You considered actual (past) spending to assess neglectedness, as is often done, but should the focus be on expected future spending?
For the left-of-boom spending, nothing is changed, because by definition there is no spending of this kind after the start of a nuclear war.
However, expected right-of-boom spending is dominated by the spending in the event of a nuclear war. For example, assuming a probability of 0.1 % of full scale nuclear war in 2023, and a right-of-boom spending of 1 T$ if it materialises, the expected right-of-boom spending would be 1 G$. This is 1.95 and 59.8 times your estimates for the left-of-boom and right-of-boom spending between 2012 and 2022. So there is a sense in which left-of-boom spending is more neglected.
I can see how right-of-boom spending may be more effective before nuclear war. Coming up with good interventions in little time may be harder, since some thinking may not be parallelizable. On the other hand, right-of-boom spending after the start of the nuclear war might be more effective due to more information being available (e.g. about the countries involved, and their motives for starting the war). I do not know what is the overall multiplier accounting for all of this, and I am not confident it favours right-of-boom spending at the current margin.
I suppose the same argument applies to other causes. For example, for global warming, it is true interventions targetting warming higher than 3 ºC are neglected if we look at past spending. However, this underestimates the respective expected future spending. Resources would start flowing towards such interventions if our best guess for warming by 2100 moved upwards. Does Founders Pledge’s robust diversification approach account for this?
Great analysis, Christian!
You considered actual (past) spending to assess neglectedness, as is often done, but should the focus be on expected future spending?
For the left-of-boom spending, nothing is changed, because by definition there is no spending of this kind after the start of a nuclear war.
However, expected right-of-boom spending is dominated by the spending in the event of a nuclear war. For example, assuming a probability of 0.1 % of full scale nuclear war in 2023, and a right-of-boom spending of 1 T$ if it materialises, the expected right-of-boom spending would be 1 G$. This is 1.95 and 59.8 times your estimates for the left-of-boom and right-of-boom spending between 2012 and 2022. So there is a sense in which left-of-boom spending is more neglected.
I can see how right-of-boom spending may be more effective before nuclear war. Coming up with good interventions in little time may be harder, since some thinking may not be parallelizable. On the other hand, right-of-boom spending after the start of the nuclear war might be more effective due to more information being available (e.g. about the countries involved, and their motives for starting the war). I do not know what is the overall multiplier accounting for all of this, and I am not confident it favours right-of-boom spending at the current margin.
I suppose the same argument applies to other causes. For example, for global warming, it is true interventions targetting warming higher than 3 ºC are neglected if we look at past spending. However, this underestimates the respective expected future spending. Resources would start flowing towards such interventions if our best guess for warming by 2100 moved upwards. Does Founders Pledge’s robust diversification approach account for this?
Update. Christian replied to this comment here.