Catalin Alexandru is a behavioural game economist with over 16 years of experience designing economies and ecosystems. He has worked as a game designer, analyst and advisor across every business model for the biggest global game companies and brands. His focus is building sustainable economies based around user generated content (UGC), which he thinks will become the dominant industry business model in the future.

In recent years, Alexandru has been deeply involved in the Web3 gaming space, working on virtual worlds based on tokenized economies. This led to his seminal article “True Ownership – The Game Economy Trilemma,” in which he outlined structural issues with the concept of “true ownership” in Web3 and offered a model to approach a solution.

In his newsletter Intrinsic Economies, Alexandru writes in-depth essays on the economics, strategy and psychology of video games. Catalin also created Figonomics, a Figma plugin that turns mockups into interactive game economy simulators.

I sat down with Catalin to discuss his views on the Web3 gaming space, the challenges it faces, and what’s next for virtual economies.

Web3 and the Challenges of Creating Open Virtual Economies

Dirk Songuer: Hello Catalin, thank you so much for taking the time, I really appreciate it!

Recently I had the pleasure of interviewing many people around economies in games and virtual worlds, and often there was this question how to differentiate an economy in a virtual world from an economy in reality. And if it even makes sense to differentiate the two. For example Edward Castronova argued that virtual worlds behaved just like any other market from an economics perspective and that the only differentiation is the goals the designers set out to achieve with it – real-world economists remove friction, while virtual world economists introduce friction as part of the experience.

I am very interested in your perspective on this differentiation, especially with your background in Web3, which is often portrayed as trying to blur the lines between real and virtual.

Catalin Alexandru: Oh, wow, I mean this is like a10-hour conversation at least. Obviously Edward Castronova is a huge pioneer in this in this space and has produced a lot of seminal work. Personally, I’ve engaged much more with the work of his students. One of the more prominent ones is Christopher Kaczmarczyk-Smith. He is the chief economist at Star Atlas, and he’s on a podcast with Phil Black and Eric Guan. Kaczmarczyk-Smith has also established the IGDA Game Economics Special Interest Group. I’ve been very aware of what they are doing and the conversations that they are pushing in game economics.

My main issue, and the reason I bring all of this up, is that they very much encapsulate a way of pure economics thinking. And that is indeed about removing constraints, making economies more efficient, and monetizing and financializing game systems as much as possible. It’s about monetizing every single bit of content to the largest extent possible. This aligns very well with the Web3 movement because that’s also built around financializing and monetizing as much as possible.

But for me, because I have an education as a game designer, this thinking hits a wall very quickly. In traditional economics, you have all these goods, all these raw materials in the ground or as cargo that needs to be moved from place to place. If you can affect policy or make proposals that can make this frictionless, or transport cargo faster, then that’s always a net positive.

But in game economies the demand and supply are not presupposed. It’s not as default as traditional economists like to believe. Because games are not systems where for thousands of years game swords have been as useful and deeply embedded into culture as for example wheat or gold in the real world. There are no real-world equivalents or analogues to manufacturing game swords. The demand for game swords is created out of thin air.

So, this idea that you can come in and take this manufactured demand and then just optimize it to get the game sword to the player faster, and get more money for it, or split it into 400 types of game swords where one of them is blue and another has a funny shape, that’s not something that can scale. The reason it can’t scale is that the initial demand for that sword depends on a closed system, the game itself. And traditional economies are not closed systems.

Dirk Songuer: Why do you think that all games right now are closed economic systems?

Catalin Alexandru: Purely from an economics perspective, all successful games right now are closed systems. I’m happy to debate it, but I don’t think it’s debatable. The definition of an open economy is one where you accept currencies from other economies and get goods and currency from your economy into other economies.

Right now, I don’t think there are any games that are successful in doing that. And by success, I mean that they can keep their player base, that they can make a profit, and at the same time trade with other with other game economies.

Dirk Songuer: What I mean is that Real Money Trade exists, where you can sell or buy virtual items from virtual worlds for real world currencies. Doesn’t RMT make a virtual economy an open one, isn’t that kind-of participating in trade with other economies?

Catalin Alexandru: I think game economies are closed because you rely on another country’s currency to affect those trades between economies. There is no game where I can trade the currency in my game, let’s say gold, for EVE Online PLEX directly. Or maybe there is, but this would be something so experimental and tiny that it’s not really on the public’s radar. There is no currency exchange market for that. I need to go through fiat.

And the way that works a lot of the time is through grey markets. It is done under the threat of banning your accounts and with a lot of enforcement around it. If that was the way we would do trade in the real world, that would be something like piracy or smuggling. That’s not how open economies interact with each other in the real world.

Dirk Songuer: And that is something that Web3 games specifically wanted to change, correct? Based on interoperable blockchains that every game could use and seamlessly exchange currencies and goods on.

How did you personally get into Web3 gaming?

Catalin Alexandru: I’ve had a very varied career. I’ve been in mobile, console, web, free to play, premium, you name it. And back in 2020 or 2021 I ended up in Web3. Even before that I was working on some projects that were very much like Web 3. I was working on MMOs that wanted to have tokens as a means of real money trading in their game. But that was very experimental, very early. Long before the term and things like NFTs really entered the public consciousness.

At the time I was at an agency, building a team that was trying to create the bedrock of theory and models to help our clients with game economy design. And because I was working in the Web3 space, I had conversations with probably most of the big Web3 game developers. A lot of the issues that these developers were facing in terms of economy design around ownership and open economies in games were old problems that had been attempted before.

We knew that these problems weren’t necessarily solved and weren’t even necessarily solvable. But it became very clear that the Web3 community thought that blockchain technology would solve all these issues. Since you have all this “amazing” tech, people believed that it would for example guarantee ownership. And, well, this belief can bring a version of that into reality.

A lot of “technology” is just believing that it can do a thing. And it ends up doing the thing because you believe that it does. That’s how real currency works. That’s how our financial system works. A lot of it is based on belief. And people in Web3 thought that believing in this technology is enough, and that it all in the end would just kind of sort itself out.

Dirk Songuer: But in March 2023 you wrote an article called “True Ownership – The Game Economy Trilemma,” in which you outline structural issues with this concept of “true ownership” in Web3 gaming.

“The real problem with “true ownership” is the one also known as the Diablo Auction House problem. This is the issue that has kept game economies built around secondary markets from exploding into mainstream gaming so far despite considerable investment most infamously from Blizzard and most recently from the glut of Web3 game startups riding the 2021 NFT craze  like Axie, Sandbox, Decentraland & co. What makes it so hard to overcome is the fact that under the “true ownership” idea that launched a thousand Discords, both audiences and companies want a system that achieves three separate goals, any two of which are contradictory to the remaining one.

This problem format is popularly known to most as a trilemma and to economy geeks as the ‘Impossible Trinity’.”

/ Catalin Alexandru, “True Ownership” – The Game Economy Trilemma

Dirk Songuer: How did you come up with this model?

Catalin Alexandru: The market started crashing. Reports showed that most of the big Web3 games had not that many players and were basically run by bots. For me, it all fell into place at that moment. I realized that “Oh, wait, these problems weren’t actually solved!”

But then I asked myself “Why did everyone think that this problem, that was publicly discussed at length with the Diablo3 Auction House debacle, would be solved by technology alone?” Because Blizzard has smart people, they have economics and monetization specialists, they have technologists. If anyone could solve these problems through just technology, they would have solved it. It was fundamentally an economic problem, one of demand.

That’s when I started looking at different models that are actually successful at doing this, like Counter-Strike: Global Offensive (CS:GO) and all of the other Web2 giants that have a real money economy. I was trying to understand why they worked. And from there, I worked my way to understand how that trilemma worked.

I had many good conversations with Eric Guan as well. He was helping me to flesh it all out and acted as a kind-of editor. But the main thing was that this problem has been obvious for many, many years. And with the article, I wanted to answer why technology didn’t solve it.

Dirk Songuer: In the article you condense a lot of best practices into a very simple model: “Asset price floor, mass market user base, progression utility – your virtual world can only have two.”

Oversimplified, “Asset price floor” is the price stability for resources over time, avoiding for example that acquired items become worthless, or that items get so expensive and exclusive that it prevents new players from entering the virtual world. “Mass market user base” is the notion that the virtual world can scale to millions, maybe even billions of users. And “Progression utility” is resources that help players achieve the goals of the game, or what they perceive to be the goals.

Looking at these three, you said that “mass market user base” seems to be the one that Web3 games are struggling with.

Catalin Alexandru: One mention, the trilemma refers to formal progressions and specifically excludes informal progressions because they are too vague and very rare in mainstream game industry products. Also a mass market  userbase is what everyone wants. That’s what Web3 is promising. I mean, who is starting to develop an MMO and goes: “It’s a niche MMO and I’m just gonna have 50K players and that’s it.”

It’s in the name: “MMO” – Massive multiplayer.

Dirk Songuer: But here we can also look back to the very first examples of MMOs. Arguably the first massive virtual world that even sort-of coined the category “MMORPG” was Ultima Online (UO). And it was a famous example of an economy that failed multiple times before they managed to stabilise it.

Some of those failures were specifically around scale. The UO team first created a simulated, closed-loop world with a fixed number of resources, but then had to cope with a fluctuating number of users, which also started hoarding basic resources in their inventory under the assumption that they might need them at some point. This in turn destabilized the prices, and suddenly consumables that were required to play and progress in the game became a valuable speculative asset, derailing the entire game.

Ultima Online launched in 1997, and as early as 1999 we got academic papers analyzing what worked and what didn’t work in its virtual economy. That’s almost 25 years. I guess my question is why haven’t we figured this out yet? Earlier, you said all this was obvious. But is it?

Catalin Alexandru: Well, if we’re talking about Web3, that’s an easy question to answer: Because people were only interested in taking VC money, put it into something that seemed attractive, pump up the token price, then sell it and pocket the proceeds. There are very few people who really wanted to build scalable economies, who were in it to push the media forward and revolutionise game economics.

This was an expensive lesson for me to learn. For a time, I really thought that it’s all just a matter of finding the right people that want to figure out a new economic model and move from free or play into what I see as the next revolution of monetization. Which is user generated content that’s monetized directly and incorporated into the game.

So, the answer when it comes to Web3 is money. There was no other way to make money, and there still isn’t to my knowledge. I kept up with most of the popular Web3 games and there is nothing that’s even close to scaling. Or to having what I would call a sustainable model.

Dirk Songuer: Why is scale such a big problem for keeping the price for resources or the power-level of items relatively stable over time? Ultima Online eventually managed it, and games like World of Warcraft achieved massive scale with somewhat healthy economies. Is this something Web3 specific?

Catalin Alexandru: The fundamental problem is: How do you preserve the fun aspects of a game? I wrote about this in another piece for my newsletter Intrinsic Economies called “User Generated Content, Real Money and the Investment vs Learning in Games Dilemma.”

When you want to have an investable game, then there is a conflict between gameplay and stability. This is a fundamental, kind of wicked, probably cursed problem: Gameplay needs to be predictably learnable.

But if you make anything that involves investing real money to get a return predictably learnable, then the house will lose and get completely wiped out. Because when players can learn how to extract money out of a system, then there is no way to stop them. Or rather, the only way to stop them from extracting money is to make the system unfair.

The most common way to make this system unfair is to replace skill with randomness so that they can’t learn it. It’s rolling the dice if I will get money this time. And now you turned your game into a casino which is what many Web3 games did. There are many, many Web3 games that look like RPGs, or they look like MMOs, but they are actually random drop casinos, and legislation is catching up with them.

The other way to keep the game part predictably learnable is to turn it into “fun work.” You know, as Raph Koster described it in his book “A Theory Of Fun For Game Design:” “Learning equals Fun”. But then the question is: “What kind of work?” What work needs to be done inside the game that is fun and deserves money?

I’ve had conversations with Daniel Paez about this. He had an economic model to around people in emerging countries earning money by being trainers and guides, or farming loot for affluent people in wealthier countries. And to me that’s just such a dystopian view of what games should be or can be. I’ve never seen that as an answer because, you know, we’re seeing that it’s not a great way to make a living. Nobody’s really producing anything or contributing anything to society or to an economy. They’re just wasting their life away.

Dirk Songuer: This also seems like a lesson already learned. Julian Dibbell has written about negative externalities in gold farming in his 2006 book “Play Money: How I Quit My Day Job and Made Millions Trading Virtual Loot“: If you have real money trade and items that are truly worth something, then the most economic way to extract value from this system is labour exploitation in third world countries. And then the question becomes if it is desirable to turn virtual worlds into exploitative structures.

Catalin Alexandru: Well, this is it. Why would you want that? I mean, at the moment we have a version of that and nobody’s happy with it. Except maybe World of Warcraft gold selling sites, I guess. But that’s about it.

So, essentially that’s the fundamental problem: How do you preserve a game? How do you make it learnable and predictable and fun, while introducing a profit motive? And I strongly believe that there is a way to do it. But outsourcing the “work” part to people in third world countries to then ship the “fun” part to wealthier players, yeah, I don’t believe that’s the way.

I believe Fortnite is starting to approach a better model, but it’s still very top down. It’s still built around the old model of doing things in free to play. The work you’re doing is still very commoditized. There is no real stake for you to make your own island in Fortnite. It disappears into a sea of a million islands. If you want other people to play it, you need to further spend money promoting it. And it’s very unlikely you will make that money back.

All the winners are islands that are basically clones of other games. They are popular for a while and then they fall off. Or islands from people who are famous in another medium, for example celebrity Twitch streamers or whatever, and their community is the reason that their island takes off. But none of this is on its own merits. At the end of the day, all this output is not producing the amount of economic value it should be producing.

Dirk Songuer: These types of experiences seem more like a media channel than a virtual world.

Catalin Alexandru: Yeah, it’s essentially an App Store where you can put your app. Is that a virtual world? You can probably make that argument, but I don’t think you could defend it very well. Because it’s not contiguous, there is no real coherence to it at all. You know, virtual worlds are virtual stores, but not all virtual stores are virtual worlds.

Companies right now want a certain type of economy, you know, certain value chains and things that that intersect with each other. We know how that’s done, you can go and make a complex and good economy like that. It seems like we’re going to have a cycle of at least 5 to 10 years where it’s going to be these “Unreal Engine for Fortnite” and Roblox-type of platforms. But these are very closed down, very top down, and very controlled.

And some people are already moving away from this top-down understanding and instead are looking for another good approach that could be the bedrock for an economy more open and based around user generated content.

Dirk Songuer: One question around Web3 games that has been nagging at me for a long time is related to this. One of the learnings we had from Ultima Online was concerning its initial closed-loop economic system. Due to its dependence on a fixed supply of resources, it imposed constraints on the design team, with player behaviour eventually leading to extreme volatility. Since then, the industry regarded internally open, dynamic economic systems with faucets and sinks as best practice, just like we went from the gold standard to fiat & monetary policy in the real world.

But then, Web3 and blockchain games in general are based on cryptocurrencies, which by definition are always finite supply. Wasn’t the learning that building economies around fixed supply resources is hard? And going back to your trilemma, maybe even impossible to stabilize at scale? Why did Web3 start with that self-imposed challenge?

Catalin Alexandru: Many Web3 proponents think that it will combat mudflation. I’m not talking about inflation of the token price, that’s obviously because the token is so scarce that the price will keep going up if demand goes up. I’m talking about power creep as you progress in the game, the inflation of progression utility or rather the inflation of item prices via which power is expressed. Usually  and famously that’s a big MMO problem. When you progress in the game, your power increases but you end up having the necessity to buy more and better gear, because all your previous gear becomes useless. Also called mudflation.

From what I’ve seen, the thinking in the Web3 space is that if your progression depends on a scarce token, and that token keeps increasing in value due to scarcity, then you have to spend more and more real money to progress. And if you have to spend tokens, which you buy for real money, that will counteract power creep and mudflation because token price inflation and token gating acts as a deterrent from constantly replacing gear.

That’s not even getting into NFTs and putting NFTs on secondary markets and having to buy new NFT every time you level up or get access to a new area and need better weapons.

All that just defeats the entire purpose. I mean, power creep is a very easy thing to combat for anyone who’s ever designed an MMO. But this is the exact thing: Most people in Web3 have never designed one. Power creep / mudflation is a fundamental effect of a functioning formal progression. What is a lot harder to solve is what to replace the progression with once you get rid of it and replace it functionally with a series of crypto paywalls.

Instead, they come into these systems, this genre that someone else has invented, and then they bolt tokens onto it, and that supposedly makes everything better. But all it does is replace the power creep / item cost inflation problem with a different worse problem.

And they also disregard that tokens aren’t anything new. We’ve had tradable currencies in virtual worlds for a while now, and we’ve had to limit them because it damaged the online economies that did it.

Dirk Songuer: What do you think is next for Web3?

Catalin Alexandru: I would even question how useful it is to keep talking about what’s going wrong with Web3.

I mean, at this point I don’t know that there’s a lot more to learn from it. Web3 has become an evolutionary dead end for games. Sure, it needs to play itself out, but it doesn’t really have anything to do with the games industry anymore. It has more to do with the cryptocurrency industry, with the price of Bitcoin, with financial regulation.

All of those externalities radiate into the price of the tokens and the liquidity of these companies. So, at this point there is no connection between the token valuation of a game and how good it is, how successful it is, you know, any of that. Just like three or four months ago, the price of Bitcoin started going up again. There was no new utility, no new technological breakthrough. It was literally just Bitcoin going up in price and suddenly Web3 is back, and everybody is talking about Web3 gaming again.

More money came into the space and now everybody is trying to milk a little bit more out of this comatose space. And once Bitcoin drops off again, it’s done again.

To me, it seems that Web3 has completely failed its potential. The money was not going towards research into developing a new economic model, which is what we would have needed.