779. N. Kaldor to Harrod , 13 June 1938 [a]
[Replies to 776 , answered by 780 ]
The London School of Economics, Houghton Street, W.C.2 #
13 June 1938
My dear Harrod,
I am sorry that I couldn't answer your letter before.
From your last letter it appears that our difference is reduced to a single point. You say:--There is asymmetry, for a change in marg[inal]. real wages cannot alter roundaboutness unless it is accompanied by a change in the marg. cost of borrowing, whereas a change in the marg. cost of borrowing can alter roundaboutness even if marg. real wages remain the same.  If you agreed that a change in both is necessary to alter the degree of r[ound].a[bout].ness you would presumably also agree that there is no asymmetry, but perfect symmetry.
Now this is precisely the point where I differ. In equilibrium, the marg. cost of borrowing must be equal to the marg. rate of return; and a change in the marg. rate of return presupposes a change in marg. real wages. If marg. real wages are constant, it is the marg. cost of borrowing which must adjust itself (via a change in the scale of output) to the standard set by real wages; whereas if the marg. cost of borrowing is constant, it is marg. real wages which must adjust themselves (in the same way) to the standard set by the rate of interest. Which is the "active" & which is the "passive" factor, in any given case, depends on which factor is independent (or more <+> independent) of the scale of output. 1
You appear to argue that the marg. rate of return could fall, even if marg. real wages are constant. I don't see how it could. There can be diminishing returns to labour (and other prime factors) if equipment is constant, or to equipment, if labour is constant, but how can there be dim. returns if prices are given & both are increased in the same proportion? 2 It is not a question of a rising marg. cost curve, since this presupposes given equipment.
You could take refuge with the dim. returns due to entrepreneurship. But I don't think this would save the situation. For if you assume that the costs of the ind. firm are rising relatively to the costs of the industry, in equilibrium each firm must still be operating on optimum scale; and if this optimum scale remained unchanged, the increment in total output must come about by an expansion of the number of firms & not by an incr. in output of the ind. firm. (In other words, the firm's marg. rate of return must be equal to the average rate--since the average is maximized when firms are producing on an optimal scale--& if therefore this average is given, the marg. rate of return is also given). In order that the ind. firm's optimal scale should change, & hence its equi. output should change, it is necessary that some factor-price should have risen in terms of output prices: if only the price of entrepreneurial services. This in turn is a particular case of a rise in real wages. (I regard this whole concept of rising costs to the firm due to entrepreneurship as a practically unimportant factor in the determination of the ind. firm's output, but this is a different matter).
I disagree also with the last para. of your letter where you state that if the supply schedule of capital is given "this gives unequivocally the proportion in which waiting on the one hand and labour etc. on the other should be mixed, regardless of what the real rate of wages etc. are".  No; for unless real wages are given, you don't know what point of the schedule is relevant. If on the other hand you assume that the schedule is infinitely elastic, & hence the marg. cost of borrowing is the same for all outputs, I could argue that if real wages are the same for all outputs you have thereby determined equally unequivocally the "proportion", regardless of the supply schedule of capital.
No, I am afraid I still maintain that there is nothing in this asymmetry-business. It appears [b] to me exactly parallel to the [c] age old controversy, which at last appears dead & buried, whether supply or demand determines prices. In arguing that it is the rate of interest (or the marg. cost of borrowing) which determines the degree of roundaboutness, you seem to be taking the same line as the man who argues that marg. cost of prod. determines price. Both propositions are true on the assumption that the supply curves in question are infinitely elastic. But there is more reason to assume infinitely elastic supply curves plus falling demand curves in the case of commodities, <than> there is in the specific case under discussion. For in our case, it appears equally likely (or rather unlikely) that the dem. curve is infinitely elastic as that the supply curve is such.
But quite apart from such theoretical niceties, I was mainly interested in the practical question : what will be the typical firms' reaction to (a) a fall in the rate of interest, (i.e. yield of Consols); (b) a fall in prices, relatively to costs? Will it be widening (or "narrowing", as the case may be) or deepening? The answer, to which I think you agree, is [d] that it can be either, or both, in both cases, depending on the factors that limit the scale of investment of the typical firm.
2. See letter 780 , [jump to page] .
3. Letter 776 , [jump to page] .
- a. ALS, six pages on three leaves, in HP IV-669-688. Photocopy in NKP NK/3/30/86/113-18.
b. Ms: «It is appears».
c. Ms: «to the to the».
d. Ms: «agree that».
2. Harrod marked this passage with a cross in the margin. 
top of page
Return to index of this section
Go to previous page
Go to next page