For six years, I ran a consulting business that catered to gaming firms, offering advice and support the design and management of the in-game economies, markets, and monetization initiatives of large-scale, massively multiplayer online games (MMOs). As online and mobile games have grown increasingly complex — at times hosting millions of players concurrently, all of whom are at various stages of progression and many of whom are pursuing different goals — designing stable, scalable, and flexible economic systems within which players can interact has gone from an amenity to a critical feature. More than a few heavily capitalized gaming projects have floundered owing to poorly designed game economies.
The following are four major, if general, lessons about virtual economics that I took away from that experience.
1. You will never fool your player base.
Many times the introduction of new content (“patches”), “nerfing” of certain game factors or effects, or “hotfixes” will lead to hasty changes on the back end of the game code, and there is occasionally a strong temptation among game designers or managers to keep changes out of release notes and thus unknown to players.
But a sufficiently large and engaged player base will, over time, function like a massively self-organized information processor, an emergent calculator. Disparate players gather slivers of information — a new price, an unanticipated effect of a combat engagement, a noticeable change in the amount of time or energy it takes to complete a familiar task — share them with other players, and through a wholly unorganized (and often disarmingly fast) process of information aggregation, organization, and rationalization, essentially reverse engineer an aspect of the game that is, by any other means to them, unknowable. This may occur formally (through officially organized groups of players) or informally (among players on game-community message boards, in chat networks, etc.).
On several occasions in my personal experience working on massive online games, I witnessed changes made to aspects of games appear, in somewhat crude but surprisingly accurate form — usually in spreadsheets — posted to the game’s community message board not long after they were made.
This phenomenon, of course, is wholly representative of the phenomenon known as spontaneous order, albeit expressed within a digital entertainment medium. Spontaneous order, sometimes called emergence, arises when scale-free, decentralized networks organize absent hierarchies.
In The Fatal Conceit, in fact, Hayek makes a specific reference to the unique case of games as a “a clear instance of a process wherein obedience to common rules by elements pursuing different and even conflicting purposes results in overall order.”
This observation has bearing on the next lesson as well.
2. Digital space leaves room for community growth.
In game design, as in life, the risk of overparameterization is ever present. Designers generally seek to present specific in-game challenges periodically as either the storyline unfolds or the players’ characters develop, with some latitude for exploration and interaction along the way. Too often, although always intending to make for good entertainment and brisk business, developers err on the side of proscription; where opportunities (for better or worse) to induce players to interact could be fostered, rigidity — a form of paternalism — is instead adopted. In other words, players can be entrusted to address issues that code otherwise might.
I’ve worked periodically on a massive-scale mobile game in which, among other things, players were required to gather resources appearing in random spots across the broad virtual landscape. Gathering said resources, however, was perilous as it necessarily exposed the gatherer to attack by other players. Rather than adding rules and features to the game that would either protect players from attack by others while acquiring resources or make resources more widely available, an interim solution was chosen: to make resources spawn regularly, but at different rates and in different concentrations across the game map.
The results were nothing short of fascinating. Various player groups began communicating with nearby player groups to apportion territories; certain groups began entering into verbal non-aggression pacts via the in-game chat network. In other places, players orchestrated what might best be described as “virtual conservation” policies, agreeing to systematically draw upon virtual resources at a rate that both wouldn’t completely exhaust them and ensure that most players - even new, low level players - had an opportunity to access them.
Thus a virtual public good was, without any external influence, transformed by players into a virtual common pool resource by means of voluntarily arrived-at apportionment, oversight, and sanction measures.
There is a growing genre of games that seek, intentionally, to encourage ruinous conflict. But for the vast majority, with sufficient value at risk — which is to say, skin in the game — cooperation becomes vastly favorable to conflict. Less formal (which is to say, game developer) interdiction in a game tends to make players more more interactive and inventive, and sets the stage for grander intrigues as events unfold and the player population increases.
Rigid order versus abject chaos — in both life and complex games — is a false dichotomy.
3. The structure of an incentive is at least as important as the incentive itself.
Rewarding players for undertaking in-game tasks shouldn’t be, or become, a rote checking-the-box activity. Properly designed incentives — in games, as in life — tend to have several features: they are aligned with the long-term goals of the venture, organization, or undertaking; they are structured, rather than discretionary; and they are engineered to discourage perverse outcomes.
Feedback loops are a major vehicle for structuring game incentives, but can work both ways. In positive feedback loops, poorly performing players quickly get mired down, while skilled players progress too easily. Both lead to declining interest and, as that implies, ineffective monetization. Positive feedback reinforces behaviors, for better or worse. Negative feedback loops, on the other hand, make play increasingly difficult for adept players and provide a boost of sorts to struggling players. Negative feedback loops are thus more effective for promoting both long-term gaming experiences and social interactions.
4. Give them what they came for.
The concept of needs and wants is core to economics, although it has its roots in psychology. Unlike arcade games and other simple playing environments, there are essentially four reasons, all rooted in self-actualization, for people to seek to play massive-multiplayer online games: competition, among those who want to win or achieve; adventure or curiosity, among those who want to explore; social experience; and meddlesomeness (variously known by other names), among those who simply want to interfere with the previous three types of player.
Creating a game experience — and in particular, a game economy — that allows the fulfillment of all of the above desires, without persistently, structurally subordinating the fulfillment of one particular group to the other three, all within a compelling playing environment run and managed by a profitable business enterprise, is the stuff that legendary MMOs are made of.
And as a “bonus”,
Economics applies not just to flesh-and-blood communities, but communities of digital avatars, their actions and interactions. Designing a game economy is more art than science, and testifies to the Hayekian notion that economics demonstrates how little we know about what we imagine we can design.