RE: Savings in a Fast-Growth Economy

Robin Hanson (hanson@econ.berkeley.edu)
Thu, 22 Apr 1999 09:54:20 -0700

Billy Brown wrote:
>I think the question is, given:
>
>1) A future in which lifespans become very long (>200 years).
>2) At least reasonably good (by current standards) economic conditions.
>3) A population in which a majority of people over age 50 wish to retire.
>
>Is there anything we can say about the likelihood of their being able to do
>so? We can readily see that compound interest will allow individuals to
>amass substantial sums of money over the course of several decades, so with
>decent interest rates the answer would seem to be yes. However, it was
>suggested that such a large pool of savings might cause interest rates to
>collapse, due to a relative shortage of profitable investment
>opportunities.

To an economist, this is still awkwardly asked. If they wished to retire badly enough, they would, and just live poorly. But retirement is not directly what people want. They want to consume stuff, and to avoid boring work, just as they always have. If we took today's life cycle and expanded it by a factor of five, from 60 years to 300 years, and personal discount rates and interest rates fell by a factor of five, then you would expect the same life cycle behavior, spread out over a longer time. People would work till ~300 then retire.

The question is what, if anything, would break this scaling hypothesis. Wealth probably doesn't much. We still work most of our lives, even though we are vastly richer than our ancestors. Personal discount rates might well not scale though. Evolution seems to have hard coded in us a discount rate of a factor of two per historic generation time of ~25 years. This suggests interest rates won't fall below ~3%, unless our subjective sense of time is changed somehow. This is what suggest earlier retirement.

Over 200 years, the economy could change a lot. But assuming we don't run out of technological progress, then there should continue to be lots of profitable investment projects.

With rapid population growth as now, the 100+ year old folks would be a small fraction of the population, so folks could retire at 100, and live off savings for the next 200 years, without distorting the economy much. If most of the population were retired, however, we get the problem of what exactly did their savings buy to let them consume so much. Most income (~70%) is now from labor, so unless we get AI, or the economy changes drastically to more favor other forms of capital, then the old folks need to "own" lots of labor income of younger folks to make this work. Absent slavery, you'd need young folks paying off debt from loans used to develop their human capital, or some such.

As usual, the right thing to do here it make one's assumptions explicit, put them in a simple economic model, and see what happens that way.

Robin Hanson

hanson@econ.berkeley.edu     http://hanson.berkeley.edu/   
RWJF Health Policy Scholar             FAX: 510-643-8614 
140 Warren Hall, UC Berkeley, CA 94720-7360 510-643-1884 after 8/99: Assist. Prof. Economics, George Mason Univ.