Wow this looks excellent. Great work!! I have only skim read it because it is long and my time is short but there are a few unanswered questions that jump out at me in this that it would be great to see a bit more detail on:
Q1. Full consideration of tail risks.
The most convincing arguments I have heard as to why donating to say Cool Earth is better than say AMF is that when you consider the tail risk of global annihilation the cost per life saved of mitigating emissions drops considerably. You discuss tail risk on p1 but on p2 when you are looking at the costs and benefits of different strategies it does not appear that you consider the tail risk, even though that is where most of the expected deaths are likely to lie.
Is there anyway you can quantify the tail risk and build it into the expected costs benefit analysis on p2?
Q2. Indirect tail risks
The tail risk section only looks at the risk of rapid climate change directly causing deaths but are there not also lots of indirect risks that arise from climate change?
(Sorry I am not expert in climate change and this issue confuses me and maybe there is a simple answer, maybe you wrote it and I missed it; but this confuses me)
My understating is that CO2 is the main problem and CO2 stays in the atmosphere for 1000′s of years. Therefore reducing emissions in order to prevent climate change simply delays the problem, unless we actually stop emissions ever happening. So even if we cut our global coal consumption useage to like 25% its current usage but that means that we will take say 100 years rather than 25 years to empty out coal mines and burn all the coal there, this does not prevent climate change, just slows it down. If cool earth protects rainforest now but we expect that over the next 1000 years that rainforest will be destroyed anyway then it has not helped prevent climate change. Does that make sense?
Q4. Quantifying lobbying
Can you quantify the value of lobbying (even roughly) or give tips on how to do this for people who are considering donating to climate lobbying groups and want to make a best guess estimate of how this compares to other charities.
Q5. Where to lobby?
In what countries did you consider lobbying organisations? Is the US the best place for lobbying action?
Q6. The value of putting time into this?
(EA as a whole has been very bad at research into how best to spend spare hours to make the world a better place. Maybe not a question for you but more an 80000Hours or EA Action issue.)
If people have time to give rather than money to give what is the case that lobbying actions in the climate change arena are the best use of a few hours a week of spare time and what is the best way of an individual spending their time in this manner.
Sorry for all the questions. That was more questions than I expected when I started writing and some of them do not seem particularly relevant—I guess I just spewed out all my uncertainties about climate change.
Anyway this is truly a magnificent piece of research (to my layman’s eyes at least) so well done :-)
I haven’t read the articles yet, though I did study climate change as part of my undergraduate and externally, so I’ll have a crack at answering your technical question (Q3).
The point of mitigation is to reduce greenhouse gas emissions (including carbon dioxide and methane) or to capture and store them (number of ways to do this, underground gas to liquid storage, growing trees etc.). CO2 actually has a much shorter residence time in the atmosphere, but it does then get stored in the ocean for up to centuries. Methane is also a big problem, because it has a larger impact on warming (around 20 times greater), but has a shorter residence time (around 8 years in the atmosphere). For this reason, some people propose that mitigation, at least in the short term, should focus on reducing methane gas, as it has a larger short term effect, and if we hypothetically cut methane to zero (not going to happen as there are still natural sources, but for arguments sake), we would see the impact of methane disappear quite quickly, compared to cutting carbon.
To answer the rest of your question, the fact that carbon is transferred from the atmosphere to the oceans, soil and organic matter over time means that there is some particular amount of greenhouse gas (GHG) that can be put in the atmosphere without warming, as the Earth as a whole can absorb it (assuming no man-made carbon capture). This point is called the equilibrium state. We have well and truly exceeded the equilibrium point through man-made emissions, which is why the planet is warming. If we could reduce emissions to the equilibrium emission rate and below that, the planet would, eventually, start to cool. Even if we expect we will only get below equilibrium in 100 years, any reduction in emissions now will mean the atmosphere will have warmed less by the time we get there.
Thanks! Glad you liked it. It’s currently just a preview and not actually published yet, so that’s why some links and functionality may not work (and the post on the model I used is still yet to go up).
In regards to Q1 - I would like to, yeah. When it comes to the probabilities of different levels of warming though, it’s super uncertain. The ~1% chance of 10 degrees of warming is only under one of several possible probability distributions and we really just don’t have any clue which of those distributions is accurate. And in addition to the uncertainty there, we know very little about just how bad those high levels of warming would be for us as there’s minimal research on it, so giving expected values would be a major challenge—one which I’m not sure I’m up to, but which definitely warrants some more research in future. There’s also the values and risk profiles of different donors to think of too—many want direct measurable benefits rather than minor reductions in existential risk somewhere in the future—so even if we got a decent estimate of the expected impact of emission reduction including tail risks, it’d have to be given separately.
Q2 - Largely the same issue as above—seriously difficult to estimate. But as far as mentioning them, that’s certainly something that can be added in before we publish. Cheers for that!
Q3 - Yeah, so reducing present emissions just temporarily won’t really help much (unless it gives us longer to adapt). But when I’m talking about emissions reduction, I mean permanently preventing that quantity of emissions from ever being emitted (e.g. through deforestation). Reducing (or preventing) emissions in this way should not only delay the point at which we reach a given temperature but also reduce the eventual peak temperature. And Michael’s spot on with his reply too (we haven’t looked at methane emissions specifically here but it does seem like it might be possible to reduce methane emissions at roughly $5/tCO2eq through ACE’s recommended animal charities—highly uncertain though).
Q4 - I’ve also just finished an evaluation of the most promising lobbying organisation we’ve found. It should be up sometime soon. We think it might be a slightly better option for donors with a greater appetite for risk, but for others it still seems like Cool Earth is the better option.
Q5 - The US. That seems pretty certain, as they’re not only a massive emitter (2nd worldwide) but it’s pretty widely accepted that a lot of action elsewhere won’t happen if the US doesn’t get the ball rolling. This’ll get mentioned in that other evaluation though.
Q6 - I’m not sure. It probably wouldn’t be a bad use of most people’s time, and the advocacy charity we’ve been looking at does use a lot of volunteers. Then again, they’re currently getting hundreds of thousands of volunteer-hours each year already so it might be more effective to volunteer elsewhere or in a different cause area. I really don’t know though.
Wow this looks excellent. Great work!! I have only skim read it because it is long and my time is short but there are a few unanswered questions that jump out at me in this that it would be great to see a bit more detail on:
Q1. Full consideration of tail risks.
The most convincing arguments I have heard as to why donating to say Cool Earth is better than say AMF is that when you consider the tail risk of global annihilation the cost per life saved of mitigating emissions drops considerably. You discuss tail risk on p1 but on p2 when you are looking at the costs and benefits of different strategies it does not appear that you consider the tail risk, even though that is where most of the expected deaths are likely to lie.
Is there anyway you can quantify the tail risk and build it into the expected costs benefit analysis on p2?
Q2. Indirect tail risks
The tail risk section only looks at the risk of rapid climate change directly causing deaths but are there not also lots of indirect risks that arise from climate change?
Eg. It is likely we will attempt geoengineering but this could go wrong and mess up the environment further. Instability arising from climate change could lead to WW3 or other x-risks (see Broad and narrow strategies for shaping the far future at http://lesswrong.com/lw/hjb/a_proposed_adjustment_to_the_astronomical_waste/)
Q3. What is the point of mitigation?
(Sorry I am not expert in climate change and this issue confuses me and maybe there is a simple answer, maybe you wrote it and I missed it; but this confuses me)
My understating is that CO2 is the main problem and CO2 stays in the atmosphere for 1000′s of years. Therefore reducing emissions in order to prevent climate change simply delays the problem, unless we actually stop emissions ever happening. So even if we cut our global coal consumption useage to like 25% its current usage but that means that we will take say 100 years rather than 25 years to empty out coal mines and burn all the coal there, this does not prevent climate change, just slows it down. If cool earth protects rainforest now but we expect that over the next 1000 years that rainforest will be destroyed anyway then it has not helped prevent climate change. Does that make sense?
Q4. Quantifying lobbying
Can you quantify the value of lobbying (even roughly) or give tips on how to do this for people who are considering donating to climate lobbying groups and want to make a best guess estimate of how this compares to other charities.
Q5. Where to lobby?
In what countries did you consider lobbying organisations? Is the US the best place for lobbying action?
Q6. The value of putting time into this?
(EA as a whole has been very bad at research into how best to spend spare hours to make the world a better place. Maybe not a question for you but more an 80000Hours or EA Action issue.)
If people have time to give rather than money to give what is the case that lobbying actions in the climate change arena are the best use of a few hours a week of spare time and what is the best way of an individual spending their time in this manner.
Sorry for all the questions. That was more questions than I expected when I started writing and some of them do not seem particularly relevant—I guess I just spewed out all my uncertainties about climate change.
Anyway this is truly a magnificent piece of research (to my layman’s eyes at least) so well done :-)
Have also posted it in https://www.facebook.com/groups/1509936222639432/ so hopefully you will get some more feedback.
Disclaimer: i have no expertise in climate change.
Also the link to https://www.givingwhatwecan.org/post/2016/04/modelling-climate-change-cost-effectiveness does not work.
I haven’t read the articles yet, though I did study climate change as part of my undergraduate and externally, so I’ll have a crack at answering your technical question (Q3).
The point of mitigation is to reduce greenhouse gas emissions (including carbon dioxide and methane) or to capture and store them (number of ways to do this, underground gas to liquid storage, growing trees etc.). CO2 actually has a much shorter residence time in the atmosphere, but it does then get stored in the ocean for up to centuries. Methane is also a big problem, because it has a larger impact on warming (around 20 times greater), but has a shorter residence time (around 8 years in the atmosphere). For this reason, some people propose that mitigation, at least in the short term, should focus on reducing methane gas, as it has a larger short term effect, and if we hypothetically cut methane to zero (not going to happen as there are still natural sources, but for arguments sake), we would see the impact of methane disappear quite quickly, compared to cutting carbon.
To answer the rest of your question, the fact that carbon is transferred from the atmosphere to the oceans, soil and organic matter over time means that there is some particular amount of greenhouse gas (GHG) that can be put in the atmosphere without warming, as the Earth as a whole can absorb it (assuming no man-made carbon capture). This point is called the equilibrium state. We have well and truly exceeded the equilibrium point through man-made emissions, which is why the planet is warming. If we could reduce emissions to the equilibrium emission rate and below that, the planet would, eventually, start to cool. Even if we expect we will only get below equilibrium in 100 years, any reduction in emissions now will mean the atmosphere will have warmed less by the time we get there.
Hi Sam,
Thanks! Glad you liked it. It’s currently just a preview and not actually published yet, so that’s why some links and functionality may not work (and the post on the model I used is still yet to go up).
In regards to Q1 - I would like to, yeah. When it comes to the probabilities of different levels of warming though, it’s super uncertain. The ~1% chance of 10 degrees of warming is only under one of several possible probability distributions and we really just don’t have any clue which of those distributions is accurate. And in addition to the uncertainty there, we know very little about just how bad those high levels of warming would be for us as there’s minimal research on it, so giving expected values would be a major challenge—one which I’m not sure I’m up to, but which definitely warrants some more research in future. There’s also the values and risk profiles of different donors to think of too—many want direct measurable benefits rather than minor reductions in existential risk somewhere in the future—so even if we got a decent estimate of the expected impact of emission reduction including tail risks, it’d have to be given separately.
Q2 - Largely the same issue as above—seriously difficult to estimate. But as far as mentioning them, that’s certainly something that can be added in before we publish. Cheers for that!
Q3 - Yeah, so reducing present emissions just temporarily won’t really help much (unless it gives us longer to adapt). But when I’m talking about emissions reduction, I mean permanently preventing that quantity of emissions from ever being emitted (e.g. through deforestation). Reducing (or preventing) emissions in this way should not only delay the point at which we reach a given temperature but also reduce the eventual peak temperature. And Michael’s spot on with his reply too (we haven’t looked at methane emissions specifically here but it does seem like it might be possible to reduce methane emissions at roughly $5/tCO2eq through ACE’s recommended animal charities—highly uncertain though).
Q4 - I’ve also just finished an evaluation of the most promising lobbying organisation we’ve found. It should be up sometime soon. We think it might be a slightly better option for donors with a greater appetite for risk, but for others it still seems like Cool Earth is the better option.
Q5 - The US. That seems pretty certain, as they’re not only a massive emitter (2nd worldwide) but it’s pretty widely accepted that a lot of action elsewhere won’t happen if the US doesn’t get the ball rolling. This’ll get mentioned in that other evaluation though.
Q6 - I’m not sure. It probably wouldn’t be a bad use of most people’s time, and the advocacy charity we’ve been looking at does use a lot of volunteers. Then again, they’re currently getting hundreds of thousands of volunteer-hours each year already so it might be more effective to volunteer elsewhere or in a different cause area. I really don’t know though.