Sunday, June 22, 2008

Learning Economics in World of Warcraft

Tony Forster

Licensed Creative Commons, share alike, non commercial, by attribution


World of Warcraft is an online computer game with 10 million subscribers (as at June 08). It has an ingame free market economy which allows for market research and the teaching of economic theory by hands on activities. This paper describes how the ingame auction house could be used to teach economic theory with hands on activities to explore the relationship between supply, demand and price.

About the game

World of Warcraft (commonly known as WoW) is a massively multiplayer online role-playing game (MMORPG). People control a character avatar within a persistent game world, exploring the landscape, fighting monsters, performing quests, building skills, and interacting with NPCs, as well as other players. The game rewards success with in-game money, items, experience and reputation, all of which in turn allow players to improve their skill and power. Players can level up their characters from level one to the next.

World of Warcraft uses server clusters (known as "realms") to allow players to choose their preferred gameplay type and to allow the game to support as many subscribers as it does. There are four types of realms: Normal (also known as PvE or player versus environment), PvP (player versus player), RP (a roleplaying Normal/PvE server) and RP-PvP (roleplaying PvP server):

When creating a character in World of Warcraft, the player can choose from ten different races in two factions: Alliance and Horde. Race determines the character's appearance, starting location, and initial skill set, called "racial traits".

* The Alliance currently consists of Humans, Night Elves, Dwarfs, Gnomes and Draenei.
* The Horde currently consists of Orcs, Tauren, Undead, Trolls and Blood Elves.

The game has nine character classes that a player can choose from, though not all classes are available for each race. During the course of playing the game, players may choose to develop side skills for their character(s). These non-combat skills are called professions.

To play, there is an initial purchase price and a monthly fee as shown below

Suggested Retail Price Monthly Fee

Europe E14.99 E11-E13

United Kingdom £9.99 £7.70-£9

North America/Oceania US$20 $13-$15

Video game stores commonly stock the trial version of World of Warcraft in DVD form priced at A$2 or \2 including VAT, which include the game and 14 days of gameplay.

Derived from (retrieved 30/4/08)

About in game trading

To be able to trade effectively, it is estimated that the character would need to be levelled up to level 10. The trial version may not allow enough time for this, it is also uncertain whether the trial version has the full functionality including the ability to use the auction house.

Estimated playing time to reach various levels is shown below.

Level 10, 4 hours playing time

Level 25, 25 hours playing time

Level 55 200 hours playing time

WoW has 10 million subscribers and has a market economy where in game items can be traded for in game currency. There are approximately 150 independent servers or realms, so within one realm there is a market economy of approximately 70,000 players. Goods are traded at an auction house with market forces determining prices. This makes WoW an ideal testbed for testing economic theories of free market behaviour.

Tradeable goods are best achieved through the professions. For example, with professions of herbalism and alchemy, the trader can collect herbs (known as materials or mats) and transform them into potions (pots). There is a market for both the materials and the potions. For example, the player could search for Mageroyal and Stranglekelp and either transform them into Lesser Mana Potion which could be traded or just trade the materials direct. You will find the online database invaluable in helping level up and finding materials.

Alternative professions such as mining and engineering may be better, players suggest that mining ores such as copper give easy access to tradeable commodities

The player is able to test theories of supply and demand and influence market prices by trading on the open market.


  • Discuss, how well does the market mirror the real world?

  • What determines the supply of money?

  • Is there any equivalent to monetary and fiscal interventions which are practised in real economies?

  • If you were Blizzard (the makers of WoW) what market interventions might you do?

  • Observe long term price trends e.g. at

  • Is there any long term inflationary or deflationary trend?

  • Observe the trading history of some commodity which you could trade. Can you estimate the elasticity of supply and demand.

  • Having estimated market elasticity, predict the effect of your trade on the market.

  • Do a large amount of trading of one commodity. Are the price effects what you predicted? If not, why?

  • Classic market theory works on the assumption that the market is perfectly informed. How well informed is the market?

  • How is the market informed?

  • Are there delayed effects as the market adjusts? What do you think the mechanism is?

Useful references invaluable in helping level up and finding materials market information and statistics Derek's Gold Mastery Guide World of Warcraft Gold Farm be Rich (auction house) World of Warcraft auction house

Sample trades

I am told that for most items, they are sold at the buyout price, rather than the auction price. The item chosen for trade was Stranglekelp. It was commanding a good price at auction.

Typical WoW wide prices were listed at at 20s to 30 s. (100 copper = 1 silver, 100 silver = 1 gold)

The auction house listed 19 units for sale at 21s to 173s I had no information as to whether it was actually selling at these prices or in what volumes. I listed at a number of units at a range of prices.




20 May

3 @ 16s

3 @ 30

3 @ 33

6 @ 40

5 @ 51

10 @ 52

5 @ 54

21 May

25 @ 53

10 @ 80

10 @ 100s

At this stage I realised that Stranglekelp was selling as high as 153s for a patient seller and my cheaper sales were being bought and immediately relisted at a higher price.




22 May

11 @ 11s

90 @ 100

23 May

6 @ 80

40 @ 100

20 @ 21

60 @ 100

12 @ 91

20 @ 95

4 @ 60

4 @ 125

26 May

100 @ 1.00

29 May

74 @ 40

148 @ 100

2 June

19 @ 75

20 @ 100

95 @ 50

I adopted this strategy and was able to maintain a market price of 100s per Stranglekelp and sell most if not all the market volume. The volume was approximately 460 units in 11 days or 40 units a day. The market seemed to be unresponsive to price. Purchasers would buy at any price so that high priced stock would eventually clear when no cheaper stock was available. Sellers would sell much more cheaply than the market could support.

Labels: , , ,

Friday, June 13, 2008

Sharing with Jabber

I described how you could evaluate the activity sharing features of the OLPC using the Live CD. The trouble is, sharing is boring unless there's someone to share with on your LAN.

You can connect to the wider network using a Jabber server. You can change what server your OLPC is connected to using the terminal activity and running control panel from the command line.

$sugar-control-panel -s jabber jabber.server.url
using a jabber server url from the list of community jabber servers or you could set up your own jabber server.

Now there are people to activity share with.

Labels: ,

Monday, June 09, 2008

Sharing with Sugar

Sugar is the operating system for the $100 laptop or OLPC. It is designed as a constructionist learning environment. Its most notable feature is enabling collaboration through application sharing on the OLPC's mesh wireless network.

Though earlier Live CDs may not application share, I just tried out the most up to date image
XO-LiveCD_080321.iso from

Download the ISO image and burn it to CD, then put it in a PC and boot. It boots ok if you have >256M RAM

Though the Live CD doesn't seem to support WiFi, I was able to use ethernet on the PC and WiFi on my OLPC and application share. That means that the educational advantages of Sugar's application sharing can be evaluated without needing multiple OLPCs.

Labels: ,

Home PCs lower education results?

A new study with very strong credentials from Ofer Malamud (University of Chicago and NBER) and Cristian Pop-Eleches (Columbia University, Harvard University’s BREAD, and NBER) suggests that home PC ownership lowers educational outcomes.

"This result is large and statistically significant … To summarize, the evidence on GPA, college plans and the school behavior grade presented in Table 5 suggests that, if anything, computer ownership has a negative impact on child academic and behavioral outcomes. …"

This study uses regression discontinuity analysis. The Romanian government offered vouchers to low income families to buy computers. The idea is that for the two communities, those above the qualification income and those below, two regression lines can be drawn. At the income breakpoint, the two line ends represent two identical families in all respects except access to the vouchers.

If you examine the graphs at the end of the report, you will see a serious problem with the use of regression discontinuity which I believe makes the study invalid. The two regression lines for these 3 indicators have opposite slopes.

For regression discontinuity to be valid for straight line regression lines then the underlying distribution, pre-treatment, should be linear:

"The conclusion is a simple one. If the true model is curved and we fit only straight-lines, we are likely to conclude wrongly that the treatment made a difference when it did not. This is a specific instance of the more general problem of model specification."

This is how the data should look if linear regression discontinuity is to be used:

And here, roughly, is how the study's graphs of GPA, college plans and the school behavior grade look:

And to show how the discontinuity in the two regression lines does not mean anything for the two populations, here is the underlying data for the two lines above, a continuous curve.

On this basis, I believe the statistical technique used in the study is not valid.

Read more about the implications of the study at

Labels: , , , ,