Virtual Economy

Hi everyone,

I’m in the process of writing up a mod with the purpose of establishing a virtual economy in a game. I’ve been thinking about how I can do all of this and I have a number of ideas. I have written two articles on this problem which I will post below here, but I must first give you the background information since these articles assume you are a player of the game and know what is already there.

This game is called “The Specialists”. A few years ago, a number of people wrote mods for it to establish a “roleplay” environment, whereby you take control of a character and play as if you are that person. The game mode that I am working on is thus called “TSRP” (The Specialists RolePlay). Anyway, none of these mods gave any real thought to the economy. Right now, there’s a “salary” indicator which gives you a certain amount of money every 60 seconds with which you can go up to “non-playable characters” to buy items. You don’t have to do anything to get this money. There is an infinite supply of these items and basically an infinite supply of money.

TSRP is flooded with problems such as players DMing (deathmatching; i.e. fighting each other when they should be roleplaying), abusive administrators and perhaps too many plugins/mods (some people argue that this detracts from the imagination required to play, although I disagree).

I launched my own mod of this type but I wrote all of the economy code before I even knew about the Austrian School. In case you’re wondering, I discovered it near the end of Ron Paul’s presidential campaign.

Anyway, there are a number of problems with the system I have devised. For starters, it’s very difficult to duplicate real life. Whereas you can invent things in real life, I have to give up that idea (unless you can think of a good, non-abusable way of doing that) and instead focus on capital accumulation. Another problem is that there are not a heck of a lot of items in the first place. Most basic consumer needs are satisfied, but I would have to make more items. The final major problem I have is that my current system revolves around players purchasing “blueprints” which they can then use to refine ore, make parts and ultimately create finished products. Where will they get these blueprints from? I was thinking of having an “engineering organization” or something similar which sells these, but then I would just be creating a monopoly since it would involve the administration regulating it.

On top of this, the blueprint idea relies on the administration setting requirements for each item. For example, to produce a hamburger, I might need a bun, patty and sauce, but on another server, I might need different components. I’m also in the process of writing a “harvesting mod” which will allow you to harvest raw resources from the map.

One of the ideas that I have been given and which I think may be useful is to have “skills”, where you have to spend time learning something to be able to create something that is relevant to that. For example, even with a blueprint, a random person would likely not be able to manufacture a car. Instead, you would have to spend time learning how to make a car first and have a blueprint before you could do this.

Anyway, I’m looking for any comments on my ideas. If anyone has any questions about anything that is unclear (as I may have left something out that you need to know), please feel free to ask. I have posted two articles just below here which summarize my views. I have based these on my learnings of the Austrian School. Please tell me what you think and suggest any improvements/revisions.


“TSRP Economics”

I wrote this article over two of my last lunch breaks at work because I haven’t had the ARP code or an editor/compiler for it handy. I don’t plan on really changing the ARP structure at all so none of these things will ever be incorporated into it, but this is my template for the new Source mod I’m thinking of launching.

Contrary to what everyone else says about TSRP, I think it could have been great and technically still can. I’m not going to bother with “revolutionizing” it, though, simply because of the ancient engine it runs on and the fact that Source is much more interesting, capable and is growing much more. Also contrary to what everyone else says, I think that the way people act in that there’s hardly any RPing going on is a symptom and not a cause. There is a fundamental problem in TSRP that no one ever addresses, and that is that there is no real economy. For this reason, I’m going to explain why TSRP is as it is from an economics perspective.

For starters, people often blame a cast-in-stone set of problems for the failures TSRP faces. Among these are poor administration, a poor police force, too many plugins, excessive pandering, corruption, DMing and bureaucracy. I sincerely believe that whenever anyone blames any of these, they are messing up causality. These are symptoms of the one central problem of a lack of a real economy. Similar to how price increases are not the same as inflation, so too is a poor police force not the same as a hopeless RP environment.

The problem started originally as no one gave much thought to money, items or property. HWRP and ARP are the epitome of this. Whereas classic RP had a limited economy (i.e. you could not amass a fortune without ever talking to anyone, effectively negating the mode of production), the two most recent attempts at RP environments not only failed to encourage actual RPing; but they actively discouraged it. Without a reason to communicate with others, player could become rich and powerful simply by playing a lot and sucking up to admins among other things. There are many other problems with this system, as well. For example, if you were to measure the inflation rate of an average TSRP server, it would be somewhere around 10^8%/year. The major problem with this is that the goods stay at the same price with an infinite supply. Where’s the encouragement to produce something?

In accordance with some of the laws of economics, I’m going to explain what I think could be done in the future as well as what was done wrong now and in the past.

[list]
[*]The law of supply and demand. TSRP is a perfect example of the failures of central planning. Prices are artificially set according to what the administration thinks people will pay for them, despite many of them becoming useless or excessively expensive because of this. As the price goes up, demand should fall. As the price goes down, demand should go up. However, TSRP is lacking a price mechanism and as such demand is constantly increasing while supply remains permanently infinite. Since goods in TSRP are not sparse, there is no need for a price mechanism. If they were to be appraised realistically, all goods would be worth $0. This is ridiculous. The way to combat this problem is by creating a means of production whereby people can produce goods and sell them in limited supply and for a chosen price, effectively creating a price mechanism.
[*]The law of diminishing returns. With an infinitely large pool of goods, there’s no productivity going on at all. This means that TSRP never realizes this law. The most common way of obtaining goods is by purchasing them from an NPC with an infinite supply of them. Why not make people produce them and then sell them on the free market?
[*]The law of one price. Without the ability to compete with NPCs, players more often than not will simply ask admins for items and then sell them for a slightly reduce price. The NPCs have the advantage of advertising, infinite supply and an apparent standard (i.e. everyone assumes they are the only ones selling the item). The NPCs will never adjust their prices according to what the market demands, so this law does not hold up in TSRP. The way to combat this is, again, allowing people to produce goods and sell them for a competitive price.
[*]The law of diminishing marginal utility. With an infinite supply of many items, they eventually become useless. For example, if you were to install Medical Tools (“Pain Mod”) and simply hand out the healing items to doctors, your server is not going to last long because that infinite supply will discourage the doctors from being productive. Instead of healing people at a good price and with good quality, they don’t make any money anyway so they simply do the bare minimum. This can be solved by forcing factions to buy all the items that they need rather than just handing them out.
[*]The law of unintended consequences. Admins often get in the way of RPing despite their best efforts not to. The purpose of the administration is to ward off DMing [deathmatching] (which I would argue doesn’t even need to happen in the first place to a large extent). When they start interfering in the economy, they will screw things up. They don’t interfere in it because it’s natural, though. Admins interfere because they see problems and think they can solve them. According to this law, when they do this, they often cause more harm than good. Even approving organizations, deciding maps, etc. should be left to the server population. There’s no reason these can’t be left for the market to decide. Admins will not have incentive to intervene if the economy runs itself well.
[*]Wagner’s law. An economy doesn’t grow because of government spending; it grows despite it. As the economy grows, people naturally want more and thus turn to the government to provide it. This is one of the most visible problems in TSRP servers. Too often, the administration is overburdened with pointless tasks that could be automated or left to the market to decide. As the server grows in popularity, the administration will take on a greater role. This is usually when the players stop joining. They will instead move to a server where there is a small, effective administration. With a good market economy, the administration would never have to do this and could instead focus all of its efforts on development and DM control.
[*]Foldvary’s law of inequality. As a direct result of the law of unintended consequences, there is often a massive wealth barrier between the users in a server. Since there is no inflation and an infinite supply of goods, the admins can simply “print” as much money as they and “produce” as many goods as they want and ultimately it will never have any impact on the economy. The best way to eliminate this is by having some sort of dollar index, which tracks the amount of money currently in circulation with regards to the value of all available goods.
[*]Say’s law of markets. With an entirely consumption driven economy, TSRP fails to realize this law. Instead of encouraging production, handing out money through “salaries” coming from an infinitely large bank account discourages people from doing anything other than consume. This is the cause of the very materialist view that most RP players have. They hoard wealth and power because it’s all they’re encouraged to.
[*]The law of subsidization. When you subsidize something, you encourage more of it. We subsidize various parts of TSRP including organizations, essential items (medical tools, for example) and the police force, but why not try directly subsidizing RPing? This would be unnecessary in my ideal environment because people would RP without any government incentive, but for now a perfectly valid idea is to have your admins hand out rewards whenever they see a good RP session going on.
[*]Law of time preference. We discourage savings through the constant flow of money from salaries and admins. People thus naturally consume right when they get their money. If their organization needs a subsidy to buy a property, they blow all of their money then go to the admins and ask for more. Not only would the properties be affordable, but also people would have encouragement to save in a realistic economy.
[*]Law of the market. There’s no strong trading mechanism other than the “salesman” plugin, which is in limited use. Since no one has any reason to trade, there is no real market for goods. Scamming is a big problem with people often walking off with money or items. Admins are usually used to rectify this which is a perfectly valid purpose for them.
[*]The law of wages. In a real economy, players would receive salaries according to how well they performed their jobs as well as how much their task is in demand. In a manufacturing job, they might receive a piecemeal salary instead. Unfortunately, in TSRP, you get paid for standing around and doing nothing. This is the cause of moneyfarming. If players actually had to do something to be paid, there would be no money farmers.
[*]The law of economic rationality. This law explains why people DM. While there are a select few who do it just to piss off the server populace, for the most part DMing occurs because people are simply bored. People are rational, and without incentive to do something (even if it’s just by providing them tools to do it), they will resort to rebellion. DMing is a form of this. This can be proven easily by simply looking at anyone who runs an organization or has a decent job in a server. More often than not, these people will never DM or only did in their first few days. On the other hand, it’s not always players who join right away who will DM, but those who are waiting for applications to be processed, can’t get a job, or can’t find anyone to RP with. This can be solved by allowing the market to do everything (i.e. form organizations without government approval using private capital, create new items using templates, etc.) instead of the slow and bureaucratic administration.[/list]

That’s pretty much it. I think I’ve more or less explained every problem TSRP faces. Many people will probably say it’s facing more social problems, but I think social policy is always dictated by economics. I was going to back up each of my points with real life examples, but I figure that it would just make the article too long.

If you want to read more about free-market economics, visit the Ludwig von Mises Institute at:

If you have any questions, comments or ideas, feel free to post.


Money, Value and Labor

There’s been a recent onslaught of people either trying to extend on or counter my views. I am surprised by how much everyone is getting everything all wrong. This article is an extension of my TSRP Economics article, written to discuss some of the things that I left out or want to expand upon.

I’d first like to start with money, which is the area that most people have disagreements with me in. Money, contrary to what everyone seems to think, is simply a medium of exchange. Money also arises naturally from the market place. The idea that the government can print paper and force people to use it is nonsense. In real life, we’re seeing the beginning of the end of the dollar system. The only reason it has worked in TSRP for so long is because the supply side (coincidentally the name of a school of economic thought) is done by the server. The server, by virtue of the fact that everything is virtual, has the ability to create and destroy any item in any amount and from any person’s inventory at the snap of its fingers.

If we want to introduce a real economy, we have to completely revise everything that we think about money. The “Wallet:” and “Bank:” values in your HUD must be thrown away as an inviable system. They don’t work and never will. To explain why, I’m going to have to diverge for a moment.

Let’s set up a hypothetical economy consisting of me, John and Joe. I am a baker; I produce bread to feed myself and others. John is a shoemaker; he produces shoes for us to wear. Joe is a homebuilder; he makes homes for us to live in. Let’s say, for example, that John and Joe will produce goods for me (John a pair of shoes; Joe a house) each for 5 loaves of bread. They each think that this is a fair arrangement (excuse the abuse of the subjective theory of value). So I produce 10 loaves of bread and then give 5 to each of them. Everybody is happy. The entire economy is more prosperous.

This is, however, not suitable for a larger economy. What if John wanted to buy a house from Joe, but Joe wanted 5 loaves of bread? Well, the solutions is simple: we introduce money. I say that each of my loaves of bread is worth $1. Using $5 that John has from selling his shoes, he gives Joe $5 in exchange for his house. He then comes to me and pays me $5 for my loaves. Everybody is happy and the economy is more prosperous.

Now, what if this money had a highly volatile value? Let’s say that someone else comes into the picture: Mark. He “creates out of thin air” $3 worth of money. He then comes to me and buys 3 loaves of bread immediately. My price goes up because I am experiencing higher demand. John comes to me and pays $5, expecting 5 loaves, but I only have 2. The price thus becomes $2.50. But uh oh, even if he purchases them, this is not enough to support him while he works on the pair of shoes he wants to sell. The economy collapses and nobody has anything to show for it. By this argument, I have been trying to show how money in TSRP works right now. Mark is just any random new player who has been given a salary by the server.

Back to my point, now: this system works fine when there’s infinite supply of all goods and does not visibly falter when there are no long-term projects that must be supported by loans. Inflation, however, always encourages consumption at the cost of savings and throws off time preference, shifting it towards now. Unfortunately, with the economic model that is soon to become a reality, this system will no longer work. For this reason, we have to devise a new one.

The most obvious way to solve this is by going back into history and looking at what has been money when the free market decides. The answer is commodities. Whether it be gold, silver, oil or tobacco, money has always been something that has a purpose in one form or another. So what’s the solution? We have to let the market decide which item they want to use as money. This item will arise naturally and its supply will expand and contract in accordance with economic expansion. To explain this, I will again have to diverge for a moment.

When economic growth happens, the total value of all constituents of an economy increases. If the amount of money in circulation stays the same, we have price decreases. There is less money tracking more goods, so the money becomes more valuable. If the economy is expanding at the same rate as the money supply, we have price stability. This is not, however, as great as it sounds. In a truly free-market economy, prices should constantly go down as capital is created and allocated toward processes that were previously less profitable. For example, let’s say I can produce 10 toys per day when I first start up a factory, but then 20 toys per day when I get new equipment. I have accumulated capital. Not only does the price of toys go down because of higher supply, but also because I am paying less to produce each one. This means that, despite economic and money supply growth, prices should always be going down.

Back to the money supply problem, let’s assume that the market has chosen gold to be its money. Let’s say you can buy fifty cigarettes for one unit of gold (likely an ounce). This is a flawed measurement, but bear with me. The economy grows and the supply of cigarettes increases. Now, you can get sixty cigarettes for one unit of gold. Entrepreneurs will see this and begin mining gold, increasing the supply of it. When they mine and sell enough to bring cigarettes back to one fiftieth of a unit of gold, the economy is back in equilibrium. At this point, it will likely take an equal amount of time to manufacture fifty cigarettes and mine one ounce of gold. It is no longer profitable to have preference of one over the other, so the supply of gold has increased at the same rate as the economy. The same thing will happen during economic contraction; the constituents of an economy will sell off their gold in favor of other goods, thereby increasing its supply and decreasing its price.

You might ask why we can’t simply let the government track economic growth, and the answer is very simple. The price mechanism exists for a reason. The price mechanism is the ultimate means of allocating capital. As I showed in this example, entrepreneurs will flock to something that is in short supply and bring it back into equilibrium. Government bureaucrats cannot know this information because there is a theoretically infinite demand for all goods. Prices simply show how much society is willing to forgo consumption of one good in favor of another. The government setting prices or flooding supply will never be as accurate as the market.

The next point I’d like to address is another comment I get repeatedly: that is that TSRPers are “little kids” and won’t understand this system or accept it. In reality, it’s quite simple. In fact, the current TSRP system is much more confusing and nonsensical. As it stands right now, you get paid to stand around and do nothing. All I’m saying is that we have to introduce the production side.

The reality is that human nature is the same everywhere. Greed is the driving force for everyone and this should be embraced. What counter-balances greed is fear; that is, fear of losing your money. If you go to a casino with some money, you can make a lot or lose it all. Almost everyone knows it would be insane to go to a casino with your life savings. Despite the greed of the person who goes to the casino and wants more money than they have, the fear of losing it all keeps it in check and they will generally bet conservatively.

This same concept applies to TSRPers. They are no different. They are driven by greed and are kept in check by fear. They may be less intelligent or less wise, but the entire point of capitalism is that good decisions are rewarded and bad decisions are punished. They cannot learn without making mistakes. So we can’t perfectly replicate life. Oh well. Almost anything would be better than the current system.

Moving on to my final point: a free-market economy adjusts rapidly and efficiently to changing conditions. I’ve had people throw arguments at me like “oh, if there’s no item deterioration, your economy can’t work because no one will have the job of repairing them!” This is the most ridiculous argument I have heard yet. Anyone who says this clearly has no understanding of basic economics.

The heart of this argument resides around the idea that labor creates value. This is nonsense. If my watch breaks and I have to get it repaired, sure, I’m giving the repairman a job. But what if I had a watch that didn’t need to be repaired? Contrary to what you may think, the repairman will not be out of a job. Instead, his labor will be shifted to a more productive purpose. For example, he could move instead to producing new watches, which is the creation of new wealth. There’s another angle to this argument. Until now, I’ve completely ignored that I must spend money to get my watch fixed. If my watch had not been damaged, I could have spent that money on something more productive. For example, I may have needed a computer to write programs. Instead, I have to forgo consumption of this computer in order to get my watch repaired. Hell, under this logic, why don’t we run around smashing everyone’s watches since it’s such a great economic stimulus? This concept very closely mirrors Bastiat’s parable of the broken window.

That’s it for now. If you’re interested in more similar articles, check out the Ludwig von Mises Institute. If you have any questions or comments, feel free to post them here.

I can’t help but wonder what most frequenters of this site likely think of it (probably too long to bother with), but as a fellow gamer who has dabbled in designing systems to emulate or demonstrate the actual functioning of a real economy, I found it interesting.

Probably the key difficulty in designing the framework for an in-game economy is recognizing that the primary economic actor in any game is not the character, but the player. The player might imagine that their character has certain needs and desires, but the REAL need is that of the player, who is presmuably there to have fun. That, I think is the reason people “deathmatch” instead of “roleplaying”; most gamers enjoy competition, and when “deathmatching” is really the only opportunity for competitive play the system offers, that is what people do. They aren’t “rebelling”; they’re simply creating their own value in the only way the system offers. GMs try to stop their fun because this isn’t what they had in mind when they made the game.

Spoilsports. :wink:

An experiment I’ve always wanted to see is a game system that explicitly lacks an official “money”. Players, to acquire the equipment necessary to augment their abilities, apply skills to resources found in the game environment (or to previously produced items). There are no “NPCs” that automagically produce “weapons” “armor” and “items”, rather the first players that enter the game enter into a wilderness with nothing more than the clothes on their backs… if that. From there, they’d have to produce their first crude tools, with which they could produce progressively finer tools… once they’d figured out the necessary combinations.

In a way, it’d be like one of those old text adventures. They’d enter an area, “use flinting (a skill) on rock (an item found in the game world)” (this would get them their first knife or spearhead), use knife on wood (this would get them their first staff), use spearhead on staff (this would get them their first spear) and keep doing that to figure out what each item does to other items. You could have as many base materials as you want, and with enough development, you could make it theoretically possible to go from sticks and stones to flying cars (given enough effort on the part of the playerbase to implement the possibilities). Over time, recipes would spread among the players–a key component in economic development.

NPC merchants could be used to provide static points of exchange throughout the game world, but they would have nothing more than their player “masters” had given them, and accepting nothing less in return than their masters instructed them to.

(Come to think of it, precoded exchanges could be but one of the things NPCs (possibly some sort of “beastman” who can be “harvested” using a taming skill just like any other “natural resource”) could do. You could allow their actions to be scripted by their “owners” (include a robust scripting language, or perhaps just allow for plugins to be written in the same language as the main program’s source code) to do everything from precoded exchanges, to serving as guards for persons, items, or areas, or anything else the scripting language could be bent to do. Their relative strength at various activities could be limited in the same way as players: skillstats which develop through use. In this way, you utlize your playerbase as a self-interested part of the development process, particularly in that stickiest of ares: AI development.)

My prediction is that, over time, players would specialize in producing certain components, to minimize the amount of time they spend wandering between different locations or sorting through piles of complicated inventory (not to mention the skills system, and the limited utility of particular capital developments. In addition, a few items, useful in many processes, would likely end up being commonly accepted by most players (and their NPC exchange points) in exchange for their produce (even if that particular player has not immediate use for it).

This would be “real” money, as opposed to “gold pieces” which I regard as a form of “voodoo economics” when used in a game. “If it works in the real world, it must work in the game?” Not necessarily. The “physical” realities in game do not match real world physical realities, and in most games, “gold pieces” are anything but useful, or rare, or possibly even durable. Their only quality is that the NPCs accept them for their bottomless pit of static inventory. Get rid of this, and you get rid of possibly the biggest barrier to simulating a “real” economy.

I would leave in “deathmatching”, and even allow the winner to make off with the loser’s inventory (even if the loser is one of the “beastman” NPCs I mentioned earlier). I would create no artificial “safe” places. The primary reason for this is to model one of the most difficult problems of economics: politics, the question of “What is the most efficient way to keep people from killing and stealing from producers rather than producing, themselves.” Programmable NPCs would obviously play a role in this process.

The GMs would remain strictly neutral on this question, focusing their efforts on helping people with glitches (“Help me, I’m stuck!”), keeping an eye out for bugs, and ever expanding the chain of item production and skill acquisition (new skills available only to new characters?) beyond its current base. As players discover the new processes, innovation is modeled.

Alternative, you could give GMs territory in which they are allowed to intervene (even code in new program rules which are executed locally), give them the title “god kings”, but leave at least a third or so of the territory unmodded, as the “control” to the various GM experiments.

One thing the GMs of such a game should keep in mind is that, sometimes, game expansions will disrupt game balance. It happens, even in the real world, and the proper response is not a nerf, but rather to observe with fascination how the in-game economy reponds to this imbalance.

One thing to note is that you are not going to “win” economics arguments with fellow gamers. Gamers are generally fairly intelligent, generally young, and if there’s one thing intelligent young people are sure of, its their own uninformed opinions. It would be far better to use these arguments primarily to identify people who might be interested in helping make the game. I wish I was a better programmer than I am… I’ll help you, if I can (I do have knowledge of basic flow control, arithmetic functions, and I can write… or at least, I hope that’s the impression one gets from my posts).

Thank you for your excellent reply. Your ideas have greatly influenced me already and I plan on trying to implement some of them.

Unfortunately, as much as I like the idea of starting as cavemen, this is not logistically possible. The biggest possible server in the game I am working on only has 32 player slots. On top of this, I am the only real developer of the programming side so it would take quite some time in a game that is already dying out.

I really like your point about thinking about the problem from the “player” side as opposed to the “character” side. I recognized this before but I don’t think I articulated it very well. Also, I agree with your statement about how an implicit form of money should be being missing.

You’re also right about one thing that I threw away a while ago as being unnecessary and logistically difficult, and that is skill development. This has been tried in the game in the past and it has never gone well. After thinking about it, though, I think that writing off the previous skills mod for its problems is like socialists writing off capitalism because of our current situation. The mod was done terribly, was massively imbalanced and created monopolies (you had to talk to a “trainer”, who was a player that charged you to learn new skills, among other problems).

Are you interested in advising me at all or perhaps even getting into developing this? I could really use some help.