MMORPG Monetarism

“Inflation is always and everywhere a monetary phenomenon.” Milton Friedman, the most important and second most influential economist of the last century, died last week. You can see the fundamental truth of his work every day in some of the oldest and most popular online games.

Most online games print currency the way that third-world kleptocracies do: whenever it is convenient and in demand, with little thought for long-run consequences. This works in the extreme short run and leads quickly to ridiculous inflation.

In-game currency usually becomes greatly devalued in the first year the game exists. Your game may track wealth in gold pieces, pyreals, ISK, dollars, or something else; you will start by scrimping to get 100 of them but quickly need 100,000,000 to get anything useful. This happens at several levels for a few clear reasons.

First, let us set aside the problem of twinking and the low-level game. Your game scales, and a level 50 earns as much gold in one fight as a level 10 does in a day. If you kill a dragon and give the loot to your alt, he will have enough money to last him for twenty levels. In EVE, I borrowed money to buy my first ship and skills, quadrupled that in a few days, then repaid Ethic. The low-level game is usually irrelevant to the economy.

So we advance to the case where the base cost of everything is measured in the millions. Maybe your game hides that behind a tiered currency, where one platinum piece is million copper pieces, but you are still using numbers that are vastly higher than what you started with. The vendor cost of a Sword of Leetness is 12,500,000 (gold, ISK, whatever). Strangely enough, that same vendor will only pay 5,000 for it, so he must get a heck of a markup.

Around this point, the vendors start to become irrelevant, and the game switches to a de facto player-based economy. The vendors do not sell anything you want, except for basic supplies that you need at all levels. Odds are that the vendors will never sell that Sword of Leetness, even at 250,000% markup; it instead gets deleted when you get your money. If you want to buy a Sword of Leetness, you need to talk to a player, and you will need to give him something he wants.

How much will someone need to pay you for your Sword of Leetness? You already get 20,000 (gold, ISK, whatever) per fight. Is it worth 100 fights to you? 1000? Your second Sword of Leetness is practically useless to you, unless you can fight Two-Fisted Monkey Style, but you know it is still worth something to that guy, so maybe you can get 100 or 1000 fights worth of gold for it.

For the first few Swords of Leetness on the server, that relationship holds pretty well. The price is astoundingly high, there are various complaints, and the economy functions as you would expect it to under conditions of scarcity.

Now you are at the level cap, with all your necessary skills and some good equipment. You have enough money to twink every alt you can make on that server. Your in-game expenses are near-zero, as is the case for every level-capped character. Money exists to pile up into larger numbers. Hey guys, why don’t we farm for an hour just to see how much gold we can get?

Let everyone on the server do that for a month. Let a number of level-capped characters quit and give away all their money before they go. Hey, why does the Sword of Leetness cost 250,000,000? The old “astoundingly high” price is now what you tip newbie characters who help you with some little task.

Currency has ceased to be a store of value because it never really was. The game is not a place like your job where the money you get is received from someone else. The money you get from the enemies is created ex nihilo, *poof.* Every time someone kills something, money comes into existence.

Every time someone shops at a vendor, money goes out of existence. The game is not a place like a store where the money you spend is received by someone else. The blacksmith is not investing in an upgraded forge to help him produce better items. Better items *poof* into existence when dragons die, so no one needs to make them (and no one can make them anyway).

Games use this creation and destruction of money to simulate the circulation of currency in a real economy. The world works as if the blacksmith donates all that money to needy goblin orphans, who are then slain as their parents were so that a new generation of heroes can buy spell components.

That simulation breaks down when more money enters the economy than leaves it. A potentially unlimited number of players can spend every waking moment slaying an endless supply of immaculately conceived monsters, each of which is a walking bag of xp and gp.

Do I need to belabor this point? You have seen it a dozen times. Players exchange money by the wheelbarrow. If wallets were not weightless, every hero would be crushed by the mass of gold that he carries around. Asheron’s Call used to have a currency that was neither weightless nor spaceless; you would frequently see characters literally paralyzed by fortune.

Players then implement a new de facto currency that is somewhat less subject to inflation. This is generally some in-game item that is either consumable, useful, and portable (similar to how salt was once used in the real world) or else so hard to get that the economy will not be flooded (for a while). This new item will have a floating exchange rate against the official currency, and you can watch its price rise into the millions over time, then sink as people successfully farm it until a new de facto currency comes along.

Alternately, just watch eBay. The first currency sales for a game might be $20 for 100,000 of the official currency. A year later, the gold farmers are selling 10,000,000 for $9.95.

Alternately, watch the new gold sinks that are implemented. Developers frequently create new ways to destroy currency, often for limited gain, but what else are you going to do with it? When City of Heroes implemented bases, you paid for them by ceasing to earn money while playing. Do mounts in your game cost 1000 gold? Someone needed to fix a big hole in the economy.

Of course, some of this then encourages the gold farmers. Would you rather farm trash mobs for twenty hours or work at your job for a half hour and just buy it for cash? As many governments have discovered, big interventions in your economy create big distortions elsewhere.

Can we really say that inflation is under control if the fix was to make the new stuff cost 100,000,000? We are still tossing around those multi-million numbers, but this time the price-gouging jerks are the ones who are already getting your $13.95/month.

So what do we learn from all this? The spontaneous creation of currency is a dangerous thing. Games try to control this with money sinks, but no one has yet perfected that, usually causing both excessively tight money and massive inflation at various points of the game.

We should not expect developers to get this right. Real world governments and economists have been trying to manipulate currency and markets for centuries, and most interventions have increased suffering while decreasing efficiency. Developers have the added advantage of controlling the laws of physics, but few of them have strong economic backgrounds.

A few games have tried interesting experiments. A Tale in the Desert has no official currency. Instead, players can build mints and make their own, which game physics protect against counterfeiting. There are competing currencies, usually backed currencies similar to a real-world gold standard, and players work out prices without NPC interference.

Actually, having any monetary policy for a game could be a big step forward. I have never seen “stable central banking” as a box feature for any game.

A difficulty in shifting away from the “gold miraculously appears on monsters” model is that players will quit and disappear. If you have a finite amount of currency, and Bob cancels his account while holding 500 units of it, you have started a deflationary cycle. Do that enough times and there will not be much currency left.

Some models reduce the amount of in-game currency and have a parallel system based on real world money. Yohoho! Puzzle Pirates and Bang! Howdy used paired currencies, one magically generated in-game and another bought from the developers (and traded between players). Second Life has a similar market exchange for converting $ to L$, and EVE Online does something similar with time cards. This creates a hard currency in the game that only expands when someone is willing to pump real dollars into the in-game economy.

Are there games with a real finite pool of money? Project Entropia has some ideas along those lines but, well, it’s Project Entropia, so who knows? Horizons flirted with a vaguely ATitD-like economy at some point in its development. I sense that I am in the weeds of niche games and obscurity, but there we can find the testing grounds for ideas to be adopted by the great powers.

I have a new variant model in mind, but I will push it to its own post after I think it through a bit more. Your responses there will help me find the huge flaws in the idea.

Do we have enough experiments yet on the viability of in-game economies not involving gold that pops into and out of existence? What are your thoughts on the success or failure of gold sinks?

: Zubon

11 thoughts on “MMORPG Monetarism”

  1. I’m going to go out on a limb and say that it’s not possible to avoid spiraling inflation in an item-centric game. As long as you start with a copper sword and end up with a Uber Claymore of Supreme Hacking +L33t, you’re going to have either out of control inflation or a scaling money sink that makes little, if any, sense. This also ties into time invested. Someone who plays 80 hours a week is going to have a lot more funds than someone who only plays 20. If you try to keep the 80 hour player from inflating the economy, you end up crippling the 20 hour player (Unless, again, you have a scaling money sink the makes no sense).

    There might be ways around this, but I can’t see it being easy or fun for the players.

  2. The anti-inflation tactics really seem to move in the wrong direction in looking for a solution from inflation. Item binding is one thing. Taking items out of the system after one person’s used them means you’ve got item inflation, as well. If that’s necessary for your game to work, it likely means that too much of the character’s power is based upon their equipment. *gives a quick scowl at WoW* A better solution would be to make no repair perfect. Have the item slowly degrade either in quality or maximum durability, _or_ make degredation slow and repairs expensive enough that standard gear would just get given away/bargain binned as it got close to threadbare. Perhaps repairs of more coveted gear would require additional crafting components.

    Cash sinks are a bigger issue. They aren’t a necessary until people have been playing the end-game for a while. Let’s take a look at some possible solutions. They could easily be avoided by raising maintenance costs to better suit income (micro-cash-sinks rather than macro-), but that favours the efficient farmer over the content-exploring gamer. Another option is to make the raid/monster/dungeon drops based more on loot than on currency. Get rid of vendor trash. Have the end-game players trading upwards of 10,000x the value of a starting character’s sword for the aforementioned Uber Claymore. If you’re looking at WoW as an example, that’s a 100g:1s ratio. For the most powerful equipment in the game, if *really* scarce (maybe one drop/manufacture per 3-6 months), I could live with that kind of disparity, but a lvl 30 ‘steel sword’ should not cost 300x as much as the lvl 10 ‘iron sword’. The items are always already restricted based on level or some other limiter. Go with the AdventureQuest theory of “You can’t own anything you can’t use.” and you’d even solve the Pally rolling on the wand problem.

  3. Here’s my totally-not-thought-through idea:

    1. Make a game where all items are built by PCs. This necessitates PvP but that’s a different story for a different time. The important point for this argument’s sake is that all goods come from resources manipulated by PCs – no NPCs with an infinite number of torches for sale.
    2. Set a Real Dollar price on a natural resource – i.e., any given tree is worth $0.02. A player can chop down the tree, cut it into pieces, and craft the wood into items usable by players. The tree will not “pop” back into the world until another $0.02 is invested into the world.
    3. When a player signs up and pays his $14.95, the entire Real Dollar amount of that transaction goes towards populating the game with resources.
    4. During subsequent months, a given fraction – let’s say half – of the monthly fee is set aside to populate the world with resources.
    5. Thus, the world becomes “richer” the more people settle there and the more time people spend in the game. On the other hand, this additional burden of higher population spending more time in-game require these additional resources.
    6. The developers can, of course, “cash in” on the money invested into their virtual goods – but only by destroying them. That is, if an item that cost $4.00 in virtual raw material to make wore out or otherwise got caught in the “gold sink”, the devs could cash a $4.00 cheque.

    I know absolutely diddly-squat about real economics, but it seems to me that would stabilize the in-game economy against the Real Dollar. While inflation is certainly not unheard of in the real world, it tends to be a fair share better than in its virtual counterpart.

    Alright, that’s it – hack it apart! :)

  4. @Lachek: That idea involves a 1 month to six weeks lifespan on any particular amount of a resource, to avoid inflation, which is ridiculous. In addition, farmers will own the world and casual gamers will be completely destitute. The world would be bare of resources by the 7th day of each month.

  5. As I said, totally-not-through-through. :)


    If a given resource has a short lifespan and regenerate spontaneously, it infuses wealth into a closed system, causing inflation. How do you get around that problem without lengthening the lifespan of the resource and trigger its repopulation to some fixed real-world quantity?

    Now that I think about it, the idea of “investing” real money into a virtual world came up during an implementation discussion on the Mud-Dev mailing list for how to start an MMO centered around RMT – with bidirectional conduits between in-game and out-of-game currencies – without taking on massive amounts of liability. The quantity linked to repopulation of resources doesn’t have to be “dollars” at all if your game doesn’t center around RMT. A better quantity may be “total number of in-game hours spent by all characters this week”, since in-game activity by necessity involves harvesting, transformation and destruction of resources.

    Still, it seems to me that a game with relatively long periods between repopping of resources and having that repop directly tied to the activities of players that transform or destroy those resources would be a very effective way of dealing with inflation, and if something that’s ridiculous works and something that’s sane doesn’t, I’d go for the ridiculous myself.

    As for farmers owning the world, here’s an eye-opener – they already DO, under ANY system. In fact, if resources are abundant, repop frequently and at semi-fixed intervals – and yes, by resources I’m also referring to cash-dropping WoW mobs, not just ore veins – then the macro-farmers will be the ones holding the cash. If resources are rare, repop rarely and at unpredictable intervals, MMORPG-farms in Latin/South America and Asia who can cycle sweatshop labourers to keep them online waiting for the repop 24/7 will have the advantage. Under the former system, you can make considerable quantities of money with minimal expenditures if you’re in a popular game. Under the latter system, your overhead will be much higher, but you can still make some cash if the RMT market for the currency is strong. Under the former system, RMT demand translates directly to in-game inflation as gold farmers increase their production capacity. Under the latter system, the developers have every opportunity to limit inflation despite RMT demand.

    It’s a very complex question and I don’t have any clear-cut answers. However, I do believe that combating concentration of wealth by flooding the in-game market with gold in an effort to keep casual gamers interested is a poor design choice and exactly why we’re having this discussion. It seems to me that an obvious solution to the problem is to severely restrict the amount of resources entering the game. Will that lead to an in-game merchant class that makes a ton of cash and effectively lords over the game’s economy? You bet. What better way to make a believable fantasy sim? :)

  6. DAoC used the use/repair/degrade model for its gear. I never got high enough to find out how well that works at the high end. Realm PvP gear (wood for doors, catapults, etc) was its big money sink: high-end raiders would spend tons of gold on PvP.

    Horizons had a pretty cool crafting model, where basically _everything_ was PC crafted. You could even craft recipes IIRC. Unfortunately its combat model wasn’t different enough to make it compelling, and its XP system actively discouraged grouping, AFAICT.

    I am definitely not an expert on this, but from what I can tell real-world economies solve these sorts of problems in a number of different ways, some of which are:

    – increasing/decreasing taxes. The analog would be increasing/decreasing loot per drop and increasing/decreasing the cost of in-game money sinks. AFAICT, without banking this is the only mechanism a game has to balance the economy.

    – increasing/decreasing the amount of money loaned within the economy, by increasing or decreasing the interest rate. This is by far the main mechanism in real-world economies for manipulating the money supply. Unfortunately there’s no MMORPG analog for the way banks create and destroy money through loans.

    – increasing or decreasing lending institutions’ reserve requirement. This one doesn’t get used much in RL, because it’s such a crude mechanism compared to manipulating interest rates.

    If you coupled player-crafted items only with high barriers to entry in crafting and adventuring (gear purchase), and added lending institutions to serve as the main control of the server money supply, you would then have all the relevant RL mechanisms for monetary and fiscal policy.

  7. The main problem is that there is no cause and effect in MMOs when it comes to resources and crafting. Resources sprout out of nowhere, in infinite amounts, with never a change, no matter how much is harvested. Attempts to balance this neverending fountain of goodies, developers implement crafting systems that encourage mass creation of products that are simply wasted, vendors that buy anything and everything (and then whatever they buy is thrown into the void) and numerous gold sinks. This just doesnt work.

    What we need is an integrated system wherein everything is interconnected. If a vendor buys a sword, he should keep that sword to sell. Resources should be limited in such a way that there is no effective farming of them. Every item a crafter succesfully makes should be useful, not just a filler used to increase skill. Gold shouldnt just fall out of the pockets of each and every thing you slay.

    It’s not an easy task, and certainly not one suited to the Amusement Park type MMOs, but if you want a living, breathing world for your MMO, you’ll have to start thinking beyond the MMO concepts that are prevalent today.

Comments are closed.