I wonder if anyone is working on the possibility of internalizing catastrophic risks into current economies? Or else: why wouldn’t it be useful?
After all, I think that debates on computing and internalizing carbon costs have made climate change policy a more tractable issue (and have helped foster an economic literature relevant for longtermism)… I have some hypothesis on why this wouldn’t apply to other major GCR and x-risks, but I wonder if someone has already written about it.
A paper that Kian Mintz-Woo is working on is relevant: “Incentives for the Long-Term(ist)”
From the abstract: “To address long-term externalities, I propose internalizing long-term externalized costs: according to our best estimates of the long-term costs of an activity or product, this cost should be added.”
You would have to ask him directly where he’s currently at with his draft.
I’ll do this. I think this discussion is getting a lot of track when it comes to climate change. Precisely because low-prob catastrophic scenarios are neglected by climate change IAMs—because they lead to a collapse of the cost-benefit analysis, according to Weitzman (2009) - Stiglitz, Taylor & Stern have been advocating for target-consistent pricing. I wonder if something analogous would be feasible for x-risks.
For instance, there have been many advocates for a robot tax on automation to internalize possible damages to job markets; I never quite bought this idea (I’m not sure we should distinguish types of unemployment when it comes to social insurance), but I find it more attractive if it could be used to fund differential progress.
Owen has done some related work here and here on pricing research externalities.