I think the "normative" discount rate (the one that everyone should use) can only be 0 for experiantial utility. This argument is valid even if money should be discounted, as money doesn't give me experiental utility, but opportunity to buy it, and the earlier the money, the more opportunity I have. This argument is also valid if you observe experiments where people want an apple rather now than next year (because of trust issues as mentioned in your cited paper). Additionally, the reason I said "normative", is because people make mistakes by overvaluing the present which results in observed discount rates. If we assume interpersonal comparisons with future generations, this does not neccesarily mean we should mitigate climate change or existential threats all the times like crazy, as you can assume concave utility and future generations will be much happier in general and any marginal improvement in their life will be almost infinite times less than improvement in mine.
This is what Frank Ramsey (who developed the first temporal utility model) wrote on the topic in his paper in 1928: “It is assumed that we do not discount later enjoyments in comparison with earlier ones, a practice which is ethically indefensible and arises merely from the weakness of the imagination.”
if im interpreting table 9 right in the Busse et al paper the discount rates dont seem ludicrously high. Leaving out the negative(!) ones, the discount rates implied by new car prices are mostly around 0-6% unless demand elasticity is assumed very large. Higher discount rates implied by used car prices maybe rationalizable by credit constraints as you suggest
Objection 1: The reason for programs that transfer wealth to the elderly is the same as the reason for programs that transfer wealth to disabled working-age people or to children exist: through no fault of their own, they don't have the ability to trade labor for productive income, and we don't want them to suffer as a result. If 85 year olds were as productive as 45 year olds, the ones who couldn't live off of retirement savings could just get jobs instead of needing everyone else to take special care of them.
Objection 2: There's a difference between the effects of a one-time surprise transfer payment and the effects of a known policy of implementing welfare payments. If you know that all your wealth is going to be given away at 70, you act differently than you would if you didn't expect that to happen, and I suspect that these kinds of payments would really mess with the prevailing market interest rates and people's apparent time preferences.
Objection 3: Large involuntary transfer payments from older people to younger people do, in fact, happen all the time in the real world. They're called inheritances.
Objection 4: Suppose you offer me a choice between a dollar yesterday and a dollar twenty years ago. This seems like the same question as whether I'd prefer to have a dollar tomorrow or 20 years in the future, but it's not quite the same. One argument for preferring to have gotten the dollar earlier in life is market interest rates: choosing to get the dollar 20 years ago strictly dominates getting it yesterday, because I could have invested that money 20 years ago and ended up with more than one dollar yesterday.
Now suppose that whenever I get a dollar from a time traveling economist, I use it to buy a tasty Diet Coke instead of drinking tap water. Since I will not have invested the $1, my actual choices are between a Diet Coke 20 years ago and a Diet Coke yesterday, instead of a choice between "the opportunity to have chosen between a Diet Coke 20 years ago or a Diet Coke plus investment income yesterday" and a Diet Coke yesterday. The dominance argument fails to apply, and it's not at all obvious to me that market interest rates imply that I should prefer to have the memory of consuming a Diet Coke on Feb 28, 2005 instead of the memory of consuming a Diet Coke on Feb 27, 2025.
I SPECIFICALLY had no children based on the premise of being Self Sufficient until I die. So YOUR children or THEIR children mean NOTHING to me. As a matter of fact, I'd be perfectly comfortable REMOVING ANY child or mother or father who seek to disenfranchise me. 😑 Just sayn 🤗
Where do estate taxes fall here? It seems like a way of taxing a person at their oldest possible point. Conversely, not having them seems like a way to have private consumption become even further into the future than an individual lifespan.
For long-term decisions, we really need to distinguish the social rate of time preference from the opportunity cost of capital investment. We should also take account of the fact that (hopefully) those future consumers will be far richer on average than we so their consumption should count for less than ours.
Really enjoyed this. Important stuff.
“Empty personhood” is I believe a philosophical concept you’re reaching for re: agents
I think the "normative" discount rate (the one that everyone should use) can only be 0 for experiantial utility. This argument is valid even if money should be discounted, as money doesn't give me experiental utility, but opportunity to buy it, and the earlier the money, the more opportunity I have. This argument is also valid if you observe experiments where people want an apple rather now than next year (because of trust issues as mentioned in your cited paper). Additionally, the reason I said "normative", is because people make mistakes by overvaluing the present which results in observed discount rates. If we assume interpersonal comparisons with future generations, this does not neccesarily mean we should mitigate climate change or existential threats all the times like crazy, as you can assume concave utility and future generations will be much happier in general and any marginal improvement in their life will be almost infinite times less than improvement in mine.
This is what Frank Ramsey (who developed the first temporal utility model) wrote on the topic in his paper in 1928: “It is assumed that we do not discount later enjoyments in comparison with earlier ones, a practice which is ethically indefensible and arises merely from the weakness of the imagination.”
if im interpreting table 9 right in the Busse et al paper the discount rates dont seem ludicrously high. Leaving out the negative(!) ones, the discount rates implied by new car prices are mostly around 0-6% unless demand elasticity is assumed very large. Higher discount rates implied by used car prices maybe rationalizable by credit constraints as you suggest
I erred, sorry.
the heckman estimates are huge though
Objection 1: The reason for programs that transfer wealth to the elderly is the same as the reason for programs that transfer wealth to disabled working-age people or to children exist: through no fault of their own, they don't have the ability to trade labor for productive income, and we don't want them to suffer as a result. If 85 year olds were as productive as 45 year olds, the ones who couldn't live off of retirement savings could just get jobs instead of needing everyone else to take special care of them.
Objection 2: There's a difference between the effects of a one-time surprise transfer payment and the effects of a known policy of implementing welfare payments. If you know that all your wealth is going to be given away at 70, you act differently than you would if you didn't expect that to happen, and I suspect that these kinds of payments would really mess with the prevailing market interest rates and people's apparent time preferences.
Objection 3: Large involuntary transfer payments from older people to younger people do, in fact, happen all the time in the real world. They're called inheritances.
Objection 4: Suppose you offer me a choice between a dollar yesterday and a dollar twenty years ago. This seems like the same question as whether I'd prefer to have a dollar tomorrow or 20 years in the future, but it's not quite the same. One argument for preferring to have gotten the dollar earlier in life is market interest rates: choosing to get the dollar 20 years ago strictly dominates getting it yesterday, because I could have invested that money 20 years ago and ended up with more than one dollar yesterday.
Now suppose that whenever I get a dollar from a time traveling economist, I use it to buy a tasty Diet Coke instead of drinking tap water. Since I will not have invested the $1, my actual choices are between a Diet Coke 20 years ago and a Diet Coke yesterday, instead of a choice between "the opportunity to have chosen between a Diet Coke 20 years ago or a Diet Coke plus investment income yesterday" and a Diet Coke yesterday. The dominance argument fails to apply, and it's not at all obvious to me that market interest rates imply that I should prefer to have the memory of consuming a Diet Coke on Feb 28, 2005 instead of the memory of consuming a Diet Coke on Feb 27, 2025.
I SPECIFICALLY had no children based on the premise of being Self Sufficient until I die. So YOUR children or THEIR children mean NOTHING to me. As a matter of fact, I'd be perfectly comfortable REMOVING ANY child or mother or father who seek to disenfranchise me. 😑 Just sayn 🤗
Where do estate taxes fall here? It seems like a way of taxing a person at their oldest possible point. Conversely, not having them seems like a way to have private consumption become even further into the future than an individual lifespan.
For long-term decisions, we really need to distinguish the social rate of time preference from the opportunity cost of capital investment. We should also take account of the fact that (hopefully) those future consumers will be far richer on average than we so their consumption should count for less than ours.