杏吧原创

Edward Castronova has got to be the only person in the world to rescue a failing career by calculating the gross domestic product of a place that doesn鈥檛 exist. Just four years ago he was teaching at a college that didn鈥檛 grant PhDs, no one knew his research, and his wife lived out of town. Then one lonely night when Castronova was playing the online game Everquest, he noticed something that piqued his economist curiosity. Normally gamers derive status or virtual goodies through their skill; here were players with great status and wealth but no idea at all of how to play. It turned out they had paid real-world cash to buy virtual artefacts or characters. Castronova decided to put a rough figure on the economic activity of this synthetic world. It was startling: Everquest鈥檚 GDP per capita nearly equalled Russia鈥檚. Almost overnight, Castronova became a leading authority on the implications of large-scale online games. He reckons that leakage from the real world makes virtual worlds ideal for testing economic or social theories. And, as he told Alison George, he is about to build his own game.

To be good for modelling economic or social theories, do these 鈥渟ynthetic worlds鈥� you work on have to be big and complex?

They are. They involve massive multiplayer online role-playing games, in which a large number of players interact with each other in a virtual world. The biggest is World of Warcraft, which has 6.5 million users worldwide. The populations of these worlds are growing exponentially. Right now it looks as if the user numbers are growing according to Moore鈥檚 law and doubling roughly every two years.

You鈥檝e just received a $240,000 grant from the MacArthur Foundation to build your own game. What are you going to do with it?

I鈥檓 working on the idea of using virtual worlds as Petri dishes to conduct experiments about social-science issues. We鈥檙e going to develop a fantasy world based on the plays of Shakespeare. I want to use it to test an economic theory called the quantity theory of money supply, which is about how the supply of money affects inflation 鈥� but this role will be invisible to the user of the game, who will be playing for fun. There will be two parallel worlds, which will be exactly the same except for the amount of money in them. So you will have two exactly equal virtual macro-economies in which you can vary one thing, something which is impossible to do in the real world. I think that鈥檚 an unprecedented research opportunity.

What made you think you could manipulate virtual worlds to test economic theories?

I was playing games like Everquest, which is an online fantasy role-playing game a bit like Dungeons and Dragons, when I stumbled across this area of the game where there were hundreds of people collecting around torches, like an open-air bazaar, to buy and sell virtual items. I noticed how many players were prepared to pay real money to buy these virtual items, so I decided to write a paper about it. Instead of the World Bank report on Poland, for instance, it would be Castronova鈥檚 report on Everquest. I would treat Everquest like a real country and try to collect all the standard macroeconomic statistics: GDP, inflation, productivity and wages. The more information I got, the bigger the economy of Everquest appeared to be.

What did you find?

I calculated that a single worker could make about $3.50 an hour, if the fruits of their labour were converted to real money. According to that measure, a citizen of Everquest could produce as much wealth as someone living in, say, Bulgaria, and four times as much as someone in China: this game world is small but hosts a lot of productive economic activity. At the time of that research, in 2001, I predicted that annual revenues in virtual worlds would top $1.5 billion by 2004.

Today it must be even bigger, mustn鈥檛 it?

I haven鈥檛 repeated the study. It got reported as 鈥渁n economist says that a fantasy economy is bigger than the economy of Bulgaria鈥�, which it isn鈥檛. So I have shied away from that number. What I鈥檝e done since then is to track the health of the virtual economies, keeping an eye on exchange rates and currencies. For example, today you can buy a gold piece to use in World of Warcraft for 9 cents. I have seen auctions for $10,000 for a set of player accounts to use in Everquest, but I don鈥檛 know if they sold. I spent $500 myself for an account, just to see if it was real. It was.

Why do people pay hard cash for fantasy items?

In the fantasy world, if you own a horse you go faster: this makes for a better game, and you can travel more easily to see your friends in another location, just like you do in the real world. It鈥檚 also about reputation and prestige 鈥� 鈥渕y horse is bigger than yours鈥�. But when you start out in the virtual world, you start with nothing and it鈥檚 hard work to build up capital. That鈥檚 why people are prepared to pay real dollars for virtual items.

Have you ever sold anything that you鈥檝e built up in online games?

No 鈥� I鈥檝e made virtual money but not real money. I鈥檓 not that good a gamer and I鈥檓 also in two minds about it. As an economist I understand why this trading happens but as a gamer and someone interested in human culture I think it鈥檚 deplorable that people use real money to get ahead in a game. I hate that. One of the fundamental design features of these games is that everybody starts out with nothing. The idea of a world of equal economic opportunity is a long-standing human dream. And it exists, right there, if only these jerks would stop spending real money to get ahead.

鈥淚t鈥檚 deplorable that people use real money to get ahead in a game鈥�

Do you think it was you that put these virtual worlds on the map?

When I published my economic analysis of this 鈥淐yberian frontier鈥� in 2001, I told the readers that I felt like an explorer and that I鈥檇 just discovered a new world. And I do feel like Captain Cook, to some extent. If you told me in 1995, when I was working on the welfare state, that I was going to become a video games scholar, I would have leapt for joy but not believed you. But the subject suddenly became salient at the same time as I was writing about it. That, combined with the conservatism of most researchers in the social sciences, left a huge chunk of academic activity open for me to explore.

How was your research viewed by other economists when you started working on virtual worlds?

My first efforts to publish papers about this in normal economic journals were turned down. They weren鈥檛 convinced that virtual reality mattered at all.

Should virtual worlds be policed, like the real world is?

It鈥檚 a dilemma. Within a game, if I trick you into giving me your goods, that鈥檚 one thing. But if I hack into the server, then that鈥檚 considered a crime. A big percentage of computer crime in Korea is people hacking into online games and switching ownership of items. If they could claim they were acting within the rules of the game, then the law gets really weird. I think it鈥檚 going to take an explicit statement on the part of real-world governments to wall these places off and surround them with rules and regulations. The responsible thing to do is to seal off that economy from the real economy. Otherwise people will increasingly use these spaces for commercial activity, even crime, instead of the pure fantasy for which they are intended.

Do you think politicians will have to find ways of dealing with this?

There have already been court cases in Korea and China on these issues. Suppose there鈥檚 a breakdown in the server, and somebody logs in one day and all their equipment in the game is gone. They complain to the company, which says that they have to start over again. That would cost someone a lot of money and they could sue. Could a government emerge within those spaces to address these issues? Maybe. If it did would it be another nation state? That鈥檚 the brave new world that I can鈥檛 get my mind around.