7 ms·

A Virtual Weimar: Hyperinflation in a Video Game World

72 points by abtinf 13y ago

ordinary 13y ago
While it's true there was a significant upset in the Diablo 3 economy after the 1.0.8 patch came out, things are settling down again, and we're certainly not seeing the kind of delta that would make me call out a hyperinflation. If anything, the prices remind of me a bubble forming and consequently popping. Worse, the article is not just inaccurate, it's dishonest: it hides the data that contradicts it, despite being publicly available on a day-by-day basis. [1] is a wonderful resource. The eye-catching table near the end stops at peak prices. The prices since that peak are more-or-less hidden in the second half of an otherwise fairly unremarkable paragraph. We've seen a downward tendency, with prices moving back to pre-patch prices. We can illustrate that by looking up more recent data for the same examples as listed in the article: Radiant star amethyst: 20m before patch (april 29), high of 60m, 22m now. Radiant square ruby: 200k before patch, high of 528k, 266k now. Star emerald: 760k before patch, high of 1.6m, 770k now. Flawless square topaz: 3k before patch, high of 16k, 4.5k now. Tome of jewelcrafting: 1.2k before patch, high of 6.5k, 1.5k now.[1] All these items have a downward price trend, with the exception of the (relatively inconsequential) tome of jewelcrafting, which is stable. I would not be surprised to see these and other (harder to track) items reach month's end at roughly pre-patch prices. It's also worth mentioning at this point that the prices of each rank of gem is linked directly to the price of the rank below it. For example, a star gem costs 80k gold plus 3 radiant square gems of the same type to craft. This is an ingame feature. All gem prices are therefore functions of the price of flawless square gems, the highest rank that monsters can drop. The article mixes the prices of different gem ranks, which is either ignorance of the way the game's economy and mechanics works, or an intentionally and artificially pretense of diversity. It's a bad sign either way. Finally, and most important of all: the European market has been fairly stable throughout this whole debacle. The only difference between the European and the North American markets is that the latter suffered the (disastrous) bug described by the article, while the European market did not: Europe gets its patches 6-9 hours after North America does and the bug that it introduced was fixed before the patch hit Europe. This lends credit to the idea that all we're seeing here is the aftershocks of massive destabilization due to a simple bug. To me, the fact that prices seem to be returning to normal testifies to the relative stability of the Diablo 3 market, and long term price trends show the same. Huzzah, Blizzard! All in all, while interesting in a general sense, the article is flawed when it comes to the specific case it uses to promote its message, the turbulence of Diablo 3's economy over the last few weeks. This is not hyperinflation. [1] http://www.diablohub.com/price-tracker/ http://www.diablohub.com/price-tracker/
- Nursie 13y ago
  >> Worse, the article is not just inaccurate, it's dishonest: it hides the data that contradicts it, despite being publicly available on a day-by-day basis. [1] is a wonderful resource. What do you expect from mises.org? It has an agenda to push, and is relentless in doing so.
GotAnyMegadeth 13y ago
I remember when the money in Phantasy Star Online became worthless.
te_platt 13y ago
This makes me wonder to what extent online games could be used to model/explore/study economic and social theories. Here we have an example of hyperinflation happening. Could you reliably reproduce the results? Could you reliably adjust conditions to lead to expected results? If so how accurately? I think you could get results much better analyzing surveys because of the seriousness people put into games.
- idunno246 13y ago
  valve hired an economist do do something like that. http://blogs.valvesoftware.com/economics/it-all-began-with-a-strange-email/ http://blogs.valvesoftware.com/economics/it-all-began-with-a...
  - zokier 13y ago
    And EVE online has had economist onboard since 2007 http://www.scientificamerican.com/article.cfm?id=virtual-world-economists-on-real-economies http://www.scientificamerican.com/article.cfm?id=virtual-wor...
rosser 13y ago
There's a delicious irony to dollars being the hedge against virtual gold hyperinflation while real gold is the canonical hedge against dollar (hyper-) inflation...
n1ghtm4n 13y ago
Good to see the Austrian School weirdos at the Ludwig von Mises Institute have given up real-world economics and now just play video games. Hopefully in the virtual world their incredible track record of dead-wrong predictions will cause less harm.
- digitalengineer 13y ago
  (Okay, I'll bite). Von Mises was the mentor of Hayek. You know, the one who won the Nobel Price? F.A. Hayek... he has also led the way in attacking the mathematical models and the planning pretensions of the would-be "scientists," and in integrating economics into a wider libertarian social philosophy. Perhaps one reason is the evident and galloping breakdown of orthodox Keynesian "macroeconomics," which leads even the most hidebound economists to at least consider alternative theories and solutions. Everybody is drowning in debt, but hey let's just print more and we'll all get wealthy. http://mises.org/daily/4082 http://mises.org/daily/4082
  - jacques_chester 13y ago
    The problem is that the Ludwig von Mises Institute don't really sell Mises. They sell Rothbardianism, mostly. It's Rothbard acolytes and a series of increasingly more-anarcho-capitalist-than-thou writers. von Mises, who was actually more sensible and more measured than Rothbard, is just the name above the door. And von Hayek, who was more sensible and more measured than von Mises, is nowhere to be seen. At all. Except possibly as the subject of jokes. These days I find the Hayekian account of economics to be thoughtful and compelling [1]. I find the Rothbard account to be a breeding ground of ivory towers and scenes from Life of Brian. [1] http://chester.id.au/2012/12/07/review-the-essence-of-hayek-1/ http://chester.id.au/2012/12/07/review-the-essence-of-hayek-...
  - vwfwerwetqwerqw 13y ago
    Pointing out flaws in other analysis doesn't actually justify your own preferred explanations. See creationism.
  - nhaehnle 13y ago
jacques_chester 13y ago
Bear in mind that inflation of the money supply is not the only possible cause for price bubbles. I used to think so too (because hey, I used to read mises.org religiously). But it's not, as they say, that simple. Here's are some good pieces on bubbles generally, and how money supply by itself can't explain real-world phenomena: http://skepticlawyer.com.au/2013/05/15/bubble-trouble-not-an-easy-money-problem/ http://skepticlawyer.com.au/2013/05/15/bubble-trouble-not-an... http://skepticlawyer.com.au/2013/04/19/bubble-trouble-about-asset-booms-and-busts/ http://skepticlawyer.com.au/2013/04/19/bubble-trouble-about-...
- jacques_chester 13y ago
  s/here's are/here are/ You can see the join where I changed the sentence :(
- nhaehnle 13y ago
  Also, the focus on "the money supply" is a red herring. What matter for real world prices is supply and demand of real world goods. When you seriously study hyperinflationary periods, you realize that - unlike the story of this Diablo auction house bug - it's basically never a case of governments going crazy with printing money. There's always an underlying cause that came before the hyperinflation - usually a combination of a collapse of production capacity (in Zimbabwe, caused by drastic and ill thought-out economic reforms) and crushing foreign-curency denominated debt. See this paper, for example: http://papers.ssrn.com/sol3/papers.cfm?abstract_id=1799102 http://papers.ssrn.com/sol3/papers.cfm?abstract_id=1799102 It is true that excessive spending by anyone can lead to inflation (not just by government, since most money is actually "created" by private banks). However, what Austrian-types like the Mises people blind themselves to is the fact that the economy has two possible reactions to an increase in demand: There can be price increases, and there can be increases in production. To end up at their stated position, Austrians have to completely deny the possibility of production increases. It's no wonder they got basically all of their inflation-related predictions wrong for many years. They are unfortunately closer to religious dogma than serious economic analysis.
  - chii 13y ago
stormbrew 13y ago
The problem is that in order to make the experience fun for new players, the faucets have to be relatively open and the sinks not too punitive. I'm not sure how you can successfully balance a truly working virtual economy with making it fun for any but the most hardcore trader types, especially once real money comes officially into the equation.
- Ntrails 13y ago
  Eve Online is the MMO that has got this closest to right. The economy has been running for 10 years now and inflation, whilst existing, has not been rampant. Interestingly, there is not a RMAH, but there is a system to exchange ISK (gold) for subscription time that someone else bought. In other words the conversion of cash to ISK is facillitated - but not ISK to cash.
- kybernetyk 13y ago
  > successfully balance You could try to balance faucets with omnipresent unavoidable sinks aka. taxes. But how fun would it be to pay taxes in a game? "You loot 100 gold. Minus 30% gold drop tax. 10% dungeon upkeep tax and 15% financial transaction tax." :) So the thing is: As soon as you try to model a "real" economy (possibly with real money exchanges) you can't have artificial faucets and sinks. Even Eve Online - which has the most advanced game economy - struggles with inflation as there are too many faucets. For such an economy to work you need to have a steady amount of money circulating the economy. And the only source (and destination) for money for players has to be other players. But then you get a lot of other problems. Like what if a player hoards all the gold? Or if players just deactivate their accounts with significant gold sums on them? How do new players get their first gold? etc.
  - sbov 13y ago
    At least one problem with gold in games is that the management of large sums of gold is unrealistic. If you really had a huge portion of gold available in the game the storage and transportation of said gold would be both costly and risky. These concepts work mostly for fantasy themed MMOs though, not as much in something like Eve Online.
sharpneli 13y ago
"Furthermore, inflation is not simply an increase in the supply of money within an economy; it is the increase in that portion (if any) not backed by a commensurate increase in specie: most common in history, market commodities like gold or silver. " Either the author of that article is misenterpreting, I am misinterpreting or Austrian school of Economics is tad stupid. I'm hoping that the one at fault here is me. Basically even with increase of the amount of gold or silver you'd still get inflation, check what happened with Spain at 1500 or so when they got a huge influx of gold from the new world. You can also imagine what would happen if suddenly we'd get huge influx of new gold. As the amount of stuff in the world does not increase the amount you can buy with that new gold must decrease.
- nhaehnle 13y ago
  Actually, a huge influx of gold probably would not affect our economy today that much, because the total amount of gold is quite small relative to the size of the world economy, and gold is not used as a currency anyway. It would obviously destroy the price of gold, though. That's really one of the two main problem with inflation scare mongers: they don't usually talk about the scale of things. A gigantic influx of new money would certainly cause inflation. However, when the Austrians come out of their holes, one is usually talking about low single-digit percentages of GDP in spending, which is really not that much. The second main problem of Austrian inflationistas is that they deny the possibility that economic production can increase in response to an increase in spending.
- digitalengineer 13y ago
  There is a difference between inflation and hyperinflation. Hyperinflation is not too much money or not enough money to spend. Hyperinflation is a loss of confidence in the currency (ie, people immediately exchange the currency for something real). Money’s component as a store of value is what prevents inflationary episodes from happening. Imagine, if everyone believed that whatever cash that they hold would lose 50% of purchasing power by the end of the year, they would more than likely go ahead and spend it now instead of when 50% of the value is lost. That's the nature of hyperinflation. The other component necessary for hyperinflation is a competing currency. In Zimbabwe the locals switched to dollars when their native gov’t currency failed. From the article: Most of the gold generated by the ruthlessly productive, rapidly adapting bots found its way to third party vendors in a black market which undercut the prices in the sanctioned, in-game auction houses.
mihaifm 13y ago
Nicely written article, but it doesn't seem to factor in the fact that the item supply is also increasing over time. Since players are only equiping stuff only once in a while, the items have nowhere to go, thus decreasing in value. While players (or bots) farm for gold, they're also farming for items. It would be interesting to see how item prices evolved over time...it might be that only prices for high end items have skyrocketed, while decent low end items are almost free.
- muraiki 13y ago
  The experience of a friend and myself when we returned to D3 about three months ago was that for a relatively small amount of money (250k gold perhaps) you could get gear good enough to play on some of the lower monster power levels of Inferno. Inferno is the hardest difficulty level, and there's 10 settings of monster power to tailor the difficulty to your preference / gear. I think that he played on MP3 or 4, but I usually stuck around at MP 1-2. However, I stopped playing because I couldn't even begin to afford the gear to go higher. I figure if I wait a few more months then the gear that I'd need for MP5 shouldn't be so expensive anymore. As such, even though my gold is technically becoming worth less in terms of top tier items, it's becoming worth more in terms of decent low-mid level items. As such, I think that the D3 situation is more complex than just hyperinflation.

72 points by abtinf 13y ago

While it's true there was a significant upset in the Diablo 3 economy after the 1.0.8 patch came out, things are settling down again, and we're certainly not seeing the kind of delta that would make me call out a hyperinflation. If anything, the prices remind of me a bubble forming and consequently popping. Worse, the article is not just inaccurate, it's dishonest: it hides the data that contradicts it, despite being publicly available on a day-by-day basis. [1] is a wonderful resource. The eye-catching table near the end stops at peak prices. The prices since that peak are more-or-less hidden in the second half of an otherwise fairly unremarkable paragraph. We've seen a downward tendency, with prices moving back to pre-patch prices. We can illustrate that by looking up more recent data for the same examples as listed in the article: Radiant star amethyst: 20m before patch (april 29), high of 60m, 22m now. Radiant square ruby: 200k before patch, high of 528k, 266k now. Star emerald: 760k before patch, high of 1.6m, 770k now. Flawless square topaz: 3k before patch, high of 16k, 4.5k now. Tome of jewelcrafting: 1.2k before patch, high of 6.5k, 1.5k now.[1] All these items have a downward price trend, with the exception of the (relatively inconsequential) tome of jewelcrafting, which is stable. I would not be surprised to see these and other (harder to track) items reach month's end at roughly pre-patch prices. It's also worth mentioning at this point that the prices of each rank of gem is linked directly to the price of the rank below it. For example, a star gem costs 80k gold plus 3 radiant square gems of the same type to craft. This is an ingame feature. All gem prices are therefore functions of the price of flawless square gems, the highest rank that monsters can drop. The article mixes the prices of different gem ranks, which is either ignorance of the way the game's economy and mechanics works, or an intentionally and artificially pretense of diversity. It's a bad sign either way. Finally, and most important of all: the European market has been fairly stable throughout this whole debacle. The only difference between the European and the North American markets is that the latter suffered the (disastrous) bug described by the article, while the European market did not: Europe gets its patches 6-9 hours after North America does and the bug that it introduced was fixed before the patch hit Europe. This lends credit to the idea that all we're seeing here is the aftershocks of massive destabilization due to a simple bug. To me, the fact that prices seem to be returning to normal testifies to the relative stability of the Diablo 3 market, and long term price trends show the same. Huzzah, Blizzard! All in all, while interesting in a general sense, the article is flawed when it comes to the specific case it uses to promote its message, the turbulence of Diablo 3's economy over the last few weeks. This is not hyperinflation. [1] http://www.diablohub.com/price-tracker/ http://www.diablohub.com/price-tracker/

Nursie 13y ago
>> Worse, the article is not just inaccurate, it's dishonest: it hides the data that contradicts it, despite being publicly available on a day-by-day basis. [1] is a wonderful resource. What do you expect from mises.org? It has an agenda to push, and is relentless in doing so.

GotAnyMegadeth 13y ago

I remember when the money in Phantasy Star Online became worthless.

te_platt 13y ago

This makes me wonder to what extent online games could be used to model/explore/study economic and social theories. Here we have an example of hyperinflation happening. Could you reliably reproduce the results? Could you reliably adjust conditions to lead to expected results? If so how accurately? I think you could get results much better analyzing surveys because of the seriousness people put into games.

idunno246 13y ago
valve hired an economist do do something like that. http://blogs.valvesoftware.com/economics/it-all-began-with-a-strange-email/ http://blogs.valvesoftware.com/economics/it-all-began-with-a...
- zokier 13y ago
  And EVE online has had economist onboard since 2007 http://www.scientificamerican.com/article.cfm?id=virtual-world-economists-on-real-economies http://www.scientificamerican.com/article.cfm?id=virtual-wor...

rosser 13y ago

There's a delicious irony to dollars being the hedge against virtual gold hyperinflation while real gold is the canonical hedge against dollar (hyper-) inflation...

n1ghtm4n 13y ago

Good to see the Austrian School weirdos at the Ludwig von Mises Institute have given up real-world economics and now just play video games. Hopefully in the virtual world their incredible track record of dead-wrong predictions will cause less harm.

digitalengineer 13y ago
(Okay, I'll bite). Von Mises was the mentor of Hayek. You know, the one who won the Nobel Price? F.A. Hayek... he has also led the way in attacking the mathematical models and the planning pretensions of the would-be "scientists," and in integrating economics into a wider libertarian social philosophy. Perhaps one reason is the evident and galloping breakdown of orthodox Keynesian "macroeconomics," which leads even the most hidebound economists to at least consider alternative theories and solutions. Everybody is drowning in debt, but hey let's just print more and we'll all get wealthy. http://mises.org/daily/4082 http://mises.org/daily/4082
- jacques_chester 13y ago
  The problem is that the Ludwig von Mises Institute don't really sell Mises. They sell Rothbardianism, mostly. It's Rothbard acolytes and a series of increasingly more-anarcho-capitalist-than-thou writers. von Mises, who was actually more sensible and more measured than Rothbard, is just the name above the door. And von Hayek, who was more sensible and more measured than von Mises, is nowhere to be seen. At all. Except possibly as the subject of jokes. These days I find the Hayekian account of economics to be thoughtful and compelling [1]. I find the Rothbard account to be a breeding ground of ivory towers and scenes from Life of Brian. [1] http://chester.id.au/2012/12/07/review-the-essence-of-hayek-1/ http://chester.id.au/2012/12/07/review-the-essence-of-hayek-...
- vwfwerwetqwerqw 13y ago
  Pointing out flaws in other analysis doesn't actually justify your own preferred explanations. See creationism.
- nhaehnle 13y ago

jacques_chester 13y ago

Bear in mind that inflation of the money supply is not the only possible cause for price bubbles. I used to think so too (because hey, I used to read mises.org religiously). But it's not, as they say, that simple. Here's are some good pieces on bubbles generally, and how money supply by itself can't explain real-world phenomena: http://skepticlawyer.com.au/2013/05/15/bubble-trouble-not-an-easy-money-problem/ http://skepticlawyer.com.au/2013/05/15/bubble-trouble-not-an... http://skepticlawyer.com.au/2013/04/19/bubble-trouble-about-asset-booms-and-busts/ http://skepticlawyer.com.au/2013/04/19/bubble-trouble-about-...

jacques_chester 13y ago
s/here's are/here are/ You can see the join where I changed the sentence :(
nhaehnle 13y ago
Also, the focus on "the money supply" is a red herring. What matter for real world prices is supply and demand of real world goods. When you seriously study hyperinflationary periods, you realize that - unlike the story of this Diablo auction house bug - it's basically never a case of governments going crazy with printing money. There's always an underlying cause that came before the hyperinflation - usually a combination of a collapse of production capacity (in Zimbabwe, caused by drastic and ill thought-out economic reforms) and crushing foreign-curency denominated debt. See this paper, for example: http://papers.ssrn.com/sol3/papers.cfm?abstract_id=1799102 http://papers.ssrn.com/sol3/papers.cfm?abstract_id=1799102 It is true that excessive spending by anyone can lead to inflation (not just by government, since most money is actually "created" by private banks). However, what Austrian-types like the Mises people blind themselves to is the fact that the economy has two possible reactions to an increase in demand: There can be price increases, and there can be increases in production. To end up at their stated position, Austrians have to completely deny the possibility of production increases. It's no wonder they got basically all of their inflation-related predictions wrong for many years. They are unfortunately closer to religious dogma than serious economic analysis.
- chii 13y ago

stormbrew 13y ago

The problem is that in order to make the experience fun for new players, the faucets have to be relatively open and the sinks not too punitive. I'm not sure how you can successfully balance a truly working virtual economy with making it fun for any but the most hardcore trader types, especially once real money comes officially into the equation.

Ntrails 13y ago
Eve Online is the MMO that has got this closest to right. The economy has been running for 10 years now and inflation, whilst existing, has not been rampant. Interestingly, there is not a RMAH, but there is a system to exchange ISK (gold) for subscription time that someone else bought. In other words the conversion of cash to ISK is facillitated - but not ISK to cash.
kybernetyk 13y ago
> successfully balance You could try to balance faucets with omnipresent unavoidable sinks aka. taxes. But how fun would it be to pay taxes in a game? "You loot 100 gold. Minus 30% gold drop tax. 10% dungeon upkeep tax and 15% financial transaction tax." :) So the thing is: As soon as you try to model a "real" economy (possibly with real money exchanges) you can't have artificial faucets and sinks. Even Eve Online - which has the most advanced game economy - struggles with inflation as there are too many faucets. For such an economy to work you need to have a steady amount of money circulating the economy. And the only source (and destination) for money for players has to be other players. But then you get a lot of other problems. Like what if a player hoards all the gold? Or if players just deactivate their accounts with significant gold sums on them? How do new players get their first gold? etc.
- sbov 13y ago
  At least one problem with gold in games is that the management of large sums of gold is unrealistic. If you really had a huge portion of gold available in the game the storage and transportation of said gold would be both costly and risky. These concepts work mostly for fantasy themed MMOs though, not as much in something like Eve Online.

sharpneli 13y ago

"Furthermore, inflation is not simply an increase in the supply of money within an economy; it is the increase in that portion (if any) not backed by a commensurate increase in specie: most common in history, market commodities like gold or silver. " Either the author of that article is misenterpreting, I am misinterpreting or Austrian school of Economics is tad stupid. I'm hoping that the one at fault here is me. Basically even with increase of the amount of gold or silver you'd still get inflation, check what happened with Spain at 1500 or so when they got a huge influx of gold from the new world. You can also imagine what would happen if suddenly we'd get huge influx of new gold. As the amount of stuff in the world does not increase the amount you can buy with that new gold must decrease.

nhaehnle 13y ago
Actually, a huge influx of gold probably would not affect our economy today that much, because the total amount of gold is quite small relative to the size of the world economy, and gold is not used as a currency anyway. It would obviously destroy the price of gold, though. That's really one of the two main problem with inflation scare mongers: they don't usually talk about the scale of things. A gigantic influx of new money would certainly cause inflation. However, when the Austrians come out of their holes, one is usually talking about low single-digit percentages of GDP in spending, which is really not that much. The second main problem of Austrian inflationistas is that they deny the possibility that economic production can increase in response to an increase in spending.
digitalengineer 13y ago
There is a difference between inflation and hyperinflation. Hyperinflation is not too much money or not enough money to spend. Hyperinflation is a loss of confidence in the currency (ie, people immediately exchange the currency for something real). Money’s component as a store of value is what prevents inflationary episodes from happening. Imagine, if everyone believed that whatever cash that they hold would lose 50% of purchasing power by the end of the year, they would more than likely go ahead and spend it now instead of when 50% of the value is lost. That's the nature of hyperinflation. The other component necessary for hyperinflation is a competing currency. In Zimbabwe the locals switched to dollars when their native gov’t currency failed. From the article: Most of the gold generated by the ruthlessly productive, rapidly adapting bots found its way to third party vendors in a black market which undercut the prices in the sanctioned, in-game auction houses.

mihaifm 13y ago

Nicely written article, but it doesn't seem to factor in the fact that the item supply is also increasing over time. Since players are only equiping stuff only once in a while, the items have nowhere to go, thus decreasing in value. While players (or bots) farm for gold, they're also farming for items. It would be interesting to see how item prices evolved over time...it might be that only prices for high end items have skyrocketed, while decent low end items are almost free.

muraiki 13y ago
The experience of a friend and myself when we returned to D3 about three months ago was that for a relatively small amount of money (250k gold perhaps) you could get gear good enough to play on some of the lower monster power levels of Inferno. Inferno is the hardest difficulty level, and there's 10 settings of monster power to tailor the difficulty to your preference / gear. I think that he played on MP3 or 4, but I usually stuck around at MP 1-2. However, I stopped playing because I couldn't even begin to afford the gear to go higher. I figure if I wait a few more months then the gear that I'd need for MP5 shouldn't be so expensive anymore. As such, even though my gold is technically becoming worth less in terms of top tier items, it's becoming worth more in terms of decent low-mid level items. As such, I think that the D3 situation is more complex than just hyperinflation.