>>the underlying preference for current consumption has weakened substantially
I have a feeling this is a rabbit hole I don't want to go down :-) - but has the underlying preference for current consumption weakened substantially because of (the author's claim) that people are living longer after retirement, or is it because people feel current consumption isn't giving them their money's worth? When you hand in your dollar, you expect to receive something worth that dollar.
In other words, if people suddenly woke up tomorrow and started accepting gold as currency, will the preference for current consumption be as weak? Or will it return to the previous levels because it is easier to see if you are getting your "unit of currency"'s worth? To be clear, I don't know the answer. But if it is the latter, then people's time preferences may not have really changed.