address a topic I have never addressed directly before: why RMT is bad.  But first, I must define the term real money trading.  At the end of 2013, I defined RMT as, at its most basic level, the exchange of virtual goods, including in-game currency, and services for real world currency.  I then, relying on the work of two researchers from the Helsinki Institute for Information Technology, divided real money trading into a primary and secondary market.

The HIIT's Tuukka Lehtiniemi in 2007 defined the primary market as trade "in which the operators of certain services sell virtual items to users for real money. Such services are usually free to use. Typically players buy the service’s own internal currency with real money, and use that currency for microtransactions inside the service." Services like this include the cash shops that often appear in games today.  Sony Online Entertainment's Station Cash and ArenaNet's Gem Store for Guild Wars 2 are two obvious examples of game operators becoming directly involved in selling in-game items.  A growing model is the PLEX-model, in which game companies sell items that players can either redeem for game time or sell to others on the in-game market in exchange for in-game currency.

Dr. Vili Lehdonvirta, who will soon head a major 5-year research project funded by the European Research Council, defined secondary markets as markets in which "virtual world users buy and sell virtual assets between each other without the operator’s involvement."  The secondary market is what most players think of when the discussion turns to RMT.  Activities such as purchasing in-game currency and items from 3rd party websites along with power leveling services make up the bulk of the transactions.  Game companies usually frown on these activities and will ban those players who engage in them.  Because these activities usually violate the EULA and/or terms of service for games, I usually refer to this type of real money transaction as illicit or unsanctioned RMT.

But why do the operators of MMORPGs frown on secondary RMT activities to the point they ban paying customers from playing their games?  Dr. Edward Castronova of Indiana University, one of the founders (if not the founder) of the academic study of virtual world economies, came to the conclusion that RMT is bad for games.  In explaining why, he introduced RMT into a game of Monopoly...
"There is, however, a failure in the RMT market, because some welfare effects of the trade fall on others. To see this clearly, imagine playing a game like Monopoly where two of the other players engage in a side deal using their real dollars. Player A is losing to Player C, but owns Park Place. Player B is also losing but owns Boardwalk. A pays B $US50 for Boardwalk; with this monopoly, A wins the game. Doubtless, A and B are better off – A now wins the game, and B, who was losing anyway, has $US50. Yet players C and D, who were not party to the transaction, are worse off. First, Player C would have won had the trade not happened. Second, and both C and D experienced a degradation in the nature of the game itself; it became less fun. This loss of fun, more broadly, can be seen as an externally-imposed disturbance of the game, a perturbation away from the gameplay as intended by the designers. The costs are borne both by users, who get less utility from the product, and by the designers, who realize fewer revenues from the sales of games. In this, RMT is like a pollution of a service that the designers are attempting to provide to their customers. Such uncompensated interdependencies can be clearly recognized as market failures potentially worthy of some sort of policy response."
Dr. Castronova posited five reasons why real money trading is bad for MMORPGs.  The first reason is that real money trading disrupts a game's fantasy atmosphere.  Just as the introduction of real money disrupted Dr. Castronova's example game of Monopoly, the activities introduced into an online game by RMT can alter players' game experiences away from that intended by a game's developers.  When designing a game, the developers have to assume that players will act within certain boundaries, which is why every game comes with end user license agreements, terms of service, and/or other types of rules of conduct that players agree to abide by.  No matter how well designed the game, when enough people chasing real world profit trample those limits, the game play of others is impacted.  Dr. Castronova's next three reasons help explain why.

The second reason is that RMT induces activity that causes inflation in a game's economy.  The virtual economies found in MMORPGs is usually based on a faucet and sink model.  Game currency is introduced through faucets like killing NPCs and completing quests while currency is removed through sinks like user fees for selling items in the auction house, skill learning, and item repair.  Those engaged in RMT will engage in currency faucet activities far longer than the regular player.  Ideally, at least in the eyes of the RMT operations, characters will run 24 hours a day, whether through shift work in a Chinese gold farm or through the use of bots.  If the number of gold farmers becomes too great, then the economic calculations of the game developers fail and inflation is the result.  If the inflation rate becomes too high, the average player may wind up discouraged, and either quit or, worse, contribute to the problem by purchasing game currency from an RMT operation.

Next, Dr. Castronova noted that real money trading induces gold farmers to occupy territory and utilize system resources at a high level.  Elder Scrolls Online suffered such a plague at launch, with bots coming to life to monopolize boss spawns in public dungeons and relentlessly harvesting materials.  Also, with a lot of characters in an area, the frame rates of the computers of players can slow down to a point the game becomes unplayable.  Through in the effect that poorly coded bots can have on the servers, as CCP demonstrated following its Unholy Rage anti-bot operation in 2009, and gold farmers and bots can negatively impact the play of many players.

The fourth reason is that real money trading results in those engaged in the practice to appropriate game communication systems to advertise their product.  In the worst cases, local and global chat channels fill up with RMT spam so fast that they become unusable for regular players looking for help or to contact others.  In-game mail boxes can fill up with spam, which is bad if finishing quests can result in players receiving rewards through the in-game mail system.  And RMT spam bots can repeatedly directly contact players to the point of annoyance.  If a game cannot prevent such activity from making the game communications systems unusable, the company is likely to lose a lot of players who leave in disgust.

The final impact that RMT has on online games is an increased cost in operating the game.  Or, as Dr. Castronova put it, "an increase in service provision costs, relative to the costs of a hypothetical counterfactual game in which RMT did not happen."  These increased costs represent not only the additional customer service staff required to handle complaints from regular players about the activities of those involved in real money trading operations, but for the additional staff required to handle criminal activities committed against the game, such as hacking accounts and credit card fraud.

Once again using the launch of Elder Scrolls Online as an example, two-and-a-half weeks after launch, lead game designer Matt Firor published a "State of the Game Address" to the player base in which he acknowledged the impact of the RMT operations on the Zenimax staff:
"The scope of the black market activity accounts for up to 85% of Customer Service emails/calls. Because of this huge influx of contact relating to this one issue, our CS team has been slower to react to other problems than planned – our sincere apologies if you have been held up for a long period of time waiting for CS to respond to you. Again, our goal is to keep this activity away from you so you don’t have to contact Customer Service in the first place."
Activision/Blizzard also acknowledges the effects of hackers on the staffing requirements for World of Warcraft:
"The overwhelming majority of purchased gold comes from stolen player accounts, in which character inventories are stripped of value, liquidated into gold and sent off to be sold. This halts the victims’ ability to play the game and contribute to their guilds.
"It’s also costly for Blizzard, as we task hundreds of Game Masters and account specialists to assist compromise victims, help secure accounts, and restore their characters. These efforts require time and focus that could otherwise be spent addressing player requests more quickly."
Now, if all RMT were to magically disappear tomorrow, I don't believe that the game companies would keep all of the staff currently handling hacked accounts and other RMT-related complaints.  However, I do believe the complaints would drop significantly, thus requiring less staff.  Could the money saved have gone to save some of the jobs cut due to layoffs over the past few years?  Probably.

I realize this post ran rather long.  But the reasons that real money trading is bad for online games don't fit into a nice little post.  In the coming weeks I will use this post as the basis for traveling down the rabbit hole into what I hope is an in-depth and worthwhile look into the subject of real money trading and the associated issues involved with the practice.


  1. I look forward to this future posts in this series. One of the things I’ve occasionally pondered is how companies like CCP are quite willing to allow customers to turn real life cash into in game virtual goods and currency but squeamish about moving in the other direction by enabling (or even allowing) players to turn in game virtual goods and currency into real life cash.

    From an in game perspective it seems cashing out would have limited economic affect if supported by CCP since it would sink out the virtual goods and virtual currency being exchanged for real life money.

    After a little thought however, allowing players to extract real life cash from in game activites would mean that all in game goods and currency would have a genuine, identifiable and exchangeable real life value meaning virtual shenanigans would generate real life economic harm (and profit). Queue real life police, attorneys and courts investigating in game actions, agreements, transactions . . .
    Hmmmm. What a mess.

  2. The reasons posited by these academics are not that highly compelling. Take inflation. Presumably the game designers can control both their faucets and sinks, and manage to balance them. So that's a red herring of sorts. It's true that the net effect here is to make it hard to grind in-game money; but so long as the game remains compelling I don't see a problem. Same with people "occupying territory"; presumably the game designers want people to do this. If they design properly, it should be a feature that their game-world is well-lived in, and not a problem. If bots cause problems, it's a failure of game design.

    The problem of controlling criminal exploitation of players is not a game design issue, but one of normal computer security. It is indeed a cost, but one which I doubt is that high if properly designed for. I mean, banks exist online. Stock trading businesses. Thousands to millions of dollars are at stake. They manage.

    Spam is another problem of game design, not fundamental. EVE, for example, does not have spam because it costs ISK to mail people. Yes, everyone turns that off, but the default is on, and if there were any serious spam problem, we'd all turn it back on.

  3. Here is my article on why secondary RMT is bad for EVE: EVE and RMT. Sample:

    By making the accumulation of ISK easy, and having nothing to spend it on but spaceship violence, CCP induces lots of spaceship violence in their game of spaceship violence. And that's a good thing, because lacking spaceship violence nobody would play EVE.

  4. And don't forget tax, bank and stock market laws: as soon as people can earn money through a service, the financial authorities get really interested.

  5. "but as long as the game remains compelling" - that is the problem. There are few ways of grinding in-game value which are truly compelling; the only reason that people bear it is the fact that they know that all other players are in the same boat (ok, some folks are strange enough to enjoy grinding for the sake of grinding). There is a reason why even you call it "grinding" and not "having fun".

    Introduce bots and illicit RMT in this place, and suddenly you have players who gain all the benefits of grinding 24/7 without putting in any of the effort, making it hard, if not impossible, for regular players to compete through normal gameplay with bearable amounts of grinding.

    "failure of game design" - you like writing that, but I rarely see you giving examples of how the design in specific cases could have been done better. For example, how would you have designed the public dungeons of ESO?

  6. Dire and Druur, thanks for this. I'll add the topic to my list to write about. Project Entropia actually had to deal with the issue.

  7. Reamde is an interesting novel and one I thought about reviewing. But I think a game would need to be designed from the ground up to handle full-blown RMT. As far as I know, only one game, Project Entropia/Planet Calypso, is. I'm skeptical of the ability of games not so designed to make such a change successfully.

  8. Of course grinding is not fun. Grinding is necessary in a game like EVE, to produce real loss. There's no reason why grinding needs to be particularly profitable. Indeed, a money-making activity can be too profitable, not from the player's POV of course but from the game designer's. Witness the factional warfare exploitation of two summers ago.

    I don't think people grind in EVE based on any notion of equality with others. They do it for money. I think most people are well aware that the grinding they do is not as profitable as ratting in null. Highsec is full of miners grinding for 10m/hour or less.

    I never played ESO, so I don't understand what the problem really was. Give me a problem that robots induce in a game that I know, and I am happy to opine.

    I did give the example of CSPA charges in EVE as an example of good game design solving a problem.

  9. CSPA does indeed tackle the mail spam problem, but would do nothing about chat channel spam (which in EVE luckily happens only in the trade hubs' local)

    While I don't play ESO either, but I read a bit about the public dungeons: they are non-instanced dungeons which can be entered by anybody, so people can band together to defeat the final boss. Unsurprisingly, from what I gathered, bots started dwelling in those dungeons ("occupying the territory") to farm the bosses upon respawn before actual players could get to them.

  10. So I don't have the reference on me, but I believe there is an emergent money laundering market using MMOs globally and that CCP is trying to get in before it starts in Eve.

  11. I heard Markee Dragon tell a story about the Russian mafia using UO to launder money using credit card fraud. There's probably several such stories if I dig hard enough.