(For the avoidance of doubt: my previous comment wasn't trying to claim that the "statistical value of a life", measured in such a way, is a good measure of how much a person values their life in non-monetary terms, or anything like that. I was just objecting to the specific claim I was responding to, namely that the "statistical value of a life" is measuring only its value as a future income stream.)
Very very crudely, the value of a small gain in money is something like inversely proportional to the amount you have already, which means that the value of a given amount of wealth increases something like logarithmically with the amount of money, and either of these means that financial institutions like markets care about people's interests roughly proportionally to their wealth. Everything in this paragraph is pure handwaving, but I find it a useful intuition.
Hm, if one did, then perhaps one could use this as a basis for expressing how much people value other things?
Though, I suppose there's the question of like, how would someone trade off risk of dieing within N years vs dying within 2N years. Like, if something would increase one's chances of dieing within 2N years by some quantity, but decrease one's chances of dieing withing N years by some other quantity, at what values would people make or not make that choice? (or, in the other direction).
Depending on how that works out, maybe that would mean that this isn't a great foundation for translating between "how much do they value this (in dollars), at their current income level" and "how much do they value this (in some other unit which is hopefully more natural in some sense)".
Is there something else that would be a better basis?