The Design of Online Economies

Part 3: Items

Phat l00t.

That is what many online gamers think of when they're about to embark on their next "epic" trip to kill some hapless monsters. The need to acquire new and more powerful stuff is a strong urge with many players, and for this reason a lot of online economies place a very large emphasis on items. Economies, in fact, are basically all about "stuff" -- acquiring, creating, trading, selling and buying "stuff".

In this part of my series on online economies, I'm going to talk about items and their role in an economy.

What Is An Item?

For the sake of pedantry, I'll define an item. An economic item is any in-game object that can be owned or manipulated by a player. Clothes, food, weapons, iron ore, books, and armor are items. Monsters, unless they are owned and controlled by a player (pets), are not items. Pieces of the scenery -- ladders, doors, boulders -- that cannot be manipulated or owned are not considered items for this discussion.

Items and Currency

As I discussed in Part 1: Currency and Part 2: NPC Merchants, currency in most online economies is just a proxy form of items. If an item can be converted to currency, and if currency can be used to reacquire that item, then currency is a generic representation for an item. This is obvious -- the whole point of currency is to remove the need for a direct item-for-item exchange (barter) and replace it with something more generic -- yet many designers still make the common mistake of treating items and currency differently, especially when it comes to economic balance.

They're the same thing. There is no functional difference short of arbitrary ones enforced by designers.

Item Sources

Items can come into the game world using different types of fiction, limited only by the imagination of the designers. Mechanically, items appear using one of only two core methods: spontaneous instantiation ("spawning") or exchange.

Spontaneous instantiation is when an item just shows up (another item is not removed in the process). Common examples of spontaneous instantiation include:

  • newbie items (created with your character)
  • quest tokens/gifts (created by an NPC quest giver in response to some phrase or trigger)
  • spawning (item just appears somewhere, including loot drops off monsters)
  • harvesting (item appears after successful skill use, e.g. lumber, ore, or foraging)
  • conjuration (casting a spell creates an item)

Items can also be created by exchange:

  • combining multiple items to create a new item (crafting)
  • quest token exchange (give quest NPC some items, receive others in return)
  • purchase (currency is given to an NPC merchant and an item is received)

Of these, spontaneous instantiation is the most common and influential to an economy. If items show up continuously and arbitrarily, the world becomes flooded, creating item inflation.

Item Inflation

Item inflation is the bane of all persistent worlds -- it is partly what makes old servers feel, well, "old". When a new world/shard is launched, most PCs will be newbies, and everyone gets to struggle along the treadmill together. On day one, it is highly unlikely that someone will acquire the Magical Power Sphere from the Grim Dragon Groblo.

But six months later? Yeah, well, everyone has the Magical Power Sphere, in fact, it's pretty old news, and now all the cool kids want the Mystical Cube of Pain. Hell, the Magical Power Sphere is so common that even newbies just drop them.

But why does this happen? This is an almost inescapable part of wealth advancement, even with real human societies -- as a society progresses and becomes more technologically advanced (analogous to finding more loot in an adventure world), the entire population benefits through "trickle down" dynamics (although the trailblazers will almost always still enjoy the brunt of the benefits).

This is, in fact, a natural by-product of an open economy (where the introduction of items is not gated by the exit of existing items). It's not only not broken, it's absolutely expected.

Psychological Impact of Trickle Down Dynamics

The wealth advancement of an online world imposes significant problems, both mechanically (balance is now out of whack) and psychologically (player response to a changing world). From the point of view other players, it is considered grossly "unfair" when others benefit from the generosity of in-game friends or, even worse, different characters used by the same player. Many players consider online worlds competitive and desire a level playing field, so when they see someone below their own level zipping about in high level equipment that wasn't "earned", they suffer from the equivalent reaction to a blue-collar worker watching some college freshman driving around in a new a BMW that Daddy bought them. "It's just not fair!"

Justified or not, this engenders resentment towards those players ("twinks") and the game itself. It's not necessarily a devastating effect, but I have seen players quit a game because they weren't part of the "in" crowd and didn't want to deal with the same social annoyances online that they encounter in the real world.

The societal and psychological impact of item inflation and the wealth advancement curve can be insidious, due in large part to its completely intangible nature. Mechanical problems can be quantified, mined, and analyzed, but intangible problems such as player sentiment are hard to isolate and identify.

Mechanical Impact (Balance) of Trickle Down Dynamics

In theory, designers sit down and work out the perfect types of encounters for different players based on their expected level and equipment. If the level six orcs are a challenge when fighting players armed with items of power X, six months later, the orcs will be slaughtered by facing players armed with items of power X + C (the trickle down effect in action).

This is a serious problem, because the game now no longer presents the intended challenge for some players. For other players, it's still just as challenging because they lack high level benefactors, or they just refuse to be "twinked" on general principle. Most players fall somewhere in between -- they may receive one or two powerful items through trade, luck, or generosity, but by and large they're still using the gear expected for their level.

This is a difficult problem to fix. If the designers just universally jack up the stats on all the monsters then all players are impacted, even non-twinks. A dynamic difficulty adjustment system is too cumbersome and error prone to be practical. Level limits on items can be imposed, but this has its own set of drawbacks.

Level Limits

Level limits are artificial. They completely break the immersion for a lot of players, because for some unknown, strange reason, this sword "doesn't work". You can't wield it, it just stops working or is so degraded in power that it's useless. Okay, that sort of fixes the power problem, but now you have this awesome item the game is telling you is off-limits...for your own good.

And that irritates players. They have the item, they acquired it somehow, it's in their inventory, they can look at it and touch it and smell it, but for some reason they can't use it. Not only that, but the item might have been acquired "fairly" -- as payment from a higher level player for an important sevice, or in exchange for many lower power items, or just through saving up a lot of gold to buy this one thing. The player feels cheated, because as far as he's concerned, he earned that item, and still cannot use it. It's like getting a Christmas bonus at work you can't spend until you've retired.

Not only that, but it's not like a player will toss an item he can't use right this second. It stays in a safe place until the very moment it can be used. Or he trades it to someone for an item he can use or even just gold, which he uses to buy something else. So level limits don't even really pull the item out of the economy or reduce power definitely, they just make it less convenient for the player.

Level limits just don't work. They're a bandage, at best, addressing the problem tangentially, pissing off players and ruining many key characteristics of a vibrant online economy. They're a simplistic number that attempts to enforce disparate ideas of "fairness", but they don't address the root cause of all this: the item should have left the economy a while ago.

Stopping item proliferation is a difficult task, achieved by either removing items or simply preventing their creation.

Removing items, via item sinks, solves the problem directly. Don't want junior to get that Sword of God Killing? Well, then, just delete it! That's easy. But doing it in a way that doesn't aggravate players is a bit more involved.

Item Decay

The first suggestion is usually something like wear-and-tear. Every time an item is used, its durability drops a bit -- or hell, just set a timer on the item, and when that timer expires, poof!, the item disappears. While a rather brute force approach, it can be effective at removing items, but the reaction to this is predictable and bad -- players end up stockpiling items that they deplete, which leads directly to contention for resources.

What if, instead of just letting items die, we allow them to be repaired for some fee? This has the benefit of introducing a money sink as well, but, again, there's a downside. For starters, if PCs are performing the fixing, it's not necessarily a money sink -- money just moves around, but it doesn't disappear. For money to disappear it has to get back into the system, usually via an NPC. But if NPCs are performing the fixing, or if NPCs are the sole supplier for components necessary for repair, then you annoy PC crafters that feel that item repair should be strictly in their domain.

That said, item repair is a viable method for removing items, because it not only destroys items directly (through wear), but it also removes them indirectly as players sell back loot in order to make money to pay for repairs.

If an item's initial "wear factor" factor is set at a point just slightly large enough to get through a particular power band, it should naturally wear out about the same time the player "ougrows" it -- limiting its lifespan and value when handed down to someone else. If the repair cost is high enough an item's new (low level) owner won't be able to afford its repair -- sort of like inheriting a 1983 Porsche 911 from your grandfather, but you find out that routine maintenance is so expensive that you'd probably never drive it anyway.

Aside from standard decay or wear and tear, you can have "garbage collection" where NPCs or the system at large go through and start removing unclaimed items from the land (litter). This is an effective item drain, unfortunately it is only removing items that someone has decided has almost no economic utility, so instead of making the economy tighter, it's literally just cleaning up messes.

Item Exchange

Alternatively you can get rid of items by forcing their destruction to create new items or perform services (such as reagents consumed by spell casting). The most obvious example of this is crafting -- four ingots of steel are consumed to make a steel helmet. A steel helmet, two gems, and a magic potion make a Mystical Helm, etc. This doesn't do much to get rid of high powered items, but it's heading in the right direction. Sometimes the exchange will fail (botched crafting attempt) at which point an item simply leaves the system.

If instead of just finding looted items all the time we make the best items quest rewards and/or crafted items, then we have a handy recycling system built in. If the best sword for your particular character and level is available only by cashing in X number of other items, then we get X:1 compression.

This can work reasonably well, however it involves a lot of planning and work on the part of the designers. It also requires an extreme amount of fictionalization, since quests effectively replace drops for the most powerful items. And, of course, there will be the rampant problems that result when players start trying to get higher level items for their lower level characters.

Wealth Expansion and Cap

Finally, there's the problem of a "wealth cap". At some point a player may have every conceivable item he could possibly obtain or want -- when that happens, all his wealth starts "spilling" into the economy because he has no need for it. Luckily enough for game designers, as long as there is anything better to acquire, players will go to great lengths and cost to acquire it even if the additional utility or value of a better item is marginal. We see this in real life all the time -- people that spend 50% more on a house because it's "in the right zip code" or they spend double for a car because it is marginally faster.

However, even when there is something more to be purchased, there is a balance problem due to disproportionate perception of wealth. If you're saving up to buy a $250,000 car, then spending $500 on a jacket doesn't seem like much. Hell, you'd spend that much on a gift for a friend's birthday without thinking about it. This happens in online economies as well -- it's just human nature. Powerful and wealthy players won't think twice about "twinking" a friend if the amount of wealth they're donating is an insignificant fraction of their total net worth -- what the wealthy player makes in a day of adventuring is likely more than his low level friend would make in a week. This can, for obvious reasons, throw off a lot of a designer's assumptions about new character wealth.

Item Farming

One common activity in online worlds is "item farming", where players stay around one area or a specific static spawn and "farm" it for a particular item, over and over. This is a natural, but annoying, side effect of a healthy player economy -- the item has value, therefore players want as many of that item as they can get their gauntlets on.

Preventing Item Creation: Closed Economies

If removing items from circulation is such a chore, then why not just prevent their creation by implementing a closed system (i.e. economies that have a fixed number of resources such that no new items are introduced until old items are recycled, destroyed or otherwise removed)?

A closed economy is conceptually simple. The designer decides that for his population that there will be one hundred swords in the world, period. This is enforced scarcity, providing inherent value. With, say, 150 characters, there are always people that want a sword, so those without swords need to find something of equal value to trade.

Unfortunately, this doesn't work. The enforced scarcity leads to hoarding, and hoarding in turn leads to reduced recycling, which eventually freezes the introduction of new items. Because Ed the pack-rat refuses to part with any goods he acquires (because he might never see those goods again) the entire economy is now frozen. You can encourage Ed to put items back into the system (sell them, destroy them, whatever), but that requires his cooperation. Or you can use item decay to force items back into circulation, but then you inherit all of its problems as well.

Closed economies don't work, in theory or practice.

NODROP

One notable hack to get players to play the way designers intended is to flag some items as NODROP -- they can not be dropped or traded, only destroyed or otherwise removed from circulation. Like level limits, NODROP is another design sledgehammer, intended to address a wide variety of perceived problems...but without actually fixing any of them.

Example 1: certain items are "too powerful" for low-level characters, and so designers worry that the item is getting into the hands of the wrong players. These items are flagged NODROP, preventing their distribution to other characters who have not "earned" the item.

Example 2: certain items are used as quest tokens, i.e. they are "evidence" that a specific quest was completed (this is a cheap, more general form of quest management than hardcoding flags into a character's core data structure). However abuse becomes rampant when the quest tokens can be purchased or acquired through means unintended by the designers.

Example 3: some items are valuable enough that it encourages players to "harvest" the item for later trade or sale.

NODROP introduces two substantial problems, quite possibly even of greater magnitude than the problems being "solved". First, it significantly impacts player trade, which hurts both the economy and player immersion. If Clara the Cleric has a Brilliant Scimitar of Light and Frank has a Club of Seal Bashing and they want to do an even trade, NODROP would render this impossible if either weapon was tagged as such. Second, it inadvertently encourages competition for key items, leading to overcrowding in areas where that item can be acquired. Clara the Cleric can't even use the Brilliant Scimitar of Light she acquired during a normal day of adventuring, and worse yet she can't trade it to someone else for an item she can use -- so now Frank has to go get it for himself. If they had simply traded with each other the two players would have been happy and many other players, busy killing things in search of the Brilliant Scimitar of Light, wouldn't be irritated when Frank shows up looking for it as well.

LORE Items

In Everquest parlance, a LORE item is something that a character may only possess a single instance of (the item is supposed to have some kind of lore associated with it, i.e. it's supposed to be unique due to some fiction). This prevents item harvesters from stocking up on the same item over and over in an effort to trade them (see Example #3 earlier), but at the same time still allows player trade. So, in effect, LORE items are a gentler attempt at solving some of the same problems as NODROP.

Pragmatically, LORE items don't have too many downsides, other than player inconvenience. For example, if you can wield two weapons at once, you won't be allowed to wield two of the same LORE item, you'll have to have a different secondary weapon. Slightly more aggravating is that you can't even temporarily hold two of the same item, e.g. if you have an Orb of Power and someone wants you to give their Orb of Power to someone else, you can't act as delivery boy since you can't hold both. Not the end of the world, but annoying. It's not like grabbing an extra burger at the drive-thru.

Player/Item Tags

One solution is to control the ability to own or loot an item as a function of the character's history. Each time a character picks up a lore item, a flag is set in the character's record, preventing him from looting that item ever again. This seems reasonable at first, but in practice ends up becoming cumbersome due to many reasons:

  • what if the player legitimately lost the item and is seeking to replace it?
  • what if the player sold the item and now wants to reacquire it?
  • how should the loot system handle drop determination when there's a party of adventurers, not just a single character?

There are even more corner cases, and none of the ostensible solutions are perfect. The larger problem of item farming is sort of fixed, but player convenience plummets, and hoarding will often end up becoming the norm as players become worried they'll never be able to own the item again.

Tension Between Trade and Item Inflation

There is a fundamental synergy between player trade and item proliferation. It is impossible to discourage one without discouraging the other as they operate hand in hand. If an item is worth having the other players will want it -- which means that some players will find it very advantageous to introduce as many of those items into the economy as possible. Level limits, NODROP, item upgrades, and LORE all discourage item farming and twinking, but at the exact same time discourage or prevent player trade.

I'm unaware of any solution that allows free player trade without inadvertently pushing item proliferation and farming.

Crafters vs The Environment

When similar items enter the economy through the actions of PCs (crafting and trade skills) or arbitrarily (spawning) then competition between craftsmen and "the environment" develops. This can have broad ramifications on the entire economy and player enjoyment, and will be covered in more depth in a later article. For now I just want to point out the obvious: if worthwhile items can be acquired "for free" from the environment, PC crafters will find themselves marginalized.

Summary

Items and, by proxy, currency are the fundamental units of measure for an economy. Players need and use a wide range of items, and controlling the influx and outflow of items is a tricky task. Economies can quickly reach a point of autocatalyzation at one extreme, or grind to a halt at the other, depending on how well the economy's "faucets" and "drains" are managed. Even small changes to items and trade mechanics can have drastic ripple effects on an entire world.

Discuss this article in our forums!