STANFORD UNIVERSITY 
STANFORD, CALIFORNIA 94305 
DEPARTMENT OF ECONOMICS 
7 March 1981 
% 
.<4 ^ 
.»% 
a 
r *' 
* / % 
i V 
T V -0 
■' ,/ ■/ 
Si tXap 
J >X 4 
.,q 
Dear Josef, 
Thank you very much for your letter and for its enclosures. 
A 
I too look forward to our meeting in Sils. It was funny to see you refer „ 
to me as a voracious reader, since I think of myself as being the very 
opposite — except as far as your writings are concerned. 
Luckily, I got the 'flu just in time to be able to read your two 
papers and found both very interestisng & enjoyable. May I, in that 
connection, ask a few guestions? 
First, I am puzzled by the meaning of your 'reinvestment coefficient’. 
Does it mean anything else than Harrod's warranted groth rate? Why does it 
hinge on savings being generated inside the firm and why should not the 
firm's gross saving give a reinvestment coefficient above one? To the extent ■-* 
that the firm's depreciation allowances allow not only for physical wear & 
tear but also for obsolescence, they can finance net investment not only on 
the Harrod-Badhuri argument in a growing economy, they can finance the net 
investment implied by the replacement of obsolete by more efficient eguipment 
as well (that argument is made by Maurice Scott in his 1976 OEP article). 
At the same time, however, I don't see why savings generated within the firm 
should be conducive to investment. In the US in the present depression, there 
has been a major shift of real income from consumers to the oil companies; 
but the oil companies use their tremendous profits not/for investment as for /so much 
the buying up of already existing companies, which of course has no stimulating 
effect on the economy at all. 
Later on in the same paper, you discuss the stimulating and the depressive 
effects of innovation, each proportionate to the capital stock in existence 
but owing to the lag with which the depressive effect makes itself felt 
(presumably because investment creates add'l capacity with a lag), the stimulating 
effect outweighs the depressing effect in a growing economy. Aren't those 
two effects very similar to Harrod's demand generating and capacity generating 
effects? If so, should not the lag of the capacity generating effect be 
introduced into the Harrod model too? 
I find especially interesting & suggestive your argument (pp.l6ff.) that 
the stimulating effect of innovation is diffused & general, while the depressing 
capacity effect is localized & therefore possibly smaller. But your alternative 
scenario (p.17) I find unconvincing. If the innovator's competitors imitate 
him, won't that lead to an even greater capacity expansion and therefore to 
an even more severe depressing effect, though piesunably with an additional 
time lag? I expected you here to introduce an idea from your other paper, 
product as against process innovation, which presumably has no depressing 
capacity effect. All things considered, however, I like both your papers very 
much. 
'Ll 
You will hear from me again as soon as our summer plans have jelled. 
Until then, all the best, 
Yours, 
P.S. Please see other side for a 
post scriptum. 
£y /U w-W. f ’ v ^ 
\f\ 
w 
v 
tf/l'-Xi- H 
lTV w ie 
£\\ 
y . ; .. •' 
A*