Re: poly: Why interest rates may stay low

From: Hal Finney <>
Date: Tue Mar 03 1998 - 14:28:59 PST

I found Robin's example very interesting and helpful, but there is
still something important I'm having trouble grasping, which is
whether economic growth will increase as the rate of technological
progress increases. Logically, it seems that it should, but Robin's
examples would suggest otherwise.

It appears that Robin suggests two reasons why economic growth rates may
remain near 2-3% as they have historically. This is my interpretation
and any errors here should not be taken to reflect on him.

First, any increase in return on investment beyond this range will attract
large amounts of investment capital, making it difficult to drive interest
rates higher than this. By itself this would not constrain growth but it
would suggest that the increased profits would go elsewhere (to labor
rather than capital?).

More importantly, the prospect of future increases in growth will attract
capital prematurely, forcing wasteful expenditures on immature technologies
in a desperate effort to win the race to new technologies. This premature
investment will dissipate some of the future potential profits, as
R&D funds are spent inefficiently.

It's not clear to me, though, that the effect of this second factor will
limit growth, but rather that it may spread it out. Consider a business
where we have had 3% growth historically, but where we can forecast that at
some future time growth will increase to 10% per year, and this will last
for 20 years. Using a log scale for output, we would have something like

           | OOOOOO
           | OOOOOO
           | OO 3%
           | OO
 (log | OO 10%
  scale) | OO
           | OOOOOO
           | OOOOOO
           | OOOOOO 3%
           | OOOOOO


If this increase is anticipated, then my interpretation of Robin's reasoning
is that investment will flow to the company at an earlier point, such that
the net result is that the stock price increases at 3% a year. We are left
with the following price, shown as P on the same chart:

           | PPPPPP
           | PPPPPP
           | 3% PPPPPP OO
 (log | PPPPPP OO
  scale) | PPPPPP OO 10%
           | PPPP OOOOOO
           | OOOOOO
           | OOOOOO 3%
           | OOOOOO


The price rises as soon as the future increase is anticipated, to maintain
a steady 3% growth rate over time. There may be a > 3% increase as the
information first becomes available about the future increase, but that
will occur at a relatively early point when the price is relatively low and
the change may be lost in the noise.

In this interpretation, the anticipatory effect does not reduce the actual
rate of growth due to technology, but rather the future growth is reflected
ahead of time as increases in the price of productive assets. The net
result is that these assets do roughly increase in price at historical
rates, but that the actual economic output can increase much faster.

One problem with this, though, is the assumption that the high growth
rates will occur for only a limited time. If the high rates were expected
to continue indefinitely, then it would seem that current prices would
be arbirarily high, using this simple graphical model.

(A year or two ago I made a similar point on Extropians, that if we believe
that growth rates will eventually become higher than our personal discount
rates, and stay that way, then rationally we should give stock prices an
infinite value today. My idea was not very well received, but this
graphical approach suggests the same conclusion.)

Received on Tue Mar 3 23:01:45 1998

This archive was generated by hypermail 2.1.8 : Tue Mar 07 2006 - 14:45:30 PST