Edward Castronova on the Exodus to the Virtual World
Jan 7 2008

Edward Castronova, of Indiana University and author of Exodus to the Virtual World, talks about his provocative thesis that a growing number of people around the world will be spending more and more time playing multiplayer games in virtual reality both as a form of escape and as a search for meaning. He talks with EconTalk host Russ Roberts about how this trend might affect government, religion, and our happiness.

Megan McArdle on Internet Shaming and Online Mobs
Author and journalist Megan McArdle of Bloomberg View talks with EconTalk host Russ Roberts about how the internet has allowed a new kind of shaming via social media and how episodes of bad behavior live on because Google's memory is...
Bruce Schneier on Power, the Internet, and Security
Bruce Schneier, author and security guru, talks with EconTalk host Russ Roberts about power and the internet. Schneier argues that the internet enhances the power of the powerless but it also enhances the power of the powerful. He argues that...
Explore audio transcript, further reading that will help you delve deeper into this week’s episode, and vigorous conversations in the form of our comments section below.


Jan 7 2008 at 6:02pm

I have played World of Warcraft (and other MMORPGs) for a number of years.

Economics actually plays out incredibly well in the Auction Houses in this game. In these Auction Houses, people offer items for E-Bay style auctions, including a Buy-It-Now option.

This is good economic fun in itself. However, the real fun comes when you start hearing people complain about how people always offer their wares at prices that are too high. “Why don’t these people sell their products for a reasonable amount” is a common complaint.

The response is always a quick instruction on how prices are determined in a micro-economic environment.

“Why don’t you just go out and earn it instead of buying it” is the most common response, followed by “If it’s not worth it, it won’t sell and the seller loses his deposit.”

Personally, I find the Auction House one of the most interesting parts of the game.

Jan 7 2008 at 6:55pm

The topic of floating cars came up about 10 minutes from the end. The context was feature creep in software. However, for those of you that are addicted to surfing the Internet like Russ Roberts, you might want to go out of context and Google “Amphicar,” an amphibious car from the 1960’s.

Jan 7 2008 at 8:42pm

The need to escape reality goes hand-in-hand with the human ability to develop intellectual abstraction. Anyone who engages in any kind of intellectual pursuit enters a virtual world of ideas. In fact, it is often argued that the best way to learn math is to think of mathematical objects as actual, “real”, objects, and interact with them, almost at an emotional level. So one could say that we’ve always had games, societies of gamers, virtual worlds, etc… except most of the action use to take place in our brains, now it can more quickly be displayed on the internet.

Bruce Boston
Jan 7 2008 at 11:04pm

Love the podcast today! Dr. Castronova has been doing some really fun things with his research on Virtual Worlds, and for anyone that wants to know more, the blog he is connected with, http://terranova.blogs.com/ has a number of academic professionals who have documented a ton of stuff. Teranova is also pretty well known among hard-core gamers and game developers who have responded to much of this on the blog so you can really get a full picture of what’s going on.

One of the key efforts that I think Dr. Castronova really does a good job at, in almost every discussion I’ve listened to of his, is the re-examination of what ‘Economics’ both the study and the practice is really about.

Personally, I think it will lead to a reexamination of what ‘human needs’ really are. Virtual Worlds make the perfect place to study them because they bring to light how small a role ‘physical goods’ play in the satisfaction of most human needs, or at least most of the human needs that are unfulfilled once you have basic food, shelter and health care.

In the business world, I am often reminded of the following example.

No one needs a screwdriver, what they want is a screw-driven. No one needs a screw-driven, what they want is a bookshelf. No one needs a bookshelf, what they want is a place to store their books. No one needs a place to store their books, what they want is their books. No one needs their books, what they want is the cooking recipes within the books. No one needs the cooking recipes within their books, what people want is the cake they bake with the recipe. No one needs the cake, what they want is the social experience they share with friends. No one needs social experiences they have with friends, what they want is the social community that gets built overtime through these experiences, etc, etc, etc, etc.

So, you can basically take every ‘need’ outside of basic water, calories, oxygen, shelter, and basic health care, and play economic ring around the roses with it, jumping from need to want, to need to want, to need to want, over and over, and over again, until you find a nice neurological need to land one, which is again, very easy to jump off of.

What Virtual Economies allow you to do, having been the economist in a small virtual world at one time, is speed-up the rate at which data is collected around what works and what doesn’t.

Want to know the exact number of seconds between each of the 10 sips that a person gets in their virtual drink? Data is all there. Want to know the effect of having 3 people of the opposite sex text chatting with the person? Data is all there. Want to know the effect of adding voice chat? Data is all there. Want to know the effect of adding a card game? Data is all there. Want to know the effect of adding music? Data is all there. The effect of lighting? Color? Stimuli? Random events? Consistency? etc, etc, etc. The data is all there, and the experiments can be set-up in minutes, or better yet, you can do what SecondLife does which is basically set-up market mechanisms that encourage the creation of millions of experiments where the best results get rewarded and repeated.

Not only is all the data there, for each and everyone of these individual events, but its also all there for months/years of past data, as well as in aggregate, and by demographic, as well as in near real time, as well as in correlation to tens of thousands of other events, all in a very structured form that is ripe for the crunching.

And, these are not sample sets, this is every detail of every millisecond of everyone in a multi-million person community. These are data sets the size of which would make the freakiest of all freakonomists totally freak out. As well as being the richest data set ever imagined. Google has a ton of data, but they have very little around the type that can be used to examine the effects of social/audible/spacial/motivational/etc/etc changes that can be made within the virtual world.

The key question around the dream machine isn’t ‘would people use it’, I think the key question is ‘what do we learn about who we are as the dreams are perfected’?

One other point I might stress, again having help build a virtual economy, is that the rules/tricks used to make an all artisan economy, are not as difficult as most people might think at first glance. Basically, it requires a very high level of very well structured data that people within the community have access to. Information about every layer, of every step, in every process, of every product in the entire economy.

Players in World of Warcraft, know the time it takes to collect 10 clumps of raw ore, the cost to transport the ore to the market, the cost to list the ore for sale, the maximum price that the ore is likely to sell for, the value of transforming the ore into a metal, the cost of re-listing the metal on the market, the maximum price that the metal will sell for, the cost of transforming the metal into a part, the value of the part, the cost of transforming the part into a component, the value of the component, the cost of combining the components into a product, the value of the product, the cost of combining products into sets of products, the value of the product sets, the cost of the product set, the value of obtaining a product set and in using it to go hunt for more things, that have very specific values on the market, etc, etc, etc. Consumption happens all over the place, but it is either very closely linked to emotional value, or economic value, and people within the community have a pretty good idea of the expected value of both via feedback systems within the community.

Setting up a similar system in the real world is not a technological problem for which we don’t already have the technology to do so, what it takes in the political will(read: community will) to do so. I think what Dr. Castronova is showing in his wittings is that Virtual Worlds are laying the groundwork that could very easily transform into political will as the efficiency of virtual economies continue to rise at a pace unknown in the physical world.

At the end of the day, where you have talent, trust and trade, you have an economy, and virtual worlds are on the fast track of optimizing all 3 pieces; again at a rate unknown in the physical world.


[Link fixed–Econlib Ed.]

Jan 8 2008 at 1:14am

meatspace was used by william gibson in some of his science fiction books.

kiran mova
Jan 9 2008 at 4:36pm

love your show. i am a programmer, not much of a video/computer game lover, but i did check out second life.

in the show, i felt one major aspect of second life was not mentioned, wonder if it is mentioned in the book.

second world is not all ‘fun’ and not just role playing like a star bucks worker becoming an star ship pilot.. second life is providing a different platform that can be used in place of emails/phones for ‘real’ people for ‘real’ social interactions.

for example, econtalk could be hosted in second life and people can come to this conference room to listen and could interact with other people who have come to listen.

i have heard of experiments to build virtual class rooms.

Jan 10 2008 at 9:33am

Doug, I agree with you that one of the more interesting aspects of the economies in these games, is in fact the player trade through venues such as the auction houses. I find it interesting, in that when a game is first released there really is no market, but as the days and weeks march on, players start to assign value to their various virtual goods. People who make or search for those goods can flood the market causing the price to drop, or cut back on the supplies causing the price to rise. Also, as the game ages and the majority of the player population has moved away from the early sections of the game, the demand for say a certain piece of armor may decrease, as the players start looking for even better pieces of armor. Its an interesting study in the evolution of markets and in supply/demand economics.

Scott Rupert
Jan 11 2008 at 1:13am

With respect to the games that create an environment that favour apetitive behaviour.

I understand why you refer to the artisan economy as such. I will go at some length in an attempt to persuade you not to refer to it as an artisan economy. I believe that it would be more accurate to call the virtual economy demand driven, and the real world economy supply driven. At least from the point of the consumer.

Consider the following:

You want a certain item in a vitual world. You learn that there are 2 ways to aquire the item. One is to kill a dragon and he will “drop” the item. The second is to kill 10000 rabits, sell their fur and purchase the good from a specialist that kills dragons for the item, or a store in the game that makes it available by a slight of hand from the programmers.

In the real world you want something like an iPod. There are two ways to get this item. You can build one or you can get a 9 to 5 and save the money to buy one.

Because of the difference in opportunity costs in both scenarios a rational individual may choose to slay dragons in the virtual world but get a 9 to 5 in the real world.

I think the argument made by both of you is that most of us would rather build our own iPod, it would be more exciting and rewarding. Unfortunately in order to do that one would have to devote their entire life to perhaps only play a small role in the making of the iPod. In the virtual world the time scale is compressed and you can probably within a week or so be able to gain the experience and knowledge to kill the dragon yourself. Because the cost of an entrepreneurial action is so much lower the virtual world provides a venue for someone to follow the more rewarding path.

There is also something else going on the virtual model is optimized to be a demand driven labour market and the producer and consumer roles are compressed into a single individual.

If achieving something requires coordination the virtual world eliminates the two obstacles of having a truly demand driven economy. One – The object or objective already exists (no one has to dream up the iPod before they go make their own). Two – There is no coordination cost. Getting other people to coordinate killing the dragon comes at almost no cost.

Which gets me back to the point I wanted to make. An artisan economy is not characteristically demand driven, however most virtual worlds are demand driven. Further more, an artisan economy does not have the characteristic of being composed of consumer-producer agents. In the artisan economy they are separate.

In the real world only part of markets that resemble the virtual markets discussed are the innovation markets where first goods are created, and entrepreneurial markets.

Daniel D
Jan 11 2008 at 7:29pm

Virtual worlds are by no means a new phenomenon, but are now more sophisticated and graphically enticing than ever and are reaching the mainstream.

Text-based MUDs (Multi-User Dungeons) were fairly popular in the late 70s and 80s.

Sherry Turkle of MIT pioneered research on social aspects of virtual worlds and “virtual life”. Check out her books:

The Second Self: Computers and the Human Spirit (1984)

Life on the Screen: Identity in the Age of the Internet (1995)


Jan 12 2008 at 1:05pm

Great podcast, It really allowed me to think about different possibilities of how life and economies work.

I was wondering about a couple of things: How long will it take for desease (possibly viruses)or crime to find a way into a portion of these games? Could this gaming environment be used to conduct economic experiments to test hypothisi?

The aspect that I would like to see discussed more is in regard to the amount people will spend in the game environment. The amount people play is an economic question. People will play to maximize their overall value(happiness). Some portion of the people involved must be living in the meat world to sustain the meat world economic needs. Some players may not need an outside job, but on average the players will need to interact enough to sustain their meat world life.

Gordon Styles
Jan 13 2008 at 6:49am

Riveting Podcast.

In 2000 I got sucked into Command & Conquer: Red Alert for about 3 months. I would lose 4 hours, sometimes an entire day – would not eat. In the end I smashed the CD’s.

I was struck by your analysis of why people were ‘migrating’ to the virtual World. A comment was made about how difficult it is to ‘vote with your feet’ in the real World when disgusted with ones environment in the real World. Clearly millions do it every year – one only has to look at the developing ghettos of Leicester (UK) to see it in action.

Until July 2005 I was living and struggling in the ‘meat’ UK. I am a very competent 5-axis CNC machining specialist, yet I couldn’t earn a meaningful living in the UK. The UK has been completely trashed by 3 decades of political incompetence, and Industry and Manufacturing (the real wealth creating sector of any ‘meat’ economy) has been murdered.

Having just presided over the closure of a World Class manufacturing facility due to competition from China I would sit in my apartment in Middlesbrough looking out at the miserable weather. I would stay in of an evening because the nightclubs and bars are boring or dangerous or filled with impolite obnoxious women or all at the same time. Nothing to do; almost nowhere to go; unwelcome at sports halls because I’m not that roughty toughty type. Basically feeling somewhat desperate, devoid of hope, and angry with the World.

The result: I voted with my feet. In June 2005 took a 2 week trip to China’s Guangdong Province (the factory of the World) and decided I would make my life there. I wrote a business plan, purchased a one-way ticket and went there to live. I now employ or create employment for 200 people in China and make a substantial living in my own business.

My life is so full of great things to do now that I would find it impossible to even find the time to sign up for Second Life, although I can understand the tantalising attraction of it having spent most of my life in Middlesbrough.

This weekend has comprised the following: happy friendly girls, happy friendly mates, playing cards in an Irish Bar, drinking German beer in a German Bar, playing in a blues band in the local club (after a 14 year gap), playing basket ball (never done before I came to China), cycling, walking in the park with 22C weather in January, learning Chinese, eating healthy ‘real’ Chinese food (not the rubbish we get in the west). I’m worn out – literally – I need a rest and my guys are asking me if I want to go ‘fishing’ tonight (and I ain’t talking fish).

I now live in a fun environment which raises the spirits: the Chinese really understand this and when they get city planning right they cater for it in abundance.

I also live in the fastest growing major economy on Earth with Industry growing at 16% per annum and, in my estimation, the Government is doing a fabulous job at managing it and managing down its structural problems.

The bottom line is this: it is possible to have the life you want in the ‘meat’ World, but we sometimes have to run out of all hope before we are prepared to act to smash through to the other side (Napoleon Hill, Robbins, Tracey et al)

The virtual world is not the answer for me. Each person must make their own judgement though.

I vote ‘MEAT’!

T L Holaday
Jan 13 2008 at 7:58pm

Russ, the URI for the Amazon link to your book is malformed. ttp: instead of http:.

[Oops! It’s fixed now. Thanks!–Lauren (Econlib Editor)]

Karl Mamer
Jan 14 2008 at 7:34am

I’ve heard a lot about Second Life but never really had much desire to check it out until now. A sign you’re getting older, when you’re getting info about cool new video games from economists 🙂

Mark Gilleland
Jan 14 2008 at 12:28pm

Interesting economic event in Second Life:

“A run on a virtual bank in the online world of Second Life produced some real-life losses for their virtual customers…”


Russ – keep up the great work, your podcasts are fantastic!

Jan 29 2008 at 11:10pm

I plan to migrate to the virtual world in the future. I think it’s more cost efficient in terms of getting cheap satisfaction.

Feb 13 2008 at 12:34pm

Synthetic worlds – real community, real money. Edward Castronova and Mark Bell in receiver magazine


Comments are closed.


About this week's guest:

About ideas and people mentioned in this podcast:Books:


Web Pages:

Podcasts and Blogs:



Podcast Episode Highlights
0:36Intro. What are large-scale, online games? Video games involve a screen, dragons, but set up on a central computer where many people log in at the same time. Characters can be run by other people. Becomes a virtual world. Even voice-over-IP. Go back to the 1970s, have become sophisticated. Millions now play. Will use term "real world" to refer to non-virtual world. World of Warcraft, Second Life. Persistent environment on a a computer. You buy the game for your computer and set up a monthly account that allows you to log in, say $15/mo. Small marginal cost for the company, though $75 million or so in development costs. Second Life is not a game environment made a tool people could manipulate to build things. Nine to ten million people have signed up for free accounts. You can, however, buy land there--an island--which does cost something. Can advertise, be found by search. Suppose you design a flower. Can sell it for pretend money, linden dollars, but those can be sold in a second market for real money. LindeX.
9:03What is the magnitude of people buying things in the virtual world and selling it in the real world, even on e-bay. That kind of trade has been going on for ten years. Trade always develops in these world. Might need a motorcycle in Second Life, but you don't have time to build one. To earn Linden dollars is also time-consuming. But you can go to LindeX and use real money. Need a magic wand or a horse in World of Warcraft. Buy it on e-bay, send a check; counterpart transaction takes place in World of Warcraft. Illegal in World of Warcraft, though legal in Second Life to make that exchange. Monopoly is less fun if people make side-payments in real money--outside the rules. Estimates are that trading of real dollars for gold pieces is more than a billion dollars a year. Astounding. Harvard.edu as a domain name is a virtual concept that is worth real money.
16:03Analogy to a foreign country: might people migrate, spend increasing amounts of time in those virtual worlds? Is it like television when it first started? It's a matter of what media people choose to devote attention to, point their eyeballs at. Virtual world has every aspect of human sociality, markets, romance, power, politics. Once you go into one of these worlds you become less skeptical. Unlike the real world, there are developers with complete freedom to structure the environment, and they are going to do so to make people happy. People have fun, become addicted because they find the environment more attractive. Why is it more attractive? It's an economy that's fair, everyone starts out with nothing. It's a community; people are isolated. Modal group of online gamers is older women, age 35-50. What about the person who poured the Starbucks latte this morning? Starship captain in a virtual world sounds like fun. Everyone wants to be a hero. Ask: If they made a perfect fantasy world for you, would you go there? Robert Nozick's Experience Machine, Dream Machine: hook yourself up, machine is preprogrammed to make you experience your greatest fantasy--you discover the cure for cancer or play in the greatest rock band. When your dream is done you are dead. How many people would want to do this? Appealing, but for many people it might have a revulsion or be unpalatable. For how many people would it be an acceptable sacrifice to renounce physicality? People would never have predicted that people would have spent 8 hours a day watching TV. Many of us spend time surfing the internet.
26:53What are some of the competitive pressures will that migration will put on the real world? Real world migrations: consider migration from Europe to America. Virtual world inhabitants will still have an effect on the real world, e.g., voting. Warriors, wizards, policy decisions to treat them equally, may have parallel to groups and expectations for government in real world. Role of myth and lore, longing for myth and lore. Religion could be threatened by virtual reality, religious figures guide real lives. Living in Middle Earth is comforting, a source of meaning. We do not promulgate it as much in our lives any more. Myth and lore in virtual worlds are attractive in ways that real life religions are not. Tolkien invented Middle Earth in a book (before computer virtual worlds) to mirror religious systems, good and evil, to speak to the hunger to act in a world that promotes goodness. Video games mimic Tolkien's creations. Powerful aesthetic, saying to the people at the top of religions, governments, that people want things fixed. Larry Iannaccone podcast.
36:50Capitalism. We feel guilty talking about it in inspiring ways, but it does very well. John Mueller, U. of Rochester, asked if there were examples in movies, etc., of a business person who is not a bad guy? In the 19th century, yes. In the 20th century "Miracle on 34th Street," original of "Sabrina" with William Holden. Mueller investigated P. T. Barnum, who was extremely honest. In virtual worlds, so long as you start everyone with zero, no one cares after the fact about redistributing income. Virtually everybody is self-employed, an artisan. Production in the virtual world suggests that people want to be self-dependent. In video game you get reinforcement every few seconds--not like real life jobs where you have to work hard for a year before getting a bonus. Cooperative activity. Measurability is the problem. How do you measure value added? Calculus teachers would be poor in a virtual world where people drop a gold piece in a canister if they enjoyed the virtual lecture.
44:04Role of fun. Story of Fred, who wants a car, trade stocks, wants a bumper sticker that says "Brought to you by Dow Jones." Set a personal goal that you will achieve a certain level of income. Survey in public policy samples him, discovers he hasn't achieved his goal, government gives him the $25,000. But he still isn't happy--perhaps because his goal was the earn the money himself? A dollar of welfare money may not feel as good as a dollar of (earned) income. How you get close to your goals matters. Standard models ignore the route. Gerard Manley Hopkins. I am my work. Can earn less in virtual world but find it more satisfying. What makes it so appealing? Games are about really deep motivations in the brain. In the savannah, see and want fruit, appetitive behavior, positive reinforcement. Aversive behavior, running from a tiger. Motivating those two systems is the goal of the game. You can't talk to your dog, but you can play with your dog. Motivate with a weak dragon and a lot of gold pieces, but it ramps up. Optimal psychological state, uncertain system over which you have some control, keeps you in a state of flow. Addiction quote from book. "A mother's love is addictive." Cathedrals. Addiction is not a word that captures the appeals involved with virtual worlds. Immersion, loss of self, fun in an ongoing way. Transcendence. There is pressure on the real world, too, to be more playful and more fun as we get wealthier. Pressure in the real world to try to understand human motivation. Where does fun make our lives better, and where is it a distraction?
59:58Feature creep. Wouldn't it be cool if...? Adding features kills a lot of games. Not fun to play a game with a lot of unpolished features. iPhone may be counterexample, but mostly you want something to do one thing well. Game companies test features first, and only then decide which ones to implement. Real world governments, however, implement features and laws without testing. Game companies are confident that they will retain their citizens if they release features that work. 1980 Volker switch to monetary policy, monetary discipline, worked, tried and true. Game designers have competitive pressure. For governments, it's hard to vote with your feet across national lines. Virtual worlds could become sovereign states that compete with real world governments. No other real world government could control it if it were on a distributed network. But you do need the real world productivity to buy the computer, keep your heat going, etc. But maybe we'll do less of some of the sillier real world things we do. Can be uplifting to do something heroic in a virtual world. In virtual worlds, families aren't busted up because of job conflicts, connection still possible. Social technology that brings people together. Exercise, Nintendo Wii, future virtual world could have physical exercise mode, swing sword around physically or creature will get you. Ability of human mind to make technology transport us, cry over endings of books or movies.