Curious if anyone in the community has worked on economic models before. I would love to know their insight into all the limitations and gold sinks that Diablo has in effect and how that will drive the value of virtual goods in D3.
Wow is a good starting place to figure that one out. Gold sinks are just tools used to counter the fact that an infinite amount of gold can drop from enemies. Of course some players can learn how to play the economy in their favor and control large chunks of gold despite the sinks. Blizzard is usually good at balancing things though. I'm not too worried.
Gold does appear to actually be a currency in Diablo 3. It was not a currency in Diablo 2. So unlike in Diablo 2 there's not a ton of pressure to move to an alternative currency regime immediately.
Currency supply is controlled by a negative feedback loop from the auction house fee. Essentially, as people accumulate more gold, prices rise on the auction house, which sinks gold more quickly. Napkin model tells me MV is more or less determined by the gold drop rate and is fixed.
Total goods available for sale should increase over time since goods are all durable and are not sunk in any way. So expect prices in gold to drop; more precisely, value will concentrate more and more in the very best drops as time goes on.
On the RMAH, you have an external currency. Demand for game goods will start high and decline with time. Supply of game goods will start low and increase with time. Expect prices on the RMAH to start sky high and quickly crater.
As prices on the RMAH crater, more items will be pushed onto the GAH. This will drive down the gold price of those items further, as more and better items compete for a more or less fixed pool of gold.
Short version: expect gold to hold value better than items.
This is what I expect as well though might not have a great theoretical understanding of why it happens. I expect the RMAH to move geometrically in a reverse asymptote (away from infinity) towards the baseline of fees imposed by paypal and blizzard.
the straightforward approach to profiting off of this would be to sell any items of decent quality immediately on the RMAH on release day.
Would a backwards approach being to sell everything for astronomical prices from the beginning on the GAH and save that currency for the point when the RMAH to GAH ratio favors gold?
Interesting notes on the GAH, and I expect new expansions to start this process all over again starting from infinity...
I also have no idea what this means "Napkin model tells me MV is more or less determined by the gold drop rate and is fixed". is there a wikipedia link that you could reference to make light of the jargon you use.
Thanks for the less heard insight on the game too!
Am I the only one that thinks there are too many variables and not enough late-game information to make accurate predictions at this point? We can only speculate about the value of gold because we do not know how late-game gold drops will work. Not all the items have been set yet. Blizzard can also "tweak" the economy by changing or adding items via patches.
Am I the only one that thinks there are too many variables and not enough late-game information to make accurate predictions at this point? We can only speculate about the value of gold because we do not know how late-game gold drops will work. Not all the items have been set yet. Blizzard can also "tweak" the economy by changing or adding items via patches.
I think this is a great place for us to start!
lets try and come up with a finite list of all the tools blizzard could use to change the economy.
heres what I see right now
-$ flat fee
-gold drop rate
-total item variation and item count
-the data they give us for last ten trades and last day of trades for gold to $
-Item drop rate (including gems)
-Max auction values (capping all auction items at a specific $ amount)
-More endgame content that is item dependent (something like D2 ubers)
-Difficulty of farming
Would a backwards approach being to sell everything for astronomical prices from the beginning on the GAH and save that currency for the point when the RMAH to GAH ratio favors gold?
I don't think that will work out exactly like that - the big piles of gold that you'd want to sell high ticket items for just won't exist early on in the game's life cycle. It will take a while for people who will eventually have the big war chests to build up to that. Also, for very high ticket items, remember that the RMAH, as it stands, has a flat fee, while the GAH takes a percentage. Even if you just want gold, for a big enough ticket item it will make sense to sell on the RMAH and buy gold with the proceeds.
I also have no idea what this means "Napkin model tells me MV is more or less determined by the gold drop rate and is fixed". is there a wikipedia link that you could reference to make light of the jargon you use.
I'm mostly just writing down the equation of exchange (an accounting identity) and making some guesses about how terms will move in relation to each other. Start here:
- What makes you believe so in a world where nothing decays and where you get endless supply of gold just by playing?
'Real world' money doesn't decay either - if it did, it wouldn't be money.
You get a potentially endless supply of gold just from playing, for sure, but you also get an endless supply of items just from playing. The economy is based on exchanging one for the other. There isn't enough information just in stating those facts to make predictions about how things will play out.
- The rising prices phenomenon in real life is called inflation, which means that the money is losing its buying power and becoming worthless.
Take a step back for a moment. What do we really mean when we say that there is inflation? Take a really basic two good model - there is 'money', and there is 'everything else'. At any point in time, there's some supply and demand equilibrium between money and everything else. Not at some future point in time, there will be a different equilibrium. Holding all else equal, if there is greater demand for everything else, or a lower supply of everything else, or a greater supply of money, or a lower demand for money, then the price of everything else in money will go up, and you will have inflation. If instead, holding all else equal, there is lower demand for everything else, or a higher supply of everything else, or a smaller supply of money, or a greater demand for money, then the price of everything else in terms of money will go down, and you will have inflation.
In reality you have a balance of all four of those factors moving somewhat independently, giving you the overall change in price level. Money never becomes worthless; if you have massive inflation or deflation, the instability, in reality, makes your currency stop functioning as money, and people migrate to new money.
Am I the only one that thinks there are too many variables and not enough late-game information to make accurate predictions at this point? We can only speculate about the value of gold because we do not know how late-game gold drops will work. Not all the items have been set yet. Blizzard can also "tweak" the economy by changing or adding items via patches.
That's the reason I'm being pretty conservative about the sorts of predictions I'm willing to make. The kinds of questions you're asking will absolutely have a large effect on what the actual price levels of various items in the game are. Will end game items cost 100,000 gold or a million gold or hundred million gold? We don't know. I can't possibly tell you that.
What we can do is look at the systems and make good predictions about the shape of the curves things will make. We may not know how much items will cost in an absolute sense, but it's not a stretch at all to say that there will be more top tier items, per player, on the auction house a year after launch than a week. How many more? I don't know. But you know that will trend up.
I am extremely happy with this discussion and glad people out there take this theory into consideration.
I am curious if there are observable and recordable data points that one person or a group of people could collect that the game denies you
obviously there is the gold to dollar exchange rate over time, that data doesn't seem to be available in the beta except for the past tent trades and the last days worth of trades. A larger data set could be used for pattern recognition.
The lack of an API for the AH's means that a full set or automated collection of data will never be available unless Blizzard just puts it out there (and monkeys might fly our of my butt).
So what could we as a community collect that would be worth analyzing?
...If I am getting it right, it seems, that the ultimate value in the game is Time, the more time you put in the more probabilistic gear and gold/$ you will get. You can buy time away from others through the dollar value which will change over time because of the supply and demand of in game items and gold. <--open to critique and correction! I am a complete laymen to economics and trying to get a basic understanding through rad peeps on the forums who are much much smarter than me.
- Alright, let's take a parallel to real life. By doing a thing X you get all your needs (=items) fulfilled and lots of money (=gold). Now if you have all your needs fulfilled you have no need for money anymore.
If you acquire all the items you ever want just by playing then you have no need for trade. You can stockpile a giant pile of gold for yourself. Since you do not trade with other players, none of the items you find or that giant pile of gold influences the player market in any way. As your presence in game consumes no rival resources, your net effect on the external player economy is zero.
If I *did* want to include you in monetary aggregates, I'd have to add your gold to M, and subtract the fact you don't spend it from V, which on the whole would have...no effect. I could include the items you found and kept in Y, but when you add that to the overall accounting balance you treat those items as being sold to yourself...so again, you're just putting a +C onto it.
This isn't terribly complicated. If you don't use the auction house you don't have an impact on the player economy. That should be obvious.
- Mistake here is thinking money is goods. You have no need for money, you have a need for things money can buy..
Money can absolutely be treated as a very particular kind of good - a good that provides a liquid medium of exchange, a divisible unit of account, and a store of value. You can say that you don't have a 'need' for money, but that's different from there being a demand for money - of which there is ample. It's an essential feature of all but the most trivial of economies. There is value in a medium of exchange that you can buy and sell with, because it overcomes the matching problem of buyers and sellers. Without money, you can only trade what you want to sell when you find a buyer who *also* has something to sell that you want to buy. With money, that transaction is broken in half - you only have to find a buyer, and then only have to find a seller, and they can be different people. This is a huge step forward in facilitating trade, and the more trade you have in your economy, the higher the demand for money for this purpose.
Money also needs to be a differentiable unit of account, so you don't have to find items of equal value to exchange. You can price everything in terms of very granular money, making it easy to 'make change' from a trade, to add extra or take a bit back. Not only can you find different buyers and sellers with money, you can also shift the size of the trades, splitting up big sales into smaller buys or smaller sales into one big buy. Also, as a store of value - you can sell without needing to buy immediately! If you're bartering, you can only sell when someone can buy; with money, you can sell whenever you have a buyer, and complete the other half of your exchange whenever you see fit.
So yeah, you don't strictly need money. You can't eat money. But money is what makes trade work, and was a huge, huge invention in the progression of mankind. You don't sell for money because you need money - you sell for money because you need other goods, and are confident that you can exchange your money for things you do need.
There's nothing fundamentally different about in-game money and US dollars or Euros or Renminbi or what have you. They are mediums of exchange that govern particular currency areas and goods bought and sold in those areas. While concerns about exchange rates and inflation / deflation are completely warranted and perfectly rational, I can assure you there is no reason to think gold in Diablo 3 will function any differently in the Diablo 3 economy than US Dollars do in the US economy.
I am curious if there are observable and recordable data points that one person or a group of people could collect that the game denies you
Gold exchange rate should be pretty easy to monitor as a 3rd party; you'll have to write a bot to query the bid / ask price on gold frequently, but that should not be difficult and I would expect to see that data available fairly quickly after the game goes live.
I would really like to be able to track item sale prices on the auction house. The stats geek in me would really love to beat that data set over the head with MRA and build a dynamic price guide for items.
The lack of a specced API isn't fatal, it just means whatever software you use to gather data has to go through the human interface elements, which won't be nearly as efficient.
...If I am getting it right, it seems, that the ultimate value in the game is Time
From a very high level view this isn't a bad way to look at it. You can buy people's labor (time) directly, or you can buy the goods they produced in that time, or you can buy machinery that represents an investment of time and that produces virtual time. In the real world there's also non-renewable resources in the equation, but even a lot of those you can measure in time (I'll joke with friends about a barrel of oil being equivalent to about 6 slave years). In the Diablo 3 universe it all starts from zero, however, so everything represents the output from time input, and yes, you're absolutely trading that time around in your exchanges.
The lack of a specced API isn't fatal, it just means whatever software you use to gather data has to go through the human interface elements, which won't be nearly as efficient.
I suggest this being a very bad idea to anyone out there, Warden (if warden is working for diablo as it does for WoW in catching bots) will most likely catch and suspend your account and you will be violation of the EULA. I believe blizzard to have this game on total lockdown for bots and modding and data collection because of the effects it would put into play on the economy.
would any rational agent have a need to use auction house without barter option.
I don't understand this question at all. Do people *need* to trade with each other? No. Do they want to? Yes. Is it rational to want to trade with each other? Yes, since it's a value-adding activity. Is using a currency to facilitate trade rational? Yes. Is a 15% tax rate enough friction to drive a sizable fraction of players to look for an alterative scheme? I doubt it.
I take it for granted that money has value, yes. I also take it for granted that if there is no money, people will invent money, spontaneously, to facilitate trade.
History has many examples where hyperinflation has rendered currencies useless and people have reverted back to bartering to make living.
I do not take for granted that any particular currency will have value. Currencies collapse all the time and are replaced quickly with other currencies. There is no contradiction here. People will use a given currency as long as it functions as money; if it ceases to function as money, people will abandon it and find a new currency. We've seen this many times before (and are starting to see it in Greece today, actually, in the face of widespread deflation and currency shortages).
Gold is not exactly the same as real money, since gold can't get you any good items in Diablo III.
'Real money' can't get you everything either. Individual currencies are only predominant in localized currency areas, and cover only goods, services, and assets within that currency area. To purchase goods in another currency area, you have to exchange your currency for the local currency, with which you purchase what you want. This leads to exchange rates and relative currency valuations depending on the economic situation in each currency area.
Gold is no less a currency because you can't buy items on the RMAH any more than US Dollars are not a currency because you cannot use them to buy land in China.
This problem had extreme proportions in Diablo 2. Nobody would trade their Stone of Jordan for any gold you'd give them, because there was no need for gold.
Ah, now I see where you're coming from. Gold in Diablo 2 had a distinct problem. For the first year of the game, a Stone of Jordan had a very clear price, in gold - roughly 5 million gold.
There was no reasonable way for a player to trade for 5 million gold. Even the expansion only raised the gold cap to 2.5 million gold; it was a million if I recall correctly pre-expansion. How could you trade a SoJ for gold if you couldn't store that much gold, let alone carry it (which, I believe, is still limited to a few hundred thousand).
So gold in Diablo 2 wasn't a currency. You couldn't store it in large enough quantities to use it as a medium of exchange. So the currency was abandoned. People needed a new currency - something uniform and compact enough that you could store it reasonably enough to pay for things. Perfect gems were tried at first, but weren't valuable enough to overcome the storage constraints. Which is why people turned to Stones of Jordan. They weren't nearly as differentiable, which made it hard to make change (perfects kept some value for a while for this reason); and they did still have a storage constraint (though much better than any other option available). But they were uniform (giving them consistent value) and readily available (you could gamble them up for ~3.5 million gold each) - and there was nothing better available - so they became the de facto standard.
The *only* reason gold failed as a currency in Diablo 2 was the gold cap. If there was never any storage or carry limit on gold I am confident that people would still be trading items for (millions and millions) of gold today. Everything else about the economy of that game was set up sufficiently to support gold as a stable currency, but that one innocent mistake wrecked it.
I don't see a comparable mistake in Diablo 3, so I have no reason to expect gold not to function as money.
I said this on Reddit and I'll say it here. If we as a community can agree not to hyper inflate the economy it will be a much funner experience. Think about it. I'll use an example from WoW. On Blackrock US a few friends of mine, myself included played the AH a lot. Keep in mind this was at the peak of BRs most recent flood of pvpers and best of the best guilds. We all noticed a major change in prices, they all skyrocketed in almost every market. We devised a plan to lower all the prices and talked a lot of the people into helping us. Long story short, Everyone could afford anything give or take a few of the rarest mounts and stuff like that. We never really got to finish the experiment but we did come to the conclusion that if we had got more people to participate it could have started a server revolution. More people having more affordable items in turn making items sell faster for less. Nobody ever did the math, but in the market I was in I might not have been selling each item for a high amount but I was making more gold faster than I was selling at a high price. To create a flourishing economy is not that hard, it's when those people who think they will come in and get rich quick gets involved is when it starts to get destroyed.
But as of this second, nobody will know the outcome of this until the end game value on gold is determined.
Cool thing is.. We as the community are the ones that will be determining it.
Am I the only one that thinks there are too many variables and not enough late-game information to make accurate predictions at this point? We can only speculate about the value of gold because we do not know how late-game gold drops will work. Not all the items have been set yet. Blizzard can also "tweak" the economy by changing or adding items via patches.
I agree with you actually. Developers have said many times the one thing they need more of in later levels is gold.
As well as virtual value goes, if a lot of people play and get addicted to their characters, you could see thousands of people willing to buy items. We don't known yet as this has never been attempted on a AAA game from a AAA developer that is doing this legit.
In short, could be huge or could be meh.
Also, item values will vary with supply and demand. It will probably take a month before enough demanders get to level 60 for example, so items won't sell much before that. Likewise as I said above, is this going to be a game where a vast amount play to high levels or will most quit after they finish it on normal? Blizzard appears to really be pushing this as a game where you play to end game (unlike d3 where most quit after normal) and if they are successful that will offer a huge potential market....
My initial aproauch would be get a estimated gold stock series and see (running an unitary root test) if gold stocks are estationary or not.
Since theres no data available the approach need to be purely theorycal. As I see it sinks truely play a important role. It's not an matter of taking coin out of the economic, as most people think. Independent of the gold drop rates and sinks (which is a inverse drop rate), inflation exists and existe at the same level (unless ALL the gold is actually sinked).
But i have the feeling the higest the utility for gold, lower the inflation. The actual model is bit hard to elaborate once you need to creat a economy with gold and this gold has both a means of trade and actual usage utility.
Anyone raised the idea that drop rates will be dynamic? i mean that the drop rates of items will be allowed to vary on a global scale, controlled by blizzard. I wonder what this will cause economically...
In the absence of any gold sink, the rate of inflation would be incredibly high at the beginning of the game, and gradually taper off asymptotically over the life of the game. This is simply because, at a fixed gold drop rate, there's a much larger relative change in the amount of gold early on, when people do not have a lot of gold yet, than later on, when the economy is flooded. More rigorously, a first order approximation of inflation is dM/M (where M is the amount of gold in the economy), and M is just the integral of dM from the beginning of the game until now. M goes to infinity as the time T goes to infinity, so dM/M goes to zero.
If you have something resembling a percentage tax rate (for example, 20% of gold drops go to repair bills, on average), then the economy behaves exactly the same way, albeit with a slightly lower price level. The rate of inflation is unchanged - if the tax rate is r, then the rate of money growth would be (1-r)*dM; and since the total amount of money is the integral of dM, the total money would be (1-r)*M; and inflation would be (1-r)*dM / (1-r)*M, or once again dM/M.
A fixed rate of expenditure would function the exact same way to a first order approximation. However, a more nuanced view shows that it gives higher inflation. When costs are fixed, the percentage of your income going to the gold sink will drop as you get better at the game and your income increases; so as players advance further, your gold sinks do less, and inflation accelerates. This is what a lot of traditional gold sinks in MMOs (like having to pay for armor and weapon upgrades from vendors) look like - and they end up making the inflation problem worse.
What you actually need to stabilize your economy is a gold sink that takes very little gold early on (because you need to accumulate some amount of money in the economy if it's going to function at all), but increases as people progress in the game, until eventually your gold sink takes 100% of gold drops in the very late game.
Taxes are incapable of doing this in any reasonable fashion. Instead your game needs to provide a service uniquely attractive to veteran players on which they will happily spend all their money. Gambling in D2 actually did a really good job of this pre-LoD; all your gold converts to items, which simultaneously sinks gold and generates more items (which pushes prices down as well) and thus fights inflation on both fronts. Crafting could also do this, in theory, but if gold is the limiting reagent in crafts then crafting materials are worthless.
D3 accomplishes this through the 15% transaction tax on the auction house - by taxing expenditure in the player market, instead of gold drops, they eventually sink 100% of excess gold drops through that tax alone. Essentially, as more gold drops, prices increase - which increase the total gold sunk - until you eventually hit an equilibrium price level where the inflationary pressure from gold drops exactly equals deflationary pressure from auction house taxes.
The other gold sinks honestly do not matter; if all you care about is keeping prices in line in the player market, the auction house is sufficient.
Apologies for this being a bit long winded, but at least it thoroughly covered the basics to the best of my knowledge. If you remember nothing else, remember it's not the quantity of gold sinks you have, but the quality, because a poor gold sink does nothing while a well designed one can curb inflation on its own.
Perfect versions of items are worth considering. Using D2 items, if I made an ebotdcb and it rolled low stats, it gets written off as "junk", but it is still a version of the best sword in the game.
Long term, when more of the perfect versions of items are being found and sold on the RMAH, the costs of these items will help to determine the cost of their non-perfect, "lesser" counterparts.
If too many perfect versions of an item are posted for reasonable prices, people will not want to pay real money for the non-perfect items, and thus it would be more advantageous to sell the non-perfect item on the GAH, then turn around and sell that gold on the RMAH, depending on the exchange rate, of course.
Curious if anyone in the community has worked on economic models before. I would love to know their insight into all the limitations and gold sinks that Diablo has in effect and how that will drive the value of virtual goods in D3.
None... Inflation will of course occur regardless as time goes along.
Gold sinks just reduce the rate at which it increases.
However, inflation only means that the value of gold per $ is less. Not much else. It is a natural occurrence in game and in life, and it in no shape of form would mean that the currency is going to collapse.
Rollback Post to RevisionRollBack
One is never hurt by being given additional choices, only by taking them away. A QUADRILLION MAGIC FIND is worthless if you can't kill shit!
Curious if anyone in the community has worked on economic models before. I would love to know their insight into all the limitations and gold sinks that Diablo has in effect and how that will drive the value of virtual goods in D3.
im working on something to record various prices per item / per gold over time to see how it goes from release to a stable situation. looking forward to having some fun making models and stuff based on the fluctuations in market pricing and stuff haha
hit me up if anyone wants to collaborate or discuss
edit: oh, i'm a rising senior majoring in economics
Rollback Post to RevisionRollBack
To post a comment, please login or register a new account.
Gold does appear to actually be a currency in Diablo 3. It was not a currency in Diablo 2. So unlike in Diablo 2 there's not a ton of pressure to move to an alternative currency regime immediately.
Currency supply is controlled by a negative feedback loop from the auction house fee. Essentially, as people accumulate more gold, prices rise on the auction house, which sinks gold more quickly. Napkin model tells me MV is more or less determined by the gold drop rate and is fixed.
Total goods available for sale should increase over time since goods are all durable and are not sunk in any way. So expect prices in gold to drop; more precisely, value will concentrate more and more in the very best drops as time goes on.
On the RMAH, you have an external currency. Demand for game goods will start high and decline with time. Supply of game goods will start low and increase with time. Expect prices on the RMAH to start sky high and quickly crater.
As prices on the RMAH crater, more items will be pushed onto the GAH. This will drive down the gold price of those items further, as more and better items compete for a more or less fixed pool of gold.
Short version: expect gold to hold value better than items.
the straightforward approach to profiting off of this would be to sell any items of decent quality immediately on the RMAH on release day.
Would a backwards approach being to sell everything for astronomical prices from the beginning on the GAH and save that currency for the point when the RMAH to GAH ratio favors gold?
Interesting notes on the GAH, and I expect new expansions to start this process all over again starting from infinity...
I also have no idea what this means "Napkin model tells me MV is more or less determined by the gold drop rate and is fixed". is there a wikipedia link that you could reference to make light of the jargon you use.
Thanks for the less heard insight on the game too!
I think this is a great place for us to start!
lets try and come up with a finite list of all the tools blizzard could use to change the economy.
heres what I see right now
-$ flat fee
-gold drop rate
-total item variation and item count
-the data they give us for last ten trades and last day of trades for gold to $
any more?
-Max auction values (capping all auction items at a specific $ amount)
-More endgame content that is item dependent (something like D2 ubers)
-Difficulty of farming
I don't think that will work out exactly like that - the big piles of gold that you'd want to sell high ticket items for just won't exist early on in the game's life cycle. It will take a while for people who will eventually have the big war chests to build up to that. Also, for very high ticket items, remember that the RMAH, as it stands, has a flat fee, while the GAH takes a percentage. Even if you just want gold, for a big enough ticket item it will make sense to sell on the RMAH and buy gold with the proceeds.
I'm mostly just writing down the equation of exchange (an accounting identity) and making some guesses about how terms will move in relation to each other. Start here:
http://en.wikipedia.org/wiki/Equation_of_exchange
'Real world' money doesn't decay either - if it did, it wouldn't be money.
You get a potentially endless supply of gold just from playing, for sure, but you also get an endless supply of items just from playing. The economy is based on exchanging one for the other. There isn't enough information just in stating those facts to make predictions about how things will play out.
Take a step back for a moment. What do we really mean when we say that there is inflation? Take a really basic two good model - there is 'money', and there is 'everything else'. At any point in time, there's some supply and demand equilibrium between money and everything else. Not at some future point in time, there will be a different equilibrium. Holding all else equal, if there is greater demand for everything else, or a lower supply of everything else, or a greater supply of money, or a lower demand for money, then the price of everything else in money will go up, and you will have inflation. If instead, holding all else equal, there is lower demand for everything else, or a higher supply of everything else, or a smaller supply of money, or a greater demand for money, then the price of everything else in terms of money will go down, and you will have inflation.
In reality you have a balance of all four of those factors moving somewhat independently, giving you the overall change in price level. Money never becomes worthless; if you have massive inflation or deflation, the instability, in reality, makes your currency stop functioning as money, and people migrate to new money.
That's the reason I'm being pretty conservative about the sorts of predictions I'm willing to make. The kinds of questions you're asking will absolutely have a large effect on what the actual price levels of various items in the game are. Will end game items cost 100,000 gold or a million gold or hundred million gold? We don't know. I can't possibly tell you that.
What we can do is look at the systems and make good predictions about the shape of the curves things will make. We may not know how much items will cost in an absolute sense, but it's not a stretch at all to say that there will be more top tier items, per player, on the auction house a year after launch than a week. How many more? I don't know. But you know that will trend up.
All these predictions are of that sort.
I am curious if there are observable and recordable data points that one person or a group of people could collect that the game denies you
obviously there is the gold to dollar exchange rate over time, that data doesn't seem to be available in the beta except for the past tent trades and the last days worth of trades. A larger data set could be used for pattern recognition.
The lack of an API for the AH's means that a full set or automated collection of data will never be available unless Blizzard just puts it out there (and monkeys might fly our of my butt).
So what could we as a community collect that would be worth analyzing?
...If I am getting it right, it seems, that the ultimate value in the game is Time, the more time you put in the more probabilistic gear and gold/$ you will get. You can buy time away from others through the dollar value which will change over time because of the supply and demand of in game items and gold. <--open to critique and correction! I am a complete laymen to economics and trying to get a basic understanding through rad peeps on the forums who are much much smarter than me.
Thanks dudes and dudettes
If you acquire all the items you ever want just by playing then you have no need for trade. You can stockpile a giant pile of gold for yourself. Since you do not trade with other players, none of the items you find or that giant pile of gold influences the player market in any way. As your presence in game consumes no rival resources, your net effect on the external player economy is zero.
If I *did* want to include you in monetary aggregates, I'd have to add your gold to M, and subtract the fact you don't spend it from V, which on the whole would have...no effect. I could include the items you found and kept in Y, but when you add that to the overall accounting balance you treat those items as being sold to yourself...so again, you're just putting a +C onto it.
This isn't terribly complicated. If you don't use the auction house you don't have an impact on the player economy. That should be obvious.
Money can absolutely be treated as a very particular kind of good - a good that provides a liquid medium of exchange, a divisible unit of account, and a store of value. You can say that you don't have a 'need' for money, but that's different from there being a demand for money - of which there is ample. It's an essential feature of all but the most trivial of economies. There is value in a medium of exchange that you can buy and sell with, because it overcomes the matching problem of buyers and sellers. Without money, you can only trade what you want to sell when you find a buyer who *also* has something to sell that you want to buy. With money, that transaction is broken in half - you only have to find a buyer, and then only have to find a seller, and they can be different people. This is a huge step forward in facilitating trade, and the more trade you have in your economy, the higher the demand for money for this purpose.
Money also needs to be a differentiable unit of account, so you don't have to find items of equal value to exchange. You can price everything in terms of very granular money, making it easy to 'make change' from a trade, to add extra or take a bit back. Not only can you find different buyers and sellers with money, you can also shift the size of the trades, splitting up big sales into smaller buys or smaller sales into one big buy. Also, as a store of value - you can sell without needing to buy immediately! If you're bartering, you can only sell when someone can buy; with money, you can sell whenever you have a buyer, and complete the other half of your exchange whenever you see fit.
So yeah, you don't strictly need money. You can't eat money. But money is what makes trade work, and was a huge, huge invention in the progression of mankind. You don't sell for money because you need money - you sell for money because you need other goods, and are confident that you can exchange your money for things you do need.
There's nothing fundamentally different about in-game money and US dollars or Euros or Renminbi or what have you. They are mediums of exchange that govern particular currency areas and goods bought and sold in those areas. While concerns about exchange rates and inflation / deflation are completely warranted and perfectly rational, I can assure you there is no reason to think gold in Diablo 3 will function any differently in the Diablo 3 economy than US Dollars do in the US economy.
Gold exchange rate should be pretty easy to monitor as a 3rd party; you'll have to write a bot to query the bid / ask price on gold frequently, but that should not be difficult and I would expect to see that data available fairly quickly after the game goes live.
I would really like to be able to track item sale prices on the auction house. The stats geek in me would really love to beat that data set over the head with MRA and build a dynamic price guide for items.
The lack of a specced API isn't fatal, it just means whatever software you use to gather data has to go through the human interface elements, which won't be nearly as efficient.
From a very high level view this isn't a bad way to look at it. You can buy people's labor (time) directly, or you can buy the goods they produced in that time, or you can buy machinery that represents an investment of time and that produces virtual time. In the real world there's also non-renewable resources in the equation, but even a lot of those you can measure in time (I'll joke with friends about a barrel of oil being equivalent to about 6 slave years). In the Diablo 3 universe it all starts from zero, however, so everything represents the output from time input, and yes, you're absolutely trading that time around in your exchanges.
I suggest this being a very bad idea to anyone out there, Warden (if warden is working for diablo as it does for WoW in catching bots) will most likely catch and suspend your account and you will be violation of the EULA. I believe blizzard to have this game on total lockdown for bots and modding and data collection because of the effects it would put into play on the economy.
I don't understand this question at all. Do people *need* to trade with each other? No. Do they want to? Yes. Is it rational to want to trade with each other? Yes, since it's a value-adding activity. Is using a currency to facilitate trade rational? Yes. Is a 15% tax rate enough friction to drive a sizable fraction of players to look for an alterative scheme? I doubt it.
I take it for granted that money has value, yes. I also take it for granted that if there is no money, people will invent money, spontaneously, to facilitate trade.
I do not take for granted that any particular currency will have value. Currencies collapse all the time and are replaced quickly with other currencies. There is no contradiction here. People will use a given currency as long as it functions as money; if it ceases to function as money, people will abandon it and find a new currency. We've seen this many times before (and are starting to see it in Greece today, actually, in the face of widespread deflation and currency shortages).
'Real money' can't get you everything either. Individual currencies are only predominant in localized currency areas, and cover only goods, services, and assets within that currency area. To purchase goods in another currency area, you have to exchange your currency for the local currency, with which you purchase what you want. This leads to exchange rates and relative currency valuations depending on the economic situation in each currency area.
Gold is no less a currency because you can't buy items on the RMAH any more than US Dollars are not a currency because you cannot use them to buy land in China.
Ah, now I see where you're coming from. Gold in Diablo 2 had a distinct problem. For the first year of the game, a Stone of Jordan had a very clear price, in gold - roughly 5 million gold.
There was no reasonable way for a player to trade for 5 million gold. Even the expansion only raised the gold cap to 2.5 million gold; it was a million if I recall correctly pre-expansion. How could you trade a SoJ for gold if you couldn't store that much gold, let alone carry it (which, I believe, is still limited to a few hundred thousand).
So gold in Diablo 2 wasn't a currency. You couldn't store it in large enough quantities to use it as a medium of exchange. So the currency was abandoned. People needed a new currency - something uniform and compact enough that you could store it reasonably enough to pay for things. Perfect gems were tried at first, but weren't valuable enough to overcome the storage constraints. Which is why people turned to Stones of Jordan. They weren't nearly as differentiable, which made it hard to make change (perfects kept some value for a while for this reason); and they did still have a storage constraint (though much better than any other option available). But they were uniform (giving them consistent value) and readily available (you could gamble them up for ~3.5 million gold each) - and there was nothing better available - so they became the de facto standard.
The *only* reason gold failed as a currency in Diablo 2 was the gold cap. If there was never any storage or carry limit on gold I am confident that people would still be trading items for (millions and millions) of gold today. Everything else about the economy of that game was set up sufficiently to support gold as a stable currency, but that one innocent mistake wrecked it.
I don't see a comparable mistake in Diablo 3, so I have no reason to expect gold not to function as money.
It really needs to be the materials; if the bottleneck is gold, then the materials will be essentially worthless.
But as of this second, nobody will know the outcome of this until the end game value on gold is determined.
Cool thing is.. We as the community are the ones that will be determining it.
I agree with you actually. Developers have said many times the one thing they need more of in later levels is gold.
As well as virtual value goes, if a lot of people play and get addicted to their characters, you could see thousands of people willing to buy items. We don't known yet as this has never been attempted on a AAA game from a AAA developer that is doing this legit.
In short, could be huge or could be meh.
Also, item values will vary with supply and demand. It will probably take a month before enough demanders get to level 60 for example, so items won't sell much before that. Likewise as I said above, is this going to be a game where a vast amount play to high levels or will most quit after they finish it on normal? Blizzard appears to really be pushing this as a game where you play to end game (unlike d3 where most quit after normal) and if they are successful that will offer a huge potential market....
http://www.wowhq.com
Since theres no data available the approach need to be purely theorycal. As I see it sinks truely play a important role. It's not an matter of taking coin out of the economic, as most people think. Independent of the gold drop rates and sinks (which is a inverse drop rate), inflation exists and existe at the same level (unless ALL the gold is actually sinked).
But i have the feeling the higest the utility for gold, lower the inflation. The actual model is bit hard to elaborate once you need to creat a economy with gold and this gold has both a means of trade and actual usage utility.
If you have something resembling a percentage tax rate (for example, 20% of gold drops go to repair bills, on average), then the economy behaves exactly the same way, albeit with a slightly lower price level. The rate of inflation is unchanged - if the tax rate is r, then the rate of money growth would be (1-r)*dM; and since the total amount of money is the integral of dM, the total money would be (1-r)*M; and inflation would be (1-r)*dM / (1-r)*M, or once again dM/M.
A fixed rate of expenditure would function the exact same way to a first order approximation. However, a more nuanced view shows that it gives higher inflation. When costs are fixed, the percentage of your income going to the gold sink will drop as you get better at the game and your income increases; so as players advance further, your gold sinks do less, and inflation accelerates. This is what a lot of traditional gold sinks in MMOs (like having to pay for armor and weapon upgrades from vendors) look like - and they end up making the inflation problem worse.
What you actually need to stabilize your economy is a gold sink that takes very little gold early on (because you need to accumulate some amount of money in the economy if it's going to function at all), but increases as people progress in the game, until eventually your gold sink takes 100% of gold drops in the very late game.
Taxes are incapable of doing this in any reasonable fashion. Instead your game needs to provide a service uniquely attractive to veteran players on which they will happily spend all their money. Gambling in D2 actually did a really good job of this pre-LoD; all your gold converts to items, which simultaneously sinks gold and generates more items (which pushes prices down as well) and thus fights inflation on both fronts. Crafting could also do this, in theory, but if gold is the limiting reagent in crafts then crafting materials are worthless.
D3 accomplishes this through the 15% transaction tax on the auction house - by taxing expenditure in the player market, instead of gold drops, they eventually sink 100% of excess gold drops through that tax alone. Essentially, as more gold drops, prices increase - which increase the total gold sunk - until you eventually hit an equilibrium price level where the inflationary pressure from gold drops exactly equals deflationary pressure from auction house taxes.
The other gold sinks honestly do not matter; if all you care about is keeping prices in line in the player market, the auction house is sufficient.
Apologies for this being a bit long winded, but at least it thoroughly covered the basics to the best of my knowledge. If you remember nothing else, remember it's not the quantity of gold sinks you have, but the quality, because a poor gold sink does nothing while a well designed one can curb inflation on its own.
Long term, when more of the perfect versions of items are being found and sold on the RMAH, the costs of these items will help to determine the cost of their non-perfect, "lesser" counterparts.
If too many perfect versions of an item are posted for reasonable prices, people will not want to pay real money for the non-perfect items, and thus it would be more advantageous to sell the non-perfect item on the GAH, then turn around and sell that gold on the RMAH, depending on the exchange rate, of course.
None... Inflation will of course occur regardless as time goes along.
Gold sinks just reduce the rate at which it increases.
However, inflation only means that the value of gold per $ is less. Not much else. It is a natural occurrence in game and in life, and it in no shape of form would mean that the currency is going to collapse.
A QUADRILLION MAGIC FIND is worthless if you can't kill shit!
im working on something to record various prices per item / per gold over time to see how it goes from release to a stable situation. looking forward to having some fun making models and stuff based on the fluctuations in market pricing and stuff haha
hit me up if anyone wants to collaborate or discuss
edit: oh, i'm a rising senior majoring in economics