Hi everyone, I just watched the talk by Daniel Bressler at the global EA conference:
I am thinking that this research might be politically dangerous because it seriuosly underestimates death-rate estimates until 2100 with 4°C of warming:
- Climate change will greatly increase current death rates from (preventable) poverty, hunger, and disease, currently roughly 10 million per year (depending on definition of poverty, cause of death and so on). That is because climate change will seriously affect food and clean water supplies, which is probably the most important effect. Billions of people depend on water from glaciers or harvests that depend on monsoons, for example. The second most important effect may be direct heat (wet-bulb temperatures exceeding skin temperature).
- Climate feedbacks and tipping points were not mentioned, but they are the main problem from a physical climate-science perspective. That's how climate change will get out of control.
- Extrapolations will not work if the prediction is about an unprecedented situation. For that purpose, we need a more top-down approach.
One might add a fourth reason. If the population of Africa approaches 4 billion as predicted at the same time as food and water supplies stagnate, billions of premature deaths due to climate change are possible in Africa alone.
On this basis one might reasonably predict one billion premature deaths globally for 2°C of warming, spread out over a century, plus one additional billion for each additional 1°C, total 3 billion for 4°C. This is very approximate, of course, but that is another point: more precise estimates are hardly possible, especially due to the uncertainty of climate feedbacks. That casts doubt on the usefulness of applying sophisticated economic modeling to this question. I explained these things in more detail here:
The implication is that emissions must now be reduced very fast, and much faster that decided in Paris in 2015. This point could be more important than anything currently being discussed in EA, in terms of the number of human lives that depend on it.
I hope you find these ideas interesting. Let me know what you think.