Community is something of a mystical beast hunted by game studios around the globe. In this article in particular I will be focusing on MMO game communities, as there are probably few other places where community is as important as in online games. And with good reasons.

Being part of a community means that players identify with other players supporting the game, as well as the game itself. A strong community will often mean that players will promote the game, they will try to get others to play it and keep them playing - all of their own free will. They will create guides, articles, art, cosplay, podcasts, streams, player run competitions, radio stations, mods, addons and provide support to fellow players. All that without pay, at least in traditional way.

If the community is good it will make new players feel welcome, make early learning easier and make finding groups a pleasant experience. But let’s be fair, not all communities are like that. There is a reason for the rise of term “toxic community”. And while toxic community is better than no community, a nice community is better than toxic one. The factors which influence what type of community is formed can be identified, interpreted and explained using economics.

That is right, I said economics.

Economics gives us enough tools to understand and affect average individual and group decision making without trying to understand every individual. By looking at averages and groups we can start building a fuller picture about the popular decisions players make. This fuller picture then helps to identify player motivations and how they are interacting with each other. Supplementing it with other social and medical sciences, like psychology or neurobiology, will then allow for a wider array of tools and less categorical thinking. So I am not advocating disregarding every other discipline, just underlining the ease of use of ready developed economic theory when looking at community formation. One of my favourite quotes is by Adam Smith, who had great influence on classical economics, he wrote it in 1776.

“Wealth of nations was based not on gold but on trade: That when two parties freely agree to exchange things of value, because both see a profit in the exchange, total wealth increases”.

(Image source: Adam Smith Muir portrait)

While this insight was originally about macroeconomics, it scales well. We can compare nations to communities, and the trade is interaction between players. In the context of game communities, wealth is less about currency than about the value for players. The easiest way to think about this is to call it “fun”. Developers want to make fun games, and players want to have fun. This is the true wealth of any title. So, to paraphrase Adam Smith:

“The Fun level of a community was not based on the game itself but on the interactions between players: That when two players freely agree to interact, because both see a benefit in the exchange, the overall Fun increases”

But so what? What is the point? The point is that economic theory is relevant even though it was developed long before MMO games. And trust me, there has been plenty of development of the science since 1776, not to mention that it was happening long before Adam Smith as well. There are hundreds of years of research ready to use.

Before we go further down this rabbit hole, let us backtrack a little to why a nice community is better than toxic one. After all a toxic community is still a community and gives a lot of the benefits of a community. I suspect that most would be intuitively able to answer this, but let’s explore it nonetheless. The two games with nicest communities I have personally played are the now shut down City of Heroes (CoH) by Cryptic Studios and later Paragon Studios; and the still live The Secret World (TSW) by Funcom. If you google one of these titles and toxic community there will be little to no relevant hits. On the other hand searching for League of Legends (LoL), by Riot Games, and toxic community will lead to a lot of relevant hits. The games are obviously very different and to be fair Riot Games has been working a lot in the past few years to improve the community. Equally obvious is that it does not define success - LoL is massively successful, while TSW and CoH are niche games. But looking specifically at the communities, what does toxicity do? It divides.

You are either in or you are out. And if you are not in, the community discourages you from even approaching. It creates continuous bad publicity - as I said a simple search will yield plenty of references. Even within the community there is continuous discord and strife between players. Looking back to our Adam Smith paraphrased quote, this means that often the mutually beneficial interaction does not happen, so no fun. I could draw an analogy to trade versus robbery. If the community in LoL was nice from the get go it would have arguably have lead to a bigger success than what it is. When that is said, without a community at all we probably would have never heard of LoL.

In CoH, the community was regarded as one of the strongest elements of the game. Indeed, the community spirit is so strong that it has persisted even after the close of the game. There are now at least 3 titles in development aiming to be the spiritual successor to CoH. All these titles were started by old CoH players, and all are hoping they will be able to recreate the feeling of community from CoH. These projects are City of Titans by Missing Worlds Media, Valiance by Silverhelm Studios and Heroes and Villains by Plan Z Virtual Studio.

So why can nice community create such a powerful effect which will compel people to play long after the replayability value of content is gone? Or even go as far as recreating a game after original was sunset? And why should a developer care when they can keep players playing long enough on now much better explored addiction mechanics?

(Image source: Extra Credits: Community Management)

First off a developer can have both. We aren’t talking heroin kind of addiction, but having gameplay which compels players to keep returning. Then we can amplify that by having a community which also encourages players to keep returning. Seeing how many people bear the title community manager, the industry understands this. However having managers for their community does not guarantee greatness. It is possible to have a great community with no community managers, although people often appreciate there being a face (or name) which they can address. Just so that I do not downplay importance of community managers too much, have a look at the Extra Credits video above, it clearly explains the benefits of having great community managers.

But what is then the basis for a nice community? To answer that we can start by looking at our brains. Starting with neurobiology, a research team from Departments of Psychology in University of Pittsburgh and University of California found that giving support has same brain activity as receiving support, just more consistently. This is one example of how we have neural pathways which provide a neural reward for behaviours like gifting, sharing, helping others and so on. The same neural pathways and transmitters are involved as in motivation; attachment and bonding; good food; and chocolate. So our brains already condition us to act in a communal fashion. If you’d like to know more, then there are more links in the further reading section - some of it can be pretty heavy, but it’s a fascinating subject.

So when players experience an environment where being nice is also economically more logical than not, there is a strong foundation for creating a nice community. And because it is much more rewarding to us as humans to partake in generosity, we feel stronger attachment to the community. We are then more motivated to repeat such activities, and overall then experience an increase in our emotional wellbeing. This makes a nice community physically better for us than a toxic one. It also makes it a community we want to be part of even if the game has lost appeal as we become more conditioned towards the neurobiological reward. This means that good communities grow strong and have a tendency to last.  

(Image source: Agents of SHIELD 2014)

Another mechanism we have hardwired is social categorization - our Friend or Foe recognition system. This is very oversimplified I know, but it should be enough for the purpose of this article. It is natural state for us to want to avoid hostiles or act aggressively towards them. But isn't that the point in a pvp game? So what is the problem? The problem is that it is not just black and white, categorization is not just one dimensional. In real life we are fully capable of being the bitter enemy of our best friends on a football field. This is achieved by how we categorize others in many many different categories. In a good community, more players will be flagged as friendly in many categories creating an environment again we want to be part of. Same as we like to hang with friends. In contrast in toxic communities far fewer will be flagged as friendly in many categories. So many more players will be flagged as red to you in many situations and as many players know “Red is Dead”.

So if nice community is better, more motivating, provide more additional fun on macro and individual level, should not all communities naturally develop to be nice? This will be based what rules of interaction emerge, these rules are best described through economic game theory. These are social games we play with each other without thinking, where every player wants to optimize their utility. Utility is the sum of all gain or loss, material and emotional. As the game is played we might see that there are multiple solutions which are equally valid. While all of these are valid strategies for players to pursue on individual level, only a subset of them will create a solution which is optimal for all players. Solutions where it is impossible to make one player better off without decreasing utility of another are called “Pareto optimal”, and in our case leads to a nice community. Game theory by itself is just a way to mathematically model decisions and those do not happen in a vacuum. When used in context of economics, political science, evolution or philosophy it becomes a powerful tool to study behaviors. It can even be used to model decision making in AI.

If we want to achieve our Pareto optimal nice community, we can either rely on chance, or we can make sure that we improve probability as much as possible. This means that game developers need to think how their systems and rewards interact, so that in the end the optimal path encourages niceness. As we have already established that doing nice things is emotionally rewarding, that part of utility function we can leave alone at least for now. What needs attention however is how progression is structured - if a player’s ability to get ahead by being toxic outweighs the benefits of being nice, then we will get toxic community.

(Image source: Spy vs Spy)

Let’s run a simple thought example about PVP. If the sole way to achieve progress is your own kill score, then every other player in the game is competition - even your own team. As a player you know that for as long as opposing team does not fully coordinate, you will be able to get better score by simply disregarding your team and raking in the kills. It is then in your interest to demoralize and prevent coordination between players in general to get maximum utility. Because you can rationally assume that others will come to same conclusion and will try to do same to you, it is better get them before they get you. Even if they act irrationally and do want to cooperate, the progression benefits will outweigh the emotional benefits of being nice. This is of course just an oversimplification, and even in this scenario just one team coordinating can force others to coordinate or be denied progression altogether. But it will not change community to nice if it has already settled on toxic. At least not that alone.

If progression is reliant on others from start however, being toxic will not give bigger utility, but will instead be punished. That still does not mean you have to be nice, just that you must not be toxic. Optimally being nice should have some long lasting benefit in progression. But we know that it has emotional benefit, so it might just be enough to show players how to get that benefit. This starts the next challenge, teaching players the rules for optimal play. In real life we spend our childhood learning and practicing playing social games, also known as social rules or norms of our society. So that when adults are interacting we can often predict how others will act as we expect them to play optimally. I am sure everyone can think of an example of what happens when someone does not play by those rules. In MMOs however, we do not have a luxury of teaching players proper behaviour over several years - it has to be done fast. For example a player comes to your community from a similar game where the community is toxic. It is likely that this player will assume the same rules of the game and create discord in your community. So early teaching becomes essential, with lots of opportunities to practice.

An example is from CoH before the introduction of their Auction House (AH). The game rewarded players a lot of “inf” (Influence/Information/Infamy was abstracted currency of the game) as they progressed. But since it had a limited use in game, lots of long-term players had massive amounts stockpiled. This made it very easy for them to give inf away to newer players who really needed it in early progression. This process facilitated early teaching of “be nice” rules of the community. It is interesting to note that this started to collapse after the AH’s introduction, and we can conclude that finding a way to make sharing a low cost and high reward is best.

In our PVP thought example, I was only talking about players who had same goals, but in a lot of MMOs there will be different player groups with conflicting goals. Be that simple PVE focused players not caring for changes desired by PVP focused ones, or crafters disliking the ability of those who farm to get same or better items. So the social games become complex quickly. However, as long as taking toxic path remains economically unsound the foundation for a nice community remains stable. If this foundation is going to remain that way, it must not rely on goodwill of few individuals, central control or policing. The need for keeping the players in check is called cost of anarchy, if the cost is high the whole social game will likely collapse when control wobbles. On the other hand if cost of anarchy is really low the social game and community it is played in will be very robust. This why I said that a great community does not really need community managers.

But let us return to teaching. Both teaching people the rules of social games and just providing relevant information ties us in with information economics. This branch of economics studies the influence of information on economic decisions and the economy. In our game setting it is important due to the economic context it provides why someone is for example a troll, or why they go to the forums to claim anyone who is playing AH is out to steal their hard earned in-game currency (IGC) of choice. And as stated it links naturally to learning rules of the game from game theory. There are a few terms which have especial relevance to community building: Moral Hazard, Information Asymmetry and Adverse Selection.

(Image source: Moral Hazard by WilliamBanzai7)

Moral Hazard - have to love how sinister it sounds. This term is the primary explanation of toxic behavior by people who otherwise are nice in their daily lives. Moral Hazard is often explained with examples from insurance, where a person once insured acts more recklessly than before because consequence of person’s actions no longer has same effect. So we get perfectly adequate people misbehaving online because repercussions from their actions have such limited effect. Moral Hazard’s achilles's heel is the ability of the other party in an interaction to obtain relevant information and punish transgression.

When one player has more information than the other, it is called Information Asymmetry and in MMOs we can do a lot to avoid it through good UI design. One of the biggest problem information asymmetry poses to the community building is degradation of trust. This is effectively what has happened if player says that they do not want to participate in player trade, because there is no way to be sure if player will get cheated or not. As even suspicion that in a game other player has more information that you and conflicting interest, can change the game to yield negative utility.

We then end up in Adverse Selection situation. And while moral hazard is behaviour which is changed as reaction to change in circumstance, adverse selection is exploitation of information advantage at expense of someone. Over time leading to colapse of trade, as players will expect others to cheat. Possibility to cheat in any situation will lead to adverse selection if not stoped. Just to round up importance of information economics and how disregarding lessons from it can hurt community, there is a great paper by George Akerlof from 1970, called “The Market for Lemons” which goes at length to explain why information asymmetry is bad. I have linked an article from The Economist in further reading discussing it. The cure? Good and freely available information so that there is little to no advantage.

(Image source: Fallout 3 G.O.A.T. Test by Bethesda)

I used player trade as example in above paragraph because player trade is important. It is a social interaction, and still some games remove it - even when it is counterproductive to community building. Creating more ways to players to interact with each other should be the focus, not reducing them. Good trade specifically has a lot in common with sharing, and the biological benefits of it. As example let us think of a situation of a simple barter, two siblings are going out on the town, one of them has a nice new necklace and the other a new watch. However both watch and necklace would look better with other sibling’s outfit, so they trade for indefinite time. As long that both see trade as fair and that one has no incentive or ability to trick the other, both will gain from the trade. The principles of described trade are same as the ones described by Adam Smith earlier in the article. They are also the ones which information economics says are beneficial. We will see them coming back again as we look at perfect market competition. But most importantly, all of us have known this since we were playing in the sandbox as toddlers. Everyone likes a good trade, even if they do not know it yet.

And what science is best suited to create good trade in a game? Economics is probably on the short list for that answer. We have for example already established that economics can easily explain how great and relevant information goes a long way to help trade to occur and prosper, and establish trust in a community. It provides excellent tools to analyze trade in the game. It has great explanatory power to understand what causes the patterns seen in data. Knowing the causes allows us to make changes to remedy any problems. Utilizing economics early in design will make it possible to have an economy in game which is a natural part of it and helps players to make more positive interactions with the community.

Most of us have, at the least, heard about the invisible hand of marcket forces, sorting out the balance of supply and demand, and perhaps even about perfect market competition. Those are important theoretical concepts from classical economics, but are often used in odd ways in games. It is as if hearing about these concepts people think “Free competition is good because it means players will sort it out themselves”. However this omits some important points we have already highlighted in this article. Perfect market competition means that everyone sells identical product, sellers cannot control market price, there are a lot of sellers, both buyers and sellers have complete information about product and prices on the market, and there are no barriers to entry or exit on the market. When those conditions are met, the price, demand and supply will change almost instantly in reaction to one and other. In an MMO it should be easy to check next to all of those conditions for perfect market competition. But then things go wrong.

When we wish to build a nice community, perfect market competition for items traded between players is an extremely desirable state. It creates the situation where markets are easy to understand, it is easy to predict changes in demand and supply, and players can trust that they get a fair deal. It is a situation hitting all the right buttons to help establish a nice community. So why are there so many games out there where we see information intentionally removed or made difficult to read? Such as useless items new players need to get their heads around before understanding which ones are of any use, or complete lack of historical information for prices, to name two of the most common ones.

(Image source: THE ULTIMATE COMBO by David Soames)

This might be a desire to reward players who spend their time trading. But instead of handicapping others a better option is to lean on the good old mantra of “easy to learn, but hard to master”. Provide advanced tools for those who really want to play with market ways to make a name for themselves, like bundling or quantity discounts as example. All the while making it as easy as possible for a new player to understand what is a fair price and why, where to go to trade and where to get items worth trading. Making it easy for players to understand how the economy is hanging together makes it also easier to setup interdependence between players.

Earlier in the article I was talking about game theory and mentioned conflicting interests. These can sometimes be just playstyles of different people - some like PVP, some raid, others RP and so on. In our quest of creating a nice community, we want to prevent conflicting interests from developing toxicity, and one way of doing it is interdependence. In economics interdependence is described as division of labour, central to growth experienced during the industrial revolution. Obviously we do not want to turn all players into unskilled factory workers, but rather specialists. When different content in the game provides different resources of value, a player can choose to do all content or specialize and trade the surplus resource the player gets for others which she or he lack.If this is to help nice community to emerge, trust is essential, so information and teaching tools are vital. This has the added benefit of incentivising the use of different content as price for resources from underused content will rise and attract players. The invisible hand will point the way.

Does this mean that it is all about classical economics and free markets? Of course no. When things get out hand, be that because of a runaway bug or simply some unforeseen emergent gameplay, we can rely on Keynesian economics to get us through some short term crisis and make sure our community does not receive any long term harm. Then there is all of the part of the economy outside of player control, for example production facilities for crafting. Price and availability of those fall right into realm of Socialist economics. How those elements of the economy outside of player control are set up and function will create the framework for the player to player interactions I have been mostly focusing on in this article. So "comrades" remember that the state controls everything in the game world! By that I mean everything from item properties and drop rates, to PVP combat rules, are controlled by the developer, all of it affects what players seek when interacting with each other. Amongst it all the parts open to player influence and control seem tiny. So while this might be less about free markets then we would like to admit, the bright side is that creation of that nice community is much more under our control than we perhaps are aware of.

(Image source: Meercats by LInda Silvestri)

In the end if it still sounds odd that we are talking about community, being nice and economics, it is then important to remember that value and money is not one and same. In his 2002 article  “An economic analysis of altruism: who benefits from altruistic acts?”, Klaus Jaffe describes a simulation where agents live in a simple agricultural society. His findings show that when looking at the growth indicators which we are used to, such as GDP, greed is good and altruism has no benefit. This is what most people expect, so all is as should be in the world. However as soon as he allows for “non economic benefits” from altruism, in this case neighbour’s willingness to save you from life threatening circumstances, the society as the whole had higher wealth growth. I have included link to the article in further reading, it is not very long but very interesting read. Economics has always been about value and growth, sometimes we are limited by what we can measure in money, but the science itself was never about money. So it is not at odds with being nice, or all those soft values which we intuitively know to be important. It is all about how value is increased for our communities, be that GDP for a nation or fun for a MMO community.

To summarise we have barely scratched the surface of how economics is linked to ensuring the growth of communities. It can be used to guide the direction that a community develops in, with emphasis on the development of a nice community. The benefits of nice communities are many, and we humans have some predispositions to be nice on biological level. But to tap into this natural inclination, we must understand the social games being played. Game theory is then the tool to model the social games and economic theory to understand what outcomes we can expect and how we can change them. Labels like Classic Economics or Socialist Economics are schools of thought or umbrella terms for different theories, like division of labour or free market forces. In the end it is all economics and all about interaction of rational people, as even governments are still run by those. Economics does not replace other social sciences, with them it will just become even more powerful.

Further reading:
I have tried to categorize the reading to hopefully make it easier to navigate. Wikipedia articles will often have a lot further references which can be followed. The relative to others abundance of links in neurobiology is both because I find it fascinating and to have multiple sources. As this is a field I found to have a lot of resources making bold claims, but with no research referenced in articles.

Neurobiology
The Neurobiological Basis of Thanksgiving, or Neuroscience of Giving and Receiving
The Neurobiology of Giving Versus Receiving Support: The Role of Stress-Related and Social Reward–Related Neural Activity
The Neuroscience of Giving
The Best Present: The Neurobiology of Giving
The two faces of oxytocin
What is dopamine for, anyway? Love, lust, pleasure, addiction?
Neurotransmitters—an Introduction
Beyond the Reward Pathway
The unsexy truth about dopamine
The Mysterious Motivational Functions of Mesolimbic Dopamine
"Neurobiology of Aversive States" Erin N. Umberg1 and Emmanuel N. Pothos (2011) - (author manuscript)
"The neurobiology of human category learning" F. Gregory Ashby and Shawn W. Ell TRENDS in Cognitive Sciences Vol.5 No.5 May 2001

Game Theory
Stanford Non-Zero-Sum games 
Beyondintractability Positive-, negative-, zero-sum Ross, Don, "Game Theory", The Stanford Encyclopedia of Philosophy (Spring 2016 Edition), Edward N. Zalta (ed.)
Stirling WC, Felin T (2013) Game Theory, Conditional Preferences, and Social Influence. PLoS ONE 8(2): e56751. doi:10.1371/journal.pone.0056751

Classical Economics
Wikipedia - Classical Economics
Investorperdia - Perfect market competition
Ivestorpedia - Free market

Information Economics
Investorpedia - asymmetric information
The Economist - Secrets and agents
Boundless. “Asymmetric Information: Adverse Selection and Moral Hazard.” Boundless Economics. Boundless, 08 Aug. 2016.
Wikipedia - Information Economics

Altruism
Okasha, Samir, "Biological Altruism", The Stanford Encyclopedia of Philosophy (Fall 2013 Edition), Edward N. Zalta (ed.)
Wikipedia general on altruism
Klaus Jaffe - An economic analysis of altruism: who benefits from altruistic acts?

Other
Division of labour (Wikipedia) - many economists have studied this subject at different points in time, so it does not really belong in a any specific economics school of thought. Although different schools will often have differing political views on the subject.
Gamasutra, Ian Bogost : Persuasive Games: Familiarity, Habituation, and Catchiness - An interesting article seeking to look at what actually makes the “Easy to learn, but hard to master” mantra work.
Wikipedia - Keynesian Economics
Wikipedia - Socialist Economics
Wikipedia - Behavioral Economics