771. N. Kaldor to Harrod , 28 May 1938 [a]
[Replies to 769 , answered by 773 ]
15, Mecklenburgh Square, [London] W.C.1 #
28 May 1938
My dear Harrod,
I am sorry that my last letter was difficult to follow. The diagrams of course were only intended for making clear what my position is; not for demonstrating propositions.  For the latter, a more complex geometrical apparatus would be necessary.
From your last letter I am by no means sure how far you accept my argument & how far you reject it. It would be best, perhaps, if I briefly re-stated my position & if you could indicate where you disagree:--
I. A reduction in the rate of interest 1 may induce entrepreneurs (i) to expand the scale of operations, without "deepening" the method of production; (ii) to expand the scale of operations, and "deepen" the method of production, at the same time (You appear to argue that it will "deepen" the method, but leave the scale of operations unchanged--but I am not sure whether you really meant that. By "scale of operations" I mean the amount of output, per unit of time, the entrepreneurs plan to produce). Whether (i) will happen or (ii) depends on the factors that limit the entrepreneurs' scale of operations.
II. A rise in real wages (cet. par, i.e. the rate of interest being given) may induce entrepreneurs (i) to restrict the scale of operations, without "deepening" the method of prod[uction].; (ii) to restrict the scale of operations, and "deepen" methods at the same time. Whether (i) will happen or (ii), again depends on the nature of the factors that limit the ind. firm's scale of operation.
Moreover, changes in scale & changes in method are to some extent "compensatory" in both cases, in the sense that for a given change in real wages or the rate of interest the change in scale will be all the greater the smaller the change in method, & vice versa.
By "real wages" I mean the relation of output & input prices, for any given level of output.
By the "rate of interest" I mean the market rate of interest, or the supply schedule of capital to the individual firm.
Neither a change in real wages nor a change in the rate of interest can leave the scale of operations unchanged; while both of them can, in certain circumstances, leave the method of production unchanged.
Now as far as I can make out from your last letter, you wouldn't quarrel with this. But you insist that a change in method must always involve a change in the marg. supply price of capital, i.e. in the marg. cost of borrowing. This is perfectly true; & I could say that in the same sense it also involves a change in "marg. real wages", i.e. in the price ratio between the last unit of output & the last unit of input. For marg. cost of borrowing = marg. rate of return, & the change in marg. return involves a change in marg. real wages (at any rate if we assume a homog. prod. function). But neither of these statements would be very helpful. Quite apart from the fact that the "marg. cost of borrowing", in my sense, includes risks, both lenders' and borrowers', & borrowers' [b] risk could hardly be treated as the marg. supply price of capital, the sense of talking about changes in the rate of interest, or of real wages, causing this or that, lies in distinguishing between different kinds of changes in data. When I speak of changes in real wages I mean shifts in the demand curve for the product or in the supply curve of labour, or both, which in the special case of perfect competition becomes change in rel. prices, rather than rel. price schedules. In the same way when I speak of changes of the rate of interest, I mean changes in the supply schedule of capital, which in the special case when this schedule is horizontal becomes a change in the rate of interest.
In the sense in which people ordinarily speak of the rate of interest, or of "the structure of interest rates" (the yield of Treasury Bills, Consols, etc.) I do maintain that it is perfectly possible that a reduction in the rate (or in the "rate-structure") should merely have the effect of expanding the scale of production (to "widen" capital) without "deepening" capital. It is quite possible that if banks, through open market operations, enforce such a reduction, and as a result the whole price-level of securities rises, shares of all kinds included, the stimulus thereby given should merely lead to an expansion in the number of "machines" employed, without changing the type of "machines". Provided you agree on this, I am quite prepared to admit that in these very same circumstances the rate of interest in your sense hasn't changed at all, for in the new equilibrium the marg. cost of borrowing for ind. firms, incl. risks, will still be the same as before and so will "marg. real wages". But I should still remain careful & employ inverted commas whenever I speak of changes in the rate--in your sense.
P.S.--Moreover, in precisely the same way as you can argue that in the case contemplated by me (when the method remains the same, despite changes in r. of i., because the marg. rate of return is constant) the change in the rate of interest does not really imply a change in the marg. cost of borrowing, I could argue that in the case contemplated by you (when the method remains the same, despite changes in real wages because the marg. cost of borrowing is constant) the change in real wages did not really imply a change in "marginal real wages". For in that case, entrepreneurs will restrict the scale of operations to precisely that extent that in the new equilibrium, the ratio between the price of the last unit of output and that of the last unit of input (or rather, the ratio between the "marg. revenue" of the last unit of output and the "marg. cost" of the last unit of input) will remain exactly the same as it was before. (You may object that the ratio be[twee]n the marg. revenue of output & the marg. cost of input is not exactly the same thing as "marginal real wages"; but in the same way, the marg. cost of borrowing is not the same thing either as the rate of interest on the marg. unit of capital!) The important point is that if marginal real wages are constant (independent of scale) there cannot be a change in the marginal cost of borrowing, whatever happens to the rate of interest (i.e. to the supply schedule of capital); and in the same way, if the marg. cost of borrowing is constant (independent of scale) there cannot be a change in marg. real wages, whatever happens to the demand schedule for the product or the supply schedule of labour. A change in both is essential (as between two equilibrium positions) for there to be a change in the degree of "deepening". The case where "real wages are raised for manual workers and there will not be a tendency to use more capital per unit of manual workers" is a special case of exactly the same order, as the case where the rate of interest is reduced and there will not be a tendency to use more cap. per unit of labour.
P.S.2. I quite agree with you that in order that real wages, for the system as a whole, should rise, the amount of "other factors" employed must have been increased.  But this is precisely the way in which, in the older cap. theory, the change in real wages was supposed to come about. If you look up Wicksell's description of the process of real capital accumulation (Lectures, vol. I App. on Åkermann  ) you will find that he contemplated that (i) saving first leads to an increase in the number of "machines" of the same type; (ii) this in turn will lead to a rise in wages and a fall in profits (for "machines" have increased relatively to "men") (iii) this in turn makes it profitable to shift the more "durable" (i.e. roundabout) machines, which will involve that "the labourers loose part of, but not all of, this recent increases in wages and the capital goods regain part of, but not all, the value they have just lost". Hence saving leads first to a "widening" of capital, this in turn to a rise in wages, which in turn will lead to a "deepening". This analysis will be correct of course only in the special case of constant marg. returns (to the indiv. firm).
2. Letter 769 , [jump to page] .
3. K. Wicksell, Lectures on Political Economy. Volume One: General Theory, London: Routledge, 1934. Kaldor refers to Appendix 2, "Real Capital and Interest", pp. 258-99.
- a. ALS, ten pages on five leaves, in HP IV-669-688. Photocopy in NKP NK/3/30/86/136-45.
b. Ms: «lenders and borrowers, & borrowers risk».
top of page
Return to index of this section
Go to previous page
Go to next page