> But a society is not an individual.
Of course, that's why it's an analogy. It's to see more clearly the implications that flow from the differences between the thing and its analog.
> Presumably the income of a whole society increases at a fairly
> constant rate, unlike what happens to an individual from age 20 to 45
> or 50. So this analogy makes no sense. You could make an exception,
> maybe, for a rapidly industrializing society, but that's not relevant
> to the U.S. or any other mature rich economy.
What I tried to say is that the U.S. economy is like a Dorian Gray kind of individual. Not a regular individual but one that -- at least for the time being -- looks as if it will stay young forever. Some individual members of society will die and other others will be born. So the nation as a whole will reproduce itself continuously.
Now, if you look at the demographic evolution of the U.S., you can see that because of immigration the aging of the population is somewhat contained -- in contrast with Japan or some European countries. So the Dorian Gray comparison is not entirely inept.
A demographically mature society will not necessarily shrink to zero. Chances are it will stabilize. But whatever, for the time being, we don't expect the U.S. labor force to evaporate in the course of a few generations.
> If you run, say, a deficit of 2% of GDP (revs = 18%, exp = 20%), in
> an economy growing at 5% a year, and roll over the old debt and issue
> new to fund the deficit, you end up with a debt/GDP ratio of 30%
> after 25 years (if you start year 0 with no debt). Debt service
> climbs steadily from 0% to 1.5% of GDP. Aren't there better uses for
> 1.5% of GDP?
You need no Ponzi. Roughly, the average growth rate of the public debt (new debt plus rollover of old debt included) has to be less than the growth rate of GDP (assuming gov't revenues are a fixed percentage of GDP). Otherwise it's Ponzi. A lot depends on expectations of economic growth, fiscal standing, and interest rates. In principle, an economy can have a growth rate of the public debt up to the expected growth rate of GDP after you adjust both for risk. In practice, given the fluctuating nature of the economy, the effective risk premium is sizable.
> But bond interest is an upward redistribution - see the Reagan years.
That also depends on how the gov't spends the borrowed money. How did Reagan spend the borrowed money? Star Wars? Giving taxes back to the rich? If he had re-built infrastructure or spent in basic education and health care, the results could have been very different.