Letter from Bob Solow

[pdavidso <pdavidso@xxxxxxx>]

While going through my papers and correspondence being sent to the Duke
University Archives, I came across a letter from Bob Solow (dated May 21,
1985) to me.  This letter was in response to a letter of mine commenting on
his article "Economic History and Economics" in the May 1985 AER Papers and
Proceedings in which he claime most economic processes were non-stationary.

Bob wrote:

"On the main point, let me first say that I have always admired that article
of yours on non-ergodic processess, and thought it was right on the button. I
usually think of it in terms of Knightian uncertainty and all that, but yours
is a good way of putting it. Nevertheless when I said "stationarity" I meant
exactly that. My point was that the rational-expectations people like to model
economic agents as statistical processors of the observations they have; and
most, maybe all, of the standard methods for dealing with times series
statistically require stationarity (at least to the second moments). I was
only after tha narrow issue> If time averages and phase averages don't
coincide, that only strengthens the point."

Thought you might find this interesting.

Paul Davidson
Editor, Journal of Post Keynesian Economics
University of Tennessee
SMC 523
Knoxville, Tennessee 37996-0550
phone # (865)974-4221; fax #(561)737-8262;
email pdavidson@xxxxxxx
http://econ.bus.utk.edu/davidsonextra/Davidson.html