Could climate change lead to the end of civilisation?
Across the world, over half of young people worry that, as a result of climate change, humanity is doomed.1 They feel angry, powerless, and — above all — afraid about what the future may hold.2
Climate change matters so much, to so many, not just because of the suffering and injustice it’s already causing, but also because it’s one of the few issues that has obvious potential to affect our world over many future generations. We think safeguarding future generations is a key moral priority, and should be a crucial consideration in prioritising problems on which to work.
If climate change could lead to the end of civilisation, then that would mean future generations might never get to exist – or they could live in a permanently worse world. If so, then preventing it, and adapting to its effects, might be more important than working on almost any other issue.
So – what does the science say?
The Intergovernmental Panel on Climate Change (IPCC) Sixth Assessment Report is, to our knowledge, the most authoritative and comprehensive source on climate change. The report is clear: climate change will be hugely destructive. We’ll see floods, famines, fires, and droughts — and the world’s poorest people will be affected the most.3
But even when we try to account for unknown unknowns,4 nothing in the IPCC’s report suggests that civilisation will be destroyed.
This isn’t to say society shouldn’t do far more to tackle climate change.
That’s because climate change’s impacts will still be significant – it could destabilise society, destroy ecosystems, put millions into poverty, and worsen other existential threats such as engineered pandemics, risks from AI, or nuclear war. If you want to make climate change the focus of your career, we include some thoughts below on the most effective ways to help tackle it.
So yes, climate change is scary. And people are right to be angry that too little is being done.
But we’re not powerless.
And we’re far from doomed.
Climate change is going to significantly and negatively impact the world. Its impacts on the poorest people in our society and our planet’s biodiversity are cause for particular concern. Looking at the worst possible scenarios, it could be an important factor that increases existential threats from other sources, like great power conflicts, nuclear war, or pandemics. But because the worst potential consequences seem to run through those other sources, and these other risks seem larger and more neglected, we think most readers can have a greater impact in expectation working directly on one of these other risks.
We think your personal carbon footprint is much less important than what you do for work, and that some ways of making a difference on climate change are likely to be much more effective than others. In particular, you could use your career to help develop technology or advocate for policy that would reduce our current emissions, or research technology that could remove carbon from the atmosphere in the future.
Our overall view
We’d love to see more people working on this issue, but — given our general worldview — all else equal we’d be even more excited for someone to work on one of our top priority problem areas.
We think work to materially reduce the probability of the worst outcomes of climate change would have a large positive impact. However, climate change seems hundreds of times less likely to directly cause human extinction than other risks we’re concerned about, like catastrophic pandemics. As a result, if climate change does have catastrophic and potentially long-lasting consequences for human civilisation, this will likely be through aggravating other problems, such as conflict between great powers. This indirect risk brings the scale of climate change as a problem closer to other extinction risks, although it still seems more than 10 times less likely to cause extinction than nuclear war or pandemics. Our guess is that more people should seriously consider aiming at those issues directly.
Overall, climate change is far less neglected than other issues we prioritise. Current spending is likely over $640 billion per year. Climate change has also received high levels of funding for decades, meaning lots of high-impact work has already occurred. It also seems likely that as climate change worsens, even more attention will be paid to it, allowing us to do more to combat its worst effects. However, there are likely specific areas that don’t get as much attention as they should.
Climate change seems more tractable than many other global catastrophic risks. This is because there is a clear measure of our success (how much greenhouse gas we are emitting), plus lots of experience seeing what works — so there is clear evidence on how to move ahead. That said, climate change is a tricky global coordination problem, which makes it harder to solve.
This is one of many profiles we've written to help people find the most pressing problems they can solve with their careers. Learn more about how we compare different problems, see how we try to score them numerically, and see how this problem compares to the others we've considered so far.
If you have any feedback on this article – whether there’s something we’ve got wrong, some wording we could improve, or you want to tell us you loved reading it – we’d really appreciate it if you could tell us what you think using this form.
Could extreme climate change directly lead to the extinction of humanity?
We’re going to review the three most common ways people say climate change might directly cause human extinction: high temperatures, rising water, and disruption to agriculture.
Worst case climate scenarios look very bad in terms of lives disrupted and lost. We’re focusing on extinction because, for reasons we discuss here, we think reducing existential threats should be among humanity’s biggest priorities – in part due to their significance for all future generations.
In short, most scientists think it’s pretty close to impossible for climate change to directly cause the extinction of humanity.
In this generally grim area, this is a piece of good news people don’t always appreciate.
That said, we shouldn’t be unconcerned about climate change — not only does it pose grave dangers short of extinction, we also think climate change indirectly raises the risk of extinction via making other threats worse, which we’ll cover in the next section.
How hot could it get?
The hotter the Earth gets, the worse we can expect effects from climate change to be.
So, to figure out whether climate change could directly cause extinction, we need to know how much we expect temperatures to rise. To figure that out, we first need to have some idea of how much greenhouse gas we’re going to emit, as well as how much of a temperature rise that will produce. We’ll look at each in turn.
How much greenhouse gas could we emit?
The IPCC Sixth Assessment Report considered many illustrative scenarios, including:
- The world meets the goals of the Paris Agreement from COP-21, to limit warming to 1.5°C. (SSP1-1.9)
- The world takes enough action to limit warming to 2°C. (SSP1-2.6)
- There are modest mitigation efforts, with slightly lower emissions than what current policies might suggest. (SSP2-4.5)
- There is a reversal of some current policies, which increases warming. For example, this could happen if countries are competing against each other for growth. (SSP3-7.0)
- There is significant policy reversal. The world decides to use fossil fuels to cause rapid development even if they are more expensive than renewable energy. (SSP5-8.5)
The IPCC projections of future emissions until 2100, from the Sixth Assessment Report.
These are the most likely scenarios. But what about the most extreme scenario? What if we tried to burn literally all the fossil fuel in the ground?
The IPCC estimates that there are 18,635 gigatonnes of carbon in the Earth’s fossil fuel reserves.5
Luckily, fossil fuel extraction methods don’t allow you to extract all the fossil fuels in a deposit — especially with coal. So the question is not how much fossil fuel there is, but how much might ultimately be recoverable with future technology.
The highest estimate we’ve seen on the quantity of recoverable fossil fuels is 2,860 gigatonnes of carbon.6
Releasing 3,000 gigatonnes of carbon would take us to carbon dioxide concentrations in the atmosphere of around 2,000 parts per million (for comparison, pre-industrial carbon dioxide concentrations were 278 parts per million, and current concentrations are 415 parts per million).7
How much warming could happen as a result?
The warming caused by our greenhouse gas emissions will occur in the decades and centuries after the emissions, and is effectively caused by the total amount of carbon emitted.
But actually predicting the total amount of warming caused by some quantity of greenhouse gas is difficult, because there are feedback loops.
Here’s an example of one of those feedback loops: when you heat a metal enough, it glows red. Things at much lower temperatures glow in infrared — that’s why you can see people at night using infrared cameras. The hotter something is, the more energy it releases through this glow (known as black-body radiation). So as the Earth’s temperature increases, it radiates more infrared radiation back into space. This reduces the effect of emissions on global temperatures.
But there are also feedback loops that could make things worse, which we’ll go through here. In the worst cases, these are associated with tipping points, where once a certain amount of greenhouse gas has been released, it triggers some feedback loop that results in an extremely significant and permanent increase in temperature.
The runaway greenhouse effect
We could theoretically see very extreme temperature rises through a runaway greenhouse effect.
We know this is possible because it appears to have happened before: on Venus. Soon after it formed, Venus may have been habitable, with a large water ocean. But Venus formed closer to the Sun than the Earth did, and this slight increase in temperature led to the gradual evaporation of its ocean. Water vapour is a greenhouse gas, so this led to further heating, and further evaporation, eventually moving Venus from a habitable world to one where surface temperatures reach 462°C (864°F), hot enough to melt lead.
Luckily, most models suggest that it’s just not possible, even in principle, for anthropogenic carbon dioxide emissions to reach levels high enough to trigger a runaway greenhouse effect on a Venus-like scale.8
And even if we did eventually lose all the oceans to space, this would take hundreds of millions of years. So we’d be very likely to be able to stop the process or find other ways to survive (if something else doesn’t kill us in the meantime).
One study found that if carbon dioxide concentrations in the atmosphere reach around 1,300 parts per million (which is unfortunately plausible under worst-case scenarios), clouds that shade large parts of the oceans and reflect light back into space could break up.
Many scientists think that the model involved is far too simple to be plausible.
However, there isn’t consensus on this. If the modelling in this study is right, cloud feedbacks would cause an extra 8°C of warming on top of the expected 6–7°C associated with that quantity of emissions. That would bring us into the extreme regions of the effects listed below.
Methane clathrate is a substance formed of methane within a crystal of solid water molecules — basically ice with methane trapped in it. There are large quantities of methane clathrates under the ocean floor.
When oceans warm, they could melt these clathrates, releasing further methane into the atmosphere.
The IPCC’s Sixth Assessment Report estimates that there is the equivalent of 1,500–2,000 gigatonnes of carbon dioxide trapped in methane clathrates (approximately twice as much as we have emitted so far). However, they expect any release of clathrates to occur over centuries or millennia, which would give us substantially more time to adapt to any changes.
So the IPCC thinks it’s unlikely that clathrate emissions will cause substantial warming in the next few centuries.9
Research into methane clathrate release appears underdeveloped, so there’s a lot we don’t know — and this contributes to our overall uncertainty on how hot it will get.
There are permanently frozen layers in the Arctic and other cold regions of Earth. The IPCC estimates that there are 1,460–1,600 gigatonnes of carbon dioxide (or equivalent quantities of other greenhouse gases) stored in permafrost.10
Some of this permafrost is already melting, releasing greenhouse gases.
Abrupt thawing could cause up to half of these trapped greenhouse gases to be released immediately, with the rest released more gradually over decades.
The IPCC report says that for each 1ºC of warming, permafrost emissions will increase by 18 gigatonnes of carbon dioxide, with a 5% to 95% range of 3 to 42 gigatonnes. At the upper end of models, we could see up to 600 gigatonnes of carbon dioxide released from permafrost, warming the planet by about an extra degree (on top of the 6ºC we’d already see from human emissions in these scenarios).
Summing up: what is the range of possible temperature increases from climate change?
Given some amount of greenhouse gas we emit, we need to know what the possible temperatures are that Earth could reach.11
There’s clearly a minimum possible temperature change; it’s virtually certain that temperatures will go up.12
The Sixth Assessment Report gives estimates of how much temperatures will go up under the emission pathways we looked at earlier (which range from meeting the terms of the Paris Agreement to an extreme fossil-fuel burning scenario).
The IPCC projections of future temperature rises by 2100, from the Sixth Assessment Report. The top of each bar is the IPCC’s median estimate, with the whiskers showing the 90% confidence range.
While these estimates do incorporate uncertainty around which feedback loops are plausible (more on that below), they show 90% confidence intervals. That is, (roughly) there’s a 10% chance that temperature changes in each scenario could be higher than the top of the thin lines, or below the bottom of the thin lines (representing the confidence intervals).
In a scenario worse than the IPCC’s worst case, where all the recoverable fossil fuels are burned, there’s a 1 in 6 chance we’d see greater than 9ºC of warming by 2100.13 And it looks like there’s an extremely small, but real, chance that this would be enough to cause the worst-case cloud feedback loops.
In total, that would lead to something like 13°C of warming relative to pre-industrial levels. We’d reach that 13°C in the years or decades after we trigger the cloud feedback tipping point — and then there could be additional warming in the centuries and millennia after that. This 13°C of warming would be a humanitarian disaster of unprecedented scale.
As far as we can tell, reaching 13°C is very unlikely, and about as hot as our models suggest it could possibly get on short timescales we might not be able to adapt to.14
Now we’ll turn to whether this warming could directly cause human extinction — via pure heat stress, sea level rise, or agricultural collapse.
Could climate change simply make it too hot for humans to survive?
On the hottest and most humid days, you’d walk outside and it felt immediately like someone pressed a hot wet towel, like you sometimes get on airplanes, over your entire head. I wear glasses, and they’d immediately fog up. You sweat instantly. People just avoid being outside in any way they can. In the summers, my friends and I would become nocturnal as a way to beat the heat.
John Hagner, on living in Dharan, Saudi Arabia (one of the most heat-stressed inhabited places in the world)
If temperatures rise high enough, it becomes too hot for humans to survive for more than a few hours — even in the shade. In places with high humidity, like the tropics, it’s harder to cool through sweating, so this effect is even worse.
This could make significant parts of the planet uninhabitable (at least outdoors, or without air conditioning) for significant portions of the year. This map shows the number of days per year we’d get surface temperatures greater than 35°C (95°F) in various regions on Earth, if we had around 7°C of warming. This is a good illustration of the sorts of areas that might become too hot for humans to survive with more than 7°C of warming.15
If there were 12°C of warming, a majority of land where humans currently live would be too hot for humans to survive at least a few days a year.16 An increase of 13°C would make working outdoors impossible for most of the year in the tropics, and around half the year in currently temperate regions.
But even with the cloud feedback loop, it would take decades for global temperatures to reach this level, and while this worst-case scenario would cause extraordinary suffering and death, it seems very likely that we could adapt to avoid extinction (for example, by building better buildings and widespread air conditioning, as well as building more in the cooler areas of the Earth).
Moreover, it would be hard for this to lead directly to extinction even if we didn’t adapt, given that a large chunk of land on Earth would remain habitable, even with 13°C of warming. We would have to live in a much smaller area, but civilisation would survive.
Could the world sink under water?
The IPCC’s Sixth Assessment Report projects a sea level rise of around one metre by 2100 if we see 5°C of warming above pre-industrial levels — in worst-case scenarios, this could be up to two metres.
The IPCC projections of sea level rises by 2100, from the Sixth Assessment Report.
This map shows the areas that will be below high tide by 2100 if there were 5°C of warming, assuming a 95th percentile worst-case scenario. (That is, there are various proposed effects that could help reduce the amount the seas will rise, and this map assumes that these largely won’t occur.)
Modelling sea level rise is difficult, so there’s large uncertainty on how bad it could get. And over centuries, sea level rise could be much higher. The IPCC says that, in the worst emissions scenario we’ve considered with around 6°C of warming, “sea level rise greater than 15m [by 2300] cannot be ruled out.”
We haven’t seen any modelling on sea level rise with 13°C of warming. As an upper bound, we can consider what would happen if the polar ice caps melted completely. The highest estimate we’ve seen is that this would produce sea level rise of around 80 metres. Fifty of the world’s major cities would flood, but the vast majority of land would remain above water.
Land remaining after an 80-metre sea level rise, as calculated by academics at UPenn. © 2017 Richard J. Weller, Claire Hoch, and Chieh Huang Atlas for the End of the World.
As with heat stress, we’ll probably be able to adapt to these changes, particularly through building new infrastructure like homes or flood defences. And the fact that it will take centuries for sea levels to rise completely means this adaptation will likely be much easier.
It seems that a one-metre sea level rise would, without adaptation, displace around half a billion people from their homes. But with adaptation (like building flood defences), the number of people displaced would be much smaller: the IPCC estimates that hundreds of thousands of people would in reality be displaced due to a two-metre sea level rise, far fewer than half a billion.
We know this adaptation is possible because we’ve seen adaptation to rapid sea level rise before. Tokyo is sinking into the ocean, and experienced effectively four metres of sea level rise in the 20th century.17 This rise happened at a rate of about 40 millimetres per year, which is similar to what we’d expect with the IPCC’s worst-case projections.
So sea level rise will cause substantial suffering and disruption to our society, particularly in developing countries. And our adaptation will be expensive — the IPCC projects that, with 4–5°C of warming, we’ll be spending over 1% of our GDP adapting to floods.
But, as with heat stress, sea level rise does not pose an extinction risk.
Could climate change destroy global agriculture?
The IPCC’s Special Report on Climate Change and Land reports that hundreds of millions of additional people will likely be at risk of hunger by 2050 as a result of climate change.
On top of extreme events like hurricanes or droughts disrupting agriculture, we can expect temperature changes, changes in rainfall, and other weather-related changes to significantly harm our ability to grow food.
There may also be some positive effects of climate change on agriculture — for example, we’ll be able to grow crops in areas that are currently too cold. It’s possible that these effects would be enough to completely mitigate the negative effects on agriculture.
With more extreme warming, higher temperatures will directly affect agriculture.
Mean maximum temperature for leaf initiation, shoot growth, root growth, and lethality for rice, wheat, and maize. Source: Sanchez et al., 2013
The chemical reactions plants need to survive (including photosynthesis and respiration) can’t operate if temperatures are too high.
As a result, more than 10°C of warming would be likely to destroy agriculture in India and regions with similar climates.
We could also see substantial changes in precipitation levels that, under extreme scenarios, would significantly harm agriculture. This map shows changes in precipitation by 2050 in various regions, under a high-emissions scenario.18
(In general, predictions about precipitation and other weather changes, like the frequency of extreme weather events, are difficult to make and vary significantly between models — so take all these figures with a grain of salt!)
But even with all these likely disruptions, we should still be able to adapt — due to increasing agricultural productivity. Over the past few centuries, food prices have fallen as technology makes it cheaper and cheaper to produce large quantities of food.
So it is against this backdrop of rapidly improving productivity that climate change will act — and even if temperatures rise a lot, it’ll take some time (decades or maybe centuries) for that to happen. As a result, the IPCC expects (with high confidence) that we’ll be able to adapt to climate change in such a way that risks to food security will be mitigated.
One expert we spoke to did say that their best guess is that a 13°C warmer world would lead — through droughts and the disruption of agriculture — to the deaths of hundreds of millions of people. But even this horrific scenario is a long way from human extinction or the kind of catastrophic event that could directly lead to humanity being unable to ever recover.
How would extreme climate change affect biodiversity?
It’s possible that climate change could lead to ecosystem collapse. Many ethical views put intrinsic value on biodiversity — and even if you don’t, ecosystem collapse could affect people and nonhuman animals in other ways.
Estimates on the proportion of species that could go extinct from climate change vary, but in the worst cases, models predict up to 40% of species could be “committed to extinction” by the middle of the century.19
So extreme climate change could have significant negative effects on biodiversity. What about the instrumental importance of biodiversity? Could reduced biodiversity exacerbate the effects of extreme warming on agriculture? In order for this to happen, we’d have to see something crucial to our food chain go extinct. One plausible possibility here is that pollinators — whose populations are already in decline — could go extinct. But models suggest our agricultural production would drop by only around 10% if we didn’t have pollinators.20 Kareiva and Carranza at the Cambridge Centre for the Study of Existential Risk looked into this in more detail and concluded that ecosystem collapse is extremely unlikely to pose risks to human existence.21
There are, of course, many other benefits to biodiversity, like the development of new medicines. But overall, biodiversity loss seems like it won’t cause the collapse of civilisation.
Summing up: why climate change almost definitely won’t directly cause human extinction
If we follow current policies, we’ll probably end up seeing around 2–3°C of warming by 2100. It’s also possible that we’ll see a reversal of current attempts to reduce emissions. This could happen if sectors of the economy that we can’t decarbonise grow rapidly, such as if we develop new technology that uses large quantities of energy, or if something like a large war incentivises high-carbon activities.
In a worse scenario, we’ll burn fossil fuels even though they’re more expensive than renewable energy. And in the worst-case, but extremely unlikely, scenario, we could burn all the recoverable fossil fuels and reach 7°C of warming.
There’s also a very small chance that in these unlikely scenarios where we rapidly burn far more fossil fuels than we are currently on track to, that cloud feedback tipping points could be reached. This could lead to something like 13°C of warming.
Though this would be a humanitarian disaster of unprecedented proportions, humanity would still have land cool enough to live on, it won’t all be submerged in the ocean, and we will still be able to grow food in many places, though not all. In other words, humanity would survive.
But does this conclusion adequately take uncertainty into account?
After all, any time we try to use what we’ve discovered so far to make predictions about the future, we have to be aware that there could be things we don’t know which could make things worse than we expect.
We saw above that one source of uncertainty is the possible emission pathways that we will follow in the future. We tried to take that into account by considering a wide range of scenarios – including us burning all ultimately recoverable fossil fuels.
We’ve also seen structural uncertainty: that is, uncertainty in our predictions because there are things we don’t know about how the system works — for example, whether methane clathrates will cause substantial warming in the next few centuries.
The IPCC’s Sixth Assessment Report, building on Sherwood et al.’s assessment of the Earth’s climate sensitivity attempts to account for structural uncertainty and unknown unknowns. Roughly, they find it’s unlikely that all the various lines of evidence are biased in just one direction — for every consideration that could increase warming, there are also considerations that could decrease it.22
This means we should expect unknowns mostly to cancel out, and be surprised if they point in one direction or the other.
There are a few caveats:
- The higher our emissions are, the further they get from the sorts of baseline assumptions the IPCC used to come to this conclusion. So if we’re really very wrong about the amount of carbon emissions we’re likely to emit, things could still get very bad (but it seems unlikely we’re very wrong about that).
- There is much more uncertainty about how other things will change. For example, it’s hard to predict how high sea levels will rise or how precipitation patterns will change (although even then we don’t think these things will change in ways that increase the direct risk of extinction).
But overall, despite our lack of knowledge about some relevant feedbacks, this makes for a very small chance that model uncertainty means things could be radically worse.
As a result, it’s extremely unlikely (we’d guess less than a 1 in 1,000,000 chance) that we’ll see the temperature changes necessary for climate change to have the kinds of effects that would directly lead to extinction.
How climate change could cause extinction indirectly anyway
We’ve argued that climate change is very, very unlikely to directly cause human extinction.
But climate change seems to contribute to the risk of human extinction anyway, by making other existential threats worse.
Here we’ll go over the common factors people put forward to argue that climate change might increase extinction risk, and how big a contributor we think each factor really is.
Climate change will likely increase migration, which could lead to instability
As we’ve seen, higher temperatures and rising sea levels will significantly affect where people are able to live. And other factors (like changes to agriculture) will affect where people are able to make a living, also leading to more migration.
According to the IPCC’s Fifth Assessment Report, a half-metre sea level rise (without governments implementing adaptive measures) implies the displacement of 72 million people; a two-metre sea level rise (something like the IPCC’s worst-case scenario) would displace 2.5% of the world’s population. These figures assume we wouldn’t act to prevent this displacement — measures like building protective dikes could reduce this to less than half a million people.
If there’s more extreme warming, we’ll see more extreme migration. At 6°C of warming, warmer areas without air conditioning could become unlivable, causing migrations of potentially hundreds of millions of people.
It’s often claimed that displaced populations can increase resource scarcity and the risk of conflict in countries that they move to. Forced displacement also arguably increases the spread of infectious diseases and general political tensions. But it’s very difficult to estimate the size of these effects — and from there, to estimate the implications of these effects for the rest of society.
How might this increase extinction risk? The biggest route is through increasing conflict and therefore the risk of great power war, which seems like a significant risk factor for extinction.
We’ll turn directly to that factor now.
Will climate change increase global conflicts?
Climate change’s clear ability to create economic shocks, migration crises, and resource scarcity makes it completely plausible that there will be (as there have already been23) conflicts at least partially caused by climate change.
Lots of this conflict is likely to be civil conflict in areas that are already unstable and are particularly vulnerable to climate change (the ICCP’s Fifth Assessment Report focuses on civil war in Africa).
There’s also the possibility of much larger wars. If climate change significantly affects the fortunes of Russia, China, India, Pakistan, the EU, or the US, this could cause a great power war. Migration crises, heat stress, sea level rise, changes to agriculture, or broader economic effects on these countries could all contribute to the chances of conflict.
This is all pretty speculative, but we still think it’s worth taking seriously.
Conflict makes it harder to solve coordination problems. For example, it incentivises dangerous arms races, which are even more dangerous when they’re between great powers. Because of this, it stands to reason that conflict — especially between powerful nations — increases existential risk.
Climate change could make society less stable in other ways
There are many other proposed pathways for climate change to make our society generally less stable. For example:
- Reduction in tax revenues because of changes to the economy (e.g. if a country’s agricultural land becomes less productive) can make people in power less able to act. This changes the relative strengths of political factions, making changes in governments more likely.
- Climate change could hurt people’s economic prospects, which can create desperation and violence. This can be a key cause of civil unrest and civil war.
- When climate change causes hardship, populations may (correctly or incorrectly) blame their governments, increasing political instability.
It’s also possible that we could be driven to develop destabilising technology to change our climate with the intention of averting catastrophe — e.g. solar geoengineering. But this poses its own risks, as it will be near impossible to carry out experiments on the global scale we’d need to act in order to verify the safety of our technology. And technology to change the weather could in turn lead to conflict between (or within) states over things like induced droughts or rainfall.
Summing up: how climate change makes global catastrophic risks worse
Risks to humanity (like nuclear war or pandemics) don’t just affect particular groups or countries, so we shouldn’t be surprised if many of the most promising solutions require global cooperation.
Fortunately, if we have the ability to cooperate to reduce these risks, we expect that we will. After all, if we don’t, the consequence is global catastrophe! But actually having this ability is vital.
Unfortunately, it seems like climate change will reduce our ability to cooperate.
For example, it’s been suggested that increased resource scarcity (in particular water scarcity) caused by climate change could increase the risk of conflict in Kashmir, one of the most important flashpoints for great power and potentially nuclear war (in this case, between India and Pakistan, even though both sides have an interest in avoiding war). We’re not sure this is right, but it doesn’t seem impossible.
General instability also increases the risk of individual actors like terrorist groups unilaterally acting to cause a catastrophe. And this sort of deliberate harm is one of the key ways we could succumb to a global catastrophic biological risk.
We think that the 21st century could plausibly be humanity’s most important due to rapid technological progress, especially in artificial intelligence. If that’s true, we’re going to want to be very careful to ensure it goes well. Lots of unpredictable things will happen, and climate change will be a key cause of many of them. And the worse climate change becomes, the more unpredictable these things will be. That in itself might be a strong reason to dedicate your career to working on climate change.
That said, we still think this risk is relatively low. If climate change poses something like a 1 in 1,000,000 risk of extinction by itself, our guess is that its contribution to other existential risks is at most a few orders of magnitude higher — so something like 1 in 10,000.24
So yes, climate change makes other existential risks worse. But humanity is still much, much more likely to survive climate change than not.
What about global catastrophes that aren’t extinction?
Even if climate change is very unlikely to cause humanity to go extinct (directly or indirectly), could it still cause a global catastrophe on such a scale as to cause the deaths of a significant proportion of the population (say, more than 10%)?
We haven’t thought about this possibility as much, but the same reasons we think climate change won’t lead to extinction suggest it won’t lead to a catastrophic event of this size. In short: even in the worst-case warming scenarios, a lot of humans will still be able to live on the land and grow food.
Even in the top 1% of worst scenarios, our guess is that it is extremely unlikely for premature deaths due to climate change to exceed a billion people, and this loss would likely be gradual (e.g. over a century) and due to things like declining economic productivity, rather than an all-at-once catastrophic collapse. This is still an immense amount of death and suffering, and we hope global leaders will ensure this does not come to pass.
However, gradual problems in general seem easier to adapt to, meaning the risk that humanity doesn’t ever recover from the effects of catastrophic (but not immediate-extinction-level) climate change seem very low — lower than, for example, an all-out nuclear war.
Again, the indirect threat from climate change seems greater here. For example, perhaps international tensions worsened by climate-related stressors will lead to such a war.
All in all, we think the risk of a sub-extinction-level global catastrophe that kills a billion or more people from climate change is still very low.
How else could climate change affect the long-term future?
Even if you don’t agree with our focus on existential risks (perhaps because you think we’re nearly guaranteed to survive the next few centuries), you might still wonder how else climate change could affect generations long into the future.
Carbon dioxide can stay in the atmosphere for thousands of years, which means that warming can continue for hundreds of thousands of years after we stop emitting — and this warming could continue to have negative effects on our society.
For example, if we have a one-metre sea level rise in the next 100 years, we can expect to see 10 metres of sea level rise over the next 10,000 years.
This means that, if we avoid existential catastrophe and humanity continues to live on Earth, future generations could be dealing with the negative effects of climate change for a long time.
Moreover, if climate change gets very bad, that probably means we burned through our fossil fuel reserves. This isn’t an effect of climate change per se, but rather an effect of us not doing enough to prevent it by reducing fossil fuel use. Besides causing climate change and everything that that entails, using up our fossil fuel reserves would mean that if humanity does suffer a (different) global catastrophe that leads to a civilisational collapse, it might be harder to rebuild.
This is because fossil fuels are one of the densest and most accessible forms of energy. Imagine, for example, we were in a nuclear winter after a global nuclear war, and needed to get everything back up and running after losing the technology and know-how we’ve built over the last 100 years. It would be extremely useful in that case to be able to temporarily burn fossil fuels in order to redevelop society. But if we’ve burned them all already, we won’t be able to do that. (Listen to our podcast with Luisa Rodriguez on how we might recover from a global civilisational collapse for more.)
Should you work on climate change or another global issue?
There are lots of global issues that deserve more attention than they currently get. This includes climate change, but also others that seem to pose a more material risk of extinction — like catastrophic pandemics or nuclear war.
If you want to make the biggest difference you can with your career, and like us, you think that reducing existential threats is a top priority, which should you focus on?
Reasons to work on climate change
Even if it’s not an extinction risk, our discussion above shows that climate change could still be hugely important for the present and future of life on Earth. The worse it gets, the more likely it is to reduce biodiversity, displace people around the world, destroy people’s livelihoods, and destabilise society. That alone is a reason to work on it.
But something being important isn’t necessarily a decisive reason to work on it. Our framework for comparing global problems suggests you should also consider:
- How solvable is climate change?
- How neglected is working on climate change?
- And how does it compare to other issues you could work on instead?
Climate change seems unusually solvable for a global issue: there is a clear measure of our success (how much greenhouse gas we are emitting), plus lots of experience seeing what works — so there is clear evidence on how to move ahead.
There’s also a lot of opportunity to work on climate change. Europeans and many Americans agree that climate change is one of the most important issues of the 21st century, and there are therefore many opportunities to work on the issue in government, business, and academia.
This means that if you are able to get into a position of power, you can leverage a lot of resources.
Because lots of people think climate change is important, the easiest ways of making a big difference have likely already been taken (more on that just below).
However, some important work on climate change does appear to be relatively neglected. Not much research is focused on how climate change interacts with other potential catastrophic risks. And working on clean energy tech also seems neglected relative to its importance for solving the problem, though it still gets a lot of resources.
Your work also might have positive side effects. For example, reducing our reliance on burning fuels might also reduce air pollution, which causes millions of deaths per year. Working on extreme climate change could indirectly help promote positive values, such as caring about future generations, and it’s possible that finding effective ways of mitigating climate change could serve as a blueprint for future efforts to tackle global threats.
Reasons not to work on climate change
It’s not as neglected as other issues
Climate change as a whole gets a lot of attention and funding. In particular, it gets much more attention than many other pressing global issues.
The US federal budget included about $23 billion of climate change spending in the 2021 fiscal year. The UK spent about £4 billion in the 2021–22 financial year. And several hundred million dollars are spent each year by foundations.25 Philanthropic spending on climate change is $5 to $10 billion a year. On top of this, many businesses and universities around the world work on general climate change research or technologies designed to reduce emissions. The Climate Policy Initiative counted over $600 billion in climate-related spending in 2020.
In comparison, biosecurity in general receives around $3 billion per year, preventing catastrophic pandemics in particular receives around $1 billion, and reducing risks from artificial intelligence receives between $10 million and $50 million.26
If a problem is less neglected, it will be harder for an additional person to make as much of a difference working on it.
Other existential threats seem considerably greater
Another consideration is that, for all that climate change is a serious problem, there appear to be other risks that pose larger threats to humanity’s long-term thriving.
Experts studying risks of human extinction usually think nuclear war, great power conflict in general, and certain dangerous advances in machine learning or biotechnology all have a higher likelihood of causing human extinction than climate change.
This seems roughly right to us.
You might think that, for the reasons discussed above, climate change is a substantial enough factor contributing to other risks to be worth prioritising. That wouldn’t be an unreasonable view.
But these other, often more direct threats to humanity often also act as contributing factors. For example, pandemics can increase geopolitical tensions, and thus risks of conflict. And some of them — especially engineered pandemics and risks from misaligned AI — seem to be direct extinction threats themselves on top of that.
So, if you agree that we face substantial more direct existential threats, then to think climate change is more important you need to think that climate change is a much larger risk factor than other things — so much larger that this outweighs the difference in direct risk. We think it probably doesn’t.
That said, there are other factors that determine what you should work on — in particular, it’s important to consider your personal fit for jobs in an area (and we’ve definitely spoken to people who are best suited to working on climate change rather than any other issue). But in our experience, it seems like many people underestimate their ability to, with a bit of training, work on issues they’re less familiar with. People also seem to underestimate the range of positions — and therefore the different kinds of work they could excel at — in different cause areas.
As a result, while we agree that it’s crucial we work on reducing existential threats to humanity, and agree that climate change increases those threats, we usually recommend people interested in safeguarding humanity’s future focus on bigger and more direct existential threats if they can.
But given that many people will work on climate change in their careers (and in absolute terms, we hope that more people do, even if we also hope many of our readers will prioritise more direct risks), we’d like to say something about how to do that as effectively as possible.
What are the best ways of working to solve climate change?
Many popular approaches to working on climate change and other environmental problems probably aren’t actually that helpful.
For example, the increased land use from organic farming means it could actually increase emissions compared to regular farming. And eating locally produced food — a popular idea for reducing your carbon footprint — is far less important than what food you choose to eat, because transportation is such a minor part of the emissions from food.
The same applies in government policy. Governments across the world have attempted to disincentivise the use of disposable plastic bags (in favour of non-disposable alternatives), but this may have actually increased emissions:
Other things work but could be really expensive (costing $100 or more per tonne of carbon dioxide removed from or prevented from going into the atmosphere). For example, planting trees sounds like it could be effective, but trees’ slow growth, risks of things like wildfires, and the high cost of land may make this a particularly expensive way of reducing greenhouse gases.
We’ll talk about ideas for cutting emissions that seem more cost effective below.
But in general we want to note that your personal carbon footprint is largely a distraction. If you cut your emissions by 50% you’ll save 2–10 tonnes of carbon dioxide per year, whereas a carefully considered donation (for example to the Founders Pledge Climate Change Fund) of only $10 could do more to reduce emissions than that.
And spending time working on the issue could be even better.
Some considerations to help you figure out what to work on
A few key ideas shape what we think is most effective to work on to address climate change.
First, emissions in Europe and North America are declining,27 but emissions are rising elsewhere.
Developing countries use much less energy per person and will need to keep their energy consumption growing in order to raise living standards — something people in poorer countries desperately need.28
So we need to reduce emissions across the world, but without harming living standards. This puts some constraints on which interventions are most important to carry out.
Second, solutions that require coordination are difficult to achieve. This is true on both an individual level and a country level.
Reducing emissions benefits everyone more than it benefits any one individual. For example, if the US became net zero, all countries would benefit from reduced harms from climate change, but the US would only get a fraction of the overall benefit from their actions (while bearing the entire cost). So you should expect individuals and countries to be doing less than would be best for the world.
For this reason, focusing on developing and deploying new technology seems more likely to succeed (and has fewer downsides, and faces fewer coordination issues) than seeking to encourage individuals to voluntarily reduce their energy consumption. This is because it doesn’t cost the innovator much; they can benefit from selling their inventions.
This means low- or no-emissions technology is likely one of the biggest levers there is.
Third, spending on climate change is gigantic but may neglect key things.
We argued earlier that climate change seems less neglected overall than other areas, with spending at $640 billion a year.
This means that, if you can identify important but neglected sub-areas within climate change, advocating for more resources for those areas or shifts in the prioritisation of existing resources could be hugely impactful. Small proportional shifts can move large amounts of money.
Reducing net greenhouse gas emissions — especially through green tech innovation
We think one of the most promising ways to reduce greenhouse gas emissions is working on research and development in green energy.
Green energy has an incredible track record of returns, could help solve problems in more than just one country, and doesn’t require convincing other people to act. For example, asking people not to drive asks them to make a personal sacrifice, whereas developing emissions-free cars solves the problem without relying on that choice.
Renewable energy is now often cheaper than fossil fuels — this could be a key reason why emissions are falling in Europe and North America.
The prices of renewable energy sources like solar and wind have fallen below fossil fuels. Source: Our World In Data.
To really maximise your impact, focus on tech that is less widely known. Why? You could propel forward a field that otherwise wouldn’t get off the ground, or wouldn’t get off the ground for a long time.
For example, emissions from cars are only about four times higher than emissions from cement, but there’s much more than four times the focus on electric cars. That means there could be better opportunities to move the needle by greening cement production. We think that means working on the latter could plausibly be better — there might be low-hanging fruit you can pick (take a look at our career review of engineering).
By the same token, working on hot rock geothermal could be higher impact than working on solar or wind energy — though we don’t know because so few people are looking into it.
There’s also value in technology that increases energy efficiency, for example by reducing the costs of building better-insulated buildings. And it’s also important to look at the barriers in deploying and scaling technology that’s already been developed to find potentially neglected ways to lower costs.
You could also work on policy advocacy and leadership. While each country’s emissions are small on their own, successful policy can spread across the world, helping to reduce net emissions.
Unfortunately, we’ve too often advocated for ineffective or intractable policy.
Economists have advocated for carbon pricing for decades. From a simple economic perspective, pricing in the negative externality of emissions should optimally produce the efficient solution. But despite these decades, the net global carbon price is actually negative $10.49 per tonne29 — we’re still subsidising carbon overall.
We haven’t looked into this in detail, but it seems like there could be significant political barriers to carbon pricing that are difficult to overcome.
Instead, we should focus on tractable policies with a track record of success. For example, the UK has almost completely removed its use of coal for power generation through a mix of implementable regulations and subsidies. Other countries like Sweden and France have had huge success in deploying nuclear power, and it’s possible that with appropriate advocacy this could happen elsewhere too.
If you’re interested in advocacy, you might want to read our reviews of policy careers and communications careers more broadly.
Research into carbon removal technology (but not solar geoengineering)
Carbon removal technology, like negative emissions tech or carbon capture and storage, seems pretty neglected compared to green energy (read a popular overview), and could be a crucial way of reducing the effects of our emissions on the climate.
Removing carbon in this way is a form of geoengineering — deliberate intervention in the climate. The other primary form of geoengineering is solar geoengineering (deliberately deflecting sunlight away from Earth to cool the planet down). Solar geoengineering poses potential risks to humanity in itself, given the unprecedented scale of the intervention and the fact that, once in use, solar geoengineering can’t be left untended without disastrous effects. These risks could be larger than the risks from climate change itself, so we think it’s potentially harmful to do work that could advance solar geoengineering.
Geoengineering research of all kinds is mainly done in academia. The Oxford University Geoengineering Programme conducts research into the social, ethical, and technical aspects of geoengineering.
Research on extreme risks from climate change
We don’t think that climate change is likely to cause a catastrophe by itself that would collapse society or kill a substantial proportion (>10%) of the population. But, as we’ve argued, the indirect effects of climate change could contribute to increasing existential threats.
But as we saw above, it’s hard to say exactly how much they contribute — and how to best mitigate those effects. This is because most research hasn’t been focused on extreme-risk scenarios or the interaction between climate and other existential threats.
So, increased investment in some areas in climate research may be able to better inform policymakers, as well as the public at large, about the likelihood of the extreme risks of climate change (both direct and indirect), as well as uncover strategies to reduce those risks.
What research there is on extreme climate change occurs mainly in academia and is funded by basic science funders like the US National Science Foundation. The Centre for the Study of Existential Risk at the University of Cambridge and the Global Catastrophic Risk Institute are conducting research into extreme climate change and possible responses.
For more on what this kind of work might be like, read our career review of academic research.
Key questions we’re unsure about
It’s really difficult to come to robust conclusions about climate change, especially when you’re focusing on the worst possible outcomes.
As a result there are several questions where we’re uncertain about the answers — and, if we had a different answer, our advice could change substantially. These include:
- How hot could it really get? Is our research (or our reasoning) wrong? Would more research into this question give us more certain answers, or is it just too hard?
- How important are extra feedback loops and tipping points that aren’t usually included in climate models, or that we just haven’t thought of yet?
- How big is climate change as an indirect risk factor? Through what pathways? Again, would more research into this question give us more certain answers, or is it just too hard?
- Which areas within climate work (e.g. extreme risks, links to other risks, or specific kinds of green tech) are most neglected relative to their impact?
- CarbonBrief has a range of excellent content and updates on climate change
- Good news on climate change argues that more recent research into climate change has reduced our uncertainty about the chances of extreme risk (take a look at the comments too for some healthy disagreement)
- Founders Pledge’s full report on high-impact climate philanthropy (PDF) (2021)
- Chapters 4 and 6 of Toby Ord’s book The Precipice: Existential Risk and the Future of Humanity (2020)
- Podcast: Mark Lynas on climate change, societal collapse & nuclear energy
- Podcast: Kelly Wanser on whether to deliberately intervene in the climate
- Podcast: Toby Ord on the precipice and humanity’s potential futures, which discusses climate change in Chapters 10 and 11 (2020)
- John Halstead’s informal report on climate change and existential risk
- Climate Shock: The Economic Consequences of a Hotter Planet by Gernot Wagner and Martin Weitzman (2015)
- Chapter 13 of Global Catastrophic Risks, edited by Nick Bostrom and Milan Cirkovic (2008)
- Podcast: Prof Yew-Kwang Ng on ethics and how to create a much happier world — Yew-Kwang Ng is a visionary economist who anticipated many key ideas in effective altruism decades ago and has written about the importance of climate change reduction
- Podcast: Lewis Dartnell on getting humanity to bounce back faster in a post-apocalyptic world
Huge thanks to Goodwin Gibbins, Johannes Ackva, John Halstead, and Luca Righetti for their extremely thoughtful and helpful comments and conversations.
This work is licensed under a Creative Commons Attribution 4.0 International License.
Thanks a lot for this. This is good and detailed work. From my point of view it's a clear improvement on what I've seen from other EA orgs on climate change previously! Keep it up :)
Regarding "key questions you are unsure about": I think the main weakness in this text and your assessment is your lack of critical engagement with the IPCC reports. You state higher up that "nothing in the IPCC’s report suggests that civilisation will be destroyed". This is correct, but it does not follow from this that IPCC suggests that civilisation will NOT be destroyed, or that IPCC reports can be used to argue that point. There are two main reasons:
My prior is therefore that outcomes will almost certainly not be better than what IPCC predicts, but that there is a very good chance that outcomes will be worse. This prior is supported by the fact that the "surprises" in the climate system seem to be in one direction only - things consistently appear worse than researchers have commonly assumed (40 degrees in London, ice melt which appears to move much faster than researchers have modelled, etc). I cannot be completely sure about this, of course, but so far I have found very little evidence which points in the other direction, and cannot see strong reasons to update these priors (even though I would sleep better at night if things weren't as bas as I fear).
Then, regarding what to do. You ask towards the end on things you are unsure about: "Which areas within climate work (e.g. extreme risks, links to other risks, or specific kinds of green tech) are most neglected relative to their impact?"
This is an important question, and I think you slightly miss the mark in your actual recommendations. Regarding green tech and renewable energy etc, there is actually no lack of research or engagement at all. There are LOTS of people doing research on this. One reason is that this is potentially very lucrative, so people want to be in on it. There could arguably be more research on nuclear power, which unfortunately hasn't been favored by policy makers, but here as well there are already many people doing important work.
Where there is NOT a lot of research or money or engagement is in two other areas:
On the first point: If we want to reduce emissions from long distance aviation, one alternative is to come up with new fuels which don't pollute as much. But another alternative is simply to reduce long distance aviation. On the first alternative there is already a lot of well-funded research. Of course the airline companies are interested in keeping going or increasing the number of flights. On the second alternative there is very little research. How would an international transport system with less long-distance flights look like, and more importantly, how do we achieve this, politically and socially? We don't know much about this. There is some research, but not a lot. (see this recent article for example: Banning super short-haul flights: Environmental evidence or political turbulence? - ScienceDirect)
This also applies to other issues - like achieving just transitions for workers in fossil-based industries, driving less, moving away from meat, etc. The reason is probably that there are few funding bodies which have clear interests in lifestyle change - but there are many funding bodies which have an interest in continuing as we do today, only with less emissions.
On the meat issue there is already lots of EA engagement of course - but similar engagement could also be applied to the issue on lifestyle changes in other domains, in developed countries at least. Your comment that focusing on one's carbon footprint "is a distraction" is almost certainly wrong, by the way. There is by now much research which shows that carbon footprint has a huge signaling effect: People simply trust climate advocates more when we focus on our own carbon footprint. Here is one such article, but there are many more: Climate change communicators’ carbon footprints affect their audience’s policy support | SpringerLink
Within EA circles one is well aware of the importance of signaling. This applies just as much to climate change as to other issues.
The second point regarding activism and impact is also important. We actually know very little on how to change things regarding climate change. What activism actually works? EA James Ozden is one of the few who actually does research on this. Radical tactics can increase support for more moderate groups - EA Forum (effectivealtruism.org)
Following up on that, there is also an argument for actually supporting and engaging directly in the kind of activism that we have reason to think may work - non-violent civil disobedience, advocacy groups who work towards lifestyle change, etc. Green advocacy groups are massively underfunded compared to fossil fuel interests, and could well use some influx of smart EAs.
Ok, that's it for this comment. Thanks again for working on the issue.
The link in "This map" is now broken, can it be changed to this one?