GDC24 – What are all those flights costing the planet, and what should be done about it?

GDC24 – What are all those flights costing the planet, and what should be done about it?

I did way too much flying last year. While looking at places to live as part of a move interstate, I went between Sydney and Melbourne more times than I care to admit. Because of the woeful state of Australia’s intercity train network (which I wrote about here and what it told me about low-carbon travel possibilities), I usually had to fly or drive. So I skipped GDC this year, which meant avoiding being responsible for somewhere around 1800 kg of greenhouse gasses from international flights, but on the other hand, I did miss GDC and all the people I so enjoy connecting with each time I go.

But I wanted to estimate how much CO2e might be emitted by all those flights of all the other attendees as well – so I cracked out a new spreadsheet and tried to make some guesstimates using the data we have. Like most carbon calculations there are a couple of ways to estimate flights, each with different levels of abstraction from the physical process itself (i.e. the combustion that happens inside a plane’s jet engine). The simplest method, which we’ll use here for now, involves a web-based calculator that takes two inputs: an origin, and a destination, and then draws a line between the two points around the sphere of the earth. There are several things this process doesn’t capture: like differences in the efficiency of different plane engines, local fuel quality, time spent in high-consumption takeoff/landing vs at cruising altitude, routes between the two airports that don’t follow straight lines, as well as more contentious points like the impact of “radiative forcing” (basically: the difference it makes that planes cause these emissions way up in the atmosphere rather than at ground level). Despite these limitations, for a little thought experiment like this, a web-based calculator will suffice, as long as we keep in mind all that it leaves out.

To plug the origin details into the calculator, we need to know how many people flew there, and where they came from. In 2023, there were just over 28,000 GDC attendees (which was almost back to the pre-pandemic peak of 29k attendees) so I’ll stick with that number. For simplicity's sake, I’m assuming that everyone flew from somewhere (despite the attendance of SF locals and people who drive from nearby).

Thanks to the 2024 State of the Industry survey with its 3000 responses, we know which percentage of the overall attendees came from which continents. 10% is a huge sample size, so it's a fairly safe bet that we’re not far off the actual percentages if we collected data from everyone. Here are the percentages and number of attendees implied (if this year has the same attendance numbers):

Continent Percentage of attendees Estimated attendees
North America 62 17,360
Europe 26 7,280
Asia 6 1,680
South America 3 840
Australia/New Zealand 3 840
Africa <1 <280
Not listed 1 280

Because continents are Quite Large, it can make quite a bit of a difference where in a continent one is coming from. London to San Francisco is almost 1000 km shorter than Warsaw to SF even though both are in Europe. Even more extreme is Tokyo, which is almost 5000km closer than Jakarta even though they would both be considered part of Asia. A lot of room for variability and error is introduced by not knowing a country of origin – but we work with what we have.

Because we don’t know where they’re all coming from in each continent, I’ve tried to make an educated guess for each based on where most of the games industry is located. Is 30% of North American attendees coming from Montreal too much, or too little? I’ve been to six GDCs in the past 14 years so I’ve tried to pick what seems like a reasonable mix to me – but I could be way off! For each continent, I’ve picked a few cities that are major hubs for game development (Montreal, Austin, etc), and given them a percentage of attendees. For the “remainder” I’ve used an average US passenger flight distance and emissions for that length of flight (600 kg CO2 per flight) as a placeholder. Its really just a guess – for South America, for instance, if more come from Mexico City (200 kg) than Sao Paolo (781 kg) that will shift the South American total quite a bit – so keep in mind that all these numbers could be out by an unknown amount. The same figure would also be an overestimate if a lot of North American attendees were flying from LA to SF.

Those caveats aside, I’ve added up the totals for attendees from each city plus the “remainder” using our default distance/emissions and arrived at a (highly error-prone!) rough guesstimate of the emissions total for flights to GDC, based on the continent. When it's all tallied up, it's about 21,000 tonnes of CO2 emissions from flights, with my current assumptions. That's about the same as about 1,210 average American households' annual emissions.

Here’s a visualisation of the results of my calculations (keeping in mind that there are a lot of assumptions baked into this) with a focus on the continental totals.

FYI the above map is interactive.

I think that's a reasonably robust breakdown. If anything, European emissions as a percentage might only be higher because the average US attendee's emissions could be lower than this accounts for. It shows us that Europeans going to GDC, even though they only make up about 26% of all attendees probably make up almost half (43%) of the emissions. Why? Because it’s a long way from Europe to San Francisco, is the short answer, and because there's a lot of them.

Asia is only 6% of attendees, but due to similarly long distances travelled makes up 10%. The proximity to the West Coast US for many East Asian originating flights – like Tokyo to San Francisco – also helps. Australia and New Zealand are way over-represented in emissions, despite being merely 3% of attendees (though we do send a lot of people to GDC) it comes out at just shy of 7% of emissions.

Really what we’re doing with this process is modelling the average distance from GDC as a proxy for CO2 emissions from flights that cover longer distances. So there's a penalty to being in the southern hemisphere (tell me about it!!) and nowhere is that more true than for South African GDC attendees. If they are travelling from Johannesburg, they’re probably travelling the furthest distance of any attendee: 16,944 km, for emissions on the order of 1,385 kg, one way. Double it for return. Yikes.

So what did my decision not to go to GDC actually do? Well it avoided me being responsible for a certain amount of emissions technically, and yes that seems like an important moral and ethical decision to try and uphold. But did it prevent a single plane from flying, or prevent any CO2 from going into the atmosphere?

The planes I would have taken still flew albeit, with a few kg's less luggage and person on them – though maybe not even that. So I think, if I’m honest, I would have to admit that no, probably my decision didn't do anything much. On my own, I am not a systemically important actor. This is particularly true because of the way that airlines dynamically price their flights to fill as many seats as possible. What I probably did was make the ticket marginally cheaper for someone else – and maybe not even that. Such is the limited consequence of any individual action. And yet, the opposite conclusion isn't true either: there's simply no way to support going back to flying as much as we used to either.

So what is there to be done?

Is there a point at which enough of “us” (whoever “we” are) opted out of going to GDC that prevented a plane from flying? It seems extraordinarily hard to know, any least with any real certainty. I think there's something more fundamentally wrong with this way of thinking though.

I think what this analysis points us towards, however, is something else – it points to the limits of certain types of approaches to managing the climate crisis. It’s something I’ve been thinking about a lot over the past few days, though it's a debate that has been bubbling along in different parts of the eco-socialist left for several years. It’s especially on my mind again after reading this fantastic interview with Christopher Shaw in the Polycrisis newsletter. Shaw is the author of a (relatively new) book Liberalism and the Challenge of Climate Change which came out last year and which I’ve since started reading.

Shaw levels his criticism at the managerial approach to the climate crisis that is dominant in the liberal democratic world – above all in Europe – but it's embedded in elite institutions like corporate boards as well. It’s a critique aligned with Mann and Wainwright’s (2018) analysis of “Climate Leviathan” the term they give to the dominant global response to climate change, embodied in the annual Conference of the Parties conference and the Paris Agreement’s 1.5ºC target. Shaw puts a bit more of a pointed edge on his critique, describing the institutional “net zero” vision of the future as “this world, but without the emissions” and the way that the liberal order’s principal policy responses are unable to countenance the kind of truly foundational reconfiguration of power relations that are needed:

Shaw: What is the most that a working-class person could hope for from a net-zero future? At present, in the vision being broadly promoted, it’s the same hard work, the same exploitation, but with a heat pump instead of a gas boiler. …Much of the discourse around net zero seeks to replicate all the comforts of middle-class life—for the middle class …But there is just no understanding of working class experience. It’s all, “Come on, care! Be concerned about this heat pump, get behind dropping meat from your diet one day a week, be part of this transformation.” And for what? The same as now. Nothing changed about the status quo, the structures, the norms, what it’s possible to hope for and aspire to.

Though Shaw’s critique sits in an awkward place for me, I think he is broadly correct. He directs his ire at the individualism baked into much of (neo)liberal-free-market envisioned responses – carbon taxes and corporate targets. And I think for many of us, it’s all too easy to fall into the same individualist trap.

So thinking about the decision to fly–or–not to GDC is almost beside the point – can we lift our horizon of possibility a bit higher than the level of individual decisionmaking? Even for the most important event on the game-business calendar? Can we imagine responses to the demands of our climate-changed world that envisage a properly transformational approach? If my staying home or going to GDC doesn't much make a difference, how about shifting focus to the outcomes we want?

Here I think our thought exercise looking at continental emissions becomes quite critical. With the predominance of European flights to GDC in mind, are there ways of planning GDC differently, arranging the same or similar networks and relationships, engaging in business facilitation, seeing talks, engaging speakers, corralling social gatherings and making connections, in other words keeping all the great aspects of GDC that also result in lower greenhouse gas emissions?

One approach that achieves the outcome we want might could involve European alternatives to GDC – at a minimum, replacing some of the social aspects of the conference (which many describe as “the real GDC” anyway) somewhere highly connected by rail. Is there a possible event, schedule, or something that can replace some of the serendipity and chance connections that happen at GDC? Perhaps it’s not all possible, not without getting the “entire game industry” together in the one city for a week (ignoring the fact that it's only one very specific construction of the entire games industry – especially odd when considering the hundreds of thousands of game devs who aren’t there every year). Is it the status of GDC as an “obligatory passage point” for so much game business that’s the issue? Should that cause us to see GDC as a barrier to sustainability?

Or perhaps it could be a partner in another way. Perhaps a shift in the venue of the conference, every other year moving perhaps: to Europe one year, to Asia the next, and back to San Francisco? It would lower European travel emissions at least (and if it’s in a good rail-connected location, perhaps substantially). But we would need to be careful too, as this might come at the cost of increasing American emissions, so it would need to be planned carefully.

One objection to this might be that “Europe already has games conferences, like Develop, etc” and that is true. Perhaps GDC is not an obligatory passage point after all, or becomes less so? Perhaps an ideal outcome might be European games communities forgoing GDC and going to local events instead. What is lost by this? Access to American publishers, perhaps. This was an issue highlighted in the Die Gute Fabrik climate report we did, where travel to promote the game and connect with potential publishers was critical to the studio's survival. We have since seen the halting of production of Die Gute Fabrik's next game, with the team now looking for work, largely as a result of the inability to find a publisher. This was not an imagined threat, this issue is quite real.

We also had two years of remote conferences during the covid years – and before you roll your eyes, just know that I am right there with you and equally tired of second-tier online events, with lag, no eye contact, staring at screens, getting distracted, etc. But we have had decades to perfect in-person events, and we had months to try and iterate online/hybrid ones. If you never attended the Freeplay Zone, an 8-bit styled social space that also facilitated event streaming and group watching of events, then you missed out on easily one of the most important and promising experiments in hosting online events.

Slow travel might be another answer – Jiri and Maria of the SGA tried valiantly to go over-seas from Europe to GDC and were simply unable to find passage from the Canary Islands to the US. The wrong time of year and no one is sailing. An enterprising ferry business could be started there probably – it looked from the billboards covered with flyers like there was enough demand.

Some of these things might even be in the best interest of many major games' companies to think about already – for those companies with ambitious net zero targets rapidly approaching, I have some bad news for you about the viability of so-called "sustainable aviation fuels". If they are going to stick within their budgets and hit the targets, travel is going to either be a lot less common, a lot slower, or both.

All this is to say that I think it's important to start from the world we want and work backwards. Starting with the very real and concrete biophysical limits of the planet, paired with and a vision of a better, fairer world that also has zero emissions. Starting with “what can I do?” in the here and now isn’t wrong per se, but it starts from where we are, deep in the valley. Some of these things look decidedly impossible from here. But they also have the virtue, to borrow a phrase from Mann and Wainwright, of being necessary. Sometimes impossible will just have to get out of the way.