On the economic models in games

Forget for a moment the externalities in the economies of a given game, a MMORPG. Lets think just about how its own economics work, as if isolated from the real life money or players' actions outside the game.
Another article I wrote talk about the manufacture of conflict that artificially change the odds of combat, and that takes part in why the economic model is chosen. The Sink/Faucet model is the economy in which money sinks and money faucets control how much money circulates in the game.
While it is a model similar to real life controls like basic rate, bonds issuing, and other such measures, it works in real life as accessory to the reality of the monetary base. The games often remove the monetary base because it is a complex control for the market. It requires that the game itself account for its own "manufature of conflict".
I swear this is only about games, regardless of how it reads relating to real life.
Manufacturing conflict is inflationary. The people needing more and more resources to keep their position, by the artificial fight the developers make to counter the natural advantage of conquest, will demand more resources, and as resources are "infinite" as long as they have a faucet, the more one manufactures conflict, the more expensive the Dreamworld, the World of the Game, becomes.
One of the solutions found by developers is to create bound currencies, which work similar to certain titles in the real World economy. You have a certain paper you can save but cant sell or transfer. It will become money or something in time, or grant you certain rights as long as you have it. Again, this only works for stable economies because it is in the money systems that runs a monetary base, not because it works by itself.
When you insert those currencies in a game, instead, you are adding yet another thing you have to control closer. You see, adding that to a sink/faucet base economy only works as tools to get more of the currency you can trade. You have the "seals of the mystic arcanum", or some such non-sense, and those are used for better gear. That better gear can be used for one thing among other things: Get more transactional currency.
We need also to account to the other side of the spectrum. If we have these bound currencies, bound gear, and irrelevant creating professions, we are making the "deescalation of conflict", given that PvP will often be optional if you want a persistent World that has enough population. You will require your game to have trade of some sort, and have means to interact in such a way resources change hands for a price.
Now add back the externalities, namely, RMT, microtransactions, and other real life money influence in the game.
Buying gold is by far the least economically invasive force in the game. It causes inflation, but not as much as the idea that selling power in the game will work to manufacture conflict in a good way. It will manufacture conflict, but the type of conflict that divides players, and soon exhausts the population that makes the buying of such powers, even cosmetics, worth the effort.
One can argue that the developer does make money from that, but making money with a moderately successful game is easy. What we are talking about here is how much money you make.
If the market allows for monetary base, and you can farm materials but not money, then you have a strong, yet complex, economy, and therefore you can mitigate the problems you cause, except the manufacture of conflict.
The solution is to create what governments do in real life. You may have hubs in which conflict is regulated, and people conflict averse can work, and generate wealth, the kind that is chased by all. The result is that you have thematic realism created by the economic control of your Dreamworld.
You can even not care about the gold selling in your game, because gold sellers will soon realize the new framework of currency discourages unproductive people to hoard gold. You have introduced the price of money. In faucet/sink base frameworks, making in game currency costs time in game, and time might cost IRL money to pay "players" to make that gold. If you use monetary base, you also incur in market dynamics, and those will hinder the ability to generate gold by farmers.
Imagine you get gold from certain activities as a faucet. Those activities will make certain that 1h gives you 1000 gold. That makes certain that your 5 IRL currency paid for hour generates 1000 gold. Removing any source of game currency, or gold as it is generic used, will start to make the market decide what generates more gold on farming. For that to work, however, there must be no activity that generate gold by itself, thus, monetary base framework. The amount of gold existing in the game World must be determined, and at all times, the game "bank" must have at least 50% of that amount in its vault. That is a security measure to avoid the situation in which the game cannot "buy" the circulating market back as a whole and start over. That concept will be worked out in another article.
Now we can manufacture conflict in another way, by creating the search for resources and the need for them. The whole game will shift to the search for those important resources, and people might even create a dynamic set of objectives. 
Suppose the players need volcanic glass to build end game weapons, and thus, as the players reach end game level, they start farming it. The industrious types would store such materials to sell. As it becomes not so rare material, it starts to devalue, and other materials may rise. 
By controlling the material bill for technology, and sources of such materials, the game can regulate the conflict and the drive for it.
It could also, like IRL, create NPC run businesses that sell materials for a given price, and buy materials for a given price, serving as buffer for the game internal coffers, and price marker for the player market.
If a NPC store sells volcanic glass for 20 and buys if for 16, the is the minimal prices players might want to trade those, not having any association or deal, as if someone is selling volcanic glass for 30, it would not make sense.
However, here, the game needs to implement something games often dont: Information.
In many games there are vendors selling items for a price, in the effort to mark that price up, but the game itself has no information readily available as to where to buy them, so people can effectively use the player driven market for overpriced sales.
It is in the interest of the Dreamworld economy that this information is widely known, otherwise the skewed economy will damage it.
On that note we come to another point in which I have to stress I am not talking about real life.
One of the worse problems regardless of the system is the conditions that allow the "super wealthy" to exist. In the creation of the monetary base I said it is important for the game to keep over 50% of the monetary base so it can "buy the whole economy" any time, or keep liquidity. If a player, or a association of coordinated players can successfully approach the 25% sustainable liquidity, they are able to maneuver the economy of the game with force that threatens the liquidity of the game itself. And that might work even more damaging effects in a sink/faucet economy, in which that in itself allow the player to extend its influence over the other players, as any effort to make the whole playerbase better will also make those super wealthy players better as well.
At the end of the day, manufactured conflict and creating simplistic economies in games can damage a lot in terms of the ability of the game to keep resources and gameplay as interesting as they want.