Gaming’s Great Downsizing and the Garage-Band Game Economy

Brendan Keogh explains the post-2008 collapse of the Australian games industry and how it might prefigure changes coming for the rest of the world

Last month I wrote this essay called The Games Industry is Deprofessionalizing, and it got some people a little worked up. Dudes in suits are fussing at me on LinkedIn, which is really what I deserve for logging into LinkedIn.

But I’ve been having fun too, especially after Bryant Francis from Game Developer interviewed me about it for his own article that wound up going viral on Reddit. This is the dirty thrill of blogging. Sometimes you throw a rock in a pond and it ripples in strange ways. On Monday I opened up Bluesky to find an Australian games professor getting into arguments about the idea. That led to me talking with him for an hour about the collapse of the Australian games industry.

I’ll get to that in a sec.

First, a quick recap: My argument was, basically, that we’re seeing some sort of mass fragmentation happening at the big-budget, triple-A end of the games industry. Many major games studios (particularly Western ones) are struggling. They’re delivering work that’s failing to excite players, missing sales targets, and laying people off en masse. But also, some small game dev teams are crushing it. Some are seriously competing with the big dogs on a level that wouldn’t have been possible five or ten years ago. Both of these trends seem set to continue and intensify going forward. And though I wouldn’t say one trend is causing the other, the two phenomena do seem interconnected.

That’s kind of it—just an observation of a set of things we can all clearly see happening. Now, there’s a temptation to append an easy moralizing takeaway to this observation: Here’s why it’s good. Or and it’s bad, actually. But I’m not so sure.

It is not obvious to me that the correct and morally justified final form of the games industry is one where individual games publishers have thousands of employees. I worked at these companies for the better part of a decade. They can serve as an incubator for talent, but just as often they mismanage the productive, creative years of hundreds of people on projects that won't even get released.

So there are a lot of interesting debates you could have about what’s happening in games right now. But one particularly interesting reaction came from Brendan Keogh, an Associate Professor at Queensland University of Technology who wrote this about the Bryant Francis piece:

The follow-up post in Keogh’s thread was even more widely-shared:

I had a mixed reaction to this the first time I read it. Yes, platform holders like Valve take a 30% cut on basically all PC games now, while individual devs are often forced to self-finance. I also agree that the 30% platform take is so low-risk that it qualifies as rent. But the comparison to Uber drivers is imperfect. To nitpick: Uber drivers earn at a relatively predictable rate based on the amount of time they spend driving, whereas pay for indie game developers is massively variable based on sales. Independent game dev is more like other types of content creation—you have a shot at making a ton of money, but most people make nothing or close to nothing.

But that’s just a quibble, and Keogh is onto something with his larger point. Another post in his thread won me over and made me laugh out loud:

I know exactly what he was feeling here. You try to point out a negative outcome of a systemic change and everybody rushes to balance the scales by pointing to the good outcomes. There’s this tendency for all conversations on social media to reduce down to a cheap takeaway like “this is good,” or “this is bad.” Or, “I’m for it,” or “I’m against it.”

I share Keogh’s frustration. Almost all changes, economic or otherwise, have negative and positive externalities. The impacts of the deprofessionalization of the games industry are bound to be kaleidoscopic, and we can only see the patterns if we turn the tube.

Luckily, Keogh has thought more deeply about this topic.

As it turns out, a few years back Keogh wrote a book called The Videogame Industry Does Not Exist (2023), published by MIT Press. Keogh interviewed over 400 game developers for the book, and he wound up identifying—in an academic style—a lot of the trends you might categorize under “deprofessionalization.”

The book’s abstract summarizes its core argument well:

The videogame industry, we're invariably told, is a multibillion-dollar, high-tech business conducted by large corporations in North America, Europe, and East Asia. But, in reality, most videogames today are made by small clusters of people working on shoestring budgets, relying on existing, freely available software platforms, and hoping, often in vain, to rise to stardom—in short, people working like artists. Aiming squarely at this disconnect between perception and reality, The Videogame Industry Does Not Exist presents a more accurate and nuanced picture of how the vast majority of videogame-makers work.

The book starts off by describing the game dev scene in Keogh’s native Australia, then pulls together stories about the ways game development is already deprofessionalized in much of the world.

One of the data points that anchors The Videogame Industry Does Not Exist is the surprising fact that, even as far back as 2021, more game developers worked in studios with 5 or fewer employees than in the 250-person-plus mega-studios we often think of when we talk about the games industry.

From GDC’s 2021 State of the Game Industry Report

If the games industry is actually deprofessionalizing—fragmenting, becoming more like the creator economy, however you want to put it—Keogh called it early. And that’s because he saw it happen in Australia first.

What follows is an edited transcript of a conversation I had with Keogh this week, in which he offers an extremely sharp and well-informed analysis of what happened when the Australian games industry deprofessionalized. He also has a very interesting hypothesis for why western AAA studios are struggling to generate demand for new games.

---

PUSH TO TALK: Could you talk a little about how you came to write The Videogame Industry Does Not Exist?

BRENDAN KEOGH: I really started thinking about the core idea in the early 2010s, and then did fieldwork and interviews from 2018 to 2020. I interviewed about 150 game developers in Australia, and some more in Montreal, Amsterdam, Berlin, and in San Francisco during GDC one year. This was all pre-COVID, you know, the end of the 2010s. I was meant to do some in Southeast Asia, too, but travel restrictions ended that. But I managed to do a few online.

And it really started with thinking, really looking at the Australian games industry, which at the time seemed quite different and exceptional to a lot of the research and business analysis I was reading. There was this implicit presumption that the game industry is made up of all these large companies with lots of money. And this was very different from what I was seeing, which was a lot of young people around Melbourne just trying to make cool stuff.

I was friends with the team from House House who were making Untitled Goose Game, which went viral in the late 2010s. I just knew them as these guys who were part-time, making a game called Push Me Pull You. They were just part of the local scene.

And so I was thinking: what does the game industry actually look like? How do you account for these small teams who are kind of professional and kind of not. Some of them were simply hobbyists until they went viral. And so suddenly it became a full-time job for them. I suppose that can happen to musicians and painters—anyone can go viral these days. But I was just really just trying to figure out: what's going on here?

The book starts out examining the Australian game scene. You make this point about how the 2008 financial crisis sort of wiped out the local games industry in Australia at the time. I would love for you to talk through what you think happened.

Yeah, I'm glad you brought that up. When I was reading your newsletter and some of the conversations about deprofessionalization, I was like: hasn't this been happening for ages? And it got me thinking about how much the current crisis is just North American game developers encountering what the rest of the world has been dealing with for a long time.

I try to make a really clear point in the book that the AAA games industry is an anomaly, both historically as well as geographically. Apart from these specific parts of North America, Western Europe, and Eastern Asia, the vast majority of games are made in very small studios under very precarious conditions. And yet I think because the American gaming industry specifically is so visible, that's kind of where our attention is drawn.

So in Australia, before the global financial crisis (GFC), the Australian games industry was essentially a hub for outsourcing. We called it work for hire, but we had a lot of mid-size studios—think anywhere from 20 to 200 employees, I'd guess—who were basically cheap English-speaking labor for American IP owners. So, you know, we would make Shrek games or Star Wars games or Olympics games. A lot of Bioshock and Bioshock Infinite were made in Australia. L.A. Noire was made entirely in Sydney. And there was a whole lot of shovelware. So it was more commercial than creative. And so what happened with the GFC was essentially the American economy tanked, our economy didn't, partly because we were selling a bunch of coal to China, partly because of various other things.

The Reserve Bank of Australia has a bunch of interesting charts and info about its history of coal exports. Keogh's summary above checks out.

What that meant was the strain of the American dollar hit parity, which was great for buying stuff online, but it was bad if you were an export industry, and the Australian games industry was an export industry. We exported game development. So it essentially was no longer cheap to send your game development to Australia. And the industry collapsed. That's pretty much it in a nutshell.

The GFC was the nail in the coffin for contracts coming to Australia. So post-GFC what do we have? We have a whole bunch of unemployed, very technically skilled game developers. We have these new platforms like the iPhone, Steam, and a few years later, digital distribution for consoles. We have affordable tools like Unity and later Unreal. So what's everyone gonna do? They're gonna start their own businesses. They're gonna go indie.

And some of them are saying, "screw the man, I'm gonna go indie because I want to actually do something creative." But what's kind of hidden behind that narrative is, you literally didn't have a choice. It was go indie or go home. It was: If you wanna make video games in Australia, you will be indie.

A lot of other countries are in a similar spot. If you go to Brazil, or if you go to Greece, or if you go to Malta, maybe you can find some studios there that are larger—but not that many. You're more likely to find indies or hobbyists. And how are they gonna get games onto the global market? Well, historically they just probably didn't. But now Steam exists, the iPhone exists. So going indie is your only choice in the majority of the world.

I think what's happening in America now is similar, except now we've got an additional 20,000 or 30,000 game developers wondering whether they should go independent.

For a lot of the people who had been doing work-for-hire game dev stuff in Australia, do you have a sense of how many ended up leaving the games industry altogether?

I don't have a percentage sadly. I've got my qualitative data. But a lot of people in games went into tech and programming. Or just retrained and went into other areas. A lot of people went overseas to join AAA games studios and have never come back, because the skills you need on large-scale games are very different from what you need in indie. So there are still a lot of Australian game devs all over the place. Some are kind of looking for a pathway back to Australia and still can't see it.

There was this really interesting thing that happened after the GFC. There were these first few teams that kind of had their breakout hits, then started to scale rapidly again. Probably the best example is Halfbrick Studios, which most famously did Fruit Ninja and Jetpack Joyride. My understanding is that they scaled quite quickly after releasing those games. And they released a bunch of other games since, but I don't know how many of those made a profit compared to those two games which were really made by one or two people.

Well, and the top charts in the App Store ossified completely. Fruit Ninja was there for like a decade.

Yeah. A lot of people who went into games after the GFC assumed that the model of how you run a successful studio is you rapidly scale. Except actually, scaling doesn't really work for indie game development. And so many collapsed again.

But if you look at, for instance, Team Cherry or House House, they had a very successful game and then stayed at the same size. And that's because that's how this model of game development was meant to work. Now there's a massive gap between what you need to make an indie game and even a smaller AAA game. You either need like five people, or you need 500 people..

There's this no man's land in the middle. You don't transition from Indie to AAA, except in very exceptional circumstances. Even a successful Indie game company is a different type of company. And so what that has meant in Australia for the last 15 years or so is us trying to figure out: How do you build an industry around companies that don't scale?

And so for the government doing public funding for games, this is a big question. For many years they thought they should fund games because there would be more jobs. And that’s what the industry was telling them, of course. And in some ways that's true, but by giving an indie developer money to make a game, you're not planting the seeds of a AAA studio.

You're not creating a bunch of jobs.

Right. So the answer is that it's more like a scene, right? I keep telling government funding bodies here in Australia and others that historically, the model we've thought of when we look at games is that it's like Hollywood or it's like the tech industry—in terms of how we think about scale and growth and resources. But I think a much better analogy, especially in this kind of deprofessionalized, fragmented, indie-centric games industry, is the music industry.

There's a bunch of bands. There's a bunch of kids in garages making stuff. And some of those will be very successful. But when they are successful, they're not gonna hire 50 more drummers. They're going to make another album. They're gonna buy some better equipment. They're maybe going to get their mate's band to be their support act and help them up as well.

So then the natural question, I guess, is how do you start a music scene? Britain did pretty well with this in the 60s because the welfare state was so strong that you could just live on welfare and make music. But… I don't know. Is that the right model?

Yeah, I mean, actually on that point, when I did interviews in Australia, what you hear is, "I'll try to go indie for my 20s and I'll see what happens." And then when I interviewed Americans, they were like "I need health insurance."

Right, right.

I will die if I don't get health insurance. So the idea of just spending your 20s figuring stuff out becomes a lot harder to do when you don't have the social support structures that allow you to do that. If you had things like free health care or a free education or a universal basic income, suddenly people could take a lot more risks creatively. And maybe you will get a music scene or an art scene because people can spend their 20s f—ing around rather than frantically trying to get a job before they die.

That's an amazing point. And it makes the possibility of America going the Australia route in terms of things breaking apart sort of scarier, right? Because it's like, oh well, we might not even get the art scene.

Yeah, totally. And there's often a stereotype about the challenging situations for good art or whatever. And sure, you get punk music when you have a totalitarian leader or whatever, when you have Thatcher. But if you make it less risky, people are gonna innovate more.

The ebook version of Keogh’s book is free to download on the MIT Press website.

This point you're making about scaling is really interesting, by the way, because it's a problem we run into with venture capital funding for games. The VC model only really works for people who are trying to actually scale a company, because the entire idea is that most investments aren't going to return, so you're trying to buy equity in companies that have a real shot at going 100x to offset the losses for the other investments. And that doesn't work if there's a dead-zone between small indie and super-sized studios.

On this point: you're saying the venture capital model is kind of pray-and-spray. You get a bunch of stuff out there—

Well, I mean, it works according to a power law distribution. It's less spray-and-pray and more like you're just making a bunch of hopefully informed bets on people you think are best positioned to make the 100x thing. That way you have tolerance for the bets that inevitably go south.

For sure. I feel like there are two really interesting observations here. First you have what researchers like to call intentional overproduction in media industries. Media companies have been doing that forever. Universal music can sign up 20 pop stars and hope one of them breaks out. And Disney can fund 10 movies and hope one of them at least will be popular enough, that it'll cover the cost of the others, right? That's what I was kind of trying to get at in that Bluesky post that blew up with the Steam graph: What platforms have done very successfully is they're still relying on a model of overproduction.

But they've outsourced that overproduction onto the indies themselves, right? So Steam's model and Apple's model is intentional overproduction. But they don't need to overproduce. They just create the framework for that overproduction. And then when everybody has no choice but to go indie, the platform very successfully gets whichever one of those is the breakout hit. They take their 30% and it covers the costs of hosting all the others.

And the other interesting one, and I think this goes maybe more for venture capital and publishers generally, but if you were to look at, say, pitching to a publisher in '95 versus today, in many ways it's a buyer's market. There are so many aspirational game developers trying to get their game out there.

So now your game has to be so much further along before anyone's gonna pay any attention, right? Like, you can't just have, "here's my game design document, what do you think?"

Yeah. Everybody is looking for traction.

So what you see in both of those is a kind of shared logic I guess. The risk is being outsourced onto the developers themselves both from the platforms and from investors of various kinds.

Both of those parties very much benefit from a model where a whole bunch of people are trying to make it.

I'd love to get your thoughts on how this connects to the big publishers. It seems like—at least in American and European markets, a lot of those are actually struggling right now.

Totally. This is separate from the platformization issue. The main way large scale game development, AAA if you will, produced or created demand through the 90s and 2000s and even 2010s was by constantly releasing a game that they could convince you was a better version of a game that came out last year. Now there's more polygons, now it's bigger. This real kind of technology driven idea of what a quality game is. And that was pretty easy when you were going from the Super Nintendo to the Nintendo 64. It's a lot harder when you're going from the PlayStation 4 to the PlayStation 5.

So the costs have kept going up. You know, the amount of people paying for games has barely moved at all. The spending has gone down in real dollars. People aren't necessarily convinced that this year's Call of Duty is better than last year's Call of Duty, or that Final Fantasy VII Rebirth looks any better on PS5 than it would have on PS4.

That's a great way of putting it—that the tech advances were a way of generating demand for the product.

On the AAA blockbuster professional side, if you want to use that word, you had increasing scale and the idea that tomorrow's game will always be better than yesterday's game.

Remember Todd Howard being like, "finally we have the technology to make Starfield?" [Laughs] As if that wasn't the most breathtakingly unambitious game. This absolutely worked in the 90s and 2000s. But that treadmill that worked then doesn't work any more. PS5 games just straight-up don't look better than PS4.

Yeah, there's very little difference. Did you watch that presentation for the PS5 Pro?

It doesn't look better at all. So you can't build demand that way anymore. But there's still that inertia of "everything has to be bigger than the last one."

That’s it for this week. Go buy Brendan Keogh’s book, or download it for free from MIT.

I’m gonna go play some weird Australian Shrek games.