Russ Roberts

Varoufakis on Valve, Spontaneous Order, and the European Crisis

EconTalk Episode with Yanis Varoufakis
Hosted by Russ Roberts
Glenn Reynolds on Politics, th... Leigh Steinberg on Sports, Age...

Yanis Varoufakis of the University of Athens, the University of Texas, and former economist-in-residence at Valve Software talks with EconTalk host Russ Roberts about the unusual structure of the workplace at Valve. Valve, a software company that creates online video games, has no hierarchy or bosses. Teams of software designers join spontaneously to create and ship video games without any top-down supervision. Varoufakis discusses the economics of this Hayekian workplace and how it actually functions alongside Steam--an open gaming platform created by Valve. The conversation concludes with a discussion of the economic crisis in Europe.

Size: 29.1 MB
Right-click or Option-click, and select "Save Link/Target As MP3.

Readings and Links related to this podcast

Podcast Readings
About this week's guest: About ideas and people mentioned in this podcast:


Podcast Highlights
0:33Intro. [Recording date: February 1 4, 2013.] Russ: We're going to talk to begin with about Valve, which I found out about because a number of listeners and readers of mine told me that I had to do a podcast about Valve. And I started to look into, and they were right: it's a really good idea, because Valve is a very unusual corporation. So, describe first what they produce, what's their output? And tell us about their structure, which is particularly unusual. Guest: Well, they are a video game company. It was started by Gabe Newell, who was a Microsoft high-ranking software development officer who cashed in his shares in 1996 I believe, and together with Mike Harrington, a colleague of his, they set up this video game company. But what is astonishing about this video game company is that at some point, a couple of years after its initial inception, it went into the world of multi-player, internet-based games. And this is of significance because when you create single-player games that are on CD ROMS and you just sell them, you are just like any other company. But the moment you create multi-player, online, real-time video games, effectively you put together a community of players. And you create a virtual environment for them to have fun within. But you don't control what they do. So they start doing all sorts of interesting things with one another, like for instance transacting and creating spontaneously without anyone imagining that that will happen. Social economies. So, that's what Valve does. And the way that it has pushed aside the great divide between consumers and producers, designers and shapers of communities has had a major impact, also, on the way that Valve Corporation employees relate to one another. Russ: So, describe that relationship. There doesn't seem to be anyone in charge. Guest: The most astonishing aspect of life at Valve is that indeed there are no bosses. It contains no explicit hierarchy, and it is based on what several members of the company have described to me as the principles of 'anarcho syndicalism'--effectively free association of employees with one another. As you know, Google prides itself for allowing its employees 10% of their time to work on whatever project it is that they want to work on. At Valve they take enormous pride at having pushed that 10% up to 100%. So, it is a big disconcerting for people who enter Valve, because there is no one there to tell them what to do. So it's a flat management, spontaneous order kind of operation, which creates a very interesting phenomenon from a managerial perspective, and actually from the perspective of people who try to live and work within it. Russ: And let's start by talking about the hiring process. Obviously they are very careful about who they hire. Do you know anything about how they go about that, what the process is? Guest: Yes, I did participate in a few hirings. They are very careful to be faithful to their principles of not allowing themselves to fall prey to the lure of hiring. So, the way it works is very simple. Let's say you and I have a chat in the corridor or the conference room and the result of this chat is that we converge to the view that we need an additional software engineer or animator or artist or hardware person. Or several of them. What we can do is we can send an email to the rest of our colleagues at Valve inviting them to join us, forming a subcommittee that actually looks for these people. Without seeking anyone's permission in the hierarchies, because there is no hierarchy. And then we form spontaneously the search committee. And we interview people, first by Skype; then we bring them, if they pass that test, to the company for a sort of face-to-face, personalized interview. And anyone who wants to participate does participate. And then during that day--usually a day-long event--emails are fired all over the place with views whether this person should be hired or not, until some consensus is reached where there is effectively no one vetoing the hiring of that person. Russ: And so there's no human resources department, as there would be, say, in a traditional corporation. But what do they pay that person? Who decides that? Guest: This is a haphazard process. The payment mechanism is to a very large extent bonus-based. So, the contract usually has a minimum pay segment in it, which is more or less established by tradition. And then the--the interesting part in this contract is how much is left to the peer-review process. Which is very complicated; it involves various layers of mutual assessment. In companies like Microsoft or elsewhere, usually the bonus is something between 8 and 15, 20% of the basic salary. In Valve, there is no upper limit to bonuses. Bonuses can end up being 5, 6, 10 times the level of the basic wage. Russ: So, who designed that complicated peer review process? Did that emerge, also, from interactions among the employees? Or was there somebody, like Gabe Newell who started the company, did he decide what that peer review process would look like? Guest: I wouldn't be surprised to hear that it's a combination of the two. The anarcho syndicalists structure of the corporation owes a great deal to Gabe Newell, because Gabe had this view from the beginning--he wanted a community of partners. He didn't want to be the boss of anyone or to be bossed around by anyone. But at the same time, he couldn't imagine the kind of social conventions that govern--the relationship [?] within Valve--would not have seen that far ahead. They evolved, organically, within the firm, the corporation.
8:33Russ: And, can you get fired? Guest: Oh, yes. Absolutely. Russ: And how would that happen? Guest: Firing people is hard in an anarcho syndicalist environment. But it does happen. I've seen it happen. It's never petty. But it involves various communications, at first, when somebody is not performing or somebody doesn't seem to fit in with the rest of the company--in many occasions people simply don't fit in not because they are not productive or good people, but because they just can't function very well in a boss-less environment. And then there are serious discussions between co-workers and the person who is being canvassed or discussed. And at some point if there is no way that a consensus emerges that this person can stay, some attractive offer is made to that particular person and usually there is an amicable parting of ways. Russ: But in this company, you can work on whatever you want to work on. So, if you are not getting along with, say, the 3 people or the 8 people on your team, I assume you can just leave at any time. You can join any team; you can leave at any time. So, one way to get fired, I suppose, is to leave a lot of teams and not be able to find one. Can you hide? Would someone notice if you weren't involved? Since you don't have a boss, would someone notice that you weren't on any teams? Guest: Yes. Yes. It may take longer than in a normal hierarchical corporation to work out that somebody has fallen out of the system. But it is important to understand that such spontaneous-order based enterprises rely to a large extent on individuals who actually believe in the social norms that govern their existence. So by the very nature of the beast, you don't have people in there who try to hide and who try to somehow create a smokescreen around the fact that they are not very good at what they do. Most of the people who are there, all of them, have been hand-picked to be excellent at what they do. They are usually on top of their game elsewhere before they join the corporation. And the mobility within the corporation is a great asset, and everybody recognizes it. Everybody's desk is on wheels. There are only two plugs that need to be unplugged in order to shift from one team to another. There have been, I am sure--I haven't been with the company long enough to notice--but there must have been situations where somebody didn't fit in and eventually was edged out of the company. But the vast majority of such moves simply contribute to the overall efficiency. And to the private joy of working there. Russ: And of course in a traditional corporation you can hide from the boss in other ways. You can loaf and take it easy and take credit for things you didn't do. There's a lot of opportunities in a traditional corporation for that kind of behavior. Guest: And the worst part of traditional corporations of course is that you create a lot of work for people under you so that you seem important. Work that is unnecessary. Russ: Correct. In this case, though, as well-meaning as some people can be--you know, when you have a baby, whether you are a man or a woman, and the baby enters your house, it's sometimes difficult to sleep at night; there's times that things have to get done during the day. It gets a lot harder to be a reliable citizen. So there must be some temptations occasionally for that kind of behavior. Guest: That's right. Certainly. And there are frictions, too. One of the tasks that's very difficult to find people--women--to do them is managing other people. Because, as I said, the vast majority, if not all the people at Valve, are highly creative people who want to produce stuff that actually improves the experience of the customer base. So taking leave from this creative process and organizing, managing, the work of others, as sometimes one needs to do, is a chore. And it's very hard to find people who want to do it within Valve. And this is one of the great impediments that is being faced by anyone in the company.
13:49Russ: So, one of the issues that you think about--somebody put the Valve Employee Handbook online. And I assume it's real. It's very charming and whimsical. So, one excerpt says: "Method to working without a boss. Step 1. Come up with a bright idea. Step 2. Tell a co-worker about it. Step 3. Work on it together. Step 4. Ship it." That's sounds nice. It's simple. But somebody's got to decide when to ship. Maybe that's the team. What about advertising? How are those decisions made? An anarcho syndicalist company--everybody doesn't make all of the decisions together. Right? They don't all sit around and make all these decisions in long, dreary meetings. Something else must be there to make these decisions. Guest: You'd be astounded to see how it's done in a totally decentralized fashion. So, if we are working on a particular game, you and me, and we come up with an update we want to ship, we ship it. And we don't tell anyone. Regarding advertising and marketing, the interesting thing here is that while you have this anarcho syndicalist, bossless environment within the company, the company itself, looking at it from a political economy point of view, has created a great deal of monopoly power in the marketplace. The creation of the platform, the electronic platform trading video games called Steam, was a major breakthrough for the company. Because Steam is like an electronic market for video games. It handles about 70% of all [?] of video games, not just of Valve's--everybody's around the world. There are about 55 million people that have online accounts on Steam. So, I'm saying this in order to answer your question about advertising and marketing. Valve doesn't need this. It doesn't do it. It doesn't indulge at all in any of those activities. It has created an international, global marketplace. It has an amazingly good name among customers, primarily because it serves them very well and it does not try to exploit them. It is not using to its full, nowhere near its full capacity to monetize activities that happen in these video games, as well as on Steam. So customers have a very healthy respect for Valve, and this is the best marketing tool that Valve could possibly have. So you have a very interesting combination of a kind of a kind of anarchy structure within the company. A distribution of income which is almost socialist within the company. And of vertical market structure that is[?] founded on pure monopoly principles, really. Russ: You say it's socialist within the company. Aren't some people making much larger bonuses than others? Guest: Yes, but those bonuses are not determined on the basis of ownership of shares. They are determined on the basis of what everybody thinks everybody else should be getting. Russ: It's not egalitarian, is a better way to say it. Guest: How this works--two years ago, when I first met Gabe, he said: We usually get our vision song contest principle. He said to me, remind listeners who are not familiar with this awful European institution. Terrible. Russ: Yeah, what is it? Guest: The Eurovision Song Contest. Terrible idea. It's been going on for 50 years now. What happens is every country is [?] with one awful song, usually. Then the audience votes in the best song, and that country wins. But, no British citizen, for instance, is allowed [arrived?] to vote in favor of a British song. We can vote for all the other songs, but not for their own. So I was told that this is the principle. So what happens when it comes to determining bonuses, you can vote in favor of somebody else getting a bonus, but you can't vote for yourself. Russ: That's a good idea.
18:07Russ: Tell us a little about the actual products. I'm not a video gamer. What are some of their most successful games, and where do they stand in terms of the marketplace? You said they are widely recognized as a quality firm, producing quality games. But tell us a little more about their market position. Guest: Well, I have a terrible confession to make here: I'm not a video gamer either. Actually, I'm pathetic at playing video games. And I'm not that interested in them, either. Russ: We're too old. Guest: I'm too old for that, that's right. Old dogs and new tricks don't mix very easily. But because of my involvement with Valve, passed around, nephews and nieces, younger people in their twenties and thirties, forties even, who know about these markets, these products, they tell me that Valve's games are exemplary. The successful stories are games like Counter-Strike, Team Fortress, Team Fortress 2, Half-Life, which was one I think of their first. There's one which is very interesting from the perspective that it involves team play called Dota, and it stands for Defense of the Ancients. Various games. And of course there is Portal, which is aesthetically very pleasing in terms of the architecture of the environment which you find yourself if you play it. So Valve hasn't cornered the market in terms of its actual games. It's more like a niche player. Its games are popular but not the most popular amongst video games. But those who actually--I even know--they think they are just like the BMWs or the Maseratis of the video game world. But the greatest success from an entrepreneurial point of view is Steam, because Steam is, as I said, the global trading place where you buy all sorts of games--not just Valve's games but everybody else's. Russ: I want to understand it--did they create Steam? Guest: Yes. Russ: Okay. Now, is Valve growing? In terms of employees? Guest: It is growing steadily. Around 30 or 40 people a year. And in terms of revenues I've heard remarkable numbers, ranging from 30 and 50%, every year. Russ: And so the question I think you've wondered about and others have is: How big can it get? Is there a limit with this kind of non-hierarchichal structure? Will it maybe at one point level off? Or keep growing? What are your thoughts? Guest: Let me inform you that the founders of the company, Gabe and Mike, when they started the company in 1996 thought that their natural limit would be around 30 and 50 people. This is what they tell me. Now it's more than 300. So, the [?] rate is much greater than they anticipated. The fear that that had about 30 or 50 people would not be consistent with this anarcho-syndicalist model of management has not been realized yet. So the big question is: What is that limit? And there is a secondary question, but it's sort of of secondary importance, and that is: How would a company like that handle contraction? Because at the moment we have a very significant growth rate. Well, corporations prove their effectiveness and their resilience at times of recession. And even though Valve has experienced a major recession around it, its own market has been going from strength to strength.
22:20Russ: Now, how did you get involved? And how much time do you spend on this? I love your title: you are called 'Economist in Residence.' Most companies don't have an economist in residence. They have a chief economist. Or an economist period. How did you get involved and what's your interaction with them over the course of a year? And is it fun? Guest: Well, in November of 2011 I was about to visit the United States--I was living in Greece at the time--on a book tour. And I got an email from Gabe. Which I nearly deleted, because I thought it was spam. What do I know about video games and why would a video game chief executive want to talk to me? So I thought it was--but it struck me that he was interested in my work on the Eurozone and my criticisms of the way that the Euro Crisis was being handled by the powers that be in Europe. And he drew what I thought was a fascinating parallel between the problems they were having at Steam, when different communities of video gamers met one another on that electronic platform and started trading with one another. His fear was that something like the Eurozone would emerge on Steam. That was enough to get my attention. Russ: That's incredibly cool. Guest: That's right. So he invited me to go to Seattle and chat with them. And when I went I thought, right, okay, I'll go there for two days, have a peaceful vacation with them, consult with them, and then that's it. But then they offered me the position of an economist. They offered me an incredible opportunity to look into their data set. Because the amazing thing about the social economy emerging on Steam is that it is a very large barter economy where you have perfect information about whatever happened in that economy. So they were interested in my; I was interested in them. And also, they allowed me to retain my academic frame of mine, in the sense that I felt no pressure to produce opinions that were more suited to the company's interests. Because I am an academic and I want to retain my academic freedom. And also, even though I agreed to be an economist in residence in the sense of being stationed there, Jason[?], Gabe, and I came to an understanding that I need to be at the university; and so that's why I am now stationed at Austin and I am teaching at the U. of Texas. So I'm no longer in residence but I keep working with Gabe Newell and consulting with them. And having a great deal of fun doing that. Russ: And a while back, a little while ago, you wrote an essay, a blog post, on the nature of spontaneous order and the intellectual antecedents of Valve structure. Talk about that. Talk about what Valve has to do with Hume, and Smith, and Hayek. And then I think we'll get to Coase. And those are all names we've talked about here. Guest: Okay. We have about 20 hours. Russ: We have plenty of time. Guest: I know we don't. Okay. The basic idea that Adam Smith got from David Hume was that you don't really need a central authority to tell the members of a society or community or village, whatever, what to do. The Prisoners' Dilemma problems that Thomas Hobbes was very concerned about in The Leviathan will somehow resolve themselves in an evolutionary manner as people learn to follow conventions that in the end select themselves on the basis of some kind of social selection process. To me that developed the basic idea of David Hume into the invisible hand. The idea being that needs can be provided for better when no central authority is trying to provide for them. By allowing the free flow of private initiative to remain unrestrained by central authority, private vices are going to be combined, as if by an invisible hand, in order to produce public virtues. Now that was the beginning, if you want, of political economics. With Adam Smith. Of course, after that, we had major debates, occasioned by theoretical objections to this view. Also by economic events. So, periods of recession, periods of crisis where free markets failed, different circumstances both in Europe and the United States gave rise to different treatises. Like for instance David Ricardo objected to some of the aspects of Smith's optimism. Later another liberal, John Maynard Keynes, following in 1929 the Great Depression. So we economists can go on and on and on about those things. But the interesting aspect of the economies that emerge spontaneously on Steam is that they come as close as one can get to this idea that Smith had of creating public virtues out of private vices. Which later was elevated to a different level by Friedrich Hayek, Mises, and the Austrian tradition. Now, I am not like you a Hayekian or an Austrian--I have a good deal of respect for their views--when it comes to the real economy. But it seems to me that these digital economies, that they were created for the purpose of giving us a real life example of Hayek and von Mises and Adam Smith at work. Because they lack two problematic markets that the real economy has. They lack the markets that throw a [?] in the works of spontaneous order and of free competition. And these markets are the market of capital and the market for labor. In these digital economies you have pure exchange economies and you can see how well they work. Even though they have a tendency to create bubbles in a way that Hayek--no one understands very well, which then get deflated, and that rejuvenates these digital economies. But they lack a labor market. It's not that they lack labor. Or production. Indeed, I was fascinated to find out that there is a lot of labor involved in these markets. And a lot of production. Let me give you a figure, which is actually rather astonishing: There are hundreds of thousands of people around the world today, as we speak, in the United States, in China, in the Ukraine, in Russia, who make a very good living out of being producers of Steam. And by producers what I mean is they are designers--they design electronic assets. Like for example they design a hat or ear mufflers you can wear, you can put on your avatar. So they design these; they put them on Steam; there will be an app producer who will put it on iTunes and share the revenue with Apple. Similarly there are designers who do that--they produce assets for Valve and share the revenue with Valve. So there is a lot of production. And there is a lot of labor. But there is no labor market. Labor markets are very specific. Russ: Everybody is self-employed. Guest: Right. Effectively you work, you produce something, and you sell that something. You are not hiring or renting your labor out to a capitalist, to somebody who owns property and therefore commands your labor in order to utilize it so as to produce some other output commodity which is alienated from you and then it is sold on. And it seems to me that, if you are Hayekian, the economy of Steam should be a place that holds a great deal of interest to you and a place that you really want to study as well as you can.
31:13Russ: We did an interview a long time ago with Edward Castronova, who is an economist interested in the virtual world of these online massive multi-player games. He was very optimistic about their impact and what we would learn from them. I don't think that promise has been realized completely. But I do think, as you point out, there are hundreds of thousands of people right now working on those markets in ways that most economists--there's a revolution going on in some dimension that most economists aren't thinking about. But let's stick with Valve, in its employee structure, which is a different part of this, I like to say emergent order--I know 'spontaneous order' is the standard term. Steam, which is the place where these online players interact and buy and sell and play, that's one market or emergent order. Another is Valve itself, which, because of the lack of hierarchy is sort of what I would call the 'un-corporation'--it's the, Coase and others have pointed out how ironic it is that in capitalism, corporations don't act like capitalists. Typically they don't use prices to allocate goods and people; they use command and control, they use top down. But Valve is a corporation which is trying to use as little top down as possible--which is extraordinary. And, as you point out, you can have emergent order without prices. There's a lot of interesting emergent orders that have prices in them. There are a lot of them. Standard markets for things where people buy and sell are examples of that. But Hume--and I think Smith--you suggest not in your essay, but we'll come to that probably--certainly Hume and Smith, I think Hayek also, understood that there are emergent orders that function without prices. So, talk about Valve as an emergent order without prices--it's employee relationships. How does that work? What's doing the ordering when there aren't prices? Guest: Well, it's social norms. It's a Humean world. I did mention it before, that disciplinary devices in Valve are far less important--formal disciplinary devices or processes--than self-regulation on the basis of social conventions that emerge in precisely the way that you were describing. Artificial virtues in the history of human nature. And so in a social economy like that within Valve, you have a number of rewards that are not related to money. The way in which these people interact--remember, I said before, these people are at the top of their game, top engineers, top software developers, top animators, people who worked on the Lord of the Rings movies. The sheer satisfaction of having shipped something that then gets rave reviews by the customer base on Steam is not to be scoffed at. It is a major reward system. Given also that you have an abundance of money, in a company like Valve, when more than $1 billion dollars revenue, you have 300 something people to share it, then you are not operating exactly under conditions of scarcity. Especially when you are not operating under conditions of scarcity, all the other social rewards that one gets from great reviews of one's output. Russ: And the respect of your peers at your workplace. Guest: Of course. Status, if you will. Status and the respect of the people that you respect. That way you overcome the master-slave, Hegelian paradox. Guest: You get the respect of people you respect. Russ: But that's very Smithian. It's just not the Smith of the Wealth of Nations. It's the Smith of The Theory of Moral Sentiments. Guest: Indeed, indeed. But you see we economists always make the mistake of reading The Wealth of Nations without The Theory of Moral Sentiments. And Smith never expected us to do that. He thought we would build, upon, the Theory of Moral Sentiments the invisible hand. But economists don't do that. To their great detriment. Russ: Yeah, I agree. But I just wanted to add that footnote, because Smith wasn't so interested in creating public virtues out of private vices. He was aware of it. But he was also interested in creating private virtues and hoping that would lead to even more public virtue. I think. Guest: Yes. That's quite right. Now the other thing that would interest you about Valve is the way in which the internal ethos is spreading to the Steam community, to the community of customers. What I believe is exceptionally important is the very fast removal of the barrier, the hard barrier separating the customers from the producers. The customers evolve from the people who work at Valve. And that diminution, that removal of this borderline, takes many different shapes and forms. Let me give you just one example. I've already mentioned the example where customers are already turning into designers. And now people at Valve realize that there is no way they can compete with their customer base in terms of productivity of electronic assets. Quite significant. Russ: Say that again? Can't compete with what? Guest: With their customer base in terms of producing electronic assets for their games. So in-house, they try to develop new shields and new guns and new stuff that you would want to put on your avatar if you are a player. Right? But the in-house production of these assets fades in its significance to what the community can now produce. So the strict separation of producer and consumer is gone, effectively. And Valve is now in the process of blending production and consumption across this divide. Let me give you another example that Gabe Newell told me once, and he keeps repeating it because he is fascinated by this: There are thousands of employees of other companies, and some of them are competing companies, competing with Valve, who make more money producing these assets for the Valve ecosystem than the salaries that they draw from the company that employs them. In their spare time. So also the dividing lines are receding.
39:02Russ: So, the other effect that I think would happen is it would put some pressure on other companies if they want to keep their designers, to either become like Valve or do something different. I'm going to read another excerpt here from the Employee Handbook. It's subtitled "The Office." It says:
Sometimes things around the office can seem a little too good to be true. If you find yourself walking down the hall one morning with a bowl of fresh fruit and Stump-town-roasted espresso, dropping off your laundry to be washed, and heading into one of the massage rooms, don't freak out. All these things are here for you to actually use. And don't worry that somebody's going to judge you for taking advantage of it--relax! And if you stop on the way back from your massage to play darts or work out in the Valve gym or whatever, it's not a sign that this place is going to come crumbling down like some 1999-era dot-com start-up. If we ever institute caviar-catered lunches, though, then maybe something's wrong. Definitely panic if there's caviar.
So if you are reading that, now that this is online, and you work somewhere that there isn't the massage room--of course, I'm sure there are other places that have darts and maybe espresso--this looks like a very fun place to work. And I would think there might be some copying going on eventually, if this is a successful model. Guest: Yes, there will be. And already there have been a number of corporations, companies, that have been in contact with Valve to try to pick their brains as to how they can move in that direction. But let me tell you what my view on this is: It's not very easy to do. The point I want to make is this: if you've already created a company which is hierarchical, it is exceptionally hard to push it in the direction of the flat management model of Valve. You can use [?] rooms, you can introduce special machines, but once you have some people who learn how to exercise, retain power on others, and others who learn how to live under the authoritarian power of their bosses, to create the social conventions of equality that are necessary to run the Valve model is absolutely impossible. I mean, to deprogram these people. And once power infects their mind, it's very difficult to find a disinfectant. Russ: Yeah, I think that's true. It's very difficult to create corporate culture generally, and it's certainly hard to change it. And it usually requires somebody like a founder, like a Gabe Newell, who creates it to start with. Imposing it from the outside is almost impossible. I think you are right. But it will be interesting to see. Guest: It's hard enough to deprogram those who come into the company. And that's why a company like Valve cannot expand exceptionally, because if you bring 40 people into a department who are used to having hierarchical ways of working, then this may be detrimental to the whole equilibrium in the company. You have to bring them in slowly in order to assimilate them, and to deprogram them. Russ: And I assume that happens. As you point out, some people come and they just don't fit; and they eventually leave, or presumably voluntarily, or they are fired. Because it's an alien environment. Guest: That's right. Some people just are not functioning. Russ: It's a little bit like an economics department. Academic life has a certain boss-less nature to it. Once you get tenure. It does have a little bit of some of that. Of course, economists have their own culture and their own norms of what's considered community behavior. I suspect that the Valve employees do a little bit better. But maybe not. I don't know. Guest: This is precisely right. And it's the reason why I was interested in being part of this environment, because it resembled so well the world of academia that I know and love. The fact that I don't have a boss; once I was tenured I no longer had a boss. And that doesn't mean that you don't have responsibility; you take responsibility for what you do. And you do it at a collegial level with your colleagues. So, yes, Valve operates like a good academic department. One that does not have a lot of dead wood. Russ: Yeah, that's the problem. Guest: You get rid of them. Russ: Yeah. The difference is that the peer review system in academic life doesn't work quite as effectively, with the bonus tied to the peer review, as it does at Valve. Guest: That's right. And also you don't have the quantity measures that you do have at Valve. Because let's face it--you do something, ship it, you see immediately the quantities going up and down, both in terms of revenues and also in terms of the satisfaction of your customers. Whereas in academia it is impossible to have such measures. And we shouldn't have such measures, fast measures, either, because our purpose, as academics, is not to please our audience. Sometimes our purpose is to upset them. Russ: Right. You could have an article that isn't downloaded very often but is still incredibly influential. But there's a relationship. There could be some quantitative aspects to evaluating academic life, but I agree with you--I wouldn't push it too far. Guest: There are, but I just don't like them. I think they are detrimental to good academic life. When in academia we try to replace judgment with calculation, we mess things up. Russ: Well, yeah. I think calculation is over-rated generally. But that puts me in a minority in the profession. The profession is going in the other direction don't you think? Guest: Of course. Catastrophically so. And there is a fundamental difference between the social sciences and the real sciences. In the real sciences, if you tell me that Person X has published 30 articles in Nature or Science, I am prepared to take it for granted that they are very good physicists. But in our neck of the woods, because we are effectively [?] to the equations, that's not necessarily the case. Russ: I agree with you. But we are in the minority. Why do you think that has happened? Guest: Oh, it goes back a long way. It goes back to the 1879s and 1880s, when economists entered the universities. Because they needed to prove that they are the scientists of society. And they had, they created a false science whereby classical mechanics of the 19th century were copied into mathematical models of exceptional complexity--and aesthetic beauty--and the only way of closing them, of solving the mathematics that was created, was by making assumptions which ensure that these models have nothing to do with really existing capitalism [?]. So the more successful you were in creating these models which gave you discursive power and academic power in the corridors of economics departments, the more irrelevant you were in explaining the real world. It is a very peculiar failure. Russ: Yeah. I have a little bit of mixed feelings about it. I think we've gone down the wrong path. And my preference would be to go back to Adam Smith. I think he had the right path, which is descriptive, narrative. I think we're more like historians than physicists and we should admit it. Guest: Indeed. We should go back to the classical economics tradition. Whatever your political or ideological or methodological take is, I think it is exceptionally [?] to all intellectuals, who think things through, that today, Adam Smith, David Ricardo, John Stuart Mill, Karl Marx, Friedrich Hayek, even John Maynard Keynes, would never get a job at Harvard University's Economics Department. Never. Under no circumstances. Russ: It's true. Guest: They wouldn't even be allowed to pass outside the department. Russ: That's an interesting idea for a movie. I like that idea. I'm not going to say anything else. Well, I'm not going to make this movie, so I will say it: The idea of any of those people standing outside the door and trying to get in and saying: But I'm the guy who started it. You are right. They wouldn't pass the exams. Guest: That's right. It's like Jesus Christ not being allowed in Church. Russ: Yeah. I don't know if you want to push it that far. But I get the analogy. Might be a little bit overreaching for economists there to make that point.
48:22Russ: We've got a few minutes. Let's turn briefly to Europe, which changed your life, just like I think the crisis has changed mine. We live in very interesting times. What is going on? Where do you think we stand now with respect to the crisis in Europe, particularly in Greece, where you come from? Guest: It's getting worse, both in Greece, and in Europe. And yet our elites are celebrating the end of the crisis. Which is quite preposterous. Look, in brief, we created a common currency that was utterly ill-designed to sustain a major financial crisis like that of 2008. So when the crisis hit in 2008, this currency area started disintegrating. And of course this disintegration begins at the weakest link. And Greece happened to be the weakest link. You and I would not be talking about Greece today if Greece in 1999 by some miracle of politics and rationality had stayed out of the Eurozone. That is the reason why it is such a disaster; and it's why it's so significant in the world economy and pipsqueak Greece has been dominating for three years. The headlines of [?] which is a sign that something is definitely wrong with the international economy. And the reason for that was that Greece was in the Eurozone. The tragedy of course is once you are in, you can't get out. You are trapped. And so on and so forth. Russ: Why? Why are you trapped? Guest: People often ask me: Why can't Greece do an Argentina, by [?] with the U.S. dollar and when its own [?] value defaulted. And the answer is: Because Argentina had its own currency. It had pegged with the dollar; but it's very easy to cut a peg. You just do it. All it needs is political will to do it. But in Greece, we don't have our drachma which is pegged with the deutschmark. If we did then we would just sever the peg. We have the deutschmark; we call it the 'euro'. We don't have our own currency. So, to actually get out means to create a currency from scratch. And it takes around 8 months; even at best it takes 8 months. So this is equivalent to announcing 8 months in advance the devaluation of your currency. Now, in those 8 months every Tom, Dick, and Harry is going to sell out in anticipation of the devaluation and take everything out of the country. So you need to have the army mounting the borders and the airports and the ports, which means that Greece will have to get out of the European Union since you can't re-erect [?] borders if you are in the European Union. So, given that Greece's survival depends on free trade with the European Union, it will be a major catastrophe. We will be going back to the stone age. Russ: So, either they can't do it or it's very hard to do. What's likely to happen then? Because what I read--and I don't read it very carefully--too much to read already--but what I read is they are muddling through. Like you, my first thought is that that's probably not true; they are whistling in the dark. But if you right and the crisis is not getting better, it's getting worse, what's going to be the resolution, or do you think will be the resolution? Guest: Well, a Great Depression, in the periphery of Europe. Which already is happening in places like Greece and Spain. [?] unemployment can reach 30% soon. It's unbelievable in this day and age that we should have such unemployment rates. And this is a self-reinforcing depression in places like Greece and Spain. Either the Euro is going to break up, or you can have a large part of Europe being turned into Kosovo-like protectorates. [?] if possible, or which the United States, Nato waged the war in 1999, has the Euro as its currency, but it's not really a country and it's not a member of the European Union or the Eurozone. It's got a mafia-ridden political system. It is more or less ruled by decree from Brussels. And Nato troops keeping the peace, so to speak. And the only export of this protectorate is it's people, migrating in droves. So that is a possibility. But I think that the Eurozone is not going to survive that because the political pressure in places like proud nations like Italy and Spain are going to be such that the whole thing is simply going to go belly up. But the reason why we are having all this, just to cut to the chase, is because in 2010 when a country like Greece, like the Greek state went bankrupt, was that Europe refused to allow it to default within the Eurozone. This was just a crime against humanity. Greece had a debt that was unsustainable; and so as to make sure that Deutschebank [Bundesbank?--Econlib Ed.]and [?] wouldn't suffer more than they would have to, even though they were themselves insolvent, and relied on the generosity of the German and the French taxpayers, the largest loan in history was piled upon the shoulders of the weak and bankrupt Greek state. This was effectively a con job. The German electorate was being told that this was an act of solidarity towards the Greeks. When in reality all that was happening was that the losses of Deutschebank were being piled on the shoulders of the German taxpayers. Russ: Yeah, I hear what you are saying. Guest: The German taxpayers realized that. And when the German taxpayers realized that, they hated the Greeks; and the Greeks hated the Germans. And we've lost the plot completely. The only beneficiaries of that are the Nazis. Russ: Well, it's as you said. I think--in a "normal" world, what would have happened is that Greece would have defaulted. Although they might not have been able to get in such a large debt if they hadn't been part of the Union. Guest: That's right. Russ: So it's kind of a weird simultaneous unstable system.
55:11Russ: What worries me--I'm worried about a much larger thing than people seem--you know, most people worry about interest rates, unemployment. I'm worried about those things; I'm worried about civilization. These kind of things usually lead to war and revolution. They don't lead--I mean a depression is bad enough. But when you get unemployment at the levels that we are talking about in Greece and in Spain and possibly elsewhere, you don't just have people disappointed with how the economy is doing and vote for somebody else. They take up, they go to violence. And I wonder if democracy is going to survive in Europe, based on this. Guest: Well, it's not by accident that the last word I uttered just before you asked me this question was 'Nazis.' Because in my country, a country that by the way fought very hard against Nazis in the 1940s, tooth and nail; and we had on a per capita basis had the largest death rate during the 1940s as a result of resisting the Nazis. In my country, that has this proud anti-Nazi position of resistance to the Nazis, the Nazi Party--it's not the Neo-Nazi Party, it's the Nazi Party, a fully fledged Nazi Party--is the third largest party in the Greek Parliament as we speak. This is precisely what you are talking about. Europe and its elites are liable for what is happening now, in Central Europe in particular, central Europe and southern Europe, which is that in viable[?] of the 1930s in the most worrying ways, and pushing populations into a corner with racism and Nazism being the only beneficiaries. Russ: Is there anything attractive to be done? Guest: There is a lot that we can do. Europe is rich. It's full of intelligent young people. Innovative people. We have excellent educational institutions. We have amazing culture. And actually, we don't have that much debt either, compared to Japan and to the United States either. What we have is an idiotic set of rules and regulations and institutions for running our common currency area. We have a Central Bank without a state behind it. And we have states without a central bank behind them. We have banks that are insolvent, and states that are insolvent. And we insist that the insolvent states borrow in order to bail out the insolvent banks. And when this is failing then we reduce aggregate demand further by increasing taxes, until the whole thing has spins out of control. There are three steps you can take, very, very simple steps, without the need of constitutions, federation, anything like that, that will put an end to the tailspin. It's just not in the interest of the political elite, individually and collectively to go ahead with those. Russ: And they are? Guest: Very simply, separate the banking crisis from the sovereign debt crisis. Instead of having the Spanish state borrowing on behalf of the Spanish banks, the Greek state on behalf of the Greek banks, the bailout funds that we have now created to deal with the crisis should inject capital straight into the banks on condition that the shareholders and the bondholders are expropriated. New governors are introduced. And a large number of these banks are liquidated and cleansed. And then once this happened, like it happened in 1990 in Sweden. Then these banks can be sold straight back to the private sector, once they've been cleansed. That way you decouple the banking crisis from the debt crisis, which you then kill off completely by having the European Central Bank [ECB] take on its books not by monetizing but by issuing bonds, that would be the equivalent of U.S. Treasury Bonds, a part of the debt which is consistent with the Maastricht Treaty. That is, a debt that every country had the right to have should be transferred centrally and managed centrally by the ECB. That's straightforward to do if you want to do it. And that overcomes the moral hazard problem because you don't transfer all the debt; you transfer the legal debt and you leave the remaining part of the debt to the sovereign state to deal with it. And the third thing that you need is you need some kind of investment [?] in Europe. And we have the European Investment Bank, which operates on simple straightforward free market banking principles. It has, as we speak--[?] because I've spoken to them--they have very interesting investment business plans waiting to put them in place. Like for instance super-fast trains in Eastern Europe and so on, and they are not allowed to carry on with them or go through with them or push through with them because of the silly rule which says that 50% of each one of these investment projects has to come from the states--the national government--of the place where it's going to take place. And of course national governments are bankrupt, so it doesn't go ahead.
1:01:09Russ: Well, I like some of those ideas. Not so crazy about others. But the part--put my opinion to the side. The most important thing that stops those ideas from coming to fruition is the political power of the financial sector, which would be unhappy with that first part of your plan. We've had this same problem in the United States, if I understand it correctly, both what you said and my understanding of the U.S. law. We had a law called the Federal Deposit Insurance Corporation Improvement Act (FDICIA), and that was supposed to do I think very similar things to what you just said to American banks that got in trouble. But instead it was not invoked in the Crisis of 2008. And instead banks were made whole, and the people who made the mistakes were allowed to keep their jobs to a large extent. Guest: You are precisely right. Russ: Which is immoral and destructive. It's bad enough that it's immoral. But it's destructive. Guest: Indeed. And especially Europe. Imagine if you had exactly the same situation in America, and on top of that you didn't have a Federal Government. And you had the State of New York trying to salvage Wall Street in that way, and the State of Nevada trying to salvage the banks of Nevada. What will happen is everybody will simply sink, states and banks together. This is what is happening in Europe.

COMMENTS (28 to date)
Ed Hanson writes:

A terrific interview with tons to digest, which I am attempting to find the time.

One correction I believe is important enough to mention. In the Podcast Highlights, the 55:11 section it is written, "a fully fledged Nazi Party--is the largest party in the Greek Parliament as we speak." If I hear the interview correctly, Yanis Varoufakis states it is the third largest party, a significant difference.


[Thanks! Good catch, Ed. I've fixed it now.--Econlib Ed.]

l0b0t writes:

Agreed; great guest, great interview. Thanks Russ and company. Listeners may be interested in this talk by Valve co-founder and managing director Gabe Newell at the LBJ School.

George writes:

After listening to this podcast and the podcast with Cathy O'Neil I got to wondering: Are there any notable examples of Regulatory Capture being reversed by peaceful means and without people having to take to the streets?

Sumeet writes:

Loved the interview, as most other interviews. In the transcript, there is a question mark on Deutsche Bank, wondering if he meant the Bundesbank. I think he did mean Deutsche Bank, as a proxy for investment banks that would have incurred losses in case of a Greek default.

Jim Feehely writes:

Hi Russ,

Another excellent discussion.

Valve Corporation, as Yanis describes its structure, is the way in which many more corporations should be structured. The entrenched corporate structure of top down hierarchy is all about obedience and imagined risk management, not performance. Only the rhetoric of corporations is about performance, underpinned by perverse incentives. That is why, in my view, corporations reached the limits of performance long ago.

What Valve does, however, challenges the very core of capitalism; i.e. the alienation of the product of labour from the owner of labour-power as the source of surplus value acquired by the capitalist (Marx).

Valve is harnessing the power of collaboration and cooperation. And humans are much more pre-disposed to collaboration and cooperation, than they are to competition. That is not to say that competition does not have a place, particularly in fair markets and price setting.

Russ, I also support your view of economics as a study of society, not a 'science'. If it is indeed a science, it has been proved wrong many times and not adjusted its assumptions in the face of that proof. As a science, conventional economics is a fraud.

Yanis also made a telling, abeit a throw away, point in respect of Smith, Mill, Marx, Hayek etc not getting a job in any American university because they are heretical to the dogma of neo-classic market economics. That is what is seriously wrong with the content of economics courses here in Australia, and I suspect in America. Alternative theories to the hegemonious and religious dogma of neo-classic market economics are simply not given any oxygen. And that seriously limits the options we consider in relation to economic problems.

Marx's technical analysis of capitalism remains, in my view, the truest and most rigorous. But Marx is ignored because of predictions he is alleged to have made (and did not) and the failure of the Soviet Union incorrectly associated with some imaged prescription Marx made for a society. For example, Marx did not advocate the state directed capitalism of the Soviet Union although most people, even very educated people, think that is the case.

As far as I have been able to find, Marx's fundamental analysis of capitalism has not been successfully rebutted. But we ignore Marx. It is time that we started listening to the economic heretics, Marx included.

Thank you again for another very stimulating discussion.

Jim Feehely

Dr. Duru writes:

This was a great podcast. Another example of a topic I may never have learned about if I were not subscribed!

Quick questions:

1. Varoufakis rightfully notes that Valve's model has yet to get tested during a contraction. I am wondering whether the founder has ever worried about this and whether he worries even more since he never thought his company could grow as big as it has. I imagine handling a contraction amongst 30 employees is a much simpler task than 300 or 400+.

2. Thanks for waking us (me) up to the fact that despite pronouncements from the ECB to contrary, the eurozone is not yet on the mend. The commentary was well-timed for what looks like yet another kick-off of dire eurozone headlines. I think it would be good to do a podcast that solely focused on the economic arrangements of the eurozone (or has one already happened that you can point me to?).

john thurow writes:

Excellent discussion! Thanks!

One thing that confused me about Valve is why base bonuses on peer review instead of based on a percentage of the profit the team or person's idea made for Valve? That is, if the team comes up with an idea why not allow them to keep some of the return for their bonus and get rid of peers deciding pay. The notion of peer review for bonuses seams foreign to Valve's ideals or this type of organization.

Michael Tolhurst writes:

As a gamer, I can fill in some of Valve's history.

Valve's first self-produced hit was in the mid nineties, a game called 'half life.' This game however was able to be modified by the user community into new games that developed a large online following. The first of these was 'counter-strike.' Instead of rejecting the community of 'modders' (ie. people who took the original product and changed it to a new purpose) they embraced it. Even today Valve encourages community development on the games and offers a forum for players to exchange player created content with each other.

The success of these early games put Valve as one of first entries into online gaming, and they expanded their distribution system to distribute other games. They traditionally have been very good about offering a 'marketplace' for smaller game designers to sell their products instead of simply selling the games made by large producers. They can do this because by 'shipping games' it simply means information transmitted over phone lines. So a small producer can try to sell their game on Steam, and if it does not sell, they and Valve aren't left with warehouses full of unsold merchandise.

As an aside, large producers like EA games have tried, unsuccessfully in my view, to mimic Valve's electronic marketplace.

Jake McKenzie writes:

As someone who has been a fan of both Valve and EconTalker since since their relative inceptions, this talk was surreal. I can confirm what Yanis said about Valve's platform steam. It's a pure monopoly and I'm fine with that because I couldn't see myself using any other platform to play games. From how robust the application is, to how easy it is to use, to how aggressive a marketplace it is, Steam has no match in the gaming space. It's no contest at the moment. PC gaming before Valve created this platform was dying, the 'home consoles' were forcing it out of the marketplace but Steam created a stable marketplace for developers of games to have a reliable place to sell games. Valve saved the platform really. I could talk about how good of a platform Steam is but as non-hierarchical Valve is as a company I want to stress the importance of the cult of personality around Gabe Newell. I don't know any other way to describe it other that to say as someone who plays games, and has for a long while, he feels like one of us. Russ the only equivalent I can think for yourself is hearing you yourself talk about Steve Jobs. He's ruthless, meticulous and he says everything I want to hear before I thought I wanted to hear it.

Also as for firing, they just laid off a lot of people companies would be falling over to get including Jeri Ellsworth who's a computer chip designer. Absolute legend but they laid her off so the anarcho syndicalist company structure is robust enough to incorporate firing. So many topics to talk about I guess I'll move to barter economy. I remember Steam's near liquidity crisis all too well and it's still a serious problem. In some countries like South Korea I believe there are obtaining so many electronic goods in their games that can be traded outside that the government of South Korea are treating it as income, and trying to tax it. Gabe talked about this at his recent Austin talk where he basically had to create forms for Koreans to fill out to pay a tax on it. Much of the liquidity crisis has been worked out via third parties run by fans who saw the problem, and decided to act. Basically creating an ebay of sorts so you could auction away your goods. They don't go for much often times but it allows breathing room for the market. Valve quickly saw the external auction sites and I believe that's around the time they added Steam wallet so that you could just keep money on your account to trade privately, circumventing these third party trading sites.

Russ a trend with Valve not brought up here is how they're now adopting what is known as "free-to-play" structure to pricing. Just google the term I won't bore you with all the details but what's interesting on that note is how they're finding ways of creating scarcity on free goods much like Apple. The most recent game by Valve, DOTA 2 (which was brought up briefly here) is gifted to users who buy products on Steam. The product is sold in the store and can be bought for 30$ but there are so many free copies in the wild there would be no need for any user to ever buy it if you have friends who buy games. There's two really important things going on with that right there. They're incentivizing community and they're creating a huge profit. Currently DOTA 2, a game tied behind this gift structure is the most played game with 600,000 concurrent users playing it and it's not even a finished product. This free product is in direct competition with a game called League of Legends which has been out for years. To unlock all the things in League of Legends that DOTA 2 offers for free, it would cost the user upwards of hundreds of dollars. Valve is slowly but effectively pricing League of Legends out of the market. I say they're in direct competition because they are in essence the same games. Since these games are considered sports(they have thousands of users watching competitive matches almost every day, rivaling television baseball broadcasts in volume), an apt comparison would be to say DOTA 2 is to the MLB as League of Legends is to NCAA Baseball. They are near identical products with minor variations. The most endearing thing about what I just said was that DOTA 2 was worked on because the employees of the company were fans of the original, it's a product that arose out of all the wonderful innovations that Yanis brought up in this talk and I really think only could have arisen out of such a structure. I'm not a fan of that style of game, many are but all the attention to detail in it is incredible. I hope copycats arise for the mediums sake but I doubt they can do it on the level Valve can, just because nobody has done what Valve has. Until they do I'll remain skeptical and pessimistic.

Shayne Cook writes:

A question for Yanis (and Russ):

What would be your assessment/comment for a recommendation that Greece (and other Southern European nations) adopt a "dual currency" model - with the Euro as the mandated foreign exchange currency, and the Drachma (or other locally issued currency) as the Greek-internal currency?

The problems of the Euro-Zone always seem to be expressed as an "all-in" or "all-out" proposition. I'm not convinced it has to be that way.

Under a "dual currency" model, the Greek Government recognizes both currencies - Euro and Drachma - as "legal tender" within the internal Greek economy. The Greek central bank issues Drachma, exclusively to the Greek government which uses those Drachma to pay its internal social obligations, and only as those social obligations become due.

Conversely, the Euro (also an accepted currency within the Greek economy) remains the required currency for foreign exchange obligations - by any economic entity in Greece, including the Greek government.

In order for this to work - and it would of course only buy time for Greek government and economy - the exchange rate between the Greek Drachma and the Euro would have to be very tightly controlled/coordinated by the Greek central bank and the ECB. And that control under the foregone reality that the Drachma would steadily and predictably devalue, as it is printed to service those Greek Government social obligations.

Under that construct, the Euro continues to serve as both a "medium of exchange", AND a "store of value", while the Drachma ONLY serves as a "medium of exchange" and then ONLY within the Greek economy, with no sustainable exchange value outside the Greek economy.

I'd be interested in your comments.

emerich writes:

Hey, this was the first time I've ever heard "alienation" used in the Marxist sense in an actual discussion. Cool! I think I understand the term better now!

Let's hope that the Valve model does have some wave effects and makes our lives more interesting. The skeptic in me fears that it's only possible by putting a very small (300?) number of very untypical people together to get very untypical results. I'll be much, much more impressed when a similar organizational model works in a company of 10,000 or 100,000, say the size of a middle tier bank. Or come to think of it, if 100 companies can made the model work with 100 to 1000 employees. Taking Yanis at his word, they're a bunch of geniuses ("at the top of their game"). Put them in any organizational model and they'd presumably do fine. If the organization had a monopoly, they'd do fine and make lots of money too.

Liam writes:

@Jake Mackenzie's post already tackled a lot of what I wanted to add here. It was awesome to hear this topic on the podcast and it's one I think needs a lot more attention. The digital gaming market is a really really interesting space right now with all kinds of experimentation and innovation: Things like Free-to-play business models, the push for DRM and the eventual realization of it's failures, the competition for digital market platforms, the world of monetized gaming streams, and to me the most interesting: the unmatched development transparency and collaboration around products like DayZ.

I would like to point out that there are alternatives to Valve's Steam platform. Probably the most successful has been Good Old Games (GoG), which has found success not through direct competition, but through building a niche for itself in old, out of production games. Their model is to take old games that are very difficult to find, modify them to work on modern PCs, and then sell them for very low prices (a couple of dollars each).

I listened to one of the founders of GoG on a podcast a few years back talking about how they purposely avoided competing with Steam because of how dominant they were; how ridiculously hard it is to try to enter that market. Instead it was a deliberate strategy of filling a niche that Steam was missing, starting out small, and slowly adding old publishers' catalogues as they proved that they were a viable retailer in and of themselves.

It's probably worth noting that the largest(?) gaming publisher, EA, has their own digital marketplace called Origin, which they started about a year ago. It is in direct competition with Steam. Since they are a large publisher, they have the rights to certain popular games which they will only release on their own platform, creating an incentive for people to use it who would normally only use Steam. Considering that EA publishes some of the biggest games out there (think big summer blockbusters in movie terms), they have some market power. However, the consensus among gamers has been that EA's poor customer service and often downright disregard for their customers makes it a second tier choice, with many people just boycotting the platform as a whole. If Valve didn't have such an amazing reporte with their customers, this would probably not be the case.

Richard writes:

Excellent podcast. I had never heard of Valve and was entirely unfamiliar with this alternative form of management structure.

As usual, Russ asked logical, probing questions throughout. One that wasn't asked on which I would have liked to hear Prof. Varoufakis opine is this -- which takes the question of replicability one step further:

Valve hires software engineers "at the top of their game" -- that is, who have been nurtured elsewhere and have a proven track record in their niche. Could a corporate structure as flat as that at Valve work in firms that hire a less proven and mature workforce, and seek to invest in and develop them? Such a firm would have a wide mix of skill/productivity levels which itself is not as "flat" as that at Valve. Or is a radically flat managment structure peculiarly appropriate or workable for an employee force that is relatively homogeneous and mature with respect to developed talent?

Mort Dubois writes:

If you only look at companies which are generating a billion in revenues and substantial profits, you will start to believe that any method of operation works. It sounds like Valve benefits from:

1) No need to manage a physical supply chain or interact with other companies/independent contractors on a time-sensitive basis. So loosy-goosy management works just fine.

2) Network effects and luck - they got into a particular space first or first/best, and uncovered a fountain of cash.

There are plenty of workplaces where the boss is a visionary who lets chaos prevail. The vast majority of them fail. On the flipside, consider Apple's success. A martinet in charge of a tighly locked down company. Most of the companies that work this way are failures, too. In fact, most companies are failures, no matter how they are organized.

People pay attention to the profits, and then rationalize that the particulars of the leader or organizational structure led to the success. I have my doubts about this. Luck is a huge part of every business success story. If you want to draw conclusions about what kind of organizational structure leads to success, it appears that most surviving companies have more management structure than Valve, and less than Apple. But the boring middle doesn't make such an attractive sound bite.

Russ, I really enjoy your interviews with business people, they help me develop my thinking on how to run my own factory. Keep 'em coming.

Jake McKenzie writes:

@Liam Steam's focus on DRM as a way of creating a cohesive experience and turn a former product into an evolving service and away from just a tool to prevent piracy is why it succeeded over other DRM platforms. That said EA's Origin is seeing the same growing pains that Steam did when it first launched. As someone who was around then, I remember all too well the torrent of hatred and vitriol toward Valve for creating a DRM platform for their games. Most of those people who hated the platform then because they were tied down by it are the same people who will fight tooth and nail to defend it now. I can see the same thing happening with Origin, there's no reason why in time that platform can't bare the same fruits. EA has been pricing aggressively(not as aggressively as Steam but it's aggressive), the platform is robust and as easy to use as Steam. The only problem is that EA needs more games on their platform, but that will come in time.

Seth writes:

I personally think the appearance of a hierarchical structure in many organizations is deceiving.

Much of what goes on in them is bottoms up and emergent and works almost like Valve. But, admittedly there is a component of folks who like to lord over others who think they are doing something productive in doing so. Those folks also have a penchant for jumping on the emergent successes and changing the narrative to make them look more like designed success -- under their design, of course.

I enjoyed a movie that came out a while ago called "Waiting". It's a comedy, day-in-the-life of a themed restaurant like Bennigans. At the end, the dishwasher that happens to be burned out psychologist points out that the restaurant was about to be closed, corporate sent a new manager (who is a buffoon), but that manager happened to hire a cook who brought into work a game he played at his other workplaces.

That game then made work fun again for the workers and improved their service and engagement. More customers started coming in and the restaurant was a success.

Corporate thought the manager was a rock star. But, the success story was really deeper than that.

I find that many success stories in the organizations I've been involved with tend to have root causes more along the lines of this, but most people settle for the easier explanation. Which, of course, can't be replicated.

Jeff Kramer writes:

Morningstar Co. is the world's largest tomato ingredient processor, a company type totally unlike Valve's, and runs in much the same way (since the 70's!) -

rpl writes:

The question I wish someone would ask in one of these Valve interviews is, how do you make sure that the thankless chores that nobody wants to do actually get done? I get why Valve doesn't have to sweat too much over their core functions of making and shipping great games, but how do you ensure that the trash gets emptied and the toilets get cleaned? Sure, you can contract those things out, but that doesn't actually solve the problem because contracts have to be managed. I can tell you from painful experience that few things are more soul-crushing to a creative person (or less appreciated by their peers) than administering a contract. Do game designers at Valve really volunteer to take time out from developing games to bawl out vendors who screw up their invoicing?

On a related topic, how do you ensure that the unsung heroes (the guys who you never notice unless something goes wrong) are appropriately rewarded? The best computer sysadmin I ever worked with looked to casual observers like he just sat around goofing off all day. In reality he had automated most routine tasks, and when something broke he sprang into action and got it fixed, usually before the users noticed it was down. Would a guy like that get good peer reviews at Valve?

lloydfour writes:

Heard that some of the Euro countries are using internal devaluation to cope with their problems.

Does Valve structure work because the are small enough where they can still cherry pick employees who work best in low structure enviroments?

mariusz writes:

here is a joke for you from the good all "workers' paradise": a guy if considering a job in Siberia, he hears that pay is great, you get a car, a house, and women are pretty and friendly. his friends are not so sure. they have heard of some who went there and never returned. the fellow decides to go. his friends ask him to write often (no internet then) and they agree on a system - if all is fine he writes in blue ink, if not he writes in red ink - keep in mind you cannot write bad things. so he goes, the 1st letter arrives - he is extatic, shops are full, the car is great, the job is fantastic. the 2nd letter is the same, his friends are begining to make plans to go there, then the third letter comes - things are great, you can buy everything but the red ink :)

any case - the fact that there is a goofy manual out there does not mean that all you do is get massaged and do what you like, you have to deliver with all its attendand stresses, office politics etc,

does the company have a pension plan, health insurance, life insurance, maternity/paternity leave policy?

have they reinvented the good old consultant system all over again?

Brit writes:

Thank you so much for these shows, your guest curation is second to none. Somehow this is the most comforting (yet intellectually simulating) radio show on the internet

Lewis writes:

I wonder how flat Valve's structure really is? Does the receptionist or cleaer work on the same basis? Or is some sort o technocracy where the technical people are really the boses?

Rob writes:

Huh? Valve are making games still? Not just distribution?!

Raja writes:

This was a really interesting topic and a good guest.

I wish the interview had gone into more detail. This is the kind of situation where you really need a fully detailed step-by-step breakdown of specific cases (e.g. hiring/firing, compensation, etc) to understand how the model works. The interview still left us with the impression of a magic black box. (eg. Programmers + secret anarcho-sauce = $$$!)

I would have also liked to hear some analysis of how this structure has affected their competitive position within their industry.

Also how does this model stack up against the Valve founder simply seeding a number of different start ups in terms of incentive alignment, economies of scale, etc?

Varoufakis was a really interesting guest and I would like to hear more from him. Loved his comment to the tune of "we're a priesthood with equations." I was quite surprised to hear a European economist who was clued in to the ongoing disaster there (and here) as well as sympathetic to the Austrian view. (Although he claimed not to be in that camp.) I was 100% in agreement with him until his suggestions at the end on solving the crisis - which essentially boil down to "let's trust govt to do the right thing this time."

Russ's comments on war were apt. I doubt that Europe has war in their blood any more though. They've been fully domesticated and I suspect we have too.

I would love to hear more podcasts on this theme of innovative organizational or incentive structuring. As part of the micro-economics oriented part of the field I think it's one of the few areas where economic thinking can add value, unlike macro-economics whose primary purpose is value destroying tinkering in the lives of strangers.

Brian writes:

Sorry but I hate Steam, it is my least favorite gaming marketplace. Yanis Varoufakis does not know the online gaming market and is wrong on what peoples opinion of steam is. The only reason I use Steam is many games I want require me to have it. I think if many other gaming companies did not require the use of Steam for their games, I don't think they would be nearly as big player in the online gaming market.

David Johnson writes:

I am with Brian. For the few games I play that go through Steam, it's extremely annoying. These games are NOT Valve games, they are from other companies that use Steam as their distribution channel.

The description of Valve as some sort of Hayekian utopia is bizarre, because they appear to be nothing of the sort. While not as bad as EA, Steam's DRM culture still treats the consumer as a potential thief who has to constantly prove himself worthy of using the product he has a sales receipt for.

Alai writes:

A transcript correction: "crime against money" should be "crime against humanity". A funny mistake to make, in that context!

[Fixed. Thanks! I knew "crime against money" seemed kind of a humorous turn of phrase, but that's how my ear heard it and it seemed possible, so I didn't take the time to re-listen to that patch.--Econlib Ed.]

R.A. writes:

A very interesting topic.

But I really wonder how much Valves organization is dependent on the fact, that their work consists of producing things they really like. Making games is more a hobby than "real work" for most programmers. Many do it without being paid at all.

So if a Valve-like company would be successful in producing something boring like toasters or chemicals, the story would be much more convincing. (Some commentators tackle a similar question when talking about who cleans the toilets at Valves office and the like).

Concerning Greece's crisis I think Yanis misses one important fact: The fresh money provided by the other european states is only partly used to pay for old debts (which he rightfully describes as a subsidy from the taxpayer to the banks).
The larger part is used to cover normal government expenses - which are still higher than the tax revenues.

So if Greece would have declared normal bankruptcy, the impact on the population would have been much worse - thus the Nazis and Communists would have profited much more from the problems.

Comments for this podcast episode have been closed
Return to top