The Social Discount Rate is Plausibly Negative
We must make tradeoffs between consumption now, and consumption later. We could choose to not save anything, and consume a lot now, but we would have to consume much less in the future. If we could perfectly predict the future, we would choose the savings rate that maximizes our total amount of consumption. Since we cannot, we prefer consumption which happens now to that which happens later, and the percentage which is needed for us to prefer future to present consumption is referred to as the “discount rate”. A discount rate of zero is equivalent to valuing future consumption 1 to 1 with present consumption; as it rises, we value present consumption more.
We don’t only have individual discount rates. We could be said to have a social discount rate as well. When we are wondering how much we should mitigate climate change, or save for our children, or pay for innovation, we are implicitly answering what that social discount rate should be. But how might we find it?
A simple answer might be that we don’t care about our children at all. We would have a rather high social discount rate in that case. One might also take an egalitarian view, and say that the consumption of people now is just as meaningful as those in the future. This would be a discount rate of zero — $100 now would be traded for $101 with certainty a hundred years from now. That we are uncertain, of course, suggests valuing present consumption now — a bird in the hand is worth two in the bush — although this intuition actually requires presupposing risk aversion. Why we would have risk aversion is unclear.
Another way is to imagine everyone who will ever exist meeting together, and picking what discount rate they want behind a veil of ignorance. They do not know when and where they will be born. Given that many more people will be born in the future, they may well favor social rules which benefit the future more than the present. This is strengthened if we remove the veil of ignorance.
Imagine we could negotiate with the future. How much would they pay us (I assume for ease of exposition that there are no transaction costs, and no Myerson-Satterthwaite problems, so there’s truthful revelation of preferences) for us to save and act in their benefit? Given that they will be much wealthier, collectively and on average, it would seem that their willingness to pay is far higher than ours today. Thus, the efficient thing to do would be to act as though those payments were made, even though it is obviously impossible. This is like Coase’s “The Problem of Social Cost” — the optimal use of a resource with externalities is whatever use people would use it for, if side-payments are frictionless.
I am therefore concerned that abortion is inefficient. We would prefer not to die. I would imagine we would pay a great sum to continue living. Indeed, we do, putting 17% of our GDP toward healthcare. I would also pay a great deal of money to keep myself from being aborted. The efficient thing is to do what would happen if side payments did happen, even if they can’t. Some mothers would be willing to outbid someone, but not many.
Of course, we would be myopic if we focused on fetuses alone. I would also pay quite a lot to be conceived! If abortion prevents mothers from having a child before they’re able to support it, and as a consequence they have many more babies, then abortion has been an unambiguous good. Hypothetical people matter just as much as real people, because we are all in some sense hypothetical.
It may not be healthy to think too hard about morality — I’m not even sure I really believe it. I just try and do the right thing.